The Intervertebral Disc

Irving M.
Shapiro
Makarand V. Risbud
Editors
The
Intervertebral Disc
Molecular and Structural

Studies of the Disc in Health
and Disease
123
The Intervertebral Disc
Irving M. Shapiro Makarand V. Risbud
Editors
The Intervertebral Disc

Molecular and Structural Studies
of the Disc in Health and Disease
Editors
Irving M. Shapiro Makarand V. Risbud
Department of Orthopaedic Surgery Department of Orthopaedic Surgery
and Graduate Program in Cell and Graduate Program in Cell
and Developmental Biology and Developmental Biology
Jefferson Medical College Jefferson Medical College
Thomas Jefferson University Thomas Jefferson University
Philadelphia Philadelphia
PA PA
USA USA
ISBN 978-3-7091-1534-3 ISBN 978-3-7091-1535-0 (eBook)

DOI 10.1007/978-3-7091-1535-0
Springer Wien Heidelberg New York Dordrecht London
Library of Congress Control Number: 2013943012
Springer-Verlag Wien 2014

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this
legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically
for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work.
Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publishers location, in its current version, and permission for use must always be obtained from Springer. Permissions
for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution
under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of publication, neither
the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may
be made. The publisher makes no warranty, express or implied, with respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

Foreword
Current understandings of the biology of the intervertebral disc is predicated on comprehen-

sion of the processes of aging in the human body. Its generalized nature has often led to flawed
surgical strategies with focal attention on what is now understood to be a widespread and uni-
versal process throughout the spine. The global nature of aging and degenerative changes of
the spine has led to an epidemic of disease and disability affecting primarily the aged. As our
population achieves greater longevity, these changes become increasingly manifest and the
demands for improved understanding become more vital. Age changes in the disc are not dis-
similar from those found throughout the connective tissues of the body including the skin,
aorta, ligaments, and peripheral joints. Biomechanical weakness of the annulus can lead to
local breakdowns in the integrity of the disc with nuclear herniation and nerve root compres-
sion. For those discs that are unresponsive to medical therapy, surgical intervention can be
extraordinarily successful providing dramatic and usually lasting relief. When the annular
changes are more generalized and the facet joints become incompetent, instability can occur
provoking a condition known as degenerative spondylolisthesis. Rather than nerve root com-
pression, this provokes instability and back pain and again is reasonably amenable to surgical
stabilization. Generalized mechanical insufficiency can provoke arthritic changes, narrowing
of the spine, and neurologic symptoms associated with the entity known as spinal stenosis.
Again, if localized, this is amenable to surgical decompression and possible stabilization.
Diffuse age-related disc degeneration is provocative of axial spine pain and substantial dis-
ability and yet is the least responsive to surgical and, indeed, nonsurgical intervention. This
wide spectrum of degenerative changes are predicated on aging and matrix degeneration that
only later provoke mechanical disorders.
Thus, a true understanding of spinal disease requires the study of the molecular and cell
biology of the normal, aging, and pathologic spine and the pathogenesis of neurologic and
non-neurologic pain syndromes and careful controlled studies of both surgical and nonsurgical
interventions.
In this authoritative book, the reader is privileged to benefit from an elegant and careful
educational process moving from basic biology to common disease entities and the intended
hope for ultimate biologic regeneration. This should be a mandatory reading for all of those
who hope to understand and effectively treat what is todays leading cause of skeletal dis-
ability, i.e., diseases of the intervertebral disc.
Richard H. Rothman, MD, PhD

James Edwards Professor and Founder,
The Rothman Institute at Thomas Jefferson University,
Philadelphia, PA, USA
v
Introduction
The major reason for embarking on the daunting task of editing a book on the intervertebral
disc is that the available literature is either too clinical for most newcomers to the field or too
fragmented for the expert. Accordingly, there is a critical need for a book that will provide a
holistic overview of advanced topics in molecular, mechanical, developmental, and cellular
aspects of intervertebral disc research, while serving as a useful text for graduate students,
postdocs, and fellows. It should be noted that the target audience is not confined to basic sci-
entists as the import of much of the information contained in the book impacts the need of
many neurosurgeons and neurologists, orthopedic surgeons, pain and rehabilitation specialists,
physicians, chiropractors, and kinesiologists. These individuals work tirelessly to understand
and expand the repertoire of treatment modalities available to relieve the pain and loss of func-
tion associated with the degeneration of the intervertebral disc, a widespread condition that
afflicts a huge percentage of the global population.
Indeed, in comparison with the plethora of books on the clinical management of spinal
disease, only one book has been published within the past 50 years that is entirely focused on
the disc. To provide continuity with the past, the preface of the current book is written by one
of the authors, the eminent surgeon-clinician scientist, Dr. Richard Rothman. A comparison of
his earlier book with the current volume shows just how far knowledge of the disc and the
pathogenesis of disease has advanced in the past half century. However, even now, despite the
explosion of interest in degenerative disc disease, from an investigative point of view and in
comparison with cartilage and bone research, discal biology in health and disease can be
regarded as the last frontier of connective tissue research. Obviously, there is still much to be
accomplished and hopefully this book will energize investigators and clinicians to define the
limits of current knowledge and to develop novel insights into the many outstanding questions
concerning the function of the intervertebral disc and the cause of disease.
To assemble this book, the editors have brought together an international galaxy of experts,
many of whom have influenced their own discipline in a very significant way. The contributors
were asked to summarize the current state of their field while at the same time to critique cur-
rent research strategies, approaches, and the interpretation of accepted and projected therapies.
So as to preserve their voice, the editors have tried to maintain, almost intact, each of the
contributions. While this has allowed the reader to view the topic through the mind of the con-
tributor, it has resulted in some duplication of the written word that hopefully will be
forgiven.
The book is divided into three parts. The first section deals with the basic biology, develop-
mental biology, and biomechanics of the normal healthy disc, topics that have received consid-
erable study although some glaring gaps in knowledge still exist. For example, there is debate
concerning cell type in the aging disc: does it contain viable and functional notochordal cells
or are the embryonic cells replaced by other cell populations? The second part is devoted to
intervertebral disc disease and disc herniation, raising questions concerning epidemiology and
pathogenesis. The authors question whether discogenic pain is directly linked to the degenera-
tive state. Following on from the discussion of discogenic pain are chapters dealing with surgi-
cal and nonsurgical modes of treatment. Authors of these chapters ask the following question:
are the advances in knee and hip surgery transferable to the disc? Since all clinical treatments
vii
viii Introduction
are dependent on the availability of model systems, the last section of the book discusses the
value of in vitro systems and small and large animal models to mimic the environment of the
human disc. Closely aligned with these concerns are discussions of the use of transgenic mouse
models, stem cell biology, and gene therapy to promote disc repair. In the chapter on the future
use of tissue engineering systems to effect biological repair, it is stated that the technology will
likely be personalized to the individual and influenced by the extent of disease. Concomitant
with the development of tissue engineering approaches for biological repair, the important
concept is reiterated that it is necessary to delineate the mechanism(s) leading to back pain so
that the twin goals of all new therapies are to ameliorate pain and maintain function.
The editors have a number of people to thank for their help in preparing this book. First and
foremost, they must acknowledge the responsiveness of the individual contributors who have
risen to the challenge of preparing each of their chapter with such flare and imagination. The
editors also thank F. Michael Angelo, MA, University Archivist for the reproduced images of
rare books, courtesy of the Archives and Special Collections at Thomas Jefferson University,
Philadelphia. Second, the authors thank Katrina Lenhart, Wilma McHugh, and the team at
Springer for all of their help. Lastly the editors must thank their own families for giving up
time for completion of this endeavor.
IMS dedicates this book to two unique individuals: Dr. Michael Bush OBE, punnist extraor-
dinaire and a great friend and cousin, who devoted much of his life to preventing and treating
AIDS in Africa. His untimely death brought great sadness to his family and to the patients who
depended on him. And to Dr. Susan H. Shapiro who traded her spinal pain for resolution, hope,
and courage. The fortitude with which she endured the pain now empowers her to follow her
heart and intuition. MVR dedicates this book to his parents Swati V. and Vinayak Y. Risbud for
their unconditional love. In addition, MVR thanks Rashmi, the love of his life, for her constant
encouragement and support and Aditya and Akshay for the fun and happiness they generate
each day.
Finally, appealing to our inner futurist, it is fun to contemplate the focus of a similar book
on the disc published 50 years hence. Indeed, will there be a book or will new knowledge
transmit directly from the bench into ganglia and neurons of the central nervous system? Will
surgery be regarded as an inhuman and savage approach to treating a chronic health problem?
Will the new molecular therapies be based on those described in this book or will there be
approaches that are as yet unimaginable at this time? Or will nutrient excess and lack of exer-
cise promote devolution of the human spine to the notochord of a sessile wormlike creature
discussed in Chap. 1? If we can avoid the last scenario, there is reason to hope that the pain of
disc disease will be like quinsy and polio, remnants of a bygone era eliminated by the ground-
breaking work of many of the clinicians and scientists who contributed to this book.
March 2013 Irving M. Shapiro

Philadelphia, PA, USA Makarand V. Risbud
Contents
Part I Biomechanical and Molecular Studies of the Intervertebral Disc
1 Introduction to the Structure, Function, and Comparative Anatomy

of the Vertebrae and the Intervertebral Disc . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Irving M. Shapiro and Makarand V. Risbud
2 The Intervertebral Disc: Overview of Disc Mechanics. . . . . . . . . . . . . . . . . . . . 17
Daniel H. Cortes and Dawn M. Elliott
3 Development of the Intervertebral Disc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Megan K. Cox and Rosa Serra
4 Proteoglycans of the Intervertebral Disc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
James Melrose and Peter Roughley
5 Collagen and Other Proteins of the Nucleus Pulposus, Annulus Fibrosus,
and Cartilage End Plates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Fackson Mwale
6 Microenvironmental Control of Disc Cell Function: Influence
of Hypoxia and Osmotic Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Makarand V. Risbud and Irving M. Shapiro
7 The Effects of Mechanical Forces on Nucleus Pulposus
and Annulus Fibrosus Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Jeffrey C. Lotz and Adam H. Hsieh
8 The Role of the ADAMTS Proteins in the Intervertebral Disc . . . . . . . . . . . . . 125
Jason C. Ho, James Wylie, and Suneel S. Apte
Part II Intervertebral Disc Disease: Pathogenesis and Current

Treatment Modalities
9 Epidemiology of Lumbar Disc Degeneration . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

Yue Wang and Michele C. Batti
10 Genetic Basis of Intervertebral Disc Degeneration. . . . . . . . . . . . . . . . . . . . . . . 157
Anita Yee and Danny Chan
11 Pathogenesis of Intervertebral Disc Degeneration . . . . . . . . . . . . . . . . . . . . . . . 177
Stephen M. Richardson, Anthony J. Freemont, and Judith A. Hoyland
12 Imaging Modalities for Studying Disc Pathology . . . . . . . . . . . . . . . . . . . . . . . . 201
Ari Borthakur and Ravinder Reddy
13 Surgical Indications for Lumbar Degenerative Disease . . . . . . . . . . . . . . . . . . . 213
Ravi R. Patel, Jeffrey A. Rihn, Ravi K. Ponnoppan, and Todd J. Albert
ix
x Contents
14 Spinal Motion Restoration Devices for the Degenerative Disc . . . . . . . . . . . . . 225

Daniel G. Kang, Melvin D. Helgeson, and Alexander R. Vaccaro
15 The Nonsurgical Treatment of Back Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Isaac L. Moss, Howard S. An, Francis H. Shen, Zemin Li,
Gunnar B.J. Andersson, and Yejia Zhang
16 Back Pain and Disc Degeneration: Are They Really Linked? . . . . . . . . . . . . . . 261
Kjell Olmarker
17 Clinical Features and Pathobiology of Chordoma . . . . . . . . . . . . . . . . . . . . . . . 277
John A. Abraham, Brian Neuman, and Francis J. Hornicek
Part III Models of Disc Disease and Biological Regeneration
18 Large Animal Models of Disc Degeneration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291

Shyam A. Patel, Christopher K. Kepler, Thomas P. Schaer, and D. Greg Anderson
19 Intervertebral Disc Herniation: Pathophysiology and Emerging Therapies . . 305
Beth A. Winkelstein, Kyle D. Allen, and Lori A. Setton
20 The Sand Rat (Psammomys obesus obesus) Model of Spontaneous,
Age-Related Intervertebral Disc Degeneration. . . . . . . . . . . . . . . . . . . . . . . . . . 327
Helen E. Gruber and Edward N. Hanley Jr.
21 Use of Knockout and Transgenic Mouse Models in Disc Research. . . . . . . . . . 341
Laura Mangiavini, Rita Gerard-ORiley, and Ernestina Schipani
22 Intervertebral Disc Culture Models and Their Applications
to Study Pathogenesis and Repair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
Svenja Illien-Jnger, Benjamin A. Walter, Jillian E. Mayer, Andrew C. Hecht,
and James C. Iatridis
23 Use of Stem Cells for Regeneration of the Intervertebral Disc . . . . . . . . . . . . . 373
Daisuke Sakai and Joji Mochida
24 Gene Therapy Approaches for Disc Regeneration . . . . . . . . . . . . . . . . . . . . . . . 385
Zulma Gazit, Nadav Kimelman-Bleich, Olga Mizrahi, and Dan Gazit
25 Enhancing Disc Repair by Growth Factors and Other Modalities . . . . . . . . . . 401
Won C. Bae and Koichi Masuda
26 Tissue Engineering of the Intervertebral Disc . . . . . . . . . . . . . . . . . . . . . . . . . . 417
Rita Kandel, Paul Santerre, Eric Massicotte, and Mark Hurtig
About the Editors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
Part I
Biomechanical and Molecular
Studies of the Intervertebral Disc
Introduction to the Structure, Function,
and Comparative Anatomy of the Vertebrae 1
and the Intervertebral Disc
Irving M. Shapiro and Makarand V. Risbud
The spine is composed of vertebrae, and it extends from the head down to the loins. The vertebrae are all
perforated, and, above, the bony portion of the head is connected with the topmost vertebrae, and is
designated the skull.
Aristotle
Contents 1.1 Introduction

1.1 Introduction ..................................................................... 3
1.1.1 Evolutionary Considerations ............................................. 3 1.1.1 Evolutionary Considerations
1.1.2 Development of the Vertebrae and Intervertebral Disc ..... 4
1.2 Anatomical and Molecular Structure The goal of this introductory chapter is to provide an
of the Intervertebral Discs .............................................. 5 overview of the design, evolution, and basic characteris-
1.2.1 Form and Function of the Intervertebral Discs ................. 5
1.2.2 Spinal Curvature................................................................ 6 tics of the disc and the vertebrae that comprise the human
1.2.3 Gross Morphology and Dimensions of the Disc ............... 8 spine. As with any survey, the state of current knowledge
1.2.4 Tissues of the Intervertebral Disc...................................... 8 reflects the work of earlier cohorts of individuals whose
1.3 Vertebral Structure ......................................................... 10 insightful observations relied almost entirely on observa-
1.3.1 Cervical Vertebrae ............................................................. 10 tion, argument, and inductive reasoning. Over the centu-
1.3.2 Thoracic Vertebrae ............................................................ 11 ries, sequential observations by men like Aristotle,
1.3.3 Lumbar Spine .................................................................... 11
1.3.4 The Sacrum and Coccygeal Bones.................................... 12
Vesalius, Hunter, and Winslow have all contributed to
understanding how the oversized human head can restric-
1.4 Vertebrae and Intervertebral Discs of Animals ........... 12 tively swivel on the multiple bones of the vertebrate spine
1.4.1 Anatomical Considerations ............................................... 12
1.4.2 Conservation of Vertebral Number ................................... 13 and in doing so provide our species with its huge biologi-
cal advantage.
1.5 Summary of Critical Concepts Discussed
in the Chapter .................................................................. 14
It needs to be acknowledged that the spine as we know it
with the intervening intervertebral discs is a relatively late
References ..................................................................................... 14
phylogenetic development in the animal kingdom. It was
preceded by the appearance of a stiff rodlike structure, the
notochord. In animals that lack backbones, the notochord
provides rigidity and some resilience to the organism, pro-
motes formation of an extended shape, and protects the
overlying spinal cord. The defining characteristic of verte-
brates, the backbone, first appeared in the fossil record
about 500 million years ago, during the Ordovician period.
While details of the transition (notochord to spine) are
missing, the 500-million-year-old tiny Middle Cambrian
I.M. Shapiro (*) M.V. Risbud
Department of Orthopaedic Surgery and Graduate Program fossil chordate, Pikaia, possessed a notochord that sepa-
in Cell and Developmental Biology, rated the distinct head and tail regions; Haikouichthysa
Jefferson Medical College, Thomas Jefferson University, small early Cambrian fish-like fossilexhibited well-
1025 Walnut Street, Suite 511 College Bldg,
developed eyes as well as muscle blocks typical of early
Philadelphia, PA 19107, USA
e-mail: makarand.risbud@jefferson.edu; vertebrates (Shu et al. 2003). Box 1.1 shows the metameric
irving.shapiro@jefferson.edu structure of Pikaia.
I.M. Shapiro, M.V. Risbud (eds.), The Intervertebral Disc, 3

DOI 10.1007/978-3-7091-1535-0_1, Springer-Verlag Wien 2014
4 I.M. Shapiro and M.V. Risbud
are to allow movement of the individual vertebrae, transmit

Box 1.1 forces between vertebrae, and serve as hydrodynamic shock
absorbers.
In humans and other primates, the spine permitted adop-
tion of a vertical posture facilitating the transition from arbo-
real to terrestrial locomotion. The upright bipedal stance
afforded evolutionary advantages including extended three-
dimensional vision: enhanced depth perception would be
expected to enhance manual dexterity, which in turn would
promote skills linked to tool creation. That these same
influences also promoted weaponization added to the unique-
ness of the human race and its determination to limit its own
growth and development. Away from the appendicular skel-
eton, the stable, strong, flexible, and vertical spine permitted
Fossil of Pikaia gracilens, a 505-million-year-old creature, evolutionary changes in the bones of the skull, allowing
found in the Burgess Shale fossil beds in Canada. There is
evidence of a notochord, a dorsal nerve, and myotome- marked cranial growth and development. Thus, over time,
like structure. Pikaia was originally thought to be a chor- the head, albeit balanced precariously at the tip of the verti-
date which would thus make it an early vertebrate ancestor cal spine, together with the bones of the arms, ribs, and legs,
(Courtesy of Smithsonian Institution) would undergo phenotypic alterations that characterize pri-
mates and humans. Moreover, the change in the biomechani-
cal status of the appendicular skeleton would impact the size,
The appearance of the spine probably signaled the most shape, and depth of the pelvis. These evolutionary changes
critical event in evolution of higher organisms. The stimulus provided animals with an enormous biological advantage,
for invertebrate chordates to develop the complex mineral- moving the organism away from the wormlike characteris-
ized metameric structure that characterizes the vertebrate tics of our distant ancestors to the frenetic and often random
subphylum is still unknown; even less understood is how activities of modern day bipedal hominids.
evolutionary pressures prompted the development of the Other chapters of this book will ask the following ques-
intervertebral disc, an event that permitted rapid locomotion tions: Why did these transitions take place, and what or how
and flexion. Remarkably, evidence is mounting that this evo- are the biomechanical forces accommodated by the skeleton
lutionary jump might be the result of the development and and the musculature? What gene clusters are altered to sup-
expression of microRNAs (Iwama et al. 2013). Rather than port this critical evolutionary change, and what is the fate of
viewing this type of transformation as a slow evolutionary the notochord itselfcan notochordal remnants influence the
process, Garstang (1928) has proposed that both cephalo- functional and developmental biology of the spine? Hopefully,
chordates and vertebrates evolved along separate pathways insights generated by these developmental, molecular,
in prehistory. This worker proposed that as a result of neo- mechanical, physiological, and biochemical studies of the
tenic1 evolution, our ancestor may have been a sessile, ascid- spine will provide answers to questions concerning the health
ian wormlike organism. and function of the intervertebral discanswers that could
As organisms evolved a mineralized vertebrate axial skel- not be derived through extant anatomical and pathological
eton, the biological advantages offered by the spine motion analysis.
segment provided functions that profoundly influenced the
activities of other organ systems. Not surprisingly, aside
from allowing extension of the body with some flexibility, 1.1.2 Development of the Vertebrae
the vertebral bone protects the spinal cord. Other advantages and Intervertebral Disc
of the vertebral bones are that they provide sites for attach-
ment of the axial skeleton to the appendicular bones via the The vertebrae develop from individual ossification centers
pectoral and pelvic girdles; additionally, the attachment of which are well documented historically (Kerkring 1717;
muscles and ribs to vertebrae facilitates functional changes Albinus 1737; Rambaud and Renault 1864). Probably the
required for locomotion and respiration. With respect to the most detailed report in the twentieth century was by Peacock
discs that separate each of the vertebrae, specific functions (1951). The reader is urged to review the latter report for more
details; the developmental biology of the intervertebral disc
and the vertebrae is discussed in great detail in Chap. 3.
1
Term used to describe the retention in the adult of traits or phenotype The vertebral bodies are formed by fusion of sclerotome
expressed in the immature state. from two adjacent somites: thus, tissue from the caudal
1 Introduction to the Structure, Function, and Comparative Anatomy of the Vertebrae and the Intervertebral Disc 5
portion of one sclerotome fuses with cranial sclerotome of of spinal anatomy and the intervertebral disc in health and
the succeeding somite. The dense connective tissue of the disease was performed by one of the most prolific anato-
two halves of each sclerotome becomes two centers of chon- mists of the nineteenth century, Hubert von Luschka. In his
drogenesis. At each putative vertebrae, two more chondro- monograph Die Halbgelenke des menschlichen Krpers
genic centers appear laterally and grow backwards to form (1868), von Luschka described the gross and microscopic
the cartilage precursor to the neural arch. During this phase structure of the intervertebral discs from birth to death.
of development, the notochord becomes compressed by the Almost at the same time, Humphrey (1858) in his book A
pressure exerted by the cartilage and may persist for a while Treatise on the Human Skeleton provided a detailed descrip-
as a mucoid streak. However, between the developing ver- tion of each of the discs. He reported the looping fibrils of
tebrae, notochordal tissue is retained and subsequently forms the annulus fibrosus and noted the absence of blood vessels
the intervertebral disc. At these sites, notochordal cells in the nucleus and inner annulus fibrosus. Studies of age
become enclosed in a dense ring of connective tissue, the changes in the disc were subsequently noted by Henle
putative annulus fibrosus. Noteworthy, some notochordal (1872), Poirier and Charpy (1899), Fick (1904) and Petersen
cells may remain in the cartilage; at a later time, cells buried (1930), and Bohmig (1930). As far as we can tell, the earli-
in the bone of the centrum may serve as a site for tumor forest comment on the relationship of the disc to the notochord
mation (see Chaps. 3 and 17). The nucleus pulposus is thus was reported by the Austrian anatomist Schaffer early in the
formed early in fetal life from notochordal elements and twentieth century (1910).
grows rapidly in late fetal life and early infancy. By birth, it
occupies half of the intervertebral space in the lumbar region,
and by 1 year it occupies almost three quarters of the space. 1.2.1 Form and Function
It is thought that there is some remodeling of the interverte- of the Intervertebral Discs
bral space early in life (Taylor 1975).
By the seventh week of life, the cartilage undergoes endo- Depending on age, time of day, occupation, and disease state,
chondral ossification. Dorsal and ventral blood vessels invade the discs make up approximately 1520 % of the length of
the two cartilage anlagen and trigger their replacement with the spinal column. Aside from absorbing biomechanical
bone. Subsequently, the anterior and posterior portions of the forces, each disc permits movement of the spinal column.
calcified centrum fuse. Along the anterior and lateral periph- Undoubtedly, flexibility decreases with age, while spinal
ery of the vertebrae, cartilage plates appear to form the apo- movements at all stages of life can be severely limited by
physis. This is the site for insertion of the fibers of the annulus disease. Since vertebrae themselves are relatively inelastic,
fibrosus. As the centrum ossifies, the cartilage anlage of the movement in the spine is mediated notably by the tissues of
neural arch is replaced by bone. The two sides of the arch the intervertebral disc. Although the mobility of contiguous
fuse and then join together with the centrum. The process vertebrae (motion segments) can be viewed as limited, the
begins in the upper cervical region and extends caudally. The integrated motion of the 33 intervertebral discs together with
laminae are also formed in cartilagethey join together after movement at the zygapophyseal joints permits all of the criti-
birth and then fuse with the rest of the vertebrae between the cal movements of the spine without compromising nerve or
third and seventh years of life. Vertebrae growth is mediated muscle function.
by chondrogenic activity at the growth plate. Actually, as the The famous English anatomist Henry Gray (18271861)
centrum has two centers of growth, it should be labeled as a classified articulations between vertebrae as amphiarthroses
synchondrosis. Histologically, prior to closure in the 17th in which the contiguous bony surfaces are either connected
25th year, a well-defined zone of hypertrophic chondrocytes by broad flattened discs of fibrocartilage, of a more or less
is visible. Once growth has ceased, the only remaining carti- complex structure. By definition, these joints permit very
lage is the endplate. little motion. Shapiro et al. (2012) compared the structure-
function relationships of both the intervertebral disc and syn-
ovial joints. On first consideration, the intervertebral disc
1.2 Anatomical and Molecular Structure could be seen as being very different from the generic syn-
of the Intervertebral Discs ovial joint. However, on reflection, the separate tissues of the
intervertebral disc are very similar to that of the diarthrodial
Medieval anatomists were the first to recognize that the ver- joint: both types of joints are lined by cartilage, they are lim-
tebrae were separated by soft gristle-like structures, the ited by an external ligament, and the joint space contains
intervertebral discs. In his intricate drawings of the spine, molecules that promote lubrication (lubricin and hyaluronan)
Winslow (1776) provided a detailed description of the disc, and elevate the osmotic pressure (aggrecan). Indeed, even
which considering the limitations posed by the distortions the presence of a band of nucleus pulposus tissue across the
of hand lenses was remarkably accurate. Another analysis joint is not out of line with what is known of complex
diarthrodial joints that contain cartilage and fibrocartilage a c

discs and menisci. Related to the function of the nucleus pul-
posus and the inner annulus, it is not yet clear whether a dis-
tinct synovial-like membrane exists in the intervertebral disc.
Whether inner annulus is derived from the notochordal
sheath has not been determined. Nevertheless, like the cells
of the synovium, the resident disc cells do have the ability to
mount a robust defense against bacterial attack (Nerlich et al.
2002; Jones et al. 2008).
In terms of movement, the current classification of the
disc as an amphiarthrosis would indicate very limited mobil-
ity. However, biomechanical studies of the motion segment
with or without contributions from the zygapophyseal joints
indicate that there is wide range of motion between verte-
brae. Moreover, the actual movement of the cervical, tho-
racic, and lumbar vertebrae includes flexion-extension, axial
rotation, and lateral bending, as well as translatory motions.
These three-dimensional movements are more in line with
those of a diarthrodial joint rather than an amphiarthrosis
where movements are slow and motion is limited. Probably
the major difference between appendicular diarthrodial
joints and the axial intervertebral joints lies in their develop-
ment. Although the joints originate from different mesen- b
chymal elements, the nucleus pulposus is derived from a
unique embryonic tissue, the notochord; deletion studies
indicate here too there are considerable similarities in the
expression of genes that govern organ development and
maturation. Recent investigations indicate that joint forma-
tion and even function are dependent on the expression of a
number of genes including those of the Hox family, BMPs,
and GDF5 (Brunet et al. 1998; Archer et al. 2003; Pacifici
et al. 2005). Indeed, deletion of Ext1 influences not just the
development of limb joints but also the formation of the
intervertebral disc (Mundy et al. 2011). This topic is consid-
ered further in Chap. 3.
Based on the overt structural and functional similarities
between the intervertebral disc and the synovial joint and
recognizing that while some differences exist between these
articulations, it would seem logical to place the disc in the
same grouping as the diarthrodial joint. Further, since the Fig. 1.1 Spinal curvature: kyphosis and lordosis. Anterior-posterior
radiographs of the spine showing (a) kyphosis (excessive posterior
intervertebral motion segment displays movement in three bending of the thoracic motion segments) and (b) lordosis (extreme
dimensions and the spine itself provides further rotatory anterior bending of the lumbar and often the cervical spine). (c) The
movements, Shapiro et al. (2012) were of the opinion that it complete spine showing the natural curvature in the cervical, lumbar,
should be classified not as an amphiarthrosis, a slightly and sacral regions (From Bougery and Jacob (1833). Plate 6)
moveable joint, but as a complex polyaxial joint.
biomechanical forces mediated through the pull of muscles,
ligaments, and gravity. Encoded curvatures are seen in the
1.2.2 Spinal Curvature cervical, lumbar, and sacral regions of the spine. On adop-
tion of a vertical stance, and with maturation, these curva-
While the intervertebral discs and the zygapophyseal tures become more distinct (Fig. 1.1c). However, about
joints provide sites for vertebral motion, the overall shape 23 % of the population exhibit deficits in axial curvature,
of the spine as well as curvatures in specific regions of the which vary from simple bending with little functional
spine is dependent on intrinsic genetic factors as well as implications to excessive bending which impacts not just
a b c
d e f
g h i
Fig. 1.2 Human vertebrae. (ac) show vertebrae from the cervical TP transverse process, VF vertebral foramen, SP spinous process, TF
spine (below C2); (df) show vertebrae from the thoracic region of the transverse foramen, SAS superior articular surface, SCF superior costal
spine; (gi) show lumbar vertebrae. a, d, and g anterior-posterior aspects facet, TCF transverse costal facet, AP accessory process, SAF superior
of the vertebrae; b, e, and h are superior views; c, f, and i are lateral articular facet (From Bougery and Jacob (1833). Plates 8 and 9)
views of the spine. VB vertebral body or centrum, P pedicle, L lamina,
locomotor activities, but other critical functions associ- condition is termed lordosis (Fig. 1.2b). While these latter
ated with the spinal nerves. The hunchback spine, conditions are deviation in the anterior-posterior (cephalic-
kyphosis, is due to excessive posterior bending of the tho- caudal) axis of the spine, abnormal bending is also seen in
racic motion segments (Fig. 1.2a); when the cervical and the lateral (side-to-side) dimensions. Scoliosis can affect
lumbar anterior spinal curvatures become excessive, this any part of the spine; the most common regions are in the
thoracic and the lower lumbar spine. These exaggerated 1.2.3 Gross Morphology and Dimensions
musculoskeletal warps in spinal architecture are evident of the Disc
almost entirely in human populations even in royalty
(Richard III); their occurrence in rodents is infrequent. The sizes of the discs in the human skeleton have been
Thus, from an experimental viewpoint, rodents and lago- assessed by a number of investigators, especially in rela-
morphs make useful models to investigate the molecular tionship with age, underlying conditions, and responses
control of spinal curvature. to surgery. Disc thickness can be assessed by radiography
Clinical analysis of the types of abnormal spinal bending and other forms of imaging analysis. Frobin et al. (1997)
indicates that the most common form of this condition is made a determined effort to measure the disc and verte-
idiopathic, i.e., of unknown origin. Nevertheless, the etiol- brae height using archived radiographic measurements of
ogy of this condition is likely to be multifactorial as both the spine. This approach was complicated by a number of
environmental and genetic factors have been implicated. factors that included artifacts due to image distortion,
A second form of scoliosis is neuromuscular which is sec- axial rotation and lateral tilt, and even magnification. To
ondary to other conditions, such as cerebral palsy or a myo- account for these problems, algorithms were developed
pathy. In the elderly, abnormalities in axial bending are often that generated dimensionless parameters. The study
due to degenerative disc disease and spondylolisthesis. showed that lumbar vertebrae and discs were larger in
Possibly, the most intriguing form of scoliosis is congenital males and females and in males there appeared to be no
in origin, a rare condition that is evident early in childhood or little impact of age. More recently, magnetic resonance
(usually within the first 68 weeks). Radiographically, the imaging (MRI) was used to provide direct information on
spine exhibits fused vertebrae, single or multiple hemiverte- the discs of seven healthy males aged 2230 (Belav
brae, a vertebral bar, block vertebrae, and wedge-shaped or et al. 2011).
butterfly vertebrae. If left untreated, almost all of these con- Some general comments about disc dimensions are as fol-
genital anomalies result in deformities and loss of normal lows. Disc height (cephalic-rostral dimensions) varies
function. Since the anomalies occur early in development, according to the spinal region. In the cervical spine, the disc
this form of scoliosis has been linked to patterning, particu- height is about 3 mm, whereas in the lumbar spine, it is
larly during the period of somitogenesis (Chal and Pourquie 917 mm; in the thoracic spine, the thickness is about 5 mm.
2009). In the cervical spine, the discs are thicker in the anterior
As will be discussed in considerable detail in Chap. 3, region than posterior, thus helping to provide the curvature
somitogenesis occurs at a very early stage in development that is characteristic of the neck. In the thoracic spine, the
and is the process whereby the mesoderm of the developing discs are of constant thickness, whereas in the lumbar spine,
embryo undergoes a carefully timed segmentation process; they are again thickest anteriorly. Radiographs have been
somites are generated that specify skeletal muscles, dermis, used to assess disc parameters in animals most commonly
vertebrae, ribs, and annulus fibrosus. Very recent work by used in spine research (OConnell et al. 2007).
Pourquie (2011) has shown that the trigger for the rhythmic
production of somites involves three major signaling path-
ways: Notch (Jiang et al. 2000), Wnt/b-catenin (Dequeant 1.2.4 Tissues of the Intervertebral Disc
et al. 2006), and fibroblast growth factor (Benazeraf et al.
2010) which are integrated into a molecular circuit. The The major functional role of the disc is mechanical: it
oscillatory activities of this circuit generate a highly coordi- allows movements between the axial and appendicular
nated developmental event that serves as a traveling wave of skeleton and the head; it accommodates applied loads; and
gene expression along the anterior-posterior axis of the to some extent the disc protects the spinal cord and nerve
developing embryo. Pourquie (2011) refers to this synchro- roots. The discs themselves are complex tissues comprising
nized change in gene expression in the pre-mesodermal cells an outer circumferential ring of fibrocartilage, the annulus
as the segmentation clock. Clearly any activity that inter- fibrosus which encloses a central proteoglycan-rich core,
feres with coordinated gene expression and the development the nucleus pulposus. The nucleus is sandwiched caudally
of the waves of gene oscillations will impact somitogenesis and cephalically by the cartilage endplates of the contigu-
which in turn will influence the subsequent formation of the ous vertebrae. Since details of the biochemical, develop-
vertebrae and the intrinsic curvature of the axial skeleton. mental, and biomechanical aspects of each of the disc
Although this system was developed from studies in mice, it tissues are provided in considerable detail in other chapters
is most likely that these new understandings will impact our of this book, the sections below merely highlight the major
understanding and ultimately the treatment of congenital characteristics of the endplate cartilage, nucleus pulposus,
scoliosis. and the annulus fibrosus.
1.2.4.1 Annulus Fibrosus With respect to the extracellular matrix, nucleus pulpo-
As an introduction to these topics, it is worth noting that the sus cells secrete aggrecan, as well as collagens I and II. The
annulus can be divided into an inner fibrocartilagenous region matrix also contains collagens IX and XI, and collagen X
and an outer or peripheral fibrous zone (Souter and Taylor has also been reported to be present during degeneration.
1970). It was reported that the outer annulus fibrosus is com- Because of the presence of aggrecan, the disc exhibits a high
posed of very well-defined collagen I fibers that bundle to osmotic pressure; moreover, since it has no blood supply,
form long parallel concentric lamellae. Marchand and Ahmad the oxygen tension within the disc is very low. These limita-
(1990) showed that the number of fiber bundles varies from 20 tions have prompted the Risbud group to note that nucleus
to 62. The thickness of lamellae varies both circumferentially pulposus cells tune their metabolism to the available oxy-
and radially and increases markedly with age, location, and gen supply (see Chap. 6 for details). In this case, nucleus
vertebral type. The central annulus fibers are inserted into the pulposus cells evidence almost complete reliance on the
endplate cartilage, while those at the periphery are anchored to glycolytic pathway to generate metabolic energy (Agrawal
the vertebral bone. In terms of collagen organization and cell et al. 2007).
content, this region is not unlike tendon or ligament.
The inner annulus fibrosus represents roughly 50 % of the 1.2.4.3 Endplate Cartilage
total radial thickness. Designated by some workers as the The caudal and cephalic ends of the disc are covered by a
transitional zone, it differs substantially from the outer layer of cartilage, the endplate. This thin layer of hyaline car-
region. Compared with the outer annulus where the cells are tilage is maximally thick in the newborn and thins with age;
elongated and fusiform and extend in the long axis of the in the adult, the actual width is about 0.51 mm. It serves not
fibrils, the cells of the inner annulus are spherical in shape just as an interface between the soft nucleus pulposus and the
and many resemble chondrocytes. These cells are few in dense bone of the vertebrae, but as a biomechanical barrier
number with short processes. A further difference between that prevents the disc from applying pressure directly to the
the inner and outer annulus is their chemical composition. bone. It is the presence of the cartilage layer that provides the
The inner annulus contains collagens I and II. While aggre- motion segment with its joint-like characteristics. Some
can is present in both regions of the annulus, decorin and authorities believe that the cartilage also plays a role in main-
biglycan are found mainly in the outer annulus. The other taining the viability of cells of the nucleus pulposus (Dahia
protein of significance is elastin which accounts for 2 % of et al. 2009). Structurally, the endplate resembles articular
the dry tissue weight. cartilage. Thus, it contains chondrocytes embedded in an
aggrecan-rich and collagen II extracellular matrix. Although
1.2.4.2 Nucleus Pulposus the cells do not undergo terminal differentiation, collagen X
The nucleus pulposus is derived from the notochord and may be present in the central region of the endplate perhaps
notochordal cells remain in the tissue after birth and into in relationship to focal areas of endochondral bone forma-
adult life. During development, the nucleus is highly cellu- tion. The endplate transitions into bone through a region of
lar: after birth, the number of cells is reduced; in the adult, calcified cartilage.
the cell density is very low. The histology of the nucleus pul- In his review of the cartilage, Moore noted that vascular
posus cells is unique and complex: large cells arranged channels penetrate the cartilage, but at maturity the vessels
mainly in clusters and separated by an abundant extracellular become narrow, constricted, or even obliterated. It is likely
matrix. Among the large notochordal cells, much smaller that this change impacts the nutrient supply to both the carti-
cells possibly derived from the notochordal sheath can also lage and the disc (Moore 2000). Crock and Yoshizawa (1976)
be seen. The large cells appear to have numerous vacuoles, reported that the central region of the endplate where there is
which has prompted some authorities to describe them as a high concentration of channels is freely permeable to small
physaliphorous. Probably the most complete TEM analy- molecules. On the other hand, Nachemson et al. (1970) noted
sis of the nucleus of the adult rabbits was described by Gan that at the tissue periphery, the cartilage is much less perme-
et al. (2003). These workers showed that the nucleus pulpo- able to low molecular weight dyes. Clinically, it is not
sus contained cell clusters embedded in a proteoglycan- uncommon to note that the central region undergoes sclero-
collagen matrix. The cells exhibited a well-defined Golgi sis or mineralization with alterations in the mechanical prop-
system, an extensive endoplasmic reticulum, and a complex erties of the cartilage. When this occurs, nucleus pulposus
vesicular system filled with beaded structures (proteogly- tissue can be forced through the endplate into the underlying
cans). Neither necrotic nor apoptotic cells were evident. bone of the vertebrae. This phenomenon is known as
A remarkable finding was that the cells contain few if any Schmorls nodes which Schmorl himself considered to be
mitochondria. A defining characteristic of the cells was the linked to degenerative changes at the cartilage bone interface
presence of numerous cytoplasmic processes. (see Box 1.2).
1.3 Vertebral Structure

Box 1.2
Since the book is devoted to the intervertebral disc, there is Christian Georg Schmorl (18611932): To spine sur-
little need to review the detailed anatomy of each of the ver- geons, the name Schmorl is synonymous with Schmorls
tebrae. Instead, we herein provide broad brush strokes that nodes, small protrusions of nucleus pulposus tissue
delineate the general features of human vertebrae; this is fol- which herniate through the endplate cartilage. They are
lowed by a few comments about individual vertebrae and the often associated with degenerative disc disease, and
sacrum. Detailed images of each of the vertebrae are shown while they can be painless, they can cause inflammatory
in Figs. 1.2 and 1.3. changes in the underlying bone marrow. Details of the
At first sight, the architecture of the vertebrae appears to work that Schmorl performed come through the writings
be very complex, each bone being riddled with numerous of Ormond A. Beagle, an American surgeon who spent
nooks, crannies, protrusions, and extrusions. However, the time with Schmorl at the Friedrichstadt Krankenhaus in
basic organization of the 24 articulating bones of the spine Dresden, Germany. He reported that Schmorl removed
is quite simple: the vertebral structure reflects its two pri- every spine for examination at postmortem. In a 5-year
mary functions, articulation with contiguous vertebrae and period, he had collected about 7,000 spines and many
protection of the spinal cord. From an anatomical viewpoint, were preserved in the museum. Schmorl reported on
a canal is formed as bone is deposited around the cord. This many problems of the anatomy and pathology of the
canal, the vertebral foramen, houses and protects the spinal spine and the intervertebral disc.
cord. The remaining structure of the vertebrae forms in the Hubert von Luschka aka Hubert Luschka (18201875):
caudal-cephalic direction a boat-like shape (albeit designed Born in Konstanz, Germany, he is the eighth of 12 sons.
by a drunken engineer), while the anterior-posterior axis He was a student of both pharmacology and medicine
exhibits a very inexact pyramidal-like structure (albeit at the University of Freiburg and the University of
designed by a heat-affected Pharaonic architect). The base Heidelberg. Luschka is considered among the major
of the pyramid is comprised of a robust bone, the centrum or anatomists of the nineteenth century and was the author
body, while the sides of the pyramid form a bone arch or of a multivolume textbook on surgical anatomy and
lamella (the hull) that surrounds the spinal cord. The apex of numerous research publications. Testaments to the
the arch extends backwards to form the spinous process (the extent of his work are the multitude of structures named
keel). This process is very well developed in the thoracic after him, especially the recurrent nerves of Luschka
spine where it serves as a site for attachment of the powerful (meningeal branches of spinal nerves that pass through
muscles of the back. Projecting upwards and forwards from the intervertebral foramen and innervate the zygapo-
the base of the lamellae are transverse processes (retractable physeal facet joints and the annulus). He described
stabilizers) which are sites of origin of the pedicles that what are now considered to be tears of the annulus.
form a base for the articulating zygapophyseal (facet) joints: Also with regard to the spine, he found that individuals
the superior (cephalic) articulating process articulates with lose height when they stand and that the height of an
the zygapophyseal joints of the contiguous cephalic verte- individual decreases with age. He discovered the unco-
brae; projecting downwards and backwards from the lami- vertebral joints (Luschka joints) which he called half
nae is the inferior articulating process from which a facet joints present in the cervical spine usually between C3
joint is formed with the contiguous caudal vertebrae. and C6. These pseudo joints are formed between the
Portholes at the junction of the fin and the lateral sta- vertebrate bodies of contiguous vertebrae.
bilizers provide openings, intervertebral foramina, for From Tubbs et al. (2011)
the nerves that flow from and into the spinal cord. In terms
of general anatomy, other than C1 and C2, the largest por-
tion of a typical vertebra is the bony centrum, the weight-
bearing region of the vertebrae. With increasing distance 1.3.1 Cervical Vertebrae (Figs. 1.2 and 1.3)
from C3, there is a significant increase in the robustness of
the centrum and the vertebrae, thus the lumbar vertebrae In line with the generalized numbering system of the indi-
and its centrum are larger than vertebrae of either the tho- vidual regions of the spine, the cervical vertebrae are sequen-
racic or cervical spine. In cervical and even upper thoracic tially numbered from rostral to caudal (C1 to C7); C1 and
vertebrae, on the cephalic bone surface, a ring-like protu- C2, the atlas and axis vertebrae, respectively, form the joint
berance, the uncus, may be present. This ossified structure, complex that permits the spinal column to articulate with the
the uncinate process, serves to limit movement at the inter- head via the occipital condyles. Neither of these vertebrae
vertebral disc and forms the so-called uncovertebral joints have a well-defined body; indeed, the atlas can be viewed as
(joints of von Luschka, see Box 1.2). a ring of dense membrane bone. Bound to the skull by very
a b c
AT AT
AA
AA
SAF
TP
IAF TP
TF TF VF
VF
PA PA
PT
d D PT e f
D
SAS PAS
VF
VF
TP
L
SP
SP
Fig. 1.3 Human cervical vertebrae: atlas and axis. (ac) show the atlas rior arch, PA posterior arch, TF transverse foramen, TP posterior tuber-
(C1) and (df) indicate the axis (C2) vertebrae. (a, d) are superior views; cle, SAF superior articulating facet, SAS superior articulating surface,
(c, f) show inferior views; (b, e) are lateral aspects of the vertebrae. VF PAS posterior articulating surface, L lamina, SP spinous process, D dens
vertebral foramen, AT anterior tubercle, PT posterior tubercle, AA ante- or odontoid process (From Bougery and Jacob (1833). Plate 7)
strong ligaments, these vertebrae allow a range of motion in size than in the cervical spine, but smaller than those of
that permits up and down as well as rotational movements of the lumbar region. Common architectural features of the
the skull. Thus, the joint between the atlas (named after the thoracic vertebrae are that the body (centrum) and the
God who balanced the world on his shoulders) and the spinous processes are large and unlike vertebrae of the
occiput (hole in the head), the atlanto-occipital joint, per- lumbar region, the spinous processes point downwards
mits flexion and extension (basically nodding), while the (see Fig. 1.2f, i). A major function of the thoracic spine is
atlanto-axial joint (C1 and C2) allows nodding, gliding, and stability and through articulations with the ribs provides
rotation. Rotation of the head and with it the atlas is medi- protection for the lungs and the heart. Of the bones that
ated by the odontoid process or dens, a bony peg-like exten- comprise the rib cage, the seven cephalic thoracic verte-
sion of C2 into C1. The actual interaction between C1 and brae are attached to the sternum via 12 pairs of ribs. As
C2 is complex with a number of centers of movement: the such, each sternal rib articulates with two vertebrae: sites
pivoting odontoid process of the axis and the gliding facet of attachment are through joints on the inferior and supe-
joints between the axis and atlas vertebrae. Noteworthy there rior aspects of the centrum and a third facet located at the
is no disc between the occiput and the atlas or between the end of the transverse process (costal facets). The remain-
axis and the atlas; the first intervertebral disc is between the ing thoracic vertebrae are attached to the unanchored ribs
axis C2 and C3. The detailed anatomy of the axis and atlas (also known as floating ribs) by similar types of
are shown in Fig. 1.3; the anatomy of C4C7 is shown in articulations.
Fig. 1.2.
1.3.3 Lumbar Spine (Fig. 1.2)

1.3.2 Thoracic Vertebrae
Like the thoracic spine, the robustness of the lumbar verte-
In general, the twelve thoracic vertebrae have the same brae increases from L1 to L5. When compared with the ver-
functional role as the other axial vertebrae. They are larger tebrae of the other regions of the spine, the individual lumbar
vertebrae are the most massive of all: in most cases being

wider and longer. However, unlike the thoracic spine, the Box 1.3 Denition of Some Commonly Used Terms
lumbar spine curves inwards to form the concavity in Amphiarthroses A joint which allows limited
the lower back. The direction of the curve is probably due to motion
the pull of the viscera of the lower region of the body. Motion Appendicular skeleton A term reserved for the bones
around the lumbar spine is considerably greater than the tho- of the limbs and pectoral and pelvic bones
racic spine, the facet, and disc joints, permitting a significant Axial skeleton Spine
degree of flexion and extension. The lumbar body (centrum) Diarthrodial joints A freely moveable joint also
is wide in all directions and exhibits concavities on both known as a synovial joint
cephalic and caudal surfaces as well as being slightly con- Hypoxia Low oxygen tension
stricted at the sides. Like the thoracic vertebrae, the spinous Motion segment Term used to describe two contigu-
process projects backwards while the large pedicles display ous vertebrae and the intervening intervertebral disc
deep inferior vertebral notches. The L2 segment is the level Notochord A flexible rodlike structure present in
at which the spinal nerves come together to form the cauda chordates that helps to define the longitudinal axis
equina. Sclerotome The portion of the embryonic somite
that gives rise to the axial skeleton
Spondylitis Inflammation of the vertebrae
1.3.4 The Sacrum and Coccygeal Bones (Fig. 1.4) Synchondrosis A joint in which the two bones are
joined by cartilage
The sacrum is a very strong robust multibone triangular com- Vertebral formula The number of cervical, thoracic,
plex (S1S5) which is joined at its upper end to the lumbar lumbar, and coccygeal vertebrae
vertebrae at L5 while its lower portion associates with the Zygapophyseal or facet joints Synovial joints on
coccyx. The five fused bones of the sacrum integrate the two each vertebra that permit movement of the spine
halves of the pelvis. The sacrum is united to the ileum by
fibrocartilage which accommodates and transmits the weight
of the upper body mass. The inferior end of the sacrum artic-
ulates with the five fused bones of the coccyx. Intervertebral 1.4 Vertebrae and Intervertebral Discs
discs are not present in the bones of the sacrum or the coccyx of Animals
(Box 1.3).
1.4.1 Anatomical Considerations
While considerable space is devoted to the sand rat (see

Chap. 20) as well as other quadrupeds (see Chap. 18), it is
worthwhile summarizing some key features of small animals
that are used extensively in studies of the intervertebral disc.
In contrast to the vertically orientated human vertebral col-
umn, the almost horizontal spine of quadrupeds is subjected
to a different series of biomechanical forces. Discussing the
cat spine, Macpherson and Ye note, Not surprisingly, the
force vectors on all of the vertebrae differ substantially from
the human. The axial skeleton may be considered as a seg-
mented beam with the legs as pillar supports and two over-
hanging regions, the head-neck segments and the tail
(Macpherson and Ye 1998). At the rostral end of the spine,
the animals head is supported through the muscles and liga-
Fig. 1.4 Human sacrum and coccygeal bones. The ala (A) of the ments of the cervical spine. The first two vertebrae are ring
sacrum articulates with the ileum (I) of the pelvis at the sacroiliac joints
shaped and are organized to allow for controlled movements
(SIJ). The sacrum consists of five fused vertebrae (S1S5). The superior
portion of the sacrum articulates with L5 (lumbar sacrum articulation, of the head. Like the human, these vertebrae do not have the
LSA) while the inferior aspect fuses with the bones of the coccyx (C1 robust body, but exhibit all of the articulations for spinal
C5). Running through the sacrum is a continuation of the vertebral movement. Macpherson and Ye (1998) propose that the sup-
canal from which the sacral nerves emerge through both anterior (ASF)
port for the head is provided by muscles that join the spine
and posterior foramina. SAP superior articulating process, SP sacral
promontory, AS apex of sacrum (From Lizars (1857). Plate III, Bones of with the scapula. These muscles include the levator scapulae
the pelvis) and serratus ventralis, which are inserted into the transverse
Table 1.1 Vertebral formula for animals and man

Species Cervical Thoracic Lumbar Sacral Coccygeal
Man 7 12 5 5 5
Rat 7 13 6 4 2836
Mouse 7 13 6 4 2428
Fig. 1.5 Axial skeleton of the rat. Micro-CT analysis of the rat. Note
Dog 7 13 7 3 Var
the 12 thoracic vertebrae form an S-shaped curve with facets for articu-
lation with the ribs. The last lumbar vertebra articulates with the sacrum Horse 7 18 6 5 18
which articulates with the pelvis through the ilium, thereby transferring Swan 2225 ? ? 8 ?
the weight of the posterior region of the body to the femurs and the Giraffe 7 12 5 5 4
almost vertical legs. The tail is composed of 2830 vertebrae which, Frog 1 8 1 Urostyle
with increasing distance from the sacrum, exhibits a progressive loss of Snake 350 47 210 1
centrum mass and decrease in the identity of articulating surfaces, pro-
Plesiosaur 40 ? ? ? ?
cesses, and foramina. Eventually, the neural arch becomes fused with
the centrum (Figure provided with kind permission by Dr. Rasesh The number of coccygeal vertebrae in the dog is variable (var), frogs
Kapadia, Scanco Medical, Switzerland) are tailless (anurans) and the final vertebrae form a long bone-like struc-
ture, the urostyle. ? unknown
processes of C3 to T9/10, and the rhomboids which join the

scapula to the spinous process of C4 to T4. Together these up the sacrum, and 4 or 5 coccygeal bones. Details of the
muscles suspend the trunk from the scapulae much like the vertebral formula for a number of common mammals are
wires on a suspension bridge. shown in Box 1.2. In nonmammalian species, considerable
In the rat, the 12 thoracic vertebrae form an S-shaped differences exist in the vertebral formula. Snakes have a large
curve (see Fig. 1.5). These vertebrae display well-developed number of thoracic (between 100 and 200) and caudal
long spinous processes that are intermediate in size between (between 15 and 140) vertebrae; the extinct marine
cervical and lumbar, and they exhibit facets for articulation Plesiosaurus had more than 70 cervical vertebrae (Narita
with the ribs. Like the human spine, the lumbar vertebrae are and Kuratani 2005).
the most massive in the rat with very well-defined interverte- For both humans and many mammals, the number of ver-
bral discs. The last lumbar vertebra articulates with the tebrae in the cervical region of mammals appears to be con-
sacrum. The body of these composite vertebrae forms a slab stant. Galis (1999) analyzed the vertebral formula data for
of bone in which there is loss of intervertebral discs and the mammals from the Descriptive Catalogue of the Osteological
zygapophyses and lateral processes are fused (Fig. 1.5). The Series Contained in the Museum of the Royal College of
sacrum articulates with the pelvis through the ilium, thereby Surgeons of England compiled by Richard Owen in 1853.
transferring the weight of the posterior region of the body to This catalogue contains data of 133 species from 15 orders
the femurs. Thus, forces applied to the animals body are of mammals and showed that a very high percentage of ani-
transmitted across the almost horizontal sacrum (usually at mals, possibly with the exception of carnivores, expressed
S1 and often S2) to the vertical legs. seven cervical vertebrae (Table 1.1).
Composed of a variable number of vertebrae (about Galis (1999) reported that occasionally, there is a loss
2830), the tail represents the final region of the spine. While of a single cervical vertebra (C7) with a concomitant
the first few vertebrae are anatomically complete, with increase in the number of thoracic vertebrae and the
increasing distance from the sacrum, there is a change in ver- appearance of a cervical rib. Associated with this change,
tebra size and complexity. There is a progressive loss of cen- in the space between the clavicle and the rib (the thoracic
trum mass and decrease in the identity of articulating surfaces outlet), there is often nerve and blood vessel compression,
and processes and foramina. Eventually, the neural arch a condition described as thoracic outlet syndrome (TOS)
becomes fused with the centrum, while the diameter of the (Makhoul and Machleder 1992). Correlated with cervical
intervertebral foramen becomes narrowed and indistinct. rib formation, Schumacher et al. (1992) reported that there
Since some studies of the intervertebral disc are performed was an increase in childhood cancer including neuroblas-
in the caudal region of the spine, these anatomical limitations toma, Wilms tumor, Ewing sarcoma, and lymphoblastic
need to be taken into account when devising studies of the and myeloid leukemia. It is likely that this developmental
caudal intervertebral discs. anomaly is a result of aberrant Hox gene expression. Thus,
a higher incidence of cervical rib is seen in the phenotype
of Hoxa-4, Hoxd-4, Hoxa-5, and Hoxa-6 knockouts or
1.4.2 Conservation of Vertebral Number overexpression of Hoxb-7 or Hoxb-8 D (Aubin et al. 1998).
The relationship between Hox expression and develop-
The vertebral formula for humans is surprisingly constant: 7 ment of the axial skeleton in mammals is developed in
cervical, 12 thoracic, 5 lumbar, 5 fused vertebrae that make more detail in Chap. 3.
As an aside, while a vertebrae-dependent increase in rib Occasionally, there is a loss of a single cervical vertebra
number is correlated with disease, loss of a rib has biblical (C7) with a concomitant increase in the number of tho-
implications. racic vertebrae and the appearance of a cervical rib.
But for Adam, no suitable helper was found. So the LORD God
caused the man to fall into a deep sleep; and while he was sleep- Acknowledgments The authors would like to thank Dr. Chris Keppler
ing, he took one of the mans ribs and closed up the place with for the radiographs shown in Fig. 1.2, the Smithsonian Institution for
flesh. Then the LORD God made a woman from the rib he had the permission to reproduce the image of Pikaia (Box 1.1), Scanco
taken out of the man, and brought her to the man. Medical for the use of the microCT image shown in Fig. 1.5, and F.
Michael Angelo, MA, for use of the plates shown in Figs. 1.1, 1.2, 1.3,
Whether Adam had TOS or suffered from headaches due and 1.4. Lastly, the authors wish to thank the NIH and NIAMS for the
to cervical tension or loss of a rib is not known. For a discus- ongoing support through grants AR050087 and AR055655.
sion of this and other biblical possibilities including the
emergence of the baculum (ossified penis bone), please read
Gilbert and Zevit (2001).
References
Agrawal A, Guttapalli A, Narayan S, Albert TJ, Shapiro IM, Risbud
1.5 Summary of Critical Concepts Discussed MV (2007) Normoxic stabilization of HIF-1alpha drives glycolytic
in the Chapter metabolism and regulates aggrecan gene expression in nucleus pul-
posus cells of the rat intervertebral disk. Am J Physiol Cell Physiol
293:C621C631
The intervertebral disc/vertebrae were preceded phyloge-
Archer CW, Dowthwaite GP, Francis-West P (2003) Development of
netically by the notochord which provided rigidity and synovial joints. Birth Defects Res C Embryo Today 69:144155
some resilience to the organism, promoted formation of an Aubin J, Lemieux M, Tremblay M, Behringer RR, Jeannotte L (1998)
extended shape, and protected the overlying spinal cord. Transcriptional interferences at the Hoxa4/Hoxa5 locus: importance
of correct Hoxa5 expression for the proper specification of the axial
The vertebral bodies are formed by fusion of sclerotome
skeleton. Dev Dyn 212:141156
from two adjacent somites: following formation of the Belav DL, Bansmann PM, Bhme G, Frings-Meuthen P, Heer M,
neural arch, remnants of the notochord subsequently form Rittweger J, Zange J, Felsenberg D (2011) Changes in intervertebral
the nucleus pulposus of the intervertebral disc. disc morphology persist 5 mo after 21-day bed rest. J Appl Physiol
111:13041314
Synchronized change in gene expression in the pre-
Benazeraf B, Francois P, Baker RE, Denans N, Little CD, Pourquie O
mesodermal cells activates the segmentation clock. The (2010) A random cell motility gradient downstream of FGF controls
coordinated expression of a limited number of genes pro- elongation of an amniote embryo. Nature 466:248252
vides waves of gene oscillations which control somitogen- Bougery JM, Jacob NH (1833) Atlas of complete treatise on human
anatomy comprising operative medicine. C.A. Delaunay, Paris
esis. Disturbances in this system influence the subsequent
Brunet LJ, McMahon JA, McMahon AP, Harland RM (1998) Noggin,
formation of the vertebrae and the intrinsic curvatures of cartilage morphogenesis, and joint formation in the mammalian
the axial skeleton. skeleton. Science 280:14551457
The discs comprise an outer circumferential ring of Chal J, Pourquie O (2009) Patterning and differentiation of the verte-
brate spine. In: Pourquie O (ed) The skeletal system. Cold Spring
fibrocartilage, the annulus fibrosus. The annulus encloses
Harbor Laboratory Press, Cold Spring Harbor, pp 41116
a central proteoglycan-rich core known as the nucleus Crock HV, Yoshizawa H (1976) The blood supply of the lumbar verte-
pulposus and bounded by the cartilage endplates of the bral column. Clin Orthop Relat Res 115:621
contiguous vertebrae. Dahia CL, Mahoney EJ, Durrani AA, Wylie C (2009) Intercellular sig-
naling pathways active during intervertebral disc growth, differen-
As a joint, the disc is classified as an amphiarthrosis with
tiation, and aging. Spine (Phila Pa 1976) 34:456462
very limited mobility. Biomechanical studies indicate that Dequeant ML, Glynn E, Gaudenz K, Wahl M, Chen J, Mushegian A,
there is wide range of motion between vertebrae that are Pourquie O (2006) A complex oscillating network of signaling genes
more in line with those of a diarthrodial joint rather than underlies the mouse segmentation clock. Science 314:15951598
Frobin W, Brinckmann P, Biggemann M, Tillotson M, Burton K (1997)
an amphiarthrosis.
Precision measurement of disc height, vertebral height and sagittal
The vertebrae protect the spinal cord and serve as sites for plane displacement from lateral radiographic views of the lumbar
connection of the pectoral and pelvic girdles and as bone spine. Clin Biomech (Bristol, Avon) 12(Suppl 1):S1S63
for attachment of muscle and rib for functional changes Galis F (1999) Why do almost all mammals have seven cervical verte-
brae? Developmental constraints, Hox genes, and cancer. J Exp
that enhanced locomotion and respiration. Specific func-
Zool 285:1926
tions of the discs include acting as hydrodynamic shock Gan JC, Ducheyne P, Vresilovic EJ, Swaim W, Shapiro IM (2003)
absorbers as well as providing flexibility to the whole Intervertebral disc tissue engineering I: characterization of the
spine. nucleus pulposus. Clin Orthop Relat Res 411:305314
Garstang W (1928) The morphology of the tunicata, and its bearings on
The vertebral formula for primates is well conserved: in
the phylogeny of the chordata. Q J Microsc Sci 72:51187
humans 7 cervical, 12 thoracic, 5 lumbar, 5 fused verte- Gilbert SF, Zevit Z (2001) Congenital human baculum deficiency: the
brae that make up the sacrum, and 4 or 5 coccygeal bones. generative bone of Genesis 2:2123. Am J Med Genet 101:284285
Iwama H, Kato K, Imachi H, Murao K, Masaki T (2013) Human Nerlich AG, Weiler C, Zipperer J, Narozny M, Boos N (2002)
microRNAs originated from two periods at accelerated rates in Immunolocalization of phagocytic cells in normal and degenerated
mammalian evolution. Mol Biol Evol 30(3):613626 intervertebral discs. Spine 27:24842490
Jiang YJ, Aerne BL, Smithers L, Haddon C, Ish-Horowicz D, Lewis J OConnell GD, Vresilovic EJ, Elliott DM (2007) Comparison of ani-
(2000) Notch signalling and the synchronization of the somite seg- mals used in disc research to human lumbar disc geometry. Spine
mentation clock. Nature 408(6811):4759 (Phila Pa 1976) 32:328333
Jones P, Gardner L, Menage J, Williams GT, Roberts S (2008) Pacifici M, Koyama E, Iwamoto M (2005) Mechanisms of synovial
Intervertebral disc cells as competent phagocytes in vitro: implica- joint and articular cartilage formation: recent advances, but many
tions for cell death in disc degeneration. Arthritis Res Ther 10:R86 lingering mysteries. Birth Defects Res C Embryo Today 75:
Lizars J (1857) A system of anatomical plates of the human body; 237248
accompanied with descriptions, and physiological, pathological, Peacock A (1951) Observations on the pre-natal development of the
and surgical observations. Lizars, Edinburgh intervertebral disc in man. J Anat 85(Pt 3):260274
Macpherson JM, Ye Y (1998) The cat vertebral column: stance Pourquie O (2011) Vertebrate segmentation from cyclic gene networks
configuration and range of motion. Exp Brain Res 119:324332, to scoliosis. Cell 145:651663
RESEARCH ARTICLE Schumacher R, Mai A, Gutjahr P (1992) Association of rib anomalies
Makhoul RG, Machleder HI (1992) Developmental anomalies at the and malignancy in childhood. Eur J Pediatr 151:432434
thoracic outlet: an analysis of 200 consecutive cases. J Vasc Surg Shapiro IM, Vresilovic EJ, Risbud MV (2012) Is the spinal motion seg-
16:534542 ment a diarthrodial polyaxial joint: what a nice nucleus like you
Marchand F, Ahmed AM (1990) Investigation of the laminate structure doing in a joint like this? Bone 50:771776
of lumbar disc anulus fibrosus. Spine (Phila Pa 1976) 15:402410 Shu DG, Morris SC, Han J, Zhang ZF, Yasui K, Janvier P, Chen L,
Moore RJ (2000) The vertebral end-plate: what do we know? Eur Spine Zhang XL, Liu JN, Li Y, Liu HQ (2003) Head and backbone of the
J 9:9296 Early Cambrian vertebrate Haikouichthys. Nature 421(6922):
Mundy C, Yasuda T, Kinumatsu T, Yamaguchi Y, Iwamoto M, Enomoto- 526529
Iwamoto M et al (2011) Synovial joint formation requires local Ext1 Souter WA, Taylor TK (1970) Sulphated acid mucopolysaccharide
expression and heparan sulfate production in developing mouse metabolism in the rabbit intervertebral disc. J Bone Joint Surg Br
embryo limbs and spine. Dev Biol 351:7081 52:371384
Nachemson A, Lewin T, Maroudas A, Freeman MA (1970) In vitro dif- Taylor JR (1975) Growth of human intervertebral discs and vertebral
fusion of dye through the end-plates and the annulus fibrosus of bodies. J Anat 120(Pt 1):4968
human lumbar inter-vertebral discs. Acta Orthop Scand 41: Tubbs RS, Vahedi P, Loukas M, Shoja MM, Cohen-Gadol AA (2011)
589607 Hubert von Luschka (18201875): his life, discoveries, and contri-
Narita Y, Kuratani S (2005) Evolution of the vertebral formulae in butions to our understanding of the nervous system: Historical
mammals: a perspective on developmental constraints. J Exp Zool B vignette. J Neurosurg 114:268272
Mol Dev Evol 304:91106
The Intervertebral Disc: Overview
of Disc Mechanics 2
Daniel H. Cortes and Dawn M. Elliott

2.1 Introduction ..................................................................... 17
The vertebral column is the main structural element of the
2.2 StructureFunction of Intervertebral Disc Tissues...... 19
2.2.1 Osmotic Effects ................................................................. 19 spine and is composed of the vertebrae and the intervertebral
2.2.2 Nucleus Pulposus .............................................................. 19 discs. The function of the vertebral column is to provide
2.2.3 Annulus Fibrosus .............................................................. 21 rigidity to the axial skeleton while allowing limited rotation
2.2.4 Cartilaginous Endplate ...................................................... 23 and bending. The vertebrae are the osseous elements of the
2.3 Intervertebral Disc Mechanics ....................................... 23 vertebral column and the spine. Each vertebra is composed of
2.3.1 Stress and Strain in the Unloaded Disc ............................. 23 a vertebral body and posterior elements. The vertebral bodies
2.3.2 Compression Mechanics ................................................... 24
resemble boxes of cortical bone filled with trabecular bone.
2.3.3 Flexion/Extension and Lateral Bending ............................ 25
2.3.4 Torsion............................................................................... 25 They are separated by the intervertebral discs, which are
attached to the relatively flat surfaces at the top and bottom of
2.4 Effect of Degeneration on Disc Mechanics ................... 26
the vertebral body. On the posterior side of the vertebral bod-
2.5 Mechanically Induced Injury ies, a bony structure composed of pedicles and processes,
of the Intervertebral Disc ............................................... 27
2.5.1 Herniation.......................................................................... 27
known as the posterior elements, serves as anchor points for
2.5.2 Endplate Fracture .............................................................. 28 tendons and ligaments. Anatomical details of each of the ver-
tebrae that comprise the spine are presented in Chap. 1.
in the Chapter .................................................................. 28 Definition of technical terms can be found in Box 2.1.
The zygapophysial joint is an articular joint between the
References ...................................................................................... 28
inferior and posterior articular processes of adjacent vertebrae.
Like most articular joints, the zygapophysial joint comprises a
capsule filled with synovial fluid. Inside this capsule, the bones
are covered by a thin layer of articular cartilage separated by
the fibroadipose meniscoids. The zygapophysial joints play an
important role in the mechanics of the spine. These joints pre-
vent excessive axial rotation between the vertebral bodies,
resist forward sliding of the superior vertebra, limit the amount
of extension by the contact of the inferior articular process and
the lamina of the vertebra below, and contribute to the trans-
mission of a fraction of the load. The ligaments and the joints
connecting the vertebral bodies provide some passive stabil-
ity; the muscles surrounding the vertebral column, through an
active mechanism, provide most of the stability of the spine
during physical activity. A detailed description of spine mus-
cle anatomy, forces, and lines of action is outside of the scope
D.H. Cortes D.M. Elliott, PhD (*) of this chapter, but can be found in Adams et al. (2006).
Department of Biomedical Engineering,
The intervertebral disc is the soft tissue in between the
University of Delaware,
125 E Delaware Avenue, Newark, DE 19716, USA vertebral bodies. It is composed of three distinct tissues:
e-mail: delliott@udel.edu nucleus pulposus, annulus fibrosus, and the cartilaginous

18 D.H. Cortes and D.M. Elliott
endplates. Each of these tissues has a characteristic

Box 2.1 Glossary of Mechanical Terms composition and structure which provide them with special
Stress Stress is the force intensity and is calculated as mechanical properties to perform their function. Their
applied load divided by area over which it is applied. interaction enables the intervertebral disc to transmit loads
Stress is classified into normal (perpendicular to the while allowing a constrained flexibility between vertebral
surface) and shear (along the surface) stresses. bodies. In a healthy disc, the nucleus pulposus is a highly
Strain Strain is a measurement of deformation. hydrated gel-like material which is surrounded by the
Similar to stress, there are normal and shear strains. annulus fibrosus and the cartilaginous endplates (Fig. 2.1).
Normal strain is defined as change in length divided by The main function of the nucleus pulposus is to support
original length. Shear strains are related to the change mechanical loads through hydraulic and osmotic pressure.
in angle. The cartilaginous endplates are thin layers of cartilage that
Poissons ratio When a sample is stretched it contracts cover the central area of the vertebral body (Fig. 2.1a). At
in the lateral direction. The Poissons ratio is the lateral the periphery of the vertebral body, not covered by the car-
contraction divided by the longitudinal stretch. This tilaginous endplate, is the ring apophysis. The cartilagi-
mechanical property is related to the compressibility nous endplates have an important role on the exchange of
of the material. A material with a Poissons ratio of 0.0 nutrients, waste products, and other metabolites between
does not exhibit any lateral contraction, while a mate- the nucleus and the blood vessels in the vertebral bodies.
rial with a Poissons ratio of 0.5 is incompressible. The annulus fibrosus is composed of series of concentric
Stiffness The stiffness is a measure of the resistance layers with collagen fibers in alternating orientations
to deformation. It is defined as force divided elonga- (Fig. 2.1b). The outer lamellae of the collagen fibers attach
tion. The stiffness depends not only on the material directly to the vertebral bodies while the inner lamellae
itself, but also on the size of the object. attach to the cartilaginous endplates. The annulus fibrosus
Modulus The modulus is related to the stiffness, provides a lateral confinement of the nucleus pulposus,
except that the effects of object size are eliminated, supports vertical loads, and limits the amount of motion
therefore, the modulus is a material property. Modulus between the vertebral bodies.
is calculated as stress divided by strain. The intervertebral disc undergoes biochemical and struc-
Anisotropy A material is anisotropic if its mechanical tural changes due to aging and degeneration. Biochemical
properties are different depending on direction of load. changes include a decrease of proteoglycan content, an
For instance, in the annulus fibrosus or other fibrous increase in protein cross-linking, and changes in the collagen
tissues, collagen fibers create direction-dependent type and distribution. The biochemical changes during
anisotropy. degeneration are similar to those of aging; they are charac-
Viscoelasticity A material is viscoelastic when its terized by occurring in a faster rate and accompanied by
mechanical behavior changes over time or as a func- structural changes that impair disc function. Structural
tion of the speed at which the loads are applied. The changes observed during degeneration include a decrease in
viscoelastic properties of a material are usually mea- disc height, inward and outward bulging of the annulus
sured using creep or stress relaxation tests. In a creep fibrosus, and loss of its lamellar organization. The objective
test, a load is applied to the material and the deforma- of this chapter is to describe the mechanical behavior of the
tion increases over time. In a stress relaxation test, a individual tissues of the intervertebral disc and then analyze
deformation is applied and the stress decreases over how they work together in different loading scenarios. In
time. addition, the effects of degeneration on the mechanics at the
tissue and disc levels are described.
a b
Vertebral
Posterior Annulus
Nucleus body
End plates fibrosus
pulposus
Nucleus
pulposus
Fig. 2.1 Schematic Intervertebral
representations of the adult disc
intervertebral disc. (a) Midsagittal
cross section showing anatomical Anterior Annulus Posterior
regions. (b) Three-dimensional fibrosus
view illustrating AF lamellar Vertebral Anterior
structure (Adapted from Smith body
et al. 2011) Spine axis
2 The Intervertebral Disc: Overview of Disc Mechanics 19
2.2 StructureFunction and proteoglycans form a charged, porous, deformable solid

of Intervertebral Disc Tissues material which is embedded in a solution of water and ions
(Urban and Maroudas 1981). The amount of negative charges
Although many studies have shed light on the attached to the solid is quantified by the fixed charge density.
structurefunction relationships for the tissues that form At equilibrium, the balance of chemical potentials results in
the intervertebral disc, this is still an ongoing research an increase of osmotic pressure (p), which is a function of the
topic, the aim of which is to describe the mechanical fixed charge density and the ionic strength of the surrounding
behavior of healthy tissues, the effects of degeneration, fluid (Overbeek 1956). Assuming an ideal solution for the
and the implications of disc mechanics on the cell biology. interstitial fluid and external solution, the osmotic pressure
Recent findings on the structurefunction relationships of can be expressed as
the tissues of the intervertebral disc are presented in the
following sections. p = RT (c 2
fc + 4cb2 - 2cb ) (1)
where R is the universal gas constant, T is the absolute tem-

2.2.1 Osmotic Effects perature, cfc is the fixed charge density, and cb is osmolarity
of the surrounding fluid bath.
The tissues of the intervertebral disc are mainly composed of The osmotic pressure and the external applied forces
water, proteoglycans, and collagen (Eyre 1979). The relative result in deformation of the solid component of the tissue,
content of each of these components differs from tissue to which in turn alter the fixed charge density (cfc). That change
tissue. For instance, the nucleus pulposus has the highest can be quantified by
proteoglycan content, while the annulus fibrosus has the
highest collagen content (see Chaps. 3 and 4) (Eyre and Muir cfc0f0 (2)
1976). The differences in relative content of these individual Cfc =
components and their organization provide these disc tissues (J - 1 + )f
0
with their special mechanical properties. For example, it is

well known that due to its higher collagen content and fiber where cfc0 and f0 are the fixed charge density and the water
organization, the annulus fibrosus has a superior tensile load- content at the reference configuration, respectively, and J
ing capacity. In a similar way, the high proteoglycan content is the ratio between the volume at the deformed and refer-
of the nucleus pulposus provides the tissue with high com- ence configurations. The reference configuration, usually
pressive properties. However, since the tissues in the disc defined as the configuration where stresses are zero, plays
have similar components, they also share some mechanical an important role in the calculation of the osmotic
behaviors, specifically, the osmotic effects which reflect the pressure.
proteoglycan high negative charge density (Urban and
Maroudas 1981). The osmotic effects have important impli-
cations on the mechanics of the disc. For instance, the osmotic 2.2.2 Nucleus Pulposus
pressure causes a deformation of the tissue usually known as
osmotic swelling. This swelling pressure induces tensile The nucleus pulposus is the gelatinous core of the interverte-
stresses and increases the stiffness of the tissue. This osmotic bral disc and it is composed of water (7085 % of total weight),
swelling also draws water into these tissues keeping the disc proteoglycans (3050 % of dry weight), collagen (20 % of dry
hydrated. Since the osmotic effects play an important role in weight), and other minor proteins (Adams and Muir 1976;
the mechanics of all disc tissues, this section provides a brief Eyre 1979). Aggrecan is the most abundant proteoglycan in
description of the relationship between composition and the nucleus pulposus, followed by other proteoglycans such as
osmotic effects. decorin (Melrose et al. 2001). Aggrecan contains keratan and
The osmotic effects are mediated by the proteoglycan chondroitin sulfate chains which interact with hyaluronic acid
content of the tissue (Maroudas and Bannon 1981; Urban filaments forming large molecules that are trapped in the col-
et al. 1979). Proteoglycans are large molecules composed of lagen network (Kiani et al. 2002). These side chains are nega-
many glycosaminoglycan units attached to a long core pro- tively charged; consequently, positively charged Na+ ions bind
tein. Glycosaminoglycans are chains of polysaccharides that to these chains creating an accumulation of cations inside the
at a physiological pH present an excess of negatively charged nucleus pulposus. Since the glycosaminoglycans are not able
ions (Comper and Laurent 1978). The molecular structure of to diffuse out of the nucleus pulposus, there is a permanent
proteoglycans and glycosaminoglycans is discussed in great difference of the concentration of cations compared to the sur-
detail in Chap. 4. Due to their large size, proteoglycans are rounding environment. This unbalance of cations is the cause
trapped in the network of collagen fibers. Therefore, collagen of the osmotic pressure in the disc.
Collagen II is the most abundant type of collagen in the summary of nucleus pulposus values obtained using
nucleus pulposus and other compression-bearing tissues confined compression.
such as articular cartilage (Eyre and Muir 1976). Unlike The elastic behavior of the nucleus pulposus can be appor-
articular cartilage, collagen II forms an unorganized fiber tioned in terms of the contribution of osmotic (ionic) and
network in the nucleus pulposus. A recent study showed that solid tissue (nonionic) effects. The contribution of the
long fibers in the nucleus pulposus continuously connect osmotic effects to the compressive properties has been mea-
both endplates (Wade et al. 2011). In an intact disc, these sured by eliminating the osmotic effects using a surrounding
fibers are much longer than the disc height; they fold in a medium with high osmolarity or by reducing the proteogly-
rather arbitrary configuration and can withstand substantial can content via enzymatic digestion (Heneghan and Riches
tension when unfolded. Experimentally, however, since it 2008a; Perie et al. 2006b). When a high ionic concentration
was necessary to cut the annulus fibrosus to separate the end- medium was used, the compressive properties of the bovine
plates, it is unlikely that the nucleus pulposus fibers experi- nucleus pulposus were reduced to 2030 % of the value mea-
ence high levels of tension under physiological conditions. sured in isotonic (physiological) medium concentrations.
This is different from articular cartilage, where fibers are Therefore, the contribution of the osmotic effects to the stiff-
highly organized and experience substantial tension due to ness and load support of the nucleus pulposus is approxi-
the osmotic swelling (Ateshian et al. 2009; Cavalcante et al. mately 7080 %. The contribution of the osmotic effects was
2005). In the case of the nucleus pulposus, the osmotic and almost constant through a wide range of applied deforma-
hydrostatic pressure is supported axially by the endplates tions (070 % compressive strains). If the proteoglycans are
and radially by tensile (hoop) stresses in the annulus fibrosus. removed by enzymatic digestion, a reduction of 20- to
Consequently, fibers are not required to hold the nucleus pul- 30-fold was observed in the compressive properties of
posus in place as is the case in articular cartilage. the nucleus pulposus (Perie et al. 2006b). This suggests that
Due to its high levels of hydration and gelatinous consis- the proteoglycans also have a nonionic contribution to the
tency, the mechanical behavior of the nucleus pulposus has mechanics of the nucleus pulposus. Evidence of the nonionic
characteristics of both a fluid and a solid (Iatridis et al. contribution of the proteoglycans has been reported for other
1996). Consequently, the nucleus pulposus is usually treated tissues such as articular cartilage (Canal Guterl et al. 2010).
as a viscoelastic material. The mechanical properties of the Viscoelastic or frequency-dependent properties of the
nucleus pulposus have been investigated mainly through tor- nucleus pulposus have been analyzed using torsion tests
sion and compression tests (Heneghan and Riches 2008a, b; (Iatridis et al. 1997a, b). Stress relaxation tests measured an
Iatridis et al. 1997a, b; Johannessen and Elliott 2005; Perie instantaneous shear modulus around 11 kPa. However, the
et al. 2005). Confined compression has been typically used shear stress rapidly relaxed to near-zero values suggesting a
to measure several mechanical properties of the nucleus fluid-like behavior. In dynamic torsion tests, a shear modulus
pulposus such as aggregate modulus and permeability of ~20 kPa and a phase shift (the delay between strain and
coefficients (Johannessen and Elliott 2005). It is measured stress measured in terms of degrees) of ~30 were measured.
by axially compressing a cylindrical sample in a chamber For comparison, the dynamic modulus of articular cartilage
that prevents lateral expansion. Although physiologically is 6001,000 kPa, the modulus for the meniscus is 540 kPa,
the nucleus pulposus is not fully confined or fully while proteoglycan solutions are 0.01 kPa (Hardingham
unconfined, confined compression tests have been gener- et al. 1987; Zhu et al. 1993, 1994). Values of phase shift of
ally accepted to characterize its compressive behavior. For 13 for cartilage, 22 for meniscus, and 65 for proteoglycan
small deformations (around 5 %), the nucleus pulposus can solutions have been reported. Since the phase shift for the
be considered to have a constant permeability and exhibits nucleus is lower than 45, it suggests a more solid-like
a linear relationship between stresses and strains dynamic behavior.
(Johannessen and Elliott 2005). However, the properties are The studies discussed above illustrate the complexity of
strain dependent (i.e., nonlinear) for moderate and large the mechanical behavior and structurefunction relationships
strains (Heneghan and Riches 2008a). Table 2.1 presents a of the nucleus pulposus. The contribution of osmotic pressure
Table 2.1 Aggregate modulus (HA) Study HA (kPa) l NP tissue s-GAG (% dry wt.)
as a function of the stretch ratio (l)
Heneghan and 691,650 1.00.3 Bovine tail 24
and glycosaminoglycan content for
Riches (2008a)
the nucleus pulposus
Perie et al. (2005) 350520 1.00.8 Bovine tail
Perie et al. (2006a) ~600 1.00.6 Bovine tail ~35
Perie et al. (2006b) ~400510 1.00.8 Bovine tail ~42
Johannessen and 1,010 0.95 Human 44
Elliott (2005)
and the blend between exhibiting characteristics of fluid and fibrils to large collagen bundles or fascicles (Kastelic et al.
solid mechanics pose a difficult challenge to model and also 1978). Collagen fibers in these tissues have a wavy or zig-
complicate the prediction of deformations during physiologi- zag shape commonly known as crimp (Diamant et al. 1972;
cal loading. Nonetheless, it is important to understand and Kastelic and Baer 1980). When the fibers are stretched, the
characterize the mechanics of the nucleus pulposus as it fibers progressively straighten with minimal resistance; i.e.,
influences cell function and impacts predictions related to a negligible force is required to uncrimp the fibers. The
mechanically induced injuries and regeneration. amount of stretch required to straighten the fiber is known
as uncrimping stretch. Once a fiber is straight, it starts taking
load (Fig. 2.2). The uncrimping stretch is not same for all
2.2.3 Annulus Fibrosus fibers in a tissue. For small deformations, few fibers with a
small uncrimping stretch are straightened as they accommo-
Similar to the nucleus pulposus, the annulus fibrosus is com- date the load. Consequently, the stiffness of the tissue is low.
posed mainly of proteoglycans and collagen, although the Progressively, all the fibers stretch and contribute to load sup-
relative content and organization of its components are sub- port resulting in high tissue stiffness (Fig. 2.2). If the tissue
stantially different. In the healthy human annulus fibrosus, is further stretched, fibers will fail by several possible mecha-
the water content is 50 %, collagen is approximately 70 % of nisms including breakage and fiber pullout.
the dry weight, and proteoglycans make up to 10 % of the All the other components of the annulus fibrosus (except
dry weight (Eyre and Muir 1976; Eyre 1979). The annulus the fibers) are usually treated as a single material known
fibrosus is subjected to both tensile and compressive stresses as extrafibrillar matrix. Since the fibers contribute to the
during physiological loading. Consequently, it has high col- mechanics of the annulus fibrosus when they are in tension,
lagen content similar to other tension-bearing tissues such as the matrix characterizes the compressive properties of the
tendon and ligaments. Proceeding from the outer to the inner annulus fibrosus. Other properties, including permeabil-
annulus, there is an decrease in the ratio of collagens I to II, ity and diffusivity of solutes, are also attributed to matrix.
and the amount of proteoglycan rises. This profile reflects a For simplicity, the elasticity of the matrix has been consid-
change in the loading environment from more tension in the ered isotropic, which means that the elastic properties (i.e.,
outer annulus fibrosus to more compression towards the Young modulus and Poissons ratio) are same in all direc-
nucleus pulposus (Eyre and Muir 1976). In a similar way, in tions. Although the matrix includes some fibrillar compo-
the outer annulus fibrosus, collagen fibers insert directly to nents such as elastin and protein cross bridges, their content
the cortical bone of the vertebrae and not to the endplate as
in the case of inner annulus fibrosus, again probably reflecting
the higher tensile loads present in the outer annulus fibrosus
(Nachemson 1963; Wu and Yao 1976). Linear region
Noteworthy, in the annulus, collagen fibers are arranged in
concentric lamellae with alternating orientations (Fig. 2.1b).
The angle between fiber directions of adjacent lamellae changes
from ~60 to the spinal axis in the outer annulus fibrosus to
~90 in the inner annulus fibrosus (Cassidy et al. 1989; Guerin
and Elliott 2006a; Hickey and Hukins 1980). This arrangement
Stress
provides the annulus with a series of important mechanical

properties, including anisotropy (direction dependence). Since
the fibers play such an important role in the mechanics of the
Heel region
annulus fibrosus, this tissue can be analyzed as a combination
of fibers and an isotropic material known as extrafibrillar
Toe region
matrix (Spencer 1984). As its name indicates, the extrafibrillar
matrix represents all the solid components of the annulus
fibrosus, except fibers.
One of the more important characteristics of the mechan- Strain
ics of collagenous tissues is nonlinearity. The nonlinearity of
the fibers is characterized by a low stiffness region for small Fig. 2.2 The tensile stressstrain response of collagenous tissues, such
deformations, known as toe region, followed by a transition as annulus fibrosus, can be divided into several regions, corresponding
to different mechanisms. In the toe region, the contribution of fibers is
(heel) region and a much stiffer linear region (Guerin and
small due to fiber crimping. In the heel region, the increase in stiffness
Elliott 2007; Wu and Yao 1976). Collagen fibers, in most is due to fiber straightening. In the linear region, most of the fibers are
tension-bearing tissues, have a hierarchical organization from straight and contributing to the high tensile stiffness
is small and unlikely to significantly alter the assumption of ing of the annulus fibrosus (Bass et al. 2004; Gregory and
isotropy. However, transport properties such as permeability Callaghan 2011; Huyghe 2010; Jacobs et al. 2013;
and diffusivities have been shown to be anisotropic (Gu et al. OConnell et al. 2012). For this test, a rectangular thin sam-
1999; Travascio and Gu 2011), which means that there are ple is gripped on all four sides and loads are applied in two
directions where the fluid and solutes can flow or move with directions (Fig. 2.3). Two-dimensional deformations are
less resistance. optically recorded during the test. Unlike a uniaxial test,
The tensile properties of the annulus fibrosus have there is not a direct relationship between the slope of these
been characterized using uniaxial and biaxial tension curves and the elastic properties of the annulus fibrosus; the
tests (Jacobs et al. 2013; Nerurkar et al. 2010). In uniaxial forces (or stress) in one direction are affected by the defor-
tests, a strip of tissue is cut from the annulus fibrosus in a mation applied to the other direction (OConnell et al.
given orientation (circumferential, axial, radial, or along 2012). Consequently, the data from biaxial tests are ana-
the fibers), and the force required to stretch the sample is lyzed through the use of a model. The advantage of using
recorded as a function of the applied strain. The Poissons biaxial experiments to characterize the mechanics of the
ratio can be measured by recording the lateral contrac- fibers is that the values obtained through these types of tests
tion of the sample during the test. The Young modulus can be used to predict the response of the annulus fibrosus
is calculated from the slope of the stressstrain response. in uniaxial tests and with other biaxial strain ratios
A summary of these properties is presented in Table 2.2. (OConnell et al. 2012).
The Young modulus is higher in the discs circumferential Since the collagen fibers only contribute to the mechanics
direction than the axial. This is expected since the fibers of the annulus fibrosus in tension, the elastic properties of the
are oriented closer in the circumferential direction; there- extrafibrillar matrix can be measured through confined com-
fore, fibers are not stretched during axial loading so that pression tests (Cortes and Elliott 2012; Drost et al. 1995;
modulus is primarily due to the matrix. Klisch and Lotz 2000; Perie et al. 2005). This test provides
Biaxial loading is another tensile test used to quantify the aggregate modulus, measured as a function of strain.
annulus fibrosus mechanics. It is thought that biaxial load- Similar to the nucleus pulposus, the mechanical behavior of
ing more closely resembles multiaxial physiological load- the matrix depends on contributions from the osmotic pres-
sure and the nonionic extrafibrillar matrix (Cortes and Elliott
Table 2.2 Linear region moduli of nongenerated (ND) and degenerated 2012). In this manner, the mechanical properties of this
(D) annulus fibrosus tissue (Elliott and Setton 2001; Guerin and extrafibrillar matrix can be measured in tension and com-
Elliott 2006b) pression by applying osmotic swelling and confined com-
Circumferential Axial Radial pression simultaneously. The nonionic extrafibrillar matrix is
ND D ND D ND D nonlinear with a higher stiffness in compression (~50 kPa)
Inner anterior 5.610.0 5.0 1.0 than in tension (~10 kPa), and the contribution of the osmotic
Outer anterior 17.029.0 22.029.0 0.8 0.40.5 0.4 pressure in the support of the applied loads is high (~70 % of
Inner posterior 2.06.0 4.0 0.5 total) when the EFM is in compression and low (~25 %)
Outer posterior 13.019.0 8.0 when in tension.
Fig. 2.3 In a biaxial test of annulus fibrosus, a sample is loaded simultaneously in the axial and circumferential direction
Table 2.3 Shear modulus (kPa) Orientation

of human annulus fibrosus: effects
Type of test Location Circ-radial Circ-axial Radial-axial
of sample orientations and location
and type of test Simple shear compressive preload Anterior 28.92 58.56 40.16
Simple shear compressive preload Posterolateral 22.2 53.6 25.1
Simple shear tensile preload Anterior 193.6
Torsion shear equilibrium Anterior 20100
Torsion shear dynamic Anterior 100280
Shear tests have been used to determine elastic and vis- with the vertebral bodies (Fig. 2.1a). It covers most of the
coelastic properties of the annulus fibrosus. Elastic shear vertebral endplate except for a small ring in the periphery
properties have been measured by applying simple shear called the ring apophysis. The thickness of the cartilaginous
tests (Fujita et al. 2000; Hollingsworth and Wagner 2011; endplate varies: it is thinnest in the center (~0.2 mm) and
Iatridis et al. 1999; Jacobs et al. 2011). Since the fibers have thickest in the periphery (~0.9 mm) (Moon et al. 2013). The
a contribution during this test, the shear modulus is anisotro- composition of the cartilaginous endplate is similar to that of
pic with a higher modulus in the circumferentialaxial plane hyaline cartilage, which is characterized by a high proteogly-
where the fibers experience stretches (Table 2.3). On the can and collagen II content. The water content of human
other hand, torsion tests have been used to measure vis- endplate is 58 % of the wet weight, the s-GAG content is
coelastic properties of the annulus fibrosus (Iatridis et al. 17 % of the dry weight, and the total collagen content is
1999). The dynamic modulus increases with frequency. Both 6080 % of dry weight (Setton et al. 1993). The cartilagi-
equilibrium and dynamic modulus decrease with shear strain nous endplate plays an important role in the transport of
amplitude. The highly viscoelastic nature of the annulus nutrients and other metabolites into the nucleus pulposus and
fibrosus is evidenced by the threefold increase of the dynamic the inner portion of the annulus fibrosus.
modulus over the equilibrium modulus. The mechanics of the cartilaginous endplate has been
Although separating the mechanics of the annulus fibrosus measured using confined compression tests (Setton et al.
into fibers and matrix is very convenient and describes much 1993). The aggregate modulus of the baboon endplate is
of the mechanical behavior of annulus fibrosus, there are inter- 0.44 MPa. The hydraulic permeability (14.3 1014 m4/Ns) is
actions between these components. Specifically, the stiffness considerably higher than values of 0.09 1014 m4/Ns and
of the matrix increases with the stretch of the fibers (Guo et al. 0.153 1014 m4/Ns for the human annulus fibrosus and
2012). To account for these effects, several fibermatrix and nucleus pulposus, respectively. The high permeability value
fiberfiber interactions have been formulated in terms of the suggests that its main function is to allow the transport of
strain perpendicular and along the fibers (Guerin and Elliott fluids, nutrients, and waste products to the cells in the nucleus
2007; OConnell et al. 2009, 2012; Wagner and Lotz 2004). pulposus and part of the annulus fibrosus.
These interactions have more accurately described the mechan-
ical behavior of the annulus fibrosus. It has also been suggested
that shear interactions are essential to obtain a good simultane- 2.3 Intervertebral Disc Mechanics
ous prediction of uniaxial, biaxial, and shear experimental data
(Hollingsworth and Wagner 2011; OConnell et al. 2012). In the previous sections, the mechanics of individual disc tis-
While many aspects of the mechanical behavior of the sues were described separately. However, these tissues interact
annulus fibrosus have been well described, this is still an with each other providing the disc with a special mechanical
active area of research. Special attention needs be given to behavior. In a similar way, a disruption or a change in mechan-
relations between interactions and composition of the annu- ical properties of one of these tissues causes an impairment of
lus fibrosus and the contribution of these interactions to the mechanical function of the overall disc. In this section, disc
mechanics of the disc. Additionally, these interactions mechanics are presented as the contribution of individual tis-
between components should be replicated in engineered tissues during a given loading scenario. First, the residual stresses
sues that are currently being investigated as therapeutic alter- in the unloaded disc are briefly described. Then, the mechanics
natives (Mauck et al. 2009; Nerurkar et al. 2010). of the intervertebral disc are analyzed for three of the most
important loads: axial compression, bending, and torsion.
2.2.4 Cartilaginous Endplate

2.3.1 Stress and Strain in the Unloaded Disc
The biomechanics of the cartilaginous endplate has been far
less studied than other disc tissues. The endplate is the inter- Before analyzing the mechanics of the disc under different
face between the nucleus pulposus and inner annulus fibrosus types of loading scenarios, it is important to understand the
impact of internal stresses and strains on the unloaded disc. changes with posture and activity, the mechanism by which
As described above, at the tissue level, the osmotic pressure the different tissues of the intervertebral disc interact to sup-
is balanced by tensile or residual stresses. In a similar way, port this load is the same. In this section, the interaction
at the disc level, there are residual stresses and strains due to between disc tissues is described for compressive loads over
osmotic effects caused by tissue proteoglycans, are present short and long periods of time.
even in the absence of applied loads. When external loads are After a compression load is applied to the disc, the imme-
applied to the disc, additional stress builds up above that of diate mechanics are different from that measured at longer
the residual stress. There are several mechanisms, at differ- time intervals. Immediately after the load has been applied,
ent scales, contributing to residual stress (Lanir 2009). At the the tissues in the disc can be considered to be incompressible
micro-level, the interaction between proteoglycans, ions, materials; due to the low permeability of the disc tissues,
water, and the collagen network produces an osmotic pres- there is insufficient time for interstitial fluid flow (Ateshian
sure that contributes to the total stress of the tissue. At the et al. 2007). In this loading state, the interstitial fluid in the
meso-level, residual stress arises from inhomogeneities nucleus pulposus pressurizes, supporting a fraction of the
within the tissues, e.g., the gradient of proteoglycan and col- load. Since the nucleus pulposus behaves as an incompress-
lagen content from inner and outer annulus. Residual stress ible material, it tends to expand radially. However, since it is
at meso-level has been recently measured in terms of the contained by the annulus fibrosus, there is a large tensile
opening angle after a radial cut in bovine annulus fibrosus strain in the circumferential direction and outward bulging of
rings (Michalek et al. 2012). This effect is similar to that the annulus (Tsantrizos et al. 2005). The applied load is sup-
observed in aortic arteries, where differences in proteogly- ported by the lamellae through compressive stress in the axial
can content between the media and the adventitia contribute direction. As a result, the compressive load causes the lamel-
to this component of the residual stress (Azeloglu et al. 2008; lae in the inner annulus fibrosus to buckle towards the nucleus
Chuong and Fung 1986). At the disc level, residual stress is pulposus; of course, this is opposed by the outward pressure
also generated by the interaction between different tissues exerted by the nucleus pulposus. Inward buckling of the
(nucleus pulposus, annulus fibrosus, endplates, and vertebral inner annulus fibrosus is evident in the degenerate disc due to
bodies). The high proteoglycan content of the nucleus pulpo- a decrease in the internal pressure of the nucleus pulposus
sus results in a significant osmotic pressure. This pressure associated with altered osmotic pressure and permeability
has been measured in vitro and in vivo using a needle pres- changes (Sasaki et al. 2001; Sato et al. 1999; Wang et al.
sure gauge (Nachemson 1981; Panjabi et al. 1988; Wilke 2010). From this perspective, pressurization of the nucleus
et al. 1996, 1999). The radial expansion of the nucleus is pulposus is of critical importance not only to carry part of the
constrained by the annulus fibrosus through tensile stresses compressive load but also to provide stability to the lamellae
in the circumferential direction (hoop stress) and compres- in the radial direction.
sion stress in the radial direction. Similarly, the osmotic pres- During the diurnal loading cycle, the disc is subjected to a
sure in the nucleus tends to vertically separate the vertebral prolonged period of compression followed by a period of
bodies, which are held in place by tensile stresses in the low-load recovery. If the load on the disc is maintained for
annulus fibrosus in the axial direction. All these contributions some hours, the pressurized interstitial fluid will flow to
to the residual stress of the disc create a multidirectional and regions of lower pressure through the annulus fibrosus and
inhomogeneous initial state of stresses and strains that must the endplate (van der Veen et al. 2007). During this process,
be considered for the analysis of disc mechanics. the disc height decreases while the outward bulging of the
annulus fibrosus increases (OConnell et al. 2007). In addi-
tion, the nucleus pulposus depressurizes, reducing its
2.3.2 Compression Mechanics contribution to load and increasing the axial compression of
the annulus fibrosus (OConnell et al. 2007). In this relaxed
Axial compression loading of the spine is of major state, the tissues in the disc interact, as described above for
physiological importance and arises from the weight of the instantaneous loading; however, the relative contribution of
upper body and by forces exerted by the muscles in the trunk each of the tissues changes. After relaxation, the osmotic
during common daily activities. Compression loads are pressure in the healthy nucleus pulposus does not vanish
transmitted from vertebra to vertebra through the interverte- completely. In fact, due to osmotic effects the remaining
bral disc and the zygapophysial joints in proportion to body intradiscal pressure is largely responsible for hydration
posture. For instance, 84 % of the compressive load is trans- recovery and mechanics during the resting period of the diur-
mitted through the intervertebral disc in the erect standing nal cycle (OConnell et al. 2011; van der Veen et al. 2007).
posture, whereas 100 % of the load is transmitted through the Nutrient and metabolite exchange during loading and
disc in the erect sitting posture (Adams and Hutton 1980). unloading is essential for disc cell viability. In this process,
Although the compressive load to the intervertebral disc nutrients and metabolites are brought to, and waste byproducts
expelled from, the disc by diffusion and convection (Das et al. a similar pattern of strains in the disc; however, the compres-
2009; Ferguson et al. 2004; Holm et al. 1981; Shirazi-Adl et al. sion and tension regions are located in the lateral annulus
2010; Soukane et al. 2007; Urban et al. 1978, 1982, 2004). The fibrosus (Costi et al. 2007; Tsantrizos et al. 2005).
rate at which fluid leaves the disc depends on the hydraulic The stiffness and range of motion of disc segments can be
permeability and diffusivity of its component tissues. Since the obtained by applying a known force and/or bending moment.
hydraulic permeability of the endplate is higher than the annu- The range of motion is defined as the relative rotation of the
lus fibrosus, it should enhance aqueous flow through the end- vertebral bodies when a pure moment is applied in the sagit-
plate (Setton et al. 1993). Moreover, from the periphery tal or coronal plane. On the other hand, the stiffness can be
(endplates and outer annulus) to the center of the disc, there is calculated as the slope of the momentrotation curve at the
a change in metabolite concentration. Numerical simulations end of the range of motion. Due to the asymmetric shape of
have also shown that the concentrations of glucose and oxygen the disc and the effect of the posterior elements, measure-
are low close to the center of the disc, whereas lactic acid, ments of the range of motion and stiffness are different in
which is the major metabolite, has a reverse distribution flexion and extension. For instance, an increase of the com-
(Jackson et al. 2011; Soukane et al. 2007). Recent studies are pression strain from 2.7 to 6.7 % for an applied force of
aimed at improving the accuracy of the numerical models by 500 N was reported for human L2/L3 spine segments when
considering anisotropy and nonlinearity of elastic, flow, and the posterior elements were removed (Heuer et al. 2008). In
diffusion properties (Chuang et al. 2010; Jackson et al. 2008). a similar way, the range of motion increased from 5.2 to
6.9 in flexion and from 3.4 to 8.2 in extension for a pure
applied moment of 7.5 Nm (Heuer et al. 2008). The stiffness
2.3.3 Flexion/Extension and Lateral Bending of disc segments has been measured on the principal axes of
the disc and on multidirectional axes (Spenciner et al. 2006).
Flexion/extension and lateral bending are spine movements It was concluded that the experimentally measured stiffness
required for many daily activities. Flexion and extension are along the multidirectional axes do not match with the ana-
terms used when the trunk bends forward and backward, lytical predictions from the stiffness along principal axes
respectively. In a well-aligned spine, a neutral position is (Spenciner et al. 2006).
observed when standing upright. However, the natural curva-
ture of the spine changes during daily activities such as sit-
ting or lifting a weight. This curvature change is the sum of 2.3.4 Torsion
the relative rotations between each of the vertebral bodies,
each of which produce internal strains and stresses in the During a twisting motion of the trunk, torsion becomes
disc. To quantify the mechanics of the disc under this type of another important component of the loading of intervertebral
motion, the forces and moments can be estimated by moni- discs. Similar to flexion/extension, torsion is defined as the
toring muscle activity. However, this approach has been relative rotation of consecutive vertebral bodies; however,
shown to be inconsistent (Potvin et al. 1991). A better the axis of rotation is parallel to the axis of the spine.
approach to estimating forces and bending moments consists Consequently, the strains in the disc are substantially differ-
of determining the correlation between in vivo and in vitro ent. During physical activity, the rotation between vertebral
measurements (Adams and Dolan 1991). bodies is about 13 which is very small compared to the
A common kinematic characteristic of flexion/extension rotations observed during flexion or extension (Pearcy et al.
and lateral bending is that the axis of rotation is perpendicular 1984). The torsion range of motion is constrained by contact
to the axis of spine. Therefore, flexion, extension, and lateral of the zygapophysial joints, which also increases the stiffness
bending produce a similar pattern of internal deformations to of intact spine segments (Adams and Hutton 1981). In vitro
the disc. During flexion, the axial compression in the anterior and numerical studies have reported a range of motion of
portion of the annulus fibrosus is increased. Consequently, 48 for intact spinal segments. The removal of the posterior
there is an increase in the bulging of the outer region of the elements of the spine increases the range of motion twofold
annulus and buckling of the lamellae in the inner portion of (Shirazi-Adl et al. 1986).
the anterior annulus fibrosus. On the other hand, the posterior Shear strains are the main component of deformation dur-
region experiences tension in the axial direction. Additionally, ing torsion. Deformation results in tensile stretch of one of the
the nucleus pulposus is shifted to the opposite side of bend- fiber populations, and while not contributing to torsion support
ing, and there is an increase in intradiscal pressure (Nachemson due to fiber buckling, the other experiences compression. The
1981; Wilke et al. 1999). When the spine is in extension, the tensile stretch on the fibers increases radially; consequently,
reverse effects are observed: tension in the anterior annulus the maximum fiber stretch is found in the outermost lamella.
fibrosus, compression of the posterior annulus, and shifting of Removing the posterior elements of the spine resulted in an
the nucleus in the anterior direction. Lateral bending produces increase in the maximum fiber stretch from 3.1 to 11.4 % for
an applied torque of 7.5 Nm (Heuer et al. 2008). From in vitro

a Healthy
experiments, there is a reduction of the outward bulging of the
annulus fibrosus and an increase of disc height and intradiscal
pressure (Heuer et al. 2008; van Deursen et al. 2001a, b). The
decrease of outward bulging can be directly linked to the high
tensile fiber stresses in the outer lamellae. The decrease of lat-
eral bulging also explains the increase of disc height and intra-
discal pressure. Although the strains observed in torsion may
be too small to cause significant damage to the disc, a decrease
in the failure loads have been observed when torsion is com-
bined with compression and flexion/extension. b Early degeneration
2.4 Effect of Degeneration

on Disc Mechanics
*
Intervertebral disc degeneration can be defined as a post-trau-
matic cell-mediated cascade of biochemical, mechanical, and
structural changes that affect the function of the disc (Adams
and Roughley 2006) (Fig. 2.4). Compositional changes dur-
ing disc degeneration are mainly loss of proteoglycans, c Advanced degeneration
increased cross-linking, and an increase in the amount of col-
lagen I over collagen II. These changes are first noticeable in
the nucleus pulposus and later spread outwards to the annulus
fibrosus. Although the causes remain largely unclear, factors
that include structural injury, genetic heritance, age, inade-
quate metabolite transport, and loading history have been
associated with the onset and progression of disc degenera-
tion (Adams and Roughley 2006; Batti et al. 2008; Buckwalter
1995; Hsu et al. 1990; Pye et al. 2007; Rannou et al. 2004). In
this section we limit our discussion to the effect of degenera- Fig. 2.4 Magnetic resonance images illustrating different stages of
human lumbar disc degeneration. (a) Healthy disc exhibiting distinct
tion on the mechanics at the tissue and disc levels. A brief AF lamellae (AF) and central NP region (NP). (b) Disc exhibiting early
review of total disc replacements, which is of the treatments stages of degeneration, including moderate height reduction, decreased
for disc degeneration, is presented in Box 2.2. NP signal intensity, and inward bulging of AF lamellae (*). (c) Disc
Numerous studies have measured the changes in the exhibiting advanced stages of degeneration, including severely reduced
height, large fissure (*), and generalized structural deterioration.
mechanical behavior of the disc tissues at several stages of Images obtained using 7T Siemens scanner and a turbo spin echo
degeneration. The compositional changes in the nucleus sequence at 200 mm isotropic voxel resolution (Adapted from Smith
pulposus include a decrease in proteoglycan content and an et al. 2011)
increase in collagen I cross-links. These changes exert con-
tradictory effects on the mechanics of the nucleus pulposus.
On one hand, the loss of proteoglycans causes a decrease in is an increase in the hydraulic permeability (Johannessen and
the osmotic pressure and consequently a reduction in tissue Elliott 2005).
stiffness. On the other hand, the increase in cross-linking and The annulus fibrosus undergoes mechanical changes
collagen content causes tissue stiffening. Experimentally, it with degeneration. The modulus at the toe region increases
has been noted that there is an overall decrease in the com- with degeneration, probably due to changes in the water
pression properties of nucleus pulposus (Johannessen and content and the increase in collagen I levels (Guerin and
Elliott 2005). This observation is in accord with other stud- Elliott 2006a).The Poissons ratio decreases about 50 % with
ies that show that proteoglycans contribute approximately degeneration (Acaroglu et al. 1995; Elliott and Setton 2001;
80 % to the compressive properties of the nucleus pulposus Guerin and Elliott 2006a), as does the shear modulus (Iatridis
with degeneration (Heneghan and Riches 2008a; Perie et al. et al. 1999). Additionally, fiber reorientation decreases, while
2006b). However, there is also a significant increase in the interaction between fibers and extrafibrillar matrix increases
shear modulus (Iatridis et al. 1997b). Transport properties of with degeneration (Guerin and Elliott 2006a; OConnell
the nucleus pulposus are also affected by degeneration as there et al. 2009). That fibermatrix interactions increase with
to a reduction of the disc height and an increase in insta-

Box 2.2 Total Disc Replacements bility of the disc measured by an increase in the range of
Total disc replacement (TDR) is a relatively new treat- motion and neutral zone (Mimura et al. 1994; OConnell
ment for discogenic pain that may be used instead of et al. 2007). The decrease of disc height causes an increase
fusion in some patients. In a TDR, a mechanical device in the compression load in the axial direction which in turn
replaces the disc and aims to permit motion between results in buckling of the lamellae, increase of outward and
the vertebrae, as opposed to fusion where there is no inner bulging of the annulus fibrosus, and loss of organiza-
motion. Rational for TDR are related to perceived tion of the lamellae structure (OConnell et al. 2007, 2010).
advantages over fusion for reduced surgical time, The decrease in osmotic pressure also causes a reduction in
improved patient recovery, and long-term improved fluid exchange during the diurnal cycle (Massey et al. 2011).
mechanics to prevent degeneration at other sites in The fluid exchange reduction affects the transport of metabo-
the disc. An ideal artificial disc should have the same lites such as glucose and lactic acid, thereby influencing cel-
range of motion and mechanical properties as a healthy lular function.
disc. However, this is generally not the case. Metallic
spheres were the first intervertebral disc replacements
and were implanted in patients in the early 1960s. The 2.5 Mechanically Induced Injury
most common complication was subsidence, or pene- of the Intervertebral Disc
tration of the ball into the endplate and vertebral body,
and this device was abandoned. The Acroflex, from In a healthy person, the loads applied to the intervertebral
the 1970s, consists of two porous plates separated by disc are not likely to exceed its strength limits. However, in
a hyperelastic polymer. The design was revolutionary some cases such as trauma, a single high-magnitude load
in the sense that the porous plate allowed integration causes a mechanical disruption of the structure of the spine.
with the bone and the polymer provided controlled Usually, in such events, the posterior elements of the spine
flexibility. Unfortunately short-term failure of the poly- such as the zygapophysial joints are damaged before the disc
meric material stopped wide clinical use. The Charite, is affected. However, under certain conditions, damage to the
introduced in the 1980s, has two metallic endplates intervertebral disc in the form of disc prolapse, fracture of
articulating against a polymeric core similar to a ball- the vertebral endplate, or tears in the annulus fibrosus or
and-socket articulation. Although this device has been nucleus pulposus can be observed. Such catastrophic changes
very popular in Europe, long-term studies have reported can cause a permanent change on the internal distribution of
several complications including migration, subsidence, stresses and strains, thereby affecting the normal functioning
ejection, and wear of the core. Recent designs simi- of disc. In addition, such changes in the mechanical environ-
lar to the Charite include the ProDisc and FlexiCore. ment trigger a cell mediate cascade of biochemical, struc-
These devices all permit motion, but generally do not tural, and morphological changes known as degenerative
mimic normal disc mechanics in terms of have axial disc disease that further impairs disc function. Another case
compression motion, energy absorption, and resistance of abnormal loading occurs when a low-magnitude load is
to torsion. While the FDA has approved some models applied a great number of times. This repeating loading
for use, they are not widely used in patients. Total disc event, known as fatigue, is believed to be linked to the onset
replacement in the cervical spine, which requires more and propagation of tears in the disc and is a cause of hernia-
motion and less axial load, appears to be more widely tion. In this section, recent studies analyzing the relationship
applied and successful. A more detailed review of total between abnormal loading and injury of the intervertebral
disc replacements and other devices can be found in disc are discussed.
Intervertebral disc properties: challenges for biode-
vices by Costi et al. (2011).
2.5.1 Herniation
Herniation is characterized by the prolapse of the nucleus

degeneration is evident from biaxial tests (OConnell et al. pulposus through the annulus fibrosus. In vitro, herniation
2012). can be produced by a single high-intensity load or the repet-
All the degenerative changes observed at the tissue level itive application of forces with lower intensity (Callaghan
have an effect on the mechanics at the disc level. Of all of and McGill 2001; Iencean 2000). Herniation has been
the tissues in the intervertebral disc, the most mechanically induced mechanically by applying a compressive force in
affected with degeneration is the nucleus pulposus. The loss the order of 5.4 kN to the disc in an anterolateral flexion
of osmotic pressure and hydration in the center of disc leads position. This results in extrusion of the nucleus pulposus
in a posterolateral radial direction (Aultman et al. 2005). discs (Baranto et al. 2005). The rationale behind this obser-
The internal strains developed during flexion/extension vation is that healthy, hydrated discs have a higher intradiscal
and lateral bending show tensile strains in the axial direc- pressure; in degenerated discs there is a lower compression
tion and thinning of the annulus fibrosus at the opposite stress in the center of the disc, while the posterior elements
direction of bending (Costi et al. 2008; Tsantrizos et al. transmit a larger portion of the compressive load (Adams and
2005). Therefore, a radial protrusion in posterolateral Dolan 2011).
direction occurs when there is anterolateral flexion. Non-
degenerated, highly hydrated discs have higher risk of her-
niation than severely degenerated discs (Gallagher 2002; 2.6 Summary of Critical Concepts
Simunic et al. 2001). This is probably due to the reduction Discussed in the Chapter
of intradiscal pressure in the nucleus pulposus with degen-
eration. However, when mechanical disruption is caused The mechanics of the intervertebral disc is determined by
by artificially increasing the nucleus pressure (instead of the interaction between the annulus fibrosus, nucleus pul-
applying a compressive load), degenerated discs fails at a posus, and endplates in different loading scenarios.
lower rupture pressure (Iencean 2000). The osmotic pressure plays an important role in the trans-
Herniation has also been induced in vitro when cyclic mission of forces through the spine as well as in the sta-
flexion/extension motion is applied to the disc (Callaghan bility of the intervertebral disc structure.
and McGill 2001). In this case, the herniation pathway is in Nonlinearity is an important mechanical characteristic of
a posterolateral radial direction. Increasing the compression disc tissues. Nonlinearity is evident in the spines rela-
load decreases the number of cycles required to cause disc tively lax neutral zone mechanics and stiffer linear region
damage. Similarly, the application of a static torque moment response in motion segment tests. Nonlinearity is impor-
shortens the disc cycle life (Drake et al. 2005). The increase tant to permit both disc motion and stability.
in intradiscal pressure due to the applied torque may acceler- Anisotropy (see Box 2.1) is an important mechanical
ate the susceptibility of intervertebral discs to injury. The characteristic of the annulus fibrosus and comes from the
shape of the disc has also been found to influence the hernia- structural organization of collagen fibers.
tion pathway in repetitive flexion/extension bending (Yates The viscoelastic behavior of the disc can be explained in
et al. 2010). Specifically, limacon-shaped discs had a defined part by the interstitial fluid flow during loading and
posterolateral herniation pathway, whereas oval-shaped discs unloading and the intrinsic viscoelasticity of disc tissues.
had a more diffuse herniation pathway. Degeneration affects the mechanics of disc tissues, which
is then reflected on the mechanics of the entire disc. One
major effect of degeneration is an increase of the range of
2.5.2 Endplate Fracture motion.
Several modes of injury, such as herniation and endplate
Another mechanically driven injury is the fracture of the fracture, are closely related to the pressure in the nucleus
vertebral endplate. The endplate is the cortical bone on the pulposus and are more frequent in healthy discs.
superior and inferior (cranial and caudal sides, respectively)
aspects of the vertebral body. On one side, the vertebral end- Acknowledgments This chapter is funded by research grants from the
plate is in contact with the intervertebral disc through the National Institutes of Health R01 AR 050052, R21 AR061751, and
R01 EB 002425.
cartilaginous endplate; on the other side, it is supported by
the trabecular bone inside the vertebral body. The main com-
ponent of the load applied to the vertebral endplate comes
from the intradiscal pressure. However, tension and shear References
forces are also applied by traction of the annulus fibrosus
Acaroglu ER, Iatridis JC, Setton LA, Foster RJ, Mow VC, Weidenbaum
(Baranto et al. 2005). Fracture of the vertebral endplate M (1995) Degeneration and aging affect the tensile behavior of human
occurs when the strains exceed the strength of the vertebral lumbar anulus fibrosus. Spine (Phila Pa 1976) 20(24):26902701
endplate (Fields et al. 2010). Endplate strength has been cor- Adams MA, Dolan P (1991) A technique for quantifying the bending
moment acting on the lumbar spine in vivo. J Biomech 24(2):117126
related with the density of the supporting trabecular bone
Adams MA, Dolan P (2011) Biomechanics of vertebral compres-
(Adams and Dolan 2011; Ordway et al. 2007; Zhao et al. sion fractures and clinical application. Arch Orthop Trauma Surg
2009). In fact, due to lower density of the cranial trabecular 131(12):17031710
bone, there is a greater incidence of fractures in this endplate. Adams MA, Hutton WC (1980) The effect of posture on the role of the
apophysial joints in resisting intervertebral compressive forces.
The degree of degeneration of the intervertebral disc also
J Bone Joint Surg Br 62(3):358362
affects the loads at which endplate fracture occurs: a higher Adams MA, Hutton WC (1981) The relevance of torsion to the mechanical
force is required to cause endplate fracture in degenerated derangement of the lumbar spine. Spine (Phila Pa 1976) 6(3):241248
Adams P, Muir H (1976) Qualitative changes with age of proteoglycans shear strain in six degrees of freedom: motions that place disc tissue
of human lumbar discs. Ann Rheum Dis 35(4):289296 at risk of injury. J Biomech 40(11):245766
Adams MA, Roughley PJ (2006) What is intervertebral disc degenera- Costi JJ et al (2011) Intervertebral disc properties: challenges for biode-
tion, and what causes it? Spine (Phila Pa 1976) 31(18):21512161 vices. Expert Rev Med Devices 8(3):357376
Adams MA, Bogduk N, Burton K, Dolan P (2006) The biomechanics Das DB, Welling A, Urban JP, Boubriak OA (2009) Solute transport in
of back pain, 2nd edn. Churchill Livingstone Elsevier, London, pp intervertebral disc: experiments and finite element modeling. Ann N
2948 Y Acad Sci 1161:4461
Ateshian GA, Ellis BJ, Weiss JA (2007) Equivalence between short- Diamant J, Keller A, Baer E, Litt M, Arridge RG (1972) Collagen;
time biphasic and incompressible elastic material responses. ultrastructure and its relation to mechanical properties as a function
J Biomech Eng 129(3):405412 of ageing. Proc R Soc Lond B Biol Sci 180(60):293315
Ateshian GA, Rajan V, Chahine NO, Canal CE, Hung CT (2009) Drake JD, Aultman CD, McGill SM, Callaghan JP (2005) The influence
Modeling the matrix of articular cartilage using a continuous fiber of static axial torque in combined loading on intervertebral joint
angular distribution predicts many observed phenomena. J Biomech failure mechanics using a porcine model. Clin Biomech (Bristol,
Eng 131(6):061003 Avon) 20(10):10381045
Aultman CD, Scannell J, McGill SM (2005) The direction of progres- Drost MR, Willems P, Snijders H, Huyghe JM, Janssen JD, Huson A
sive herniation in porcine spine motion segments is influenced by (1995) Confined compression of canine annulus fibrosus under
the orientation of the bending axis. Clin Biomech (Bristol, Avon) chemical and mechanical loading. J Biomech Eng 117(4):390396
20(2):126129 Elliott DM, Setton LA (2001) Anisotropic and inhomogeneous tensile
Azeloglu EU, Albro MB, Thimmappa VA, Ateshian GA, Costa KD behavior of the human anulus fibrosus: experimental measurement
(2008) Heterogeneous transmural proteoglycan distribution pro- and material model predictions. J Biomech Eng 123(3):256263
vides a mechanism for regulating residual stresses in the aorta. Am Eyre DR (1979) Biochemistry of the intervertebral disc. Int Rev Connect
J Physiol Heart Circ Physiol 294(3):H1197H1205 Tissue Res 8:227291
Baranto A, Ekstrom L, Hellstrom M, Lundin O, Holm S, Sward L Eyre DR, Muir H (1976) Types I and II collagens in intervertebral disc.
(2005) Fracture patterns of the adolescent porcine spine: an experi- Interchanging radial distributions in annulus fibrosus. Biochem
mental loading study in bending-compression. Spine (Phila Pa J 157(1):267270
1976) 30(1):7582 Ferguson SJ, Ito K, Nolte LP (2004) Fluid flow and convective transport
Bass EC, Ashford FA, Segal MR, Lotz JC (2004) Biaxial testing of of solutes within the intervertebral disc. J Biomech 37(2):213221
human annulus fibrosus and its implications for a constitutive for- Fields AJ, Lee GL, Keaveny TM (2010) Mechanisms of initial end-
mulation. Ann Biomed Eng 32(9):12311242 plate failure in the human vertebral body. J Biomech 43(16):
Batti MC, Videman T, Levlahti E, Gill K, Kaprio J (2008) Genetic 31263131
and environmental effects on disc degeneration by phenotype and Fujita Y, Wagner DR, Biviji AA, Duncan NA, Lotz JC (2000)
spinal level: a multivariate twin study. Spine (Phila Pa 1976) Anisotropic shear behavior of the annulus fibrosus: effect of harvest
33(25):28018. doi: 10.1097/BRS.0b013e31818043b7 site and tissue prestrain. Med Eng Phys 22(5):349357
Buckwalter JA (1995) Aging and degeneration of the human interverte- Gallagher S (2002) Letters. Spine (Phila Pa 1976) 27(12):1378
bral disc. Spine (Phila Pa 1976) 20(11):13071314 Gregory DE, Callaghan JP (2011) A comparison of uniaxial and biaxial
Callaghan JP, McGill SM (2001) Intervertebral disc herniation: studies mechanical properties of the annulus fibrosus: a porcine model.
on a porcine model exposed to highly repetitive flexion/extension J Biomech Eng 133(2):024503
motion with compressive force. Clin Biomech (Bristol, Avon) Gu WY, Mao XG, Foster RJ, Weidenbaum M, Mow VC, Rawlins BA
16(1):2837 (1999) The anisotropic hydraulic permeability of human lumbar
Canal Guterl C, Hung CT, Ateshian GA (2010) Electrostatic and non- anulus fibrosus. Influence of age, degeneration, direction, and water
electrostatic contributions of proteoglycans to the compressive content. Spine (Phila Pa 1976) 24(23):24492455
equilibrium modulus of bovine articular cartilage. J Biomech Guerin HA, Elliott DM (2006a) Degeneration affects the fiber reorien-
43(7):13431350 tation of human annulus fibrosus under tensile load. J Biomech
Cassidy JJ, Hiltner A, Baer E (1989) Hierarchical structure of the inter- 39(8):14101418
vertebral disc. Connect Tissue Res 23(1):7588 Guerin HA, Elliott DM (2006b) Structure and properties of soft tissues
Cavalcante FS, Ito S, Brewer K, Sakai H, Alencar AM, Almeida MP in the spine. In: Kurtz SM, Edidin AA (eds) SPINE: technology
et al (2005) Mechanical interactions between collagen and proteo- handbook. Elsevier Academic Press, Amsterdam/Boston
glycans: implications for the stability of lung tissue. J Appl Physiol Guerin HL, Elliott DM (2007) Quantifying the contributions of
98(2):672679 structure to annulus fibrosus mechanical function using a
Chuang SY, Popovich JM Jr, Lin LC, Hedman TP (2010) The effects of nonlinear, anisotropic, hyperelastic model. J Orthop Res 25(4):
exogenous crosslinking on hydration and fluid flow in the interver- 508516
tebral disc subjected to compressive creep loading and unloading. Guo Z, Shi X, Peng X, Caner F (2012) Fibre-matrix interaction in
Spine (Phila Pa 1976) 35(24):E1362E1366 the human annulus fibrosus. J Mech Behav Biomed Mater 5(1):
Chuong CJ, Fung YC (1986) On residual stresses in arteries. J Biomech 193205
Eng 108(2):189192 Hardingham TE, Muir H, Kwan MK, Lai WM, Mow VC (1987)
Comper WD, Laurent TC (1978) Physiological function of connective Viscoelastic properties of proteoglycan solutions with varying
tissue polysaccharides. Physiol Rev 58(1):255315 proportions present as aggregates. J Orthop Res 5(1):3646
Cortes DH, Elliott DM (2012) Extra-fibrillar matrix mechanics of Heneghan P, Riches PE (2008a) The strain-dependent osmotic pressure
annulus fibrosus in tension and compression. Biomech Model and stiffness of the bovine nucleus pulposus apportioned into ionic
Mechanobiol 11(6):781790 and non-ionic contributors. J Biomech 41(11):24112416
Costi JJ, Stokes IA, Gardner-Morse M, Laible JP, Scoffone HM, Iatridis Heneghan P, Riches PE (2008b) Determination of the strain-dependent
JC (2007) Direct measurement of intervertebral disc maximum hydraulic permeability of the compressed bovine nucleus pulposus.
shear strain in six degrees of freedom: motions that place disc tissue J Biomech 41(4):903906
at risk of injury. J Biomech 40(11):245766 Heuer F, Schmidt H, Wilke HJ (2008) The relation between interverte-
Costi JJ, Stokes IA, Gardner-Morse M, Laible JP, Scoffone HM, Iatridis bral disc bulging and annular fiber associated strains for simple and
JC (2008) Direct measurement of intervertebral disc maximum complex loading. J Biomech 41(5):10861094
Hickey DS, Hukins DW (1980) Relation between the structure of the Mauck RL, Baker BM, Nerurkar NL, Burdick JA, Li WJ, Tuan RS et al
annulus fibrosus and the function and failure of the intervertebral (2009) Engineering on the straight and narrow: the mechanics of
disc. Spine (Phila Pa 1976) 5(2):106116 nanofibrous assemblies for fiber-reinforced tissue regeneration.
Hollingsworth NT, Wagner DR (2011) Modeling shear behav- Tissue Eng Part B Rev 15(2):171193
ior of the annulus fibrosus. J Mech Behav Biomed Mater 4(7): Melrose J, Ghosh P, Taylor TK (2001) A comparative analysis of the
11031114 differential spatial and temporal distributions of the large (aggrecan,
Holm S, Maroudas A, Urban JP, Selstam G, Nachemson A (1981) versican) and small (decorin, biglycan, fibromodulin) proteoglycans
Nutrition of the intervertebral disc: solute transport and metabolism. of the intervertebral disc. J Anat 198(Pt 1):315
Connect Tissue Res 8(2):101119 Moon SM, Yoder JH, Wright AC, Smith LJ, Vresilovic EJ, Elliott DM
Hsu K, Zucherman J, Shea W, Kaiser J, White A, Schofferman J et al (2013) Evaluation of intervertebral disc cartilaginous endplate
(1990) High lumbar disc degeneration. Incidence and etiology. structure using magnetic resonance imaging. Eur Spine J
Spine (Phila Pa 1976) 15(7):679682 doi:10.1007/s00586-013-2798-1
Huyghe JM (2010) Biaxial testing of canine annulus fibrosus tissue under Michalek AJ, Gardner-Mose MG, Iatridis JC (2012). Large residual
changing salt concentrations. An Acad Bras Cienc 82(1):145151 strains are present in the intervertebral disc annulus fibrosus in the
Iatridis JC, Weidenbaum M, Setton LA, Mow VC (1996) Is the nucleus unloaded state. J Biomech 45(7):12271231
pulposus a solid or a fluid? Mechanical behaviors of the nucleus Mimura M, Panjabi MM, Oxland TR, Crisco JJ, Yamamoto I, Vasavada
pulposus of the human intervertebral disc. Spine (Phila Pa 1976) A (1994) Disc degeneration affects the multidirectional flexibility of
21(10):11741184 the lumbar spine. Spine (Phila Pa 1976) 19(12):13711380
Iatridis JC, Setton LA, Weidenbaum M, Mow VC (1997a) The vis- Nachemson A (1963) The influence of spinal movements on the lumbar
coelastic behavior of the non-degenerate human lumbar nucleus intradiscal pressure and on the tensil stresses in the annulus fibrosus.
pulposus in shear. J Biomech 30(10):10051013 Acta Orthop Scand 33:183207
Iatridis JC, Setton LA, Weidenbaum M, Mow VC (1997b) Alterations Nachemson AL (1981) Disc pressure measurements. Spine (Phila Pa
in the mechanical behavior of the human lumbar nucleus pulposus 1976) 6(1):9397
with degeneration and aging. J Orthop Res 15(2):318322 Nerurkar NL, Elliott DM, Mauck RL (2010) Mechanical design criteria for
Iatridis JC, Kumar S, Foster RJ, Weidenbaum M, Mow VC (1999) intervertebral disc tissue engineering. J Biomech 43(6):10171030
Shear mechanical properties of human lumbar annulus fibrosus. OConnell GD, Johannessen W, Vresilovic EJ, Elliott DM (2007)
J Orthop Res 17(5):732737 Human internal disc strains in axial compression measured noninva-
Iencean SM (2000) Lumbar intervertebral disc herniation following sively using magnetic resonance imaging. Spine (Phila Pa 1976)
experimental intradiscal pressure increase. Acta Neurochir (Wien) 32(25):28602868
142(6):669676 OConnell GD, Guerin HL, Elliott DM (2009) Theoretical and uniaxial
Jackson AR, Yuan TY, Huang CY, Travascio F, Yong Gu W (2008) experimental evaluation of human annulus fibrosus degeneration.
Effect of compression and anisotropy on the diffusion of glucose in J Biomech Eng 131(11)
annulus fibrosus. Spine (Phila Pa 1976) 33(1):17 OConnell GD, Vresilovic EJ, Elliott DM (2010) Human intervertebral
Jackson AR, Huang CY, Gu WY (2011) Effect of endplate calcification disc internal strain in compression: the effect of disc region, loading
and mechanical deformation on the distribution of glucose in inter- position, and degeneration. J Orthop Res 29(4):547555
vertebral disc: a 3D finite element study. Comput Methods Biomech OConnell GD, Jacobs NT, Sen S, Vresilovic EJ, Elliott DM (2011)
Biomed Engin 14(2):195204 Axial creep loading and unloaded recovery of the human interverte-
Jacobs NT, Smith LJ, Han WM, Morelli J, Yoder JH, Elliott DM (2011) bral disc and the effect of degeneration. J Mech Behav Biomed
Effect of orientation and targeted extracellular matrix degradation Mater 4(7):933942
on the shear mechanical properties of the annulus fibrosus. J Mech OConnell GD, Sen S, Elliott DM (2012) Human annulus fibrosus
Behav Biomed Mater 4(8):16111619 material properties from biaxial testing and constitutive modeling
Jacobs NT, Cortes DH, Vresilovic EJ, Elliott DM (2013) Biaxial ten- are altered with degeneration. Biomech Model Mechanobiol
sion of fibrous tissue: using finite element methods to address exper- 11(34):493503
imental challenges arising from boundary conditions and anisotropy. Ordway NR, Lu YM, Zhang X, Cheng CC, Fang H, Fayyazi AH (2007)
J Biomech Eng 135(2):021004. doi:10.1115/1.4023503 Correlation of cervical endplate strength with CT measured sub-
Johannessen W, Elliott DM (2005) Effects of degeneration on the chondral bone density. Eur Spine J 16(12):21042109
biphasic material properties of human nucleus pulposus in confined Overbeek JT (1956) The Donnan equilibrium. Prog Biophys Biophys
compression. Spine (Phila Pa 1976) 30(24):E724E729 Chem 6:5784
Kastelic J, Baer E (1980) Deformation in tendon collagen. Symp Soc Panjabi M, Brown M, Lindahl S, Irstam L, Hermens M (1988) Intrinsic
Exp Biol 34:397435 disc pressure as a measure of integrity of the lumbar spine. Spine
Kastelic J, Galeski A, Baer E (1978) The multicomposite structure of (Phila Pa 1976) 13(8):913917
tendon. Connect Tissue Res 6(1):1123 Pearcy M, Portek I, Shepherd J (1984) Three-dimensional x-ray analy-
Kiani C, Chen L, Wu YJ, Yee AJ, Yang BB (2002) Structure and func- sis of normal movement in the lumbar spine. Spine (Phila Pa 1976)
tion of aggrecan. Cell Res 12(1):1932 9(3):294297
Klisch SM, Lotz JC (2000) A special theory of biphasic mixtures and Perie D, Korda D, Iatridis JC (2005) Confined compression experi-
experimental results for human annulus fibrosus tested in confined ments on bovine nucleus pulposus and annulus fibrosus: sensitivity
compression. J Biomech Eng 122(2):180188 of the experiment in the determination of compressive modulus and
Lanir Y (2009) Mechanisms of residual stress in soft tissues. J Biomech hydraulic permeability. J Biomech 38(11):21642171
Eng 131(4):044506 Perie D, Iatridis JC, Demers CN, Goswami T, Beaudoin G, Mwale F
Maroudas A, Bannon C (1981) Measurement of swelling pressure in et al (2006a) Assessment of compressive modulus, hydraulic per-
cartilage and comparison with the osmotic pressure of constituent meability and matrix content of trypsin-treated nucleus pulposus
proteoglycans. Biorheology 18(36):619632 using quantitative MRI. J Biomech 39(8):13921400
Massey CJ, van Donkelaar CC, Vresilovic E, Zavaliangos A, Perie DS, Maclean JJ, Owen JP, Iatridis JC (2006b) Correlating mate-
Marcolongo M (2011) Effects of aging and degeneration on the rial properties with tissue composition in enzymatically digested
human intervertebral disc during the diurnal cycle: a finite element bovine annulus fibrosus and nucleus pulposus tissue. Ann Biomed
study. J Orthop Res 30(1):122128 Eng 34(5):769777
Potvin JR, McGill SM, Norman RW (1991) Trunk muscle and lumbar Urban JP, Maroudas A, Bayliss MT, Dillon J (1979) Swelling pressures
ligament contributions to dynamic lifts with varying degrees of of proteoglycans at the concentrations found in cartilaginous tis-
trunk flexion. Spine (Phila Pa 1976) 16(9):10991107 sues. Biorheology 16(6):447464
Pye SR, Reid DM, Lunt M, Adams JE, Silman AJ, ONeill TW (2007) Urban JP, Holm S, Maroudas A, Nachemson A (1982) Nutrition of the
Lumbar disc degeneration: association between osteophytes, end-plate intervertebral disc: effect of fluid flow on solute transport. Clin
sclerosis and disc space narrowing. Ann Rheum Dis 66(3):330333 Orthop Relat Res 170:296302
Rannou F, Lee TS, Zhou RH, Chin J, Lotz JC, Mayoux-Benhamou MA Urban JP, Smith S, Fairbank JC (2004) Nutrition of the intervertebral
et al (2004) Intervertebral disc degeneration: the role of the mito- disc. Spine (Phila Pa 1976) 29(23):27002709
chondrial pathway in annulus fibrosus cell apoptosis induced by van der Veen AJ, van Dieen JH, Nadort A, Stam B, Smit TH
overload. Am J Pathol 164(3):915924 (2007) Intervertebral disc recovery after dynamic or static load-
Sasaki M, Takahashi T, Miyahara K, Hirosea T (2001) Effects of chon- ing in vitro: is there a role for the endplate? J Biomech 40(10):
droitinase ABC on intradiscal pressure in sheep: an in vivo study. 22302235
Spine (Phila Pa 1976) 26(5):463468 van Deursen DL, Snijders CJ, Kingma I, van Dieen JH (2001a) In vitro
Sato K, Kikuchi S, Yonezawa T (1999) In vivo intradiscal pressure torsion-induced stress distribution changes in porcine intervertebral
measurement in healthy individuals and in patients with ongoing discs. Spine (Phila Pa 1976) 26(23):25822586
back problems. Spine (Phila Pa 1976) 24(23):24682474 van Deursen DL, Snijders CJ, van Dieen JH, Kingma I, van Deursen LL
Setton LA, Zhu W, Weidenbaum M, Ratcliffe A, Mow VC (1993) (2001b) The effect of passive vertebral rotation on pressure in the
Compressive properties of the cartilaginous end-plate of the baboon nucleus pulposus. J Biomech 34(3):405408
lumbar spine. J Orthop Res 11(2):228239 Wade KR, Robertson PA, Broom ND (2011) A fresh look at the nucleus-
Shirazi-Adl A, Ahmed AM, Shrivastava SC (1986) Mechanical response endplate region: new evidence for significant structural integration.
of a lumbar motion segment in axial torque alone and combined Eur Spine J 20(8):12251232
with compression. Spine (Phila Pa 1976) 11(9):914927 Wagner DR, Lotz JC (2004) Theoretical model and experimental results
Shirazi-Adl A, Taheri M, Urban JP (2010) Analysis of cell viability in for the nonlinear elastic behavior of human annulus fibrosus.
intervertebral disc: effect of endplate permeability on cell popula- J Orthop Res 22(4):901909
tion. J Biomech 43(7):13301336 Wang C, Witschey W, Elliott MA, Borthakur A, Reddy R (2010)
Simunic DI, Broom ND, Robertson PA (2001) Biomechanical factors Measurement of intervertebral disc pressure with T 1rho MRI.
influencing nuclear disruption of the intervertebral disc. Spine Magn Reson Med 64(6):17211727
26(11):12231230 Wilke HJ, Wolf S, Claes LE, Arand M, Wiesend A (1996) Influence of
Smith LJ, Nerurkar NL, Choi KS, Harfe BD, Elliott DM (2011) varying muscle forces on lumbar intradiscal pressure: an in vitro
Degeneration and regeneration of the intervertebral disc: lessons study. J Biomech 29(4):549555
from development. Dis Model Mech 4(1):3141 Wilke HJ, Neef P, Caimi M, Hoogland T, Claes LE (1999) New in vivo
Soukane DM, Shirazi-Adl A, Urban JP (2007) Computation of coupled measurements of pressures in the intervertebral disc in daily life.
diffusion of oxygen, glucose and lactic acid in an intervertebral disc. Spine (Phila Pa 1976) 24(8):755762
J Biomech 40(12):26452654 Wu HC, Yao RF (1976) Mechanical behavior of the human annulus
Spencer AJM (1984) Continuum theory of the mechanics of fibre- fibrosus. J Biomech 9(1):17
reinforced composites. Springer, Wien/New York Yates JP, Giangregorio L, McGill SM (2010) The influence of interver-
Spenciner D, Greene D, Paiva J, Palumbo M, Crisco J (2006) The mul- tebral disc shape on the pathway of posterior/posterolateral partial
tidirectional bending properties of the human lumbar intervertebral herniation. Spine (Phila Pa 1976) 35(7):734739
disc. Spine J 6(3):248257 Zhao FD, Pollintine P, Hole BD, Adams MA, Dolan P (2009) Vertebral
Travascio F, Gu WY (2011) Simultaneous measurement of anisotropic fractures usually affect the cranial endplate because it is thin-
solute diffusivity and binding reaction rates in biological tissues by ner and supported by less-dense trabecular bone. Bone 44(2):
FRAP. Ann Biomed Eng 39(1):5365 372379
Tsantrizos A, Ito K, Aebi M, Steffen T (2005) Internal strains in healthy and Zhu W, Mow VC, Koob TJ, Eyre DR (1993) Viscoelastic shear proper-
degenerated lumbar intervertebral discs. Spine 30(19):21292137 ties of articular cartilage and the effects of glycosidase treatments.
Urban JP, Maroudas A (1981) Swelling of the intervertebral disc J Orthop Res 11(6):771781
in vitro. Connect Tissue Res 9(1):110 Zhu W, Chern KY, Mow VC (1994) Anisotropic viscoelastic
Urban JP, Holm S, Maroudas A (1978) Diffusion of small solutes into the shear properties of bovine meniscus. Clin Orthop Relat Res
intervertebral disc: as in vivo study. Biorheology 15(34):203221 306:3445
Development of the Intervertebral Disc
3
Megan K. Cox and Rosa Serra
Contents 3.1 Overview

3.1 Overview .......................................................................... 33
Intervertebral discs are derived from embryonic structures
3.2 Development of the Notochord and
the Formation of the Nucleus Pulposus......................... 35 called the sclerotome and notochord (Paavola et al. 1980;
3.2.1 Development and Maintenance of the Notochord............. 35 Theiler 1988; Rufai et al. 1995). The nucleus pulposus, the
3.2.2 The Notochord Sheath and Formation cushioning core of the mature intervertebral disc, is derived
of the Nucleus Pulposus .................................................... 36 from the notochord, while the annulus fibrosus, which pro-
3.2.3 Evidence for Lineage of the Nucleus Pulposus ................ 37
vides the structural properties of the disc, is derived from
3.3 Somitogenesis................................................................... 38 sclerotome (Christ et al. 2004, 2007; Christ and Scaal 2008).
3.3.1 Clock and Wavefront ......................................................... 39
The sclerotome is derived from the somites, transient struc-
3.3.2 Epithelialization ................................................................ 41
3.3.3 Segment Identity ............................................................... 41 tures that determine the segmented nature of the embryo. In
response to signals from the notochord and floor plate of the
3.4 Formation of Sclerotome ................................................ 42
neural tube, the maturing somites undergo dorsal-ventral
3.5 Resegmentation ............................................................... 42 compartmentalization establishing the dermomyotome and
3.6 Sclerotome Derivatives ................................................... 42 sclerotome, the latter forming most of the connective tissues
3.6.1 Differentiation of the Annulus Fibrosus ........................... 44 of the future axial skeleton. The development of the sclero-
3.6.2 Differentiation of Vertebral Cartilage ............................... 45
tome is characterized by proliferation and expansion of cells
3.6.3 Boundary Between Vertebral Body
and Annulus Fibrosus........................................................ 45 as well as the formation of three subcompartments: ventral,
3.6.4 Syndetome (Tendon) ......................................................... 45 lateral, and dorsal. The ventral sclerotome gives rise to the
3.7 Developmental Pathways Involved vertebral bodies and annulus fibrosus and is made up of Pax-
in Maintenance of Postnatal Disc................................... 46 1-expressing cells that have invaded the perinotochordal
space (Monsoro-Burq et al. 1994; Peters et al. 1999).
3.8 Closing .............................................................................. 46
In addition to dorsal-ventral compartmentalization, each
3.9 Summary of Critical Concepts Discussed sclerotome segment demonstrates rostral to caudal polarity
in the Chapter .................................................................. 46
(Christ et al. 2007; Christ and Scaal 2008). This polarity is
References ...................................................................................... 46 morphologically apparent in the ventral sclerotome as a
condensed caudal portion and loose rostral portion within
each segment. Due to resegmentation of sclerotome during
development, the caudal domain of the sclerotome will form
the anterior structures of each vertebra and the rostral
domain forms the posterior structures (Huang et al. 2000).
The annulus fibrosus will form from the cells in the sclero-
tome adjacent to the border, sometimes called von Ebners
fissure, between the sclerotome halves. These cells can be
M.K. Cox R. Serra (*) traced back to somitocoele cells at the center of the somite
Department of Cell, Developmental, and Integrative Biology,
University of Alabama at Birmingham,
before the formation of the sclerotome. This subcompart-
1918 University Blvd. 660 MCLM, Birmingham, AL 35294, USA ment of the sclerotome has been termed the arthrotome
e-mail: mkcox@uab.edu, rserra@uab.edu (Mittapalli et al. 2005).

34 M.K. Cox and R. Serra
Endochondral bone formation follows the expansion, basis for tissue engineering protocols (Lenas et al. 2009a,
migration, and patterning of the sclerotome to form the ver- 2011; Gadjanski et al. 2012). Recently, the concept of devel-
tebral body. As the vertebral bodies undergo chondrogenesis, opmental engineering has been used to generate chondro-
notochord cells are removed from the vertebral region and cytes from embryonic stem cells (Oldershaw et al. 2010).
expand into the intervertebral disc region to form the nucleus This concept could also be used in the future to engineer the
pulposus. The annulus fibrosus develops into a fibrocartilage intervertebral disc (Box 3.1). This chapter will cover the
structure and does not normally undergo endochondral molecular mechanisms governing the major steps of inter-
ossification. The annulus fibrosus can be further divided into vertebral disc development with an emphasis on processes
an inner portion and a more fibrous outer portion. TGF-3 is important for understanding human disease and potential for
one of the earliest markers of the developing intervertebral tissue engineering.
disc within the sclerotome (Pelton et al. 1990). Later, it is
preferentially expressed in the outer annulus. In the adult, the
annulus fibrosus is bound by the spinal ligaments, which Box 3.1: Developmental Engineering
insert into the bone to form the entheses. It is likely that the and the Intervertebral Disc
ligament is also derived from the sclerotome although this Developmental biology can provide insights into
has not been addressed directly. Tendons in the axial skeleton novel strategies for repair, regeneration, or engineer-
are derived from a subcompartment of the sclerotome call ing of tissues. The effort to develop in vitro processes
the syndetome (Brent et al. 2003; Schweitzer et al. 2001). that mimic normal development was recently termed
The adult disc is juxtaposed to the cartilage end plate of the developmental engineering (Lenas et al. 2011, 2009a,
adjacent vertebra, which consists of hyaline cartilage similar b; Gadjanski et al. 2012). The term is used to empha-
to that found in the peripheral joints. The fibrocartilage of the size the concept that it is not the tissue, but rather the
adult annulus fibrosus, the cartilage-like matrix of the nucleus process of development that needs to be engineered.
pulposus, and the hyaline cartilage of the vertebrae have dis- Noteworthy, conventional tissue engineering has been
tinct and overlapping properties. All contain collagen II and an empirical discipline, the goal of which is to generate
aggrecan. The annulus fibrosus also contains collagen I with functional tissues using scaffold, cells, and signaling
higher levels in the outer annulus. Versican and fibromodulin molecules. By incorporating information on develop-
(Fmod) are preferentially expressed in the annulus fibrosus ment, a rational methodology can be developed to gen-
relative to cartilage (Smits and Lefebvre 2003; Shi et al. erate tissues in vitro.
2003; Sohn et al. 2010). Versican is also present in ligaments Developmental engineering aims to build on the
and entheses (Shi et al. 2003) (see Chap. 4). Keratin 8, most significant aspects of embryonic development: the
Keratin 18, Keratin 19, and NCAM1 have recently been highly regulated spatiotemporal organization of cells
identified as markers to distinguish the nucleus pulposus into complex tissues. This approach would thus allow
from the annulus fibrosus and hyaline cartilage (Sakai et al. the separate sequential steps that correspond to varying
2009; Lee et al. 2007; Minogue et al. 2010). Brachyury/T is stages of development to generate tissue intermediates
also considered a marker for the notochord as well as the that can act as modular functional units of the engi-
nucleus pulposus (Kispert et al. 1994). neered tissue (Lenas et al. 2011, 2009a). This gradual
Insight into the mechanisms of pathology in the spine can and stepwise process is inherently stable and can be
be provided through an understanding of the development of controlled for quality at each step. The early stages of
the axial skeleton. Since the intervertebral disc is derived development would be given the highest scrutiny since
from the notochord, somites, and sclerotome, alterations in subsequent development could progress naturally and
the development of any of these tissues can result in human in many cases would be semiautonomous. In addition,
developmental disorders that affect the intervertebral disc. care would need to be taken not to interfere with nor-
Alterations in the segmentation of the somites can lead to mal cell-cell interactions or morphogen gradients since
congenital defects in the formation of the vertebrae and the this would disrupt the architecture of the final product.
disc resulting in fusion of vertebrae (Turnpenny 2008; Shifley Functional modules can then be used to generate more
and Cole 2007). Genetic variations involved in increased complex organs (Lenas et al. 2011).
susceptibility to intervertebral disc degeneration can be asso- Early forms of developmental engineering have
ciated with subtle developmental abnormalities (Jin et al. been used to direct human embryonic stem cells to
2011; Dahia et al. 2009). In addition to providing insight into chondrocytes and to recapitulate endochondral bone
the etiology of disorders of the spine, an understanding of formation using mesenchymal stem cells (Oldershaw
developmental biology can provide a basis for treatment, et al. 2010; Scotti et al. 2010). In both cases, the step-
repair, or regeneration strategies. Information about how the wise approach mimicking normal development was
axial skeleton develops in the first place would also be the
3 Development of the Intervertebral Disc 35
The notochordal plate buds off of the endoderm to finally

used. Although the intervertebral disc is more complex form the notochord, which lies between the roof of the primi-
than cartilage or bone, the use of a stepwise processes tive gut and the floor of the developing neural tube (Moore
based on information about embryonic development and Oersaud 2003).
would facilitate the disc tissue engineering process.
Special consideration would have to be given to gradi-
ents within the tissue and integration of the annulus 3.2.1 Development and Maintenance
fibrosus and the nucleus pulposus with each other and of the Notochord
with the surrounding end plate and ligaments. The
pathways controlling these aspects of intervertebral Mutations in 19 genes resulting in absence of the notochord
disc development are starting to be elucidated. have been catalogued by the Mouse Genome Informatics
(MGI) database (www.informatics.jax.org). The most well
characterized is the T gene, encoding the brachyury pro-
tein (Kispert et al. 1994). Brachyury is the prototype T-box
3.2 Development of the Notochord and transcription factor. It is also considered a marker of primi-
the Formation of the Nucleus Pulposus tive mesoderm and notochord. It starts out being expressed
in all mesodermal cells. Later, it becomes restricted to
The notochord is a transient rodlike mesodermal structure the notochord where expression is maintained. Mice with
that is located under the neural tube and spans most of the homozygous mutations in T have defects in the formation
length of the embryo (Stemple 2005). It serves as a primitive of mesoderm, and the trunk notochord is not established
support structure for the embryo and as a signaling center, (Dobrovolskaia-Zavadskaia 1927). Subsequent malforma-
directing development of surrounding structures including tions in the spine and allantois lead to the early demise
the sclerotome. In mice and humans, the notochord will of the embryo. Mice with a dominant-negative mutation
eventually develop into the nucleus pulposus of the interver- in T (Tc) survive but demonstrate abnormal nucleus pul-
tebral disc (Fig. 3.1). The notochord forms during embryonic posus morphology (Stott et al. 1993). Brachyury is also
gastrulation. Cells that migrate through the primitive streak a marker for chordomas, rare malignant tumors along the
will form the endoderm and mesoderm. Some of the migrat- spine thought to arise from persistent remnants of the
ing cells will form the notochordal process, which integrates notochord (Vujovic et al. 2006). Duplications in the T
transiently with the endoderm to form the notochordal plate. gene in humans have been associated with susceptibility to
a b c
Fig. 3.1 Development of the
notochord and nucleus pulposus.
(a) The nucleus is derived from
the notochord, a rod-shaped
embryonic structure that lies
under the neural tube. Sclerotome
condenses around the notochord
to form the vertebrae and annulus
fibrosus of the intervertebral disc.
Brachyury/T, Sd, Shh, and
Sox5/6/9 are required for the
formation of notochord and
notochord sheath. (b) Once the
vertebrae and disc start to form,
the notochord contracts from the
vertebral body and expands into
the area of the future disc. The
notochord sheath and collagen II,
which helps to maintain osmotic
pressure in the vertebral cartilage,
are required for the expansion of
the notochord and formation of
Notochord derived
the nucleus pulposus. (c) Growth
and maintenance of the nucleus Sclerotome vertebrae
pulposus is in part controlled by
the Skt gene Sclerotome annulus
chordoma (Yang et al. 2009). For a complete discussion of Lefebvre 2003; Choi and Harfe 2011). Sox5 and Sox6 are
chordomas, see Chap. 18. Sry-related HMG box transcription factors that cooper-
While mice with mutations in T fail to establish the noto- ate with Sox9 to mediate chondrogenesis (Lefebvre 2002).
chord, Danforths short-tail (Sd) mutation results in failure Inactivation of Sox5 and Sox6 result in dramatic chon-
to maintain the notochord (Paavola et al. 1980). Sd mice drodysplasia with impaired chondrocyte differentiation
arose from a spontaneous mutation in a yet unknown gene. (Smits et al. 2001). Sox5/6 is also required for the forma-
Sd mice have abnormal nucleus pulposus morphology likely tion of the notochord sheath and subsequent survival of the
due to defects in the formation and maintenance of the noto- notochord and development of the nucleus pulposus (Smits
chord early in development. In Sd mice the notochord is and Lefebvre 2003). The sheath contains many matrix pro-
discontinuous and becomes increasingly fragmented, even- teins that are also found in cartilage including collagen II
tually disappearing in mice homozygous for the mutant and proteoglycans containing sulfated glycosaminoglycans
allele. In heterozygous mice, the intervertebral disc forms like aggrecan and perlecan. The aggrecan and perlecan con-
but the nucleus pulposus is absent and the disc is occupied tent in the notochord sheath from Sox5/6 mutant mice was
with fibrous tissue similar to the annulus fibrosus (Semba dramatically reduced, whereas collagen II was not affected.
et al. 2006). It was suggested that Sox5/6 promotes the formation of the
Mutations in the Sickle tail gene (Skt), on the same chro- notochord sheath by regulating matrix gene expression in
mosome as Sd, have nucleus pulposi; however, the nuclei are the notochord cells. Failure to maintain the sheath matrix
shifted to the periphery of the disc (Semba et al. 2006). The resulted in premature death of notochord cells and an aber-
boundary between the nucleus pulposus and the annulus rant nucleus pulposus.
fibrosus is altered, and the annulus exhibits thin fibrous lay- Shh, a secreted signaling molecule important for several
ers relative to control mice. The sequence of the Skt gene was aspects of development, is also required for the formation
identified using the gene-trapped ES clone from which it was and maintenance of the notochordal sheath (Choi and Harfe
derived. The protein product is predicted to have a proline- 2011). Shh is secreted by the notochord as well as the nucleus
rich region in the N-terminus and a coiled-coil domain in the pulposus in embryonic and postnatal life (Dahia et al. 2009;
middle. The coiled-coil domain was similar to those found in DiPaola et al. 2005). Shh binds to its receptor, Ptc, on nearby
a large number of scaffold proteins including keratins. It was cells, thus relieving repression of Smoothened, Smo, a
suggested that while Sd is required at early stages of noto- transmembrane protein that is required to transmit the Shh
chord development, Skt is required later in development for signal (Ingham and McMahon 2001). In mice containing a
proper growth, differentiation, and maintenance of the germ line null allele of Shh, the notochord was formed, but
nucleus pulposus. Furthermore, specific polymorphisms in was not maintained (Chiang et al. 1996). The embryos died
the human SKT gene have been significantly associated with shortly thereafter preventing analysis of development of the
lumbar disc herniation in Japanese and Finnish case- intervertebral disc. To address the role of Shh in the mainte-
controlled populations (Karasugi et al. 2009). nance of the notochord and the subsequent formation of the
nucleus pulposus, a conditionally deleted allele of Smo in
mice was used (Choi and Harfe 2011). In these experiments
3.2.2 The Notochord Sheath and the investigators removed Smo from all Shh-expressing cells,
Formation of the Nucleus Pulposus including the notochord. The notochord sheath was missing
in the Smo-deleted mice. Furthermore, the nucleus pulposus
A sheath of extracellular matrix containing collagens and did not expand into the intervertebral disc region, and noto-
glycoproteins, including collagen II and laminin, surrounds chord cells were scattered throughout the vertebral column.
the mesodermal cells of the notochord (Gotz et al. 1995). Notochord remnants in the vertebral body could be marked
It has been proposed that the notochord sheath functions by ROSA26 in ShhCreERT2 mice. If Smo was removed after
to contain and direct internal hydrostatic pressure within the notochord sheath formed, the nucleus pulposus was not
the notochord. The notochord is the forerunner of the axial affected indicating the importance of the sheath in this tissue
skeleton in non-vertebrate chordates (Box 3.2). During early formation.
Xenopus development, osmotic inflation of the notochord Collagen II (Col2) is the major collagen in cartilage and
against the sheath results in lengthening and straightening an important component of the notochordal sheath (Swiderski
of the embryo (Adams et al. 1990). Similar mechanics-based and Solursh 1992; Gotz et al. 1995). The notochord is not
mechanisms involving the notochord sheath may be involved removed from vertebral bodies, and intervertebral discs do
in morphogenesis of the nucleus pulposus. not form normally in mice with a null mutation in Col2a1
It has been shown that mutations in genes encoding pro- (Aszdi et al. 1998). Cartilage proteins including collagens
teins that affect the formation of the notochord sheath also IX and XI, aggrecan, and COMP appear to be expressed in
affect the development of the nucleus pulposus (Smits and these mice, but collagens I and II are inappropriately
expressed in the cartilage of the presumptive vertebrae. As a

result, the collagen fibers in cartilage are disorganized. It was chicken and quail, lack a nucleus pulposus and there is
proposed that this disorganization would lead to a weakened limited, if any, contribution of the notochord to the
structure unable to contain the cartilage osmotic pressure. adult axial skeleton (Bruggeman et al. 2012). Since the
The reduction in internal pressure within the vertebral body contribution of the notochord to the axial skeleton and
could be responsible for the failure of the notochord to be development of a nucleus pulposus-like structure
removed from the vertebrae and expand into the interverte- occurred very early in vertebrate evolution, before
bral disc (Adams et al. 1990), as a result, the nucleus pulpo- significant contribution of sclerotome derived tissues,
sus would not form. Mutations in other genes that are it seems likely that the nucleus pulposus was lost dur-
expressed in the sclerotome and affect the development of ing the evolution of reptiles and birds. This loss could
the vertebral body including Nkx3.2 and Pax1/9 also affect have been due to either selective pressure to lose the
the removal of the notochord from the vertebral body and nucleus pulposus or lack of pressure to keep it. Perhaps
development of the nucleus pulposus, further supporting the as the sclerotome evolved, it was able to take over all
requirement of mechanical pressure for removal/expansion of the functions of the adult axial skeleton that were
of the nucleus pulposus (Peters et al. 1999; Lettice et al. previously served by the notochord.
1999; Tribioli and Lufkin 1999). The line that would lead to mammals (synapsids)
and that of reptiles and birds (saurapsids) diverged
from the basal amniotes some 320 million years ago.
Box 3.2: Evolution of the Intervertebral Disc Whatever the reason (pressures related to gait or
The notochord is the forerunner of the axial skeleton in mechanical loading), mammals retained the notochord-
early non-vertebrate chordates. In basal vertebrates derived structure. Of course, proof of this model will
including lamprey and fish (teleosts), most of the axial require additional fossil evidence, which is scarce for
skeleton is derived from the notochord (Koob and unmineralized tissue; more complete comparative
Long 2000; Ytteborg et al. 2012; Haga et al. 2009; anatomy of existing species; and further experimenta-
Dale and Topczewski 2011; Inohaya et al. 2007). tion, including lineage tracing and molecular biology,
Sclerotome is not detected in non-vertebrate chordates. on the development of the intervertebral disc in multi-
A primitive sclerotome, however, is observed in the ple species.
most basal vertebrates and teleosts (Scaal and Wiegreffe
2006; Keller 2000; Koob and Long 2000; Inohaya
et al. 2007). Development of the axial skeleton in 3.2.3 Evidence for Lineage
teleosts, including salmon, medaka, and zebra fish, of the Nucleus Pulposus
begins with patterned mineralization of the notochord
sheath where the vertebrae will form (Ytteborg et al. The lineage of the cells in the adult nucleus pulposus has
2012; Haga et al. 2009; Dale and Topczewski 2011; been the source of much debate (Risbud et al. 2010; Erwin
Inohaya et al. 2007). In adult fish, the disc consists of a 2010; Shapiro and Risbud 2010). It has been proposed that
central core, similar to the nucleus pulposus, contain- the notochord forms the initial central part of the interverte-
ing cells with large fluid-filled vacuoles derived from bral disc, but as this compartment matures, these large noto-
the notochord, a fibrous layer derived from the notochord cells are replaced with smaller cells that more closely
chord sheath, and an elastic membrane or ligament that resemble chondrocytes, perhaps recruited from the end plate
is likely derived from the primitive sclerotome or the inner annulus (Wamsley 1953; Kim et al. 2003).
(Ytteborg et al. 2012; Haga et al. 2009; Dale and However, more recently, evidence has accumulated to sug-
Topczewski 2011; Inohaya et al. 2007). gest that all of the cells within the nucleus pulposus are
Comparative anatomy suggests the sclerotome indeed derived from the notochord.
evolved further in amphibians. In some, the sclerotome Shh is highly expressed in the notochord and later in the
composes a large portion of the differentiating somite, nucleus pulposus (Dahia et al. 2009; DiPaola et al. 2005;
and segmentation is independent of the notochord Choi et al. 2008). Mice that express an inducible Cre under
(Scaal and Wiegreffe 2006; Keller 2000). By the time the control of the Shh promoter were used in fate mapping
amniotes evolved, they already had a well-developed studies of the notochord (Choi et al. 2008). To exclude the
intervertebral disc with a nucleus-like tissue derived possibility that Shh-Cre would mark cells in the notochord
from the notochord and an annulus-like fibrous tissue and then again in a new nucleus pulposus cell population,
derived from an evolving sclerotome. It was recently a tamoxifen-inducible Cre was employed. Notochord cells
shown that some adult reptiles and birds, including were pulse labeled by administration of tamoxifen at an
early stage of development and then followed over time with
-galactosidase staining through Cre-mediated activation of cells composed of an outer epithelial layer surrounding
of the ROSA26-LacZ locus. When mouse embryos were a mesenchymal core, the somitocoele (Ferrer-Vaquer et al.
pulse labeled at E8.0 days of gestation and examined at 2010). The formation of somites is tightly controlled dur-
E13.5, it was clear that all of the cells of the nucleus pulpo- ing development both spatially and temporally. Somite pairs
sus were labeled. Furthermore, when old mice (19 months) bud off from the anterior end of the PSM at time intervals
were examined, all of the cells of the nucleus pulposus were specific to the species (Pourquie 2011; Brand-Saberi et al.
labeled with no labeling observed in the annulus fibrosus. 2008). Somite nomenclature is based on the relationship of
More recently, notochord cells were traced using a Noto- the somite to the anterior end of the PSM with the closest
Cre mouse (McCann et al. 2012). Noto is a highly conserved somite labeled number SI followed by SII, SIII, etc., count-
transcription factor with expression limited to the node and ing toward the anterior end of the embryo (Christ and Ordahl
early notochord. Again, when cells were marked by activation 1995). Future somites in the presomitic mesoderm are labeled
of the ROSA26-LacZ locus by Noto-Cre, all of the nucleus S0 and SI. These somites have not budded off the PSM, but
pulposus cells in the adult, after dramatic changes in cellular can be seen histologically or with molecular markers and are
morphology, were still labeled although only a subset of adult frequently referred to as somitomeres. It is important to note
nucleus cells stained for K8, a putative marker for nucleus that the first five somites to bud off are destined to fuse and
pulposus cells. The fate mapping studies strongly suggest the form the occipital bone (Couly et al. 1993). The remaining
nucleus pulposus is derived completely from the notochord. somites will progress into the components of the axial skel-
Evidence that the nucleus pulposus is derived from the noto- eton and skeletal muscle. Disruptions to somitogenesis and
chord also comes from gene profiling studies. Recent studies segmentation of the somites result in birth defects that can
used microarray technology to compare the gene expression severely affect the function of the spine by disrupting the for-
patterns in human, bovine, canine, and rodent nucleus pulpo- mation and shape of both vertebral bodies and intervertebral
sus, annulus fibrosus, and articular cartilage (Minogue et al. disc (Box 3.3; Turnpenny 2008).
2010; Lee et al. 2007; Sakai et al. 2009). Brachyury/T, K8,
K18, and K19 were highly expressed in the nucleus pulpo-
sus. Brachyury/T is known as a marker of the notochord and Box 3.3: Human Pathologies Associated with
also a marker of chordoma, notochord tumors, suggesting that Segmentation Defects: Klippel-Feil Syndrome
the nucleus pulposus could be derived from these embryonic Many congenital defects of the spine are caused by
cells. More importantly, gene expression patterns significantly problems in segmentation of the somites (Turnpenny
overlapped, and expression of the above nucleus pulposus 2008; Shifley and Cole 2007). Some of these are linked
markers was similar in the large notochord-like cells and the to teratogens such as retinoic acid (RA) which can cause
smaller chondrocytic cells within the nucleus pulposus. In con- axial skeleton abnormalities by interfering with the
trast to this result, another group showed heterogeneity in the normal retinoid signaling and thus affecting the normal
nucleus pulposus with regard to K8 expression (Gilson et al. timing and location of segment formation (Alexander
2010). Heterogeneity in K8 expression was also observed in and Tuan 2010). Various types of spondylocostal dys-
the Noto-Cre cell fate mapping studies described above, even ostosis have been linked to the clock genes Dll3, Lnfg,
though all of the nucleus pulposus cells were clearly derived Hes7, and Mesp2 (Pourquie 2011). Alagille syndrome,
from Noto-expressing cells (McCann et al. 2012). A likely which is characterized by misshapen butterfly verte-
explanation is that the notochord can differentiate into all of brae caused by dorsal fusion failure, has been linked to
the cell types including K8-expressing and K8-non-expressing mutations in Jagged1, a Notch ligand (Oda et al. 1997;
cells, within the nucleus pulposus. This idea is supported by Li et al. 1997). Segmentation defects can also affect
studies showing that notochord cells from rabbit can differ- the intervertebral disc as well as the vertebra. Most
entiate into cells with varying morphological characteristics prominently, the disc is completely absent with fused
similar to those observed in the adult nucleus pulposus (Kim block vertebrae and pushed to one side when only par-
et al. 2009). Taken together, there is now considerable evi- tial vertebral fusion occurs (Turnpenny 2008; Shifley
dence indicating that all of the cell types in the adult nucleus and Cole 2007). Unfortunately, many clinical observa-
pulposus are derived from the notochord. tions fail to include discal analysis, most likely due to
inability to visualize it by earlier imaging methods.
Klippel-Feil syndrome (KFS) is one of the clear-
3.3 Somitogenesis est examples of a disc development defect in humans.
KFS is primarily characterized by congenital cervical
Somites are transient structures formed from the presomitic synostosis due to an absence of the disc and resulting
mesoderm (PSM) that define the anterior-posterior seg- in fusion of adjacent vertebrae. First identified in 1912
mented pattern of the embryo. They are essentially balls
occur via a clock and wavefront mechanism (Fig. 3.2).

by Klippel and Feil, it was described as limited head This model proposed that the tightly controlled size and
range of motion, low posterior hairline, and absence of number of somites is due to the cycling of expression for
neck (Klippel and Feil 1975; Willard and Nicholson specific genes within the PSM cells. Cycling of gene expres-
1934). KFS is currently categorized into three types as sion within individual cells provides the clock, and the
follows (Tracy et al. 2004; Kaplan et al. 2005): presence of a morphogen gradient across the anterior-poste-
Type I multiple cervical fusions rior axis of the embryo provides the wavefront. This gradi-
Type II 12 cervical fusions ent in combination with the timing of genes expressed
Type III cervical fusion combined with lumbar through the clock leads to the formation of a boundary where
fusions the somite will separate from the PSM. Commitment and dif-
After a screen of 63 affected individuals showed 6 ferentiation of cells is also modulated as they pass through
with potential deleterious mutations in PAX1, it was this boundary. Additional experimental evidence over the
proposed that defects in PAX1 function resulting in years has supported and expanded upon the clock and wave-
somite segmentation defects were the major cause of front model so that now many of the molecular details have
KFS (McGaughran et al. 2003). More recently, genetic been elucidated (Pourquie 2011; Brand-Saberi et al. 2008).
analysis of both inherited and sporadic cases of KFS
identified two missense mutations (L298P and A249E) 3.3.1.1 Clock Regulation
in the GDF6 gene and an inversion (q22.2q23.3) Two major signaling pathways are involved in regulating the
623 kb 3 of GDF6 (Clarke et al. 1995; Tassabehji clock that helps determine where and when each somite
et al. 2008). GDF6 (also known as BMP13) is a mem- will separate from the PSM. The first is the Notch pathway.
ber of the TGF- protein family that clusters with Notch1 has a critical role in the segmentation and epithelial-
GDF5 and GDF7 (BMP14, BMP12) and is structurally ization of somites (Conlon et al. 1995; Swiatek et al. 1994).
related to BMP2 and BMP4 (Rider and Mulloy 2010). Although Notch1 is expressed throughout the PSM, activated
Knockdown of GDF6 in zebra fish and Xenopus results Notch signals where segmentation will occur (Reaume et al.
in KFS-like phenotypes (Asai-Coakwell et al. 2009; 1992) and other pathway members that regulate Notch activ-
Tassabehji et al. 2008). Deletion of GDF6 in mice does ity, such as delta1(Dll1) (Bettenhausen et al. 1995), Hes1
not result in any obvious spinal phenotypes; however, (Palmeirim et al. 1997; Dequeant et al. 2006), Hes7 (Bessho
the additional deletion of GDF5 results in a KFS-like et al. 2003), and lunatic fringe (Lfng) (Evrard et al. 1998;
scoliosis phenotype suggesting some redundancy in Johnston et al. 1997; Forsberg et al. 1998; Zhang and Gridley
the function of GDF5 and GDF6 (Asai-Coakwell et al. 1998), exhibit cyclic expression patterns in the PSM. The
2009; Settle et al. 2003). It is also important to note pattern of expression has been compared to waves washing
that even with the double deletion, these mice do not up on shore. Expression starts at the posterior end of the
exhibit vertebral fusions. It has been shown that GDF6 PSM and moves in a wave to the point where the next somite
can inhibit endochondral ossification of mesenchy- will form, then expression starts back at the posterior domain.
mal stem cells in culture suggesting that GDF6 may The timing of the cycle is species dependent. Mesp2, an
normally help to specify or maintain the disc space in inhibitor of Notch signaling, also has a cyclic expression pat-
the sclerotome (Shen et al. 2009). Furthermore, it has tern in the PSM, and its expression becomes restricted to,
been shown that GDF6 prevents the effects of annular and maintained in, the anterior portion of the SI somite at
injury in an ovine model and may act as a protective each cycle (Morimoto et al. 2005). This sets up a sharp
agent for the annulus fibrosus (Wei et al. 2009). These boundary of Notch activity that is high in the posterior end of
only cover a few of the many abnormalities associ- the S0 somite and low in the anterior end of the future SI
ated with segmentation, but future genetic screening somite, thus defining the segmentation site for the S0 somite
promises to link more cases with the genes involved in (Morimoto et al. 2005; Saga 2007; Sasaki et al. 2011; Saga
somitogenesis. et al. 1997).
The second signaling pathway shown to be involved in
the clock system and in regulating its cyclic pattern of gene
expression is that of Wnt. Axin2, a negative regulator of the
3.3.1 Clock and Wavefront Wnt pathway, has an oscillating pattern of expression that is
not in synchrony with genes of the Notch family (Aulehla
The first step in the formation of the somite from the PSM is et al. 2003; Dequeant et al. 2006). Disruption of the Notch
the determination and timing of the segmentation site. This pathway does not disrupt Axin2 expression, although disrup-
anterior-posterior segmentation and patterning of the PSM tion of Wnt3a disrupts Lnfg, an inhibitor of Notch signaling
was originally hypothesized by Cooke and Zeeman (1976) to (Aulehla et al. 2003, 2008). This suggests that cyclic Wnt
Fig. 3.2 Somitogenesis RA

(segmentation). Somites (green)
a
bud off at evenly spaced time
intervals, the period of which is SII
determined by species, from the
anterior PSM. Counter gradients
SI SI SI
of RA and FGF/Wnt signaling
determine the location of the
determination front near the
S0 S0 S0
posterior boundary of somite SI.
The two main clock signaling
pathways cycle on and off S-I S-I S-I
synchronously with somite
formation yet slightly out of phase
with each other. (a) Blue stripes
are representative of Wnt pathway
gene cycling within the PSM
during the period of one somite
generation. (b) Red stripes are
representative of Notch pathway
gene cycling within the PSM
during the period of one somite
generation. The star indicates the FGF/Wnt
position of a single cell during the
period of formation for one somite
RA
b
SII
SI SI SI
S0 S0 S0
S-I S-I S-I
FGF/Wnt
signaling is occurring upstream of the cyclic Notch signaling. decreases toward the anterior PSM (Dubrulle et al. 2001).
This cycling of gene expression seems to allow the cells of More recently, FGF4 was seen expressed in a similar pattern
the PSM to respond to the wavefront signals at the appro- (Naiche et al. 2011). Experiments implanting beads soaked
priate time initiating the cellular shape and adhesion changes in Fgf8 have demonstrated the ability of Fgf8 concentration
that permit the somite to bud off from the PSM. to regulate the size of the developing somites (Sawada et al.
2001). Deletion of Fgfr1, the only FGF receptor expressed in
3.3.1.2 Wavefront the PSM, results in loss of cyclic gene expression and failure
The concept of the wavefront, a point of abrupt transition of somite segmentation (Wahl et al. 2007). The FGF8 gradi-
leading to somite formation, has evolved to include two ent has been determined to be due to the instability of the
counteracting morphogen gradients (Brand-Saberi et al. RNA transcripts that leads to a gradual decrease in Fgf8 pro-
2008; Pourquie 2011). The first identified was that of tein (Dubrulle and Pourquie 2004), thus allowing for the
fibroblast growth factor (FGF) signaling. FGF8 was found to tight temporal and spatial control of the segmentation bound-
be present at high levels in the posterior PSM and gradually ary. In addition to FGF, Wnt signaling shows a similar
gradient, as evidenced by localization of b-catenin protein, fibronectin, cytoactin, and neural cell adhesion molecule
that acts both through and parallel to FGF signaling (Aulehla (Duband et al. 1987; Crossin et al. 1986). These molecules
et al. 2008). This Wnt signaling gradient also suggests a are known to be involved in the formation of epithelia.
mechanism of cross talk between the clock and wave- N-cadherin is necessary for the production of stable somites
front as Wnt signaling can control the Notch pathway. It has with its loss resulting in irregular and loosely attached
been suggested that the gradient and cyclic expression of somites (Radice et al. 1997) as well as lack of adhesion in
Wnt acts as a bridge between the wavefront and clock (Brand- culture (Duband et al. 1987). EphA4/Ephrin signaling is also
Saberi et al. 2008; Pourquie 2011). required for proper epithelialization of the somite. Ablation
The counter gradient to FGF is a retinoic acid (RA) gradi- of EphA4/Ephrin signaling in the form of a dominant-nega-
ent that extends in an anterior to posterior direction along the tive, truncated EphA4 results in somite boundaries, but no
developing embryo and is inhibited by Fgf8 through repres- epithelial layer formation (Barrios et al. 2003). This process
sion of Raldh2, a gene required for RA synthesis (Diez del results in epithelia forming only on the outer boundary of
Corral et al. 2003). In turn, RA is then able to repress FGF each somite sphere. It is not clear how high Notch signaling
signaling (Diez del Corral et al. 2003). Cyp26, an enzyme from the segmentation clock translates into MET, but Notch
that degrades RA, is expressed in the tail bud (Sakai et al. regulates the expression of a transcription factor called Hes1
2001) providing a mechanism to keep the diffusion of RA that regulates Ephrin expression (Glazier et al. 2008).
localized. This method of gradient formation is known as a
source-sink mechanism (Aulehla and Pourqui 2010) in con-
trast to the mRNA decay used to control the FGF gradient. It 3.3.3 Segment Identity
is currently thought that the boundary of the wavefront is
determined by the transition from high FGF/Wnt and low Hox genes control the patterning of the PSM, which will ulti-
RA signaling to low FGF/Wnt and high RA signaling. mately lead to the differences in cranial to sacral vertebrae
The bilateral symmetry of the somite pairs produced dur- identity (Iimura et al. 2009; Wellik 2009). Originally discov-
ing somitogenesis is also controlled by RA signaling ered in drosophila segmentation, Hox genes are located in
(Kawakami et al. 2005; Vermot et al. 2005; Vilhais-Neto clusters that are arranged from 3 to 5 in the order of their
et al. 2010; Vermot and Pourquie 2005; Sirbu and Duester expression from anterior to posterior domains. In mammals,
2006). During normal embryonic development, the asym- there are four clusters Hox A, B, C, and D each containing up
metric expression of various genes, such as Nodal and Pitx2, to 13 Hox genes (Wellik 2009). Hox genes with the same
occurs directing the organization of the internal organs. It number across clusters are called paralogous groups (e.g.,
has been proposed that RA acts to insulate somitogenesis Hoxa1, Hoxb1, and Hoxd1). The expression domains of indi-
from surrounding signals driving the left-right asymmetry vidual Hox genes have been shown to line up with specific
seen throughout the body cavity (Brent 2005; Brend and vertebral segments (Burke et al. 1995). Although the colin-
Holley 2009). This work links the genes involved in somito- earity of the Hox genes is easily demonstrated in drosophila
genesis directly to the control of body symmetry. with deletions resulting in anterior homeotic transforma-
tions, in the mouse, Hox deletions result in both the expected
anterior homeotic transformations and posterior homeotic
3.3.2 Epithelialization transformations (Wellik 2007). The reason for this variation
seen in mice is thought to be due to the activity of paralogous
As somites bud off of the anterior end of the PSM, the outer Hox genes. Deletion of multiple paralogous Hox genes con-
cell layer undergoes a mesenchymal to epithelial transition sistently results in anterior homeotic transformations (Wellik
(MET). The two primary transcription factors involved in 2007). The term Hox code is used to describe the phenom-
MET are Pax3 and Paraxis. In vitro overexpression of Pax3 enon in which a very specific complement of Hox expression
results in epithelialization of mesenchymal cell lines (Wiggan determines the identity of a specific vertebral segment
et al. 2002), while deletion of Pax3 in the PSM results in (Kessel and Gruss 1991; Iimura et al. 2009). In the mouse,
somites that are unable to maintain epithelial integrity the anterior boundaries of Hox gene expression are set by
(Mansouri et al. 2001). Loss of Paraxis also disrupts epithe- embryonic day 12.5 (Wellik 2007). There has been some
lialization of the somite (Burgess et al. 1996) although it indication that Hox gene expression is linked to the somite
does not affect segmentation and future differentiation. The clock (Zakany et al. 2001; Kmita and Duboule 2003) allow-
first evidence for MET during somitogenesis came from a ing additional layers of specification in the patterning of the
finding in 1978 that cells from the segmented mesoderm spine. RA has also been linked to the control of Hox gene
showed a greater adhesiveness than those of the unsegmented expression during specific stages of development with ecto-
PSM (Bellairs et al. 1978). Various adhesion molecules are pic RA causing homeotic transformations in the vertebrae
expressed in the developing somite such as N-cadherin, (Kessel and Gruss 1991). Thus, there is cross talk between
the signaling pathways in the wavefront and those required 3.5 Resegmentation
for location specification. This activity emphasizes how all
of the signaling pathways that regulate development of the Resegmentation of the sclerotome was first proposed by
axial skeleton overlap and intertwine. Remak in 1855 (Bagnall et al. 1988) after observing that
in relationship to the original somites, there was half-
segment realignment of the vertebrae. The initial studies
3.4 Formation of Sclerotome followed somite development in chick embryos through
lineage tracing using quail/chick chimeras (Bagnall et al.
Differentiation begins starting with the anterior-most somite. 1988; Goldstein and Kalcheim 1992), vital dye (Bagnall
The dorsal epithelium of the somite will form the dermo- 1992), or viral transduction (Ewan and Everett 1992).
myotome, which will later differentiate into the dermis of the The consensus from these studies was that the sclerotome
back and skeletal muscle. The somitocoele and the ventral derived from one somite divides into rostral and caudal
epithelium of the somite will undergo an epithelial to mesen- halves with the rostral half of one segment joining with the
chymal transition (EMT) to generate sclerotome, which will caudal half of the segment immediately rostral to it form-
form all of the connective tissues of the axial skeleton. The ing the vertebral body. Accordingly, the intervertebral disc
division of sclerotome and dermomyotome begins when the would be derived from the sclerotome at the junction of the
somite reaches SIII (Christ and Ordahl 1995) although the two half segments (Fig. 3.3). Since those early reports, it
sclerotome segment is still plastic until a much later stage has been determined that resegmentation occurs after the
(Dockter and Ordahl 2000). Shh secreted from the notochord sclerotome differentiates from the ventral portion of the
and ventral floor plate is the primary inductive signal con- somite. Resegmentation results in a one half-segment stag-
trolling the delamination of the epithelial somite to form the ger between the sclerotome and myotome allowing for the
sclerotome (Borycki et al. 1998; Fan and Tessier-Lavigne alternating pattern between the musculature and the verte-
1994; Murtaugh et al. 1999; Dockter 2000; Chiang et al. bral body in the axial skeleton. Phenotypes indicative of
1996; Marcelle et al. 1999). Pax3 and Pax7 are expressed alterations in resegmentation include fusion of vertebrae
in the unsegmented PSM but are downregulated in the ven- where joints should be, split vertebrae and ribs, alterations
tral somite and somitocoele as the sclerotome differentiates. in migration of neural crest through the sclerotome and
The expression of Pax1 and Pax9, markers of sclerotome, disorganization of the dorsal root ganglia. Several mouse
increases (Brand-Saberi and Christ 2000). Shh has been models demonstrate these types of resegmentation defects
shown to induce Pax1, a marker of sclerotome, when present including deficiencies in RAB23 (opb mutant) (Sporle and
ectopically (Fan and Tessier-Lavigne 1994; Johnson et al. Schughart 1998), Paraxis (Johnson et al. 2001), and Tgfbr2
1994). A second permissive signal is also required for the (Baffi et al. 2006). Ablation of the neural tube after sclero-
formation of sclerotome (Stafford et al. 2011). BMP from tome formation also results in resegmentation failure sug-
the lateral plate mesoderm normally interrupts differentiation gesting a role for signals from the neural tube in directing
of sclerotome and interferes with Shh signaling, allowing resegmentation (Colbjorn Larsen et al. 2006). Much of the
the sclerotome to differentiate only from the ventral medial molecular mechanisms defining the process of resegmenta-
side of the somite. Several extracellular BMP antagonists tion remain to be elucidated.
are expressed regionally that carefully restrict BMP activity
(Stafford et al. 2011; Rider and Mulloy 2010). Noggin (Nog)
and Gremlin1 (Grem1) cooperate to antagonize BMP signal- 3.6 Sclerotome Derivatives
ing, permitting sclerotome differentiation in the presence of
Shh (Stafford et al. 2011). Deletion of Nog and Grem1 in mice Sclerotome contains multipotent progenitor cells that differ-
results in the complete failure of sclerotome differentiation. entiate into all of the connective tissue cell types of the axial
Dermomyotome was not affected in these mice. Inhibition of skeleton (Monsoro-Burq 2005). Sclerotome can differentiate
BMP alone is not sufficient to specify sclerotome or expand into cartilage that will subsequently undergo endochondral
sclerotome differentiation in vivo suggesting antagonism of ossification to form the bony vertebrae of the spine. It will
BMP is a permissive factor for sclerotome differentiation also differentiate into the annulus fibrosus of the interverte-
(Rider and Mulloy 2010). Very recently, using notochord bral disc and the tendons that link the vertebrae to the mus-
deficient Sd mice, it was shown that the floor plate alone cle. Which cell type is specified is determined by the location
is sufficient for the development of the sclerotome (Ando of cells within the sclerotome and complex interactions of
et al. 2011). Shh from the floor plate could replace Shh from growth factors (Fig. 3.4). How these factors interact with
the notochord to allow differentiation of sclerotome in the each other to specify cell lineage in the axial skeleton is just
absence of notochord. beginning to be determined.
NT
A P
M
V
A P
D L
Neural tube Somite
Dermomyotome Sclerotome
Fig. 3.3 Sclerotome resegmentation. Resegmentation begins with the sclerotome then associates with the posterior region of the sclerotome
dorsal-ventral division of the somite into the dermomyotome (brown) directly anterior to form what will later become an individual vertebral
and sclerotome (red and blue), respectively. The sclerotome then pro- body. In this way the vertebral body and muscle segment will be stag-
ceeds to divide into anterior and posterior regions that can be seen as gered by one half segment. Arrow indicates direction of cell fate
alternating loose and condensed mesenchyme. The anterior region of one determination. NT Neural Tube
a b c
Somite and somitocoel Differentiation of vertebrae, Final structure

AF, and tendon from sclerotome
Somite Vertebral body
D
D
A
A D
L R
L R A
P
P L R
V
V P
V
Neural tube Sclerotome
Notochord Dermomyotome Vertebral body
Pre-sclerotome Intervertebral disk Muscle
Pre-dermomyotome Syndetome Intervertebral disk
Somitocoel Tendon
Fig. 3.4 Derivatives of the sclerotome. (a) Sclerotome is derived from bodies and will become the annulus fibrosus (AF) of the intervertebral
the ventral medial portion of the somite. In response to Shh, cells from disc. Cells at the dorsal border of the sclerotome differentiate into ten-
the somite and somitocoele migrate toward the notochord and differ- don in response to factors from the myotome. (c) In the final structure,
entiate into sclerotome. (b) Cells from the somitocoele end up at the the disc is situated between the vertebral bodies, which are staggered by
border between the rostral and caudal halves of each segment. After one half segment from the muscle and connected to the muscle though
resegmentation, these cells lie in between the developing vertebral the tendons
3.6.1 Differentiation of the Annulus Fibrosus marked within the same somite and precise cell transplantation
techniques were not required.
The compartment of the sclerotome that will form the While the cells undergoing chondrogenesis in the devel-
annulus fibrosus of the intervertebral disc can be traced oping vertebral body adopt a round cell morphology, the
back to the somitocoele cells of the somite (Mittapalli et al. cells in the presumptive annulus are fibroblastic and organize
2005). As described above, somites are transient structures in concentric circles around the developing nucleus pulposus
that organize the segmented pattern of the embryo. They (Peacock 1951; Rufai et al. 1995; Hayes et al. 2011). This
consist of an outer epithelial ball with a central core of orientation provides the template for collagen deposition that
mesenchymal cells. This mesenchymal core is the somito- will ultimately form the radial-ply, lamellar structure typical
coele. When the sclerotome forms, Shh from the notochord of the annulus fibrosus. This cellular template is the product
causes an epithelial to mesenchymal transition in the ven- of the orientation of the cellular actin cytoskeleton linked
tral portion of the epithelial somite, these cells, along with to adherens junctions connecting adjacent cells (Hayes et al.
the mesenchymal cells of the somitocoele, migrate ven- 1999). Highly organized matrix deposition follows this cel-
trally to surround the notochord and form the sclerotome lular orientation phase. The organized deposition of the col-
(Monsoro-Burq 2005). The cells from the somitocoele end lagen matrix may be similar to that seen in other connective
up in the caudal half of the sclerotome adjacent to von tissues like the tendon: collagen fibrils self-assemble in the
Ebners fissure, the domain that marks the border between cell, then larger fibers are formed in membrane-bound com-
the two sclerotome halves (Williams 1910; Christ et al. partments between cells (Birk and Trelstad 1986); small
2000). After resegmentation, this is the area that will end leucine-rich proteoglycans (SLRPs) control collagen fiber
up in between the vertebrae as the annulus fibrosus. The formation as well as regulate the availability of growth fac-
importance of this compartment for the formation of the tors, including TGF-. Fibromodulin is strongly expressed
intervertebral disc as well as the zygapophysial joints (syn- in the annulus fibrosus relative to the vertebral cartilage and
ovial joints of the vertebrae) was determined using clas- may play an important role in its development through its
sical developmental biology techniques in chick embryos control of collagen fiber formation and growth factor bio-
(Mittapalli et al. 2005). Somitocoeles were microsurgically availability (Hayes et al. 2011; Smits and Lefebvre 2003;
removed and replaced with an inert bead. After an incuba- Sohn et al. 2010).
tion period of 6 days, about half of the operated embryos Members of the TGF- superfamily are secreted sig-
lacked intervertebral discs. In addition, adjacent articular naling molecules that regulate many aspects of cell physi-
processes were fused due to loss of the synovial joints of ology (Serra and Chang 2003; Patil et al. 2011). TGF-s
the vertebrae. Recently, it was shown that although these signal through heteromeric serine/threonine kinase recep-
cells can contribute to the annulus fibrosus, they are not tors (Wrana et al. 1994). Mice with targeted deletion of the
committed to the annulus fibrosus cell fate while in the epi- Tgfbr2 gene in the sclerotome demonstrate defects in the
thelial somite (Senthinathan et al. 2012). When somitocoele development of the axial skeleton including a reduced or
cells were marked with GFP and transplanted between the absent disc (Baffi et al. 2004, 2006). The annulus was most
neural tube and notochord, GFP-expressing cells were not affected. Expression of fibromodulin, a matrix protein that
restricted to the annulus fibrosus or vertebral body later in is highly enriched in the annulus fibrosus, was reduced or
development. Furthermore, expression of annulus fibrosus missing. The mature form of collagen IIB, indicating verte-
markers was not found in the vertebral bodies. The results bral development, was ectopically expressed in the presump-
suggest that the annulus cells are likely specified by their tive annulus fibrosus and peanut agglutinin, which normally
location within the sclerotome. only stains the presumptive vertebrae, stained the length
It is clear that the annulus fibrosus is derived from cells at of the developing spine. Furthermore, a global analysis of
the border of sclerotome halves; however, there is some dis- gene expression comparing wild-type and Tgfbr2 mutant
pute about whether the annulus arises from the rostral or cau- intervertebral discs indicated that the annulus fibrosus from
dal side of this border. Somitocoele cells normally end up on Tgfbr2 mutant mice more closely resembled wild-type ver-
the caudal side of this boundary (Mittapalli et al. 2005), and tebrae than annulus fibrosus (Sohn et al. 2010). Additional
previous fate mapping and peanut agglutinin binding studies microarray experiments also showed that genes that are
suggest the annulus fibrosus arises from the caudal domain enriched in the annulus fibrosus were stimulated in TGF-
(Bagnall and Sanders 1989; Huang et al. 1994). In contrast, -treated sclerotome (Sohn et al. 2010). A separate study
experiments using chick-quail grafting experiments and, using rat annulus cells in culture showed that TGF- could
more recently, dye-labeling studies suggest the annulus arises upregulate proteins associated with fibrocartilage including
from the rostral sclerotome (Goldstein and Kalcheim 1992; collagen I and II (Hayes and Ralphs 2011). Together the
Bruggeman et al. 2012). The advantage of the dye-labeling results suggest that TGF- could (a) allow formation of the
studies was that two discrete cell populations could be intervertebral disc by preventing differentiation of vertebral
cartilage in the disc space and/or (b) directly stimulate fibrosus of the intervertebral disc, and the loose mesenchyme
annulus differentiation. is where the vertebral cartilage will form. A sharp bound-
Klippel-Feil syndrome (KFS; OMIM#118100) is a con- ary exists between the two compartments. High concentra-
genital malformation characterized by cervical synostosis tions of BMP activity are required to generate the vertebral
due to lack of intervertebral disc resulting in fusion of adja- cartilage. Deletion of BMP receptors results in an overall
cent vertebrae. It was recently shown that KFS is associated disruption to chondrogenesis and endochondral bone forma-
with mutations in the GDF6 gene (Tassabehji et al. 2008). tion (Yoon et al. 2005). Nevertheless, BMP mRNA is synthe-
GDF6/BMP13 is a member of the BMP subgroup within the sized in the cells that will become the disc (Zakin et al. 2008,
TGF-/BMP family of secreted signaling molecules. The 2010). To generate the morphogenetic field that defines disc
mechanism of GDF6 action in development of the interverte- and vertebrae within the sclerotome, BMP activity must be
bral disc is not known, but it has been speculated to regulate relocalized and concentrated. Two BMP interacting proteins,
specification of where the disc will form within the sclero- Crossveinless-2 (Cv-2) and Chordin (Chd), are required
tome (Box 3.3). (Zakin et al. 2008, 2010). Loss of Cv-2 or Chd results in
small vertebral bodies and slightly increased intervertebral
space. Cv-2 mRNA is made in cells in the presumptive ver-
3.6.2 Differentiation of Vertebral Cartilage tebrae. The protein is attached to cells via heparin sulfate
proteoglycans and resides in the vertebral compartment.
Vertebral chondrocyte cell fate is specified early when the Chd mRNA is made in the presumptive intervertebral disc;
sclerotome first forms (Murtaugh et al. 1999). Shh secreted however, most of the protein is localized to the vertebral
from the notochord, which induces delamination of cells from compartment. Chordin binds to and inactivates BMP in the
the somite and initial differentiation of the sclerotome, can developing disc, but it also moves BMP to the developing
also prime the cells to respond to the chondrogenic actions vertebral body. Once in the vertebral compartment, Chd
of BMP. Shh induces expression of the transcription factors binds to Cv-2, the Chd is cleaved by a tolloid-like protease,
Pax1 and Pax9, which can also be considered markers of releasing BMP. Strong BMP activity in the vertebral body is
the early sclerotome (Muller et al. 1996). Shh and Pax1/9 indicated by the presence of phospho-Smad1, which is low
regulate the expression of a transcriptional repressor called to absent in the intervertebral disc compartment. The move-
Nkx3.2 (Zeng et al. 2002; Rodrigo et al. 2003). Nkx3.2 is ment and concentration of BMP activity within the sclero-
one of the earliest markers of prechondrogenic cells in the tome help to set up the compartments and boundaries that
axial skeleton, and it induces the expression of Sox9, a mas- define where the vertebrae and disc will form.
ter regulator of chondrogenesis, in a BMP-dependent manner In addition, examination of mouse embryos with a condi-
(Zeng et al. 2002; Tribioli and Lufkin 1999). Since Nkx3.2 tional deletion of Tgfbr2 in the sclerotome indicated that
is a transcriptional repressor, induction of Sox9 is likely Tgfbr2 was required to maintain the sharp boundary between
mediated by derepression, with Nkx3.2 inhibiting an as yet the developing vertebrae and annulus fibrosus (Baffi et al.
unidentified repressor of Sox9. Nkx3.2 and Smad1/Smad4 2006). Pax1/9 is important for specifying and maintaining
interact directly to recruit histone deacetylase/Sin3a to DNA, boundaries in developing tissues, and the expression domain
thus acting to inhibit gene expression (Kim and Lassar 2003). of Pax1/9 was expanded in Tgfbr2 mutant mice. As TGF-
These results demonstrated that Smads act as transcriptional can antagonize BMP activity, it is possible that loss of TGF-
repressors in the context of specific binding partners like results in inappropriate BMP signaling in the presumptive
Nkx3.2. In this way, BMP and Nkx3.2 cooperate to regulate intervertebral disc where BMP mRNA is synthesized (Candia
chondrogenesis. Once Sox9 expression and chondrogenesis et al. 1997; Li et al. 2006). TGF- may also interfere with the
are initiated in the axial skeleton, chondrocyte differentiation relocalization of BMP activity through Chd and Cv-2.
and endochondral bone formation occur in the vertebral bod-
ies in a manner similar to that seen in the limbs. When the
bony vertebra is formed, hyaline cartilage is maintained at 3.6.4 Syndetome (Tendon)
the end plate, adjacent to the intervertebral disc.
The tendons of the axial skeleton are formed from a com-
partment of the sclerotome called the syndetome (Brent
3.6.3 Boundary Between Vertebral Body et al. 2003). Since tendons link the vertebrae to the mus-
and Annulus Fibrosus cle, localization during development is critical. When the
sclerotome separates from the somite, the epithelial dermo-
After resegmentation, the sclerotome is organized into a pat- myotome is formed from the most dorsal part of the somite.
terned structure of alternating loose and dense mesenchyme. The future muscle cells then separate to form the myotome,
The dense mesenchyme represents the future annulus which expresses master regulators of muscle development,
MyoD and Myf5. The myotome is immediately dorsal and 3.9 Summary of Critical Concepts Discussed
adjacent to the sclerotome and secretes fibroblast growth in the Chapter
factors (FGF) 4 and 6. Receptors for FGF are located on
the sclerotome and FGF signaling stimulates expression The intervertebral disc is derived from two embryonic
of Scleraxis (Scx), marking developing tendon. High lev- structures, somites and notochord.
els of Shh on the ventral side of the sclerotome from the Notochordal cells form the nucleus pulposus.
notochord prevent tendon formation and promote the for- Somites are transient structures that determine the seg-
mation of the vertebral bodies; thus, the tendon forms in mented pattern of the embryo.
between and adjacent to the vertebrae and muscle (Brent Somites differentiate into sclerotome and dermomyo-
et al. 2003, 2005). tome. The sclerotome will form all of the connective tis-
sues of the spine.
The sclerotome resegments after it differentiates from the
3.7 Developmental Pathways Involved somite so that the muscle and vertebral segments are stag-
in Maintenance of Postnatal Disc gered by one half segment.
The annulus fibrosus is derived from a subcompartment
Several signaling pathways that regulate development are of the sclerotome known as the arthrotome, which can be
also involved in maintaining the intervertebral disc structure traced to the somitocoele of the somite.
in the adult. For example, Tgfb is expressed in the interver- Tendons are derived from a subcompartment of the scle-
tebral disc and end plate cartilage into maturity (Dahia et al. rotome known as the syndetome.
2009). Mice expressing a dominant-negative Tgfbr2 (DNIIR) Defects in formation or maintenance of the notochord,
show kyphoscoliosis by 3 months of age (Serra et al. 1997), segmentation of somites, formation of sclerotome, reseg-
while postnatal deletion of the Tgfbr2 in the annulus fibrosus mentation, and differentiation of annulus fibrosus result
of the intervertebral disc and the end plate cartilage results in disorders of the spine that affect intervertebral disc.
in signaling changes that suggest accelerated degeneration Understanding how the intervertebral disc develops will
(Jin et al. 2011). Together, these studies indicate that func- lead to novel strategies for developmental engineering of
tional TGF- signaling is necessary for the maintenance of this complex organ.
a healthy intervertebral disc well after the original develop-
ment has ended. Another example is Wnt signaling: inappro- Acknowledgments Spine research in Dr. Serras laboratory is sup-
priate activation of Wnt/-catenin signaling via disruption ported by a grant from the National Institutes of Health, R01AR053860.
MC is supported by NIDCR Training Grant (DART) T32-DE0176707.
of the inhibitors Axin1 and Axin2 results in scoliosis and
The authors would like to thank Zak Kosan for the help with the
fusions in lumbar vertebrae (Dao et al. 2010). Likewise, figures.
transient activation of Wnt signaling by expression of a con-
stitutively active -catenin resulted in postnatal deterioration
of the annulus fibrosus (Kondo et al. 2011). Many other path-
ways important to intervertebral disc development continue References
to be active in the postnatal state including Shh, Wnt, FGF,
and BMP (Dahia et al. 2009). Future studies will elucidate Adams DS, Keller R, Koehl MA (1990) The mechanics of notochord
molecular mechanism through which these developmentally elongation, straightening and stiffening in the embryo of Xenopus
laevis. Development 110(1):115130
important pathways maintain the postnatal disc.
Alexander PG, Tuan RS (2010) Role of environmental factors in axial
skeletal dysmorphogenesis. Birth Defects Res C Embryo Today
90(2):118132
3.8 Closing Ando T, Semba K, Suda H, Sei A, Mizuta H, Araki M, Abe K, Imai K,
Nakagata N, Araki K, Yamamura K (2011) The floor plate is
sufficient for development of the sclerotome and spine without the
Understanding of signals involved in the embryonic devel- notochord. Mech Dev 128(12):129140
opment of the axial skeleton provides insight into mecha- Asai-Coakwell M, French CR, Ye M, Garcha K, Bigot K, Perera AG,
nisms of spinal pathology. For this purpose, it is necessary to Staehling-Hampton K, Mema SC, Chanda B, Mushegian A,
Bamforth S, Doschak MR, Li G, Dobbs MB, Giampietro PF, Brooks
understand how cells with specialized functions are derived
BP, Vijayalakshmi P, Sauve Y, Abitbol M, Sundaresan P, van
from undifferentiated cells and how cells interact with each Heyningen V, Pourquie O, Underhill TM, Waskiewicz AJ, Lehmann
other and their environment to form tissues and organs. In OJ (2009) Incomplete penetrance and phenotypic variability charac-
addition, a clear understanding of how the skeletal system terize Gdf6-attributable oculo-skeletal phenotypes. Hum Mol Genet
18(6):11101121
develops directly impacts regeneration and tissue engineer-
Aszdi A, Chan D, Hunziker E, Bateman JF, Fssler R (1998) Collagen
ing strategies as well as understanding the pathogenesis of II is essential for the removal of the notochord and the formation of
diseases of the spine. intervertebral discs. J Cell Biol 143(5):13991412
Aulehla A, Pourqui O (2010) Signaling gradients during paraxial Burgess R, Rawls A, Brown D, Bradley A, Olson EN (1996)
mesoderm development. Cold Spring Harb Perspect Biol 2(2) Requirement of the paraxis gene for somite formation and musculo-
Aulehla A, Wehrle C, Brand-Saberi B, Kemler R, Gossler A, Kanzler skeletal patterning. Nature 384(6609):570573
B, Herrmann BG (2003) Wnt3a plays a major role in the segmenta- Burke AC, Nelson CE, Morgan BA, Tabin C (1995) Hox genes and
tion clock controlling somitogenesis. Dev Cell 4(3):395406 the evolution of vertebrate axial morphology. Development
Aulehla A, Wiegraebe W, Baubet V, Wahl MB, Deng C, Taketo M, 121(2):333346
Lewandoski M, Pourquie O (2008) A [beta]-catenin gradient links Candia AF, Watabe T, Hawley SH, Onichtchouk D, Zhang Y, Derynck
the clock and wavefront systems in mouse embryo segmentation. R, Niehrs C, Cho KW (1997) Cellular interpretation of multiple
Nat Cell Biol 10(2):186193 TGF-beta signals: intracellular antagonism between activin/BVg1
Baffi MO, Slattery E, Sohn P, Moses HL, Chytil A, Serra R (2004) and BMP-2/4 signaling mediated by Smads. Development 124(22):
Conditional deletion of the TGF-beta type II receptor in Col2a 44674480
expressing cells results in defects in the axial skeleton without alter- Chiang C, Litingtung Y, Lee E, Young KE, Corden JL, Westphal H,
ations in chondrocyte differentiation or embryonic development of Beachy PA (1996) Cyclopia and defective axial patterning in mice
long bones. Dev Biol 276(1):124142 lacking Sonic hedgehog gene function. Nature 383(6599):407413
Baffi MO, Moran MA, Serra R (2006) Tgfbr2 regulates the maintenance Choi K-S, Harfe BD (2011) Hedgehog signaling is required for forma-
of boundaries in the axial skeleton. Dev Biol 296(2):363374 tion of the notochord sheath and patterning of nuclei pulposi within
Bagnall KM (1992) The migration and distribution of somite cells after the intervertebral discs. Proc Natl Acad Sci 108(23):94849489
labelling with the carbocyanine dye, Dil: the relationship of this dis- Choi KS, Cohn MJ, Harfe BD (2008) Identification of nucleus pulpo-
tribution to segmentation in the vertebrate body. Anat Embryol sus precursor cells and notochordal remnants in the mouse: impli-
(Berl) 185(4):317324 cations for disk degeneration and chordoma formation. Dev Dyn
Bagnall KM, Sanders EJ (1989) The binding pattern of peanut lec- 237(12):39533958
tin associated with sclerotome migration and the formation of the Christ B, Ordahl CP (1995) Early stages of chick somite development.
vertebral axis in the chick embryo. Anat Embryol (Berl) 180(5): Anat Embryol (Berl) 191(5):381396
505513 Christ B, Scaal M (2008) Formation and differentiation of avian somite
Bagnall KM, Higgins SJ, Sanders EJ (1988) The contribution made by derivatives. Adv Exp Med Biol 638:141
a single somite to the vertebral column: experimental evidence in Christ B, Huang R, Wilting J (2000) The development of the avian ver-
support of resegmentation using the chick-quail chimaera model. tebral column. Anat Embryol (Berl) 202(3):179194
Development 103(1):6985 Christ B, Huang R, Scaal M (2004) Formation and differentiation of the
Barrios A, Poole RJ, Durbin L, Brennan C, Holder N, Wilson SW avian sclerotome. Anat Embryol (Berl) 208(5):333350
(2003) Eph/Ephrin signaling regulates the mesenchymal-to-epithe- Christ B, Huang R, Scaal M (2007) Amniote somite derivatives. Dev
lial transition of the paraxial mesoderm during somite morphogen- Dyn 236(9):23822396
esis. Curr Biol 13(18):15711582 Clarke RA, Singh S, Mckenzie H, Kearsley JH, Yip MY (1995) Familial
Bellairs R, Curtis AS, Sanders EJ (1978) Cell adhesiveness and embry- Klippel-Feil syndrome and paracentric inversion Inv(8)(Q22.2q23.3).
onic differentiation. J Embryol Exp Morphol 46:207213 Am J Hum Genet 57(6):13641370
Bessho Y, Hirata H, Masamizu Y, Kageyama R (2003) Periodic repres- Colbjorn Larsen K, Fuchtbauer EM, Brand-Saberi B (2006) The neural
sion by the bHLH factor Hes7 is an essential mechanism for the tube is required to maintain primary segmentation in the sclerotome.
somite segmentation clock. Genes Dev 17(12):14511456 Cells Tissues Organs 182(1):1221
Bettenhausen B, Hrabe de Angelis M, Simon D, Guenet JL, Gossler A Conlon RA, Reaume AG, Rossant J (1995) Notch1 is required for
(1995) Transient and restricted expression during mouse embryo- the coordinate segmentation of somites. Development 121(5):
genesis of Dll1, a murine gene closely related to Drosophila Delta. 15331545
Development 121(8):24072418 Cooke J, Zeeman EC (1976) A clock and wavefront model for control
Birk DE, Trelstad RL (1986) Extracellular compartments in tendon of the number of repeated structures during animal morphogenesis.
morphogenesis: collagen fibril, bundle, and macroaggregate forma- J Theor Biol 58(2):455476
tion. J Cell Biol 103(1):231240 Couly GF, Coltey PM, Le Douarin NM (1993) The triple origin of skull
Borycki AG, Mendham L, Emerson CP (1998) Control of somite pat- in higher vertebrates: a study in quail-chick chimeras. Development
terning by Sonic hedgehog and its downstream signal response 117(2):409429
genes. Development 125(4):777790 Crossin KL, Hoffman S, Grumet M, Thiery JP, Edelman GM (1986)
Brand-Saberi B, Christ B (2000) Evolution and development of distinct Site-restricted expression of cytotactin during development of the
cell lineages derived from somites. Curr Top Dev Biol 48:142 chicken embryo. J Cell Biol 102(5):19171930
Brand-Saberi B, Rudloff S, Gamel AJ (2008) Avian somitogen- Dahia CL, Mahoney EJ, Durrani AA, Wylie C (2009) Intercellular sig-
esis: translating time and space into pattern. Adv Exp Med Biol naling pathways active during intervertebral disc growth, differen-
638:4257 tiation, and aging. Spine (Phila Pa 1976) 34(5):456462
Brend T, Holley SA (2009) Balancing segmentation and laterality dur- Dale RM, Topczewski J (2011) Identification of an evolutionarily con-
ing vertebrate development. Semin Cell Dev Biol 20(4):472478 served regulatory element of the zebrafish col2a1a gene. Dev Biol
Brent AE (2005) Somite formation: where left meets right. Curr Biol 357(2):518531
15(12):R468R470 Dao DY, Yang X, Flick LM, Chen D, Hilton MJ, OKeefe RJ (2010)
Brent AE, Schweitzer R, Tabin CJ (2003) A somitic compartment of Axin2 regulates chondrocyte maturation and axial skeletal develop-
tendon progenitors. Cell 113(2):235248 ment. J Orthop Res 28(1):8995
Brent AE, Braun T, Tabin CJ (2005) Genetic analysis of interactions Dequeant ML, Glynn E, Gaudenz K, Wahl M, Chen J, Mushegian
between the somitic muscle, cartilage and tendon cell lineages dur- A, Pourquie O (2006) A complex oscillating network of sig-
ing mouse development. Development 132(3):515528 naling genes underlies the mouse segmentation clock. Science
Bruggeman BJ, Maier JA, Mohiuddin YS, Powers R, Lo Y, Guimaraes- 314(5805):15951598
Camboa N, Evans SM, Harfe BD (2012) Avian intervertebral Diez del Corral R, Olivera-Martinez I, Goriely A, Gale E, Maden M,
disc arises from rostral sclerotome and lacks a nucleus pulposus: Storey K (2003) Opposing FGF and retinoid pathways control ven-
implications for evolution of the vertebrate disc. Dev Dyn 241(4): tral neural pattern, neuronal differentiation, and segmentation dur-
675683 ing body axis extension. Neuron 40(1):6579
DiPaola CP, Farmer JC, Manova K, Niswander LA (2005) Molecular Huang R, Zhi Q, Brand-Saberi B, Christ B (2000) New experimental
signaling in intervertebral disk development. J Orthop Res evidence for somite resegmentation. Anat Embryol (Berl)
23(5):11121119. doi:10.1016/j.orthres.2005.03.008 202(3):195200
Dobrovolskaia-Zavadskaia N (1927) Sur la mortification spotanee de la Iimura T, Denans N, Pourqui O (2009) Chapter 7, Establishment of
chez la souris nouvau-nee et sur lexistence dun caractere (facteur) Hox vertebral identities in the embryonic spine precursors. In:
herededitaire, non-viable. CR Soc Biol 97:114116 Olivier P (ed) Current topics in developmental biology, vol 88.
Dockter JL (2000) Sclerotome induction and differentiation. Curr Top Academic, New York, pp 201234
Dev Biol 48:77127 Ingham PW, McMahon AP (2001) Hedgehog signaling in animal devel-
Dockter J, Ordahl CP (2000) Dorsoventral axis determination in the opment: paradigms and principles. Genes Dev 15(23):30593087
somite: a re-examination. Development 127(10):22012206 Inohaya K, Takano Y, Kudo A (2007) The teleost intervertebral region
Duband JL, Dufour S, Hatta K, Takeichi M, Edelman GM, Thiery JP acts as a growth center of the centrum: in vivo visualization of
(1987) Adhesion molecules during somitogenesis in the avian osteoblasts and their progenitors in transgenic fish. Dev Dyn
embryo. J Cell Biol 104(5):13611374 236(11):30313046
Dubrulle J, Pourquie O (2004) fgf8 mRNA decay establishes a gradient Jin H, Shen J, Wang B, Wang M, Shu B, Chen D (2011) TGF-beta sig-
that couples axial elongation to patterning in the vertebrate embryo. naling plays an essential role in the growth and maintenance of
Nature 427(6973):419422 intervertebral disc tissue. FEBS Lett 585(8):12091215
Dubrulle J, McGrew MJ, Pourquie O (2001) FGF signaling controls Johnson RL, Laufer E, Riddle RD, Tabin C (1994) Ectopic expression
somite boundary position and regulates segmentation clock control of Sonic hedgehog alters dorsal-ventral patterning of somites. Cell
of spatiotemporal Hox gene activation. Cell 106(2):219232 79(7):11651173
Erwin WM (2010) The enigma that is the nucleus pulposus cell: the Johnson J, Rhee J, Parsons SM, Brown D, Olson EN, Rawls A (2001)
search goes on. Arthritis Res Ther 12(3):118 The anterior/posterior polarity of somites is disrupted in paraxis-
Evrard YA, Lun Y, Aulehla A, Gan L, Johnson RL (1998) Lunatic deficient mice. Dev Biol 229(1):176187, S0012-1606(00)99969-X
fringe is an essential mediator of somite segmentation and pattern- [pii]
ing. Nature 394(6691):377381 Johnston SH, Rauskolb C, Wilson R, Prabhakaran B, Irvine KD, Vogt
Ewan KB, Everett AW (1992) Evidence for resegmentation in the for- TF (1997) A family of mammalian Fringe genes implicated in
mation of the vertebral column using the novel approach of retrovi- boundary determination and the Notch pathway. Development
ral-mediated gene transfer. Exp Cell Res 198(2):315320 124(11):22452254
Fan C-M, Tessier-Lavigne M (1994) Patterning of mammalian somites Kaplan KM, Spivak JM, Bendo JA (2005) Embryology of the spine and
by surface ectoderm and notochord: evidence for sclerotome induc- associated congenital abnormalities. Spine J 5(5):564576
tion by a hedgehog homolog. Cell 79(7):11751186 Karasugi T, Semba K, Hirose Y, Kelempisioti A, Nakajima M, Miyake
Ferrer-Vaquer A, Viotti M, Hadjantonakis A-K (2010) Transitions A, Furuichi T, Kawaguchi Y, Mikami Y, Chiba K, Kamata M, Ozaki
between epithelial and mesenchymal states and the morphogenesis K, Takahashi A, Makela P, Karppinen J, Kimura T, Kubo T, Toyama
of the early mouse embryo. Cell Adh Migr 4(3):447457 Y, Yamamura K, Mannikko M, Mizuta H, Ikegawa S (2009)
Forsberg H, Crozet F, Brown NA (1998) Waves of mouse Lunatic fringe Association of the tag SNPs in the human SKT gene (KIAA1217)
expression, in four-hour cycles at two-hour intervals, precede somite with lumbar disc herniation. J Bone Miner Res 24(9):15371543
boundary formation. Curr Biol 8(18):10271030 Kawakami Y, Raya A, Raya RM, Rodriguez-Esteban C, Belmonte JC
Gadjanski I, Spiller K, Vunjak-Novakovic G (2012) Time-dependent (2005) Retinoic acid signalling links left-right asymmetric pattern-
processes in stem cell-based tissue engineering of articular carti- ing and bilaterally symmetric somitogenesis in the zebrafish embryo.
lage. Stem Cell Rev 8(3):863881 Nature 435(7039):165171
Gilson A, Dreger M, Urban JP (2010) Differential expression level of Keller R (2000) The origin and morphogenesis of amphibian somites.
cytokeratin 8 in cells of the bovine nucleus pulposus complicates Curr Top Dev Biol 47(47):183244
the search for specific intervertebral disc cell markers. Arthritis Res Kessel M, Gruss P (1991) Homeotic transformations of murine verte-
Ther 12(1):R24 brae and concomitant alteration of Hox codes induced by retinoic
Glazier JA, Zhang Y, Swat M, Zaitlen B, Schnell S (2008) Coordinated acid. Cell 67(1):89104
action of N-CAM, N-cadherin, EphA4, and ephrinB2 translates Kim DW, Lassar AB (2003) Smad-dependent recruitment of a histone
genetic prepatterns into structure during somitogenesis in chick. deacetylase/Sin3A complex modulates the bone morphogenetic
Curr Top Dev Biol 81:205247 protein-dependent transcriptional repressor activity of Nkx3.2. Mol
Goldstein RS, Kalcheim C (1992) Determination of epithelial half- Cell Biol 23(23):87048717
somites in skeletal morphogenesis. Development 116(2):441445 Kim KW, Lim TH, Kim JG, Jeong ST, Masuda K, An HS (2003) The
Gotz W, Osmers R, Herken R (1995) Localisation of extracellular origin of chondrocytes in the nucleus pulposus and histologic
matrix components in the embryonic human notochord and axial findings associated with the transition of a notochordal nucleus pul-
mesenchyme. J Anat 186(Pt 1):111121 posus to a fibrocartilaginous nucleus pulposus in intact rabbit inter-
Haga Y, Dominique VJ, Du SJ (2009) Analyzing notochord segmenta- vertebral discs. Spine (Phila Pa 1976) 28(10):982990
tion and intervertebral disc formation using the twhh:gfp transgenic Kim JH, Deasy BM, Seo HY, Studer RK, Vo NV, Georgescu HI, Sowa
zebrafish model. Transgenic Res 18(5):669683 GA, Kang JD (2009) Differentiation of intervertebral notochordal
Hayes AJ, Ralphs JR (2011) The response of foetal annulus fibrosus cells through live automated cell imaging system in vitro. Spine
cells to growth factors: modulation of matrix synthesis by TGF- (Phila Pa 1976) 34(23):24862493
beta1 and IGF-1. Histochem Cell Biol 136(2):163175 Kispert A, Herrmann BG, Leptin M, Reuter R (1994) Homologs of the
Hayes AJ, Benjamin M, Ralphs JR (1999) Role of actin stress fibres in mouse Brachyury gene are involved in the specification of posterior
the development of the intervertebral disc: cytoskeletal control of terminal structures in Drosophila, Tribolium, and Locusta. Genes
extracellular matrix assembly. Dev Dyn 215(3):179189 Dev 8(18):21372150
Hayes AJ, Isaacs MD, Hughes C, Caterson B, Ralphs JR (2011) Klippel M, Feil A (1975) The classic: a case of absence of cervical
Collagen fibrillogenesis in the development of the annulus fibrosus vertebrae with the thoracic cage rising to the base of the cranium
of the intervertebral disc. Eur Cell Mater 22:226241 (cervical thoracic cage). Clin Orthop Relat Res 109:38
Huang R, Zhi Q, Wilting J, Christ B (1994) The fate of somitocoele Kmita M, Duboule D (2003) Organizing axes in time and space; 25
cells in avian embryos. Anat Embryol (Berl) 190(3):243250 years of colinear tinkering. Science 301(5631):331333
Kondo N, Yuasa T, Shimono K, Tung W, Okabe T, Yasuhara R, Pacifici Muller TS, Ebensperger C, Neubuser A, Koseki H, Balling R, Christ B,
M, Zhang Y, Iwamoto M, Enomoto-Iwamoto M (2011) Intervertebral Wilting J (1996) Expression of avian Pax1 and Pax9 is intrinsically
disc development is regulated by Wnt/beta-catenin signaling. Spine regulated in the pharyngeal endoderm, but depends on environmen-
(Phila Pa 1976) 36(8):E513E518 tal influences in the paraxial mesoderm. Dev Biol 178(2):403417
Koob TJ, Long JH (2000) The vertebrate body axis: evolution and Murtaugh LC, Chyung JH, Lassar AB (1999) Sonic hedgehog promotes
mechanical function. Am Zool 40(1):118 somitic chondrogenesis by altering the cellular response to BMP
Lee CR, Sakai D, Nakai T, Toyama K, Mochida J, Alini M, Grad S signaling. Genes Dev 13(2):225237
(2007) A phenotypic comparison of intervertebral disc and articular Naiche LA, Holder N, Lewandoski M (2011) FGF4 and FGF8 com-
cartilage cells in the rat. Eur Spine J 16(12):21742185 prise the wavefront activity that controls somitogenesis. Proc Natl
Lefebvre V (2002) Toward understanding the functions of the two Acad Sci 108(10):40184023
highly related Sox5 and Sox6 genes. J Bone Miner Metab Oda T, Elkahloun AG, Pike BL, Okajima K, Krantz ID, Genin A,
20(3):121130 Piccoli DA, Meltzer PS, Spinner NB, Collins FS, Chandrasekharappa
Lenas P, Moos M, Luyten FP (2009a) Developmental engineering: a SC (1997) Mutations in the human Jagged1 gene are responsible for
new paradigm for the design and manufacturing of cell-based prod- Alagille syndrome. Nat Genet 16(3):235242
ucts. Part I: from three-dimensional cell growth to biomimetics of Oldershaw RA, Baxter MA, Lowe ET, Bates N, Grady LM, Soncin F,
in vivo development. Tissue Eng Part B Rev 15(4):381394 Brison DR, Hardingham TE, Kimber SJ (2010) Directed differenti-
Lenas P, Moos M, Luyten FP (2009b) Developmental engineering: a ation of human embryonic stem cells toward chondrocytes. Nat
new paradigm for the design and manufacturing of cell-based prod- Biotechnol 28(11):11871194
ucts. Part II: from genes to networks: tissue engineering from the Paavola LG, Wilson DB, Center EM (1980) Histochemistry of the
viewpoint of systems biology and network science. Tissue Eng Part developing notochord, perichordal sheath and vertebrae in
B Rev 15(4):395422 Danforths short-tail (sd) and normal C57BL/6 mice. J Embryol
Lenas P, Luyten FP, Doblare M, Nicodemou-Lena E, Lanzara AE Exp Morphol 55:227245
(2011) Modularity in developmental biology and artificial organs: a Palmeirim I, Henrique D, Ish-Horowicz D, Pourqui O (1997) Avian
missing concept in tissue engineering. Artif Organs 35(6):656662 hairy gene expression identifies a molecular clock linked to verte-
Lettice LA, Purdie LA, Carlson GJ, Kilanowski F, Dorin J, Hill RE (1999) brate segmentation and somitogenesis. Cell 91(5):639648
The mouse bagpipe gene controls development of axial skeleton, Patil AS, Sable RB, Kothari RM (2011) An update on transform-
skull, and spleen. Proc Natl Acad Sci U S A 96(17):96959700 ing growth factor-beta (TGF-beta): sources, types, functions and
Li L, Krantz ID, Deng Y, Genin A, Banta AB, Collins CC, Qi M, Trask clinical applicability for cartilage/bone healing. J Cell Physiol
BJ, Kuo WL, Cochran J, Costa T, Pierpont ME, Rand EB, Piccoli 226(12):30943103
DA, Hood L, Spinner NB (1997) Alagille syndrome is caused by Peacock A (1951) Observations on the prenatal development of the
mutations in human Jagged1, which encodes a ligand for Notch1. intervertebral disc in man. J Anat 85(3):260274
Nat Genet 16(3):243251 Pelton RW, Dickinson ME, Moses HL, Hogan BL (1990) In situ
Li TF, Darowish M, Zuscik MJ, Chen D, Schwarz EM, Rosier RN, hybridization analysis of TGF beta 3 RNA expression during mouse
Drissi H, OKeefe RJ (2006) Smad3-deficient chondrocytes have development: comparative studies with TGF beta 1 and beta 2.
enhanced BMP signaling and accelerated differentiation. J Bone Development 110(2):609620
Miner Res 21(1):416 Peters H, Wilm B, Sakai N, Imai K, Maas R, Balling R (1999) Pax1 and
Mansouri A, Pla P, Larue L, Gruss P (2001) Pax3 acts cell autono- Pax9 synergistically regulate vertebral column development.
mously in the neural tube and somites by controlling cell surface Development 126(23):53995408
properties. Development 128(11):19952005 Pourquie O (2011) Vertebrate segmentation: from cyclic gene networks
Marcelle C, Ahlgren S, Bronner-Fraser M (1999) In vivo regulation of to scoliosis. Cell 145(5):650663
somite differentiation and proliferation by sonic hedgehog. Dev Radice GL, Rayburn H, Matsunami H, Knudsen KA, Takeichi M,
Biol 214(2):277287 Hynes RO (1997) Developmental defects in mouse embryos lacking
McCann MR, Tamplin OJ, Rossant J, Seguin CA (2012) Tracing noto- N-cadherin. Dev Biol 181(1):6478
chord-derived cells using a Noto-cre mouse: implications for inter- Reaume AG, Conlon RA, Zirngibl R, Yamaguchi TP, Rossant J (1992)
vertebral disc development. Dis Model Mech 5 Expression analysis of a Notch homologue in the mouse embryo.
McGaughran JM, Oates A, Donnai D, Read AP, Tassabehji M (2003) Dev Biol 154(2):377387
Mutations in PAX1 may be associated with Klippel-Feil syndrome. Rider CC, Mulloy B (2010) Bone morphogenetic protein and growth
Eur J Hum Genet 11(6):468474 differentiation factor cytokine families and their protein antagonists.
Minogue BM, Richardson SM, Zeef LA, Freemont AJ, Hoyland JA Biochem J 429(1):112
(2010) Characterization of the human nucleus pulposus cell pheno- Risbud MV, Schaer TP, Shapiro IM (2010) Toward an understanding of
type and evaluation of novel marker gene expression to define adult the role of notochordal cells in the adult intervertebral disc: from
stem cell differentiation. Arthritis Rheum 62(12):36953705 discord to accord. Dev Dyn 239(8):21412148
Mittapalli VR, Huang R, Patel K, Christ B, Scaal M (2005) Arthrotome: Rodrigo I, Hill RE, Balling R, Munsterberg A, Imai K (2003) Pax1 and
a specific joint forming compartment in the avian somite. Dev Dyn Pax9 activate Bapx1 to induce chondrogenic differentiation in the
234(1):4853 sclerotome. Development 130(3):473482
Monsoro-Burq AH (2005) Sclerotome development and morphogene- Rufai A, Benjamin M, Ralphs JR (1995) The development of
sis: when experimental embryology meets genetics. Int J Dev Biol fibrocartilage in the rat intervertebral disc. Anat Embryol (Berl)
49(23):301308 192(1):5362
Monsoro-Burq AH, Bontoux M, Teillet MA, Le Douarin NM (1994) Saga Y (2007) Segmental border is defined by the key transcription fac-
Heterogeneity in the development of the vertebra. Proc Natl Acad tor Mesp2, by means of the suppression of notch activity. Dev Dyn
Sci U S A 91(22):1043510439 236(6):14501455
Moore KL, Oersaud TVN (2003) Before we are born, 6th edn. Saunders, Saga Y, Hata N, Koseki H, Taketo MM (1997) Mesp2: a novel mouse
Philadelphia gene expressed in the presegmented mesoderm and essential for
Morimoto M, Takahashi Y, Endo M, Saga Y (2005) The Mesp2 tran- segmentation initiation. Genes Dev 11(14):18271839
scription factor establishes segmental borders by suppressing Notch Sakai Y, Meno C, Fujii H, Nishino J, Shiratori H, Saijoh Y, Rossant J,
activity. Nature 435(7040):354359. doi:10.1038/nature03591 Hamada H (2001) The retinoic acid-inactivating enzyme CYP26 is
essential for establishing an uneven distribution of retinoic acid L-Sox5 and Sox6 are essential for cartilage formation. Dev Cell
along the anterio-posterior axis within the mouse embryo. Genes 1(2):277290
Dev 15(2):213225 Sohn P, Cox M, Chen D, Serra R (2010) Molecular profiling of the
Sakai D, Nakai T, Mochida J, Alini M, Grad S (2009) Differential phe- developing mouse axial skeleton: a role for Tgfbr2 in the develop-
notype of intervertebral disc cells: microarray and immunohis- ment of the intervertebral disc. BMC Dev Biol 10(1):29
tochemical analysis of canine nucleus pulposus and anulus fibrosus. Sporle R, Schughart K (1998) Paradox segmentation along inter- and
Spine (Phila Pa 1976) 34(14):14481456 intrasomitic borderlines is followed by dysmorphology of the axial
Sasaki N, Kiso M, Kitagawa M, Saga Y (2011) The repression of skeleton in the open brain (opb) mouse mutant. Dev Genet
Notch signaling occurs via the destabilization of mastermind- 22(4):359373
like 1 by Mesp2 and is essential for somitogenesis. Development Stafford DA, Brunet LJ, Khokha MK, Economides AN, Harland RM
138(1):5564 (2011) Cooperative activity of noggin and gremlin 1 in axial skele-
Sawada A, Shinya M, Jiang YJ, Kawakami A, Kuroiwa A, Takeda H ton development. Development 138(5):10051014
(2001) Fgf/MAPK signalling is a crucial positional cue in somite Stemple DL (2005) Structure and function of the notochord: an essential
boundary formation. Development 128(23):48734880 organ for chordate development. Development 132(11):25032512
Scaal M, Wiegreffe C (2006) Somite compartments in anamniotes. Stott D, Kispert A, Herrmann BG (1993) Rescue of the tail defect of
Anat Embryol (Berl) 211(Suppl 1):919 Brachyury mice. Genes Dev 7(2):197203
Schweitzer R, Chyung JH, Murtaugh LC, Brent AE, Rosen V, Olson Swiatek PJ, Lindsell CE, del Amo FF, Weinmaster G, Gridley T (1994)
EN, Lassar A, Tabin CJ (2001) Analysis of the tendon cell fate using Notch1 is essential for postimplantation development in mice.
Scleraxis, a specific marker for tendons and ligaments. Development Genes Dev 8(6):707719
128(19):38553866 Swiderski RE, Solursh M (1992) Localization of type II collagen, long
Scotti C, Tonnarelli B, Papadimitropoulos A, Scherberich A, Schaeren form alpha 1(IX) collagen, and short form alpha 1(IX) collagen
S, Schauerte A, Lopez-Rios J, Zeller R, Barbero A, Martin I (2010) transcripts in the developing chick notochord and axial skeleton.
Recapitulation of endochondral bone formation using human adult Dev Dyn 194(2):118127. doi:10.1002/aja.1001940205
mesenchymal stem cells as a paradigm for developmental engineer- Tassabehji M, Fang ZM, Hilton EN, McGaughran J, Zhao Z, de Bock
ing. Proc Natl Acad Sci U S A 107(16):72517256 CE, Howard E, Malass M, Donnai D, Diwan A, Manson FDC,
Semba K, Araki K, Li Z, Matsumoto K, Suzuki M, Nakagata N, Takagi Murrell D, Clarke RA (2008) Mutations in GDF6 are associated
K, Takeya M, Yoshinobu K, Araki M, Imai K, Abe K, Yamamura K with vertebral segmentation defects in Klippel-Feil syndrome. Hum
(2006) A novel murine gene, Sickle tail, linked to the Danforths Mutat 29(8):10171027
short tail locus, is required for normal development of the interver- Theiler K (1988) Vertebral malformations. Adv Anat Embryol Cell
tebral disc. Genetics 172(1):445456 Biol 112:199
Senthinathan B, Sousa C, Tannahill D, Keynes R (2012) The generation Tracy MR, Dormans JP, Kusumi K (2004) Klippel-Feil syndrome: clin-
of vertebral segmental patterning in the chick embryo. J Anat ical features and current understanding of etiology. Clin Orthop
220(6):591602 Relat Res 424:183190
Serra R, Chang C (2003) TGF-beta signaling in human skeletal Tribioli C, Lufkin T (1999) The murine Bapx1 homeobox gene plays a
and patterning disorders. Birth Defects Res C Embryo Today critical role in embryonic development of the axial skeleton and
69(4):333351 spleen. Development 126(24):56995711
Serra R, Johnson M, Filvaroff EH, LaBorde J, Sheehan DM, Derynck R, Turnpenny PD (2008) Defective somitogenesis and abnormal vertebral
Moses HL (1997) Expression of a truncated, kinase-defective TGF-b segmentation in man. Adv Exp Med Biol 638:164189
type II receptor in mouse skeletal tissue promotes terminal chondro- Vermot J, Pourquie O (2005) Retinoic acid coordinates somito-
cyte differentiation and osteoarthritis. J Cell Biol 139:541552 genesis and left-right patterning in vertebrate embryos. Nature
Settle SH Jr, Rountree RB, Sinha A, Thacker A, Higgins K, Kingsley 435(7039):215220
DM (2003) Multiple joint and skeletal patterning defects caused by Vermot J, Gallego Llamas J, Fraulob V, Niederreither K, Chambon P,
single and double mutations in the mouse Gdf6 and Gdf5 genes. Dolle P (2005) Retinoic acid controls the bilateral symmetry of somite
Dev Biol 254(1):116130 formation in the mouse embryo. Science 308(5721):563566
Shapiro IM, Risbud MV (2010) Transcriptional profiling of the nucleus Vilhais-Neto GC, Maruhashi M, Smith KT, Vasseur-Cognet M, Peterson
pulposus: say yes to notochord. Arthritis Res Ther 12(3):117 AS, Workman JL, Pourquie O (2010) Rere controls retinoic acid sig-
Shen B, Bhargav D, Wei A, Williams LA, Tao H, Ma DD, Diwan AD nalling and somite bilateral symmetry. Nature 463(7283):953957
(2009) BMP-13 emerges as a potential inhibitor of bone formation. Vujovic S, Henderson S, Presneau N, Odell E, Jacques TS, Tirabosco
Int J Biol Sci 5(2):192200 R, Boshoff C, Flanagan AM (2006) Brachyury, a crucial regulator
Shi S, Ciurli C, Cartman A, Pidoux I, Poole AR, Zhang Y (2003) of notochordal development, is a novel biomarker for chordomas. J
Experimental immunity to the G1 domain of the proteoglycan ver- Pathol 209(2):157165
sican induces spondylitis and sacroiliitis, of a kind seen in human Wahl MB, Deng C, Lewandoski M, Pourqui O (2007) FGF signaling
spondylarthropathies. Arthritis Rheum 48(10):29032915 acts upstream of the NOTCH and WNT signaling pathways to con-
Shifley ET, Cole SE (2007) The vertebrate segmentation clock and its trol segmentation clock oscillations in mouse somitogenesis.
role in skeletal birth defects. Birth Defects Res C Embryo Today Development 134(22):40334041
81(2):121133 Wamsley R (1953) Development and growth of the intervertebral disc.
Sirbu IO, Duester G (2006) Retinoic-acid signalling in node ectoderm Edinb Med J 60:341363
and posterior neural plate directs left-right patterning of somitic Wei A, Williams LA, Bhargav D, Shen B, Kishen T, Duffy N, Diwan
mesoderm. Nat Cell Biol 8(3):271277 AD (2009) BMP13 prevents the effects of annular injury in an ovine
Smits P, Lefebvre V (2003) Sox5 and Sox6 are required for notochord model. Int J Biol Sci 5(5):388396
extracellular matrix sheath formation, notochord cell survival and Wellik DM (2007) Hox patterning of the vertebrate axial skeleton. Dev
development of the nucleus pulposus of intervertebral discs. Dyn 236(9):24542463
Development 130(6):11351148 Wellik DM (2009) Chapter 9, Hox genes and vertebrate axial pattern.
Smits P, Li P, Mandel J, Zhang Z, Deng JM, Behringer RR, de In: Olivier P (ed) Current topics in developmental biology, vol 88.
Crombrugghe B, Lefebvre V (2001) The transcription factors Academic, New York, pp 257278
Wiggan O, Fadel MP, Hamel PA (2002) Pax3 induces cell aggregation Zakany J, Kmita M, Alarcon P, de la Pompa JL, Duboule D (2001)
and regulates phenotypic mesenchymal-epithelial interconversion. J Localized and transient transcription of Hox genes suggests
Cell Sci 115(Pt 3):517529 a link between patterning and the segmentation clock. Cell
Willard DP, Nicholson JT (1934) The Klippel-Feil syndrome. Ann Surg 106(2):207217
99(4):561567 Zakin L, Metzinger CA, Chang EY, Coffinier C, De Robertis EM (2008)
Williams LW (1910) The somites of the chick. Am J Anat 11:55100 Development of the vertebral morphogenetic field in the mouse:
Wrana JL, Attisano L, Wieser R, Ventura F, Massague J (1994) interactions between Crossveinless-2 and Twisted Gastrulation. Dev
Mechanism of activation of the TGF-beta receptor. Nature Biol 323(1):618
370(6488):341347 Zakin L, Chang EY, Plouhinec JL, De Robertis EM (2010)
Yang XR, Ng D, Alcorta DA, Liebsch NJ, Sheridan E, Li S, Goldstein Crossveinless-2 is required for the relocalization of Chordin
AM, Parry DM, Kelley MJ (2009) T (brachyury) gene duplication protein within the vertebral field in mouse embryos. Dev Biol
confers major susceptibility to familial chordoma. Nat Genet 347(1):204215
41(11):11761178 Zeng L, Kempf H, Murtaugh LC, Sato ME, Lassar AB (2002) Shh
Yoon BS, Ovchinnikov DA, Yoshii I, Mishina Y, Behringer RR, Lyons establishes an Nkx3.2/Sox9 autoregulatory loop that is maintained
KM (2005) Bmpr1a and Bmpr1b have overlapping functions and by BMP signals to induce somitic chondrogenesis. Genes Dev
are essential for chondrogenesis in vivo. Proc Natl Acad Sci U S A 16(15):19902005. doi:10.1101/gad.1008002
102(14):50625067 Zhang N, Gridley T (1998) Defects in somite formation in lunatic
Ytteborg E, Torgerson J, Baeverfjord G, Takle H (2012) Chapter 14, fringe-deficient mice. Nature 394(6691):374377
The Atlantic salmon (Salmo salar) vertebra and cellular pathways to
vertebral deformities. In: Carvalho ED, David GS, Silva RJ (eds)
Health and environment in aquaculture. Intech, Rijeka
Proteoglycans of the
Intervertebral Disc 4
James Melrose and Peter Roughley

4.1 Introduction ..................................................................... 53
4.2 Glycosaminoglycan Structure and Function ................ 53 Proteoglycans are present within the extracellular matrix
(ECM) of the intervertebral disc and on the surface of its
4.3 Aggrecan .......................................................................... 55
4.3.1 Aggrecan Protein Structure and Function ......................... 55
cells. The disc possesses many matrix proteoglycans, with
4.3.2 Aggrecan Gene Organization, Expression, most also being present in hyaline cartilages. The best studied
and Mutations.................................................................... 57 of these are aggrecan and versican (members of the hyalec-
4.3.3 Aggrecan Degradation in the Degenerate Disc ................. 59 tan/lectican family) and decorin, biglycan, fibromodulin,
4.4 Versican ............................................................................ 59 lumican, PRELP, and chondroadherin (members of the small
4.4.1 Versican Protein Structure and Function ........................... 60 leucine-rich repeat protein/SLRP family). More recently, the
4.4.2 Versican Gene Organization and Mutation ....................... 60 disc has also been shown to contain perlecan and lubricin,
4.4.3 Versican Degradation in the Degenerate Disc ................... 61
which were previously thought to be characteristic of base-
4.5 Perlecan ............................................................................ 61 ment membranes and the surface of articular cartilage, respec-
4.5.1 Perlecan Protein Structure ................................................. 63
4.5.2 Perlecan Gene Organization and Mutation ....................... 64
tively. Much less is known about the disc cell-associated
4.5.3 Perlecan Degradation ........................................................ 64 proteoglycans, though it is likely that several members of the
syndecan (Tkachenko et al. 2005) and glypican (Fransson
4.6 Lubricin............................................................................ 64
4.6.1 Lubricin Protein Structure ................................................. 65 et al. 2004) families will be present, together with other unre-
4.6.2 Lubricin Gene Organization and Mutation ....................... 66 lated proteoglycans such as NG2 (Akeda et al. 2007; Stallcup
4.7 The Small Leucine-Rich Repeat 2002). While little is known concerning the specific function
Proteoglycans (SLRPS)................................................... 67 of the cell-associated proteoglycans within the disc, it has
4.7.1 Decorin .............................................................................. 68 been shown that syndecan-4 expression is increased when
4.7.2 Biglycan ............................................................................ 69 elevated levels of interleukin-1 (IL-1) or tumor necrosis fac-
4.7.3 Asporin .............................................................................. 69
4.7.4 Fibromodulin..................................................................... 69 tor-a (TNF-a) are present (Wang et al. 2011), and that it may
4.7.5 Lumican ............................................................................ 70 play a role in promoting aggrecanase-mediated proteolysis
4.7.6 PRELP............................................................................... 70 within the disc. However, due to the scarcity of information
4.7.7 Chondroadherin ................................................................. 70 on disc cell-associated proteoglycans, this chapter will focus
4.7.8 SLRP Knockout Mice and Gene Mutations ...................... 70
on the matrix proteoglycans, in particular aggrecan.
in the Chapter .................................................................. 71
References ...................................................................................... 72 4.2 Glycosaminoglycan Structure
and Function
J. Melrose (*)
Kolling Institute (B6), Level 10, Proteoglycans can be considered as specialized glycoproteins
Royal NorthShore Hospital, and are a ubiquitous component of all tissues. They are distin-
St. Leonards, New South Wales 2065, Australia
guished from other glycoproteins by the substitution of their
e-mail: james.melrose@sydney.edu.au
core protein with sulfated glycosaminoglycan (GAG) chains
P. Roughley
(Box 4.1), though they may also possess more typical O-linked
Genetics Unit, Shriners Hospital for Children,
1529 Cedar Avenue, Montreal, QC, Canada H3G 1A6 and N-linked oligosaccharides (Nilsson et al. 1982). The sul-
e-mail: proughley@shriners.mcgill.ca fated GAGs can be divided into three families chondroitin

54 J. Melrose and P. Roughley
sulfate/dermatan sulfate (CS/DS), keratan sulfate (KS), and are most commonly associated with the GAG chains, but
heparan sulfate/heparin (HS/Hep) (Jackson et al. 1991). splicing variations (Fulop et al. 1993) or the use of alterna-
While many proteoglycans possess GAGs from only one tive transcription start sites (Muragaki et al. 1990) may also
family, some proteoglycans possess GAGs from different influence the structure of the core protein of some proteogly-
families. cans. In contrast, degradative changes occurring within the
CS is a copolymer of glucuronic acid and matrix are most commonly associated with processing of the
N-acetylgalactosamine, with the latter commonly being core proteins by proteinases, although modification of GAGs
sulfated at the 4 or 6 position. DS is initially synthesized by extracellular sulfatases or glycosidases has been reported
as CS, but during processing in the Golgi, some of the (Vlodavsky et al. 1999). Irrespective of their origin, all types
glucuronic acid is epimerized to iduronic acid, which may of structural change have the potential to influence proteo-
be sulfated at the 2 position. KS is a copolymer of galac- glycan function.
tose and N-acetylglucosamine and may be sulfated at the 6 Proteoglycans have been classified by the type of GAG
position on either residue. HS is a copolymer of glucuronic chain that they possess or by their location within the tissue.
acid or iduronic acid and N-acetylglucosamine, in which In terms of location, the division is commonly between pro-
the iduronic acid may be sulfated at its 2 position and the teoglycans that reside in the ECM and those associated with
N-acetylglucosamine may be sulfated at the 3 and 6 positions. the cell. Matrix proteoglycans are often substituted with CS,
On some glucosamine residues, N-sulfation may replace the DS, or KS, whereas cell-associated proteoglycans are often
N-acetyl group. In heparin, the presence of iduronic acid and linked with HS. Heparin is usually only defined in relation-
O- and N-sulfation is high. As there is no template for GAG ship to the serglycin proteoglycan present within mast cells
synthesis, GAG chain length, degree and position of sulfa- (Humphries and Stevens 1992) but can be structurally simi-
tion, and degree of epimerization can vary enormously, both lar to regions of highly sulfated and epimerized HS present
between different proteoglycans and on the same proteogly- on other proteoglycans (Girardin et al. 2005). CS and HS
can at different sites. may replace one another at the same site on some proteogly-
GAG chains have traditionally been considered structural cans, as they share the same amino acid substitution motif
electrorepulsive entities of connective tissues. This is due to (Ser-Gly) and linkage oligosaccharide (Xyl-Gal-Gal-GlcA).
their repeating charged disaccharide and sulfated sugar In contrast, KS has two different substitution motifs and
motifs or as agents which provide a high fixed charge density linkage oligosaccharides, which it shares with O-linked and
in the tissues. As such due to GAG-associated counterions N-linked oligosaccharides.
and the Donnan equilibrium effect, they are responsible for
the water-regain properties of tissues. With the emerging
concept of the sugar code, with the realization that dynamic Box 4.1: A Historical Perspective of Glycosaminoglycans
structural changes in HS produce a characteristic (nonran- and Proteoglycans
dom) heparanome, these charged sugars may also be involved The existence of glycosaminoglycans has been known
in information storage and transfer (Cummings 2009). The since the 1860s when chondroitin sulfate was first
biological importance of chondroitin sulfation during mam- described in cartilage. Discovery of the other gly-
malian development and growth factor signaling is poorly cosaminoglycans did not occur until the twentieth cen-
understood (Caterson 2012; Caterson et al. 1990), although tury, with many owing their discovery to the work of
chondroitin 4-O-sulfation is required for proper CS localiza- Karl Meyer. Meyer described the existence of
tion and modulation of distinct signaling pathways during hyaluronic acid in the vitreous humor of the eye in
growth plate morphogenesis (Kluppel et al. 2005). On this 1934, dermatan sulfate in skin in 1941, and keratan
basis, chondroitin sulfation has emerging biological roles in sulfate in the cornea in 1953. Heparin was first
mammalian development. The elucidation of the intimate described in 1916 because of its anticoagulant activity,
interplay of GAG chains with a variety of specific bioactive and the structurally related heparan sulfate in 1948.
binding partners triggering cell signaling, cell proliferation, However, modern terminology was not in use by the
matrix production, and differentiation underscores the range 1950s when the glycosaminoglycans were referred to
of functionalities which may all be affected (Turnbull 2010). as mucopolysaccharides which formed the ground
Accordingly, GAG chains are versatile tools for information substance of connective tissues. This name persists
storage and transfer and represent a new paradigm in devel- today with the mucopolysaccharidoses, a group of
opmental biology. heritable disorders due to gene defects in specific lyso-
Due to differences in either synthesis or degradation, each somal glycosyltransferases, glycosidases, and sul-
proteoglycan does not possess a unique structure: both the fatases responsible for glycosaminoglycan assembly
core protein and the GAG chains may vary with site, age, and catabolism. Similarly, the original terminology for
and pathology. Synthetic changes occurring within the cell
4 Proteoglycans of the Intervertebral Disc 55
properties of aggrecan are related to its abundance, degree of

some of the mucopolysaccharides was also different, sulfation, and ability to form proteoglycan aggregates. The
with dermatan sulfate being referred to as chondroitin swelling is driven principally by the sulfate groups on the
sulfate B and heparan sulfate as heparitin sulfate or GAGs, which attract water into the disc by osmosis. As more
heparin monosulfate. Once the structure of all the water enters the disc, the osmotic properties of the aggrecan
mucopolysaccharides was established, and it was decrease, and an equilibrium is attained in which the swell-
appreciated that they were all copolymers of a sugar ing of the disc is counterbalanced by the tension induced in
and an amino sugar, the term glycosaminoglycan came the collagenous framework of the tissue. On subjecting the
into use. In addition, until the 1950s it was not appreci- disc to compressive load, water is displaced, effectively
ated that the sulfated glycosaminoglycans were increasing the aggrecan concentration and its swelling poten-
attached to protein, and considerable effort was made tial. On removal of the load, the increased swelling potential
to purify them from this contamination. It was in is dissipated by re-imbibition of water into the disc and res-
1958 that Helen Muir proved that chondroitin sulfate toration of the equilibrium state. In addition to the symmet-
in cartilage was covalently attached to protein via a ric loading due to gravity, the disc experiences asymmetric
serine residue, and the proteoglycan era was born. compressive loading upon bending. Under asymmetric load-
However, initially the term protein polysaccharide was ing it is essential that the aggrecan cannot diffuse from the
used. In 1966, Lennart Roden described the structure site of compression if optimal restoration of disc height is to
of the trisaccharide which links the chondroitin sulfate occur following straightening. The diffusion of aggrecan is
with its serine core protein attachment point. This related to its size and is minimized by the formation of large
attachment region is now known to be common to all proteoglycan aggregates. This topic is discussed further in
the sulfated glycosaminoglycans, with the exception of Chap. 5.
keratan sulfate, irrespective of the proteoglycan core to Aggrecan is located throughout the disc, though its abun-
which the glycosaminoglycan is attached. dance at different sites varies greatly with age. In the fetal
human spine, aggrecan is prominently immunolocalized in
the cartilaginous vertebral rudiment cartilages and in the
developing intervertebral disc space (Fig. 4.1) (Smith et al.
4.3 Aggrecan 2009). The fetal vertebral rudiment cartilages are transient
developmental scaffolds which are transformed into ossified
Aggrecan is a KS/CS proteoglycan that was originally iso- structures in the adult spine, while the discs remain as per-
lated from hyaline cartilage and the gene was cloned from manent cartilaginous entities. In the human, aggrecan pre-
chondrosarcoma cells (Doege et al. 1987). It was later shown dominates in the nucleus pulposus in the young juvenile. The
to be present in the intervertebral disc and to be synthesized aggrecan content of the nucleus pulposus increases during
by disc cells. On a weight basis, aggrecan is the most abun- juvenile growth and reaches a maximum in the late adoles-
dant proteoglycan in both the disc and cartilage and has cent/early adult period. During adult life the aggrecan con-
probably been more extensively studied than any other pro- tent of the nucleus pulposus declines (Fig. 4.2a). Aging is
teoglycan. Aggrecan belongs to the family of hyaluronan also associated with an increase in aggrecan abundance in
(HA)-binding proteoglycans, together with versican, neuro- the annulus fibrosus, particularly the inner annulus, and in
can, and brevican (Margolis and Margolis 1994). All family the mature adult the aggrecan content of the annulus fibrosus
members possess an amino terminal globular domain respon- may surpass that of the nucleus pulposus. Thus, in the mature
sible for interaction with HA and a carboxy terminal globu- adult, the annulus fibrosus is as important as the nucleus pul-
lar domain containing a lectin homology domain. These posus for resisting compression.
common features give rise to the alternative family names of
hyalectins and lecticans. The interaction with HA permits
the formation of proteoglycan aggregates (Morgelin et al. 4.3.1 Aggrecan Protein Structure and Function
1988), and it was this ability to form proteoglycan aggre-
gates that led to the name aggrecan. The aggrecan core protein possesses about 2,300 amino
Aggrecan provides the disc with its ability to resist com- acids, which form three disulfide-bonded globular regions
pressive loading on the spine, causing the disc to swell and with two intervening extended regions (Fig. 4.3a) (Sandy
keep the vertebrae apart. The acquisition of a biped vertical et al. 1990; Watanabe et al. 1998). The amino terminal glob-
posture resulted in loading of the spine due to gravity. Partial ular region (G1) responsible for interaction with HA pos-
removal of this load at night and the imbibition of tissue fluid sesses three disulfide-bonded loops. The first loop allows
result in swelling of the disc, which accounts for the diurnal interaction with a link protein (LP) that stabilizes the proteo-
variation in disc height (Botsford et al. 1994). The swelling glycan aggregate (Neame and Barry 1993), and this is
Aggrecan Versican
Aggrecan a OAF d OAF g
500 m
Versican b IAF e IAF h
Toluidine blue c NP f NP i
Fig. 4.1 Immunolocalization of aggrecan (a), versican (b), and toluid- (OAF), inner annulus fibrosus (IAF), and nucleus pulposus (NP) are
ine blue-stained proteoglycan (c) in a 14-week gestational age fetal also presented in selected areas of the aggrecan (df) and versican (gi)
human intervertebral disc and adjacent cartilaginous vertebral body immunolocalizations
rudiment cartilages. Higher-power images of the outer annulus fibrosus
followed by a pair of loops responsible for the interaction resisting disc compression. While the CS chains are confined
with HA (Watanabe et al. 1997). The second globular region to the CS1 and CS2 domains, KS chains may also be present
(G2) possesses two disulfide-bonded loops that share struc- on the G1, IGD, and G2 regions (Barry et al. 1995). The CS2
tural homology with the HA-binding loops of the G1 region. domain is followed by the carboxy terminal globular domain
However, they do not facilitate interaction with HA (Fosang (G3), which possesses disulfide-bonded loops having homol-
and Hardingham 1989), and their function is presently ogy to epidermal growth factor (EGF), C-type lectin, and
unclear. The G1 and G2 regions are separated by a short complement regulatory protein (CRP) sequences. The G3
interglobular domain (IGD). After the G2 region, there is a region facilitates transit of the aggrecan through the cell dur-
long extended region to which the majority of GAG chains ing synthesis (Zheng et al. 1998) and via its lectin domain
are attached. There may be over 100 GAG chains attached to also facilitates the interaction with other components of the
each core protein, accounting for about 90 % of the molecu- extracellular matrix, such as fibulins and tenascins (Day
lar weight of aggrecan. et al. 2004). It is not clear if these G3 region interactions are
The GAG-attachment region may be divided into three of functional significance in vivo, but they could potentially
domains. The domain closest to the G2 region is responsible link proteoglycan aggregates to one another.
for the attachment of KS (KS domain), and this is followed Both the abundance and structure of aggrecan change
by two domains responsible for the attachment of CS (CS1 with age, due to variations in intracellular synthesis and
and CS2 domains). The KS and CS chains provide the aggre- extracellular degradation. Apart from possible variations in
can with osmotic properties essential for its function in gene expression, the synthesis changes are confined to
a one fragment possessing a G1 region that remains bound to

100 HA (aggregated) and one fragment that is no longer bound to
NP HA (non-aggregated) and is free to diffuse within the disc. In
articular cartilage the latter fragments are rapidly lost into
Aggrecan content mg/g tissue
AF
80
the synovial fluid, but in the disc they accumulate as their
diffusion is impeded by the vertebral end plates and the outer
60 fibrous layers of the AF. With increasing age, the abundance
of the non-aggregated fragments may exceed that of the
40 aggregated fragments (Fig. 4.2b), and ongoing proteolysis
further decreases the size of both the aggregated and non-
aggregated fragments. Ultimately, the aggregated fragments
20
are cleaved to their G1 region, which appears to be relatively
resistant to proteolysis. As the size of the non-aggregated
0 fragments declines, they are eventually lost from the tissue,
Age, years 24 50 87 and the total aggrecan content declines. The G1 fragments
may also be eventually lost from the tissue as the size of the
b Aggregated aggregates decreases due to depolymerization of the HA by
50 Large non-aggregated extracellular hyaluronidases (Durigova et al. 2011b) or free
radicals (Roberts et al. 1987). The average half-life of both
Small non-aggregated
the aggregated and non-aggregated aggrecan fragments
Aggrecan aggregation %
40
within the disc is about 20 years (Sivan et al. 2006).
Aggrecans from different species do not possess identi-
30 cal structures, and there are differences in the structure of
both their core proteins and GAG chains. The major core
20 protein differences relate to the number of repeats in both
the KS and CS1 domains (Barry et al. 1994; Doege et al.
1997). Of particular note is the absence of an extended KS
10
domain in the mouse and rat, though it is unclear whether
this is of functional importance. The largest species differ-
0 ences are in the GAG chains, which can differ enormously
NP AF in chain length and degree and position of sulfation. In
24 years addition, there is variation in the abundance of aggrecan
Fig. 4.2 Variation in aggrecan content (a) and aggregation (b) in the within the discs of different species. While changes in
human intervertebral disc. Aggrecan content declines with age through- aggrecan structure and abundance are likely to have func-
out the disc but at a faster rate in the nucleus pulposus than the annulus tional consequences, it is unclear whether such changes
fibrosus. Aggrecan aggregation is lower and the proportion of small render some species more susceptible to disc or cartilage
aggrecan fragments is greater in the nucleus pulposus (NP) than the
annulus fibrosus (AF) degeneration.
4.3.2 Aggrecan Gene Organization,

posttranslational modification of the core protein, particu- Expression, and Mutations
larly the synthesis of KS and CS (Brown et al. 1998; Roughley
and White 1980). With age, the chain length of KS increases, The human aggrecan gene (ACAN, AGC1, CSPG1) resides
while that of CS decreases. This could be viewed as a com- on chromosome 15 (Korenberg et al. 1993) and is composed
pensation mechanism for maintaining the sulfation of aggre- of 19 exons (Valhmu et al. 1995). Exon 1 encodes the
can and its swelling properties. The sulfation position of CS 5-untranslated region (UTR), exon 2 encodes the signal pep-
also changes with age, with the level of 4-sulfation decreas- tide, exons 36 encode the G1 region, exon 7 encodes the
ing and 6-sulfation increasing. It is not clear whether this IGD, exons 810 encode the G2 region, exons 11 and 12
variation in sulfation position has any functional significance. encode the GAG-attachment region, exons 1318 encode the
Currently, there is no evidence for the extracellular degrada- G3 region, and exon 19 encodes the 3-UTR. The G3 region
tion of CS or KS, and degradative changes in aggrecan are does not possess a unique structure, as the exons encoding its
confined to the proteolytic cleavage of its core protein two EGF-like sequences and its one CRP-like sequence may
(Roughley et al. 2006). Each proteolytic cleavage generates undergo alternative splicing (Doege et al. 1991; Fulop et al.
Fig. 4.3 Schematic depiction G1 G2 G3

of the structural organization a
of aggrecan (a) and versican
(b) drawn to scale
KS CS1 CS2
G3
G1
b
GAG- GAG-
Scale
Key: 50 kDa
G1, G2, G3 Globular domain
CS side chain KS side chain Core protein Disulphide stabilised loop
1993). All alternatively spliced forms of aggrecan do, however, nucleus pulposus cell gene expression, permitting the disc
possess a G3 region with a lectin-like sequence, and hence cells to function normally under low oxygen tension (Agrawal
may participate in ECM interactions. It is not clear whether et al. 2007). In addition, both TonEBP and HIF-1a regulate
absence of the EGF and CRP domains influences the function the expression of the glucuronic acid transferase responsible
of the G3 domain in vivo, but this has been suggested (Day for CS synthesis (Gogate et al. 2011; Hiyama et al. 2009).
et al. 2004). Exon 12, which encodes the CS1 and CS2 Thus, both the osmotic and hypoxic environment of the disc
domains, exhibits a unique length polymorphism within the cells participate in maintaining normal aggrecan synthesis
sequence encoding the CS1 domain in the human (Doege and structure.
et al. 1997). The human CS1 domain is composed of repeats Mutations in the aggrecan gene and genes involved in
of 19 amino acids, each of which bears consensus sequences GAG sulfation give rise to a variety of chondrodysplasias,
for the attachment of two CS chains. The number of repeats which affect not only the hyaline cartilages but also the
has been reported to vary between 13 and 33, with most indi- intervertebral disc. In humans some forms of spondyloepi-
viduals possessing 2628 repeats. This type of polymorphism physeal dysplasia (SED) and spondyloepimetaphyseal dys-
can influence the number of CS chains present on each aggre- plasia (SEMD) are associated with mutations in the aggrecan
can molecule, and it has therefore been suggested that this gene (Gleghorn et al. 2005; Tompson et al. 2009). A non-
may influence aggrecan function; it is predicted that those sense mutation is responsible for nanomelia in chickens (Li
aggrecan molecules bearing less repeats are functionally infe- et al. 1993), and a 7bp deletion in exon 5 causing a frame-
rior (Roughley 2006). This led to the prediction that individu- shift and a premature stop in exon 6 is responsible for carti-
als possessing aggrecan with a low number of CS1 repeats lage matrix deficiency (cmd) in mice (Watanabe et al. 1994).
would be more susceptible to degeneration of both the inter- Aggrecan is deficient in the extracellular matrix of the
vertebral disc and articular cartilage. While some evidence mutant tissues, probably due to a combination of nonsense-
does support this conclusion (Kawaguchi et al. 1999), it is mediated decay of the message and impaired secretion and
likely that other predisposing factors must also be present. intracellular degradation of any truncated product. Mutations
Aggrecan gene expression is regulated by a number of in the DSDST sulfate transporter gene are responsible for
factors that relate to the unique environment within the disc. diastrophic dysplasia, atelosteogenesis type II, and achon-
TonEBP, an osmoregulatory protein present in nucleus pul- drogenesis type 1B in humans (Karniski 2001; Superti-Furga
posus cells, interacts with two conserved TonE motifs within et al. 1996), and a mutation in the APS kinase, responsible
the aggrecan gene promoter and promotes aggrecan synthe- for sulfate donor (PAPS) synthesis in cartilage, gives rise to
sis, thereby allowing the nucleus pulposus cells to adapt to brachymorphism in mice (Kurima et al. 1998). Chondrocytes
their hyperosmotic environment (Tsai et al. 2006). HIF-1a and disc cells require large amounts of sulfate for aggrecan
also enhances aggrecan promoter activity and increases synthesis, and when absent, an undersulfated product is
formed. These disorders add credence to the view that disc

function requires both a high tissue content of aggrecan and Box 4.2: The Therapeutic Use of Chondroitin Sulfate
a high degree of sulfation. and Glucosamine
For the past two decades, nutraceutical companies have
been promoting oral supplements of glucosamine and
4.3.3 Aggrecan Degradation chondroitin sulfate (CS) for the treatment of osteoarthri-
in the Degenerate Disc tis. The original theory behind this treatment was that
CS was a building block for aggrecan and that glu-
The interglobular domain of aggrecan is particularly suscep- cosamine was a building block for CS and that their
tible to proteolysis and is cleaved by most proteinases in vitro supplementation would therefore promote aggrecan
(Fosang et al. 1992). Analysis of in vivo degradation prod- production or their presence in the bloodstream would
ucts indicates two predominant naturally occurring cleavage somehow tolerize the body to these components, thus
sites, which could be cleaved by aggrecanases (ADAMTS4 preventing autoantibody production which is prevalent
and ADAMTS5) and matrix metalloproteinases (MMPs) in some forms of immune-driven inflammatory arthri-
(Sztrolovics et al. 1997). Both aggrecanases and several tis. As loss of aggrecan is associated with a deteriora-
MMPs have been detected in the disc (Roberts et al. 2000), tion in the functional properties of articular cartilage in
and it is currently unclear as to which family members are osteoarthritis, it seemed reasonable that an increase in
predominantly responsible for causing aggrecan damage aggrecan production would be beneficial. Indeed it may
in vivo. However, in vitro studies indicate that ADAMTS5 is be, but the question is whether CS and glucosamine aid
more efficient than ADAMTS4 at cleaving within the aggre- in this process. Aggrecan synthesis does not involve the
can IGD (Gendron et al. 2007) and that MMP-3, MMP-7, addition of intact CS chains to its protein core, and CS
and MMP-12 are the most efficient MMPs (Durigova et al. synthesis does not utilize glucosamine to produce its
2011a). Aggrecanases are also able to cleave within the CS2 N-acetyl galactosamine component. Commercial
region of aggrecan, and five cleavage sites within this region sources of glucosamine are derived from the chitin
have been identified (Tortorella et al. 2002). MMPs may also component (poly-N-acetyl glucosamine) of crustacean
cleave at sites outside the IGD, but at present the extent of shells, and while it may make more sense to administer
their cleavage is not fully understood. One of the initial galactosamine as a therapeutic agent, there is no readily
events in aggrecan degradation, following its secretion into available commercial source of this material.
the extracellular matrix, is removal of the G3 region (Flannery Furthermore, it is likely that much of the CS entering
et al. 1992), and it is unclear whether aggrecanases or MMPs the circulation would be degraded to its constituent
are responsible. monosaccharides by the liver. It is therefore not surpris-
Proteolytic cleavage of aggrecan and its loss are detri- ing that there is much skepticism concerning the ability
mental to disc function and are thought to be directly of CS and glucosamine to promote cartilage repair, par-
involved with intervertebral disc degeneration (Roughley ticularly given the high doses of these components
2004). Not only does degradation and loss of aggrecan lessen required to elicit a positive response. There is however
the ability of the disc to swell, it may also predispose it to some evidence that CS and glucosamine therapy can aid
mechanical damage. Indeed, there may be a vicious circle in in the relief of joint pain in arthritic patients, although
which overloading of the disc stimulates aggrecan degrada- the mechanism for this effect is not clear. If this is true,
tion via the production of proteinases by the disc cells, which then CS and glucosamine therapy may be an attractive
in turn renders the disc susceptible to material damage. Such alternative to more conventional nonsteroidal anti-
material damage may be irreversible and distinguish disc inflammatory drug (NSAID) therapy, as the former have
degeneration from normal aging (Adams and Roughley few side effects. It does however remain to be shown
2006). Loss of aggrecan can also promote angiogenesis whether all formulations of CS plus glucosamine are
(Johnson et al. 2005) and may be a prelude to blood vessel equally effective and whether CS plus glucosamine for-
and nerve invasion of the disc with the onset of discogenic mulations are more effective than glucosamine alone.
pain (Johnson et al. 2002). Structural changes in aggrecan
are also associated with the discs present in the scoliotic
spine. This may also be a consequence of abnormal loading,
but this time in an asymmetric manner. Indeed, aggrecan 4.4 Versican
changes do vary between the concave and convex sides of
the scoliotic disc. It has been suggested that dietary Versican was originally identified in fibroblasts and recog-
supplementation with CS and glucosamine may help prevent nized to encode a CS proteoglycan (Zimmermann and
aggrecan loss in articular cartilage (Box 4.2), and if true this Ruoslahti 1989). Versican has a much wider tissue distribu-
may also be relevant to the disc. tion than aggrecan and together with HA has been suggested
to provide tissue hydration and viscoelasticity (Hasegawa proteolytic modification (Sztrolovics et al. 2002), which has
et al. 2007). In the fetal intervertebral disc, versican is hampered its purification from aggrecan. As a result, there is
prominently expressed throughout the tissue, but not in the little information on the structure of its CS chains and
adjacent cartilaginous vertebral body rudiments, and it whether they may change in structure with age in the disc.
prominently demarcates the margins of the developing disc Although versican is commonly referred to as a CS proteo-
from adjacent structures (Fig. 4.1) (Smith et al. 2009). In glycan, the presence of DS cannot be discounted at all ages.
the mature intervertebral disc, versican is present through- The spectrum of versican core protein sizes within the disc is
out the tissues, being diffusely distributed within the nucleus of a similar range to those of aggrecan, varying from free G1
pulposus, and most prominent between the lamellae of the regions to intact molecules (Sztrolovics et al. 2002).
annulus fibrosus (Melrose et al. 2001). Although versican is
less abundant than aggrecan in the disc, its abundance is
greater than in articular cartilage (Sztrolovics et al. 2002). 4.4.2 Versican Gene Organization and Mutation
It is unclear whether versican serves a unique function
throughout the disc, but it could provide viscoelastic prop- The human versican gene (VCAN, CSPG2) resides on
erties to the outer annulus fibrosus where aggrecan is chromosome 5 (Iozzo et al. 1992) and is composed of 15
depleted. However, the function of versican may not be exons (Naso et al. 1994). Exon 1 encodes the 5-UTR, exon
restricted to a structural role within the extracellular matrix, 2 encodes the signal peptide, exons 36 encode the G1
as it is also known to influence cell function, particularly in region, exons 7 and 8 encode the GAG-attachment region,
cancer (Ricciardelli et al. 2009). The versican G3 region exons 914 encode the G3 region, and exon 15 encodes the
has also been shown to influence disc cell function (Yang 3-UTR. The region encoded by exon 7 is referred to as
et al. 2003). GAGa and that encoded by exon 8 is referred to as GAGb.
The exons encoding the GAG-attachment region may
undergo alternative splicing (Dours-Zimmermann and
4.4.1 Versican Protein Structure and Function Zimmermann 1994), giving rise to four versican mRNAs.
The presence of both the GAGa and GAGb regions gives
Versican is structurally related to aggrecan, possessing ter- rise to the V0 form of versican, the presence of only the
minal domains analogous to the G1 and G3 regions of GAGb region gives rise to the V1 form, the presence of only
aggrecan (Fig. 4.3b), although there is no evidence for alter- the GAGa region gives rise to the V2 form, and the absence
native splicing in the versican G3 region (Grover and of both the GAGa and GAGb regions gives rise to the V3
Roughley 1993). There is also no analogous IGD or G2 form. The V1 form of versican appears to be the most com-
region, and the central GAG-attachment region is common form in most tissues, including the disc (Sztrolovics
pletely different in its amino acid sequence and GAG orga- et al. 2002). Whether the different forms of versican serve
nization. Likewise, a domain for the attachment of KS does unique functions is unknown, but it is likely that the V3
not exist, although KS may be present on the G1 region, and form, being devoid of CS, may differ in function from the
there are many fewer CS chains. The G1 region of versican other forms. The V0 form of versican is analogous to PG-M,
has functional HA-binding and LP-binding domains, and is which was identified in chick limb bud mesenchyme (Ito
able to form proteoglycan aggregates, but it is uncertain et al. 1995).
whether versican and aggrecan can reside on the same Mutations in the human versican gene give rise to the
aggregate. Interestingly, the four hyalectan genes reside in dominantly inherited Wagner syndrome, and in the original
tandem with an LP gene (Spicer et al. 2003), suggesting that index case, this was due to a base substitution at the exon 8/
coordinated gene expression may occur. Somewhat surpris- intron 8 splice junction affecting the splicing of exon 8
ingly, versican interacts best with the LP adjacent to the (Kloeckener-Gruissem et al. 2006). Additional mutations
aggrecan gene, whereas aggrecan interacts best with the LP affecting the intron7/exon 8 splice junction have also been
adjacent to the versican gene (Shi et al. 2004). As with reported in other families with Wagner syndrome
aggrecan, the lectin domain within the G3 region of versi- (Mukhopadhyay et al. 2006). Defective splicing of exon 8
can has the ability to interact with fibulins and tenascins results in a decrease in the V1 form of versican and an
(Olin et al. 2001). The presence of multiple protein-binding increase in the V2 and V3 forms. Wagner syndrome is
motifs and the resulting possibility of versatility in function classified as a vitreoretinopathy, and its ocular features are
led to the name versican. probably associated with perturbation in the role played by
The versican core protein possesses different splice vari- versican in gelling of the human vitreous. However, in accord
ants, which alter the size of its GAG-attachment region. The with the widespread tissue distribution of versican, Wagner
core protein of the common V1 form of versican is of a simi- syndrome patients also exhibit extraocular features, includ-
lar length to that of aggrecan, whereas that of the V0 form of ing skeletal defects similar to those reported in Stickler syn-
versican is much larger than aggrecan, having a core protein drome. Accordingly, it is likely that perturbation in
with about 3,400 amino acids. Versican undergoes extensive intervertebral disc formation and function will also occur.
4.4.3 Versican Degradation rounding the ossification center of the developing vertebral
in the Degenerate Disc body and in terminally differentiated growth plate chondro-
cytes (Smith et al. 2009). In the neonatal and adult disc, per-
The V1 form of versican can be cleaved by ADAMTS1 and lecan is cell associated in the pericellular matrix of the
ADAMTS4 to yield a product of 441 amino acids that includes nucleus pulposus, inner and outer annulus fibrosus, cartilagi-
the G1 region (Sandy et al. 2001). Thus, aggrecanases could nous end plates, and vertebral growth plates (Fig. 4.5).
potentially play a role in versican degradation within the disc However, its relative abundance diminishes with the decline
in vivo. However, the size of this product appears to be larger in cell number evident in the aging intervertebral disc.
than that of the free G1 region present in vivo, suggesting that Perlecan interacts with a number of growth factors and
other proteinases are also active. The MMPs would be the morphogens, including FGF-1, FGF-2, FGF-7, FGF-9, and
most likely candidates to fulfill this role. As with aggrecan, it FGF-18; platelet-derived growth factor (PDGF); vascular
is likely that premature or excessive proteolytic degradation endothelial cell growth factor (VEGF); hepatocyte growth
of versican is associated with intervertebral disc degenera- factor; BMP-1, BMP-2, BMP-4, and BMP-7; hedgehog
tion. Peptide sequences within both the versican and aggre- (Hh); and Wnt, and through these molecules influences cell
can G1 regions have also been associated with the development proliferation and differentiation and matrix production
of autoimmune spondyloarthropathies (Shi et al. 2003). (Whitelock et al. 2008). Perlecan also interacts with a num-
ber of cell attachment proteins, including laminin, fibronectin,
thrombospondin, and a5b1 and a2b1 integrin, and thereby
4.5 Perlecan plays an important role in cell attachment and recruitment
during tissue development and remodeling. Through its abil-
Perlecan was named for its appearance when first visualized ity to interact with a number of matrix components, includ-
by rotary shadowing electron microscopy, where it appeared ing PRELP; WARP; types IV, VI, XIII, and XVIII collagen;
as multiple globular domains considered to resemble a string fibrillin-1 and fibrillin-2; nidogen-1 and nidogen-2; latent
of pearls on a chain. Perlecan has a widespread distribution transforming growth factor-beta binding protein-2 (LTBP-
throughout the developing human fetal intervertebral disc 2); and tropoelastin, perlecan modulates extracellular matrix
and vertebral cartilaginous rudiment cartilages, where it disassembly and stabilization (Hayes et al. 2011c; Iozzo 1994,
plays a pericellular localization pattern. However, it is also 1998; Melrose et al. 2008b). Via these interactions, perlecan
prominent in the territorial and interterritorial matrix of the participates in disc development and in conversion of the
developing disc (Fig. 4.4) (Smith et al. 2009). Perlecan cartilaginous vertebral rudiment cartilages into bone during
expression is elevated in hypertrophic chondrocytes sur- spine development (Smith et al. 2009). This is consistent
Biglycan Fibromodulin Rabit lgG ve Perlecan

a b c control
d
Fig. 4.4 Immunolocalization of biglycan (a), fibromodulin (b), nonimmune rabbit IgG negative control (c), and perlecan (d) in a 14-week
gestational age fetal human intervertebral disc and adjacent cartilaginous vertebral body rudiment cartilages
VGP CEP
Blood vessel
50 m
100 m
NP AF
50 m 100 m
Fig. 4.5 Immunolocalization of perlecan in the newborn ovine lumbar intervertebral disc. Perlecan is present as a pericellular proteoglycan in the verte-
bral growth plate (VGP), cartilaginous end plate (CEP), a blood vessel within the CEP, and the nucleus pulposus (NP) and annulus fibrosus (AF)
with roles recently ascribed to perlecan as an early chondro- FGF-2 maintains chondrocytes in the permanent cartilages
genic marker in the development of cartilaginous tissues in a delayed state of differentiation, where they are respon-
(Smith et al. 2010). sible for the replenishment of matrix components to effect
Perlecan promotes extracellular matrix production tissue homeostasis.
through its interactions with members of the FGF family. Recent studies have also shown that the HS chains of per-
Perlecan is a low-affinity co-receptor for several members of lecan are important in fibrillin and elastin assembly (Hayes
the FGF family and sequesters these molecules pericellularly et al. 2011a, c) and support earlier observations concerning
for later presentation to FGFRs. In this way, FGF can pro- basement membrane assembly. Perlecan is localized to a
mote cell signaling events which drive proliferation and number of elastin-associated proteins in the intervertebral
matrix production (Chuang et al. 2010). This sequestration disc (Hayes et al. 2011c). LTBP-2 interacts with the perlecan
process also stabilizes the FGFs, protecting them from prote- HS chains (Parsi et al. 2010) and in the disc co-localizes with
olysis in situ and extending their biological half-life. Perlecan perlecan pericellularly. The biological significance of this
has been co-localized with FGF-18 in the developing spine, localization is not known; however, LTBP-2 may have some
with an overexpression evident in terminally differentiated regulatory role to play in the microfibrillogenesis process
hypertrophic vertebral growth plate chondrocytes and in (Hirai et al. 2007; Hirani et al. 2007) by occupying sites on
cells surrounding the ossification centers in the developing fibrillin-1 that LTBP-1 normally occupies (Hirani et al. 2007;
vertebral bodies. Perlecan associates with FGF-2 in the Vehvilainen et al. 2009) or by interaction with another elas-
developing intervertebral disc interspace in the fetal human tin-associated protein. Alternatively, by acting as a competi-
spine. Thus, FGF-18 promotes terminal differentiation of tive substrate for the HS chains in perlecan domain I, it may
chondrocytes, leading eventually to bone formation, whereas regulate growth factor binding to perlecan.
4.5.1 Perlecan Protein Structure in mouse perlecan (Melrose et al. 2008b; Murdoch and Iozzo
1993). The C-terminal domain V bears homology to the LG
Perlecan is a large modular HS-proteoglycan composed of laminin-type G domain and contains three LG domains sepa-
five distinct domains with homology to growth factors and to rated by EGF-like domains. The perlecan core protein is
protein modules involved in lipid metabolism, cell adhesion, large (467 kDa) and highly aggregative under associative
and homotypic and heterotypic interactions involved in conditions, leading to the formation of higher molecular
matrix assembly and stabilization (Melrose et al. 2008b; weight forms of about 800 kDa in free solution.
Murdoch and Iozzo 1993) (Box 4.3). The N-terminal domain The perlecan core protein can contain three HS chains in the
I contains three HS attachment sites, through which N-terminal domain, and additional GAG consensus attachment
HS-mediated growth factor and morphogen interactions points have been identified in domain V; however, it has yet to
occur. Consensus regions for GAG attachment have also be definitively shown that these are occupied in the interverte-
been identified in the C-terminal domain V (Fig. 4.6a). The bral disc. Disc cells synthesize a hybrid HS/CS proteoglycan
N-terminal domain is unique to perlecan, whereas domain II form of perlecan, with at least one of the HS chains replaced by
exhibits homology to the low-density lipoprotein receptor a chondroitin-4-sulfate (C4S) chain. In the fetal and newborn
and domain III bears homology to the L4 laminin-type IV disc, this C4S chain is capped by a unique developmental CS
domain and LE laminin EGF domain. Domain IV, the largest motif identified by MAb 7D4 (Hayes et al. 2011b). However,
domain in perlecan, contains multiple immunoglobulin the abundance of this 7D4 epitope on perlecan diminishes with
repeats, although this domain is truncated by about 20 kDa aging, and it is not clear if this has any functional consequence.
Box 4.3: Perlecan and Its Interactive Components
LDL Laminin
SEA domain receptor Laminin type IV domain Immunoglobulin repeat domain type G
type A domain
HS/CS chains
I II III IV V
Laminin-1
Collagen type IV VLDL
Fibronectin LDL
FGF-7,18 nidogen-1
FGF-2, 9, 18 fibrillin-1
FGF-BP nidogen-1, 2 fibulin-2
PDFG Wnt/Ca2+
PRELP CTGF
PDFG fibulin-2 1-Integrin
Fibrillin-1
WARP fibronectin -dystroglycan
thrombospondin collagen type IV heparin
heparanase heparin sulfitides
BMP-2, 4, 7 sulfitides FGF-7
Hedgehog PDGF endostatin
Ang-3 ECM1
VEGF progranulin
IL-2 acetylcholinesterase
WARP 21 integrin
action A
follistatin
SHh
Abbreviations: FGF fibroblast growth factor, PDGF density lipoprotein, Wnt a morphogenic ligand, hybrid
platelet derived growth factor, PRELP proline/arginine- abbreviation of Int (integration-1) and Wg (wingless),
rich and leucine-rich repeat protein/prolargin, BMP bone CTGF connective tissue growth factor, FGF-BP FGF
morphogenetic protein, Ang angiopoietin, SHh sonic binding protein, WARP von Willebrand factor A domain-
hedgehog, VLDL very low density lipoprotein, LDL low related protein.
4.5.2 Perlecan Gene Organization and Mutation 4.5.3 Perlecan Degradation
The perlecan gene (HSPG2) is encoded by 94 exons located Little is known about the mechanism by which perlecan is pro-
on chromosome 1p3634 (Cohen et al. 1993; Kallunki and cessed in the intervertebral disc. Gel electrophoresis separates
Tryggvason 1992; Murdoch et al. 1992; Noonan et al. 1991), full-length perlecan from three other perlecan species in new-
with each of the structural domains being encoded by mul- born and young adult ovine discs. The latter three species are
tiple exons. smaller than full-length perlecan, devoid of GAG, and detect-
The importance of perlecan in skeletogenesis (Arikawa- able using a MAb to perlecan domain I. Thus, they represent
Hirasawa et al. 1999), vasculogenesis, and muscle and nerve fragments cleaved from the domain I carboxy terminal to the
development is evident from analyses of two naturally occur- HS chains. However, the cleavage sites themselves still await
ring mutations in the human HSPG2 gene. Schwartz-Jampel detailed characterization. Perlecan fragmentation has also been
syndrome is a relatively mild skeletal condition, which arises observed by Western blotting of newborn and 2-year-old inter-
from missense, splicing, exon skipping, and deletion muta- vertebral disc extracts using a domain I-specific MAb, with
tions. These events result in partial loss of domain IV and total one major and four minor species evident (Fig 4.6b). A similar
loss of domain V, complete loss of domain V only, or defective range of perlecan fragments has been observed in extracts of
disulfide bonding in domain III of perlecan (Arikawa-Hirasawa human knee joint articular cartilage (Melrose et al. 2006).
et al. 2002). As a result, there are reduced functional levels of In vitro digestion of endothelial cell perlecan with MMP-
perlecan in cartilaginous tissues, chondrodysplasia, myotonia, 1, MMP-3, and plasmin has demonstrated the susceptibility
impairment in the endochondral ossification process, and short of perlecan to cleavage in domains IV and V (Whitelock
stature. In the more severe condition of dyssegmental dyspla- et al. 1996). Tolloid-like metalloprotease (BMP-1) also
sia, Silverman-Handmaker type, perlecan is almost undetect- cleaves perlecan between the LG2 and LG3 domains of
able in cartilaginous tissues, and this condition is characterized domain V, within the peptide sequence HLEGSGGN--
by a severe chondrodysplasia; severe disruption in skeleto- DAPGQYGA, releasing an anti-angiogenic peptide termed
genesis; profound effects on lung, heart, muscle, and cranial endorepellin (Bix et al. 2004, 2007; Gonzalez et al. 2005).
development; synaptogenesis; and complete absence of acetyl- Endorepellin has the ability to disrupt endothelial cell a2b1
cholinesterase at the neuromuscular junction leading to dysto- integrin-based basement membrane interactions, which nor-
nia (Arikawa-Hirasawa et al. 2001a, b). The perlecan knockout mally stabilize tube formation (Bix et al. 2004, 2007;
mouse further emphasizes the essential roles of perlecan in Gonzalez et al. 2005). So far endorepellin is the only perle-
development (Arikawa-Hirasawa et al. 1999). Perlecan knock- can fragment that has been extensively characterized and its
out is a lethal condition with the majority of mouse pups dying functional properties determined.
in utero, and in those few that survive to birth, there is severe
impairment in skeletal stature, cranial and long bone deve-
lopment, and large vessel, heart, and lung development. 4.6 Lubricin
Studies with the Hspg2 exon 3 null mouse (Rossi
et al. 2003), where perlecan domain I containing the Lubricin was originally identified as the large mucinous gly-
GAG-attachment sites is ablated, are now allowing examina- coprotein present in synovial fluid, which by providing
tion of the specific role of the perlecan HS chains in skeletal boundary lubrication at the surface of articular cartilage was
development. Hspg2 exon 3 null chondrocytes are poorly responsible for friction-free joint motion (Swann et al. 1977).
responsive to FGF-2 in cell proliferation studies. Baf-32 cells This role in joint lubrication led to the name lubricin. Later,
transfected with FGFR3IIIc are also poorly responsive to knee a glycoprotein was identified in the superficial zone of artic-
rudiment cartilage perlecan predigested with heparitinase III ular cartilage (Schumacher et al. 1994) and termed superficial
to remove its HS chains, indicating the essential role of the zone protein (SZP). It is produced by superficial zone chon-
domain I HS chains for FGF-2 binding and cell signaling drocytes and has been shown to be analogous in structure to
processes (Hayes et al. 2011b). This is not the case for FGF- lubricin (Jay et al. 2001b). Unlike lubricin, SZP has been
18, which induces cell proliferation, even in the absence of shown to exist in part as a CS proteoglycan, termed proteo-
perlecan HS chains (Hayes et al. 2011b), but consistent with glycan 4 (PRG4). It is however not clear whether this dis-
a domain III FGF-18 reactive site in perlecan. The Hspg2 tinction exists at all ages or in all disease states or whether
exon 3 null mouse has a relatively mild phenotype with no the CS chain contributes to SZP function. Recently, lubricin
apparent defects in cartilage assembly. However, recent stud- has also been shown to reside in the intervertebral disc (Jay
ies have indicated that with maturation, defects become evi- et al. 2001b; Shine et al. 2009; Shine and Spector 2008).
dent in cartilaginous tissues, and its reparative ability after a Lubricin is present in all regions of the disc but appears to
traumatic challenge also appears to be impaired. Fibrillin-1 be most abundant in the nucleus pulposus (Shine et al. 2009).
assembly and deposition is also impaired in the Hspg2 exon Its core protein undergoes extensive proteolytic degradation,
3 null mutant mouse intervertebral disc (Hayes et al. 2013). with accumulation of the degradation products in the tissue;
Fig. 4.6 (a) Schematic represen- a

tation of the structural organization Scale
of perlecan. (b) SDS/PAGE 50 kDa
analysis of perlecan heterogeneity
in newborn and 2-year-old ovine
intervertebral disc samples. Note
the extensive fragmentation of disc
perlecan compared to that from
human vascular endothelial cells
(HUVEC). The multiple perlecan
species that are discernible in the I II III IV V
disc are identified by arrows in the
right hand margin. OAF outer
annulus fibrosus, IAF inner
annulus fibrosus, NP nucleus
pulposus
Key: CS side chain HS side chain Core protein
b Newborn 2 years old
Std OAF IAF NP HUVEC OAF IAF NP

Mw
(kDa)
460
268
238
171
117
71
55
it has a polydisperse size distribution comparable to other oligosaccharides (Swann et al. 1981b), which reside in a
tissue sources, such as synovial fluid or articular cartilage long central domain. This mucin-like domain is flanked by
(Fig. 4.7b). It is not known which proteinases are responsible N- and C-terminal cysteine-rich domains that resemble
for cleavage in vivo, but they are likely to be the same as domains of vitronectin. Domains possessing somatomedin
those involved in aggrecan and versican degradation. In this B homology and a heparin-binding domain may reside at its
respect, MMPs have been demonstrated to degrade lubricin N-terminus, and a hemopexin domain resides at its
in vitro (Elsaid et al. 2005). Also unknown is the precise C-terminus (Fig. 4.7a). There is a single consensus sequence
function of lubricin in the disc and how proteolysis may for the attachment of CS near the N-terminus of the mucin-
affect lubricin function. One possibility is a role in lubricat- like domain. In its SZP form, lubricin has also been reported
ing motion between adjacent annulus fibrosus lamellae. to contain KS, but it is unclear where this resides. The lubri-
cating properties of lubricin are conferred by its mucin-like
domain (Jay et al. 2001a), but the other domains provide the
4.6.1 Lubricin Protein Structure potential for extracellular interactions and cell associations.
The terminal domains appear to be important, as their reduc-
Intact lubricin has a molecular weight of about 240 kDa, of tion and alkylation impairs lubricin function (Swann et al.
which about 50 % is contributed by O-linked mucin-like 1981a).
Fig. 4.7 (a) Schematic represen-

tation of the structural organiza- a
NH2 COOH
tion of lubricin. (b) SDS/PAGE
analysis of lubricin heterogeneity
in the human IVD. The lumican
core protein is more fragmented in
the intervertebral disc than in syn- Key: Core protein
CS side chain Scale
ovial fluid or articular cartilage
50 kDa
Somatomedian B homologies Hemopexin domain
Heparin binding Mutin-like O-linked oligosaccharides

domain domains
b
MW
kDa
180
115 SF Synovial fluid

C Articular cartilage
82 AF Annulus fibrosus
NP Nucleus Pulposus
64
49
37
26
19
SF C AF NP
4.6.2 Lubricin Gene Organization and Mutation domain flanking the central mucin-like domain. This heteroge-
neity accounts for the variable size of intact lubricin.
The mRNA encoding lubricin/SZP has been shown to origi- Mutations in the human PRG4 gene give rise to
nate by alternative splicing of the megakaryocyte stimulating the autosomal recessive camptodactyly-arthropathy-coxa
factor (MSF) precursor gene (PRG4) (Flannery et al. 1999). vara-pericarditis syndrome (CACP) (Marcelino et al. 1999).
The human PRG4 gene resides on chromosome 1 and pos- Most of the originally identified mutations causing frame-
sesses 12 exons (Merberg et al. 1993). Exon 1 encodes the sig- shifts or nonsense substitution near the end of exon 6 or
nal peptide, exons 2 and 3 encode domains with somatomedin in subsequent exons result in truncation of the hemopexin
B homology, and exons 4 and 5 encode a region containing a domain. These mutations result in a lack of lubricin in syn-
heparin-binding domain. Exon 6 encodes the mucin-like ovial fluid and a lack of lubricin production by cultured syn-
domain, possessing about 80 potential attachment sites for oviocytes (Rhee et al. 2005b). Many of the features of CACP
O-linked oligosaccharides and a single potential attachment are recapitulated in the lubricin knockout mouse (Rhee et al.
site for CS. Exons 712 encode the carboxy terminus of the 2005a), supporting the concept that deficient production of
molecule, which contains a domain with hemopexin-like lubricin is responsible for the CACP phenotype. The wide
homology. Both synovial lubricin and cartilage SZP are derived range of symptoms associated with CACP show that lubricin
by alternative splicing of a combination of PRG4 exons 2, 4, function is not restricted to joint motion but also influences
and 5 to yield several messages. Thus, there is no unique tendons and the heart. While patients with CACP have spine
structure for the lubricin core protein, but all forms possess at abnormalities (Faivre et al. 2000), it is not clear whether disc
least one somatomedin B-like domain and a hemopexin-like function is also affected in these individuals.
Fig. 4.8 Schematic repre-

sentation of the structural
organization of SLRPs found
in the intervertebral disc
DCN
BGN
SO4 SO4 SO4

Y Y Y
Y Y Y
SO4 SO4 SO4 FMOD
Scale
10 kDa
SO4 SO4
Y Y
Y Y
SO4 SO4 LUM
RRR
PPP PRELP
DDD
DDD ASPN
CHAD
SO4 SO4 DDD

Key: RRR Arg cluster PPP Pro cluster Y Y Sulphated Try cluster Aspartic acid cluster
N-linked oligosaccharide Disulphide stabilised 4 cysteine or 2 cysteine cluster
CS (DS) side chain KS side chain Core protein
4.7 The Small Leucine-Rich Repeat including the CS-/DS-substituted decorin and biglycan;
Proteoglycans (SLRPS) KS-substituted lumican, fibromodulin, and keratocan; and
non-glycanated proline/arginine-rich protein (PRELP, pro-
The SLRPs are members of a large family of leucine-rich largin), chondroadherin, and asporin (Fig. 4.8). These SLRPs
repeat (LRR) proteins, which contain multiple adjacent 24 contain ten LRRs, which are flanked by amino and carboxy
amino acid domains bearing a common leucine-rich motif terminal disulfide-bonded regions.
(Hocking et al. 1998). The SLRPs have been categorized into Decorin and biglycan possess attachment sites for their
a number of subfamilies on the basis of their gene organiza- CS/DS side chains within the extreme amino terminus of
tion, number of LRRs, type of GAG substitutions, and gen- their core proteins. In decorin, there is one such site, whereas
eral structural organization (Kalamajski and Oldberg 2010). biglycan contains two GAG substitution sites (Roughley
Eight SLRPs have been identified in the intervertebral disc, and White 1989). In most connective tissues, including the
intervertebral disc, the CS chains are modified in the Golgi decorin located at LRR 7 and the binding site on asporin
by epimerization of the b-D-glucuronic acid moieties to located at C-terminal of this site (Kalamajski et al. 2009).
a-L-iduronic acid to form DS. Non-glycanated forms of While lumican and fibromodulin both utilize leucine repeat
decorin and biglycan devoid of DS chains have been domains five to seven to bind to fibrillar collagen and both
observed in disc tissues, and as with other connective tis- can regulate early collagen fibril assembly processes, only
sues, their relative abundance accumulates with age. These fibromodulin facilitates growth steps leading to mature fibril
non-glycanated molecules likely arise by proteolysis in formation (Kalamajski and Oldberg 2009). Decorin and big-
the N-terminal region. Besides the removal of a small lycan display coordinated control of collagen fibrillogenesis
N-terminal signal peptide, the mature core proteins of deco- during development and acquisition of biomechanical
rin and biglycan are generated by removal of additional properties during tendon development (Iozzo et al. 2011).
amino acid segments of 14 and 21 amino acids, respectively Alterations in the functional properties of SLRP members
(Roughley et al. 1996b; Scott et al. 2000). However, the through point mutations affecting the structure of critical
functional consequence of the removal of these pro-pep- interactive regions within the LRRs can give rise to impaired
tides is not known. tissue assembly and function.
Lumican and fibromodulin possess four N-linked oligo-
saccharide chains within their central LRRs, which may be
modified to KS, although substitution at all sites is uncom- 4.7.1 Decorin
mon (Plaas et al. 1990). Non-glycanated forms of lumican
and fibromodulin also occur in connective tissues, due to Decorin is so named since it was found to decorate the
lack of KS substitution (Grover et al. 1995; Roughley et al. surface of collagen fibrils. The decorin gene (DCN) is located
1996a). Only the small N-terminal signal peptides are on chromosome 12q21.3q23. It possesses 8 exons encoding
removed from the lumican and fibromodulin core proteins to a 36kDa core protein, which contains a single CS or DS
form the mature core proteins in situ. Fibromodulin and chain located at position 4 in the mature human core protein
lumican also have a number of sulfated tyrosine residues sequence (Roughley 2006). Decorin interacts with the d
clustered at their extreme N-termini in the mature protein, and e bands of collagen I fibrils, fibronectin, C1q, EGF
which may also contribute to the anionic nature of these receptor, TGF-b, and thrombospondin. Thus, it has roles in
SLRPs in situ (Antonsson et al. 1991; Onnerfjord et al. 2004; the regulation of collagen fibrillogenesis in vitro and fibrosis
Tillgren et al. 2009). in vivo and controls the bioavailability of TGF-b. Decorin
PRELP and asporin are non-glycanated SLRPs but con- also influences cell proliferation at certain stages of the cell
tain clusters of arginine and proline and aspartic acid, respec- cycle and has roles as a linking module in collagenous matri-
tively, at their N-termini (Bengtsson et al. 1995; Grover and ces (Iozzo and Schaefer 2010).
Roughley 1998, 2001). PRELP is unique amongst the SLRPs The regions of decorin which interact with collagen are
in possessing a cationic N-terminal region. Chondroadherin located within the leucine-rich repeats at dissimilar regions
is devoid of an N-terminal core protein region where charged to those involved in its interaction with TGF-b. Molecular
amino acids are clustered in PRELP and asporin. It has a modeling predicts that decorin possesses a horseshoe
similar LRR core protein structure (Grover et al. 1997), but conformation capable of accommodating a single collagen
differs from the other SLRPs in the structure of its C-terminal molecule at the surface of the collagen fibrils within its con-
region. cave face (Orgel et al. 2009; Scott 1996, 2003; Scott and
SLRPs have important roles to play as tissue organiz- Stockwell 2006). However, X-ray diffraction analysis of
ers based on how they affect the orientation and assembly decorin crystals indicates that it exists as a dimer with inter-
of collagenous matrices and how they interact with growth locking concave faces (Scott et al. 2004), although it is not
factors and cell surface receptors during tissue develop- clear whether decorin dimers represent the functional form
ment and matrix remodeling (Iozzo et al. 2011). They dis- in solution and how this impacts its interactions with other
play diversified and overlapping functional properties with molecules. Decorin, together with biglycan, fibromodulin,
a level of redundancy between SLRP members. Asporin and lumican, also interacts with collagen VI, XII, and XIV
binds to collagen with an affinity in the low nanomolar (Nareyeck et al. 2004; Wiberg et al. 2002), fibronectin and
ranges, as does decorin (Kalamajski et al. 2009). Asporin elastin, and growth factors and cytokines such as EGF, IGF,
and decorin bind to the same region in fibrillar collagen and TGF-b, and TNF-a. SLRP GAG chain interactions with
can effectively compete with one another for this site when growth factors sequester them in the matrix surrounding cells
applied at equimolar concentrations, whereas biglycan can- in cartilaginous matrices, with the possibility that SLRPs
not (Kalamajski and Oldberg 2010). The collagen-binding play a role in the regulation of the bioavailability of these
sites on asporin and decorin differ, with the binding site for growth factors.
The decorin core protein is susceptible to cleavage by signaling molecule, with an extensive repertoire of molecu-
MMPs in vitro (Imai et al. 1997; Monfort et al. 2006) and is lar interactions with growth factors and receptors, which
fragmented in an ovine annular lesion model of intervertebral regulate cell growth, morphogenesis, and immunity (Iozzo
disc degeneration (Melrose et al. 2007) and in an ovine and Schaefer 2010).
meniscectomy model of osteoarthritis in areas undergoing
remodeling processes (Young et al. 2005). Fragmentation of
decorin is also evident in meniscal and articular cartilages 4.7.3 Asporin
from osteoarthritic knees and hips (Melrose et al. 2008a).
Decorin is processed by three isoforms of BMP-1 (von The naming of asporin reflects its unique N-terminal aspartic
Marschall and Fisher 2010) and MT1-MMP (Mimura et al. acid cluster (Henry et al. 2001; Lorenzo et al. 2001). Asporin
2009). However, by acting as a sacrificial substrate on the is also known as periodontal-associated protein-1 (PLAP-1).
surface of collagen fibers or through steric exclusion effects The asporin gene (ASPN) is located at chromosome 9q2123
which prevent the collagenase from accessing the collagen and possesses eight exons that encode a 43 kDa non-glycan-
substrate, its presence may protect the protein from cleavage ated core protein. The asporin core protein exhibits polymor-
by collagenases (Geng et al. 2006). phism within its N-terminal region, which contains 920
aspartic acid repeats. An association of the ASPN D14 allele
has been observed with disc degeneration (Gruber et al.
4.7.2 Biglycan 2009; Song et al. 2008). Asporin is a cell-associated SLRP
and is found predominantly in the outer annulus fibrosus
The biglycan gene (BGN) is located on chromosome Xq28 (Gruber et al. 2009).
(McBride et al. 1990). It possesses eight exons encoding the
38 kDa biglycan core protein, which has two CS/DS chains
located at amino acids five and ten of the mature human core 4.7.4 Fibromodulin
protein sequence. Non-glycanated forms of biglycan have
been detected and appear to be the result of proteolysis within The fibromodulin gene (FMOD) is located on chromosome
the amino terminal region of the core protein. Mature biglycan 1q32 (Antonsson et al. 1993; Sztrolovics et al. 1994) and
is cleaved by MMPs (Monfort et al. 2006), and probiglycan by possesses three exons encoding the 42 kDa fibromodulin
BMP-1 (Scott et al. 2000). Extensive biglycan fragmentation core protein. Fibromodulin is a KS-substituted SLRP which
is also evident in models of osteoarthritis induced by menis- shares significant amino acid sequence homology with deco-
cectomy (Young et al. 2005), in an annular lesion model of rin and biglycan (Antonsson et al. 1993; Oldberg et al. 1989)
experimental disc degeneration (Melrose et al. 2007), and in but contains four N-linked oligosaccharide sites within the
pathological meniscal and articular cartilage from osteoar- LRR domains, which can potentially be substituted with KS.
thritic knees and hips (Melrose et al. 2008a). Fibromodulin also possesses a cluster of amino terminal sul-
Unlike the interaction of decorin, fibromodulin, and lumi- fated tyrosine residues which impart an anionic character to
can with collagen fibrils, the association of biglycan with the molecule and allow this region to interact with clusters of
collagen fibrils is sensitive to environmental conditions. This basic residues in a variety of heparin-binding proteins,
may explain the differences observed in the distribution of including a number of bioactive factors (Onnerfjord et al.
biglycan, which is more of a cell-associated SLRP than the 2004; Tillgren et al. 2009). Non-glycanated forms of
other SLRP members. Biglycan interacts with the lattice- fibromodulin can accumulate in tissues (Grover et al. 1995;
forming collagen VI to form chondron basketlike structures Roughley et al. 1996a), due to an age-dependent decline in
around cells in cartilaginous matrices and the intervertebral KS synthesis.
disc. Biglycan has a widespread distribution in the develop- Fibromodulin interacts with collagen I and II fibrils and
ing human fetal disc and prominently demarcates its margins inhibits fibrillogenesis in vitro (Chen et al. 2010; Ezura et al.
with the cartilaginous vertebral body rudiment cartilages, 2000). Fibromodulin also interacts with TGF-b and C1q and
where it is also prominently localized. As such, its position may have roles in TGF-b sequestration, in inflammation, in
delineates the margins of the developing intervertebral disc the regulation of the assembly of collagen I and II fibrils
interspace from the presumptive cartilage end plate (Fig. 4.4). in vivo, and in the bio-regulation of TGF-b activity. The use
Biglycan also interacts with BMPs and TGF-b. Biglycan of fibromodulin knockout mice has demarcated the significant
plays a role in the initiation of the inflammatory response functional roles of fibromodulin in cartilaginous tissues with
during tissue stress, by binding to BMP/TGF-b, and may regard to collagen assembly and in the regulation of the
modulate their activities, influencing fibrosis and skeletal fibrillogenesis (Chakravarti 2002; Goldberg et al. 2006;
cell differentiation. Biglycan also has emerging roles as a Svensson et al. 1999).
Fibromodulin is prominently immunolocalized in the car- osteoarthritis, with increased levels of lumican core protein
tilaginous vertebral rudiment cartilages of the human fetal evident (Young et al. 2005).
spine and also the developing disc, and its distribution in the
fetal spine prominently demarcates the margins of the devel-
opmental intervertebral disc with the vertebral rudiment car- 4.7.6 PRELP
tilages (Fig. 4.4). In the mature disc, fibromodulin is localized
with collagenous structures pericellularly and in the intersti- PRELP is a 55 kDa non-glycanated SLRP (Bengtsson et al.
tial matrix. It occurs throughout the intervertebral disc, but 1995) encoded by the PRELP gene which is located on chro-
predominantly in the annulus fibrosus. mosome 1q32 and possesses three exons (Grover et al. 1996).
When bound to collagen, fibromodulin is susceptible to Its name PRELP is based on the presence of a unique cluster
cleavage by MMP-13 which removes the N-terminal sul- of N-terminal arginine and proline residues. PRELP binds to
fated tyrosine cluster (Heathfield et al. 2004). Soluble perlecan and collagens and may act as a basement membrane
fibromodulin is not, however, susceptible to cleavage by anchor or as a linking molecule bringing together collage-
MMP-13, neither is it degraded by MMP-2, MMP-8, and nous tissue networks (Bengtsson et al. 2002). The N-terminus
MMP-9. The MMP-13 cleavage site is contained within a of PRELP binds to heparin and HS and thereby promotes
10 kDa N-terminal peptide, Gln19-Lys 98, which is located interactions with the HS-proteoglycan perlecan (Bengtsson
adjacent to the sulfated tyrosine cluster (Heathfield et al. et al. 2000).
2004). Fibromodulin is extensively fragmented in patho-
logical meniscal and articular cartilage from osteoarthritic
knees and hip (Melrose et al. 2008a), in an ovine annular 4.7.7 Chondroadherin
lesion model of intervertebral disc degeneration in areas
undergoing remodeling (Melrose et al. 2007), and in an Chondroadherin is a developmentally regulated SLRP (Shen
ovine meniscectomy model of osteoarthritis (Young et al. et al. 1998) closely related to decorin, biglycan, fibromodulin,
2005). lumican, and PRELP (Neame et al. 1994). The chondroad-
herin gene (CHAD) is located on chromosome 17q21.33 and
possesses 4 exons encoding the 36 kDa non-glycanated
4.7.5 Lumican chondroadherin core protein (Grover et al. 1997), which con-
tains 10 LRRs (Neame et al. 1994). Chondroadherin inter-
The lumican gene (LUM) is located at chromosome 12q21.3 acts with collagen II (Mansson et al. 2001) and a2b1 integrin
q22 (Danielson et al. 1999; Grover et al. 1995) and possesses (Camper et al. 1997; Haglund et al. 2011) and plays a role in
three exons encoding the 38 kDa lumican core protein, which the attachment of connective tissue cells to matrix compo-
has four N-linked sites within the LRR domain that poten- nents; in this way it could maintain the cellular phenotype
tially can be substituted with KS. Lumican interacts with and regulate tissue homeostasis. Depletion and proteolytic
similar partners to those outlined for fibromodulin and dis- fragmentation of chondroadherin occurs in scoliotic inter-
plays similar roles to those of fibromodulin in tissues. vertebral discs and appears to be associated with disc matrix
Lumican and fibromodulin have homologous sequences in remodeling (Haglund et al. 2009).
the 57 LRRs which compete for collagen binding
(Kalamajski and Oldberg 2009), and they bind to the same
regions of collagen I fibrils (Svensson et al. 2000). However, 4.7.8 SLRP Knockout Mice and Gene Mutations
despite the significant similarities between lumican and
decorin, they have dissimilar binding sites on collagen I A number of human diseases are associated with SLRP
(Hedbom and Heinegard 1993) and modulate collagen mutations, and mouse knockout models have identified
fibrillogenesis independently (Neame et al. 2000). pathological consequences resulting from the ablation of
As with fibromodulin and decorin, lumican is also sus- SLRP genes. Examples of pathological features induced by
ceptible to degradation by MMPs, but apparently to a lesser SLRP gene ablation and human SLRP gene mutations are
extent, and it is also cleaved by MT1-MMP (Li et al. 2004). described in Tables 4.1 and 4.2. Thus far, asporin is the only
Fragmentation of lumican is a prominent feature in patho- SLRP specifically associated with human disc disease and
logical meniscal and articular cartilages from osteoarthritic the disc phenotype. The intervertebral disc nevertheless
knees and hips (Melrose et al. 2008a) and has also been exhibits similarities in collagenous organization to tensional
observed in an ovine annular lesion model of disc degenera- and weight-bearing connective tissues which display pheno-
tion in areas undergoing remodeling (Melrose et al. 2007). types in SLRP knockout mouse models. It is highly probable
Although it does not contain KS, lumican expression is, that SLRPs have specific roles to play in disc degeneration
however, upregulated in an ovine meniscectomy model of which await detailed elucidation.
Table 4.1 Pathological consequences of targeted ablation of SLRP Table 4.2 Human diseases linked to mutations in SLRP genes
genes in mouse models
Type of Pathology/clinical
Ablated gene Pathology Phenotype References Gene mutation phenotype References
Decorin Collagen fibril Skin fragility Danielson Decorin Frameshift Corneal opacity, Bredrup et al.
structure in skin Reduced tendon et al. mutation congenital stromal (2005)
and tendon function (1997) generating a dystrophy of
abnormal C-terminal cornea
Biglycan Decreased bone Osteoporosis Heegaard truncated core
mass, structural Spontaneous et al. protein
collagen fibril aortic dissection/ (2007) and Lumican, Intronic Corneal detach- Chen et al.
abnormalities in rupture Xu et al. fibromodulin, variations, ment, choroidal (2009), Majava
medial aorta (1998) PRELP single-nucle- neovasculariza- et al. (2007),
Lumican Abnormal Skin fragility Chakravarti otide tion, high myopia andWang et al.
collagen fibril Corneal opacity et al. polymor- (2006)
organization in (1998) phisms in
cornea and dermis promoter
Fibromodulin Abnormal Lax tendons with Svensson Asporin Asporin D4 Early onset of OA Gruber et al.
collagen fibril reduced et al. allele and disc (2009) and
organization in biomechanical (1999) polymor- degeneration Song et al.
tendon function phisms (2008)
Biglycan/ Abnormal Similar to Corsi et al. affecting
decorin collagen fibril progeroid form (2002) N-terminal of
formation in of Ehlers-Danlos core protein
bone, tendon, syndrome
dermis
Biglycan/fibro Maturational Gait impairment, Ameye and swelling maintains disc height underload and is respon-
modulin structural/ ectopic Young sible for the diurnal variation in disc height associated
biomechanical calcification, (2002) with daily life.
abnormalities in premature OA
Versican and lubricin are present throughout the interverte-
collagen fibrils in
tendon bral disc, but their precise functions are unclear. The high
Lumican/fibro Abnormal Joint laxity and Jepsen abundance of versican in the immature disc may suggest a
modulin collagen impaired tendon et al. role in disc development during fetal life, and it has been
maturation and function (2002) suggested that lubricin plays a role in enhancing motion
structure
of tendons
between adjacent lamellae within the annulus fibrosus.
Asporin Reduced Diminished Tomoeda Perlecan promotes and stabilizes matrix assembly through
inhibition of negative et al. its interaction with a diverse repertoire of matrix compo-
periodontal regulation of (2008) and nents. Roles are now also emerging for perlecan in
ligament periodontal Yamada fibrillin-1 and elastin assembly in the disc, with the HS
mineralization ligament cell et al.
cytodifferentia- (2007) chains of perlecan being important. The HS chains of per-
tion by point lecan also sequester a number of bioactive growth factors
mutations and morphogens of relevance to developmental and
disrupting asporin remodeling processes in the disc through their abilities to
LRR5-mediated
interactions with regulate cell proliferation and differentiation. Furthermore,
BMP-2 the ability of perlecan to interact with a number of cell
attachment proteins may regulate cellular recruitment
during development and remodeling of the disc.
4.8 Summary of Critical Concepts Discussed The SLRPs have important regulatory roles to play in col-
in the Chapter lagen fibrillogenesis and are important for disc extracellu-
lar matrix assembly and repair. Emerging interactive roles
Role of Proteoglycans in Disc Function of SLRPs with cytokines and bioactive growth factors
Aggrecan provides the disc with its ability to swell and implicate them in cell signaling, affecting a diverse range
resist compressive loads. This property is dependent on of biological processes, including fibrosis, inflammation,
two features of aggrecan: first, its ability to form large and the immune response.
proteoglycan aggregates in association with HA, which Role of Proteoglycans in Disc Disease
limit its diffusion within the matrix, and second, the Aggrecan degradation is associated with disc degenera-
osmotic properties and water binding are due to its sub- tion in the adult and probably in the juvenile with scolio-
stitution with numerous CS and KS chains. Aggrecan sis. The proteolytic degradation of aggrecan results in
fragments that are no longer able to interact with HA; Antonsson P, Heinegard D, Oldberg A (1993) Structure and deduced
these are slowly lost as their size is further decreased by amino acid sequence of the human fibromodulin gene. Biochim
Biophys Acta 1174:204206
continuing proteolysis. This decline in aggrecan content Arikawa-Hirasawa E, Watanabe H, Takami H, Hassell JR, Yamada Y
of the degenerate disc may facilitate blood vessel and (1999) Perlecan is essential for cartilage and cephalic development.
nerve ingrowth and contribute to discogenic pain. Nat Genet 23:354358
Mutations in the aggrecan gene result in osteochondrodys- Arikawa-Hirasawa E, Wilcox WR, Le AH, Silverman N, Govindraj P,
Hassell JR, Yamada Y (2001a) Dyssegmental dysplasia, Silverman-
plasia, the features of which include developmental Handmaker type, is caused by functional null mutations of the per-
abnormalities in the disc. lecan gene. Nat Genet 27:431434
Versican and lubricin undergo extensive proteolytic frag- Arikawa-Hirasawa E, Wilcox WR, Yamada Y (2001b) Dyssegmental
dysplasia, Silverman-Handmaker type: unexpected role of perlecan
mentation throughout life, which is likely caused by the in cartilage development. Am J Med Genet 106:254257
same proteinases responsible for aggrecan degradation. It Arikawa-Hirasawa E, Le AH, Nishino I, Nonaka I, Ho NC, Francomano
is, however, unclear whether this contributes to disc CA, Govindraj P, Hassell JR, Devaney JM, Spranger J, Stevenson
pathology. RE, Iannaccone S, Dalakas MC, Yamada Y (2002) Structural and
functional mutations of the perlecan gene cause Schwartz-Jampel
Perlecan also undergoes extensive fragmentation within syndrome, with myotonic myopathy and chondrodysplasia. Am
the disc, though little is known as to how this occurs or J Hum Genet 70:13681375
whether the released perlecan fragments provide new Barry FP, Neame PJ, Sasse J, Pearson D (1994) Length variation in the
functional biological properties. An exception to this is keratan sulfate domain of mammalian aggrecan. Matrix Biol
14:323328
the C-terminal peptide, endorepellin, which has anti- Barry FP, Rosenberg LC, Gaw JU, Koob TJ, Neame PJ (1995) N- and
angiogenic properties due to its ability to destabilize tube O-linked keratan sulfate on the hyaluronan binding region of aggre-
formation by endothelial cells. In degenerate discs, if the can from mature and immature bovine cartilage. J Biol Chem
fragments are not lost or further degraded, these proper- 270:2051620524
Bengtsson E, Neame PJ, Heinegard D, Sommarin Y (1995) The pri-
ties might be expected to maintain the avascular nature of mary structure of a basic leucine-rich repeat protein, PRELP, found
the disc. in connective tissues. J Biol Chem 270:2563925644
A number of the SLRPs are also fragmented in degenerate Bengtsson E, Aspberg A, Heinegard D, Sommarin Y, Spillmann D
discs, and the loss of the CS/DS chains in decorin and (2000) The amino-terminal part of PRELP binds to heparin and
heparan sulfate. J Biol Chem 275:4069540702
biglycan may adversely affect their abilities to interact Bengtsson E, Morgelin M, Sasaki T, Timpl R, Heinegard D, Aspberg A
with cytokines and growth factors and thus their partici- (2002) The leucine-rich repeat protein PRELP binds perlecan and
pation in cell regulatory processes. Fragmentation of collagens and may function as a basement membrane anchor. J Biol
lumican and fibromodulin is also evident in the degener- Chem 277:1506115068
Bix G, Fu J, Gonzalez EM, Macro L, Barker A, Campbell S, Zutter
ate disc, but it is not known how this impacts tissue MM, Santoro SA, Kim JK, Hook M, Reed CC, Iozzo RV (2004)
function. Endorepellin causes endothelial cell disassembly of actin cytoskel-
eton and focal adhesions through alpha2beta1 integrin. J Cell Biol
Acknowledgements We would like to thank the Shriners of North 166:97109
America and the Canadian Institutes of Health Research for support of Bix G, Iozzo RA, Woodall B, Burrows M, McQuillan A, Campbell S,
the work described in this chapter. Support from the National Health Fields GB, Iozzo RV (2007) Endorepellin, the C-terminal angio-
and Medical Research Council of Australia is also gratefully static module of perlecan, enhances collagen-platelet responses via
acknowledged. the alpha2beta1-integrin receptor. Blood 109:37453748
Botsford DJ, Esses SI, Ogilvie-Harris DJ (1994) In vivo diurnal varia-
tion in intervertebral disc volume and morphology. Spine (Phila Pa
1976) 19:935940
Bredrup C, Knappskog PM, Majewski J, Rodahl E, Boman H (2005)
References Congenital stromal dystrophy of the cornea caused by a mutation in
the decorin gene. Invest Ophthalmol Vis Sci 46:420426
Adams MA, Roughley PJ (2006) What is intervertebral disc degenera- Brown GM, Huckerby TN, Bayliss MT, Nieduszynski IA (1998)
tion, and what causes it? Spine (Phila Pa 1976) 31:21512161 Human aggrecan keratan sulfate undergoes structural changes dur-
Agrawal A, Guttapalli A, Narayan S, Albert TJ, Shapiro IM, Risbud ing adolescent development. J Biol Chem 273:2640826414
MV (2007) Normoxic stabilization of HIF-1alpha drives glycolytic Camper L, Heinegard D, Lundgren-Akerlund E (1997) Integrin alpha-
metabolism and regulates aggrecan gene expression in nucleus pul- 2beta1 is a receptor for the cartilage matrix protein chondroadherin.
posus cells of the rat intervertebral disk. Am J Physiol Cell Physiol J Cell Biol 138:11591167
293:C621C631 Caterson B (2012) Fell-Muir lecture: chondroitin sulphate glycosamin-
Akeda K, An HS, Pichika R, Patel K, Muehleman C, Nakagawa K, oglycans: fun for some and confusion for others. Int J Exp Pathol
Uchida A, Masuda K (2007) The expression of NG2 proteoglycan 93:110
in the human intervertebral disc. Spine (Phila Pa 1976) 32: Caterson B, Mahmoodian F, Sorrell JM, Hardingham TE, Bayliss MT,
306314 Carney SL, Ratcliffe A, Muir H (1990) Modulation of native chon-
Ameye L, Young MF (2002) Mice deficient in small leucine-rich pro- droitin sulphate structure in tissue development and in disease.
teoglycans: novel in vivo models for osteoporosis, osteoarthritis, J Cell Sci 97(Pt 3):411417
Ehlers-Danlos syndrome, muscular dystrophy, and corneal diseases. Chakravarti S (2002) Functions of lumican and fibromodulin: lessons
Glycobiology 12:107R116R from knockout mice. Glycoconj J 19:287293
Antonsson P, Heinegard D, Oldberg A (1991) Posttranslational Chakravarti S, Magnuson T, Lass JH, Jepsen KJ, LaMantia C, Carroll H
modifications of fibromodulin. J Biol Chem 266:1685916861 (1998) Lumican regulates collagen fibril assembly: skin fragility
and corneal opacity in the absence of lumican. J Cell Biol 141: Clinical variability and genetic homogeneity of the camptodactyly-
12771286 arthropathy-coxa vara-pericarditis syndrome. Am J Med Genet
Chen ZT, Wang IJ, Shih YF, Lin LL (2009) The association of haplo- 95:233236
type at the lumican gene with high myopia susceptibility in Flannery C, Stanescu V, Morgelin M, Boynton R, Gordy J, Sandy J
Taiwanese patients. Ophthalmology 116:19201927 (1992) Variability in the G3 domain content of bovine aggrecan
Chen S, Oldberg A, Chakravarti S, Birk DE (2010) Fibromodulin regu- from cartilage extracts and chondrocyte cultures. Arch Biochem
lates collagen fibrillogenesis during peripheral corneal develop- Biophys 297:5260
ment. Dev Dyn 239:844854 Flannery CR, Hughes CE, Schumacher BL, Tudor D, Aydelotte MB,
Chuang CY, Lord MS, Melrose J, Rees MD, Knox SM, Freeman C, Kuettner KE, Caterson B (1999) Articular cartilage superficial zone
Iozzo RV, Whitelock JM (2010) Heparan sulfate-dependent signal- protein (SZP) is homologous to megakaryocyte stimulating factor
ing of fibroblast growth factor 18 by chondrocyte-derived perlecan. precursor and is a multifunctional proteoglycan with potential
Biochemistry 49:55245532 growth-promoting, cytoprotective, and lubricating properties in car-
Cohen IR, Grassel S, Murdoch AD, Iozzo RV (1993) Structural charac- tilage metabolism. Biochem Biophys Res Commun 254:535541
terization of the complete human perlecan gene and its promoter. Fosang AJ, Hardingham TE (1989) Isolation of the N-terminal globular
Proc Natl Acad Sci USA 90:1040410408 protein domains from cartilage proteoglycans. Identification of G2
Corsi A, Xu T, Chen XD, Boyde A, Liang J, Mankani M, Sommer B, domain and its lack of interaction with hyaluronate and link protein.
Iozzo RV, Eichstetter I, Robey PG, Bianco P, Young MF (2002) Biochem J 261:801809
Phenotypic effects of biglycan deficiency are linked to collagen Fosang AJ, Neame PJ, Last K, Hardingham TE, Murphy G, Hamilton
fibril abnormalities, are synergized by decorin deficiency, and mimic JA (1992) The interglobular domain of cartilage aggrecan is cleaved
Ehlers-Danlos-like changes in bone and other connective tissues. by PUMP, gelatinases, and cathepsin B. J Biol Chem 267:
J Bone Miner Res 17:11801189 1947019474
Cummings RD (2009) The repertoire of glycan determinants in the Fransson LA, Belting M, Cheng F, Jonsson M, Mani K, Sandgren S
human glycome. Mol Biosyst 5:10871104 (2004) Novel aspects of glypican glycobiology. Cell Mol Life Sci
Danielson KG, Baribault H, Holmes DF, Graham H, Kadler KE, Iozzo 61:10161024
RV (1997) Targeted disruption of decorin leads to abnormal colla- Fulop C, Walcz E, Valyon M, Glant TT (1993) Expression of alterna-
gen fibril morphology and skin fragility. J Cell Biol 136:729743 tively spliced epidermal growth factor-like domains in aggrecans of
Danielson KG, Siracusa LD, Donovan PJ, Iozzo RV (1999) Decorin, different species. Evidence for a novel module. J Biol Chem
epiphycan, and lumican genes are closely linked on murine chromo- 268:1737717383
some 10 and are deleted in lethal steel mutants. Mamm Genome Gendron C, Kashiwagi M, Lim NH, Enghild JJ, Thogersen IB, Hughes
10:201203 C, Caterson B, Nagase H (2007) Proteolytic activities of human
Day JM, Olin AI, Murdoch AD, Canfield A, Sasaki T, Timpl R, ADAMTS-5: comparative studies with ADAMTS-4. J Biol Chem
Hardingham TE, Aspberg A (2004) Alternative splicing in the 282:1829418306
aggrecan G3 domain influences binding interactions with tenascin- Geng Y, McQuillan D, Roughley PJ (2006) SLRP interaction can pro-
C and other extracellular matrix proteins. J Biol Chem 279: tect collagen fibrils from cleavage by collagenases. Matrix Biol
1251112518 25:484491
Doege K, Sasaki M, Horigan E, Hassell JR, Yamada Y (1987) Complete Girardin EP, Hajmohammadi S, Birmele B, Helisch A, Shworak NW,
primary structure of the rat cartilage proteoglycan core protein de Agostini AI (2005) Synthesis of anticoagulantly active heparan
deduced from cDNA clones. J Biol Chem 262:1775717767 sulfate proteoglycans by glomerular epithelial cells involves multi-
Doege KJ, Sasaki M, Kimura T, Yamada Y (1991) Complete coding ple 3-O-sulfotransferase isoforms and a limiting precursor pool.
sequence and deduced primary structure of the human cartilage J Biol Chem 280:3805938070
large aggregating proteoglycan, aggrecan. Human-specific repeats, Gleghorn L, Ramesar R, Beighton P, Wallis G (2005) A mutation in the
and additional alternatively spliced forms. J Biol Chem variable repeat region of the aggrecan gene (AGC1) causes a form
266:894902 of spondyloepiphyseal dysplasia associated with severe, premature
Doege KJ, Coulter SN, Meek LM, Maslen K, Wood JG (1997) A osteoarthritis. Am J Hum Genet 77:484490
human-specific polymorphism in the coding region of the aggrecan Gogate SS, Nasser R, Shapiro IM, Risbud MV (2011) Hypoxic regula-
gene. Variable number of tandem repeats produce a range of core tion of beta-1,3-glucuronyltransferase 1 expression in nucleus pul-
protein sizes in the general population. J Biol Chem 272: posus cells of the rat intervertebral disc: role of hypoxia-inducible
1397413979 factor proteins. Arthritis Rheum 63:19501960
Dours-Zimmermann MT, Zimmermann DR (1994) A novel gly- Goldberg M, Septier D, Oldberg A, Young MF, Ameye LG
cosaminoglycan attachment domain identified in two alternative (2006) Fibromodulin-deficient mice display impaired collagen
splice variants of human versican. J Biol Chem 269:3299232998 fibrillogenesis in predentin as well as altered dentin mineralization
Durigova M, Nagase H, Mort JS, Roughley PJ (2011a) MMPs are less and enamel formation. J Histochem Cytochem 54:525537
efficient than ADAMTS5 in cleaving aggrecan core protein. Matrix Gonzalez EM, Reed CC, Bix G, Fu J, Zhang Y, Gopalakrishnan B,
Biol 30:145153 Greenspan DS, Iozzo RV (2005) BMP-1/Tolloid-like metallopro-
Durigova M, Troeberg L, Nagase H, Roughley PJ, Mort JS (2011b) teases process endorepellin, the angiostatic C-terminal fragment of
Involvement of ADAMTS5 and hyaluronidase in aggrecan degrada- perlecan. J Biol Chem 280:70807087
tion and release from OSM-stimulated cartilage. Eur Cell Mater Grover J, Roughley PJ (1993) Versican gene expression in human artic-
21:3145 ular cartilage and comparison of mRNA splicing variation with
Elsaid KA, Jay GD, Warman ML, Rhee DK, Chichester CO (2005) aggrecan. Biochem J 291(Pt 2):361367
Association of articular cartilage degradation and loss of boundary- Grover J, Roughley PJ (1998) Characterization of the human proline/
lubricating ability of synovial fluid following injury and inflammatory arginine-rich end leucine-rich repeat protein (PRELP) gene pro-
arthritis. Arthritis Rheum 52:17461755 moter and identification of a repressor element. Biochem J 336(Pt
Ezura Y, Chakravarti S, Oldberg A, Chervoneva I, Birk DE (2000) 1):7782
Differential expression of lumican and fibromodulin regulate colla- Grover J, Roughley PJ (2001) Characterization and expression of
gen fibrillogenesis in developing mouse tendons. J Cell Biol murine PRELP. Matrix Biol 20:555564
151:779788 Grover J, Chen XN, Korenberg JR, Roughley PJ (1995) The human
Faivre L, Prieur AM, Le Merrer M, Hayem F, Penet C, Woo P, Hofer M, lumican gene. Organization, chromosomal location, and expression
Dagoneau N, Sermet I, Munnich A, Cormier-Daire V (2000) in articular cartilage. J Biol Chem 270:2194221949
Grover J, Chen XN, Korenberg JR, Recklies AD, Roughley PJ (1996) Humphries DE, Stevens RL (1992) Regulation of the gene that encodes
The gene organization, chromosome location, and expression of a the peptide core of heparin proteoglycan and other proteoglycans
55-kDa matrix protein (PRELP) of human articular cartilage. that are stored in the secretory granules of hematopoietic cells. Adv
Genomics 38:109117 Exp Med Biol 313:5967
Grover J, Chen XN, Korenberg JR, Roughley PJ (1997) The structure Imai K, Hiramatsu A, Fukushima D, Pierschbacher MD, Okada Y
and chromosome location of the human chondroadherin gene (1997) Degradation of decorin by matrix metalloproteinases:
(CHAD). Genomics 45:379385 identification of the cleavage sites, kinetic analyses and transform-
Gruber HE, Ingram JA, Hoelscher GL, Zinchenko N, Hanley EN Jr, ing growth factor-beta1 release. Biochem J 322(Pt 3):809814
Sun Y (2009) Asporin, a susceptibility gene in osteoarthritis, is Iozzo RV (1994) Perlecan: a gem of a proteoglycan. Matrix Biol
expressed at higher levels in the more degenerate human interverte- 14:203208
bral disc. Arthritis Res Ther 11:R47 Iozzo RV (1998) Matrix proteoglycans: from molecular design to cel-
Haglund L, Ouellet J, Roughley P (2009) Variation in chondroadherin lular function. Annu Rev Biochem 67:609652
abundance and fragmentation in the human scoliotic disc. Spine Iozzo RV, Schaefer L (2010) Proteoglycans in health and disease: novel
(Phila Pa 1976) 34:15131518 regulatory signaling mechanisms evoked by the small leucine-rich
Haglund L, Tillgren V, Addis L, Wenglen C, Recklies A, Heinegard D proteoglycans. FEBS J 277:38643875
(2011) Identification and characterization of the integrin alpha- Iozzo RV, Naso MF, Cannizzaro LA, Wasmuth JJ, McPherson JD (1992)
2beta1 binding motif in chondroadherin mediating cell attachment. Mapping of the versican proteoglycan gene (CSPG2) to the long
J Biol Chem 286:39253934 arm of human chromosome 5 (5q12-5q14). Genomics 14:845851
Hasegawa K, Yoneda M, Kuwabara H, Miyaishi O, Itano N, Ohno A, Iozzo R, Goldoni S, Berendsen AD, Young MF (2011) Small leucine
Zako M, Isogai Z (2007) Versican, a major hyaluronan-binding rich proteoglycans. In: Mecham R (ed) The extracellular matrix: an
component in the dermis, loses its hyaluronan-binding ability in overview, Biology of extracellular matrix. Springer, Berlin/
solar elastosis. J Invest Dermatol 127:16571663 Heidelberg, pp 197231
Hayes A, Lord MS, Smith SM, Smith MM, Whitelock JM, Weiss AS, Ito K, Shinomura T, Zako M, Ujita M, Kimata K (1995) Multiple forms
Melrose J (2011a) Colocalization in vivo and association in vitro of of mouse PG-M, a large chondroitin sulfate proteoglycan generated
perlecan and elastin. Histochem Cell Biol 136:437475 by alternative splicing. J Biol Chem 270:958965
Hayes AJ, Hughes CE, Ralphs JR, Caterson B (2011b) Chondroitin Jackson RL, Busch SJ, Cardin AD (1991) Glycosaminoglycans: molec-
sulphate sulphation motif expression in the ontogeny of the interver- ular properties, protein interactions, and role in physiological pro-
tebral disc. Eur Cell Mater 21:114 cesses. Physiol Rev 71:481539
Hayes AJ, Smith SM, Gibson MA, Melrose J (2011c) Comparative Jay GD, Harris DA, Cha CJ (2001a) Boundary lubrication by lubricin is
immunolocalisation of the elastin fibre associated proteins fibrillin-1, mediated by O-linked beta(13)Gal-GalNAc oligosaccharides.
LTBP2 and MAGP-1 with components of the collagenous and pro- Glycoconj J 18:807815
teoglycan matrix of the foetal human IVD. Spine (Phila Pa 1976) Jay GD, Tantravahi U, Britt DE, Barrach HJ, Cha CJ (2001b) Homology
36:E1365E1372 of lubricin and superficial zone protein (SZP): products of mega-
Hayes AJ, Smith SM, Melrose J (2013) Comparative immunolocalisa- karyocyte stimulating factor (MSF) gene expression by human syn-
tion of fibrillin-1 and perlecan in the human foetal and HS deficient ovial fibroblasts and articular chondrocytes localized to chromosome
Hspg2 exon 3 null mutant mouse intervertebral disc. Histochem 1q25. J Orthop Res 19:677687
Cell Biol 139:111 Jepsen KJ, Wu F, Peragallo JH, Paul J, Roberts L, Ezura Y, Oldberg A,
Heathfield TF, Onnerfjord P, Dahlberg L, Heinegard D (2004) Cleavage of Birk DE, Chakravarti S (2002) A syndrome of joint laxity and
fibromodulin in cartilage explants involves removal of the N-terminal impaired tendon integrity in lumican- and fibromodulin-deficient
tyrosine sulfate-rich region by proteolysis at a site that is sensitive to mice. J Biol Chem 277:3553235540
matrix metalloproteinase-13. J Biol Chem 279:62866295 Johnson WE, Caterson B, Eisenstein SM, Hynds DL, Snow DM,
Hedbom E, Heinegard D (1993) Binding of fibromodulin and decorin to Roberts S (2002) Human intervertebral disc aggrecan inhibits nerve
separate sites on fibrillar collagens. J Biol Chem 268:2730727312 growth in vitro. Arthritis Rheum 46:26582664
Heegaard AM, Corsi A, Danielsen CC, Nielsen KL, Jorgensen HL, Johnson WE, Caterson B, Eisenstein SM, Roberts S (2005) Human
Riminucci M, Young MF, Bianco P (2007) Biglycan deficiency intervertebral disc aggrecan inhibits endothelial cell adhesion and
causes spontaneous aortic dissection and rupture in mice. Circulation cell migration in vitro. Spine (Phila Pa 1976) 30:11391147
115:27312738 Kalamajski S, Oldberg A (2009) Homologous sequence in lumican and
Henry SP, Takanosu M, Boyd TC, Mayne PM, Eberspaecher H, Zhou fibromodulin leucine-rich repeat 57 competes for collagen bind-
W, de Crombrugghe B, Hook M, Mayne R (2001) Expression pating. J Biol Chem 284:534539
tern and gene characterization of asporin. A newly discovered mem- Kalamajski S, Oldberg A (2010) The role of small leucine-rich proteo-
ber of the leucine-rich repeat protein family. J Biol Chem 276: glycans in collagen fibrillogenesis. Matrix Biol 29:248253
1221212221 Kalamajski S, Aspberg A, Lindblom K, Heinegard D, Oldberg A (2009)
Hirai M, Horiguchi M, Ohbayashi T, Kita T, Chien KR, Nakamura T Asporin competes with decorin for collagen binding, binds calcium and
(2007) Latent TGF-beta-binding protein 2 binds to DANCE/fibulin-5 promotes osteoblast collagen mineralization. Biochem J 423:5359
and regulates elastic fiber assembly. EMBO J 26:32833295 Kallunki P, Tryggvason K (1992) Human basement membrane heparan
Hirani R, Hanssen E, Gibson MA (2007) LTBP-2 specifically interacts sulfate proteoglycan core protein: a 467-kD protein containing mul-
with the amino-terminal region of fibrillin-1 and competes with tiple domains resembling elements of the low density lipoprotein
LTBP-1 for binding to this microfibrillar protein. Matrix Biol 26: receptor, laminin, neural cell adhesion molecules, and epidermal
213223 growth factor. J Cell Biol 116:559571
Hiyama A, Gajghate S, Sakai D, Mochida J, Shapiro IM, Risbud MV Karniski LP (2001) Mutations in the diastrophic dysplasia sulfate trans-
(2009) Activation of TonEBP by calcium controls {beta}1,3- porter (DTDST) gene: correlation between sulfate transport activity
glucuronosyltransferase-I expression, a key regulator of gly- and chondrodysplasia phenotype. Hum Mol Genet 10:14851490
cosaminoglycan synthesis in cells of the intervertebral disc. J Biol Kawaguchi Y, Osada R, Kanamori M, Ishihara H, Ohmori K, Matsui H,
Chem 284:98249834 Kimura T (1999) Association between an aggrecan gene polymor-
Hocking AM, Shinomura T, McQuillan DJ (1998) Leucine-rich repeat phism and lumbar disc degeneration. Spine (Phila Pa 1976) 24:
glycoproteins of the extracellular matrix. Matrix Biol 17:119 24562460
Kloeckener-Gruissem B, Bartholdi D, Abdou MT, Zimmermann DR, Merberg D, Fitz LJ, Temple P, Giannotti J, Murtha P, Fitzgerald M,
Berger W (2006) Identification of the genetic defect in the original Scaltreto H, Kelleher K, Preissner K, Kriz R, Jacobs K, Turner K
Wagner syndrome family. Mol Vis 12:350355 (1993) A comparison of vitronectin and megakaryocyte stimulating
Kluppel M, Wight TN, Chan C, Hinek A, Wrana JL (2005) Maintenance factor. In: Rosenblatt S, Kost C, Wegerhoff J, Mosher DF, Preissner
of chondroitin sulfation balance by chondroitin-4-sulfotransferase 1 KT (eds) Biology of vitronectins and their Receptors. Elsevier,
is required for chondrocyte development and growth factor signal- Amsterdam, pp 4554
ing during cartilage morphogenesis. Development 132:39894003 Mimura T, Han KY, Onguchi T, Chang JH, Kim TI, Kojima T, Zhou Z,
Korenberg JR, Chen XN, Doege K, Grover J, Roughley PJ (1993) Azar DT (2009) MT1-MMP-mediated cleavage of decorin in cor-
Assignment of the human aggrecan gene (AGC1) to 15q26 using neal angiogenesis. J Vasc Res 46:541550
fluorescence in situ hybridization analysis. Genomics 16:546548 Monfort J, Tardif G, Reboul P, Mineau F, Roughley P, Pelletier JP,
Kurima K, Warman ML, Krishnan S, Domowicz M, Krueger RC Jr, Martel-Pelletier J (2006) Degradation of small leucine-rich repeat
Deyrup A, Schwartz NB (1998) A member of a family of sulfate- proteoglycans by matrix metalloprotease-13: identification of a new
activating enzymes causes murine brachymorphism. Proc Natl Acad biglycan cleavage site. Arthritis Res Ther 8:R26
Sci USA 95:86818685 Morgelin M, Paulsson M, Hardingham TE, Heinegard D, Engel J (1988)
Li H, Schwartz NB, Vertel BM (1993) cDNA cloning of chick cartilage Cartilage proteoglycans. Assembly with hyaluronate and link pro-
chondroitin sulfate (aggrecan) core protein and identification of a tein as studied by electron microscopy. Biochem J 253:175185
stop codon in the aggrecan gene associated with the chondrodystro- Mukhopadhyay A, Nikopoulos K, Maugeri A, de Brouwer AP, van
phy, nanomelia. J Biol Chem 268:2350423511 Nouhuys CE, Boon CJ, Perveen R, Zegers HA, Wittebol-Post D,
Li Y, Aoki T, Mori Y, Ahmad M, Miyamori H, Takino T, Sato H (2004) van den Biesen PR, van der Velde-Visser SD, Brunner HG, Black
Cleavage of lumican by membrane-type matrix metalloproteinase-1 GC, Hoyng CB, Cremers FP (2006) Erosive vitreoretinopathy and
abrogates this proteoglycan-mediated suppression of tumor cell wagner disease are caused by intronic mutations in CSPG2/Versican
colony formation in soft agar. Cancer Res 64:70587064 that result in an imbalance of splice variants. Invest Ophthalmol Vis
Lorenzo P, Aspberg A, Onnerfjord P, Bayliss MT, Neame PJ, Heinegard Sci 47:35653572
D (2001) Identification and characterization of asporin. a novel Muragaki Y, Nishimura I, Henney A, Ninomiya Y, Olsen BR (1990)
member of the leucine-rich repeat protein family closely related to The alpha 1 (IX) collagen gene gives rise to two different transcripts
decorin and biglycan. J Biol Chem 276:1220112211 in both mouse embryonic and human fetal RNA. Proc Natl Acad Sci
Majava M, Bishop PN, Hagg P, Scott PG, Rice A, Inglehearn C, USA 87:24002404
Hammond CJ, Spector TD, Ala-Kokko L, Mannikko M (2007) Murdoch AD, Iozzo RV (1993) Perlecan: the multidomain heparan sul-
Novel mutations in the small leucine-rich repeat protein/proteogly- phate proteoglycan of basement membrane and extracellular matrix.
can (SLRP) genes in high myopia. Hum Mutat 28:336344 Virchows Arch A Pathol Anat Histopathol 423:237242
Mansson B, Wenglen C, Morgelin M, Saxne T, Heinegard D (2001) Murdoch AD, Dodge GR, Cohen I, Tuan RS, Iozzo RV (1992) Primary
Association of chondroadherin with collagen type II. J Biol Chem structure of the human heparan sulfate proteoglycan from basement
276:3288332888 membrane (HSPG2/perlecan). A chimeric molecule with multiple
Marcelino J, Carpten JD, Suwairi WM, Gutierrez OM, Schwartz S, domains homologous to the low density lipoprotein receptor, lami-
Robbins C, Sood R, Makalowska I, Baxevanis A, Johnstone B, nin, neural cell adhesion molecules, and epidermal growth factor.
Laxer RM, Zemel L, Kim CA, Herd JK, Ihle J, Williams C, Johnson J Biol Chem 267:85448557
M, Raman V, Alonso LG, Brunoni D, Gerstein A, Papadopoulos N, Nareyeck G, Seidler DG, Troyer D, Rauterberg J, Kresse H, Schonherr
Bahabri SA, Trent JM, Warman ML (1999) CACP, encoding a E (2004) Differential interactions of decorin and decorin mutants
secreted proteoglycan, is mutated in camptodactyly-arthropathy- with type I and type VI collagens. Eur J Biochem 271:33893398
coxa vara-pericarditis syndrome. Nat Genet 23:319322 Naso MF, Zimmermann DR, Iozzo RV (1994) Characterization of the
Margolis RU, Margolis RK (1994) Aggrecan-versican-neurocan family complete genomic structure of the human versican gene and func-
proteoglycans. Methods Enzymol 245:105126 tional analysis of its promoter. J Biol Chem 269:3299933008
McBride OW, Fisher LW, Young MF (1990) Localization of PGI (bigly- Neame PJ, Barry FP (1993) The link proteins. Experientia 49:393402
can, BGN) and PGII (decorin, DCN, PG-40) genes on human chro- Neame PJ, Sommarin Y, Boynton RE, Heinegard D (1994) The struc-
mosomes Xq13-qter and 12q, respectively. Genomics 6:219225 ture of a 38-kDa leucine-rich protein (chondroadherin) isolated
Melrose J, Ghosh P, Taylor TK (2001) A comparative analysis of the from bovine cartilage. J Biol Chem 269:2154721554
differential spatial and temporal distributions of the large (aggrecan, Neame PJ, Kay CJ, McQuillan DJ, Beales MP, Hassell JR (2000)
versican) and small (decorin, biglycan, fibromodulin) proteoglycans Independent modulation of collagen fibrillogenesis by decorin and
of the intervertebral disc. J Anat 198:315 lumican. Cell Mol Life Sci 57:859863
Melrose J, Roughley P, Knox S, Smith S, Lord M, Whitelock J (2006) Nilsson B, De Luca S, Lohmander S, Hascall VC (1982) Structures of
The structure, location, and function of perlecan, a prominent peri- N-linked and O-linked oligosaccharides on proteoglycan monomer
cellular proteoglycan of fetal, postnatal, and mature hyaline carti- isolated from the Swarm rat chondrosarcoma. J Biol Chem
lages. J Biol Chem 281:3690536914 257:1092010927
Melrose J, Smith SM, Fuller ES, Young AA, Roughley PJ, Dart A, Noonan DM, Fulle A, Valente P, Cai S, Horigan E, Sasaki M, Yamada Y,
Little CB (2007) Biglycan and fibromodulin fragmentation corre- Hassell JR (1991) The complete sequence of perlecan, a basement
lates with temporal and spatial annular remodelling in experimen- membrane heparan sulfate proteoglycan, reveals extensive similarity
tally injured ovine intervertebral discs. Eur Spine J 16:21932205 with laminin A chain, low density lipoprotein-receptor, and the neu-
Melrose J, Fuller ES, Roughley PJ, Smith MM, Kerr B, Hughes CE, ral cell adhesion molecule. J Biol Chem 266:2293922947
Caterson B, Little CB (2008a) Fragmentation of decorin, biglycan, Oldberg A, Antonsson P, Lindblom K, Heinegard D (1989) A collagen-
lumican and keratocan is elevated in degenerate human meniscus, binding 59-kd protein (fibromodulin) is structurally related to the
knee and hip articular cartilages compared with age-matched mac- small interstitial proteoglycans PG-S1 and PG-S2 (decorin). EMBO
roscopically normal and control tissues. Arthritis Res Ther J 8:26012604
10:R79 Olin AI, Morgelin M, Sasaki T, Timpl R, Heinegard D, Aspberg A
Melrose J, Hayes AJ, Whitelock JM, Little CB (2008b) Perlecan, the (2001) The proteoglycans aggrecan and Versican form networks
jack of all trades proteoglycan of cartilaginous weight-bearing with fibulin-2 through their lectin domain binding. J Biol Chem
connective tissues. Bioessays 30:457469 276:12531261
Onnerfjord P, Heathfield TF, Heinegard D (2004) Identification of cytes of the superficial zone of articular cartilage. Arch Biochem
tyrosine sulfation in extracellular leucine-rich repeat proteins using Biophys 311:144152
mass spectrometry. J Biol Chem 279:2633 Scott JE (1996) Proteodermatan and proteokeratan sulfate (decorin,
Orgel JP, Eid A, Antipova O, Bella J, Scott JE (2009) Decorin core lumican/fibromodulin) proteins are horseshoe shaped. Implications
protein (decoron) shape complements collagen fibril surface struc- for their interactions with collagen. Biochemistry 35:87958799
ture and mediates its binding. PLoS One 4:e7028 Scott JE (2003) Elasticity in extracellular matrix shape modules of
Parsi MK, Adams JR, Whitelock J, Gibson MA (2010) LTBP-2 has mul- tendon, cartilage, etc. A sliding proteoglycan-filament model.
tiple heparin/heparan sulfate binding sites. Matrix Biol 29:393401 J Physiol 553:335343
Plaas AH, Neame PJ, Nivens CM, Reiss L (1990) Identification of the Scott JE, Stockwell RA (2006) Cartilage elasticity resides in shape
keratan sulfate attachment sites on bovine fibromodulin. J Biol module decoran and aggrecan sumps of damping fluid: implications
Chem 265:2063420640 in osteoarthrosis. J Physiol 574:643650
Rhee DK, Marcelino J, Al-Mayouf S, Schelling DK, Bartels CF, Cui Y, Scott IC, Imamura Y, Pappano WN, Troedel JM, Recklies AD, Roughley
Laxer R, Goldbach-Mansky R, Warman ML (2005a) Consequences PJ, Greenspan DS (2000) Bone morphogenetic protein-1 processes
of disease-causing mutations on lubricin protein synthesis, probiglycan. J Biol Chem 275:3050430511
secretion, and post-translational processing. J Biol Chem 280: Scott PG, McEwan PA, Dodd CM, Bergmann EM, Bishop PN, Bella J
3132531332 (2004) Crystal structure of the dimeric protein core of decorin, the
Rhee DK, Marcelino J, Baker M, Gong Y, Smits P, Lefebvre V, Jay GD, archetypal small leucine-rich repeat proteoglycan. Proc Natl Acad
Stewart M, Wang H, Warman ML, Carpten JD (2005b) The secreted Sci USA 101:1563315638
glycoprotein lubricin protects cartilage surfaces and inhibits syn- Shen Z, Gantcheva S, Mansson B, Heinegard D, Sommarin Y (1998)
ovial cell overgrowth. J Clin Invest 115:622631 Chondroadherin expression changes in skeletal development.
Ricciardelli C, Sakko AJ, Ween MP, Russell DL, Horsfall DJ (2009) Biochem J 330(Pt 1):549557
The biological role and regulation of versican levels in cancer. Shi S, Ciurli C, Cartman A, Pidoux I, Poole AR, Zhang Y (2003)
Cancer Metastasis Rev 28:233245 Experimental immunity to the G1 domain of the proteoglycan ver-
Roberts CR, Mort JS, Roughley PJ (1987) Treatment of cartilage pro- sican induces spondylitis and sacroiliitis, of a kind seen in human
teoglycan aggregate with hydrogen peroxide. Relationship between spondylarthropathies. Arthritis Rheum 48:29032915
observed degradation products and those that occur naturally during Shi S, Grothe S, Zhang Y, OConnor-McCourt MD, Poole AR, Roughley
aging. Biochem J 247:349357 PJ, Mort JS (2004) Link protein has greater affinity for versican than
Roberts S, Caterson B, Menage J, Evans EH, Jaffray DC, Eisenstein aggrecan. J Biol Chem 279:1206012066
SM (2000) Matrix metalloproteinases and aggrecanase: their role in Shine KM, Spector M (2008) The presence and distribution of lubricin
disorders of the human intervertebral disc. Spine (Phila Pa 1976) in the caprine intervertebral disc. J Orthop Res 26:13981406
25:30053013 Shine KM, Simson JA, Spector M (2009) Lubricin distribution in the
Rossi M, Morita H, Sormunen R, Airenne S, Kreivi M, Wang L, Fukai human intervertebral disc. J Bone Joint Surg Am 91:22052212
N, Olsen BR, Tryggvason K, Soininen R (2003) Heparan sulfate Sivan SS, Tsitron E, Wachtel E, Roughley PJ, Sakkee N, van der Ham
chains of perlecan are indispensable in the lens capsule but not in F, DeGroot J, Roberts S, Maroudas A (2006) Aggrecan turnover in
the kidney. EMBO J 22:236245 human intervertebral disc as determined by the racemization of
Roughley PJ (2004) Biology of intervertebral disc aging and degenera- aspartic acid. J Biol Chem 281:1300913014
tion: involvement of the extracellular matrix. Spine (Phila Pa 1976) Smith SM, Whitelock JM, Iozzo RV, Little CB, Melrose J (2009)
29:26912699 Topographical variation in the distributions of versican, aggrecan
Roughley PJ (2006) The structure and function of cartilage proteogly- and perlecan in the foetal human spine reflects their diverse func-
cans. Eur Cell Mater 12:92101 tional roles in spinal development. Histochem Cell Biol 132:
Roughley PJ, White RJ (1980) Age-related changes in the structure of 491503
the proteoglycan subunits from human articular cartilage. J Biol Smith SM, Shu C, Melrose J (2010) Comparative immunolocalisation
Chem 255:217224 of perlecan with collagen II and aggrecan in human foetal, newborn
Roughley PJ, White RJ (1989) Dermatan sulphate proteoglycans of and adult ovine joint tissues demonstrates perlecan as an early
human articular cartilage. The properties of dermatan sulphate pro- developmental chondrogenic marker. Histochem Cell Biol
teoglycans I and II. Biochem J 262:823827 134:251263
Roughley PJ, White RJ, Cs-Szabo G, Mort JS (1996a) Changes with Song YQ, Cheung KM, Ho DW, Poon SC, Chiba K, Kawaguchi Y,
age in the structure of fibromodulin in human articular cartilage. Hirose Y, Alini M, Grad S, Yee AF, Leong JC, Luk KD, Yip SP,
Osteoarthritis Cartilage 4:153161 Karppinen J, Cheah KS, Sham P, Ikegawa S, Chan D (2008)
Roughley PJ, White RJ, Mort JS (1996b) Presence of pro-forms of Association of the asporin D14 allele with lumbar-disc degeneration
decorin and biglycan in human articular cartilage. Biochem J 318(Pt in Asians. Am J Hum Genet 82:744747
3):779784 Spicer AP, Joo A, Bowling RA Jr (2003) A hyaluronan binding link
Roughley PJ, Melching LI, Heathfield TF, Pearce RH, Mort JS (2006) protein gene family whose members are physically linked adjacent
The structure and degradation of aggrecan in human intervertebral to chondroitin sulfate proteoglycan core protein genes: the missing
disc. Eur Spine J 15(Suppl 3):S326S332 links. J Biol Chem 278:2108321091
Sandy JD, Flannery CR, Boynton RE, Neame PJ (1990) Isolation and Stallcup WB (2002) The NG2 proteoglycan: past insights and future
characterization of disulfide-bonded peptides from the three globu- prospects. J Neurocytol 31:423435
lar domains of aggregating cartilage proteoglycan. J Biol Chem Superti-Furga A, Hastbacka J, Rossi A, van der Harten JJ, Wilcox WR,
265:2110821113 Cohn DH, Rimoin DL, Steinmann B, Lander ES, Gitzelmann R
Sandy JD, Westling J, Kenagy RD, Iruela-Arispe ML, Verscharen C, (1996) A family of chondrodysplasias caused by mutations in the
Rodriguez-Mazaneque JC, Zimmermann DR, Lemire JM, Fischer diastrophic dysplasia sulfate transporter gene and associated with
JW, Wight TN, Clowes AW (2001) Versican V1 proteolysis in impaired sulfation of proteoglycans. Ann N Y Acad Sci
human aorta in vivo occurs at the Glu441-Ala442 bond, a site that is 785:195201
cleaved by recombinant ADAMTS-1 and ADAMTS-4. J Biol Chem Svensson L, Aszodi A, Reinholt FP, Fassler R, Heinegard D, Oldberg A
276:1337213378 (1999) Fibromodulin-null mice have abnormal collagen fibrils, tis-
Schumacher BL, Block JA, Schmid TM, Aydelotte MB, Kuettner KE sue organization, and altered lumican deposition in tendon. J Biol
(1994) A novel proteoglycan synthesized and secreted by chondro- Chem 274:96369647
Svensson L, Narlid I, Oldberg A (2000) Fibromodulin and lumican Vlodavsky I, Friedmann Y, Elkin M, Aingorn H, Atzmon R, Ishai-
bind to the same region on collagen type I fibrils. FEBS Lett Michaeli R, Bitan M, Pappo O, Peretz T, Michal I, Spector L, Pecker
470:178182 I (1999) Mammalian heparanase: gene cloning, expression and
Swann DA, Sotman S, Dixon M, Brooks C (1977) The isolation and function in tumor progression and metastasis. Nat Med 5:793802
partial characterization of the major glycoprotein (LGP-I) from the von Marschall Z, Fisher LW (2010) Decorin is processed by three iso-
articular lubricating fraction from bovine synovial fluid. Biochem J forms of bone morphogenetic protein-1 (BMP1). Biochem Biophys
161:473485 Res Commun 391:13741378
Swann DA, Hendren RB, Radin EL, Sotman SL, Duda EA (1981a) The Wang IJ, Chiang TH, Shih YF, Hsiao CK, Lu SC, Hou YC, Lin LL
lubricating activity of synovial fluid glycoproteins. Arthritis Rheum (2006) The association of single nucleotide polymorphisms in the
24:2230 5-regulatory region of the lumican gene with susceptibility to high
Swann DA, Slayter HS, Silver FH (1981b) The molecular structure of myopia in Taiwan. Mol Vis 12:852857
lubricating glycoprotein-I, the boundary lubricant for articular car- Wang J, Markova D, Anderson DG, Zheng Z, Shapiro IM, Risbud MV
tilage. J Biol Chem 256:59215925 (2011) TNF-alpha and IL-1beta promote a disintegrin-like and met-
Sztrolovics R, Chen XN, Grover J, Roughley PJ, Korenberg JR (1994) alloprotease with thrombospondin type I motif-5-mediated aggre-
Localization of the human fibromodulin gene (FMOD) to chromo- can degradation through syndecan-4 in intervertebral disc. J Biol
some 1q32 and completion of the cDNA sequence. Genomics Chem 286:3973839749
23:715717 Watanabe H, Kimata K, Line S, Strong D, Gao LY, Kozak CA, Yamada
Sztrolovics R, Alini M, Roughley PJ, Mort JS (1997) Aggrecan degra- Y (1994) Mouse cartilage matrix deficiency (cmd) caused by a 7 bp
dation in human intervertebral disc and articular cartilage. Biochem deletion in the aggrecan gene. Nat Genet 7:154157
J 326(Pt 1):235241 Watanabe H, Cheung SC, Itano N, Kimata K, Yamada Y (1997)
Sztrolovics R, Grover J, Cs-Szabo G, Shi SL, Zhang Y, Mort JS, Identification of hyaluronan-binding domains of aggrecan. J Biol
Roughley PJ (2002) The characterization of versican and its mes- Chem 272:2805728065
sage in human articular cartilage and intervertebral disc. J Orthop Watanabe H, Yamada Y, Kimata K (1998) Roles of aggrecan, a large
Res 20:257266 chondroitin sulfate proteoglycan, in cartilage structure and function.
Tillgren V, Onnerfjord P, Haglund L, Heinegard D (2009) The tyrosine J Biochem 124:687693
sulfate-rich domains of the LRR proteins fibromodulin and Whitelock JM, Murdoch AD, Iozzo RV, Underwood PA (1996) The
osteoadherin bind motifs of basic clusters in a variety of heparin- degradation of human endothelial cell-derived perlecan and release
binding proteins, including bioactive factors. J Biol Chem 284: of bound basic fibroblast growth factor by stromelysin, collagenase,
2854328553 plasmin, and heparanases. J Biol Chem 271:1007910086
Tkachenko E, Rhodes JM, Simons M (2005) Syndecans: new kids on Whitelock JM, Melrose J, Iozzo RV (2008) Diverse cell signaling
the signaling block. Circ Res 96:488500 events modulated by perlecan. Biochemistry 47:1117411183
Tomoeda M, Yamada S, Shirai H, Ozawa Y, Yanagita M, Murakami S Wiberg C, Heinegard D, Wenglen C, Timpl R, Morgelin M (2002)
(2008) PLAP-1/asporin inhibits activation of BMP receptor via its Biglycan organizes collagen VI into hexagonal-like networks
leucine-rich repeat motif. Biochem Biophys Res Commun resembling tissue structures. J Biol Chem 277:4912049126
371:191196 Xu T, Bianco P, Fisher LW, Longenecker G, Smith E, Goldstein S,
Tompson SW, Merriman B, Funari VA, Fresquet M, Lachman RS, Bonadio J, Boskey A, Heegaard AM, Sommer B, Satomura K,
Rimoin DL, Nelson SF, Briggs MD, Cohn DH, Krakow D (2009) A Dominguez P, Zhao C, Kulkarni AB, Robey PG, Young MF (1998)
recessive skeletal dysplasia, SEMD aggrecan type, results from a Targeted disruption of the biglycan gene leads to an osteoporosis-
missense mutation affecting the C-type lectin domain of aggrecan. like phenotype in mice. Nat Genet 20:7882
Am J Hum Genet 84:7279 Yamada S, Tomoeda M, Ozawa Y, Yoneda S, Terashima Y, Ikezawa K,
Tortorella MD, Liu RQ, Burn T, Newton RC, Arner E (2002) Ikegawa S, Saito M, Toyosawa S, Murakami S (2007) PLAP-1/
Characterization of human aggrecanase 2 (ADAM-TS5): substrate asporin, a novel negative regulator of periodontal ligament mineral-
specificity studies and comparison with aggrecanase 1 (ADAM-TS4). ization. J Biol Chem 282:2307023080
Matrix Biol 21:499511 Yang BL, Yang BB, Erwin M, Ang LC, Finkelstein J, Yee AJ (2003) Versican
Tsai TT, Danielson KG, Guttapalli A, Oguz E, Albert TJ, Shapiro IM, G3 domain enhances cellular adhesion and proliferation of bovine
Risbud MV (2006) TonEBP/OREBP is a regulator of nucleus intervertebral disc cells cultured in vitro. Life Sci 73:33993413
pulposus cell function and survival in the intervertebral disc. J Biol Young AA, Smith MM, Smith SM, Cake MA, Ghosh P, Read RA,
Chem 281:2541625424 Melrose J, Sonnabend DH, Roughley PJ, Little CB (2005) Regional
Turnbull JE (2010) Heparan sulfate glycomics: towards systems biol- assessment of articular cartilage gene expression and small proteo-
ogy strategies. Biochem Soc Trans 38:13561360 glycan metabolism in an animal model of osteoarthritis. Arthritis
Valhmu WB, Palmer GD, Rivers PA, Ebara S, Cheng JF, Fischer S, Res Ther 7:R852R861
Ratcliffe A (1995) Structure of the human aggrecan gene: exon- Zheng J, Luo W, Tanzer ML (1998) Aggrecan synthesis and secretion.
intron organization and association with the protein domains. A paradigm for molecular and cellular coordination of multiglobu-
Biochem J 309(Pt 2):535542 lar protein folding and intracellular trafficking. J Biol Chem
Vehvilainen P, Hyytiainen M, Keski-Oja J (2009) Matrix association of 273:1299913006
latent TGF-beta binding protein-2 (LTBP-2) is dependent on Zimmermann DR, Ruoslahti E (1989) Multiple domains of the large
fibrillin-1. J Cell Physiol 221:586593 fibroblast proteoglycan, versican. EMBO J 8:29752981
Collagen and Other Proteins of the
Nucleus Pulposus, Annulus Fibrosus, 5
and Cartilage End Plates
Fackson Mwale
Contents 5.8 Collagen-Related Diseases .............................................. 88

5.8.1 Hereditary Disorders ......................................................... 88
5.1 Introduction ..................................................................... 79 5.8.2 Other Collagen-Related Diseases...................................... 89
5.1.1 Evolutionary Considerations ............................................. 79
5.1.2 Brief Overview of the Anatomical Characteristics 5.9 Summary of Critical Concepts Discussed
of the Intervertebral Disc .................................................. 79 in the Chapter .................................................................. 89
5.2 Protein Composition of the Intervertebral Discs ......... 80 References ...................................................................................... 90
5.2.1 Collagenous Protein Structure .......................................... 80
5.2.2 Classification of Collagens................................................ 82
5.2.3 Other Matrix Molecules .................................................... 84
5.1 Introduction
5.3 Biosynthesis of Collagen Proteins .................................. 85
5.3.1 Intracellular Processing
and Posttranslational Modifications .................................. 85 5.1.1 Evolutionary Considerations
5.3.2 Chaperone-Assisted Folding of Individual Chains
into a Triple-Helical Structure .......................................... 85 Characterized by the presence of at least one triple-helical
5.3.3 Mechanisms of Collagen Chain Selection,
domain, members of the collagen family are the most abun-
Intracellular Transport, and Secretion ............................... 85
dant proteins in the animal kingdom. From a phylogenetic
5.4 Extracellular Processing perspective, while the triple-helical domain is present in bac-
of Collagens/Procollagens ............................................... 87
5.4.1 Enzymes Involved in Processing of Procollagens............. 87 teria and fungi and even some viruses, collagen and colla-
5.4.2 Procollagen N and Procollagen C Proteinases .................. 87 gen-like proteins have been identified in all metazoa. With
5.5 Self-Assembly of Collagen Fibrils.................................. 87
the appearance of the phylum Chordata, the notochord pro-
5.5.1 Homotypic CollagenCollagen Interactions ..................... 87 vided a first skeleton which enabled the organism to
5.5.2 Heterotypic CollagenCollagen assume a longitudinal shape and provided support for the
and CollagenOther Molecules ......................................... 87 digestive tube and the nerve cord. This notochordal structure
5.5.3 Fibril Formation ................................................................ 87
5.5.4 Cross-Link Formation ....................................................... 87
is sheathed in collagen; postembryonic remnants of the noto-
chord in vertebrates form the nucleus pulposus of the inter-
5.6 Collagen FibrilCell Interactions .................................. 88 vertebral discs. Zhang et al. (2009) have put forward the
5.6.1 CollagenIntegrin Binding ................................................ 88
5.6.2 CollagenDiscoidin Domain Receptor hypothesis that the vertebrate chondrocytes that all express
(DDR) Binding.................................................................. 88 the type II gene may have evolved from notochordal cells. In
5.6.3 Integrin-Dependent and DDR-Dependent Signaling ........ 88 vertebrates, collagen fibrils formed the template for deposi-
5.7 Homeostasis of Collagenous Matrices ........................... 88 tion of mineral and the development of bone and cartilage of
5.7.1 Degeneration of Collagenous Matrices ............................. 88 the axial and appendicular skeleton.
5.7.2 Regulation of Collagen Expression................................... 88
5.1.2 Brief Overview of the Anatomical

Characteristics of the Intervertebral Disc
F. Mwale The discs typically are composed of three morphologically

Division of Orthopaedic Surgery,
distinct major structural regions the peripheral collagen-
Lady Davis Institute for Medical Research, McGill University,
3755 Chemin de la Cote Ste-Catherine, Montreal, rich annulus fibrosus, surrounding the proteoglycan-rich
QC H3T 1E2, Canada central nucleus pulposus, and the cartilage end plates

80 F. Mwale
interfacing the disc and the vertebral body. The junctional of the collagen fibrils in the annulus are retaining the nucleus
zone between vertebral bone and annulus fibrosus is a type of and taking up and distributing the load exerted by the disc.
enthesis. The cartilage end plates in the juvenile are respon-
sible for longitudinal vertebral growth. In the adult, the inter- 5.1.2.2 Nucleus Pulposus
vertebral discs are avascular and receive most of their The nucleus pulposus is a soft jellylike, highly hydrophilic
nutrients from the vertebral bone vasculature by diffusion tissue occupying the central region of the disc. The disc con-
through the cartilage end plates. Collagens provide the struc- tains proteoglycans (predominantly aggrecan; see Chap. 4),
tural framework of the intervertebral discs and are responsi- randomly organized fibrillar collagens (Inoue 1981), radially
ble for its biomechanical properties such as torsion and arranged elastin fibers (Yu et al. 2002), and water. The pro-
resistance to pressure or tension. Proteoglycans, such as portion and organization of water, fibrillar collagens, and
decorin, fibromodulin, and biglycan together with other proteoglycans vary not only with the position across the disc,
matrix constituents, have important influences on collagen but with age and level (Antoniou et al. 1996; Scott et al.
fibril formation. 1994; Demers et al. 2004; Mwale et al. 2004; Antoniou et al.
1996). The nucleus pulposus has a higher concentration of
proteoglycans and water than other regions of the disc,
Box 5.1 Abbreviation List whereas the highest levels of collagen are in the outer annu-
C-NC domain Carboxyl-terminal noncollagenous lus and the lowest concentration in the nucleus (Mwale et al.
domain (same as C-propeptide) 2004; Inkinen et al. 1998). The collagen content of the
COMP Cartilage oligomeric matrix protein nucleus pulposus is greatest in cervical discs and lowest in
DDR Discoidin domain receptors lumbar discs. In contrast, the proteoglycan content of the
ECM Extracellular matrix nucleus pulposus peaks in lumbar discs and is lowest in cer-
ED-A Extra domain A vical discs (Scott et al. 1994).
EGF Epidermal growth factor
ER Endoplasmic reticulum 5.1.2.3 Cartilage End Plates
FACIT Fibril-associated collagen with inter- The end plates in the human disc consist of a thin horizontal
rupted triple helices layer of hyaline cartilage, usually less than 1 mm thick, which
GAG Glycosaminoglycan separate the nucleus pulposus and annulus fibrosus from the
IVDD Intervertebral disc degeneration adjacent vertebral bone. In the adult, the cartilage end plate is
MACIT Membrane-associated collagen with narrow, and often calcified, leading to disturbances to the
interrupted triple helices nutrient supply to the nucleus pulposus (Maroudas et al. 1975;
MMP Matrix metalloproteinases Nachemson et al. 1970). The collagen fibers within the end
Multiplexin Multiple triple-helix domains and plates run parallel to the vertebral bodies, with the fibers con-
interruptions tinuing into the disc (Roberts et al. 1989).
NC Noncollagenous
N-NC domain Amino-terminal noncollagenous
domain (same as N-propeptide) 5.2 Protein Composition of the
RER Rough endoplasmic reticulum Intervertebral Discs
RUNX2 Runt-related transcription factor 2
TIMP Tissue inhibitors of metalloproteinases 5.2.1 Collagenous Protein Structure
Collagen, the most abundant fibrous protein of the disc, is the

name for a family of rope-like polypeptidic molecules
5.1.2.1 Annulus Fibrosus (Fig. 5.1) that have a linear structure containing regions of
The annulus fibrosus makes up the peripheral portion of the repeating triplets of amino acids (GlyaaXaaYaa) with glycine
disc structure (Fig. 5.1). In the mature lumbar disc, there are (Gaa) at every third position and X and Y often being proline
up to 25 lamellae that comprise regular concentric bundles of and hydroxyproline, respectively. These two amino acids
parallel collagen fibers arranged around the central gelati- constitute about 1/6 with glycine accounting for 1/3 of the
nous nucleus pulposus (Roberts 2002). Collagen fibers in a sequence. Since glycine is the smallest amino acid with no
particular lamella run in one direction, while fibers in an side chain, it is positioned at every third position in the chain.
adjacent lamella run in the opposite direction. The thickness The high glycine content stabilizes the collagen helix, facili-
of the lamellae varies from 200 to 400 mm, increasing from tating hydrogen bonding and the formation of intermolecular
inside to outside (Inoue 1973). This alternating pattern is cross-links (Eyre et al. 1984; Pokharna and Phillips 1998;
designed to withstand torsional stresses. The key functions Berg and Prockop 1973). The ring structures of proline and
5 Collagen and Other Proteins of the Nucleus Pulposus, Annulus Fibrosus, and Cartilage End Plates 81
AF parallel organized fibrillar collagen
AF
NP random organized fibrillar collagen
Fig. 5.1 Schematic illustration of the organization of fibrillar collagens in the annulus fibrosus (AF) and nucleus pulposus (NP) of a human inter-
vertebral disc
hydroxyproline occupy the outer positions where they inter- the major collagens, since they comprise 98 % of the total
act with neighboring collagen molecules. The hydroxyl connective tissue protein. There is considerable complexity
groups of 4-hydroxyproline are essential for the formation of and diversity in the 28 different types of collagen.
intramolecular hydrogen bonds and determine the thermal In the lumbar disc, there are up to 25 lamellae that com-
stability of each collagen chain (Berg and Prockop 1973). prise regular concentric bundles of parallel collagen fibers
The collagen molecule is formed from three a-chains arranged around the central gelatinous nucleus pulposus
wound together in a triple helix. These helices are twisted (Roberts 2002). Collagen fibers in a particular lamella run
together into a right-handed coiled coil stabilized by hydro- in one direction, while fibers in an adjacent lamella run in
gen bonds and woven together to form a left-handed helix the opposite direction. The thickness of the lamellae varies
(Hulmes and Miller 1981). This uninterrupted triple-helix from 200 to 400 mm, increasing from inside to outside
molecule is approximately 300 nm in length and 1.5 nm in (Inoue 1973). This alternating pattern is designed to with-
diameter a cable-like structure that provides significant stand torsional stresses. The key functions of the collagen
tensile strength to skin, tendons, ligaments, and other carti- fibrils in the annulus are retaining the nucleus and taking
laginous structures, making them tough and flexible. The up and distributing the load exerted by the disc. The fibril-
amino acid hydroxylysine is responsible for stabilizing the forming collagens I and II form an important network in
side-to-side packing of the chains into fibrils. The collagen the disc. Collagen I forms hybrids with other fibrillar col-
molecule contains three structural domains, the amino- and lagens, particularly collagen V. Co-fibril formation with
carboxyl-terminal extra-helical regions and the major triple- collagen I is believed to regulate the diameter of fibrils
helical rodlike domain. Collagens IV are considered to be because of the partial processing of the N-propeptide of
82 F. Mwale
collagen V. With disc aging and degeneration, there are

profound alterations in collagen cross-links (Pokharna and Box 5.2 Collagen Types Present in the Intervertebral
Phillips 1998). Disc
The proportion of collagen II and I varies gradually and Fibril forming
inversely across the disc, with exclusively collagen I at the Type I (COL1 a1, COL1a2)
extreme outermost layers of the annulus fibrosus and the Type II (COL2a1)
nucleus pulposus possessing mostly collagen type II (Eyre Type III
and Muir 1976). Under physiological conditions, collagen II Type V (COL5a1, COL5a2, COL5a3)
fibrils contain more water than I fibrils (Grynpas et al. 1980). FACIT (fibril-associated collagen with interrupted triple
Collagen V is a minor component and forms hybrids with helices)
Type IX (COL9a1, COL9a2, COL9a3)
collagen I, while XI forms hybrids with collagen II.
Type XII (COL12a1)
The collagen chains are synthesized as procollagens. In
Basement membrane (basal membrane)
the rough endoplasmic reticulum (ER), the procollagen chain
Type IV (COL4a1, COL4a2, COL4a3, COL4a4,
undergoes a series of processing reactions. First, as with COL4a5, COL4a6)
other secreted proteins, glycosylation of procollagen occurs Multiplexin
in the rough ER and Golgi complex. Galactose and glucose None
residues are added to hydroxylysine residues, and long oli- Other
gosaccharides are added to specific asparagine residues in Type VI (COL6a1,COL6a2, COL6a3, COL6a4)
the carboxyl-terminal propeptide, a segment at the carboxyl- Type X (COL10a1, COL10a2, COL10a3)
terminus of a procollagen molecule that is absent from Type XI (COL11a1, COL11a2)
mature collagen. Generally, the propeptides (amino and car-
boxyl terminal) are removed after secretion, and then colla-
gen fibrils form in the extracellular space.
5.2.2.2 Collagen II
Collagen II fibrils in the nucleus pulposus appear to be orga-
5.2.2 Classication of Collagens nized randomly (Fig. 5.1). In the adult, these collagen mole-
cules are joined by hydroxypyridium cross-links (Eyre and
The disc contains many different collagen types whose abun- Muir 1976). There is evidence that procollagen II can be
dance changes with age (Roughley 2004). The annulus expressed in two forms by alternative splicing of the primary
fibrosus is composed primarily of collagen I and to a lesser gene transcript. The two mRNAs either include (IIA) or
extent II, III, V, VI, IX, XI, XII, and XIV. The nucleus pulpo- exclude (IIB) exon 2, so that procollagen II can be synthe-
sus is rich in collagen II, but it also contains collagen I, VI, sized with or without exon 2 encoding the major portion of
and IX (Eyre and Muir 1976, 1977; Wu et al. 1987; Eyre the amino propeptide (Sandell et al. 1991; Ryan and Sandell
et al. 2002). Collagens I and II constitute about 80 % of the 1990). This polymorphism is thought to influence cell mor-
collagens in the disc (Eyre and Muir 1977) (for details of the phology, with the cells expressing procollagen IIA being
collagen species, see Box 5.2). narrow, elongated, and fibroblastic in appearance, while
the cells expressing procollagen IIB are large and round. The
5.2.2.1 Collagen I expression of procollagen IIB appears to be correlated with
The structure of procollagen I is similar to other fibrillar col- abundant synthesis and accumulation of aggrecan (Sandell
lagens, and it comprises three polypeptide a-chains, which et al. 1991). Procollagen IIA may function in the annulus
form a unique triple-helical structure (Fig. 5.2). It is a het- fibrosus and nucleus pulposus, particularly during
erotrimer of two a1 and one a2 chains. It contains an unin- development.
terrupted triple-helical domain flanked by short non-helical
telopeptides. The telopeptides, which do not have a repeating 5.2.2.3 Collagen VI
GlyXY structure and do not adopt a triple-helical confor- Collagen VI accounts for 1020 % of the total collagen and is
mation, account for 2 % of the molecule. Many macromole- particularly prominent in the nucleus pulposus and the end
cules such as COMP, fibromodulin, matrilin, and decorin plate cartilage (Roberts et al. 1991). It resembles collagens IX
attach to collagen I (Fig. 5.3). Collagen I molecules form and X in that it is a shorter chain collagen than collagen II. It
D-periodic (D = 67 nm, the characteristic axial periodicity of consists of a-chains that form a highly branched filamentous
collagen) cross-striated fibrils in the extracellular space, giv- network based on the formation of tetramers but does not
ing the tissue its mechanical strength and providing the major appear to form cross-links with other matrix molecules (Wu
biomechanical scaffold for cell attachment and anchorage of et al. 1987). The beaded filaments of collagen VI represent
macromolecules (Fig. 5.4). another important network in the disc matrix (Feng et al. 2006).
Fig. 5.2 Schematic illustration

11 chain
of type 1 collagen a-chains,
recognition site in the C-NC
domain, and the crystal structure
of the collagen intertwined
a-chains (the upper image was 12 chain
generated in 3D Studio Max,
while the lower image was
generated in CN3D ver. 4.3 using
data for Crystal Structure of the
Collagen Triple Helix Model
[(Pro-Pro-Gly)10]3 from the
Protein Data Bank rcsb.org
2 chain
DOI:10.2210/pdb1k6f/pdb)
Yellow area corresponds

to the putative recognition
site
Crystal structure of the

collagen triple helix model
Cross-striated collagen fibrils

Collagen fibril II are anchored by COMP
macromolecules
COMP
Side-by-side cross-linking
of collagen fibrils form
collagen fibres
Fibromodulin
Microscopic aspect of
parallel and perpendicular
collagen fibrils
Matrilin
Decorin
Fig. 5.4 Computer visualization model, schematics, and simulated

microscopic images of normal collagen fibrils showing the association
with COMP, side-by-side cross-linking of collagen fibrils, and micro-
Fig. 5.3 Schematic illustration of assemblies of cross-striated collagen
scopic aspects of collagen fibrils
fibrils and cartilage oligomeric matrix protein (COMP), fibromodulin,
matrilin, and decorin in the intervertebral disc (The figure was modified
with permission from Feng et al. (2006)) end-to-end and side-to-side associations catalyzed by proteo-
glycans such as biglycan and decorin. These ligands (collagen
It forms a tetramer of two pairs of antiparallel collagen VI mol- VI tetramers) are also bound to molecules such as PRELP,
ecules arranged such that two N-terminal ends are exposed at fibronectin, and matrilin-1, -2, or -3, which in turn are bound to
either end of the unit. Further assembly occurs both by a collagen fiber, a procollagen molecule, or aggrecan.
84 F. Mwale
5.2.2.4 Collagen IX with advanced age and degenerative disc lesions (Nerlich
Collagen IX is a fibril-associated collagen with interrupted et al. 1997; Hristova et al. 2011). Aigner et al. (1998) studied
triple helices (FACIT) collagen. Other FACIT members are the variation in the pattern of collagen X expression with age
collagen XII, XIV, XVI, and XXI; when compared with colin normal human discs, particularly the cells of the inner
lagen IX, they are less characterized in terms of structure and annulus fibrosus and the nucleus pulposus. These workers
function. Collagen IX is usually found in tissues containing showed that with age some cells from the inner annulus or
collagen II. It is extensively cross-linked to fibrils of collagen the nucleus expressed a hypertrophic chondrocyte phenotype
II in an antiparallel orientation and may also be cross-linked and secreted collagen X as a component of the disc matrix.
to other collagen IX molecules (Eyre et al. 1988; van der
Rest and Mayne 1988; Wu et al. 1992). It has a periodic dis-
tribution along the fibril of approximately 67 nm (Vaughan 5.2.3 Other Matrix Molecules
et al. 1988). The globular NC4 domain at the N-terminus of
the a1(IX) chain extends out from the fibril (Vaughan et al. 5.2.3.1 Cartilage Oligomeric Matrix Protein (COMP)
1988) and may provide a molecular link between the fibrils COMP is a 524-kD protein composed of five identical glyco-
and other interfibrillar matrix components, such as the proprotein subunits each of 100120 kD held together by a five-
teoglycan aggrecan. The amino-terminal NC4 domain is stranded coiled-coil domain in the N-terminal portion and
very basic (Vasios et al. 1988; Muragaki et al. 1990). Some exposing unique C-terminal globular domains. Each subunit
forms of collagen IX may lack the NC4 domain due to the has EGF-like and calcium-binding (thrombospondin-like)
use of an alternative promoter (Nishimura et al. 1989), and domains. It is present in the extracellular matrix of the
the expression of this variant is thought to be tissue specific nucleus pulposus and annulus fibrosus (Ishii et al. 2006; Lee
and developmentally regulated. Collagen IX of the disc is et al. 2007). It exhibits a lamellar distribution pattern in the
distinct from that of hyaline cartilage in that the disc contains annulus fibrosus region. Higher nucleus pulposus levels of
only the short form of a1(IX) that lacks the NC4 domain COMP are found in aging discs (Lee et al. 2007). COMP
(Wu and Eyre 2003). Usage of the short a1(IX) transcript in plays a role in regulating collagen fibril assembly.
disc tissue has no apparent effect on cross-linking behavior.
The precise biological role(s) of collagen IX, its assembly in 5.2.3.2 Fibronectin
relationship to collagen II, and the mechanisms that regulate Synthesis of fibronectin has been demonstrated in healthy
its synthesis and degradation remain unknown. disc tissue (Hayes et al. 2001; Anderson et al. 2010) and car-
tilage (Wurster and Lust 1984). Fibronectin is a minor com-
5.2.2.5 Collagen X ponent of the disc of young and older animals (Hayes et al.
Collagen X is a non-fibrillar short-chain protein containing 2001). Its function in the discs is unknown, although it is
three identical a1 chains [a1(X)3] with a low molecular well established that fibronectin can play a role in cell
weight of 59 kDa (Schmid and Linsenmayer 1985). Two matrix, matrixmatrix interactions as well as binding colla-
exons encode the complete primary translation product, gen and heparan sulfate proteoglycans through RGD
which contains the N-terminal region (NC2). The protein is sequences. As a result of alternative splicing, different iso-
comprised of 52 amino acids, 18 of which form the signal forms of fibronectin exist. Disc cells synthesize fibronectin
peptide (LuValle et al. 1988). The collagen X molecule also with either or both the ED-B and ED-A domains present
contains a relatively larger (162 amino acids) noncollagenous (Anderson et al. 2010), the significance of which is unclear.
C-terminal domain (NC1) (Yamaguchi et al. 1989), which Fibronectin is elevated in degenerated discs, and its frag-
plays a key role in the intracellular assembly of the triple- ments induce the cell to degrade the matrix (Oegema et al.
helical collagen X molecules and may be necessary for aggre- 2000, Anderson et al. 2004).
gation to form extracellular networks. The presence of several
functional RUNX2 binding sites within the promoter region 5.2.3.3 Amyloid
of the COL10A1 genes suggests that it is a direct transcrip- Amyloid deposits in the intervertebral disc were first described
tional target of RUNX2 during chondrogenesis (Zheng et al. by Bywaters and Dorling (1970). In the annulus, amyloid
2003). It forms pericellular mats and is also closely associ- deposits are found lying between thick bundles of collagen
ated with the fibrils of collagen II (Linsenmayer et al. 1998). fibers, while a diffuse nodular distribution is found in the
Collagen X is synthesized by hypertrophic chondrocytes nucleus pulposus, in proximity to cells (Ladefoged 1985). The
during endochondral ossification in the growth plate (Mwale incidence of amyloid deposition in discs increases with advanc-
et al. 2000; Tchetina et al. 2003). It was shown to be present ing age, although it is also found in the young disc (Ladefoged
in human lumbar discs during aging and degeneration by 1985; Yasuma et al. 1992). Amyloid deposits of different mor-
Boos et al. (1997) and Xi et al. (2004). In addition, nucleus phological types have been observed in the disc (Mihara et al.
pulposus chondrocytes express collagen X in association 1994). At the ultrastructu ral level, amyloid deposits are seen
composed of 10-nm-wide nonbranching fibrils (Mihara et al. 5.3 Biosynthesis of Collagen Proteins
1994). Changes in matrix glycosaminoglycans, particularly
strongly sulfated glycosaminoglycans such as keratan sulfate, 5.3.1 Intracellular Processing and
may play a role in the pathogenesis of localized and systemic Posttranslational Modications
amyloid deposition (Athanasou et al. 1995).
The synthesis of collagen involves a cascade of unique post-
5.2.3.4 Tenascin translational modifications of the original procollagen poly-
Tenascin has also been called hexabrachion (Erickson and peptide. The many steps in collagen biosynthesis can be
Inglesias 1984) and myotendinous antigen (Chiquet and interrupted or changed by mutant enzymes or by disease pro-
Fambrough 1984). The tenascins are a family of extracellular cesses. With procollagen synthesis on the RER, there is hydrox-
matrix glycoproteins with repeated structural domains homol- ylation of proline and lysine, an initial glycosylation step and
ogous to epidermal growth factor (EGF), fibronectin type III, formation of triple helices. Hydroxylation begins after the pep-
and the fibrinogens. It has been shown that aggrecan can tide chain has reached a certain minimum length and is still
interact with certain matrix proteins containing EGF-repeats, bound to the ribosomes. The two enzymes involved are pepti-
including tenascins, the fibrillins, as well as the fibulins, (Day dyl prolyl hydroxylase and peptidyl lysyl hydroxylase.
et al. 2004) to form higher order networks. Tenascin is a large Glycosylation of lysine occurs after its hydroxylation. Each of
extended octopus-like molecule, in which six arms radiate the collagen isoforms has differing levels of carbohydrate in
out from a central point, with each arm having an Mr of the form of galactose or glycosylgalactose linked to hydroxy-
200 kDa and composed of a single polypeptide chain. The lysine. Terminal glycosylation takes place in the Golgi appara-
domain structure of the cloned molecule supports the finding tus, and the molecule is packaged into secretory vesicles and
that it can interact with multiple ligands (Nies et al. 1991). It secreted by exocytosis. Outside of the cell, procollagen pepti-
is a hemagglutinin and can bind, directly or indirectly, with a dase removes the non-helical domains of procollagen.
variety of disc matrix molecules. In discs, it is present in
young and adult annulus fibrosus and nucleus pulposus,
where it is confined to the pericellular matrix (Gruber et al. 5.3.2 Chaperone-Assisted Folding of Individual
2002, 2006) and its expression can be modulated by mechani- Chains into a Triple-Helical Structure
cal strain (Benjamin and Ralphs 2004). It is thought to have a
role in disc aging and degeneration, possibly by modulating Assembly begins with a-chains trimerization to form a tri-
fibronectin cell interactions and causing alterations in the ple-helical protomer. Protomers are usually heterotrimers,
shape of disc cells (Gruber et al. 2002). composed of up to three different a-chains. For example,
collagen I is composed of a1 and a2 chains, forming an
5.2.3.5 Elastin a1a1a2 heterotrimer. They have in common at least one
Earlier studies reported that the elastic fiber network of the triple-helical collagenous domain of varying length and two
disc was sparse and irregular. Thus, elastic fibers were gener- noncollagenous domains C-NC and C-NC that are positioned
ally considered to play no significant role in the mechanical at the N and C ends. Thus, the N-propeptide from the N-NC
functioning of the disc. However, recent studies have shown domain and the C-propeptide from the C-NC domain act as
that the network is highly organized and that the distribution molecular chaperones. The C-propeptide at the C-NC domain
and orientation of elastic fibers varies from region to region for both a1 and a2 chains contains five subdomains (Hulmes
(Yu et al. 2002, 2005). In the annulus fibrosus, elastin fibers 2002). They are involved in the protection of nascent pro-
appear densely distributed in the region between the lamellae teins on the ribosome; folding of newly synthesized or mem-
and also in bridges across the lamellae. Elastin molecules brane-translocated proteins; protection and refolding of
are also present in the center of the nucleus, where long fibers misfolded, partially unfolded, or aggregated proteins; and
are radially oriented and anchor perpendicularly or obliquely most importantly the maintenance of procollagen in a par-
to the cartilaginous end plate (Yu et al. 2002, 2007). They tially unfolded state for binding effectors or crossing
form a network, as is fibrillin, and constitute the amorphous membranes.
component of the nucleus pulposus that is responsible for its
elastic properties. Individual elastin polypeptide chains (tro-
poelastin) are covalently cross-linked producing an insoluble 5.3.3 Mechanisms of Collagen Chain Selection,
protein and presumably forming random coil-like structures. Intracellular Transport, and Secretion
With such coupling, elastic fibers could play a significant
mechanical role even though overall elastin is less than 5 % Each chain is synthesized with an extra length of peptides
of the total dry weight of the disc. Its content correlates with called registration peptides on both the N-terminal and
degenerative grade and age (Cloyd and Elliott 2007). C-terminal end (Fig. 5.5). Registration peptides ensure that
86 F. Mwale
Procollagen triple helix (300nm)
N-terminal peptide and C-terminal

covalent crosslinking
Cleavage of propeptide C and N and

tropocolagen molecule by
N- and C proteinases
Self-assembly of tropocollagen into

microfibril catalyzed by lysyl oxidase
The molecules are staggered from each

other by about 67 nm (D)
D (67 nm)
Gap(0.53D) Overlap(0.47D)
Overlap and gap regions
Assembled mictofibrils form the

collagen fibril
Gap (lacunar) regions and overlap regions

of microfibrils give the striate aspect of
collagen fibers
Microscopic aspect of a collagen fibril
Fig. 5.5 Schematic illustration of the formation of tropocollagen, showing procollagen, cross-linking, propeptide cleavage, self-assembly, over-
laps, and GAP regions of microfibrils
the chains assemble in the correct position as a triple helix. assembly and precipitation as collagen fibrils. The C-NC
The terminal N and C domains are then excised, modified, domain is the key domain required for heterotrimer assem-
or incorporated directly into the final structure, depending bly. Disulfide bonding in the a2 C-NC domain is crucial for
on protomer and function. Subsequently, specific protomers heterotrimer formation (Fig. 5.2). As discussed previously,
oligomerize into distinct suprastructures involving interac- the domain structure consists of a central collagenous triple-
tions that form end-to-end connections, lateral associations, helical domain flanked by two noncollagenous domains,
and supercoiling of helices. An additional function of the carboxy-terminus (NC1 domain) and amino-terminus (NC2
extra peptides is to maintain the solubility of the procolla- domain) (Fig. 5.4). The terminal NC domains function as
gen molecule and prevent its premature intracellular recognition modules. This happens at the different
subdomains (Telo, Ia, Ib, II, IV, III, and V) of the C-NC 5.4.2 Procollagen N and Procollagen C
domain of the a1 and a2 chains. The selection, binding, and Proteinases
registration of the three a-chains for assembly of the triple-
helical protomers at these subdomain sites are characterized Fibrillar procollagens contain a C-propeptide that is com-
by interactions based on complementarity of shape, electro- pletely removed by a C proteinase after secretion leading to
static charge distribution, and hydrophobic selection of cog- polymerization of their triple-helical domains (Fig. 5.5)
nate a-chains (Khoshnoodi 2006). Domain I of the collagen (Prockop et al. 1997/1998). In contrast, the extent to which
molecule folds into subdomains Ia and Ib without a-helical their amino propeptide (N-propeptide) is removed by N pro-
or b-sheet conformations, inconsistent with suggestions that collagen proteinases is different. With certain types of colla-
subdomain Ia participates in trimerization by forming gens, such as collagens I and II, the N-propeptide is completely
a-helical coiled coils (McAlinden et al. 2003). Subdomain removed, whereas for collagens V and XI most of the
Ib contains the interchain disulfide bonds in the assembled N-propeptides are left attached on the triple-helical domains.
C-NC trimer. Domains II and IV fold into globular regions As such, this may allow them to regulate fibril assembly by
G1 and G2, respectively. These are linked by an antiparallel hindering the addition of molecules at fibril surfaces.
sheet assembled from domains III and V.
Folding is initiated at the C-terminus by bimolecular asso-
ciation of the a2 and a12, peptide chains near the junction of 5.5 Self-Assembly of Collagen Fibrils
domain III and domain IV-G2, and proceeds towards the
N-terminus (Malone et al. 2005). The a2 C-NC domain, 5.5.1 Homotypic CollagenCollagen
specifically domain V, provides the driving force for het- Interactions
erotrimer formation. Trimerization proceeds via a second
interaction between the a2a12 dimer and the a11 at domain The altered protein after cleavage, known as tropocollagen,
II-G1 after which the interchain disulfide bonds become is able to assemble into polymeric collagen fibrils (Fig. 5.5).
established in domain Ib. Folding of domain Ia is the last and Hydroxyproline residues contribute to the stability of the tro-
slowest folding step, but eventually it drives the folding pocollagen triple helix, forming hydrogen bonds between its
through the CteloIa junction (Malone et al. 2004). Putative polypeptide chains.
trimerization control sequences have been located within the
C-NC domains.
5.5.2 Heterotypic CollagenCollagen
and CollagenOther Molecules
5.4 Extracellular Processing of Collagens/
Procollagens Collagen fibrils aggregate spontaneously to form fibers.
Proteoglycans and structural glycoproteins play an important
5.4.1 Enzymes Involved in Processing role in the aggregation of tropocollagen to form fibrils and in
of Procollagens the formation of fibers from fibrils. Collagen may be attached
to cell membranes via several types of proteins, including
The formation and passage of correctly folded triple-helical fibronectin and integrin.
collagen molecules through the secretory pathway involve
chaperone proteins such as disulfide isomerase or binding
proteins that are involved in the recognition of the 5.5.3 Fibril Formation
C-propeptide which may be a mechanism for the retention of
incorrectly folded collagen molecules in the cell (Bottomley Formation of collagen fibrils occurs by self-aggregation of tro-
et al. 2001). The major collagen-binding protein of the rough pocollagen molecules in a staggered array (Fig. 5.5). Collagen
endoplasmic reticulum is HSP47, sometimes called gp46, or fibrils are stabilized by the formation of lysine-derived cross-
colligin (Tasab et al. 2002) and acts along with other chaper- links between tropocollagen molecules catalyzed by lysyl oxi-
ones during collagen passage from the rough endoplasmic dase. The fibrillar structure is reinforced by the formation of
reticulum to the Golgi apparatus. Outside of the cell registra- covalent cross-links between tropocollagen molecules.
tion peptides are cleaved and tropocollagen is formed by pro-
collagen peptidase (Fig. 5.5). Multiple tropocollagen
molecules form collagen fibrils, via covalent cross-linking 5.5.4 Cross-Link Formation
(aldol reaction) by lysyl oxidase which links hydroxylysine
and lysine residues. Multiple collagen fibrils form into col- Side-by-side cross-linking of collagen fibrils to form a col-
lagen fibers. lagen fiber is mediated by proteoglycans and FACIT
88 F. Mwale
collagens. For correct assembly and/or turnover of collagen than an increase in the amount of secreted collagen (Antoniou
fibrils, other molecules, such as tenascin-X, are also et al. 1996). The matrix metalloproteinase (MMP) family has
required. been implicated in the breakdown of collagen and other
matrix proteins.
5.6 Collagen FibrilCell Interactions

5.7.2 Regulation of Collagen Expression
5.6.1 CollagenIntegrin Binding
Besides the general protease inhibitors, mainly a2-macro-
Heterodimers of four integrins, namely, a1b1, a2b1, a10b1, globulin and a1-antiprotease, the activities of matrix met-
and a11b1, form a special subclass of cell adhesion receptors. alloproteinases (MMPs) are limited by specific inhibitors,
They are all collagen receptors, and they recognize their the tissue inhibitors of metalloproteinases (TIMP-14)
ligands with an inserted domain (I domain) in their a-sub- which binds to the substrate-binding site of the MMPs. In
unit. Integrin a1b1 also known as VLA-1 or CD49aCD29 is addition, TIMP-2 is a cofactor in the activation of MMP2
strongly expressed during development. Integrin a2b1 was by membrane-type MMPs. TIMPs are produced by a
originally thought to be the only collagen receptor. Ligand broader range of cell types than the proteases, and TIMP-2
binding to this receptor triggers intracellular signaling, which is often constitutively expressed. TIMP-3, in particular,
include the p38 MAP kinase activation. appears to act as a pro-apoptotic factor. The expression of
most TIMPs is additionally regulated at the transcriptional
level. In general, the same factors that induce MMPs are
5.6.2 CollagenDiscoidin Domain Receptor involved, but additional cytokines such as TNFa as well as
(DDR) Binding glucocorticoids and retinoids are active, at least in the regu-
lation of TIMP-1.
Discoidin domain receptors (DDRs) DDR1 and DDR2 are
receptor tyrosine kinases with the unique ability among
receptor tyrosine kinases to respond to collagen. They are 5.8 Collagen-Related Diseases
distinguished from other members of the receptor tyrosine
kinase family by a discoidin homology repeat in their extra- 5.8.1 Hereditary Disorders
cellular domains that is also found in a variety of other trans-
membrane and secreted proteins. While a considerable number of collagen diseases have
been documented (a number of the more common diseases
are shown in Table 5.1 and Box 5.3), in this chapter the
5.6.3 Integrin-Dependent and DDR-Dependent focus is on those with a phenotype in the intervertebral
Signaling disc.
Collagen IX: Recent reports have suggested the impor-
Integrins are receptors that mediate the attachment between tance of genetic factors in disc disease. Mutations in colla-
a cell and the tissues that surround it, such as other cells or gen IX are associated with premature disc degeneration in
the extracellular matrix. They are involved in cell signaling mice (Kimura et al. 1996) and a predisposition to disc disor-
and the regulation of cell survival, differentiation, and ders in humans. Transgenic mice expressing mutant colla-
response to environmental stimuli in the disc. However, little gen alpha 1(IX) develop progressive joint degeneration with
is known of the discoidin cell surface receptors that directly age as well as accelerated intervertebral disc degeneration
bind to and interact with these matrix proteins in the interver- (Kimura et al. 1996; Boyd et al. 2008). The changes include
tebral disc. shrinkage or disappearance of the nucleus pulposus and
fissures in the annulus fibrosus, which sometimes lead to
herniation of disc material and osteophyte formation.
5.7 Homeostasis of Collagenous Matrices Degeneration in the end plate is associated with cell prolif-
eration, cartilage disorganization, and new bone formation
5.7.1 Degeneration of Collagenous Matrices (Boyd et al. 2008).
MMP-3: Matrix metalloproteinase-3 (MMP-3, stromelysin-1)
With age, the proportion of aggrecan to water in the nucleus is involved in the pathogenesis of disc disease (Takahashi
pulposus falls steeply, while the proportion of collagen to et al. 2001). Polymorphisms in MMP3 as well as TIMP-1
aggrecan rises. A similar change is seen in degenerate discs. were associated with the radiographic progression of disc
These changes appear to arise from loss of aggrecan rather degeneration (Valdes et al. 2005).
Table 5.1 Clinical phenotypes of collagen gene mutations

Gene or protein Clinical phenotype
COL1A1, COLlA2 (collagen 1 al, a2 Family: Osteogenesis imperfecta
chains)
COL2A1 (collagen 2 al chain) Family: achondrogenesis 2, hypochondrogenesis, congenital spondyloepiphyseal dysplasia
(SEDC), Kniest, Stickler arthro-ophthalmopathy, familial osteoarthritis, other variants
COL9Al,COL9A2, C0L9A3 (collagen 9 a1, Multiple epiphyseal dysplasia (MED; two or more variants)
a2, a3 chains)
COLl0A1 (collagen 10) Metaphyseal dysplasia (Schmid alchain)
COII IAI, Co111A2 (collagen 11 al, a2 Oto-spondylo-megaepiphyseal dysplasia (OSMED); Stickler (variant), Marshall syndrome
chains)
COMP Pseudoachondroplasia, multiple epiphyseal dysplasia (MED; one form)
MATN3 (matrilin-3) Multiple epiphyseal dysplasia (MED; one variant)
Perlecan SchwartzJampel type 1; dyssegmental dysplasia
In the annulus, collagen lamellae are formed consisting of

Box 5.3 Clinical Disorders Resulting from Defects in
concentric bundles of parallel collagen fibers arranged
Collagen Synthesis
around the central gelatinous nucleus pulposus.
Disorder Defect Symptoms With age, the proportion of aggrecan to water in the nucleus
EhlersDanlos Mutations in either one Joint pulposus falls steeply, while the proportion of collagen to
type III or two separate genes hypermobility
aggrecan rises. A similar change is seen in degenerate
(also involved in
Vascular EDS and discs. These changes appear to arise from loss of aggrecan
tenascin-X deficiency) rather than an increase in the amount of secreted collagen.
EhlersDanlos Faulty transcription or Aortic and/or Collagen II fibrils in the nucleus pulposus are organized
type IV translation of type III intestinal rupture randomly joined by hydroxypyridium cross-links.
(vascular EDS)
Procollagen II is expressed in two forms by alternative
EhlersDanlos Faulty lysine Augmented skin
type VI hydroxylation elasticity, rupture
splicing with or without exon 2.
of eyeball The fibril-forming collagens I and II form an important
EhlersDanlos Decrease in procollagen Increased network in the disc. Collagen I forms hybrids with other
type VII peptidase activity articular fibrillar collagens, particularly collagen V. Co-fibril for-
mobility, mation with collagen I regulates the fibril diameter.
frequent luxation
The collagen chains are synthesized as procollagens.
Scurvy Lack of vitamin C Ulceration of
(cofactor for proline gums, Glycosylation of procollagen occurs in the rough ER and
hydroxylase) hemorrhages Golgi complex. Galactose and glucose residues are added
Osteogenesis Change of one Spontaneous to hydroxylysine residues, and long oligosaccharides are
imperfect nucleotide in genes for fractures, cardiac added to specific asparagine residues in the C-propeptide.
collagen type I insufficiency
The formation and passage of correctly folded triple-
helical collagen molecules through the secretory pathway
involve disulfide isomerase that recognizes the
5.8.2 Other Collagen-Related Diseases C-propeptide. The major collagen-binding protein in the
RER is HSP47.
See Table 5.1 and Box 5.3. Outside of the cell, registration peptides are cleaved and
tropocollagen is formed by procollagen peptidase.
Multiple tropocollagen molecules form collagen fibrils,
5.9 Summary of Critical Concepts Discussed via covalent cross-linking by lysyl oxidase.
in the Chapter Mutations in collagen IX are associated with premature
disc degeneration in mice and a predisposition to disc dis-
Collagens provide the structural framework of the interver- orders in humans.
tebral discs and are responsible for biomechanical proper- Noncollagenous fibrous proteins in the disc include
ties, including torsion and resistance to pressure or tension. COMP, fibronectin, amyloid, tenascin, and elastin.
Proteoglycans, such as decorin, fibromodulin, and bigly- Collagen cell interactions are mediated via discoidin
can together with other matrix constituents, influence col- domain receptors and integrin a2b1. Ligand binding to this
lagen fibril formation. receptor triggers p38 MAP kinase activation.
90 F. Mwale
Acknowledgments Dr. Mwale is funded by AO Spine, Canadian lecular disulfide bonds in pro-alpha2(I) blocks assembly of type I
Institute of Health Research (CIHR), and North American Spine Society collagen. J Cell Biochem 71(2):233242
(NASS). The author thanks Drs. Shapiro and Risbud for critically read- Erickson HP, Inglesias JL (1984) A six-armed oligomer isolated from
ing the manuscript and providing valuable suggestions and Dr. Ovidiu cell surface fibronectin preparations. Nature 311:267269
Ciobanu for generating the figures and helping with the manuscript. Eyre DR (1979) Biochemistry of the intervertebral disc. Int Rev Connect
Tissue Res 8:227291
Eyre DR, Muir H (1976) Types I and II collagens in intervertebral disc.
Interchanging radial distributions in annulus fibrosus. Biochem J
157:267270
References Eyre DR, Dickson IR, Van Ness K (1988) Collagen cross-linking in
human bone and articular cartilage. Age-related changes in the con-
Aigner T, Gresk-Otter KR, Fairbank JC, von der Mark K, Urban JP tent of mature hydroxypyridinium residues. Biochem J
(1998) Variation with age in the pattern of type X collagen expres- 252(2):495500
sion in normal and scoliotic human intervertebral discs. Calcif Eyre DR, Koob TJ, Van Ness KP (1984) Quantitation of hydroxypyri-
Tissue Int 63:263268 dinium crosslinks in collagen by high-performance liquid chroma-
Anderson ET, Fletcher L, Severin A, Murphy E, Baker SM, Matsuka tography. Anal Biochem 137:380388
YV (2004) Identification of factor XIIIA-reactive glutamine accep- Eyre DR, Matsui Y, Wu JJ (2002) Collagen polymorphisms of the inter-
tor and lysine donor sites within fibronectin-binding protein (FnbA) vertebral disc. Biochem Soc Trans 30(Pt 6):844848
from Staphylococcus aureus. Biochemistry 43(37):1184211852, Feng H, Danfelter M, Stromqvist B, Heinegard D (2006) Extracellular
PubMed PMID: 15362870 matrix in disc degeneration. J Bone Joint Surg Am 88(Suppl
Anderson DG, Markova D, Adams SL, Pacifici M, An HS, Zhang Y 2):2529
(2010) Fibronectin splicing variants in human intervertebral disc Gruber HE, Ingram JA, Hanley EN Jr (2002) Tenascin in the human
and association with disc degeneration. Spine 35:15811588 intervertebral disc: alterations with aging and disc degeneration.
Antoniou J, Steffen T, Nelson F, Winterbottom N, Hollander AP, Poole Biotech Histochem 77:3741
RA, Aebi M, Alini M (1996) The human lumbar intervertebral disc: Gruber HE, Ingram JA, Hanley EN Jr (2006) Immunolocalization of
evidence for changes in the biosynthesis and denaturation of the thrombospondin in the human and sand rat intervertebral disc. Spine
extracellular matrix with growth, maturation, ageing, and degenera- 31:25562561
tion. J Clin Invest 98:9961003 Grynpas MD, Eyre DR, Kirschner DA (1980) Collagen type II differs
Athanasou NA, Kokubun S, West L, Sallie B, Puddle B (1995) from type I in native molecular packing. Biochim Biophys Acta
Glycosaminoglycans in intervertebral disc amyloid deposits. Eur 626:346355
Spine J 4:308312 Hayes AJ, Benjamin M, Ralphs JR (2001) Extracellular matrix in
Benjamin M, Ralphs JR (2004) Biology of fibrocartilage cells. Int Rev development of the intervertebral disc. Matrix Biol 20:107121
Cytol 233:145 Hristova GI, Jarzem P, Ouellet JA, Roughley PJ, Epure LM, Antoniou
Berg RA, Prockop DJ (1973) The thermal transition of a non-hydroxy- J, Mwale F (2011) Calcification in human intervertebral disc degen-
lated form of collagen. Evidence for a role for hydroxyproline in eration and scoliosis. J Orthop Res 29:18881895
stabilizing the triple-helix of collagen. Biochem Biophys Res Hulmes DJS (2002) Building collagen molecules, fibrils, and
Commun 52:115120 suprafibrillar structures. J Struct Biol 137:10
Boos N, Nerlich AG, Wiest I, von der Mark K, Aebi M (1997) Hulmes DJ, Miller A (1981) Molecular packing in collagen. Nature
Immunolocalization of type X collagen in human lumbar interverte- 293:239240
bral discs during ageing and degeneration. Histochem Cell Biol Inkinen RI, Lammi MJ, Lehmonen S, Puustjarvi K, Kaapa E, Tammi
108:471480 MI (1998) Relative increase of biglycan and decorin and altered
Bottomley MJ, Batten MR, Lumb RA (2001) Quality control in the chondroitin sulfate epitopes in the degenerating human interverte-
endoplasmic reticulum: PDI mediates the ER retention of unassem- bral disc. J Rheumatol 25:506514
bled procollagen C-propeptides. Curr Biol 11:11141118 Inoue H (1973) Three-dimensional observation of collagen framework
Boyd LM, Richardson WJ, Allen KD, Flahiff C, Jing L, Li Y, Chen J, of intervertebral discs in rats, dogs and humans. Arch Histol Jpn
Setton LA (2008) Early-onset degeneration of the intervertebral 36:3956
disc and vertebral end plate in mice deficient in type IX collagen. Inoue H (1981) Three-dimensional architecture of lumbar interverte-
Arthritis Rheum 58:164171 bral discs. Spine 6:139146
Bywaters EGL, Dorling J (1970) Amyloid deposits in articular carti- Khoshnoodi J, Cartailler J-P, Alvares K, Veis A, Hudson BG (2006)
lage. Ann Rheum Dis 29:294306 Molecular recognition in the assembly of collagens: terminal non-
Chiquet M, Fambrough DM (1984) Chick myotendinous antigen. II. A collagenous domains are key recognition modules in the formation
novel extracellular glycoprotein complex consisting of large of triplehelical protomers. J Biol Chem 281:3811738121
disulfide-linked subunits. J Cell Biol 98:19371946 Kimura T, Nakata K, Tsumaki N, Miyamoto S, Matsui Y, Ebara S, Ochi
Cloyd JM, Elliott DM (2007) Elastin content correlates with human T (1996) Progressive degeneration of articular cartilage and inter-
disc degeneration in the annulus fibrosus and nucleus pulposus. vertebral discs. An experimental study in transgenic mice bearing a
Spine 32:18261831 type IX collagen mutation. Int Orthop 20:177181
Day JM, Olin AI, Murdoch AD, Canfield A, Sasaki T, Timpl R, Ladefoged C (1985) Amyloid in intervertebral discs. A histopathologi-
Hardingham TE, Aspberg A (2004) Alternative splicing in the aggre- cal investigation of intervertebral discs from 30 randomly selected
can G3 domain influences binding interactions with tenascin-C and autopsies. Appl Pathol 3:96104
other extracellular matrix proteins. J Biol Chem 279:1251112518 Lee CR, Sakai D, Nakai T, Toyama K, Mochida J, Alini M, Grad S
Demers CN, Antoniou J, Mwale F (2004) Value and limitations of using (2007) A phenotypic comparison of intervertebral disc and articular
the bovine tail as a model for the human lumbar spine. Spine cartilage cells in the rat. Eur Spine J 16:21742185
29(24):27932799 Linsenmayer TF, Long F, Nurminskaya M et al (1998) Type X collagen
Doyle SA, Smith BD (1998) Role of the pro-alpha2(I) COOH-terminal and other up-regulated components of the avian hypertrophic carti-
region in assembly of type I collagen: disruption of two intramo- lage program. Prog Nucleic Acid Res Mol Biol 60:79
LuValle P, Ninomiya Y, Rosenblum ND, Olsen BR (1988) The type X Ryan MC, Sandell LJ (1990) Differential expression of a cysteine-rich
collagen gene. Intron sequences split the 5-untranslated region and domain in the amino-terminal propeptide of type II (cartilage)
separate the coding regions for the non-collagenous amino-terminal procollagen by alternative splicing of mRNA. J Biol Chem
and triple-helical domains. J Biol Chem 263:1837818385 265:1033410339
Malone JP, George A, Veis A (2004) Type I collagen N-Telopeptides Sandell LJ, Morris N, Robbins JR, Goldring MB (1991) Alternatively
adopt an ordered structure when docked to their helix receptor dur- spliced type II procollagen mRNAs define distinct populations of
ing fibrillogenesis. Proteins 54:206215 cells during vertebral development: differential expression of the
Malone JP, Alvares K, Veis A (2005) Structure and assembly of the amino-propeptide. J Cell Biol 114:13071319
heterotrimeric and homotrimeric C-propeptides of type I collagen: Schmid TM, Linsenmayer TF (1985) Immunohistochemical localiza-
significance of the a2(I) chain. Biochemistry 44:1526915279 tion of short chain cartilage collagen (type X) in avian tissues. J Cell
Maroudas A, Stockwell RA, Nachemson A, Urban J (1975) Factors Biol 100:598605
involved in the nutrition of the human lumbar intervertebral disc: Scott JE, Bosworth TR, Cribb AM, Taylor JR (1994) The chemical
cellularity and diffusion of glucose in vitro. J Anat 120(Pt 1): morphology of age-related changes in human intervertebral disc
113130 glycosaminoglycans from cervical, thoracic and lumbar nucleus
McAlinden A, Smith TA, Sandell LJ, Ficheux D, Parry DA, Hulmes DJ pulposus and annulus fibrosus. J Anat 184(Pt 1):7382
(2003) Alpha-helical coiled-coil oligomerization domains are Takahashi M, Haro H, Wakabayashi Y, Kawa-uchi T, Komori H,
almost ubiquitous in the collagen superfamily. J Biol Chem 278: Shinomiya K (2001) The association of degeneration of the inter-
4220042207 vertebral disc with 5a/6a polymorphism in the promoter of the
Mihara S, Kawai S, Gondo T, Ishihara T (1994) Intervertebral disc human matrix metalloproteinase-3 gene. J Bone Joint Surg Br
amyloidosis: histochemical, immunohistochemical and ultrastruc- 83(4):491495
tural observations. Histopathology 25:415420 Tasab M, Jenkinson L, Bulleid NJ (2002) Sequence-specific recogni-
Muragaki Y, Kimura T, Ninomiya Y, Olsen BR (1990) The complete tion of collagen triple helices by the collagen-specific molecular
primary structure of two distinct forms of human alpha 1 (IX) col- chaperone hsp47. J Biol Chem 277:3500735012
lagen chains. Eur J Biochem 192:703708 Tchetina E, Mwale F, Poole AR (2003) Distinct phases of coordinated
Mwale F, Billinghurst C, Wu W, Alini M, Webber C, Reiner A, Ionescu early and late gene expression in growth plate chondrocytes in
M, Poole J, Poole AR (2000) Selective assembly and remodelling of relationship to cell proliferation, matrix assembly, remodeling, and
collagens II and IX associated with expression of the chondrocyte cell differentiation. J Bone Miner Res 18:844851
hypertrophic phenotype. Dev Dyn 218:648662 Valdes AM, Hassett G, Hart DJ, Spector TD (2005) Radiographic
Mwale F, Roughley P, Antoniou J (2004) Distinction between the extra- progression of lumbar spine disc degeneration is influenced by vari-
cellular matrix of the nucleus pulposus and hyaline cartilage: a req- ation at inflammatory genes: a candidate SNP association study in
uisite for tissue engineering of intervertebral disc. Eur Cell Mater the Chingford cohort. Spine 30:24452451
8:5863, discussion 6354 van der Rest M, Mayne R (1988) Type IX collagen proteoglycan from
Nachemson A, Lewin T, Maroudas A, Freeman MA (1970) In vitro dif- cartilage is covalently cross-linked to type II collagen. J Biol Chem
fusion of dye through the end-plates and the annulus fibrosus of 263(4):16151618
human lumbar inter-vertebral discs. Acta Orthop Scand 41: Vasios G, Nishimura I, Konomi H, van der Rest M, Ninomiya Y, Olsen
589607 BR (1988) Cartilage type IX collagen-proteoglycan contains a large
Nerlich AG, Schleicher ED, Boos N (1997) Volvo award winner in amino-terminal globular domain encoded by multiple exons. J Biol
basic science studies. Immunohistologic markers for age-related Chem 263:23242329
changes of human lumbar intervertebral discs. Spine 22: Vaughan L, Mendler M, Huber S, Bruckner P, Winterhalter KH, Irwin
27812795 MI, Mayne R (1988) D-periodic distribution of collagen type IX
Nies DE, Hemesath TJ, Kim JH, Gulcher JR, Stefansson K (1991) The along cartilage fibrils. J Cell Biol 106:991997
complete cDNA sequence of human hexabrachion (tenascin). A Wu JJ, Eyre DR (2003) Intervertebral disc collagen. Usage of the short
multidomain protein containing unique epidermal growth factor form of the alpha1(IX) chain in bovine nucleus pulposus. J Biol
repeats. J Biol Chem 266:28182823 Chem 278:2452124525
Nishimura I, Muragaki Y, Olsen BR (1989) Tissue-specific forms of Wu JJ, Eyre DR, Slayter HS (1987) Type VI collagen of the interverte-
type IX collagen-proteoglycan arise from the use of two widely bral disc. Biochemical and electronmicroscopic characterization
separated promoters. J Biol Chem 264:2003320041 of the native protein. Biochem J 248:373381
Oegema TR Jr, Johnson SL, Aguiar DJ, Ogilvie JW (2000) Fibronectin Wu JJ, Woods PE, Eyre DR (1992) Identification of cross-linking sites
and its fragments increase with degeneration in the human in bovine cartilage type IX collagen reveals an antiparallel type
intervertebral disc. Spine 25:27422747 II-type IX molecular relationship and type IX to type IX bonding.
Pokharna HK, Phillips FM (1998) Collagen crosslinks in human lumbar J Biol Chem 267:2300723014
intervertebral disc aging. Spine 23:16451648 Wurster NB, Lust G (1984) Synthesis of fibronectin in normal and
Prockop DJ, Sieron AL, Li SW (1997/1998) Procollagen N-proteinase osteoarthritic articular cartilage. Biochim Biophys Acta 800:
and procollagen C-proteinase Two unusual metalloproteinases that 5258
are essential for procollagen processing probably have important Yamaguchi N, Benya PD, van der Rest M, Ninomiya Y (1989) The
roles in development and cell signaling. Matrix Biol 16:399408 cloning and sequencing of alpha 1 (VIII) collagen cDNAs demon-
Roberts S (2002) Disc morphology in health and disease. Biochem Soc strate that type VIII collagen is a short chain collagen and contains
Trans 30(Pt 6):864869 triple-helical and carboxyl-terminal non-triple-helical domains sim-
Roberts S, Menage J, Urban JP (1989) Biochemical and structural prop- ilar to those of type X collagen. J Biol Chem 264:1602216029
erties of the cartilage end-plate and its relation to the intervertebral Yasuma T, Arai K, Suzuki F (1992) Age-related phenomena in the lum-
disc. Spine 14:166174 bar intervertebral discs. Lipofuscin and amyloid deposition. Spine
Roberts S, Ayad S, Menage PJ (1991) Immunolocalisation of type VI 17:11941198
collagen in the intervertebral disc. Ann Rheum Dis 50:787791 Yu J, Winlove PC, Roberts S, Urban JP (2002) Elastic fibre organiza-
Roughley PJ (2004) Biology of intervertebral disc aging and degeneration: tion in the intervertebral discs of the bovine tail. J Anat 201:
involvement of the extracellular matrix. Spine 29(23):26912699 465475
92 F. Mwale
Yu J, Fairbank JC, Roberts S, Urban JP (2005) The elastic fiber network Zhang G, Eames BF, Cohn MJ (2009) Chapter 2. Evolution of verte-
of the annulus fibrosus of the normal and scoliotic human interver- brate cartilage development. Curr Top Dev Biol 86:1542
tebral disc. Spine 30:18151820 Zheng Q, Zhou G, Chen Y, Garcia-Rojas X, Lee B (2003) Type X col-
Yu J, Tirlapur U, Fairbank J, Handford P, Roberts S, Winlove CP, Cui Z, lagen gene regulation by Runx2 contributes directly to its hypertro-
Urban J (2007) Microfibrils, elastin fibres and collagen fibres in the phic chondrocyte-specific expression in vivo. J Cell Biol 162:
human intervertebral disc and bovine tail disc. J Anat 210:460471 833842
Microenvironmental Control of Disc Cell
Function: Inuence of Hypoxia 6
and Osmotic Pressure
Makarand V. Risbud and Irving M. Shapiro
The action of this principle is exactly like that of the centrifugal governor of the steam engine, which checks
and corrects any irregularities almost before they become evident; and in like manner no unbalanced
deficiency in the animal kingdom can ever reach any conspicuous magnitude, because it would make itself
felt at the very first step, by rendering existence difficult and extinction almost sure soon to follow
Alfred Russell Wallace
Contributions to the Theory of Natural Selection 1858
Contents 6.1 Dening the Intervertebral Disc Niche

6.1 Defining the Intervertebral Disc Niche ......................... 93
The intervertebral disc is a complex structure that displays
6.2 Role of HIF Proteins in the Intervertebral
many of the characteristics of a diarthrodial polyaxial joint in
Disc Niche......................................................................... 94
6.2.1 Control of HIF-a Stability in the Disc .............................. 96 that it separates opposing cartilage-covered bones (verte-
6.2.2 Role of HIF-1 in Energy Conservation by Cells brae), permits a range of motions, and accommodates high
of the Nucleus Pulposus .................................................... 98 biomechanical forces. The nature of this joint has been
6.2.3 Role of Hypoxia and HIF in Promoting Cell Survival
described in detail in Chap. 1 and in a recent review by
and Function in the Disc ................................................... 99
Shapiro et al. (2012). While discs from the different anatom-
6.3 Contribution of TonEBP to the Maintenance of the ical regions of the spine vary in shape and volume, their
Osmotic Niche .................................................................. 101
6.3.1 Control of TonEBP Expression and Activity in the architecture is generally similar. At the disc periphery, the
Nucleus Pulposus .............................................................. 102 outer annulus fibrosus layer forms a ligamentous structure,
6.4 Niche Factors Control Matrix Synthesis by Nucleus
composed of tightly packed parallel collagen I fibrils that are
Pulposus Cells .................................................................. 103 inserted into contiguous superior and inferior vertebral bod-
6.4.1 Control of Proteoglycan Synthesis.................................... 103 ies. The inner surface of the annulus fibrosus comprises a
6.4.2 Control of GAG Synthesis by Niche Factors .................... 103 poorly organized fibrocartilage containing collagen II fibrils.
6.4.3 Hypoxia, HIF-1, and CCN2 Expression ........................... 103
The annulus and the cartilaginous endplates enclose the
6.5 Role of Niche Factors in Promoting Disc nucleus pulposus, an aggrecan-rich gel-like tissue that is
Cell Renewal .................................................................... 104 sparsely populated with cells. Although embryologically
6.6 Summary of Critical Concepts Discussed distinct, cells of the nucleus pulposus are often mistakenly
in the Chapter .................................................................. 104 compared with chondrocytes. The nucleus pulposus is
References ..................................................................................... 105 derived from the notochord, whereas annulus fibrosus and
endplate cartilage are sclerotomal in origin (see Chap. 3 for
further details concerning the ontology of disc cells).
Throughout this chapter, cells of this notochordal-derived
tissue are described as cells of the nucleus pulposus.
The interaction between the semifluid nucleus pulposus
and the tight molecular lattice of the annulus fibrosus pro-
vides the biomechanical properties necessary for spinal sta-
bility. Disturbing this relationship by compromising the
M.V. Risbud (*) I.M. Shapiro stability of the nucleus pulposus, the annulus fibrosus, or the
Department of Orthopaedic Surgery and Graduate Program endplate cartilage results in disc degeneration, a condition
in Cell and Developmental Biology, that can lead to excruciating pain and loss of function and
Jefferson Medical College, Thomas Jefferson University,
1025 Walnut Street, Suite 511 College Bldg,
which often results in costly surgical interventions. Because
Philadelphia, PA 19107, USA the degenerative process is chronic, the nucleus pulposus
e-mail: makarand.risbud@jefferson.edu; irving.shapiro@jefferson.edu cells are required to function for long time periods in what

94 M.V. Risbud and I.M. Shapiro
can be described as a suboptimal microenvironmental niche; indicating a hypoxic environment. Based on these and other
the goal of this chapter is to consider those conditions that studies, it is now widely accepted that the nucleus pulposus
enhance nucleus pulposus cell survival as well as factors that cells reside in a hypoxic tissue niche.
disregulate the disc microenvironment and promote degen- Another important environmental factor that character-
erative disc disease. izes the disc niche is the elevated osmolarity. Although
Herein, we use the term niche to describe the confines of sparse, cells in the nucleus pulposus secrete a complex extra-
the nucleus, bounded laterally and medially by the annulus cellular matrix that contains collagens and the proteoglycan
and superiorly and inferiorly by the endplate cartilage. aggrecan as well as versican. Glycosaminoglycan (GAG)
Although the concept of a niche was originally directed at chains of the aggrecan molecule provide a robust hydrody-
anatomical structures, more recently the term has been used namic system that serves to accommodate applied biome-
to describe interactions between communities of cells that chanical forces (Feng et al. 2006; Setton and Chen 2006). In
are in close proximity to each other. For example, within the the nucleus pulposus, the principle GAG is chondroitin sul-
bone marrow niche, stem cell commitment to a particular lin- fate. Bound to the aggrecan core protein and associated with
eage is dependent on local microenvironmental conditions hyaluronic acid, the chondroitin sulfate chains form a giant
that regulate the interactions between resident hematopoietic polydispersed supramolecular structure with a net negative
as well as stromal cells. charge; this serves as a driving force for binding cations
Although the clinical outcomes of degenerative disc dis- especially Na+, thereby elevating tissue osmolarity (this topic
ease are well documented, biological events that regulate is discussed further in Chaps. 2 and 4). The high osmotic
nucleus pulposus cell survival are not understood. One over- pressure of the aggregate contains the forces applied to the
riding aspect of disc cell biology is that cells of the nucleus spine (Ng et al. 2003). We have shown that nucleus pulposus
pulposus and cells residing in the inner annulus are removed cells responded to changes in osmotic pressure by upregulat-
from the blood supply. For example, blood vessels originating the transcription factor, TonEBP (tonicity enhancer-
ing in the vertebral body traverse the superficial region of the binding protein) (Tsai et al. 2006), the only known mammalian
endplates; none of these vessels infiltrate the nucleus pulpo- transcription factor that responds to changes in osmolarity.
sus. Urban, Maroudas, and colleagues pioneered the studies In the following sections, we will describe in detail the activ-
of solute transport and biophysical properties of the disc ities of this interesting protein and its importance in mainte-
(Urban et al. 1977). Several modeling studies and biochemi- nance of disc cell function. In addition to osmolarity and
cal measurements of glycolytic pathway metabolites by these hypoxia, several other morphogenic proteins including those
workers predict that the oxygen tension (pO2) within the disc of the TGF-b superfamily play an important role in niche
is low and that metabolism is primarily anaerobic (Bartels maintenance. Towards the end of the chapter, the activities of
et al. 1998), even if the oxygen tension is raised (Holm et al. TGF-b in the postnatal nucleus pulposus are briefly
1981). With respect to the annulus, Gruber et al. (2005) discussed.
pointed out that this tissue is avascular except for small dis-
crete capillary beds in the dorsal and ventral surfaces; in no
case, does the annulus vasculature enter the nucleus pulpo- 6.2 Role of HIF Proteins in the Intervertebral
sus. Microangiographical and immunohistochemical studies Disc Niche
of the human disc lend strong support to the notion that the
nucleus pulposus is avascular in nature (Hassler 1969; Rudert If the concept of a regulatory niche, composed of a number
and Tillmann 1993). Moreover, even during disc degenera- of cell types responsive to local microenvironmental condi-
tion vascular invasion of the nucleus pulposus is not seen, tions, is valid, then this begs the question: is cell survival in
suggesting that vascular in-growth is not a defining feature of the hypoxic niche hypoxia-inducible factor (HIF) depen-
disc disease (Nerlich et al. 2007). Lee and colleagues exam- dent? Thanks to the brilliant studies of Semenza and col-
ined hypoxic nature of the rat disc using 2-nitroimidazole, leagues, it is now recognized that the critical molecule
EF5, a drug that at low pO2 forms covalent product with cel- regulating energy metabolism and survival activity under
lular proteins (Lee et al. 2007). These studies revealed that hypoxia is HIF-1 (Semenza et al. 1994). HIF is a member of
the transition zone and not the nucleus pulposus exhibited the the basic helixloophelix (bHLH)PERARNTSIM
highest level of EF5 binding. The authors concluded that the (PAS) family of proteins and composed of a constitutively
disc cells adapt to the local environment by limiting the con- expressed b-subunit and an a-subunit. In most cells, the lat-
sumption of oxygen. To further investigate this concept, ter subunit is stable under hypoxic conditions but is rapidly
Schipani and colleagues generated and characterized degraded in normoxia (Wang et al. 1995). Transactivation of
hypoxia-inducible reporter mice (5XHRE-LacZ reporter) HIF-1 target genes involves dimerization of the two subunits
(Fig. 6.1). Labeling with hypoxia marker EF5 clearly showed and binding to an enhancer, the hypoxia-response element in
a robust signal in the presumptive intervertebral disc target genes. HIF-1 serves as a key transcription factor that
6 Microenvironmental Control of Disc Cell Function: Influence of Hypoxia and Osmotic Pressure 95
a b
Fig. 6.1 The axial skeleton is hypoxic and expresses HIF-1a. (a) hypoxia marker EF5 confirms that the nucleus pulposus (NP) of the
Characterization of hypoxia- inducible reporter mice (5XHRE-LacZ presumptive intervertebral disc is hypoxic (Images kindly provided
reporter). Note activation of the reporter construct in mesenchymal by Dr. Ernestina Schipani, Indiana University)
condensations by b-gal staining (black arrows). (b) Staining with
regulates the expression of enzymes concerned with glycoly- expression of Sox9, Sox5, and Sox6 is hypoxia- (5 % O2) and
sis, the activity of the TCA cycle, and oxidative phosphory- HIF-2-sensitive during chondrogenic differentiation of stem
lation (Semenza et al. 1994; Papandreou et al. 2006; Fukuda cells derived from the infrapatellar fat pads of osteoarthritic
et al. 2007). Additional target genes include those required patients (Khan et al. 2007). In contrast, using marrow mesen-
for survival, apoptosis, autophagy, and matrix synthesis chymal stem cells, Kanichai et al. (2008) showed involve-
(Schipani et al. 2001; Zhang et al. 2008; Hofbauer et al. ment of HIF-1a in regulating Sox9 expression during
2003). Details of these relationships are shown schemati- chondrogenesis under hypoxia. However, the relationship
cally in Fig. 6.2. It should be added that other isoforms exist, between Sox proteins and HIF in the hypoxic niche of the
the most important being HIF-2a. Recent evidence suggests disc is not as yet known (Box 6.1).
that HIF-1a and HIF-2a are not redundant and that the rela-
tive importance of each of the homologues, in response to
hypoxia, varies among different cell types (Sowter et al.
2003). For example, unlike HIF-1, HIF-2 regulates expres- Box 6.1: Oxygen, Hypoxia and HIF
sion of a number of unique genes including superoxide dis- Over a billion years ago, the primitive atmosphere of
mutase 2 (SOD2), catalase, frataxin, and cited2 (Scortegagna the earth held very small amounts of dioxygen. Through
et al. 2003; Oktay et al. 2007; Aprelikova et al. 2006). the extraordinary activities of photosynthetic plants
In addition to the genes mentioned above, the Sox family and organisms, over an enormous time period, the level
of transcription factors that are essential for the development of oxygen increased rapidly: a mere 200 million years
and function of the nucleus pulposus are hypoxia and HIF ago, the oxygen content increased to about 16 %.
sensitive (Smits and Lefebvre 2003; Lafont et al. 2007; Khan Today, despite a huge elevation in manmade and natu-
et al. 2007; Kanichai et al. 2008). Lafont et al. (2007) showed ral oxidative activities, the gas level has held constant
that HIF-2, but not HIF-1, regulated the expression of Sox9 around 21 %.
and the phenotype of primary human chondrocytes. Similarly,
Fig. 6.2 Functional activity

of HIF target genes. Critical Nucleus
functions include energy
metabolism, angiogenesis, cell HIF-1/2
Poll
survival, autophagy and P300/ complex
apoptosis, matrix synthesis, CBP
proliferation, self-renewal and
differentiation, radical dismuta- HIF-1 HIF-1/2
tion, and pH regulation. Many
of these functions are critical for
survival and functioning of the HRE
nucleus pulposus cells in the
avascular niche of the interverte-
bral disc. Hypoxia-/HIF-sensitive
proteins that are identified in the
nucleus pulposus cells are shown
in parentheses (Reproduced from
Cell proliferation
Risbud et al. (2010). With self renewal Radical dismutation
permission from Elsevier) (Hes1, Hey1) (Frataxin, SOD2)
Cell survival,
Matrix genes
autophagy and apoptosis
Energy/glucose (Aggrecan, Col-2,
(Akt, CCN2, ERK1/2,
metabolism GlcAT-1, Sox-9)
Galectin-3, VEGF-A)
(enolase-1, Glut-1, -3,
PFKFB)
The life-supporting activities of the gas was recog- example, in the carotid body and in mitochondria.
nized by the Greeks who postulated that there were A family of enzymes, prolyl hydroxylases, monitors
only four natural elements, air along with earth, the oxemic state of the cell and enhances adaptation to
water, and fire. At the time of the American Revolution, hyperoxic and hypoxic conditions.
there was a convergence of intellectual thought on the The term hypoxia is a generalized term used to
notion that air was a mixture of gases, especially the describe the fall in oxygen to levels that are necessary
vital gas, oxygen. Among those charged with the dis- to sustain most animal life. In some tissues, the blood
covery of oxygen were Joseph Priestley, Carl Wilhelm oxygen concentration can be normoxic, but the cells
Scheele, and Antoine Lavoisier. Priestley, an experience hypoxia due to a high cell density, elevated
Englishman, who for religious and political reasons metabolic activity, or poor vascularization. If hypoxia
settled in Pennsylvania, and was a good friend of develops, a family of transcription factors is activated.
Benjamin Franklin, found that a gas generated from Hypoxia-inducible factors or HIFs serve to change the
metal oxides could keep a mouse alive longer than a metabolic activities of the cell so as to accommodate
similar volume of air. Priestley called his discovery the available oxygen concentration.
dephlogisticated air. Almost at the same time,
Scheele isolated the same gas which he termed fire
air. A Frenchman, Antoine Lavoisier, labeled the gas 6.2.1 Control of HIF-a Stability in the Disc
acid-maker or oxygen, viewing it as part of the air
that was free to combine with other elements. Of note, To return to the question raised above concerning the impor-
Priestly found that the dephlogisticated air had the tance of the HIF system, a considerable number of reports
property of changing dark venous blood to bright red now clearly show that there is a robust HIF response by cells
arterial blood. of the nucleus pulposus. The response is evident across spe-
We now know that red blood cells carry dioxygen to cies; it is seen in vivo and in vitro, and more importantly,
all of the tissues of the body so as to power metabolic HIF-1a activity is unresponsive to the oxemic state of the
reactions, particularly those concerned with mainte- tissue (Rajpurohit et al. 2002; Risbud et al. 2006a, b; Agrawal
nance of redox and high-energy intermediates such as et al. 2007). Accordingly, when compared with most other
ATP. Oxygen sensor systems exist within tissues: for tissues, there are substantive underlying differences in the
HIF status and reactivity of disc cells: HIF-1a expression
and activity is always on. This unusual response suggests

that stabilization of HIF-1a in cells of the nucleus pulposus O2 dependent
PHD1
PHD3
ensures that transcriptional activity is a major determinant of DMOG/Bips
cell function. The second HIF homologue, HIF-2a, is VHL Ub 26S
robustly expressed by nucleus pulposus cells. Like HIF-1a, PHD2
OH OH OH OH
steady-state protein levels are similar in both hypoxia and
normoxia, suggesting that it too is constitutively expressed HIF-1 HIF-1 HIF-1
(Agrawal et al. 2008).
Before leaving this topic, it is important to comment on
mechanisms of stabilization and turnover of HIF-1a and HSP70/RACK1 MG132
HIF-2a in nucleus pulposus cells. HIF-1a can be stabilized
in a number of different ways. For example, von Hippel- Autophago HIF-2
Lindau protein activity can be suppressed, or O2 sensing some
by one or more of the prolyl hydroxylase (PHD) enzymes,
members of the 2-oxoglutarate/Fe2+-dependent dioxyge- O2 independent
Lysosome
nase superfamily, could be low (Appelhoff et al. 2004).
Importantly, since the activity of PHDs depends on the
Bafilomycin A1/chloroquine
tissue oxygen tension, these molecules serve as sensors
that control the cellular abundance of HIF-a proteins. It is Fig. 6.3 A schematic model of the unique regulation of HIF-1a and
known that PHD proteins hydroxylate specific prolyl resi- HIF-2a degradation in nucleus pulposus cells. PHD2 controls a limited
dues in the oxygen-dependent degradation domain of HIF-a oxygen-dependent degradation of HIF-1a through 26S proteasome
subunits. The hydroxylated proteins are bound by the ubiq- pathway. Oxygen-independent mechanisms through 26S proteasome as
well as lysosomal pathway are active in HIF-1a turnover. In contrast,
uitin ligase von HippelLindau tumor suppressor protein HIF-2a is unresponsive to oxidative degradation and is also turned over
(pVHL), which targets them for rapid ubiquitination and though 26S and lysosomal pathway (Reproduced from Fujita et al.
26S proteasomal degradation (Maxwell et al. 1999). We (2012a). With permission of the American Society for Bone and Mineral
reported recently that expression of PHD1-3 is higher in Research)
cells of the nucleus pulposus than in cells of the annulus
fibrosus (Fujita et al. 2012a). Noteworthy, unlike other cells, there is very low cellular utilization of O2, an adaptive response
our studies clearly showed that in nucleus pulposus cells, to the hypoxic niche (Bibby et al. 2005; Lee et al. 2007). The
stability of HIF-a-oxygen-dependent degradation domain limited involvement of PHD2 in HIF-1a degradation implies
(ODD) is independent of oxemic tension. In addition, muta- that it is not a major regulator of HIF-1a turnover. Furthermore,
genesis studies suggest that hydroxylation reaction may it lends support to the hypothesis that HIF-1a levels are regu-
not control HIF-2a degradation in these cells (Kditz et al. lated primarily by oxygen-independent pathways.
2007; Fujita et al. 2012a). Again these findings are different In contrast to HIF-1a, the turnover of HIF-2a through
from articular chondrocytes that exhibit responsiveness of 26S proteasome is largely independent of PHD function;
HIF-2a degradation to PHD function in vitro suggesting a there is also a limited involvement of the lysosomal pathway.
cell type-specific response (Thoms and Murphy 2010). Relevant to this discussion, recent studies by Gogate et al.
Further investigations found that both HIF-1a and HIF-2a (2012) indicate that Hsp70 modulates HIF-1a protein stabil-
were degraded through the 26S proteasome pathway. Since all ity and transcriptional activity. In nucleus pulposus, Hsp70
PHDs mediate proteasomal HIF-a degradation, but differ in was shown to interact with HIF-1a under hypoxic conditions
their ability to hydroxylate HIF-1a in vivo, we investigated and promoted degradation through the proteasomal pathway
their individual role in HIF-a turnover in nucleus pulposus (Gogate et al. 2012). These findings strongly suggest that
cells (Minamishima et al. 2008; Takeda et al. 2006). The high nucleus pulposus cells are functionally adapted to their avas-
relative expression of PHD2 in nucleus pulposus tissue sug- cular, hypoxic microenvironment and rely mostly on oxy-
gests that this isoform may play an important role in HIF-a gen-independent pathways for controlling HIF-1a and
turnover. These studies indicate that PHD2 controls to a lim- HIF-2a levels. With respect to regulation, in hypoxia, PHD3
ited extent HIF-1a degradation even under hypoxic condi- promotes HIF-1a transcriptional activity (Fujita et al.
tions, indicating preservation of PHD2 enzymatic activity at 2012b); this observation is in line with a recent report that
low O2 tension. This finding is in marked contrast to previous describes a crucial role of the PHD3PKM2 complex in
reports demonstrating HIF-1a stabilization at low oxygen ten- enhancing HIF-1a interaction with p300, an important tran-
sions because of inhibition of PHD enzymatic activity (Epstein scriptional coactivator (Luo et al. 2011). The possible path-
et al. 2001). Moreover, this observation highlights the unique ways involved in HIF-a turnover in nucleus pulposus cells
physiology of the nucleus pulposus cells and suggests that are demarcated in Fig. 6.3.
It is important to note that, in nucleus pulposus cells, the three target genes (glucose transporter [GLUT]-1 and 3 and
expression of PHD2 and PHD3 is also induced by hypoxia in GAPDH) at 221 % O2 were evaluated, it was found that the
an isoform-specific manner (Fujita et al. 2012b). While activities are comparable and remain constant over time
PHD2 is selectively regulated by HIF-1a, PHD3 expression (Agrawal et al. 2007). Although these genes were not respon-
is controlled by both HIF-1a and HIF-2a at the transcript sive to the oxemic state of the culture, we have observed a
level. Significantly when there is inflammatory disc disease, small induction in enolase-1 and phosphofructokinase 2
expression of PHD2 and PHD3 is primarily responsive to (PFKFB) promoter activities. This result is surprising as the
TNF-a and IL-1b in HIF independent fashion (Fujita et al. later protein is regarded as the glycolytic pacemaker; how-
2012c). Moreover, unlike other tissues (DAngelo et al. ever, because this intermediary step is sensitive to a number
2003; Marxsen et al. 2004; Henze et al. 2010), hypoxic of hormones, and metabolic intermediates, it is more than
expression of PHD1 is also dependent on HIF-1a activity in likely that induction is in response to other regulatory fac-
nucleus pulposus cells (Fujita et al. 2012b). Taken together, tors. Nevertheless, the muted response does not detract from
these studies clearly indicate the existence of a regulatory the conclusion that the glycolytic flux in disc cells, even in
feedback loop between PHD2, PHD3, and HIF-1a in the normoxia, is high.
hypoxic nucleus pulposus cells. This scenario is distinct In normoxia, the basal concentration of ATP in nucleus
from the articular cartilage, a tissue functionally similar to pulposus cells is between 20 and 25 nmol/L/mg protein
the nucleus pulposus, where PHD2 also regulates HIF-2a (Agrawal et al. 2007). These values are comparable with lev-
degradation (Thoms and Murphy 2010). These results high- els reported for articular chondrocytes, another cell type that
light the unique nature and control of the HIF-PHD system uses glycolysis to generate energy (Pfander et al. 2003). In
in nucleus pulposus cells and for the first time provide a bio- the presence of 2-deoxyglucose, a potent inhibitor of glyco-
chemical rationale for normoxic stabilization and mainte- lysis, ATP generation is suppressed by almost 80 % (Agrawal
nance of steady-state levels of these proteins in the nucleus et al. 2007). The sensitivity of the cells to this inhibitor
pulposus. Whether stabilization of HIF-a proteins is related emphasizes the reliance on glycolysis for energy generation.
to the unique embryonic origins of the tissue is currently Based on this observation, it is likely that the oxemic stabil-
unknown. However, it is important to note that because discs ity of HIF-1a in nucleus pulposus cells is optimal for sur-
are hypoxic in vivo, stabilization of HIF-a expression would vival in an environment where there are frequent shifts in
serve to maintain cell metabolism and functional activities vascular supply and O2 delivery; in the intervertebral disc,
when disc integrity is breached during disc herniation (Ha these shifts may reflect minute to minute or day/night varia-
et al. 2006) or at an early stage of degeneration (Roberts tions in biomechanical forces applied to the spinal units.
et al. 2006). Although glycolysis is clearly the major ATP-generating
pathway, the possibility exists that some high-energy inter-
mediates may be produced through mitochondrial oxidative
6.2.2 Role of HIF-1 in Energy Conservation phosphorylation. However, current studies indicate that
by Cells of the Nucleus Pulposus inhibitors of mitochondrial function do not influence ATP
production nor nucleus pulposus cell viability (Agrawal et al.
Earlier classical biochemical studies have shown that when 2007). As for a role, if any, for mitochondria, little is known.
the pO2 is low, there is almost complete reliance on glycoly- Gan et al. (2003) have reported that although nucleus pulpo-
sis to generate ATP and reducing equivalents. As indicated in sus cells contain mitochondria with normal architecture, the
the previous section, one of the consequences of low oxygen total number of organelles per cell is low. Nevertheless,
tension in the nucleus pulposus is the reliance on glycolysis nucleus pulposus cells can perform mitochondrial oxidative
for energy generation (Agrawal et al. 2007; Holm et al. metabolism: thus, they oxidize fatty acid and generate ATP
1981). Glycolysis may be viewed as a relatively inefficient (Agrawal et al. 2007). Based on all of these studies, there is
process: it generates 2 mol of ATP/mol glucose; in contrast, strong support for the notion that although glucose and
mitochondrial metabolism is slow, but it creates about 30 anaerobic glycolysis represent the major fuel and pathway
mol of ATP/mol glucose. In the trade-off between rate and for energy generation, respectively, mitochondria in the
yield, the glycolytic pathway generates a small number of nucleus pulposus are functional; they retain the capacity to
ATP molecules at a very fast rate and maintains the reducing metabolize fatty acids through mitochondrial oxidative
status of the cell. In this way, glycolysis provides sufficient metabolism.
energy for both housekeeping functions and for protein The conclusion that disc cell energy metabolism is depen-
synthesis. dent on glycolysis fits well with current observations con-
One of the logical outcomes of stabilization of HIF-1a is cerning the regulatory functions of HIF-1. It is known that
the robust expression of glucose transporters and enzymes HIF-1 plays a major role in directing the interplay between
required for anaerobic glycolysis. When the expression of glycolysis and oxidative phosphorylation. HIF-1 inhibits
mitochondrial function by trans-activating the gene encod- are high in hypoxia and when serum-starved confers resis-
ing pyruvate dehydrogenase kinase 1. Because this protein tance to apoptosis (Risbud et al. 2005a, b). Activation of this
suppresses pyruvate dehydrogenase, pyruvate cannot be con- protein is of considerable interest as it has been shown to
verted into acetyl-CoA, and as a result the TCA cycle is modulate apoptosis by inactivating Bad and caspase-9 and
blocked (Papandreou et al. 2006). Moreover Fukuda et al. modulating the transcription of proapoptotic transcription
(2007) showed that HIF-1 reciprocally regulates mitochon- factors (Duronio 2008). Relevant to nucleus pulposus cells,
drial cytochrome c oxidase (COX)-4 subunit expression by activation of PI3K/Akt signaling has been shown to regulate
activating transcription of the genes encoding COX4-2 and a HIF-1a protein levels in other cell types (Kanichai et al.
protease that is required for COX4-1 degradation. Thus, HIF 2008). Like Akt, extracellular signal-regulated kinase
regulates not just the entry of reducing equivalents into the (ERK)1/2 is induced in hypoxic nucleus pulposus cells.
mitochondria but also oxidative phosphorylation. Based on Because activation of ERK has been linked to survival, pos-
these observations, it is concluded that although mitochon- sibly by regulating nitric oxide synthase and caspase activi-
drial function is retained by cells of the intervertebral disc, it ties, the it is probable that activation of ERK in concert with
is reasonable to assume that normoxic expression of HIF-1a Akt serves to maintain the viability of the disc cells at a low
by nucleus pulposus cells serves to suppress oxidative phos- pO2 (Risbud et al. 2005a, b).
phorylation and promotes glycolytic ATP generation. More Like HIF-1, one of the critical functions of Akt is regula-
than likely, nucleus pulposus mitochondria are required for tion of glucose metabolism. It is thought that Akt may pro-
non-energy-related metabolic functions, while oxidative mote cell survival by maintaining GLUT-1 transcription
phosphorylation is used to a very minor degree. under conditions of growth factor withdrawal (serum starva-
tion) (Rathmell et al. 2003). Indeed, the high level of expres-
sion of GLUT-1 protein by nucleus pulposus cells in vivo
6.2.3 Role of Hypoxia and HIF in Promoting Cell (Rajpurohit et al. 2002; Richardson et al. 2008) indicates that
Survival and Function in the Disc this tissue may adapt to its hypoxic environment by increas-
ing glucose uptake. This activity would serve to promote and
If the premise is correct that the HIF signaling network serves enhance glycolysis, thereby preventing ischemia-induced
to promote nucleus pulposus function, then the cells should injury. Taken together, these studies suggest that the PI3K-
be adapted to survive and grow in a hypoxic environment. Akt and ERK signaling pathways in conjunction with HIF-1
Studies from our lab (Risbud et al. 2005a, b; Agrawal et al. provide a mechanism by which nucleus pulposus cells remain
2007; Zeng et al. 2007) and other groups (Mwale et al. 2011; viable and maintain their specialized physiological function,
Feng et al. 2013) show that hypoxia, possibly through HIF-1, despite environmental limitations in O2 and possibly changes
enhances the expression of important matrix genes and the in nutrient availability.
phenotype of nucleus pulposus cells. Experiments have also Two proteins that have been linked to nucleus pulposus
been performed in which nucleus pulposus cells were treated survival in hypoxia are VEGF-A and galectin-3. In disc cells,
with low levels of common apoptogens and survival mea- it has been reported that HIF-1 regulates galectin-3 expres-
sured (Risbud et al. 2005a, b). Notably, when the pO2 was sion (Zeng et al. 2007). From a functional perspective, by
below 5 %, there was maximum disc cell survival. Studies of forming complexes with integrins, externalized galectin-3
nucleus pulposus and chondrocytes showed that when HIF- influences cell adhesion and spreading (Sasaki et al. 1998).
1a is partially silenced, viability is maintained in the face of Accordingly, galectin-3 is most likely involved with matrix
an O2 challenge (Fujita et al. 2012b; Bohensky et al. 2007). stability and in concert with HIF-1 provides the discal cells
On the other hand, complete deletion of HIF-1 in growth with both a mechano-transduction and a survival function.
plate chondrocytes results in massive cell death highlighting Other studies have shown that galectin-3 regulates survival
the requirement of HIF-1 for cell survival in the hypoxic by suppressing signaling through the TNF family of proteins
niche (Schipani et al. 2001). However, it is possible that (Oka et al. 2005). This finding is particularly pertinent to disc
some effects of hypoxia on nucleus pulposus survival are disease, as TNF-a together with other cytokines is known to
probably mediated by other signaling molecules in an HIF- play a major role in the etiology, as well as progression, of
1-independent fashion. The latter observation raises the the degenerative state. Based on these findings, it is possible
question: which signaling pathways are upregulated in that in the hypoxic intervertebral disc, the robust expression
hypoxia? Work from a number of labs suggests that a variety of HIF-1a serves to maintain galectin-3 levels, which then
of hypoxia-responsive proteins exist, including vascular serve to promote cell survival and disc function (Zeng et al.
endothelial growth factor (VEGF) (Fujita et al. 2008; 2007).
Agrawal et al. 2008), galectin-3 (Zeng et al. 2007), and Akt/ With respect to VEGF-A, not surprisingly, levels of this
PI3K (Risbud et al. 2005a, b). Our work has shown that protein are high in herniated discs or in degenerative
expression levels of phospho-Akt in nucleus pulposus cells discs where there is evidence of neovascularization
(Kokubo et al. 2008). In the normative state, because the disc Bohensky et al. (2009) pointed out that HIF-2 was involved
environment is avascular, it would be reasonable to assume in regulating survival by modulating autophagy. HIF-2 was
that VEGF expression is low. However, this is not the case as expressed abundantly by cells in human and murine articular
there is robust expression of this protein and its receptor in cartilage, hypertrophic cartilage, and in the endplate carti-
the nucleus pulposus (Fujita et al. 2008; Agrawal et al. 2008). lage. When HIF-2a was suppressed, ROS generation was
Most likely, the level of expression is related to both HIF-1 elevated, and there was a decrease in the activity of the ROS
and HIF-2, as both isoforms upregulate VEGF-A expression dismutating enzymes catalase and superoxide dismutase.
and promoter activity (Agrawal et al. 2008). This observa- Suppression of HIF-2a was associated with decreased Akt-
tion begs the question: in the disc, what is the function of 1, reduced Bcl-x(L) expression, and a robust autophagic
VEGF? Clearly, it cannot serve to promote angiogenesis as response, even under nutrient-replete conditions (Bohensky
this activity would promote vascularization and compromise et al. 2009). In addition, Semenza and colleagues have dem-
disc function. There is some information to indicate that onstrated important contributions of HIF-1a to autophagy
VEGF supports cell survival (Zelzer et al. 2004). Indeed, (Zhang et al. 2008). Relevant to disc disease, it is generally
Fujita et al. (2008) confirmed that VEGF and its receptors are agreed that the degenerative state is exacerbated by a decrease
expressed by nucleus cells in hypoxia and showed that this in the permeability of the endplate cartilage and the concom-
protein promoted nucleus pulposus survival. Thus, from a itant reduction in nutrient availability. It is possible that under
functional viewpoint, VEGF could serve as to maintain these nutritionally challenging conditions, increased HIF-
nucleus pulposus viability in the face of shifts in environ- 1/-2 expression may serve to maintain nucleus pulposus sur-
mental pO2. vival by promoting the induction of autophagy (Box 6.2).
As indicated above, there is growing interest in the sec-
ond HIF homologue, HIF-2a. This protein is robustly
expressed by nucleus pulposus cells. With respect to func- Box 6.2: HIF, Hypoxia, and Human Populations
tional activities, unlike most other tissues, hypoxia failed to Oxidative metabolism and the availability of oxygen
increase the transcriptional activities of SOD2 and frataxin, have powered evolutionary processes that have permit-
two common HIF-2 target genes concerned with radical dis- ted animals to populate almost every region of the
mutation (Scortegagna et al. 2003; Oktay et al. 2007). This planet. At high altitudes and in the depth of the ocean,
finding could explain why this tissue is susceptible to radical the partial pressure of oxygen is low; nevertheless,
attack associated with annular lesions or nucleus herniation. organisms have evolved mechanisms to adapt to these
Notably, there is evidence to indicate that kyphosis, scolio- nonphysiologic conditions. The question that has
sis, and radiculopathies are linked to defective radical dis- intrigued scientists is: what type of adaptation mecha-
mutation (Murakami and Kameyama 1963; Sparrow et al. nism exists to enable humans and animals to reside in
2012), whereas Friedreichs ataxia is now known to be attrib- harsh environments characterized by the high Andean,
utable to low frataxin levels and loss of antioxidant defenses Tibetan, and Ethiopian plateaus? For humans living at
(Gakh et al. 2006). It would be important to know whether high altitudes, there may be genetic selection pressures
these conditions are also linked to the inability of vertebral for phenotypes characterized by raised oxygen hemo-
tissues to mount a robust HIF-2-dependent scavenging globin saturation; one problem here is that a change in
response. blood viscosity can elicit profound medical problems.
Notably, both HIF-1 and HIF-2 are involved in survival of However, studies of Tibetans living at very high alti-
endplate chondrocytes by activation of the autophagic path- tudes show that they exhibit an SNP close to the region
way (Bohensky et al. 2009; Srinivas et al. 2009). Recent encoding HIF-2. Beall et al. (2010) have proposed that
studies suggest that autophagy is active in the nucleus pulpo- this mutation causes a blunted erythropoietic response.
sus (Ye et al. 2011; Jiang et al. 2012). The importance of this In this case, the change in viscosity would be minimal.
system for removal of misfolded proteins and damaged There is evidence that HIF-1 isoform may also be
organelles has been emphasized by a number of workers, and involved. Studying the Naqu Yak, an animal that has
its role in directing the maturation of connective tissue cells lived on the Tibetan plateau for more than million years
has been discussed by Srinivas and his colleagues (Srinivas at an altitude of 9,00015,000 ft above sea level, Wang
et al. 2009). Noteworthy, although autophagy is viewed as a et al. (2006) reported that HIF-1 is expressed at high
survival pathway, there is little doubt that continued macro- levels in the brain, lung, and kidney. Since HIF-1 tar-
molecular breakdown, while serving as a source of nutrients gets EPO (erythropoietin) as well as many other genes
and energy for the stressed cell, inevitably leads to increased concerned with erythropoiesis, these animals would
susceptibility to apoptosis (type II apoptosis). Hence, HIF have an enhanced ability to transport oxygen in the
activity and ultimately HIF targets serve as key proteins that blood stream. In summary, both animal and humans
straddle both the apoptotic and survival pathways.
Rel family member and unlike monomeric members of the

have adapted to hypoxia by expressing genes that mod- NFAT family exists as a homodimer and forms stable dimers
ify oxygen transport and erythropoiesis. with DNA. Upon activation, TonEBP binds to the tonicity-
responsive enhancer element (TonE) of genes required for
References osmotolerance and cell survival. These genes include the
Beall CM, Cavalleri GL, Deng L, Elston RC, Gao Y, betaine/g-aminobutyric acid transporter, sodium myo-inositol
Knight J, Li C, Li JC, Liang Y, McCormack M, co-transporter (Ko et al. 1997; Miyakawa et al. 1998; Rim
Montgomery HE, Pan H, Robbins PA, Shianna KV, et al. 1998), taurine transporter (Zhang et al. 2003; Ito et al.
Tam SC, Tsering N, Veeramah KR, Wang W, 2004), and aldose reductase (Lopez-Rodrguez et al. 1999).
Wangdui P, Weale ME, Xu Y, Xu Z, Yang L, Zaman By regulating levels of betaine, myo-inositol, taurine, and
MJ, Zeng C, Zhang L, Zhang X, Zhaxi P, Zheng sorbitol, these genes control the osmotic properties of the
YT (2010) Natural selection on EPAS1 (HIF2alpha) cytosol. Hsp70, a molecular chaperone that maintains cellular
associated with low hemoglobin concentration in function under hypertonic stress is also induced by TonEBP
Tibetan highlanders. Proc Natl Acad Sci U S A (Woo et al. 2002; Shim et al. 2002). Most homozygous
107(25):1145911464 TonEBP knockout mice evidence midgestational lethality. Of
Wang DP, Li HG, Li YJ, Guo SC, Yang J, Qi DL, Jin the few that survive, all exhibit severe growth retardation and
C, Zhao XQ (2006) Hypoxia-inducible factor kidney dysfunction (Lopez-Rodriguez et al. 2004). A trans-
1alpha cDNA cloning and its mRNA and protein genic mouse expressing a dominant-negative form of TonEBP
tissue specific expression in domestic yak (Bos (DN-TonEBP) in collecting duct epithelial cells demonstrated
grunniens) from Qinghai-Tibetan plateau. Biochem an absolute requirement of TonEBP for expression of the urea
Biophys Res Commun 15:348(1):310349 transporter gene and aquaporin-2 (Lam et al. 2004).
Aside from osmoregulation, TonEBP is required for T
cell proliferation and function (Trama et al. 2000; Go et al.
2004), and it is implicated in cancer cell migration and
6.3 Contribution of TonEBP to the metastasis (Jauliac et al. 2002). A study by Wang et al. (2005)
Maintenance of the Osmotic Niche showed that expression of DN-TonEBP in lens fiber cells
promoted cataract formation by causing defects in their elon-
For disc cells, the second major environmental challenge is gation. Since TonEBP is expressed by a number of cell types,
the high osmotic pressure of the niche (this topic is also dis- it is reasonable to assume that it serves a variety of physio-
cussed in Chap. 2). As pointed out earlier, the biomechanical logic functions, especially those that impact on tissue hydra-
function of the nucleus pulposus is primarily attributed to the tion and the osmotic environment (Maouyo et al. 2002).
water binding and unique osmotic properties of the tissue. TonEBP and its downstream target genes are robustly
Osmotic pressurization of the nucleus pulposus balances the expressed in the nucleus pulposus and the annulus fibrosus
loads experienced by the spine. Studies by Urban and col- (see Fig. 6.4). Importantly, TonEBP is critical for mainte-
leagues indicate that the tissue osmolarity is substantially nance of nucleus pulposus survival under hyperosmotic con-
higher than that of plasma and is in the range of 400500 mos- ditions (Tsai et al. 2006). The observed increase in apoptosis
mol/kg (Urban and Maroudas 1981; Ishihara et al. 1997). is in agreement with studies of TonEBP null mice and trans-
Until recently, the mechanism by which cells of the nucleus genic animals expressing DN-TonEBP in the thymus and in
pulposus and annulus fibrosus control their intracellular the lens (Trama et al. 2000; Go et al. 2004; Wang et al. 2005).
osmotic properties was not known. Drawing on information In both these conditions, there was an acceleration of cell
relevant to other tissues, it was noted that cellular adaptation death through apoptosis. Very importantly, recent studies in
to hyperosmotic stress is mediated by the tonicity enhancer- C. elegans demonstrating selective and pronounced expres-
binding protein (TonEBP), also called OREBP (Miyakawa sion of TonEBP in the notochord suggest that its osmoregu-
et al. 1999) or NFAT5 (Lopez-Rodrguez et al. 1999). latory function first evolved in the primitive axial skeleton
TonEBP/NFAT5 belongs to the 5-member NFAT protein sub- prior to diversification to other tissues (Jos-Edwards et al.
family (NFAT1-5) which is part of the larger Rel superfamily 2011). The relationship between TonEBP and cell death and
of proteins. Similar to other members of the Rel family, it its expression in the notochord lends strength to the notion
contains the Rel homology domain, a conserved DNA- that this transcription factor is of critical importance in the
binding domain. However, no similarities are seen between life history of cells of the nucleus pulposus.
TonEBP and NF-kB or NFAT1-4 outside of the Rel homol- One of the primary responses of cells to variations in
ogy domain. It lacks the binding site for calcineurin neces- local osmolarity is a change in regulatory cell volume. Disc
sary for NFAT1-4 dephosphorylation and subsequent nuclear cells and chondrocytes alike adapt to these osmotic shifts
translocation (Lopez-Rodrguez et al. 1999). It is the largest by remodeling their cytoskeleton and by catalyzing the
b d f
Cartilage
a
Kidney
Heart
Liver
Disc
TonEBP
TauT
c e g
HSP70
SMIT
BGT1
GAPDH
Neonatal Mature Mature
Fig. 6.4 (a) Expression level of TonEBP and other osmotically active with alcian blue, eosin, and propidium iodide (cg). Note that nucleus
genes in the intervertebral disc and other rat tissues. mRNA was pulposus cells in the neonatal (b) as well as skeletally mature disc cells
extracted from disc tissue, costochondral cartilage, heart, kidney, and (d) express TonEBP protein; much of the staining is localized to the
liver of adult rats and subjected to RT-PCR analysis. Note there is nucleus (b, arrows). Some staining is also evident in the cytosol of the
robust expression of TonEBP and its target genes: HSP-70, BGT-1, and nucleus pulposus cells of mature discs (d, arrowhead). Furthermore,
SMIT mRNA. Cartilage another aggrecan-rich skeletal tissue contains annulus fibrosus cells localized in a narrow zone of alcian blue-posi-
lower levels of TonEBP mRNA than the disc. Kidney maximally tive matrix (g, arrowhead) express TonEBP protein (f, arrow).
expresses TonEBP as well as the target genes. (bg) Sagittal and lon- Magnification: 20 (This research was originally published in Tsai
gitudinal sections of disc tissue from neonatal (b) and mature rat (df) et al. (2006). the American Society for Biochemistry and Molecular
spines that were treated with anti-TonEBP antibody or counterstained Biology)
transport of osmotically active molecules and water across 6.3.1 Control of TonEBP Expression and
the plasma membrane (Pritchard et al. 2002; Tsai et al. Activity in the Nucleus Pulposus
2006, 2007; Hall and Bush 2001). Water transport is regu-
lated by a large family of channel-forming proteins, aqua- Before leaving this topic, it is important to comment on sev-
porins (AQP) (Fu and Lu 2007). AQP2, an arginine eral mechanisms unique to the disc niche that control TonEBP
vasopressin regulated channel, plays an important role in expression and activity. The mechanism of activation of
water reabsorption by connecting tubules and collecting TonEBP is complex and not completely understood, espe-
ducts of the kidney (Verkman 2006). When activated, phos- cially whether it is mediated by protein phosphorylation
phorylation of critical serine residues in AQP2 results in its (Woo et al. 2002). In T cells and kidney cells, there is some
translocation from cytoplasmic vesicles to the apical mem- evidence to indicate that regulation may be mediated by a
brane. Intercalated with membrane proteins, AQP2 enhances phosphatase, calcineurin, which is activated by calcium ions
water influx into the cell (Verkman 2006). It has been sug- (Trama et al. 2000). Studies from our lab suggest that although
gested that in the kidney, expression of AQP2 is regulated Ca2+ is involved in TonEBP activation, the downstream
by TonEBP (Hasler et al. 2006; Li et al. 2007; Jeon et al. mechanisms and contribution of calcineurin may be cell type
2006). Studies by Li et al. (2007) suggest that calcium ions specific (Hiyama et al. 2009). Treatment of nucleus pulposus
with calcineurin-NFAT participate in regulation of AQP2 cells with cyclosporine A and FK506, inhibitors of calcineu-
expression. Related to the functional importance of this rin signaling, failed to block induction of TonEBP or change
system, Pritchard et al. (2002) have documented the pres- the promoter activity of the TonEBP target gene, taurine
ence of calcium transients in disc cells exposed to osmotic transporter. Other niche factors like hypoxia, TGF-b, and
stress. In recent studies, we clearly showed that in both the BMP-2 have been shown to modulate TonEBP expression
rat and the human, nucleus pulposus cells express AQP2 (Hiyama et al. 2010). Gogate et al. (2012) showed that
protein (Gajghate et al. 2009). Importantly, unlike kidney, hypoxia causes a small increase in TonEBP protein levels.
osmotic pressure and calcium modulate AQP2 expression Importantly, in hypoxia, there was increased phosphorylation
through TonEBP in calcineurinNFAT-independent fash- and activation of TonEBP-TAD. Likewise, BMP-2 or TGF-b
ion. This finding lends credence to the view that by regulat- increased protein levels of TonEBP and there was a significant
ing the hydration status of the disc, TonEBP maintains cell activation of TAD. Since calcium regulates TonEBP activity
function in a hyperosmotic mechanically stressed and as one effect of TGF-b is the initiation of calcium tran-
environment. sients, this raises the question: are these transients required
for TGF-b-mediated TonEBP activation? Nevertheless, it is chondroitin sulfate synthesis. Other workers have shown that
clear that irrespective of changes in Ca2+ flux, TGF-b and galactose-b1,3-glucuronysltransferase-1 (GlcAT-I) activity is
BMP-2 serve to upregulate TonEBP expression as well as its required for GAG chain synthesis (Kitagawa et al. 1996);
transcriptional activity possibly in a cell-/tissue-specific man- hence, there is the possibility that this enzyme serves as the
ner. The notion that TonEBP expression and activity is con- rate-limiting step in GAG synthesis for nucleus pulposus cells
trolled by tissue- and cell-specific niche factors in addition to as well as chondrocytes (Venkatesan et al. 2004; Bai et al.
osmolarity was supported by a recent study of smooth muscle 1999). Related to this point, it is now known that IL-1b sup-
cells by Halterman and colleagues (Halterman et al. 2011). In presses GAG biosynthesis by downregulating GlcAT-I expres-
these cells, angiotensin II promoted TonEBP nuclear translo- sion and activity (Gouze et al. 2001). A second factor
cation and activity, while PDGF-BB elevated TonEBP pro- regulating aggrecan as well as GAG synthesis is the intracel-
tein levels. lular Ca2+ concentration (Alford et al. 2003; Parvizi et al.
2002; Vijayagopal and Subramaniam 2001; Fagnen et al.
1999); it is speculated that Ca2+ ions controlled a common
6.4 Niche Factors Control Matrix Synthesis early step in the GAG biosynthetic pathway. Building on these
by Nucleus Pulposus Cells observations, we performed studies to investigate the role of
Ca2+ and TonEBP in GlcAT-I expression. Our studies clearly
6.4.1 Control of Proteoglycan Synthesis demonstrated that TonEBP regulates GlcAT-I expression and
that regulation is dependent on intracellular Ca2+ ions (Hiyama
Although a considerable number of water-binding molecules et al. 2009). We also showed that Ca2+-dependent calcineurin
contribute to the regulation of the osmotic pressure, aggrecan (Cn)-NFAT signaling serves as a negative regulator of GlcAT-I
is the major polyelectrolyte. The charged COO and SO42 expression in these cells. From this perspective, by control-
groups of N-acetylgalactosamine, glucuronic acid, and other ling GAG as well as aggrecan synthesis, TonEBP permits
substituted sugars bind hydrated Na+ ions, thereby regulating nucleus pulposus cells to autoregulate the osmotic environ-
the osmotic properties of the disc. While it is known that ment of the disc. However, in contrast to TonEBP, HIF-1
aggrecan transcription and oxemic tension as well as osmotic serves as a negative regulator of GlcAT-I expression (Gogate
pressure are linked (Ishihara et al. 1997; Risbud et al. et al. 2011). This observation was surprising as hypoxia serves
2006a, b; Wuertz et al. 2007), details of the relationship are to increase GAG synthesis by the nucleus pulposus cells
obscure. Promoter analysis of aggrecan provided a new (Gogate et al. 2011; Feng et al. 2013). Thus, TonEBP and
insight into this relationship. We showed the presence of two HIF-1 may in some instances counter each others activities.
TonE sites at 390 and 912 bp in the mouse aggrecan pro-
moter (Tsai et al. 2006). A similar motif was noted in the
human aggrecan promoter. The observation that the human 6.4.3 Hypoxia, HIF-1, and CCN2 Expression
TonE was at 890 bp probably reflects differences in species
specific organization of the aggrecan promoter sequence. Previous work has shown that CCN2/CTGF, a matricellular
The presence of these conserved motifs provides a direct link protein is expressed by nucleus pulposus cells and is critical
between aggrecan expression and tissue osmolarity. for matrix homeostasis. In nucleus pulposus cells, CCN2
Subsequent loss-of-function studies clearly showed that promotes expression of aggrecan and collagen II (Tran et al.
aggrecan promoter is responsive to TonEBP (Tsai et al. 2010; Erwin et al. 2006). Relevant to the niche of the inter-
2006). Important to this discussion of aggrecan regulation, vertebral disc, low oxygen tension is known to regulate
studies from our lab (Risbud et al. 2005a, b; Agrawal et al. CCN2 in several cell types (Higgins et al. 2004; Hong et al.
2007) and Feng et al. (2013) showed that hypoxia and HIF- 2006; Kondo et al. 2006). Interestingly, CCN2 regulation by
1a positively controls aggrecan gene expression in nucleus hypoxia is cell type specific and complex, with hypoxia pro-
pulposus cells. These findings strongly indicate that aside moting production of CCN2 in most cells; in some, CCN2
from Sox9 and other transcriptionally active proteins, expression is downregulated.
TonEBP and HIF-1 serve as regulators of aggrecan expres- Higgins et al. (2004) were the first to demonstrate that
sion, a critically functional component of the disc matrix. hypoxic induction of CCN2 in tubular epithelial cells required
two HREs in the murine CCN2 promoter located at
1558/1554 and 3745/3741 bp. It was concluded that HIF-
6.4.2 Control of GAG Synthesis by Niche Factors 1a induces CCN2 transcription in response to hypoxia and that
both of the HRE binding sites were necessary for promoter
As the water binding capacity of the proteoglycan matrix is activation. Although the latter HRE lies within an evolutionary
dependent on the GAG side chains, it raises the question conserved region (ECR) of the promoter, these HREs are not
whether disc niche factors also regulate GAG and in particular conserved in location in the human CCN2 promoter, begging
the question: are HREs in the human promoter functional and notch signaling pathway is responsive to hypoxia. In skeletal
required for controlling expression in the nucleus pulposus? tissues, disruption of notch signaling markedly increases tra-
To address this question and study the regulation of CCN2 becular bone mass: with aging, the mice become osteopenic
in nucleus pulposus cells, we analyzed the proximal 5 kb of due to a sharp reduction in mesenchymal progenitor popula-
the human CCN2 promoter with an experimentally validated tions (Engin et al. 2008; Hilton et al. 2008). Hypoxia also
HRE matrix using the JASPAR database and found three increases the expression of known notch target genes such as
putative HREs located at 640/634,2010/2006, and Hes1 and Hey1 (Gustafsson et al. 2005). Recent studies by
2264/2258 bp, of which, the 2010/2006 bp site lies Hiyama and colleagues showed that the cells of the nucleus
within an ECR. In nucleus pulposus cells, we observed that pulposus and annulus fibrosus expressed genes of the notch
hypoxia decreased CCN2 transcription. Interestingly, muta- signaling pathway (Hiyama et al. 2011). Moreover, in both
genesis of the putative HREs in both human and mouse pro- tissues, hypoxia increased notch1 and notch4 expression.
moter showed that the suppressive effect of hypoxia may not Interestingly, some tissue specificity was also noticed in that
involve direct binding of HIF-1a to these sites and suggests a Jagged1 was induced by hypoxia only in the annulus fibrosus,
complex regulation of CCN2 by hypoxia and HIF-1a in while Jagged2 expression was highly sensitive to hypoxia in
nucleus pulposus cells (Tran et al. 2013). This is not surpris- both tissues. Importantly, inhibition of notch signaling
ing given the unique stabilization of HIF-1a and a similar blocked disc cell proliferation. Relevant to disc disease, this
mode of regulation of other HIF-1a target genes in the nucleus study clearly showed increased expression of notch signaling
pulposus (Fujita et al. 2012a; Gogate et al. 2012). Moreover, genes in degenerated human discs (Hiyama et al. 2011).
TGF-b, another important morphogenic factor, robustly Central to this discussion, HIF-1a has been shown to
induced CCN2 and is active in the disc from development to interact with the intracellular domain of the notch protein and
maturation. We have shown that TGF-b can still induce CCN2 results in inhibition of differentiation of myogenic and neural
expression under hypoxia. Although the magnitude of induc- precursor cells (Gustafsson et al. 2005). Accordingly, in the
tion is decreased it supports the idea that hypoxic suppression nucleus, HIF-1a may directly interact with the notch intrac-
of CCN2 in nucleus pulposus cells may serve to prevent ellular domain and direct cell fate. Based on what is known of
excessive CCN2 production (Tran et al. 2013). The fact that cell replacement in other tissues, this HIF-1-regulated path-
the locations of HREs in the CCN2 promoter are not con- way is a critical component of cell renewal and replacement.
served in vertebrates suggests that regulation by hypoxia is From a disease viewpoint, an oxemic shift possibly medi-
not only cell type specific, but may also be species specific. ated by alterations in the vascular supply to the endplate car-
tilage or even the annulus fibrosus would be expected to lead
to a failure in progenitor cell activation and a decrease in the
6.5 Role of Niche Factors in Promoting Disc number of differentiated cells. In turn, this would lead to
Cell Renewal decrements in function and enhancement of the effect of
agents that are known to promote disc degeneration. From a
In this chapter, attention has been drawn to the critical role of therapeutic viewpoint, it should be possible to modulate the
the tissue oxygen tension on the function and survival of niche environment to enhance renewal and promote differen-
cells of the intervertebral disc. We have focused on the tiation of precursors into functional cells of the nucleus or
mechanisms by which the hypoxia-sensitive transcription the annulus. Accordingly, rather than relying on surgical and
factors HIF-1 and HIF-2 influence energy metabolism and other interventional strategies, which may themselves dam-
expression of survival proteins. In addition, we have dis- age the disc or cause infection, it should be possible to pro-
cussed how cells of the nucleus respond to hypoxia-sensitive mote tissue repair by manipulating oxemic conditions within
proteins, galectin-3, Akt, and VEGF-A. Where applicable, the niche or use proteins of the notch signaling pathway to
we have extended these discussions to include the impact of reactivate the endogenous progenitor cells in the annulus
these molecules and hypoxia on degenerating resident cells fibrosus or nucleus pulposus. Restoration of disc cell func-
in the intervertebral niche. It should be stated that in concert tion and prevention of degeneration remains the ultimate
with most connective tissues, cell turnover within the disc goal of current intervertebral disc research.
niche is slow. Moreover, like most of these tissues, progeni-
tor cells are present in the disc that can differentiate along the
mesengenic pathway to replace resident cells (Risbud et al. 6.6 Summary of Critical Concepts Discussed
2007; Sakai et al. 2012) (also see Chap. 23). Thus, tissue in the Chapter
renewal in the intervertebral disc is dependent on the ability
of progenitor cells to commit to the nucleus pulposus lineage The tissues of the intervertebral disc form a specialized
and undergo terminal differentiation. niche which is avascular and hypoxic; the osmotic pres-
The notch signaling pathway is central to these progenitor sure within the niches is raised due to the high concentra-
activities, and pertinent to the ideas discussed earlier, the tion of aggrecan and other proteoglycans.
In the disc niche, nucleus pulposus activities are controlled Bohensky J, Shapiro IM, Leshinsky S, Terkhorn SP, Adams CS, Srinivas
by HIF and TonEBP, two major transcription factors. V (2007) HIF-1 regulation of chondrocyte apoptosis: induction of
the autophagic pathway. Autophagy 3:207214
HIF-a stability in the nucleus pulposus is primarily con- Bohensky J, Terkhorn SP, Freeman TA, Adams CS, Garcia JA, Shapiro
trolled by oxygen-independent pathways. PHD2 plays a IM, Srinivas V (2009) Regulation of autophagy in human and
limited role in HIF-1a turnover, while HIF-2a turnover is murine cartilage: hypoxia-inducible factor 2 suppresses chondro-
refractory to prolyl hydroxylation. cyte autophagy. Arthritis Rheum 60:14061415
DAngelo G, Duplan E, Boyer N, Vigne P, Frelin C (2003) Hypoxia
Nucleus pulposus cells are obligate glycolytic and require up-regulates prolyl hydroxylase activity: a feedback mechanism
HIF-1 for energy generation. that limits HIF-1 responses during reoxygenation. J Biol Chem 278:
Hypoxia, HIF, and other hypoxia-sensitive protein play an 3818338187
important role in maintenance of cell survival activities in Duronio V (2008) The life of a cell: apoptosis regulation by the PI3K/
PKB pathway. Biochem J 415:333344
the disc. Engin F, Yao Z, Yang T, Zhou G, Bertin T, Jiang MM et al (2008)
In the hyperosmolar and mechanically stressed environ- Dimorphic effects of notch signaling in bone homeostasis. Nat Med
ment of the disc, both nucleus pulpous and annulus 14:299305
fibrosus cells robustly express TonEBP protein which is Epstein AC, Gleadle JM, McNeill LA, Hewitson KS, ORourke J, Mole
DR et al (2001) C. Elegans EGL-9 and mammalian homologs define
critical for the maintenance of cell survival. a family of dioxygenases that regulate HIF by prolyl hydroxylation.
Osmolarity and other niche factors such as hypoxia, Cell 107:4354
TGF-b family proteins, and Ca2+ control TonEBP expres- Erwin WM, Ashman K, ODonnel P, Inman RD (2006) Nucleus pulpo-
sion and activity in the nucleus pulposus. sus notochord cells secrete connective tissue growth factor and up-
regulate proteoglycan expression by intervertebral disc chondrocytes.
Homeostatic maintenance of proteoglycan-rich matrix in Arthritis Rheum 54:38593867
the intervertebral disc is TonEBP and HIF dependent. Fagnen G, Phamantu NT, Bocquet J, Bonnamy PJ (1999) Inhibition of
Niche factors acting through the notch pathway may pro- transmembrane calcium influx induces decrease in proteoglycan
mote disc cell proliferation and differentiation. synthesis in immature rat sertoli cells. J Cell Biochem 76:322331
Feng H, Danfelter M, Strmqvist B, Heinegrd D (2006) Extracellular
matrix in disc degeneration. J Bone Joint Surg Am 88:2529
Acknowledgements This work is supported by grants AR050087 Feng G, Li L, Liu H, Song Y, Huang F, Tu C et al (2013) Hypoxia dif-
and AR055655 from the National Institutes of Health (NIAMS). The ferentially regulates human nucleus pulposus and annulus fibrosus
authors wish to thank Dr. Ernestina Schipani for the use of the images cell extracellular matrix production in 3D scaffolds. Osteoarthritis
of the 5XHRE reporter mouse (Fig. 6.1). Cartilage 21:582588.
Fu D, Lu M (2007) The structural basis of water permeation and proton
exclusion in aquaporins. Mol Membr Biol 24:366374
Fujita N, Imai J, Suzuki T, Yamada M, Ninomiya K, Miyamoto K et al
References (2008) Vascular endothelial growth factor-a is a survival factor for
nucleus pulposus cells in the intervertebral disc. Biochem Biophys
Res Commun 372:367372
Agrawal A, Guttapalli A, Narayan S, Albert TJ, Shapiro IM, Risbud
Fujita N, Chiba K, Shapiro IM, Risbud MV (2012a) HIF-1a and HIF-
MV (2007) Normoxic stabilization of HIF-1alpha drives glycolytic
2a degradation is differentially regulated in nucleus pulposus cells
metabolism and regulates aggrecan gene expression in nucleus pul-
of the intervertebral disc. J Bone Miner Res 27:401412
posus cells of the rat intervertebral disk. Am J Physiol Cell Physiol
Fujita N, Markova D, Anderson DG, Chiba K, Toyama Y, Shapiro IM
293:C621C631
et al (2012b) Expression of prolyl hydroxylases (PHDs) is selec-
Agrawal A, Gajghate S, Smith H, Anderson DG, Albert TJ, Shapiro IM
tively controlled by HIF-1 and HIF-2 proteins in nucleus pulposus
et al (2008) Cited2 modulates hypoxia-inducible factor-dependent
cells of the intervertebral disc: distinct roles of PHD2 and PHD3
expression of vascular endothelial growth factor in nucleus pulposus
proteins in controlling HIF-1a activity in hypoxia. J Biol Chem
cells of the rat intervertebral disc. Arthritis Rheum 58:3798808
287:1697516986
Alford AI, Yellowley CE, Jacobs CR, Donahue HJ (2003) Increases in
Fujita N, Gogate SS, Chiba K, Toyama Y, Shapiro IM, Risbud MV
cytosolic calcium, but not fluid flow, affect aggrecan mRNA levels
(2012c) Prolyl hydroxylase 3 (PHD3) modulates catabolic effects of
in articular chondrocytes. J Cell Biochem 90:938944
tumor necrosis factor-a (TNF-a) on cells of the nucleus pulposus
Appelhoff RJ, Tian YM, Raval RR, Turley H, Harris AL, Pugh CW et al
through co-activation of nuclear factor kB (NF-kB)/p65 signaling.
(2004) Differential function of the prolyl hydroxylases PHD1,
J Biol Chem 287:3994239953
PHD2, and PHD3 in the regulation of hypoxia-inducible factor.
Fukuda R, Zhang H, Kim JW, Shimoda L, Dang CV, Semenza GL
J Biol Chem 279:3845838465
(2007) HIF-1 regulates cytochrome oxidase subunits to optimize
Aprelikova O, Wood M, Tackett S, Chandramouli GV, Barrett JC (2006)
efficiency of respiration in hypoxic cells. Cell 129:111122
Role of ETS transcription factors in the hypoxia-inducible factor-2
Gajghate S, Hiyama A, Shah M, Sakai D, Anderson DG, Shapiro IM
target gene selection. Cancer Res 66:56415647
et al (2009) Osmolarity and intracellular calcium regulate aqua-
Bai X, Wei G, Sinha A, Esko JD (1999) Chinese hamster ovary cell
porin2 expression through TonEBP in nucleus pulposus cells of the
mutants defective in glycosaminoglycan assembly and glucurono-
intervertebral disc. J Bone Miner Res 24:9921001
syltransferase I. J Biol Chem 274:1301713024
Gakh O, Park S, Liu G, Macomber L, Imlay JA, Ferreira GC et al (2006)
Bartels EM, Fairbank JC, Winlove CP, Urban JP (1998) Oxygen and
Mitochondrial iron detoxification is a primary function of frataxin
lactate concentrations measured in vivo in the intervertebral discs of
that limits oxidative damage and preserves cell longevity. Hum Mol
patients with scoliosis and back pain. Spine 23:17
Genet 15:467479
Bibby SR, Jones DA, Ripley RM, Urban JP (2005) Metabolism of the
Gan JC, Ducheyne P, Vresilovic EJ, Swaim W, Shapiro IM (2003)
intervertebral disc: effects of low levels of oxygen, glucose, and pH
Intervertebral disc tissue engineering I: characterization of the
on rates of energy metabolism of bovine nucleus pulposus cells.
nucleus pulposus. Clin Orthop Relat Res 411:305314
Spine 30:487496
Go WY, Liu X, Roti MA, Liu F, Ho SN (2004) NFAT5/TonEBP mutant Hofbauer KH, Gess B, Lohaus C, Meyer HE, Katschinski D, Kurtz A
mice define osmotic stress as a critical feature of the lymphoid (2003) Oxygen tension regulates the expression of a group of pro-
microenvironment. Proc Natl Acad Sci U S A 101:1067310678 collagen hydroxylases. Eur J Biochem 270:45154522
Gogate SS, Nasser R, Shapiro IM, Risbud MV (2011) Hypoxic regula- Holm S, Maroudas A, Urban JP, Selstam G, Nachemson A (1981)
tion of b-1,3-glucuronyltransferase I expression in nucleus pulposus Nutrition of the intervertebral disc: solute transport and metabolism.
cells of the rat intervertebral disc: role of hypoxia-inducible factor Connect Tissue Res 8:101119
proteins. Arthritis Rheum 63:19501960 Hong K, Yoo S, Kang S, Choi J, Kim W, Cho C (2006) Hypoxia induces
Gogate SS, Fujita N, Skubutyte R, Shapiro IM, Risbud MV (2012) expression of connective tissue growth factor in scleroderma skin
Tonicity enhancer binding protein (TonEBP) and hypoxia-inducible fibroblasts. Clin Exp Immunol 146:362370
factor (HIF) coordinate heat shock protein 70 (Hsp70) expression in Ishihara H, Warensjo K, Roberts S, Urban JP (1997) Proteoglycan syn-
hypoxic nucleus pulposus cells: role of Hsp70 in HIF-1a degrada- thesis in the intervertebral disk nucleus: the role of extracellular
tion. J Bone Miner Res 27:11061117 osmolality. Am J Physiol 272:C1499C1506
Gouze JN, Bordji K, Gulberti S, Terlain B, Netter P, Magdalou J et al Ito T, Fujio Y, Hirata M, Takatani T, Matsuda T, Muraoka S et al (2004)
(2001) Interleukin-1 beta down-regulates the expression of Expression of taurine transporter is regulated through the TonE
glucuronosyltransferase I, a key enzyme priming glycosaminogly- (tonicity-responsive element)/TonEBP (TonE-binding protein) path-
can biosynthesis: influence of glucosamine on interleukin-1beta- way and contributes to cytoprotection in HepG2 cells. Biochem J
mediated effects in rat chondrocytes. Arthritis Rheum 44:351360 382:177182
Gruber HE, Ashraf N, Kilburn J, Williams C, Norton HJ, Gordon BE Jauliac S, Lopez-Rodriguez C, Shaw LM, Brown LF, Rao A, Toker A
et al (2005) Vertebral endplate architecture and vascularization: (2002) The role of NFAT transcription factors in integrin-mediated
application of micro-computerized tomography, a vascular tracer, carcinoma invasion. Nat Cell Biol 4:540544
and immunocytochemistry in analyses of disc degeneration in the Jeon US, Kim JA, Sheen MR, Kwon HM (2006) How tonicity regulates
aging sand rat. Spine 30:25932600 genes: story of TonEBP transcriptional activator. Acta Physiol (Oxf)
Gustafsson MV, Zheng X, Pereira T, Gradin K, Jin S, Lundkvist J et al 187:241247
(2005) Hypoxia requires notch signaling to maintain the undifferen- Jiang L, Zhang X, Zheng X, Ru A, Ni X, Wu Y et al (2012) Apoptosis,
tiated cell state. Dev Cell 9:617628 senescence, and autophagy in rat nucleus pulposus cells: implica-
Ha KY, Koh IJ, Kirpalani PA, Kim YY, Cho YK, Khang GS et al (2006) tions for diabetic intervertebral disc degeneration. J Orthop Res.
The expression of hypoxia inducible factor-1alpha and apoptosis in doi:10.1002/jor.22289
herniated discs. Spine 31:13091313 Jos-Edwards DS, Kerner P, Kugler JE, Deng W, Jiang D, Di Gregorio
Hall AC, Bush PG (2001) The role of a swelling-activated taurine trans- A (2011) The identification of transcription factors expressed in the
port pathway in the regulation of articular chondrocyte volume. notochord of Ciona intestinalis adds new potential players to the
Pflugers Arch 442:771781 brachyury gene regulatory network. Dev Dyn 240:17931805
Halterman JA, Kwon HM, Zargham R, Bortz PD, Wamhoff BR (2011) Kanichai M, Ferguson D, Prendergast PJ, Campbell VA (2008) Hypoxia
Nuclear factor of activated T cells 5 regulates vascular smooth mus- promotes chondrogenesis in rat mesenchymal stem cells: a role for
cle cell phenotypic modulation. Arterioscler Thromb Vasc Biol AKT and hypoxia-inducible factor (HIF)-1alpha. J Cell Physiol
31:22872296 216:708715
Hasler U, Jeon US, Kim JA, Mordasini D, Kwon HM, Fraille E et al Khan WS, Adesida AB, Hardingham TE (2007) Hypoxic conditions
(2006) Tonicity-responsive enhancer binding protein is an essential increase hypoxia-inducible transcription factor 2alpha and enhance
regulator of aquaporin-2 expression in renal collecting duct princi- chondrogenesis in stem cells from the infrapatellar fat pad of
pal cells. J Am Soc Nephrol 17:15211531 osteoarthritis patients. Arthritis Res Ther 9:R55
Hassler O (1969) The human intervertebral disc. A micro-angiographi- Kitagawa H, Ujikawa M, Sugahara K (1996) Developmental changes in
cal study on its vascular supply at various ages. Acta Orthop Scand serum UDP-GlcA:chondroitin glucuronyltransferase activity. J Biol
40:765772 Chem 271:65836585
Henze AT, Riedel J, Diem T, Wenner J, Flamme I, Pouyseggur J et al Ko BC, Ruepp B, Bohren KM, Gabbay KH, Chung SS (1997)
(2010) Prolyl hydroxylases 2 and 3 act in gliomas as protective Identification and characterization of multiple osmotic response
negative feedback regulators of hypoxia-inducible factors. Cancer sequences in the human aldose reductase gene. J Biol Chem
Res 70:357366 272:1643116437
Higgins DF, Biju MP, Akai Y, Wutz A, Johnson RS, Haase VH (2004) Kditz J, Nesper J, Wottawa M, Stiehl DP, Camenisch G, Franke C et al
Hypoxic induction of ctgf is directly mediated by Hif-1. Am J (2007) Oxygen-dependent ATF-4 stability is mediated by the PHD3
Physiol Renal Physiol 287:F1223F1232 oxygen sensor. Blood 110:36103617
Hilton MJ, Tu X, Wu X, Bai S, Zhao H, Kobayashi T et al (2008) Notch Kokubo Y, Uchida K, Kobayashi S, Yayama T, Sato R, Nakajima H et al
signaling maintains bone marrow mesenchymal progenitors by sup- (2008) Herniated and spondylotic intervertebral discs of the human
pressing osteoblast differentiation. Nat Med 14:306314 cervical spine: histological and immunohistological findings in 500
Hiyama A, Gajghate S, Sakai D, Mochida J, Shapiro IM, Risbud MV en bloc surgical samples. Laboratory investigation. J Neurosurg
(2009) Activation of TonEBP by calcium controls b1,3-glucurono- Spine 9:285295
syltransferase-I expression, a key regulator of glycosaminoglycan Kondo S, Kubota S, Mukudai Y, Moritani N, Nishida T, Matsushita H
synthesis in cells of the intervertebral disc. J Biol Chem 284: et al (2006) Hypoxic regulation of stability of connective tissue
98249834 growth factor/CCN2 mRNA by 3-untranslated region interacting
Hiyama A, Gogate SS, Gajghate S, Mochida J, Shapiro IM, Risbud MV with a cellular protein in human chondrosarcoma cells. Oncogene
(2010) BMP-2 and TGF-beta stimulate expression of beta1,3- 25:10991110
glucuronosyl transferase 1 (GlcAT-1) in nucleus pulposus cells Lafont JE, Talma S, Murphy CL (2007) Hypoxia-inducible factor
through AP1, TonEBP, and Sp1: role of MAPKs. J Bone Miner Res 2alpha is essential for hypoxic induction of the human articular
25:11791190 chondrocyte phenotype. Arthritis Rheum 56:32973306
Hiyama A, Skubutyte R, Markova D, Anderson DG, Yadla S, Sakai D Lam AK, Ko BC, Tam S, Morris R, Yang JY, Chung SK et al (2004)
et al (2011) Hypoxia activates the notch signaling pathway in cells Osmotic response element-binding protein (OREBP) is an essential
of the intervertebral disc: implications in degenerative disc disease. regulator of the urine concentrating mechanism. J Biol Chem
Arthritis Rheum 63:13551364 279:4804848054
Lee DC, Adams CS, Albert TJ, Shapiro IM, Evans SM, Koch CJ (2007) Pfander D, Cramer T, Schipani E, Johnson RS (2003) HIF-1alpha
In situ oxygen utilization in the rat intervertebral disc. J Anat controls extracellular matrix synthesis by epiphyseal chondrocytes.
210:294303 J Cell Sci 116:18191826
Li SZ, McDill BW, Kovach PA, Ding L, Go WY, Ho SN et al (2007) Pritchard S, Erickson GR, Guilak F (2002) Hyperosmotically induced
Calcineurin-NFATc signaling pathway regulates AQP2 expression volume change and calcium signaling in intervertebral disk cells:
in response to calcium signals and osmotic stress. Am J Physiol Cell the role of the actin cytoskeleton. Biophys J 83:25022510
Physiol 292:C1606C1616 Rajpurohit R, Risbud MV, Ducheyne P, Vresilovic EJ, Shapiro IM
Lopez-Rodrguez C, Aramburu J, Rakeman AS, Rao A (1999) NFAT5, (2002) Phenotypic characteristics of the nucleus pulposus: expres-
A constitutively nuclear NFAT protein that does not cooperate with sion of hypoxia inducing factor-1, glucose transporter-1 and MMP-
Fos and Jun. Proc Natl Acad Sci U S A 96:72147219 2. Cell Tissue Res 308:401407
Lopez-Rodriguez C, Antos CL, Shelton JM, Richardson JA, Lin F, Rathmell JC, Fox CJ, Plas DR, Hammerman PS, Cinalli RM, Thompson
Novobrantseva TI et al (2004) Loss of NFAT5 results in renal atro- CB (2003) Akt-directed glucose metabolism can prevent Bax con-
phy and lack of tonicity-responsive gene expression. Proc Natl Acad formation change and promote growth factor-independent survival.
Sci U S A 101:23922397 Mol Cell Biol 23:73157328
Luo W, Hu H, Chang R, Zhong J, Knabel M, OMeally R et al (2011) Richardson SM, Knowles R, Tyler J, Mobasheri A, Hoyland JA (2008)
Pyruvate kinase M2 is a PHD3-stimulated coactivator for hypoxia- Expression of glucose transporters GLUT-1, GLUT-3, GLUT-9 and
inducible factor 1. Cell 145:732744 HIF-1alpha in normal and degenerate human intervertebral disc.
Maouyo D, Kim JY, Lee SD, Wu Y, Woo SK, Kwon HM (2002) Mouse Histochem Cell Biol 129:503511
TonEBP-NFAT5: expression in early development and alternative Rim JS, Atta MG, Dahl SC, Berry GT, Handler JS (1998) Kwon HM
splicing. Am J Physiol 282:F802F809 (1998) transcription of the sodium/myo-inositol cotransporter gene
Marxsen JH, Stengel P, Doege K, Heikkinen P, Jokilehto T, Wagner T is regulated by multiple tonicity-responsive enhancers spread over
et al (2004) Hypoxia-inducible factor-1 (HIF-1) promotes its degra- 50 kilobase pairs in the 5-flanking region. J Biol Chem 273:
dation by induction of HIF-alpha-prolyl-4-hydroxylases. Biochem J 2061520621
381:761767 Risbud MV, Fertala J, Vresilovic EJ, Albert TJ, Shapiro IM (2005a)
Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Nucleus pulposus cells upregulate PI3K/Akt and MEK/ERK signal-
Cockman ME et al (1999) The tumour suppressor protein VHL tar- ing pathways under hypoxic conditions and resist apoptosis induced
gets hypoxia-inducible factors for oxygen-dependent proteolysis. by serum withdrawal. Spine 30:882889
Nature 399:271275 Risbud MV, Guttapalli A, Albert TJ, Shapiro IM (2005b) Hypoxia acti-
Minamishima YA, Moslehi J, Bardeesy N, Cullen D, Bronson RT, vates MAPK activity in rat nucleus pulposus cells: regulation of
Kaelin WG (2008) Somatic inactivation of the PHD2 prolyl hydrox- integrin expression and cell survival. Spine 30:25032509
ylase causes polycythemia and congestive heart failure. Blood Risbud MV, Guttapalli A, Stokes DG, Hawkins D, Danielson KG,
111:32363244 Schaer TP et al (2006a) Nucleus pulposus cells express HIF-1alpha
Miyakawa H, Woo SK, Chen CP, Dahl SC, Handler JS, Kwon HM (1998) under normoxic culture conditions: a metabolic adaptation to the
Cis- and trans-acting factors regulating transcription of the BGT1 intervertebral disc microenvironment. J Cell Biochem 98:152159
gene in response to hypertonicity. Am J Physiol 274:F753F761 Risbud MV, Di Martino A, Guttapalli A, Seghatoleslami R, Denaro V,
Miyakawa H, Woo SK, Dahl SC, Handler JS, Kwon HM (1999) Vaccaro AR et al (2006b) Toward an optimum system for interver-
Tonicity-responsive enhancer binding protein, a rel-like protein that tebral disc organ culture: TGF-beta 3 enhances nucleus pulposus
stimulates transcription in response to hypertonicity. Proc Natl Acad and annulus fibrosus survival and function through modulation of
Sci U S A 96:25382542 TGF-beta-R expression and ERK signaling. Spine 31:884890
Murakami U, Kameyama Y (1963) Vertebral malformation in the Risbud MV, Guttapalli A, Tsai TT, Lee JY, Danielson KG, Vaccaro AR
mouse foetus caused by maternal hypoxia during early stage of et al (2007) Evidence for skeletal progenitor cells in the degenerate
pregnancy. J Embryol Exp Morphol 11:107118 human intervertebral disc. Spine 32:25372544
Mwale F, Ciobanu I, Giannitsios D, Roughley P, Steffen T, Antoniou J Risbud MV, Schipani E, Shapiro IM (2010) Hypoxic regulation of
(2011) Effect of oxygen levels on proteoglycan synthesis by inter- nucleus pulposus cell survival: from niche to notch. Am J Pathol
vertebral disc cells. Spine (Phila Pa 1976) 36:E131E138 176:15771583
Nerlich AG, Schaaf R, Wlchli B, Boos N (2007) Temporo-spatial dis- Roberts S, Evans H, Trivedi J, Menage J (2006) Histology and pathol-
tribution of blood vessels in human lumbar intervertebral discs. Eur ogy of the human intervertebral disc. J Bone Joint Surg Am 88:
Spine J 16:547555 1014
Ng L, Grodzinsky AJ, Patwari P, Sandy J, Plaas A, Ortiz C (2003) Rudert M, Tillmann B (1993) Lymph and blood supply of the human
Individual cartilage aggrecan macromolecules and their constituent intervertebral disc. Cadaver study of correlations to discitis. Acta
glycosaminoglycans visualized via atomic force microscopy. J Struct Orthop Scand 64:3740
Biol 143:242257 Sakai D, Nakamura Y, Nakai T, Mishima T, Kato S, Grad S et al (2012)
Oka N, Nakahara S, Takenaka Y, Fukumori T, Hogan V, Kanayama HO Exhaustion of nucleus pulposus progenitor cells with ageing and
et al (2005) Galectin-3 inhibits tumor necrosis factor-related apop- degeneration of the intervertebral disc. Nat Commun 3:1264.
tosis-inducing ligand-induced apoptosis by activating Akt in human doi:10.1038/ncomms2226
bladder carcinoma cells. Cancer Res 65:75467553 Sasaki T, Brakebusch C, Engel J, Timpl R (1998) Mac-2 binding pro-
Oktay Y, Dioum E, Matsuzaki S, Ding K, Yan LJ, Haller RG et al tein is a cell-adhesive protein of the extracellular matrix which self-
(2007) Hypoxia-inducible factor 2 alpha regulates expression of the assembles into ring-like structures and binds beta1 integrins,
mitochondrial aconitase chaperone protein frataxin. J Biol Chem collagens and fibronectin. EMBO J 17:16061613
282:1175011756 Schipani E, Ryan HE, Didrickson S, Kobayashi T, Knight M, Johnson
Papandreou I, Cairns RA, Fontana L, Lim AL, Denko NC (2006) HIF-1 RS (2001) Hypoxia in cartilage: HIF-1 alpha is essential for chon-
mediates adaptation to hypoxia by actively downregulating mito- drocyte growth arrest and survival. Genes Dev 15:28652876
chondrial oxygen consumption. Cell Metab 3:187197 Scortegagna M, Ding K, Oktay Y, Gaur A, Thurmond F, Yan LJ et al
Parvizi J, Parpura V, Greenleaf JF, Bolander ME (2002) Calcium sig- (2003) Multiple organ pathology, metabolic abnormalities and
naling is required for ultrasound-stimulated aggrecan synthesis by impaired homeostasis of reactive oxygen species in Epas1/ mice.
rat chondrocytes. J Orthop Res 20:5157 Nat Genet 35:331340
Semenza GL, Roth PH, Fang HM, Wang GL (1994) Transcriptional Tsai TT, Guttapalli A, Agrawal A, Albert TJ, Shapiro IM, Risbud MV
regulation of genes encoding glycolytic enzymes by hypoxia- (2007) MEK/ERK signaling controls osmoregulation of nucleus
inducible factor 1. J Biol Chem 269:2375723763 pulposus cells of the intervertebral disc by transactivation of
Setton LA, Chen J (2006) Mechanobiology of the intervertebral disc TonEBP/OREBP. J Bone Miner Res 22:965974
and relevance to disc degeneration. J Bone Joint Surg Am Urban JP, Maroudas A (1981) Swelling of the intervertebral disc
88:5257 in vitro. Connect Tissue Res 9:110
Shapiro IM, Vresilovic EJ, Risbud MV (2012) Is the spinal motion seg- Urban JP, Holm S, Maroudas A, Nachemson A (1977) Nutrition of the
ment a diarthrodial polyaxial joint: what a nice nucleus like you intervertebral disk. An in vivo study of solute transport. Clin Orthop
doing in a joint like this? Bone 50:771776 Relat Res 129:101114
Shim EH, Kim JI, Bang ES, Heo JS, Lee JS, Kim EY et al (2002) Venkatesan N, Barr L, Benani A, Netter P, Magdalou J, Fournel-
Targeted disruption of hsp70.1 Sensitizes to osmotic stress. EMBO Gigleux S et al (2004) Stimulation of proteoglycan synthesis by
Rep 3:857861 glucuronosyltransferase-I gene delivery: a strategy to promote carti-
Smits P, Lefebvre V (2003) Sox5 and Sox6 are required for notochord lage repair. Proc Natl Acad Sci U S A 101:1808718092
extracellular matrix sheath formation, notochord cell survival and Verkman AS (2006) Roles of aquaporins in kidney revealed by trans-
development of the nucleus pulposus of intervertebral discs. genic mice. Semin Nephrol 26:200208
Development 130:11351148 Vijayagopal P, Subramaniam P (2001) Effect of calcium channel block-
Sowter HM, Raval RR, Moore JW, Ratcliffe PJ, Harris AL (2003) ers on proteoglycan synthesis by vascular smooth muscle cells and
Predominant role of hypoxia-inducible transcription factor (Hif)-1 low density lipoproteinproteoglycan interaction. Atherosclerosis
alpha versus Hif-2 alpha in regulation of the transcriptional response 157:353360
to hypoxia. Cancer Res 63:61306134 Wang GL, Jiang BH, Rue EA, Semenza GL (1995) Hypoxia-inducible
Sparrow DB, Chapman G, Smith AJ, Mattar MZ, Major JA, OReilly factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by
VC et al (2012) A mechanism for gene-environment interaction in cellular O2 tension. Proc Natl Acad Sci U S A 92:55105514
the etiology of congenital scoliosis. Cell 149:295306 Wang Y, Ko BC, Yang JY, Lam TT, Jiang Z, Zhang J et al (2005)
Srinivas V, Bohensky J, Zahm AM, Shapiro IM (2009) Autophagy in Transgenic mice expressing dominant-negative osmotic-response
mineralizing tissues: microenvironmental perspectives. Cell Cycle element-binding protein (OREBP) in lens exhibit fiber cell elonga-
8:391393 tion defect associated with increased DNA breaks. J Biol Chem
Takeda K, Ho VC, Takeda H, Duan LJ, Nagy A, Fong GH (2006) 280:1998619991
Placental but not heart defects are associated with elevated hypoxia- Woo SK, Lee SD, Na KY, Park WK, Kwon HM (2002) TonEBP/NFAT5
inducible factor alpha levels in mice lacking prolyl hydroxylase stimulates transcription of HSP70 in response to hypertonicity. Mol
domain protein 2. Mol Cell Biol 26:83368346 Cell Biol 22:57535760
Thoms BL, Murphy CL (2010) Inhibition of hypoxia-inducible factor- Wuertz K, Urban JP, Klasen J, Ignatius A, Wilke HJ, Claes L et al
targeting prolyl hydroxylase domain-containing protein 2 (PHD2) (2007) Influence of extracellular osmolarity and mechanical stimu-
enhances matrix synthesis by human chondrocytes. J Biol Chem lation on gene expression of intervertebral disc cells. J Orthop Res
285:2047220480 25:15131522
Trama J, Lu Q, Hawley RG, Ho SN (2000) The NFAT-related protein Ye W, Xu K, Huang D, Liang A, Peng Y, Zhu W et al (2011) Age-
NFATL1 (TonEBP/NFAT5) is induced upon T cell activation in a related increases of macroautophagy and chaperone-mediated
calcineurin-dependent manner. J Immunol 165:48844894 autophagy in rat nucleus pulposus. Connect Tissue Res 52:472478
Tran CM, Markova D, Smith HE, Susarla B, Ponnappan RK, Anderson Zelzer E, Mamluk R, Ferrara N, Johnson RS, Schipani E, Olsen BR
DG et al (2010) Regulation of CCN2/CTGF expression in the (2004) VEGFA is necessary for chondrocyte survival during bone
nucleus pulposus of the intervertebral disc: role of smad and AP1 development. Development 131:21612171
signaling. Arthritis Rheum 62:19831992 Zeng Y, Danielson KG, Albert TJ, Shapiro IM, Risbud MV (2007)
Tran CM, Fujita N, Huang BL, Ong J, Lyons KM, Shapiro IM et al HIF-1 alpha is a regulator of galectin-3 expression in the interverte-
(2013) HIF-1a and CCN2 form a regulatory circuit in hypoxic bral disc. J Bone Miner Res 22:18511861
nucleus pulposus cells: CCN2 suppresses HIF-1a level and tran- Zhang Z, Ferraris JD, Brooks HL, Brisc I, Burg MB (2003) Expression
scriptional activity. J Biol Chem 288:1265412666 of osmotic stress-related genes in tissues of normal and hyposmotic
Tsai TT, Danielson KG, Guttapalli A, Oguz E, Albert TJ, Shapiro IM rats. Am J Physiol 285:F688F693
et al (2006) TonEBP/OREBP is a regulator of nucleus pulposus cell Zhang H, Bosch-Marce M, Shimoda LA, Tan YS, Baek JH, Wesley JB
function and survival in the intervertebral disc. J Biol Chem et al (2008) Mitochondrial autophagy is an HIF-1-dependent adap-
281:2541625424 tive metabolic response to hypoxia. J Biol Chem 283:1089210903
The Effects of Mechanical Forces
on Nucleus Pulposus and Annulus 7
Fibrosus Cells
Jeffrey C. Lotz and Adam H. Hsieh

7.1 Introduction ..................................................................... 109
The relationship between mechanical loading and
7.2 Organ and Tissue Levels................................................. 109
7.2.1 Physiologic Interplay in Mechanical Function
intervertebral disc health and disease has long been rec-
in the Nucleus Pulposus and Annulus Fibrosus ognized (Panel on Musculoskeletal Disorders and the
Is Critical for Tissue Homeostasis .................................... 110 Workplace 2001). While it is well established that mechan-
7.2.2 Implications of Poroviscoelasticity in Nucleus ical factors can significantly influence cell function (Nelson
Pulposus and Annulus Fibrosus Mechanobiology ............ 111
et al. 2005; Hoffman and Crocker 2009), how this process
7.3 Cellular and Molecular Levels ....................................... 113 influences nucleus pulposus and annulus fibrosus activity in
7.3.1 Studies of Disc Cell Mechanical Responses In Vitro ....... 113
the intervertebral disc is complex and confounded by mul-
7.3.2 Disc Cell Responses to Mechanical Perturbation ............. 117
7.3.3 Disc Cell Response to Pressure ......................................... 117 tiple, hierarchical mechanisms by which load affects the
7.3.4 Disc Cell Response to Stretch ........................................... 118 human body. At the whole-body level, spinal forces depend
7.4 Final Comments .............................................................. 120 on body mass, external loads, posture, muscle function, and
body resonant frequencies. At the organ level, disc loads
vary with duration, spinal level, and posture. At the tissue
in the Chapter .................................................................. 121
level, stress and strains are heterogeneously distributed
References ..................................................................................... 121 within the disc in a time-dependent manner. Ultimately, disc
cells convert local physical cues into biochemical signals
and integrate these into cellular responses. Adding further
complexity is the fact that local cellular responses influence
the entire spinal system by modifying disc material proper-
ties (thereby altering organ-level behaviors) or triggering
pain (thereby altering muscle function and whole-body
mechanics). This chapter dissects this highly complex pro-
cess and identifies patterns in disc mechanobiology at vari-
ous levels of scale, disc regions, and cell types. Clarification
of these relationships is relevant to understanding disease
J.C. Lotz (*) mechanisms and for developing treatments for patients with
Orthopaedic Bioengineering Laboratory, spinal pathology.
Department of Orthopaedic Surgery,
University of California, San Francisco,
513 Parnassus Ave, 11th Floor, S1157,
San Francisco, CA 94143-0514, USA 7.2 Organ and Tissue Levels
e-mail: jeffrey.lotz@ucsf.edu
A.H. Hsieh Discussion of mechanotransduction by nucleus pulpo-
Orthopaedic Mechanobiology Laboratory, sus and annulus fibrosus cells must begin at the level of
Fischell Department of Bioengineering, the functional spinal unit, since the transmission of loads
University of Maryland, College Park,
through the spinal column dictates the magnitude and types
Jeong H. Kim Engineering Building, Rm 3242,
College Park, MD 20742, USA of intervertebral disc stress. At each spinal level, loads are
e-mail: hsieh@umd.edu shared between the intervertebral disc anteriorly and the

110 J.C. Lotz and A.H. Hsieh
degeneration and disease (Urban et al. 2000). As such, the

disc represents a uniquely complex organ of the musculo-
skeletal system.
Based on how forces are distributed from organ to tissue
levels, several mechanotransduction themes have emerged
from the growing body of literature. The first is that the
magnitude and distribution of loads within the nucleus pul-
posus and annulus fibrosus govern disc cell function. The
second is that load frequency and duration significantly
contribute to the discs mechanical and mechanobiologic
Pressure response. And the third is that age-related matrix changes
may set off a vicious cycle of degeneration-promoting pro-
cesses. In this section, we will examine the evidence that
support these concepts.
Given the discs complex matrix structure, mechanobio-
logic interactions are influenced at multiple hierarchical lev-
els. Within each disc sub-tissue, a unique combination of
extracellular matrix fibers and ground substance confer dif-
ferent physical properties, where the extracellular matrix
molecules are adapted to manage and respond to mechanical
stress. Consequently, the extracellular matrix physical prop-
erties are dynamic and regulated, and create a microenviron-
ment that presents cells with physical and biochemical cues
that are important for maintenance of a healthy tissue. Cues
include chemical (hypoxia, pH), biochemical (growth fac-
Fig. 7.1 Static loading of the spine under a follower load configuration, tors, cytokines, neurotrophins, hormones), and physical
resulting in a primarily axial compression of the intervertebral disc (topography, fluid flow, stiffness) factors.
two facet joints posteriorly. Generally speaking, the discs

principal mechanical role is to support compressive forces, 7.2.1 Physiologic Interplay in Mechanical
while the facets resist shear and guide segmental rotations Function in the Nucleus Pulposus
during spinal movements (Shirazi-Adl and Drouin 1987). and Annulus Fibrosus Is Critical
Disc/facet interactions vary between the multiple spinal for Tissue Homeostasis
levels as the biomechanical behaviors change from the
cervical (high rotation/low compression) to lumbar (low Intervertebral disc loading depends on an individuals size,
rotation/high compression) regions. Proper biomechanical physical activity level, occupational demands, and degree of
synergy between the disc and facets is lost when degenera- rest. In the human, spinal stability requires that the direction
tion reduces disc height and compliance (Niosi and Oxland of the resultant force vectors at each vertebra passes through
2004). The disc performs its mechanical function by syner- the centers of rotation of adjacent motion segments in the sag-
gistic interactions between the hydrophilic nucleus pulpo- ittal plane, a concept called follower load path (Patwardhan
sus, fibrous annulus fibrosus, and cartilage/bone composite et al. 1999). The follower load strategy allows the spine to
vertebral end plate. Each disc sub-tissue has a different support static loads up to and significantly above typical
composition of cells, matrix fibers (collagen, elastin, retic- physical demands (Patwardhan et al. 2000), without markedly
ular), and ground substance (glycosaminoglycans (GAG), sacrificing flexibility or range of motion (Rohlmann et al.
proteoglycans, glycoproteins) that define signature physical 2001; Patwardhan et al. 2003). This suggests that during static
properties uniquely adapted for a specific role. Spinal load- conditions muscle activation patterns in vivo cause the disc to
ing is spatially filtered into regions of hydrostatic pressure be primarily loaded in axial compression (Fig. 7.1), a loading
and octahedral shear that correlate with variations in cell mode that has been extensively studied in animal models.
phenotype and extracellular matrix composition, consistent In the healthy spine, the intervertebral disc responds to
with current theories of tissue adaptation and homeostasis axial compression by nucleus pulposus pressurization facili-
(Carter and Beaupre 2001). These interdependent func- tated by lateral annulus fibrosus constraint, which in turn
tions are specified at development and regulated during generates circumferential and longitudinal annulus fibrosus
growth (Hayes et al. 2001) and when disrupted can lead to tension. Preservation of this structural interrelationship is
7 The Effects of Mechanical Forces on Nucleus Pulposus and Annulus Fibrosus Cells 111
important for tissue maintenance. This is perhaps most nucleus pulposus pressurization (and consequently annulus
clearly demonstrated in static compression of rodent discs. fibrosus tension) also trigger degenerative changes in the
In such experiments, static loads induce fluid shifts that annulus with increased MMP-2 immunoreactivity, while those
deplete nucleus pulposus volume, causing the annulus that are too small to affect disc pressurization do not (Rousseau
fibrosus to function more as a compression-bearing strut than et al. 2007; Hsieh et al. 2009; Rastogi et al. 2013).
a biaxially stretched membrane (Lotz et al. 1998). As fluid is
expelled from the nucleus, the matrix and cells consolidate
(Lotz 2004; Hsieh et al. 2005), ultimately leading to deregu- 7.2.2 Implications of Poroviscoelasticity in
lated gene expression (Lotz et al. 1998), increased metallo- Nucleus Pulposus and Annulus Fibrosus
proteinase (MMP) activation (Hsieh and Lotz 2003), and Mechanobiology
apoptosis (Chin et al. 1999; Lotz and Chin 2000). In both
mice (Lotz et al. 1998) and rats (Iatridis et al. 1999), these Although the spine is dynamically loaded over the course of
cellular effects alter disc architecture and mechanics. the day, trunk muscles stabilizing the spine generate resul-
In contrast, beneficial effects of nucleus pulposus pressur- tant loads that subject discs to sustained, time-averaged com-
ization and annulus fibrosus tension are observed when the disc pression. Due to the discs poroviscoelastic nature, this
is subjected to cyclic loading. Low load magnitudes are most sustained compression results in loss of stature, contributed
amenable to maintaining homeostatic stress environments in large part by decreases in spine length (Koeller et al. 1984;
in the nucleus pulposus and annulus fibrosus over extended Leatt et al. 1986). Even low-impact activities, such as gentle
loading durations. For example, under low cyclic loads com- walking, have been shown to decrease stature significantly
parable to normal activities, nucleus pulposus cells exhibit lit- (Hoe et al. 1994), although more strenuous exercise, work-
tle alteration in expression of genes that would lead to matrix related activities, and high body mass exacerbate this loss
remodeling, regardless of loading frequency (Maclean et al. (Fig. 7.2, top) (Garbutt et al. 1990; McGill et al. 1996;
2004). Similarly, annulus fibrosus cells undergo a low level Leivseth and Drerup 1997; Rodacki et al. 2005). These stat-
of apoptosis and exhibit depressed expression of certain cata- ure changes are primarily attributed to the redistribution of
bolic factors such as MMPs and ADAMTSs (Maclean et al. fluid as the applied stresses exceed the discs swelling pres-
2004; Walsh and Lotz 2004). High magnitudes of stress can sure (Adams and Hutton 1983; Koeller et al. 1984; Adams
be similarly accommodated for brief durations and frequen- et al. 1990; Terahata et al. 1994; Ayotte et al. 2000; Hsieh
cies, near 1 Hz. However, if applied at low frequency or long et al. 2005; Schroeder et al. 2006).
durations, high compression magnitudes lead to a shift in cell As with other compressive load-bearing tissues such as
function and tissue morphology. Under these circumstances, articular cartilage, the discs extracellular solidfluid interac-
nucleus pulposus cells respond via upregulation of genes tions define its viscoelastic behavior. Interstitial fluid in the
associated with extracellular matrix remodeling (Maclean hydrated extracellular matrix serves as a hydraulic cushion
et al. 2004; MacLean et al. 2005; Wuertz et al. 2009), leading that helps distribute forces and absorb shock. Sustained inter-
to decreases in disc height and mechanical stability (Ching vertebral disc compression consolidates the disc extracellular
et al. 2003, 2004). In the annulus fibrosus, lower loading fre- matrix as there is an efflux of interstitial fluid and a decrease
quencies also cause a shift in the balance of gene expression in water content (Fig. 7.2, bottom) (Adams et al. 1990). The
toward a more catabolic profile (Maclean et al. 2004). importance of tissue hydration in intervertebral disc mechan-
Taken together, these studies demonstrate that physiologic ics has been demonstrated in a number of studies that reveal a
load sharing between nucleus pulposus and annulus fibrosus close relationship between hydration and disc mechanical
(principally nucleus pulposus pressure and annulus fibrosus properties (Bass et al. 1997; Pflaster et al. 1997; Race et al.
tension) promotes tissue homeostasis. This is underscored by 2000; Han et al. 2001; Costi et al. 2002). Of note, varying
experiments in which specific aspects of this interplay are water content through load, by imposition of free swelling and
selectively modulated. For instance, using mouse models, disc creep compression boundary conditions, reveals the existence
bending to induce tension and compression on opposing of an optimal hydration point at which the effective modulus
regions of the annulus stimulates apoptosis and altered gene reaches a peak value (Race et al. 2000). The principles behind
expression only on the side of compression (Court et al. 2001, this have not yet been elucidated, but likely involve both phys-
2007). The restoration of annulus fibrosus tension has simi- ical relationships among subregions, consolidation of the tis-
larly been demonstrated to be potentially protective. Applying sue, and transient solidfluid interactions in the tissue.
bending to discs that had previously been subjected to degen- Nevertheless, these observations highlight the importance of
eration-inducing static compression was effective in preserv- water content in the discs mechanical function.
ing annulus fibrosus lamellar morphology, even though Movement of moisture due to poroviscoelasticity is
apoptosis could not be mitigated (Lotz et al. 2008). These particularly relevant in nucleus pulposus mechanobiology.
effects can also be observed when needle stabs that reduce In human discs, the swelling properties of the nucleus
Fig. 7.2 Top: changes in stature 4

due to various levels of activity
(Adapted from Leatt et al. 1986;
Garbutt et al. 1990; Hwang et al.
2012). Bottom: hydration profile
Normalized height loss

3
of a human intervertebral disc
before and after 4 h of sustained
loading in a flexed position
(Adapted from Adams et al. 1990) 2
Baseline Running Running Running Weight

running enchanced maximum 3 distance circuit
effort effort training
Water content
Non-loaded
Loaded
Relative A-P position
pulposus are well recognized as a prominent contributor to pulposus tissues, intervertebral disc biomechanics, and
intervertebral disc mechanics. As cut from the disc, the degenerative changes in animal models (Boxberger et al.
nucleus pulposus has a water content (WC = 1 dry weight/ 2008).
wet weight) of approximately 0.85 in juvenile discs, and Nucleus pulposus poroviscoelastic behavior, in turn, impacts
that decreases to approximately 0.75 in aged discs (Urban the loading experienced by the annulus fibrosus, which itself is
and McMullin 1988). Distribution of water also dimin- porous and viscoelastic. However, consistent with the similari-
ishes outwardly through annular lamellae, as a proportion ties to ligament tissue in terms of function and microstructure,
of the nucleus pulposus water content, by 0.07, 0.11, and the contribution to viscoelastic properties in the annulus fibrosus
0.22 from inner to mid to outer annulus fibrosus (Urban is thought to be primarily flow independent. Specifically, it
and Maroudas 1981). The marked spatial variation in fluid relies on collagen fiber bundles organized in parallel to resist
distribution is a unique hallmark of the intervertebral disc tensile stresses (Broberg and von Essen 1980; Hickey and
and plays a significant role in its function. In the nucleus Hukins 1980; Stokes and Greenaple 1985; Cassidy et al. 1989;
pulposus, a high negative fixed charge density from con- Holzapfel et al. 2005). The staggered discontinuous nature of
centrated localization of large aggregating proteoglycans the fibrillar collagen microstructure (Marchand and Ahmed
produces increases in osmotic pressures that tend to pro- 1990; Holzapfel et al. 2005) results in complex transmission of
mote tissue swelling and affects mechanically driven fluid force along lamellae and has profound impact on mechan-
exudation (Urban and Maroudas 1981; Urban and otransduction. As observed in the bovine annulus fibrosus
McMullin 1985, 1988). Studies have demonstrated that (Bruehlmann et al. 2004a, b) and similarly in the rat tail tendon
matrix alteration particularly GAG degradation (Screen et al. 2004), tissue stretch does not uniformly propagate
adversely affects the poroviscoelastic behavior of nucleus across the levels of collagen. Rather, straightening of collagen
crimp and fibril-fibril sliding leads to nonuniformly, increasing Fluid redistribution

fibril recruitment at different loads and appears to account for
the majority of tissue-level deformations. These phenomena Collagen
have been demonstrated both through intercellular measure- Pressure fibril
ments within and among collagen fibers (Bruehlmann et al. sliding
2004a, b; Screen et al. 2004) and by measuring the deformation

of photobleached lines across labeled collagen fibers
(Bruehlmann et al. 2004b). The mechanisms and kinetics of 0.6
recovery, whether passive or actively cell-mediated, are not yet Equal final
clear. Thus, not only is cell stretch a mechanism of stimulation applied stress
stress (MPa)
0.4
Applied
during annulus fibrosus deformation, but shear stress likely also
serves as a robust determinant of cell function.
0.2
Poroviscoelastic interactions between the nucleus pulposus
and annulus fibrosus cause nucleus pressurization and annulus
tension to be time dependent, modulated by the nucleus pulpo- 0
0 1,000 2,000 3,000
sus hydration state and degree of annulus fibrosus stress relax-
ation (Fig. 7.3). Thus, for any spinal load applied over a finite 600
duration, the micromechanical loading regime imposed on
pressure (kPa)
disc cells is defined both by the current stress to the motion Unequal final
Intradiscal
400 intradiscal
segment and by the history of earlier experienced applied
pressures
stresses. One demonstration of this dependence on the time
history of loading is illustrated in the study of Hwang et al. 200
(2012). This study showed that in rat discs, the effectiveness of
the nucleus pulposus to pressurize varied according to the
0
specific sequence of loading events and not just by the applied 0 1,000 2,000 3,000
endpoint stress. Meanwhile, disc height was determined only Time (s)
by the endpoint stress and independent of temporal path of
Fig. 7.3 Poroviscoelastic phenomena affect the internal mechanics of
axial loads. This load history dependence of the internal pres- the disc (top). Graphs show that two temporal paths (solid and dotted
sure and shear could then have an impact on cellular mechan- curves) of applied stress (middle) lead to markedly different intradiscal
otransduction and differentiation. pressures generated (bottom). Despite equivalent applied stresses at the
As a corollary, there is also the implicit potential for restoring end, the path with a higher intermediate load exhibited inferior final
pressures (solid, bottom) than that with the lower intermediate load
the homeostatic combination of nucleus pulposus pressure and (dotted, bottom) (Adapted from Hwang et al. 2012)
annulus fibrosus shear. Introducing detours in the temporal path
of spinal loading could have restorative effects on intervertebral
disc hydration. Results from Gabai et al. (2007) using rat discs may represent a state in which an adverse pressure-shear
suggest that interspersing repeated axial compressive stresses microenvironment is permanently maintained, resulting in a
with short, low-magnitude tensional loading allows interverte- vicious cycle of matrix catabolism and abnormal matrix syn-
bral discs to preserve the pressure-shear balance between nucleus thesis. Evidence in the literature demonstrating decreases in
pulposus and annulus fibrosus. Thus, exploiting this phenome- intradiscal pressure generation and increasing numbers of
non could be one strategy toward preventative maintenance. spikes in the stress profile with age and degeneration sup-
While very pronounced in small animal discs, such load ports the notion that pressure and shear remain in a state of
history effects in adult human intervertebral discs remain imbalance in the adult disc (Adams et al. 1996).
uncharacterized. It is possible that the temporal sequence of
loading is more significant during early growth in juvenile
discs and that the decreased baseline tissue hydration and 7.3 Cellular and Molecular Levels
calcification of end plates in aged discs reduce the impact of
load history. If this is the case, the argument could be made 7.3.1 Studies of Disc Cell Mechanical
for greater focus on spinal health in younger populations. Responses In Vitro
There are some parallels between the time load history
effects and age-related changes in the intervertebral disc that Cells sense and respond to mechanical cues via a plethora
are interesting to note. Similar to the expression of fluid out of mechanisms that typically operate in four steps: mechani-
of the intervertebral disc with sustained compression, the cal coupling, mechanotransduction, signal transmission, and
aging disc possesses diminished swelling pressures and cell response. Mechanocoupling is facilitated by cellular
impaired capability to retain water. As such, the aged disc load transducers that include integrins, ion channels,
Fig. 7.4 Cells possess several Stretch-activated

mechanisms by which they sense ion channels
mechanical signals. These include Cell surface Cell membrane
the stretch-activated ion channels receptors tension
and cytoskeletal elements that are
connected to matrix via focal
Cytoskeletal Nucleus
adhesions
filaments
Focal adhesion
complexes Crosslinking
proteins
ECM
Tissue strain
Fig. 7.5 Schematic of mechanism

by which integrins interact with
adhesion plaque proteins and
cytoskeleton to activate specific Collagen
downstream signaling

Fibronectin
Integrins ECM
Cell membrane
Talin Cytosol
Vinculin
Focal adhesion
Paxillin kinase (FAK)
Actin filament Signal transduction

to nucleus
-actinin
Nucleus
G protein-coupled receptors, and tyrosine kinase receptors loading and matrix stiffness, with higher forces and stiffer
(Fig. 7.4). These lead to cell-ECM adhesion assemblies matrices leading to enhanced attachment (Paszek et al.
that are sensitive to reciprocal tension between the cell and 2005). These types of load-dependent interactions are called
matrix. catch bonds (Friedland et al. 2009). Mechanical reinforce-
Integrins are a class of cell membrane proteins that medi- ment of catch bonds between the cell and matrix mediates
ate adhesion of cells to substrates (Fig. 7.5). They consist integrin assembly and development of clusters to form focal
of two transmembrane glycoprotein subunits with the extra- adhesions. Focal adhesions elicit a reciprocal actomyosin-
cellular domains interacting to form a functional heterodi- mediated cell contractility that is essential to amplify the
mer. Integrins functionally serve to connect the extracellular mechanoresponse (Sniadecki and Chen 2007; Na et al. 2008;
matrix with the cytoskeleton and have the capacity to bind Wang et al. 2008). The affinity of integrins for extracellular
collagen and fibronectin among other matrix constituents. matrix proteins can be modified by tissue deformation and
This integrin/matrix bond may be influenced by tensile molecular conformation force-dependent unfolding (such as
with fibronectin) that can expose binding sites (Vogel and observed during osmotic swelling), it can alter the channel
Baneyx 2003). cross section and protein configuration, leading to channel
Disc cells express integrin receptors that vary by region opening, even in the absence of direct activation by a specific
and degree of degeneration. These include fibronectin- chemical ligand (Janmey and Kinnunen 2006). Calcium
binding integrins (a5b1 and avb3), collagen-binding integrins modulates cell activity, growth and differentiation, motility,
(a1b1, a2b1, and avb1), and laminin-binding integrins (a6b1 intercellular coupling, and apoptosis. Several studies demon-
and a6b4) (Nettles et al. 2004; Xia and Zhu 2008; Chen et al. strate that disc cells can respond to mechanical loading via
2009); laminin-binding integrins are more prevalent in the activation of calcium channels that cause intracellular cal-
nucleus (Gilchrist et al. 2007; Chen et al. 2009). Within the cium transients. This type of calcium signaling has been
nucleus, fibronectin fragments are known to accumulate with shown to be important in regulatory volume fluxes that are
disc degeneration (Oegema et al. 2000) and trigger upregula- observed after hypo-osmotic or fluid-induced shear stress
tion of a5b1 integrin expression and ERK signaling of cata- (Elfervig et al. 2001; Pritchard and Guilak 2004). Calcium
bolic processes (Xia and Zhu 2011). transients can lead to F-actin remodeling that facilitates rees-
Mechanotransduction occurs when extracellular matrix tablishment of cell volume after an osmotic challenge.
forces are coupled to the cytoskeleton and induce conforma- Signals are propagated deep within the cell by the stiff
tional changes of intracellular proteins that alter substrate cytoskeletal filaments (microfilaments (F-actin), interme-
availability leading to phosphorylation (Doyle and Yamada diate filaments (vimentin, cytokeratin), and microtubules
2010; del Rio et al. 2009). Mechanosensing can also occur (tubulin)) that link the cell nucleus to the extracellular
via mechanically gated calcium channels. Transmembrane matrix via focal adhesion complexes and hemidesmosomes
ion channels influence the cells lipid bilayer organization (Alenghat and Ingber 2002). Because cells contain a con-
and tension (Fig. 7.6). When membrane tension increases (as tinuum of cytoskeletal elements, they display an integrated
Collagen
Ion

Fibronectin Ion channel
ECM ECM
Cell membrane Cell membrane
Talin Cytosol Cytosol
Vinculin
Vinculin binding site
Actin filament Load application
Nucleus
Collagen
tension
Ion

Fibronectin Ion channel Cell membrane
ECM ECM
force
Cell membrane Cell membrane
Talin Cytosol Cytosol
Actin filament Focal adhesion

Paxillin kinase (FAK)
Cytoskeletal
force
-actinin
Nucleus
Fig. 7.6 Mechanotransduction can be viewed as a series of rapid switch-like events, activated in response to applications of force
mechanical behavior, whose properties depend on the more pronounced and extends into cell processes (Errington
composition and organization of the cytoskeleton, and extra- et al. 1998; Bruehlmann et al. 2002; Li et al. 2008).
cellular matrix interactions via transmembrane proteins, cel- Mechanoresponses involve downstream intracellular sig-
lular proteins, subcellular structures, and intracellular fluid naling and transcription networks. Intracellular signal trans-
volume and composition (Ingber 2003). Predictably, the duction can occur via the cytoskeleton, small molecules
mechanical properties of disc cells display zonal variations (second messengers Ca2+, 1,4,5-triphosphate (IP3), cAMP),
in cytoskeletal composition. While disc cells demonstrate protein kinases (focal adhesion kinase, cSrc, protein kinase
an overall viscoelastic behavior, nucleus pulposus cells C, mitogen-activated protein kinase), and transcription fac-
are three times stiffer and more viscous than annulus cells tors (c-fos, c-jun, c-myc, NF-kB).
(Guilak et al. 1999). Cellular response pathways function over different times-
Microtubules form an extensive mesh throughout the cyto- cales that can define frequency-dependent cell behaviors
plasm in both nucleus and annulus cells (Li et al. 2008). (Fig. 7.7). For example, force can cause an immediate cell
Likewise, vimentin filaments are densely distributed within response (hundreds of milliseconds) through activation of
the nucleus and annulus cells and extend into annular cell ion channels or conformational changes in the cytoskeleton.
processes (Johnson and Roberts 2003; Li et al. 2008). Nucleus In contrast, other signaling pathways require polymerization
pulposus cells are also characterized by the presence of cytok- of cytoskeletal stress fibers and alterations of intracellular
eratin intermediate filaments. Pronounced differences exist in protein networks and act over minutes. Still others may take
F-actin distribution between the nucleus and annulus cells. In hours or days as they act by causing changes in gene
the nucleus, F-actin is localized to punctate regions of the cell expression that influence cytoskeletal or adhesion proteins
membrane, while in the annulus along with vimentin, it is and ultimately force-transmission pathways. Consequently,
Spine loading E
D
x(t)
Matrix conformation
Porosity
Disc poroviscoelasticity Water content
Osmolality
h1(t) pH
Cytoskeletal reorganization
Cell viscoelasticity
Fig. 7.7 A conceptual scheme of Etc.
the multiscale, multifactorial h2(t)
influences on disc cell
mechanobiology. The time- y(t)
dependent cellular response (y(t))
resulting from spinal loading
(x(t)) is dependent on dynamic
tissue and cellular processes that
may be quantified by transfer h1(t) h2(t)
x(t) y(t)
functions h1(t) and h2(t),
respectively
these time-dependent behaviors lead to frequency-dependent et al. 2005; Gruber et al. 2007). These factors will lead to
signaling, where cells function as band-pass filters (Hoffman experiment-to-experiment variability and require larger sam-
et al. 2011). To be stimulated, a particular signaling pathway ple sizes for identifying statistically significant effects. In
needs to match the timescale of the applied forces. comparison, animal tissues are more readily available, but
can be limited by the presence of persistent notochordal
cells, which are lost in adolescence in humans but are main-
7.3.2 Disc Cell Responses tained throughout life in mice, rats, rabbits, pigs, cats, and
to Mechanical Perturbation non-chondrodystrophic dogs (Risbud and Shapiro 2011;
Hunter et al. 2003). The nucleus pulposus of several species
Disc cell phenotype varies between the annulus and nucleus, has been observed to convert from notochordal to chondro-
and accordingly, these cells are considered to have distinct cyte-like cells with age in a similar fashion to humans. These
repertoires of load-induced behaviors. These differences are include cattle, horses, sheep, and chondrodystrophic dogs
likely due to spatially dependent adaptation to hydrostatic (beagles) (Miyazaki et al. 2009). Not surprisingly therefore,
pressure (nucleus) and stretch (annulus) that are developed interspecies differences have been reported for disc cell
during spinal loading. To facilitate systematic study of cell responsiveness (Minogue et al. 2010; Miyazaki et al. 2009;
mechanoresponsiveness, a number of in vitro test systems Sakai et al. 2009), particularly when distinguishing noto-
have been developed (Brown 2000). These allow the con- chordal from non-notochordal nucleus tissues (Miyazaki
trolled application of known inputs (either stress or strain) so et al. 2009). Bovine tails may represent the best compromise
that doseresponse functions can be established. Independent for in vitro investigations, as they have a non-notochordal
variables in these in vitro bioreactor studies include type, mag- nucleus and are easy and inexpensive to procure in large
nitude, frequency, and duration of mechanical stimulation. numbers. For a full discussion of animal preferences in inter-
vertebral disc research, see Chap. 18.
Hydrostatic pressure chambers are most commonly used
7.3.3 Disc Cell Response to Pressure to study disc cell responsiveness to compression loading
(Hutton et al. 2001; Liu et al. 2001; Kasra et al. 2003; Wenger
The healthy disc nucleus is largely water and, consequently, et al. 2005; Reza and Nicoll 2008). Alternatively, some use
is subjected to hydrostatic pressure during spinal loading. confined or unconfined compression of 3D gel/cell con-
For that reason, both monolayer (2D) and hydrogel (3D) cul- structs. To simplify the latter tests, loading platens are often
ture systems have been used to subject nucleus pulposus operated in displacement control, so that compressive strain,
cells to hydrostatic pressure or compression. Nucleus cells rather than compressive stress, becomes the independent test
are generally considered chondrocyte-like, while annulus variable (Chen et al. 2004; Korecki et al. 2009; Fernando
fibrosus cells are fibroblastic. Accordingly, nucleus cells pre- et al. 2011; Salvatierra et al. 2011; Wang et al. 2011). Note
fer the environment of inert hydrogels such as alginate or that, in Chap. 22, further details are provided on choices of
agarose that support a spherical configuration conducive to bioreactors for disc research.
the assumption of a stable chondrocyte-like phenotype (Chen In vitro loading parameters are typically based on values
et al. 2002). In contrast, in monolayer culture, nucleus cells obtained from human disc studies. For example, studies of
dedifferentiate as cell spreading leads to significant down- human subjects suggest that due to the inherent resonant fre-
regulation of Col2 gene expression and a change to a fibrotic quency of the torso, certain loading frequencies may be del-
phenotype with increased growth kinetics (Horner et al. eterious (between 4 and 6 Hz; Wilder and Pope 1996; Kumar
2002; Kluba et al. 2005; Rastogi et al. 2009; Wang et al. et al. 1999). Due to upright posture, static or low-frequency
2011). Conversely, annulus cells do less well in 3D culture in (1 Hz) pressures from gravity loading and daily living activi-
terms of cell morphology and survival (Horner et al. 2002). ties are also of interest given the recognized differences in
Interpretation of published studies on disc cell mechano- disc degeneration in humans when compared to quadrupeds
responsiveness is confounded by the fact that disc cells are (Lotz 2004). Accordingly, in vitro studies explore cell
isolated from many different tissue sources. These include responses over a broad range of loading frequencies, from
human surgical samples and small and large animals. While static to 20 Hz (Table 7.1).
human cells may be desirable to study disease pathogenesis, The disc is one of the most highly loaded tissues in the
they are logistically difficult to acquire without close body and routinely experiences compressive pressures in the
affiliation to high-volume surgical centers. Also, studies with 2 MPa range and as high as 3 MPa under extreme activities
human cells are complicated by individual-to-individual (Ranu 1990; Wilke et al. 1999). Diurnal fluctuations can be
variability and degeneration-related phenotypic shifts (Kluba significant, for example, during sleep, pressures decrease to
Table 7.1 Experimental in vitro studies of pressure on disc cell function

Cell type Cell source Scaffold Load Frequency (Hz) Duration Reference
AF Porcine Alginate 13 MPa 0.5 3h Wenger et al. (2005)
NP Rats Alginate 10 kPa, 20 % strain 0.5 1h Wang et al. (2011)
AF, NP Porcine Agarose 15 % strain 2 4h Salvatierra et al. (2011)
IAF, OAF Bovine PGLA 5 MPa 0.5 4 h/day, 314 days Reza and Nicoll (2008)
NP Human, bovine Col 1 0.25 and 2.5 MPa 0.1 1 h, 24 h Neidlinger-Wilke et al. (2009)
NP Human, bovine Col 1 0.25, 2.5 MPa 0.1 24 h Neidlinger-Wilke et al. (2006)
AF, NP Human Tissues 0.1, 0.3, 3 MPa Static 2h Liu et al. (2001)
AF, NP Human Alginate 0.350.95 MPa 1 2h Le Maitre et al. (2009)
AF, NP Porcine, bovine Alginate 212 % strain 3 2 h/day, 7 days Korecki et al. (2009)
AF, NP Rabbit Alginate (NP) 03 MPa 120 30 min/day, 3 days Kasra et al. (2003)
AF, NP Porcine Alginate 1 MPa 1, 3, 5, 8, 10 30 min/day, 3 days Kasra et al. (2006)
AF, NP Canine Alginate 0.35, 0.1 MPa Static 9 days Hutton et al. (2001)
IAF, OAF, NP Human Tissue 0.3, 3.0 MPa Static 2h Handa et al. (1997)
NP Rabbit Alginate 0.7, 2, 4 MPa Static 4, 24 h Sowa et al. (2011a)
AF, NP Porcine Agarose 15 % strain Static, 0.1 and 1 4h Fernando et al. (2011)
AF, NP Porcine Alginate 25 % strain Static 2, 18, 30 h Chen et al. (2004)
AF annulus fibrosus, NP nucleus pulposus, IAF inner annulus fibrosus, OAF outer annulus fibrosus
0.1 MPa and during quiet standing 0.5 MPa (Wilke et al. stress can trigger cell volume regulatory mechanisms. For
1999). Pressures applied during in vitro tests, therefore, example, when disc cells are challenged by osmotic stress,
cover the range from atmospheric (free swelling control con- calcium signaling and F-actin cytoskeletal elements play an
ditions) to as high as 5 MPa. important role in restoring homeostasis (Pritchard and Guilak
The disc cells response to pressure is dependent on the 2004). The importance of the cytoskeleton is also implicated
magnitude, frequency, and duration of loading. In general, by observations that load-induced loss of proteoglycan
in vitro studies suggest that physiologic loading conditions expression can be inhibited by exposure to an RGD inhibi-
(<1 MPa magnitude, <3 Hz frequency, and <24-h duration) tory peptide (Le Maitre et al. 2009). Given the role integrins
are anabolic, whereas regimes outside this range are cata- play in connecting the cytoskeleton to extracellular matrix,
bolic. For both nucleus and annulus cells, low pressure (0.2 this later observation further supports the importance of the
1.0 MPa) tends to increase expression of anabolic matrix cytoskeleton in the signaling cascade. Additionally, cell
genes (collagen 1, aggrecan, biglycan, decorin, lumican, membrane water channels, aquaporins, have been identified
fibromodulin, fibronectin (Chen et al. 2004; Wenger et al. in disc cells and are considered another important mecha-
2005; Sowa et al. 2011b; Wang et al. 2011)) and enzyme nism of cell volume regulatory control that may be stimu-
inhibitors (TIMP1) (Handa et al. 1997; Sowa et al. 2011a, b). lated by hydrostatic stress (Richardson et al. 2008;
In parallel, there is a reduction of catabolic factors (MMPs, Haudenschild et al. 2009).
iNOS, Cox2) (Sowa et al. 2011b). Conversely, high pressure Other bioactive factors are also considered important to
(14 MPa) tends to reduce expression of anabolic genes (col- disc cell pressure responsiveness. Hydrostatic pressure regu-
lagen and aggrecan) and increase catabolic (MMP-1, MMP-3, lates nitric oxide (NO) production by disc cells (Salvatierra
MMP-13) (Handa et al. 1997; Neidlinger-Wilke et al. 2006; et al. 2011). NO is a short-lived molecule that is produced
Le Maitre et al. 2009) and inflammatory factors (Cox2, from citrulline via the enzyme NO synthase (NOS) (Mitchell
iNOS) (Sowa et al. 2011b). et al. 1997). NO production is known to inhibit aerobic oxi-
With loading frequencies in the range of 35 Hz, nucleus dation of pyruvate after glycolysis and could regulate mito-
pulposus cells reduce aggrecan and collagen synthesis and chondrial respiration (Fernando et al. 2011; Salvatierra et al.
increase aggrecan degradation (Kasra et al. 2003, 2006; 2011). NO production by cells suppresses proteoglycan syn-
Korecki et al. 2009). This frequency effect may be age- thesis (Liu et al. 2001).
related as young and mature disc cells show opposite trends
at frequencies below 3 Hz (Korecki et al. 2009). Likewise, as
the duration of static compressive loading approaches 24 h, 7.3.4 Disc Cell Response to Stretch
catabolic and proinflammatory responses predominate (Sowa
et al. 2011b). Generally speaking, the outer annulus fibrosus in the healthy
A number of signaling mechanisms discussed above are disc is subjected to biaxial stretch as it acts to contain the
implicated in these pressure-induced behaviors. Compressive pressurized nucleus (Shirazi-Adl et al. 1984). The extent of
Table 7.2 Experimental Cell type Cell source Substrate Stretch Frequency (Hz) Duration Reference
in vitro studies of stretch
AF Rabbit Col 1 3, 6, 8 % 0.1, 0.5, 1 4, 24 h Sowa et al. (2011b)
on disc cell function
AF, NP Human 10 % 1 2 h/day, 7 days Hee et al. (2010)
AF, NP Rat Col 1 20 % 0.05 48 h Miyamoto et al. (2006)
AF Rabbit, human 0.1 G vibration 6 up to 1 h Yamazaki et al. (2003)
AF Rat 6% 0.05 4h Sowa and
Agarwal (2008)
AF Rabbit Col 1 1, 5 % 1 0.5, 24 h Rannou et al. (2003)
AF Human 10 % 0.33, 1 0.3 h Gilbert et al. (2010)
NP Rabbit 10 % 0.5 8 days Matsumoto et al.
(1999)
AF Human 10 % 1 0.3 h Gilbert et al. (2011)
AF annulus fibrosus, NP nucleus pulposus
annular stretch generated in response to in vivo compression necessary for biosynthetic processes. For example, Yamazaki
loading conditions is near 4 % strain and as high as 6 % and colleagues demonstrated that annulus cells initially
during flexion and extension (Stokes 1987). The impor- increase their resting production of ATP in response to con-
tance of stretch to annular homeostasis has been elegantly tinuous vibration, but this is only transient and becomes sup-
demonstrated by Hayes and colleagues who show that dur- pressed after 15 min (Yamazaki et al. 2003). Consequently,
ing development, pressures generated by the nucleus pulpo- when the duty cycle is limited to 2 h twice per day, 5 % strain
sus trigger stress fiber formation in annular fibroblasts and maintains collagen and proteoglycan production for 2 weeks
ultimately patterning of the complex lamellar architecture relative to unloaded controls (Hee et al. 2010) Increasing
(Hayes et al. 1999). To mimic this process in a controlled the strain magnitude to 10 % elevates cell proliferation and
fashion in vitro, several experimental systems have been causes a 25 % increase in collagen production.
described (Brown 2000). The most common system is the Loading frequency also influences the cellular response to
Flexercell (Vande Geest et al. 2004) that permits cells to be stretch. Gilbert and coworkers subjected human annulus
maintained in deformable culture plates. The strain pattern, fibrosus cells to 10 % stretch for 20 min at either 0.33 or
magnitude, waveform, frequency, and duty cycle can be con- 1.0 Hz (Gilbert et al. 2010). For healthy disc cells, 1 Hz stim-
trolled thereby permitting a broad range of study parameters. ulation maintained collagen and aggrecan gene expression
Additionally, the dishes can be coated with matrix proteins relative to control, whereas there was a shift toward catabo-
such as collagen and laminin so as to better mimic in vivo lism when the frequency was changed to 0.33 Hz. A frequen-
cell/matrix interactions. Cell stretch is generated by vacuum cy-dependent response to stretch has also been reported for
pressures that deform the substrates over various sized posts rabbit annular fibroblasts, where 1 Hz stimulation increased
and allow choice between uniaxial versus biaxial conditions. MMP-3 gene expression after 4 h of stimulation while
While the systems have been optimized for strain uniformity, TIMP-1 was downregulated with 0.1 and 0.5 Hz loading
variations in the strain field, particularly within the unsup- (Sowa et al. 2011a).
ported portion of the membranes, should be recognized Because painful disc degeneration has been linked to ele-
(Gilbert et al. 1994; Vande Geest et al. 2004). vated cytokine levels, several studies have questioned whether
Disc cells are sensitive to stretch magnitude, duration, and the cells response to stretch is modulated by proinflammatory
duty cycle (Table 7.2). Continuous cyclic stretch (CCS) at cytokines. Miyamoto and coworkers subjected rat nucleus
low magnitude (1 %) and physiologic frequencies (1 Hz) is pulposus and annulus fibrosus cells to CCS (20 % stretch at
homeostatic (maintains proteoglycan production) for annu- 0.05 Hz for 48 h) with and without inflammatory factors in
lus cells over a 24-h period (Rannou et al. 2003). However, the culture media (10 ng/mL IL-1b or TNF-a) (Miyamoto
the response turns catabolic (decreased proteoglycan produc- et al. 2006). Their data indicate that stretch and cytokines
tion, increased NO production, increased Cox2 and MMP-3 individually had comparable effect on inflammatory media-
gene expression) with increasing strains (518 %) and time tor production (prostaglandin E2, PGE2). Importantly, there
(beyond 46 h) (Rannou et al. 2003; Sowa et al. 2011a). For was a significant synergistic increase when both stimuli
rabbit nuclear cells, CCS (10 % strain at 0.5 Hz) increases were combined. The effect was more pronounced in annular
are evident in cell proliferation and collagen production dur- cells and correlated with the gene expression of COX-2, a
ing the first 12 days, but this anabolic effect is lost by 4 rate-limiting enzyme for the cellular production of PGE2.
and 8 days (Matsumoto et al. 1999). The conversion from an However, there is also evidence that CCS can be protective
anabolic to catabolic response to CCS with time may be due against inflammation-induced catabolic behaviors. When
to fatigue: the cells inability to sustain energy production measuring alternate marker genes for proinflammatory cell
responses, Sowa and colleagues reported that CCS (6 % shear stress. Fluid-induced shear stresses in tissues are not
stretch at 0.05 Hz for 4 h) decreased an IL-1-induced pro- well characterized, but are likely also important determinants
duction of several inflammation markers (iNOS, TNF-alpha, of cellular function. Thus, while extensive mechanobiology
MMP-3, and MMP-13) by 50 % (Sowa and Agarwal 2008). data are available at the tissue and cellular level, there remains
Consistent with this finding, Gilbert and co-investigators a need to develop mathematical/computational models that
(Gilbert et al. 2010) report that the anti-catabolic effect of link tissue/cellular dynamic responses so as to reconcile the
CCS (10 % stretch at 1 Hz for 1 h) on human annulus fibrosus phenomena observed at the two levels of scale.
cells (suppression of MMP-3 and ADAMTS4 gene expres- More mechanistic studies are needed to define the acti-
sion) was lost by exposure to cytokine inhibitors (IL-1Ra or vation of pathways that may underlie disease mechanisms
IL-4RAb). and thereby inform pharmacologic interventions. In addition
to those pathways discussed above, recent studies suggest
alternate mechanisms by which disc cells transduce mechan-
7.4 Final Comments ical signals. For example, caveolae are plasma membrane
invaginations that are highly enriched with cholesterol,
Information is encoded in a time-dependent manner in con- with their main constituents being caveolin-1 and caveo-
formational changes of cell membrane receptors and the lin-2 (Sinha et al. 2011). Acute increases in cell volume or
cytoskeleton. These changes can be induced by relative differ- stretch lead to a rapid loss of caveolae, and as such they are
ences in cell and matrix stiffness, matrix deformation, osmo- implicated in a membrane-mediated mechanical response
lality, with the responses tuned by inflammatory and other triggered by tyrosine phosphorylation (Alenghat and Ingber
soluble molecules. Data from both organ- and cell-level stud- 2002). Caveolae are required to buffer fluctuations in cell
ies suggest the existence of optimal loading regimens, where membrane stress induced by acute membrane tension and
both excessive and negligible pressure/stretch responses are osmotic shock, such that loss of caveolae compromises buff-
catabolic. This is consistent with clinical observations that ering of cell membrane tension (Parton and Simons 2007).
show that disc injury rates are increased under situations of Recently, caveolin-1 has been identified in nucleus pulposus
either chronic inactivity or extreme exposures, leading to the cells, with its expression decreasing with age (Heathfield
concept of a U-shaped distribution in the discs response to et al. 2008). These observations suggest caveolae may par-
load (Lotz 2011; Panel on Musculoskeletal Disorders and the ticipate in nucleus pulposus cell responses to hydrostatic
Workplace 2001). Objectively, for both prevention and treat- pressure.
ment, defining such boundaries would be very beneficial. Another potentially important component of the disc
Given that magnitude, frequency, and duration all contribute cells mechanoregulatory machinery is the glycocalyx (Fuster
to cell responses, it would be valuable to combine these vari- and Esko 2005). Glycans can exist as cell membrane-bound
ables into a single-dose parameter (Gardner 2000). Dose glycoconjugates that mediate cell adhesion as well as other
response relationships may be more easily compared between intracellular signaling events. The glycocalyx forms a por-
studies and ultimately extrapolated to humans. tion of the pericellular matrix known as the chondron that
It is hoped that this review clearly demonstrates the has been observed both in chondrocytes and in nucleus pul-
substantial effort that has been invested into elucidating posus cells (Roberts et al. 1991; Chang and Poole 1997). The
tissue-level mechanobiologic principles and cellular mecha- presence of chondrons in the nucleus pulposus has been con-
nobiologic phenomena. Yet, relatively little is understood sidered a phenotypical indicator of degeneration (Ciapetti
about how these two hierarchical levels are quantitatively et al. 2012). Recently, the glycocalyx has been shown to con-
linked. This knowledge gap exists because of complexities tribute to extracellular stiffness that mediates load-modulated
in matrix structure-function and in the nature by which the fluctuations in integrin clustering and signaling (Paszek et al.
deforming matrix affects cells function (Bruehlmann et al. 2009). Similarly to the glycocalyx pericellular stiffness,
2004a, b). For example, in situ visualization of whole-tissue extracellular matrix stiffness can influence cell behaviors by
preparations show that in highly ordered collagenous tissues, modifying integrin-mediated TGF-b activation, heterodi-
an 8 % applied strain results in displacements that approxi- merization and signaling between integrins and TGF-b
mate 1 % strain along a collagen fiber and approximately receptors, or the activity of TAZ (transcriptional co-activator
4 % strain between fibers (Screen et al. 2004). This distribu- with PDZ-binding motif), a transcriptional co-regulator that
tion suggests that cells may experience a smaller degree of is a common effector of stiffness sensing and TGF-b signal-
elongation and a greater amount of shear than tissue-level ing (Dupont et al. 2011; Hinz 2009).
strains might indicate. Moreover, pressure gradients in the Ultimately, the translation of mechanobiologic under-
disc result in interstitial fluid flux within and across tissue standing into clinical benefits for patients requires coopera-
boundaries. Wang and colleagues (2011) have demonstrated tive efforts of interdisciplinary teams that include clinicians,
that immature nucleus pulposus cells are sensitive to fluid surgeons, bioengineers, and biologists. In this way, the
necessarily diverse lines of evidence can be most efficiently Bass EC, Duncan NA et al (1997) Frozen storage affects the compressive
stitched into testable theories and research programs that creep behavior of the porcine intervertebral disc. Spine
22(24):28672876
link basic and clinical studies. Boxberger JI, Auerbach JD et al (2008) An in vivo model of reduced
nucleus pulposus glycosaminoglycan content in the rat lumbar
intervertebral disc. Spine (Phila Pa 1976) 33(2):146154
7.5 Summary of Critical Concepts Broberg KB, von Essen HO (1980) Modeling of intervertebral discs.
Spine 5(2):155167
Discussed in the Chapter Brown TD (2000) Techniques for mechanical stimulation of cells
in vitro: a review. J Biomech 33(1):314
Spinal loading is filtered into intradiscal regions of hydro- Bruehlmann SB, Rattner JB et al (2002) Regional variations in the cel-
static pressure and octahedral shear. lular matrix of the annulus fibrosus of the intervertebral disc. J Anat
201(2):159171
Disc matrix properties are poroviscoelastic, such that the Bruehlmann SB, Hulme PA et al (2004a) In situ intercellular mechanics
extracellular environment is time varying. of the bovine outer annulus fibrosus subjected to biaxial strains.
Coupled physical and biochemical cues in the tissue J Biomech 37(2):223231
microenvironment signal cells to regulate the extracellu- Bruehlmann SB, Matyas JR et al (2004b) ISSLS prize winner: Collagen
fibril sliding governs cell mechanics in the anulus fibrosus: an in situ
lar matrix dynamically. confocal microscopy study of bovine discs. Spine (Phila Pa 1976)
Microenvironment cues include pH, growth factor and 29(23):26122620
nutrient concentration, osmolality, fluid flow, and extra- Carter DR, Beaupre GS (2001) Skeletal function and form. Cambridge
cellular matrix stiffness and conformation. University Press, New York
Cassidy JJ, Hiltner A et al (1989) Hierarchical structure of the interver-
Disc cells react to mechanical cues via four steps: mechan- tebral disc. Connect Tissue Res 23(1):7588
ical coupling to matrix, mechanotransduction or conver- Chang J, Poole CA (1997) Confocal analysis of the molecular heteroge-
sion of mechanical to chemical signals, intracellular neity in the pericellular microenvironment produced by adult canine
signal transmission, and the eventual cellular response. chondrocytes cultured in agarose gel. Histochem J 29(7):515528
Chen J, Baer AE et al (2002) Matrix protein gene expression in inter-
Matrix and cellular response pathways function over vertebral disc cells subjected to altered osmolarity. Biochem
varying timescales, such that the disc/cell composite can Biophys Res Commun 293(3):932938
function as a band-pass filter. Chen J, Yan W et al (2004) Static compression induces zonal-specific
Nucleus cells response to pressure is dependent on the changes in gene expression for extracellular matrix and cytoskeletal
proteins in intervertebral disc cells in vitro. Matrix Biol 22(7):
magnitude, frequency, and duration. Physiologic pressure 573583
loading (<1 MPa magnitude, <3 Hz frequency, and <24-h Chen J, Jing L et al (2009) Expression of laminin isoforms, receptors,
duration) is anabolic, whereas pressures outside that range and binding proteins unique to nucleus pulposus cells of immature
are catabolic. intervertebral disc. Connect Tissue Res 50(5):294306
Chin JR, Court C et al (1999) Compression-induced apoptosis in the
Annulus cell response to cyclic shear at low magnitude mouse intervertebral disc is time and stress dependent. 45th annual
(<1 % strain) and at physiologic frequencies (1 Hz) is meeting, Orthopaedic Research Society, Anaheim
anabolic for up to 12 days. Ching CT, Chow DH et al (2003) The effect of cyclic compression on
Cellular responses vary by species and are significantly the mechanical properties of the inter-vertebral disc: an in vivo study
in a rat tail model. Clin Biomech (Bristol, Avon) 18(3):182189
altered by the presence of degenerated matrix fragments Ching CT, Chow DH et al (2004) Changes in nuclear composition fol-
and proinflammatory cytokines. lowing cyclic compression of the intervertebral disc in an in vivo
rat-tail model. Med Eng Phys 26(7):587594
Acknowledgements NIH AR049786 (JL), NIH AR049770 (JL), NIH Ciapetti G, Granchi D et al (2012) Ex vivo observation of human inter-
AR063357 (JL), NIH AR059325 (AH), NSF CBET 0845754 (AH), vertebral disc tissue and cells isolated from degenerated interverte-
NIST 70NANB11H191 (AH). bral discs. Eur Spine J 21(1):1019
Costi JJ, Hearn TC et al (2002) The effect of hydration on the stiffness
of intervertebral discs in an ovine model. Clin Biomech (Bristol,
Avon) 17(6):446455
References Court C, Colliou OK et al (2001) The effect of static in vivo bending on
the intervertebral disc. Spine J 1(4):239245
Adams M, Hutton W (1983) The effect of posture on the fluid content Court C, Chin JR et al (2007) Biological and mechanical conse-
of lumbar intervertebral discs. Spine 8(6):665671 quences of transient intervertebral disc bending. Eur Spine J
Adams MA, Dolan P et al (1990) Diurnal changes in spinal mechanics and 16(11):18991906
their clinical significance. J Bone Joint Surg Br 72-B(2):266270 del Rio A, Perez-Jimenez R et al (2009) Stretching single talin rod mol-
Adams MA, McNally DS et al (1996) Stress distributions inside inter- ecules activates vinculin binding. Science 323(5914):638641
vertebral discs. The effects of age and degeneration. J Bone Joint Doyle AD, Yamada KM (2010) Cell biology: sensing tension. Nature
Surg Br 78(6):965972 466(7303):192193
Alenghat FJ, Ingber DE (2002) Mechanotransduction: all signals point Dupont S, Morsut L et al (2011) Role of YAP/TAZ in mechanotrans-
to cytoskeleton, matrix, and integrins. Sci STKE 2002(119):pe6 duction. Nature 474(7350):179183
Ayotte DC, Ito K et al (2000) Direction-dependent constriction flow in Elfervig MK, Minchew JT et al (2001) IL-1beta sensitizes interverte-
a poroelastic solid: the intervertebral disc valve. J Biomech Eng bral disc annulus cells to fluid-induced shear stress. J Cell Biochem
122(6):587593 82(2):290298
Errington RJ, Puustjarvi K et al (1998) Characterisation of cytoplasm- Hoe A, Atha J et al (1994) Stature loss from sustained gentle body
filled processes in cells of the intervertebral disc. J Anat 192(Pt loading. Ann Hum Biol 21(2):171178
3):369378 Hoffman BD, Crocker JC (2009) Cell mechanics: dissecting the physi-
Fernando HN, Czamanski J et al (2011) Mechanical loading affects the cal responses of cells to force. Annu Rev Biomed Eng 11:259288
energy metabolism of intervertebral disc cells. J Orthop Res Hoffman BD, Grashoff C et al (2011) Dynamic molecular processes
29(11):16341641 mediate cellular mechanotransduction. Nature 475(7356):316323
Friedland JC, Lee MH et al (2009) Mechanically activated integrin Holzapfel GA, Schulze-Bauer CA et al (2005) Single lamellar mechan-
switch controls alpha5beta1 function. Science 323(5914):642644 ics of the human lumbar anulus fibrosus. Biomech Model
Fuster MM, Esko JD (2005) The sweet and sour of cancer: glycans as Mechanobiol 3(3):125140
novel therapeutic targets. Nat Rev Cancer 5(7):526542 Horner HA, Roberts S et al (2002) Cells from different regions of the
Gabai AS, YewAG et al (2007) Distraction helps maintain transient disc intervertebral disc: effect of culture system on matrix expression
mechanical behavior before exertion. 34th annual meeting of the and cell phenotype. Spine 27(10):10181028
International Society for the Study of the Lumbar Spine, Hong Kong Hsieh AH, Lotz JC (2003) Prolonged spinal loading induces matrix metallo-
Garbutt G, Boocock MG et al (1990) Running speed and spinal shrink- proteinase-2 activation in intervertebral discs. Spine 28(16):17811788
age in runners with and without low back pain. Med Sci Sports Hsieh AH, Wagner DR et al (2005) Dependence of mechanical behav-
Exerc 22(6):769772 ior of the murine tail disc on regional material properties: a paramet-
Gardner SN (2000) A mechanistic, predictive model of doseresponse ric finite element study. J Biomech Eng 127(7):11581167
curves for cell cycle phase-specific and -nonspecific drugs. Cancer Hsieh AH, Hwang D et al (2009) Degenerative anular changes induced
Res 60(5):14171425 by puncture are associated with insufficiency of disc biomechanical
Gilbert JA, Weinhold PS et al (1994) Strain profiles for circular cell function. Spine (Phila Pa 1976) 34(10):9981005
culture plates containing flexible surfaces employed to mechani- Hunter CJ, Matyas JR et al (2003) The notochordal cell in the nucleus
cally deform cells in vitro. J Biomech 27(9):11691177 pulposus: a review in the context of tissue engineering. Tissue Eng
Gilbert HT, Hoyland JA et al (2010) The response of human anulus 9(4):667677
fibrosus cells to cyclic tensile strain is frequency-dependent and Hutton WC, Elmer WA et al (2001) Do the intervertebral disc cells
altered with disc degeneration. Arthritis Rheum 62(11):33853394 respond to different levels of hydrostatic pressure? Clin Biomech
Gilbert HT, Hoyland JA et al (2011) The involvement of interleukin-1 (Bristol, Avon) 16(9):728734
and interleukin-4 in the response of human annulus fibrosus cells to Hwang D, Gabai AS et al (2012) Role of load history in intervertebral
cyclic tensile strain: an altered mechanotransduction pathway with disc mechanics and intradiscal pressure generation. Biomech Model
degeneration. Arthritis Res Ther 13(1):R8 Mechanobiol 11(12):95106
Gilchrist CL, Chen J et al (2007) Functional integrin subunits regulat- Iatridis JC, Mente PL et al (1999) Compression-induced changes in inter-
ing cell-matrix interactions in the intervertebral disc. J Orthop Res vertebral disc properties in a rat tail model. Spine 24(10):9961002
25(6):829840 Ingber DE (2003) Mechanobiology and diseases of mechanotransduc-
Gruber HE, Ingram JA et al (2007) Senescence in cells of the aging and tion. Ann Med 35(8):564577
degenerating intervertebral disc: immunolocalization of senescence- Janmey PA, Kinnunen PK (2006) Biophysical properties of lipids and
associated beta-galactosidase in human and sand rat discs. Spine dynamic membranes. Trends Cell Biol 16(10):538546
32(3):321327 Johnson WE, Roberts S (2003) Human intervertebral disc cell morphol-
Guilak F, Ting-Beall HP et al (1999) Viscoelastic properties of interver- ogy and cytoskeletal composition: a preliminary study of regional
tebral disc cells. Identification of two biomechanically distinct cell variations in health and disease. J Anat 203(6):605612
populations. Spine (Phila Pa 1976) 24(23):24752483 Kasra M, Goel V et al (2003) Effect of dynamic hydrostatic pressure on
Han SM, Lee SY et al (2001) Disc hydration measured by magnetic rabbit intervertebral disc cells. J Orthop Res 21(4):597603
resonance imaging in relation to its compressive stiffness in rat Kasra M, Merryman WD et al (2006) Frequency response of pig inter-
models. Proc Inst Mech Eng H 215(5):497501 vertebral disc cells subjected to dynamic hydrostatic pressure.
Handa T, Ishihara H et al (1997) Effects of hydrostatic pressure on J Orthop Res 24(10):19671973
matrix synthesis and matrix metalloproteinase production in the Kluba T, Niemeyer T et al (2005) Human anulus fibrosis and nucleus
human lumbar intervertebral disc. Spine 22(10):10851091 pulposus cells of the intervertebral disc: effect of degeneration and
Haudenschild AK, Hsieh AH et al (2009) Pressure and distortion regu- culture system on cell phenotype. Spine 30(24):27432748
late human mesenchymal stem cell gene expression. Ann Biomed Koeller W, Funke F et al (1984) Biomechanical behavior of human inter-
Eng 37(3):492502 vertebral discs subjected to long lasting axial loading. Biorheology
Hayes AJ, Benjamin M et al (1999) Role of actin stress fibres in the 21(5):675686
development of the intervertebral disc: cytoskeletal control of extra- Korecki CL, Kuo CK et al (2009) Intervertebral disc cell response to
cellular matrix assembly. Dev Dyn 215(3):179189 dynamic compression is age and frequency dependent. J Orthop Res
Hayes AJ, Benjamin M et al (2001) Extracellular matrix in develop- 27(6):800806
ment of the intervertebral disc. Matrix Biol 20(2):107121 Kumar A, Varghese M et al (1999) Effect of whole-body vibration on
Heathfield SK, Le Maitre CL et al (2008) Caveolin-1 expression and the low back. A study of tractor-driving farmers in north India.
stress-induced premature senescence in human intervertebral disc Spine (Phila Pa 1976) 24(23):25062515
degeneration. Arthritis Res Ther 10(4):R87 Le Maitre CL, Frain J et al (2009) Altered integrin mechanotransduc-
Hee HT, Zhang J et al (2010) Effects of cyclic dynamic tensile strain on tion in human nucleus pulposus cells derived from degenerated
previously compressed inner annulus fibrosus and nucleus pulposus discs. Arthritis Rheum 60(2):460469
cells of human intervertebral disc-an in vitro study. J Orthop Res Leatt P, Reilly T et al (1986) Spinal loading during circuit weight-train-
28(4):503509 ing and running. Br J Sports Med 20(3):119124
Hickey DS, Hukins DWL (1980) Relation between the structure of the Leivseth G, Drerup B (1997) Spinal shrinkage during work in a sitting
annulus fibrosus and the function and failure of the intervertebral posture compared to work in a standing posture. Clin Biomech
disc. Spine 5(2):106116 (Bristol, Avon) 12(78):409418
Hinz B (2009) Tissue stiffness, latent TGF-beta1 activation, and Li S, Duance VC et al (2008) Zonal variations in cytoskeletal element
mechanical signal transduction: implications for the pathogenesis organization, mRNA and protein expression in the intervertebral
and treatment of fibrosis. Curr Rheumatol Rep 11(2):120126 disc. J Anat 213(6):725732
Liu GZ, Ishihara H et al (2001) Nitric oxide mediates the change of Parton RG, Simons K (2007) The multiple faces of caveolae. Nat Rev
proteoglycan synthesis in the human lumbar intervertebral disc in Mol Cell Biol 8(3):185194
response to hydrostatic pressure. Spine 26(2):134141 Paszek MJ, Zahir N et al (2005) Tensional homeostasis and the malig-
Lotz JC (2004) Animal models of intervertebral disc degeneration: les- nant phenotype. Cancer Cell 8(3):241254
sons learned. Spine 29(23):27422750 Paszek MJ, Boettiger D et al (2009) Integrin clustering is driven by
Lotz JC (2011) Load and the spine. How does the Goldilocks principle mechanical resistance from the glycocalyx and the substrate. PLoS
apply? Spine J 11(1):4445 Comput Biol 5(12):e1000604
Lotz JC, Chin JR (2000) Intervertebral disc cell death is dependent on the Patwardhan AG, Havey RM et al (1999) A follower load increases the
magnitude and duration of spinal loading. Spine 25(12):14771483 load-carrying capacity of the lumbar spine in compression. Spine
Lotz JC, Colliou OK et al (1998) Compression-induced degeneration of (Phila Pa 1976) 24(10):10031009
the intervertebral disc: an in vivo mouse model and finite-element Patwardhan AG, Havey RM et al (2000) Load-carrying capacity of the
study. Spine 23(23):24932506 human cervical spine in compression is increased under a follower
Lotz JC, Hadi T et al (2008) Anulus fibrosus tension inhibits degenerative load. Spine (Phila Pa 1976) 25(12):15481554
structural changes in lamellar collagen. Eur Spine J 17(9):11491159 Patwardhan AG, Havey RM et al (2003) Effect of compressive follower
Maclean JJ, Lee CR et al (2004) Anabolic and catabolic mRNA levels preload on the flexion-extension response of the human lumbar
of the intervertebral disc vary with the magnitude and frequency of spine. J Orthop Res 21(3):540546
in vivo dynamic compression. J Orthop Res 22(6):11931200 Pflaster DS, Krag MH et al (1997) Effect of test environment on inter-
MacLean JJ, Lee CR et al (2005) The effects of short-term load dura- vertebral disc hydration. Spine (Phila Pa 1976) 22(2):133139
tion on anabolic and catabolic gene expression in the rat tail inter- Pritchard S, Guilak F (2004) The role of F-actin in hypo-osmotically
vertebral disc. J Orthop Res 23(5):11201127 induced cell volume change and calcium signaling in anulus fibrosus
Marchand F, Ahmed AM (1990) Investigation of the laminate structure cells. Ann Biomed Eng 32(1):103111
of lumbar disc anulus fibrosus. Spine 15(5):402410 Race A, Broom ND et al (2000) Effect of loading rate and hydration
Matsumoto T, Kawakami M et al (1999) Cyclic mechanical stretch on the mechanical properties of the disc. Spine (Phila Pa 1976)
stress increases the growth rate and collagen synthesis of nucleus 25(6):662669
pulposus cells in vitro. Spine 24(4):315319 Rannou F, Richette P et al (2003) Cyclic tensile stretch modulates pro-
McGill SM, van Wijk MJ et al (1996) Studies of spinal shrinkage to eval- teoglycan production by intervertebral disc annulus fibrosus cells
uate low-back loading in the workplace. Ergonomics 39(1):92102 through production of nitrite oxide. J Cell Biochem 90(1):148157
Minogue BM, Richardson SM et al (2010) Transcriptional profiling of Ranu HS (1990) Measurement of pressures in the nucleus and within
bovine intervertebral disc cells: implications for identification of the annulus of the human spinal disc: due to extreme loading. Proc
normal and degenerate human intervertebral disc cell phenotypes. Inst Mech Eng H 204(3):141146
Arthritis Res Ther 12(1):R22 Rastogi A, Thakore P et al (2009) Environmental regulation of noto-
Mitchell JA, Williams FM et al (1997) Role of nitric oxide in the dilator chordal gene expression in nucleus pulposus cells. J Cell Physiol
actions of capsaicin-sensitive nerves in the rabbit coronary circula- 220(3):698705
tion. Neuropeptides 31(4):333338 Rastogi A, Kim H et al. (2013) MMP-2 mediates local degradation and
Miyamoto H, Doita M et al (2006) Effects of cyclic mechanical stress remodeling of collagen by annulus fibrosus cells of the interverte-
on the production of inflammatory agents by nucleus pulposus and bral disc. Arthritis Res Ther 15(2):R57
anulus fibrosus derived cells in vitro. Spine (Phila Pa 1976) Reza AT, Nicoll SB (2008) Hydrostatic pressure differentially regulates
31(1):49 outer and inner annulus fibrosus cell matrix production in 3D scaf-
Miyazaki T, Kobayashi S et al (2009) A phenotypic comparison of pro- folds. Ann Biomed Eng 36(2):204213
teoglycan production of intervertebral disc cells isolated from rats, Richardson SM, Knowles R et al (2008) Aquaporin expression in the
rabbits, and bovine tails; which animal model is most suitable to human intervertebral disc. J Mol Histol 39(3):303309
study tissue engineering and biological repair of human disc disor- Risbud MV, Shapiro IM (2011) Notochordal cells in the adult interver-
ders? Tissue Eng Part A 15(12):38353846 tebral disc: new perspective on an old question. Crit Rev Eukaryot
Na S, Collin O et al (2008) Rapid signal transduction in living cells is a Gene Expr 21(1):2941
unique feature of mechanotransduction. Proc Natl Acad Sci U S A Roberts S, Menage J et al (1991) Type III collagen in the intervertebral
105(18):66266631 disc. Histochem J 23(1112):503508
Neidlinger-Wilke C, Wurtz K et al (2006) Regulation of gene expres- Rodacki AL, Fowler NE et al (2005) Body mass as a factor in stature
sion in intervertebral disc cells by low and high hydrostatic pres- change. Clin Biomech (Bristol, Avon) 20(8):799805
sure. Eur Spine J 15(Suppl 3):S372S378 Rohlmann A, Neller S et al (2001) Influence of a follower load on intra-
Neidlinger-Wilke C, Liedert A et al (2009) Mechanical stimulation discal pressure and intersegmental rotation of the lumbar spine.
alters pleiotrophin and aggrecan expression by human intervertebral Spine (Phila Pa 1976) 26(24):E557E561
disc cells and influences their capacity to stimulate endothelial Rousseau MA, Ulrich JA et al (2007) Stab incision for inducing interverte-
migration. Spine (Phila Pa 1976) 34(7):663669 bral disc degeneration in the rat. Spine (Phila Pa 1976) 32(1):1724
Nelson CM, Jean RP et al (2005) Emergent patterns of growth con- Sakai D, Nakai T et al (2009) Differential phenotype of interverte-
trolled by multicellular form and mechanics. Proc Natl Acad Sci U bral disc cells: microarray and immunohistochemical analysis of
S A 102(33):1159411599 canine nucleus pulposus and anulus fibrosus. Spine (Phila Pa 1976)
Nettles DL, Richardson WJ et al (2004) Integrin expression in cells of 34(14):14481456
the intervertebral disc. J Anat 204(6):515520 Salvatierra JC, Yuan TY et al (2011) Difference in energy metabolism
Niosi CA, Oxland TR (2004) Degenerative mechanics of the lumbar of annulus fibrosus and nucleus pulposus cells of the intervertebral
spine. Spine J 4(6 Suppl):202S208S disc. Cell Mol Bioeng 4(2):302310
Oegema TR Jr, Johnson SL et al (2000) Fibronectin and its fragments Schroeder Y, Wilson W et al (2006) Osmoviscoelastic finite ele-
increase with degeneration in the human intervertebral disc. Spine ment model of the intervertebral disc. Eur Spine J 15(Suppl
25(21):27422747 3):S361S371
Panel on Musculoskeletal Disorders and the Workplace, C. o. B. a. S. S. Screen HR, Lee DA et al (2004) An investigation into the effects of the
a. E., National Research Council (2001) Musculoskeletal disorders hierarchical structure of tendon fascicles on micromechanical prop-
and the workplace. National Academy Press, Washington, DC erties. Proc Inst Mech Eng H 218(2):109119
Shirazi-Adl A, Drouin G (1987) Load-bearing role of facets in a lumbar Urban JP, Roberts S et al (2000) The nucleus of the intervertebral disc
segment under sagittal plane loadings. J Biomech 20(6):601613 from development to degeneration. Am Zool 40:5361
Shirazi-Adl A, Shrivastava SC et al (1984) Stress analysis of the lumbar Vande Geest JP, Di Martino ES et al (2004) An analysis of the
disc-body unit in compression: a three dimensional nonlinear finite complete strain field within Flexercell membranes. J Biomech
element study. Spine 9(2):120134 37(12):19231928
Sinha B, Koster D et al (2011) Cells respond to mechanical stress by Vogel V, Baneyx G (2003) The tissue engineering puzzle: a molecular
rapid disassembly of caveolae. Cell 144(3):402413 perspective. Annu Rev Biomed Eng 5:441463
Sniadecki NJ, Chen CS (2007) Microfabricated silicone elastomeric Walsh AJ, Lotz JC (2004) Biological response of the intervertebral disc
post arrays for measuring traction forces of adherent cells. Methods to dynamic loading. J Biomech 37(3):329337
Cell Biol 83:313328 Wang Y, Shyy JY et al (2008) Fluorescence proteins, live-cell imaging,
Sowa G, Agarwal S (2008) Cyclic tensile stress exerts a protec- and mechanobiology: seeing is believing. Annu Rev Biomed Eng
tive effect on intervertebral disc cells. Am J Phys Med Rehabil 10:138
87(7):537544 Wang P, Yang L et al (2011) Nucleus pulposus cell response to confined
Sowa G, Coelho P et al (2011a) Determination of annulus fibrosus cell and unconfined compression implicates mechanoregulation by fluid
response to tensile strain as a function of duration, magnitude, and shear stress. Ann Biomed Eng 39(3):11011111
frequency. J Orthop Res 29(8):12751283 Wenger KH, Woods JA et al (2005) Matrix remodeling expression in
Sowa GA, Coelho JP et al (2011b) Alterations in gene expres- anulus cells subjected to increased compressive load. Spine (Phila
sion in response to compression of nucleus pulposus cells. Spine Pa 1976) 30(10):11221126
J 11(1):3643 Wilder DG, Pope MH (1996) Epidemiological and aetiological aspects
Stokes I (1987) Surface strain on human intervertebral discs. J Orthop of low back pain in vibration environments - an update. Clin
Res 5(3):348355 Biomech (Bristol, Avon) 11(2):6173
Stokes I, Greenaple D (1985) Measurement of surface deformation of Wilke HJ, Neef P et al (1999) New in vivo measurements of pressures
soft tissue. J Biomech 18(1):17 in the intervertebral disc in daily life. Spine 24(8):755762
Terahata N, Ishihara H et al (1994) Effects of axial traction stress on Wuertz K, Godburn K et al (2009) In vivo remodeling of intervertebral
solute transport and proteoglycan synthesis in the porcine interver- discs in response to short- and long-term dynamic compression.
tebral disc in vitro. Eur Spine J 3(6):325330 J Orthop Res 27(9):12351242
Urban JPG, Maroudas A (1981) Swelling of the intervertebral disc Xia M, Zhu Y (2008) Expression of integrin subunits in the herniated
in vitro. Connect Tissue Res 9:110 intervertebral disc. Connect Tissue Res 49(6):464469
Urban JPG, McMullin JF (1985) Swelling pressure of the intervertebral Xia M, Zhu Y (2011) Fibronectin fragment activation of ERK increas-
disc: influence of proteoglycan and collagen contents. Biorheology ing integrin alpha and beta subunit expression to degenerate nucleus
22:145157 pulposus cells. J Orthop Res 29(4):556561
Urban JP, McMullin JF (1988) Swelling pressure of the lumbar inter- Yamazaki S, Weinhold PS et al (2003) Annulus cells release
vertebral discs: influence of age, spinal level, composition, and ATP in response to vibratory loading in vitro. J Cell Biochem
degeneration. Spine 13(2):179187 90(4):812818
The Role of the ADAMTS Proteins
in the Intervertebral Disc 8
Jason C. Ho, James Wylie, and Suneel S. Apte

8.1 Overview ........................................................................ 125
8.1.1 Biological Significance and General 8.1.1 Biological Signicance and General
Features of Proteinases.................................................... 125 Features of Proteinases
8.1.2 Historical Perspective and Evolution
of Metalloproteinases ...................................................... 126
Proteinases (often referred to as proteases) are enzymes that
8.2 ADAMTS Structure and Function .............................. 127 hydrolyze the peptide bonds that hold amino acids together
8.3 ADAMTS Proteinases in Intervertebral within a polypeptide (protein) molecule. Secreted and cell-
Disc Biology and Disease .............................................. 131 surface proteinases are indispensable in several processes,
8.4 Summary of Critical Concepts such as digestion, molecular maturation, or activation of pre-
Discussed in the Chapter .............................................. 132 cursor proteins (proproteins), and in the turnover of diverse
References .................................................................................... 133 cellular products such as cell-surface receptors and extracel-
lular matrix (ECM) proteins and proteoglycans. These last
two functions are highly relevant to the skeletal system,
which comprises ECM proteins, including proproteins such
as procollagens, as quantitatively major components having
an indispensable structural role. In addition, the ECM is
increasingly being recognized for its role in regulating cell
behavior, such as in signaling through cell-matrix adhesion
molecules, and proteolytic products of ECM (matrikines).
Proteinases are required for and participate in all major
phases of vertebrate life, e.g., during embryonic develop-
ment, they are required for rapid tissue remodeling, whereas
in the adult organism, they participate in essential homeo-
static processes, such as coagulation, or adaptive responses
to biomechanical fluxes, e.g., bone and connective tissue
remodeling in response to mechanical stress. They are recog-
nized to have diverse, complex roles in the origin and/or
resolution of inflammatory, degenerative, and malignant dis-
J.C. Ho J. Wylie
Department of Biomedical Engineering, orders. Indeed, proteinases, including ADAMTS proteinases,
Orthopedic and Rheumatologic Institute, are frequently targeted for drug development because of their
Cleveland Clinic, Cleveland, OH, USA pivotal role in diseases (Fosang and Little 2008).
S.S. Apte (*) The functional core of every proteinase is its catalytic
Department of Biomedical Engineering, domain, which is the effector domain for proteolysis, and it
Orthopedic and Rheumatologic Institute,
is usually conjoined with an ancillary domain, required for
Cleveland Clinic, Cleveland, OH, USA
binding to the target protein that is cleaved (a substrate).
Department of Biomedical Engineering,
Through the combination of binding properties of the ancil-
Lerner Research Institute- ND20, Cleveland Clinic,
9500 Euclid Avenue, Cleveland, OH 44195, USA lary domain and the lock and key fit of the protease active
e-mail: aptes@ccf.org site with its substrate, a substrate specificity ranging from

126 J.C. Ho et al.
stringent to promiscuous is determined. Most proteinases of cell-surface molecules and have little reported direct
contain an N-terminal regulatory peptide segment or domain action on the ECM. ADAMTS proteinases appear thus far to
(the propeptide) that is involved in maintaining the enzyme have major roles related to proteolysis of the ECM rather
in an inactive or latent state (the zymogen) until its activity is than cell-surface substrates.
required. An intriguing aspect of proteinase biology is exis- ADAMTS and ADAM proteinase catalytic domains are
tence of natural (endogenous) inhibitors, which bind to and related to enzymes present in hemorrhagic snake venoms,
inactivate proteinases and protect against indiscriminate, such as those from rattlesnakes, since only these proteinases,
excess polypeptide destruction. The proteinases activity is but not MMPs, have the active-site zinc-binding sequence
further restricted to specific substrates by its precise spatial HEXXH + HD (single amino acid nomenclature, X is any
localization. This is determined by information within the amino acid). This sequence is referred to as the reprolysin
protease (such as a transmembrane domain or heparin- signature. ADAMTS and ADAM proteinases also have a
binding domain) that limits its activity to specific spatial similar zymogen activation mechanism, since their propep-
domains, e.g., cell-surface or pericellular matrix, or binding tides are proteolytically excised within the secretory path-
sites in ECM, which favors cleavage of the accessible sub- way, at the cell surface, or extracellularly by proprotein
strates. Additionally, spatial and temporal regulation of convertases such as furin. In contrast, only a handful of
expression, such as by transcriptional and post-transcriptional MMPs, such as the membrane-type (MT) MMPs, utilize this
mechanisms, is also crucial in proteinase regulation. mechanism. Instead, most MMPs have a cysteine-switch
Proteinases should be thought of as molecular scissors, usu- mechanism that results in exposure of the active-site cleft
ally cutting with precision, rather than indiscriminately during activation. In contrast to ADAMs, which are trans-
destructive enzymes. Thus, the biology of a proteinase is membrane proteins with an extracellular catalytic domain,
closely linked to biology of its substrates, and it is the effect the ADAMTS proteinases are secreted, but they may func-
on the substrate that is the most relevant to biologic and dis- tion as operational cell-surface or cell-proximate proteinases
ease processes. through their binding to cell-surface/pericellular molecules
(Fig. 8.1). Furthermore, despite sharing propeptide and cata-
lytic domain features, ADAMs and ADAMTSs have entirely
8.1.2 Historical Perspective and Evolution different ancillary domains (Fig. 8.1).
of Metalloproteinases Once genome sequencing of many organisms was com-
pleted, it became apparent that the repertoire of genes encod-
Proteinases are grouped into distinct classes based on the ing for extracellular matrix had expanded substantially in
chemical nature of their catalytic mechanism or their pre- vertebrates (Huxley-Jones et al. 2009), which intuitively
ferred pH optimum. Metalloproteinases (now known as suggests the reason for an observed concomitant expansion
matrix metalloproteinases or MMPs) comprise a very large of genes encoding metzincins (Huxley-Jones et al. 2007).
and diverse group of proteinases whose catalytic mechanism Furthermore, tissue inhibitors of metalloproteinases (TIMPs),
requires a metal ion, most commonly zinc. In 1962, Gross the main endogenous inhibitors of metalloproteinases,
and Lapiere identified the first metalloproteinase (an intersti- evolved from a single gene in Drosophila into the four human
tial collagenase) physiologically responsible for turnover of TIMPs studied today (Brew and Nagase 2010), further sub-
collagen, which is a triple helical rodlike molecule otherwise stantiating the significance of proteolysis in advanced bio-
resistant to cleavage by most proteinases. They used an ele- logic systems.
gant, yet simple model, i.e., digestion of collagen in vitro by MMPs have been extensively studied in orthopaedic biol-
the resorbing vestigial structures of tadpoles during morpho- ogy (Pasternak and Aspenberg 2009). For the purpose of this
genesis, to identify this protease (Gross and Lapiere 1962). chapter, we will focus only on ADAMTS metalloproteinases
Starting with characterization of the tadpole collagenase, the and summarize their emerging roles in the intervertebral
study of MMPs expanded considerably over the past half disc. ADAMs are mentioned where relevant, since they are
century to encompass closely related families: A disintegrin often wrongly thought to be the same as ADAMTS protei-
and metalloproteinase (ADAMs) (Klein and Bischoff 2011), nases yet have been neglected in the context of the disc.
followed by A disintegrin-like and metalloproteinase with The first ADAMTS protease (ADAMTS1) was identified
thrombospondin type 1 motif (ADAMTS) proteinases 15 years ago by Kuno et al. (1997), as an inflammation
(Apte 2009). Collectively, these proteinases are referred to as associated gene product and initially thought to be a variant
metzincins. In general, MMPs have the most diverse sub- ADAM since it had the reprolysin-type catalytic domain
strate repertoire, being involved in processing of multiple (Kuno et al. 1997). Subsequent molecular cloning of 18
ECM components, cytokines, and other soluble proteins, as structurally similar proteinases, facilitated greatly by rapid
well as shedding of cell-surface proteins. ADAMs appear to progress in the human genome project, demonstrated
be almost exclusively involved in ectodomain shedding the existence of this hitherto unknown metalloproteinase
8 The Role of the ADAMTS Proteins in the Intervertebral Disc 127
Fig. 8.1 Distinct domain

structures and differential cell Cell
Metalloprotease membrane
localization of ADAMTS and Disintegrin-like Thrombospondin type 1
domain
ADAM proteinases. The domain domain repeats
structures of ADAMTS5 and
of a typical ADAM proteinase
are shown in the context of the
cell surface, with the various
domains indicated. Note the
ADAMTS Cys-rich region Spacer
transmembrane insertion of
ADAMs. ADAMTS proteinases
are secreted, but several bind
near the cell surface,
potentially through components
of pericellular matrix or to Pericellular
cell-surface molecules matrix
Nucleus
EGF-like
domain
Cytoplasmic
Disintegrin domain domain
ADAM
Transmembrane
Metalloprotease Cys-rich region domain
domain
family as one that is distinct from ADAMs. It should be with a prototypic ADAM, illustrating also the different local-
noted that the enzyme designated as ADAMTS11 ization with respect to cells.
(Abbaszade et al. 1999) is now referred to as ADAMTS5, A high degree of regulation of ADAMTS proteinases is a
and the designation ADAMTS11 is left vacant. Thus, crucial aspect of their biology. Proprotein convertases such
although 20 ADAMTS numbers are assigned, there are 19 as furin cleave ADAMTS and ADAM proteinases after
ADAMTS proteinases. paired basic residues at the junction of the propeptide and
catalytic domains. Furin processing of ADAMs mostly
occurs within the secretory pathway, but some ADAMTS
8.2 ADAMTS Structure and Function proteinases, such as ADAMTS5, ADAMTS7, and
ADAMTS9, may be cleaved at the cell surface or extracel-
Once the full repertoire of ADAMTS proteinases was deter- lularly (Koo and Apte 2009; Longpr et al. 2009). Indeed,
mined, evolutionary analysis showed that they clustered into activation of stored zymogen in the ECM rather than de novo
several distinct subgroups (Apte 2004; Huxley-Jones et al. production of ADAMTS5 has been suggested as a potential
2005) (Fig. 8.2). Typically, ADAMTS proteinases within mechanism for cartilage destruction in arthritis (Malfait et al.
these subgroups have similar domain structures, high 2008; Wylie et al. 2012).
sequence similarity, and sufficient functional overlap that Of the four TIMPs, only TIMP3 is an ADAMTS/ADAM
they work cooperatively in certain contexts. The ADAMTS inhibitor; although all four TIMPs have inhibitory activity
ancillary domain consists of a disintegrin-like module, a towards MMPs, certain TIMP-protease pairings (e.g.,
thrombospondin type 1 repeat (TSR), a cysteine-rich region, TIMP2-MMP2) show higher affinity. TIMP3 inhibitory
a cysteine-free spacer region, one or more additional TSRs, activity towards ADAM17 and ADAMTS4 and ADAMTS5
as well as other modules (Apte 2009) (Fig. 8.2). The differ- implicates it as a key player in regulating inflammation and
ences in the ancillary domains allow each proteinase to bind cartilage matrix breakdown (Kashiwagi et al. 2001; Sahebjam
and act on distinct substrates, and with very few exceptions, et al. 2007). As endogenous inhibitors of the metalloprotei-
ADAMTS catalytic domains without adjoining ancillary nases, TIMPs likely play an important role in ECM turnover
domains lack proteolytic activity. The detailed structure and of the intervertebral disc, although this has yet to be
posttranslational modification of ADAMTS proteases were specifically investigated. The proteinase inhibitor a2-macro-
previously reviewed (Apte 2009). Figure 8.1 contrasts the globulin, which unlike TIMPs has an extremely broad spec-
domain structure of a typical ADAMTS protease (ADAMTS5) trum of inhibition, is also capable of blocking ADAMTS
128 J.C. Ho et al.
Protease domain Ancillary domain Structure/function

defined clades
ADAMTS4
ADAMTS5/8 1
Singnal peptide
Propeptide ADAMTS1/15 2 Proteoglycanases
Catalytic module
ADAMTS9/20 14
Disintegrin-like module
4
ADAMTS6/10 4
TSR (central)
Cysteine-rich module ADAMTS17/19 4

Spacer Procollagen amino-
ADAMTS2/3/14 3 propeptidases
n TSR (C-terminal)
ADAMTS7/12 3 4 Mucin-proteoglycans
PLAC
Procollagen N-propeptidase ADAMTS16/18 5
CUB domain
ADAMTS13 7 vWF-protease
Mucin/proteoglycan module
Gon-1 module
ADAMTS modules
Fig. 8.2 Mammalian ADAMTS proteases. The domain backbone istics that best define them; clades without a known function or a
shared by each ADAMTS protease is shown at the top. The unique defining characteristic are not named. The proteoglycanases constitute
structure of each ADAMTS protease C-terminal to the backbone is a super-clade comprising ADAMTS proteases with different domain
indicated on the right, and the key to these modules is located at left. structures (The figure is based on reference sequences obtained from
Some clades are named according to structural or functional character- GenBank)
activity (Somerville et al. 2004; Tortorella et al. 2004) and enzymes involved in excision of the amino (N)-terminal
presumably does so in the circulation. propeptide of procollagens I, II, and III (Colige et al. 1999,
Despite their brief history, functional contexts were rap- 2002; Fernandes et al. 2001) but could also have additional
idly established for several ADAMTS family members, properties unrelated to procollagens. Removal of the bulky
aided by their clustering into functional family groups procollagen N-propeptides is a prerequisite for proper col-
(Fig. 8.2), stringently established associations with heredi- lagen assembly. In the absence of ADAMTS2, an animal
tary and acquired diseases, and via several natural and engi- disorder named dermatosparaxis results, in which skin and
neered animal mutations. These are summarized in Table 8.1 other collagen-rich tissues show abnormal collagen fibrils
and described in more detail in a previous review (Apte (Lapire and Nusgens 1993), although the intervertebral
2009). Since some of these molecular functions clearly have disc was not one of the tissues specifically investigated.
potential or established relevance to the intervertebral disc, Dermatosparactic collagen forms branched and thin fibrils
they are discussed here and illustrated in Fig. 8.3. ADAMTS1 assuming a hieroglyphic or cauliflower-like pattern in
cleaves aggrecan, versican, thrombospondin-1 and throm- electron microscopy, rather than highly ordered, broad,
bospondin-2, and the cell-surface proteoglycan syndecan-4 unbranched normal fibrils. These anomalous fibrils are
(Sandy et al. 2001; Lee et al. 2006; Rodrguez-Manzaneque structurally weak, a problem most strikingly evident in the
et al. 2009). It is associated with inflammation, cancer exceedingly fragile skin of dermatosparactic cattle. A cor-
cachexia, infertility, urinary tract anomalies, and bone responding human-inherited connective tissue disorder,
metastasis and has potent antiangiogenic activity (Luque Ehlers-Danlos syndrome, dermatosparactic type, is similar
et al. 2003; Mittaz et al. 2004; Apte 2009; Lu et al. 2009). in its clinical presentation (Colige et al. 1999). However,
Thus, ADAMTS1 could be involved not only in matrix pro- neither spine nor disc was specifically reported to be anom-
teolysis but also in inflammatory disc disease. ADAMTS2, alous. In collagen II-rich tissues such as cartilage,
ADAMTS3, and ADAMTS14 are procollagen-processing ADAMTS3 is likely to have a more significant role than
Table 8.1 Functions of selected ADAMTS proteins and potential relevance to intervertebral disc biology
Known functions Potential function in IVD
ADAMTS1 Cleaves versican, thrombospondin-1 and thrombospondin-2, and syndecan-4; inhibits Potentially involved in ECM
angiogenesis. Has a role in TGFb activation. Null mice have impaired fertility, turnover, TGFb activation, and
abnormal cardiac development, and hydronephrosis angiogenesis
ADAMTS2, Removal of amino-propeptide of procollagens I, II, and III. ADAMTS2 mutations lead Potential role in procollagen I and
ADAMTS3, to dermatosparaxis in animals and EDS dermatosparactic type in humans procollagen II processing, collagen
ADAMTS14 assembly, and maintaining tensile
strength of annulus fibrosus
ADAMTS4 Cleaves aggrecan and versican. Null mice are reported to be developmentally normal. Potentially involved in proteoglycan
Combinatorial null mice (with ADAMTS1) have a thin renal medulla turnover in nucleus pulposus, end
plate, and perichondrium
ADAMTS5 Cleaves aggrecan, versican, and biglycan. ADAMTS5 null mice are resistant to induced Potentially involved in proteoglycan
cartilage degeneration. ADAMTS5 null mice lack embryonic sculpting of pulmonic turnover in nucleus pulposus, end
valve leaflets and have reduced interdigital web regression. Cooperates with plate, and perichondrium
ADAMTS9 and ADAMTS20 in interdigital web regression
ADAMTS7, Reported to bind to and cleave cartilage oligomeric protein (COMP, thrombospondin-5) Potentially involved in ECM
ADAMTS12 and granulin-epithelin precursor. ADAMTS12 null mice are developmentally normal turnover in annulus fibrosus
ADAMTS9 Cleaves aggrecan and versican. ADAMTS9 null mice die early during embryogenesis. Potentially involved in proteoglycan
ADAMTS9 haploinsufficient mice have cardiovascular defects. Cooperates with turnover in nucleus pulposus, end
ADAMTS5 and ADAMTS20 in interdigital web regression and with ADAMTS20 in plate, and perichondrium and
closure of the secondary palate in mice. ADAMTS9 is antiangiogenic regulation of angiogenesis
ADAMTS10 Binds to and possibly cleaves fibrillin-1. Promotes fibrillin microfibril assembly. Potential role in fibrillin microfibril
ADAMTS10 is mutated in Weill-Marchesani syndrome in humans assembly
ADAMTS13 ADAMTS13 is required for maturation of ultra-large forms of von Willebrand factor.
ADAMTS13 mutations or autoantibodies lead to thrombotic thrombocytopenic purpura
ADAMTS17 ADAMTS17 mutations lead to a Weill-Marchesani-like syndrome in humans and Potentially involved in fibrillin
recessive isolated ectopia lentis in dogs microfibril assembly
ADAMTS20 Cleaves versican. ADAMTS20 mutations in mice lead to a white spotting mutant Potentially involved in versican
named belted (bt) turnover
ADAMTSL2 ADAMTSL2 binds to fibrillin-1 and latent TGFb-binding protein 1. ADAMTSL2 Expressed in IVD. Potentially
mutations lead to geleophysic dysplasia. In dogs, ADAMTSL2 mutations lead to involved in regulating TGFb
Musladin-Lueke syndrome binding and/or activation
ADAMTSL4 ADAMTSL4 binds to fibrillin-1 and enhances microfibril biogenesis in cultured Potentially involved in microfibril
fibroblasts. ADAMTSL4 mutations lead to recessive isolated ectopia lentis in the eye assembly
ADAMTSL6 ADAMTSL6 binds to fibrillin-1 and enhances microfibril biogenesis in cultured Potentially involved in microfibril
fibroblasts and transgenic mice when it is overexpressed assembly
ADAMTS2 because of its expression in cartilage and dem- malian genomes to have this property. They cleave cartilage
onstrated ability to process procollagen II (Le Goff et al. oligomeric matrix protein (COMP) (Liu et al. 2006), which
2006). is an important component of cartilage ECM, as well as gran-
ADAMTS1, ADAMTS4, ADAMTS5, ADAMTS9, and ulin-epithelin precursor (GEP) (Guo et al. 2010), a growth
ADAMTS20 have the ability to cleave the large aggregating factor reported to be involved in tissue regeneration and
chondroitin sulfate proteoglycans aggrecan and versican at inflammation (Bai et al. 2009). GEP was reported to act as a
specific sites in their core proteins (Apte 2009). ADAMTS9 competitive inhibitor of ADAMTS7 and ADAMTS12 (Guo
is also an antiangiogenic protease that is widely distributed et al. 2010). Based on being a target of PTHrP and cleavage
in microvascular endothelial cells of most organs as an appar- of GEP, ADAMTS7 was proposed as a negative regulator of
ently constitutive product (Koo et al. 2010). ADAMTS4 and endochondral bone formation, but this remains to be substan-
ADAMTS5 seem to be the most important aggrecan- tiated in genetic models (Bai et al. 2009; Liu 2009).
degrading proteinases (aggrecanases) in the articular carti- ADAMTS10 is mutated in a connective tissue disorder
lage degradation in arthritis (Fosang and Little 2008) and named recessive Weill-Marchesani syndrome (WMS). The
have thus been most extensively investigated in the interver- identification of dominantly inherited fibrillin-1 mutations in
tebral disc, although much on the published analysis is of WMS (Faivre et al. 2003; Sengle et al. 2012) suggested a
gene expression and protein distribution rather than de novo functional link between ADAMTS10 and fibrillin-1. Such
functional analysis. ADAMTS7 and ADAMTS12 are unique fibrillin-1 mutations typically lead to Marfan syndrome, yet
proteinases in also being chondroitin sulfate proteoglycans WMS appears to constitute an opposite of Marfan syn-
(Somerville et al. 2004), the only known proteinases in mam- drome, featuring short stature, short digits, stiff skin and
130 J.C. Ho et al.
Fig. 8.3 Potential roles

of ADAMTS proteinases in
various processes within the
context of intervertebral disc B
components. The nucleus Versican turnover
pulposus (NP), annulus fibrosus (ADAMTS1, 4, 5, 9, 20)
(AF), cartilaginous end plate
(EP), and vertebral bone (B) are
indicated EP
Microfibril assembly Aggrecan turnover

(ADAMTS10) (ADAMTS1, 4, 5, 9)
COMP processing
(ADAMTS7, 12) Thrombospondin-1 & 2 cleavage
(ADAMTS1)
NP
TGF control
(ADAMTSL2)
Procollagen processing
(ADAMTS2, 3, 14)
AF
joints, and in the eye, lens dislocation, and glaucoma. (Bader et al. 2010). Analysis of cells from GD patients
Recent work showed that ADAMTS10 bound to fibrillin-1 has suggested that there is profound TGFb dysregulation,
and fibrillin-2 was associated with fibrillin microfibrils in likely related to the pivotal role of fibrillin microfibrils in
tissues and enhanced fibrillin-1 microfibrils formation regulating TGFb and bone morphogenetic proteins (Le
in vitro (Kutz et al. 2011). Goff et al. 2008). Indeed, ADAMTSL2 also binds to latent
A unique aspect of ADAMTS proteases is the exis- TGFb-binding protein-1 (LTBP1), which strongly sup-
tence of a closely related family of seven independent gene ports a role in TGFb sequestration in ECM or activation.
products resembling ADAMTS ancillary domains (Apte We recently reviewed the strong functional involvement of
2009). These molecules, named ADAMTS-like proteins ADAMTS proteinases in connective tissue regulation vis--
(ADAMTSL), lack catalytic domains and are therefore vis fibrillins, specifically, in regulating the cellular microen-
not proteinases; however, together with ADAMTS protei- vironment (Hubmacher and Apte 2011). ADAMTSL2 is
nases, they are thought to comprise a protein superfamily. expressed in the nucleus pulposus of the intervertebral disc
Intriguingly, like ADAMTS10, the majority of ADAMTSLs, (Koo et al. 2007; Sohn et al. 2010), and it could have a
specifically, ADAMTSL2, ADAMTSL3, ADAMTSL4, and role in disc development because of the established role of
ADAMTSL6, also bind to or influence microfibril forma- TGFb signaling in this process (Sohn et al. 2010).
tion (Le Goff et al. 2011; Saito et al. 2011; Gabriel et al. Another cluster of ADAMTS proteinases highly relevant
2012; Sengle et al. 2012) (Table 8.1). Human genetic dis- to IVD is the so-called proteoglycanase cluster, containing
orders resulting from ADAMTSL2 or ADAMTSL4 muta- ADAMTS1, ADAMTS4, ADAMTS5, ADAMTS9, and
tions, i.e., geleophysic dysplasia (GD) and isolated ectopia ADAMTS20 (Apte 2004; Huxley-Jones et al. 2005). The
lentis, respectively, are also caused by fibrillin-1 mutations, major proteoglycan substrates of these proteinases are
and both these proteins bind to fibrillin-1 (Hubmacher and aggrecan, a cartilage-specific chondroitin sulfate proteogly-
Apte 2011; Le Goff et al. 2011). The mutations in GD lead can, and its widely distributed relative in non-cartilaginous
to a short-stature, short-digit phenotype superficially resem- tissues, versican. The role of these proteinases in aggrecan
bling WMS (these conditions fall within a category named destruction in osteoarthritis has been investigated by strin-
acromelic dysplasias), although GD is much more severe gent analysis of null mice, biochemical assays, and associa-
and frequently lethal in children because of cardiac involve- tion of mRNA, protein, and catabolic fragments with
ment; however, unlike WMS, it lacks ocular involvement arthritic cartilage (Fosang and Little 2008). ADAMTS5-
(Le Goff et al. 2008). ADAMTSL2 mutations in dogs also deficient mice are protected against either a mechanical
cause dwarfism and severe skin and joint stiffness, a con- instability-induced or cytokine-induced cartilage breakdown
nective tissue disorder named Musladin-Lueke syndrome (Glasson et al. 2005; Stanton et al. 2005).
Although these five enzymes are potent aggrecanases, amounts include versican, decorin, biglycan, fibromodulin,
which of them has a role in physiological aggrecan break- lumican, and perlecan (Roughley 2004). Aggrecan is found
down, such as in skeletal development, is unclear. However, in both the nucleus pulposus and annulus fibrosus, mostly as
some of them were discovered to have a crucial role in turnover aggregates complexed with hyaluronan and link protein,
of versican in several developmental contexts, specifically, although it makes up a larger proportion of the nucleus pul-
myocardial compaction, closure of the secondary palate, posus (65 % dry weight) than the annulus fibrosus (1520 %
sculpting of heart valves, and resorption of interdigital webs dry weight) (Le Maitre et al. 2007). The CEP of the vertebral
(Stankunas et al. 2008; McCulloch et al. 2009; Enomoto bodies are similar in content and structure to articular carti-
et al. 2010; Dupuis et al. 2011). Two intriguing findings lage found in other joints and are primarily comprised of col-
emerged from these discoveries: One, that proteoglycanases lagen II fibers with aggrecan aggregates complexed to
cooperated in versican proteolysis in the context of palate hyaluronan and link protein. Previous chapters in this book
closure and web regression, and two, that a product of ver- (Chaps. 4 and 5) provide further details regarding the roles of
sican proteolysis was a matrikine with context-dependent the proteoglycans and collagens in the intervertebral disc.
function in cell proliferation or apoptosis in these processes As the intervertebral disc ages, its ECM undergoes
(McCulloch et al. 2009; Enomoto et al. 2010). Further study significant catabolic changes in which ADAMTS proteinases
of these families will no doubt find other important func- are likely to have a role. The majority of experimental stud-
tions of these proteinases in normal development and disease ies to date have focused on the levels and localization of
pathogenesis. ADAMTS4, ADAMTS5, and catabolic products of aggre-
In contrast to these functions of ADAMTS proteinases can. Using immunohistochemistry, ADAMTS4 was detected
which are directly relevant to extracellular matrix matura- in nondegrading disc cells from the nucleus pulposus and
tion, assembly, and turnover, ADAMs are primarily involved inner annulus fibrosus, with little immunoreactivity in the
in ectodomain shedding of cell-surface molecules. In partic- outer annulus fibrosus, suggesting a maintenance role, but in
ular they are identified to have a pivotal role in growth con- degenerated discs, the authors found an increase in
trol through epidermal growth factor receptor signaling ADAMTS4 that correlated with the severity of degeneration
pathways, in cell adhesion, in Notch signaling, and in and increased TIMP1 and TIMP2 but not TIMP3 (Le Maitre
inflammation through processing of pro-TNF-a and cytokine et al. 2004). It was also found that the presence of inflammatory
receptors (Klein and Bischoff 2011; Saftig and Reiss 2011). modulators such as TNF-a and IL-1 increased ADAMTS4
Thus, ADAM proteinases may play a role in disc develop- and ADAMTS5 expression and ADAMTS aggrecan degra-
ment and degeneration through modulation of these and sev- dation (Le Maitre et al. 2005; Sguin et al. 2005). Toll-like
eral other pathways. Unfortunately direct effects of ADAM receptor adaptor signaling molecule MyD88 antagonized
proteinases in the intervertebral disc have not been eluci- LPS or IL-1-mediated induction of ADAMTS4 and
dated to any significant extent, and it remains an area of ADAMTS5 (Ellman et al. 2012). An immunohistochemical
opportunity for future research. analysis of surgically resected discs showed ADAMTS4 pri-
marily in CD68-positive mononuclear cells (monocyte/mac-
rophages) in granulation tissue and adjacent disc, with a
8.3 ADAMTS Proteinases in Intervertebral higher number of stained cells associated with specific her-
Disc Biology and Disease niation patterns (transligamentous extrusion and sequestra-
tion). A recent detailed analysis investigated the role of
Three anatomic components of the intervertebral disc are rel- syndecan-4 in ADAMTS5 activity in nucleus pulposus cells.
evant to its physiology and pathology in regard to metallo- TNF-a and IL-1a increased ADAMTS4 and ADAMTS5
proteinases, namely the cartilaginous end plates, annulus expression and promoted syndecan-4 interaction with
fibrosus, and nucleus pulposus, which act together to fulfill ADAMTS5 (Wang et al. 2011). This was consistent with an
the biological and mechanical functions of the disc but also association of aggrecan degradation with increased synde-
differ in their matrix structure because each is mechanically can-4 and ADAMTS5 in the human intervertebral disc (Wang
specialized. The two major components of the ECM in the et al. 2011).
disc are the proteoglycan aggrecan and collagen, which are In addition to inflammatory modulators, studies have also
cleaved and modulated by ADAMTS proteinases. Collagen shown that mechanical loading can alter proteinase expres-
I, II, III, V, VI, IX, XI, XII, and XIV can be found in the disc sion (Maclean et al. 2004; MacLean et al. 2005). Long-term
at varying levels and will also change with age (Eyre et al. upright posture in rats demonstrated increased disc expres-
2002). Of the collagens, I and II are most abundant, with I sion of MMP13, ADAMTS5, and Col10a1 and decreased
comprising the outer layers of the annulus fibrosus and II Col2a1 and Acan (Liang et al. 2008). Static compression of
mainly located in the nucleus. Besides aggrecan in the rat disc led to upregulation of ADAMTS4, but not ADAMTS5,
nucleus pulposus, other proteoglycans found in lower with concomitant increase of aggrecan catabolic fragments
132 J.C. Ho et al.
(Yurube et al. 2012). Loading of rat caudal intervertebral An intriguing study suggested that immunity to the versi-
discs was noted to increase expression of ADAMTS7 and can G1 domain led to development of spondylitis in mice
ADAMTS12 mRNA (Yu and Zhu 2012). In human nucleus (Shi et al. 2003). Since one of the versican fragments released
pulposus cells, a dynamic compressive load led to upregula- by ADAMTS proteases primarily comprises the G1 domain,
tion of ADAMTS1, ADAMTS4, and ADAMTS5 mRNA and it is an interesting possibility that humoral immunity to
the respective proteins (Huang et al. 2012), with the response ADAMTS cleaved and released versican G1 domain could
to cyclic tensile stress being frequency dependent (Gilbert have a role in development of spinal pathology.
et al. 2010). Figure 8.3 summarizes the various ways in which
A comparative gene expression analysis of healthy and ADAMTS superfamily proteins, including ADAMTS protei-
severely degenerated discs identified downregulation of nases and ADAMTSLs, could influence intervertebral disc
ADAMTSL3 and ADAMTS10 mRNA in degenerated discs biology and disease. Expectedly, analysis of aggrecan break-
(Gruber et al. 2011). When comparing degenerated versus down has seen the lions share of attention, owing to its rel-
nondegenerated discs, it was found that ADAMTS1, evance for disc degeneration. The considerable evidence
ADAMTS4, ADAMTS5, and ADAMTS15 expression were accumulated to date strongly associates ADAMTS4 and
significantly increased in degenerated tissue and ADAMTS4, ADAMTS5 with aggrecan breakdown in the nucleus pulpo-
ADAMTS5, ADAMTS9, and ADAMTS15 immunohis- sus; however, definitive experiments using genetically
tochemical staining also increased (Pockert et al. 2009). deficient mice remain to be done. As shown in Fig. 8.3, the
A study of human discs suggested that ADAMTS5 was procollagen proteinases could have a crucial role in normal
involved in intervertebral disc degeneration and that IL-1 disc development and repair, since their function appears to
induction of ADAMTS5 was mediated by nitric oxide (Zhao be primarily anabolic. The evidence for expression of
et al. 2011); another showed multiple positive correlations ADAMTS7 and ADAMTS12 in the intervertebral disc is
between MMPs and ADAMTS4 mRNA in herniated discs. compelling, but the precise functions of these proteinases
In addition to defining the expression of proteinases in the remain unknown. The most intriguing possibilities, however,
disc itself, there appear to be significant differences among are related to the potential for microenvironmental regula-
the cell types of the disc. It was found that nucleus pulposus tion by ADAMTSLs and ADAMTS10. ADAMTSL2 and
cells express more ADAMTS1, ADAMTS2, ADAMTS17, ADAMTS10 (authors unpublished data) are both expressed
and TIMP1, whereas ADAMTS4, ADAMTS5, ADAMTS6, in the intervertebral disc, associated with inherited connec-
ADAMTS14, ADAMTS18, ADAMTS19 and TIMP3 were tive tissue disorders and functionally linked to fibrillins,
lower in nucleus cells versus articular chondrocytes (Cui which control both TGFbs and bone morphogenetic proteins.
et al. 2010). A sheep annulus fibrosus transection model Their potential role is worthy of further investigation.
showed increased ADAMTS5 mRNA and ADAMTS4 A major challenge, however, is functional overlap (referred
mRNA in the annulus, but only ADAMTS5 mRNA increased to as redundancy) between such functionally related proteins
in the nucleus pulposus, whereas ADAMTS4 mRNA that requires complex combinatorial genetics to elucidate
decreased; however, there was no change in aggrecan neo- their true roles. Finally, the ADAMs have been substantially
epitopes (Melrose et al. 2012). neglected in disc research, yet their function is highly rele-
ADAMTS1 and ADAMTS5 were identified as candidate vant to tissue inflammation and repair.
genes for lumbar disc disease in the Finnish population
(Virtanen et al. 2007). A gene-profiling study comparing
chondrocytes and nucleus pulposus cells found that NP cells 8.4 Summary of Critical Concepts
expressed higher levels of ADAMTS1, ADAMTS2, and Discussed in the Chapter
ADAMTS17, but lower levels of ADAMTS4, ADAMTS5,
ADAMTS6, ADAMTS14, ADAMTS18, ADAMTS19, and The proteolytic activities and regulatory mechanisms of
TIMP3. Such analysis could be useful in identifying differ- ADAMTS and ADAM proteinases have considerable
ences between the metabolic pathways of chondrocytes and potential relevance for the intervertebral disc.
nucleus pulposus cells. ADAMTS5 mRNA was increased in ADAMTS proteinases have been studied in the disc pri-
rat nucleus pulposus cells after inflammatory stimulation via marily at the level of mRNA and protein expression. The
a NO-dependent pathway, and the amount of ADAMTS5 published work has focused on a small number of pro-
protein appeared to increase with increasing age (Zhao et al. teases, namely, ADAMTS4, ADAMTS5, and ADAMTS7,
2011). It was also found that ADAMTS4, but not ADAMTS5, yet the activities of many other family members as well as
content was higher in Grade 4 versus Grade 2 degenerative ADAMTSLs are worthy of deeper investigation.
tissue, but there was no difference in the amount of fragments ADAM proteinases have not been extensively investi-
in each group, suggesting involvement of other proteases gated in disc disease and are likely to be highly relevant to
such as MMPs in turnover of aggrecan (Patel et al. 2007). inflammation and cell-surface proteolysis.
With further investigation into the roles of these protei- Enomoto H, Nelson CM, Somerville RPT, Mielke K, Dixon LJ,
nases into pathogenesis of intervertebral disc disease, Powell K, Apte SS (2010) Cooperation of two ADAMTS metallo-
proteases in closure of the mouse palate identifies a requirement for
future identification of potential disease-modifying tar- versican proteolysis in regulating palatal mesenchyme proliferation.
gets is viable and could capitalize on availability of sev- Development 137:40294038
eral well-characterized proteinase inhibitors. Eyre DR, Matsui Y, Wu J-J (2002) Collagen polymorphisms of the
intervertebral disc. Biochem Soc Trans 30:844848
Faivre L, Gorlin RJ, Wirtz MK, Godfrey M, Dagoneau N, Samples JR,
Acknowledgment This work was supported by NIH award Le Merrer M, Collod-Beroud G, Boileau C, Munnich A,
AR53890. Cormier-Daire V (2003) In frame fibrillin-1 gene deletion in auto-
somal dominant Weill-Marchesani syndrome. J Med Genet 40:
3436
Fernandes RJ, Hirohata S, Engle JM, Colige A, Cohn DH, Eyre DR,
References Apte SS (2001) Procollagen II amino propeptide processing by
ADAMTS-3. Insights on dermatosparaxis. J Biol Chem 276:
Abbaszade I, Liu RQ, Yang F, Rosenfeld SA, Ross OH, Link JR, 3150231509
Ellis DM, Tortorella MD, Pratta MA, Hollis JM, Wynn R, Duke JL, Fosang AJ, Little CB (2008) Drug insight: aggrecanases as therapeutic
George HJ, Hillman MC Jr, Murphy K, Wiswall BH, Copeland RA, targets for osteoarthritis. Nat Clin Pract Rheumatol 4:420427
Decicco CP, Bruckner R, Nagase H, Itoh Y, Newton RC, Gabriel LAR, Wang LW, Bader H, Ho JC, Majors AK, Hollyfield JG,
Magolda RL, Trzaskos JM, Burn TC (1999) Cloning and character- Traboulsi EI, Apte SS (2012) ADAMTSL4, a secreted glycoprotein
ization of ADAMTS11, an aggrecanase from the ADAMTS family. widely distributed in the eye, binds fibrillin-1 microfibrils and accel-
J Biol Chem 274:2344323450 erates microfibril biogenesis. Invest Ophthalmol Vis Sci 53:
Apte SS (2004) A disintegrin-like and metalloprotease (reprolysin type) 461469
with thrombospondin type 1 motifs: the ADAMTS family. Int Gilbert HTJ, Hoyland JA, Millward-Sadler SJ (2010) The response of
J Biochem Cell Biol 36:981985 human annulus fibrosus cells to cyclic tensile strain is frequency-
Apte SS (2009) A disintegrin-like and metalloprotease (reprolysin- dependent and altered with disc degeneration. Arthritis Rheum
type) with thrombospondin type 1 motif (ADAMTS) superfamily: 62:33853394
functions and mechanisms. J Biol Chem 284:3149331497 Glasson SS, Askew R, Sheppard B, Carito B, Blanchet T, Ma H-L,
Bader HL, Ruhe AL, Wang LW, Wong AK, Walsh KF, Packer RA, Flannery CR, Peluso D, Kanki K, Yang Z, Majumdar MK, Morris
Mitelman J, Robertson KR, OBrien DP, Broman KW, Shelton GD, EA (2005) Deletion of active ADAMTS5 prevents cartilage degra-
Apte SS, Neff MW (2010) An ADAMTSL2 founder dation in a murine model of osteoarthritis. Nature 434:644648
mutation causes Musladin-Lueke Syndrome, a heritable disorder Gross J, Lapiere CM (1962) Collagenolytic activity in amphibian
of beagle dogs, featuring stiff skin and joint contractures. PLoS tissues: a tissue culture assay. Proc Natl Acad Sci U S A 48:
One 5 10141022
Bai X-H, Wang D-W, Kong L, Zhang Y, Luan Y, Kobayashi T, Gruber HE, Hoelscher GL, Ingram JA, Bethea S, Zinchenko N,
Kronenberg HM, Yu X-P, Liu C-J (2009) ADAMTS-7, a direct Hanley EN Jr (2011) Variations in aggrecan localization and gene
target of PTHrP, adversely regulates endochondral bone growth expression patterns characterize increasing stages of human
by associating with and inactivating GEP growth factor. Mol Cell intervertebral disk degeneration. Exp Mol Pathol 91:534539
Biol 29:42014219 Guo F, Lai Y, Tian Q, Lin EA, Kong L, Liu C (2010) Granulin-epithelin
Brew K, Nagase H (2010) The tissue inhibitors of metalloproteinases precursor binds directly to ADAMTS-7 and ADAMTS-12 and
(TIMPs): an ancient family with structural and functional diversity. inhibits their degradation of cartilage oligomeric matrix protein.
Biochim Biophys Acta 1803:5571 Arthritis Rheum 62:20232036
Colige A, Sieron AL, Li SW, Schwarze U, Petty E, Wertelecki W, Huang M, Wang H-Q, Zhang Q, Yan X-D, Hao M, Luo Z-J (2012)
Wilcox W, Krakow D, Cohn DH, Reardon W, Byers PH, Lapire Alterations of ADAMTSs and TIMP-3 in human nucleus pulposus
CM, Prockop DJ, Nusgens BV (1999) Human Ehlers-Danlos syn- cells subjected to compressive load: implications in the pathogene-
drome type VII C and bovine dermatosparaxis are caused by muta- sis of human intervertebral disc degeneration. J Orthop Res
tions in the procollagen I N-proteinase gene. Am J Hum Genet 30:267273
65:308317 Hubmacher D, Apte SS (2011) Genetic and functional linkage between
Colige A, Vandenberghe I, Thiry M, Lambert CA, Van Beeumen J, ADAMTS superfamily proteins and fibrillin-1: a novel mechanism
Li S-W, Prockop DJ, Lapiere CM, Nusgens BV (2002) Cloning and influencing microfibril assembly and function. Cell Mol Life Sci
characterization of ADAMTS-14, a novel ADAMTS displaying 68:31373148
high homology with ADAMTS-2 and ADAMTS-3. J Biol Chem Huxley-Jones J, Apte SS, Robertson DL, Boot-Handford RP (2005)
277:57565766 The characterisation of six ADAMTS proteases in the basal chor-
Cui Y, Yu J, Urban JPG, Young DA (2010) Differential gene expression date Ciona intestinalis provides new insights into the vertebrate
profiling of metalloproteinases and their inhibitors: a comparison ADAMTS family. Int J Biochem Cell Biol 37:18381845
between bovine intervertebral disc nucleus pulposus cells and Huxley-Jones J, Clarke T-K, Beck C, Toubaris G, Robertson DL,
articular chondrocytes. Spine 35:11011108 Boot-Handford RP (2007) The evolution of the vertebrate metz-
Dupuis LE, McCulloch DR, McGarity JD, Bahan A, Wessels A, Weber incins; insights from Ciona intestinalis and Danio rerio. BMC Evol
D, Diminich AM, Nelson CM, Apte SS, Kern CB (2011) Altered Biol 7:63
versican cleavage in ADAMTS5 deficient mice; a novel etiology of Huxley-Jones J, Pinney JW, Archer J, Robertson DL, Boot-Handford
myxomatous valve disease. Dev Biol 357:152164 RP (2009) Back to basicshow the evolution of the extracellular
Ellman MB, Kim J-S, An HS, Chen D, Kc R, An J, Dittakavi T, van matrix underpinned vertebrate evolution. Int J Exp Pathol
Wijnen AJ, Cs-Szabo G, Li X, Xiao G, An S, Kim S-G, Im H-J 90:95100
(2012) Toll-like receptor adaptor signaling molecule MyD88 on Kashiwagi M, Tortorella M, Nagase H, Brew K (2001) TIMP-3 is a
intervertebral disk homeostasis: in vitro, ex vivo studies. Gene potent inhibitor of aggrecanase 1 (ADAM-TS4) and aggrecanase 2
505:283290 (ADAM-TS5). J Biol Chem 276:1250112504
134 J.C. Ho et al.
Klein T, Bischoff R (2011) Active metalloproteases of the A Disintegrin Liu C-J, Kong W, Ilalov K, Yu S, Xu K, Prazak L, Fajardo M, Sehgal B,
and Metalloprotease (ADAM) family: biological function and Di Cesare PE (2006) ADAMTS-7: a metalloproteinase that directly
structure. J Proteome Res 10:1733 binds to and degrades cartilage oligomeric matrix protein. FASEB J
Koo B-H, Apte SS (2009) Cell-surface processing of the metallopro- 20:988990
tease pro-ADAMTS9 is influenced by the chaperone GRP94/GP96. Longpr J-M, McCulloch DR, Koo B-H, Alexander JP, Apte SS, Leduc
J Biol Chem 285(1):197205 R (2009) Characterization of proADAMTS5 processing by propro-
Koo B-H, Le Goff C, Jungers KA, Vasanji A, OFlaherty J, tein convertases. Int J Biochem Cell Biol 41:11161126
Weyman CM, Apte SS (2007) ADAMTS-like 2 (ADAMTSL2) is a Lu X, Wang Q, Hu G, Van Poznak C, Fleisher M, Reiss M, Massagu J,
secreted glycoprotein that is widely expressed during mouse Kang Y (2009) ADAMTS1 and MMP1 proteolytically engage
embryogenesis and is regulated during skeletal myogenesis. Matrix EGF-like ligands in an osteolytic signaling cascade for bone metas-
Biol 26:431441 tasis. Genes Dev 23:18821894
Koo B-H, Coe DM, Dixon LJ, Somerville RPT, Nelson CM, Wang LW, Luque A, Carpizo DR, Iruela-Arispe ML (2003) ADAMTS1/METH1
Young ME, Lindner DJ, Apte SS (2010) ADAMTS9 is a cell-auton- inhibits endothelial cell proliferation by direct binding and seques-
omously acting, anti-angiogenic metalloprotease expressed by tration of VEGF165. J Biol Chem 278:2365623665
microvascular endothelial cells. Am J Pathol 176:14941504 Maclean JJ, Lee CR, Alini M, Iatridis JC (2004) Anabolic and catabolic
Kuno K, Kanada N, Nakashima E, Fujiki F, Ichimura F, Matsushima K mRNA levels of the intervertebral disc vary with the magnitude and
(1997) Molecular cloning of a gene encoding a new type of metal- frequency of in vivo dynamic compression. J Orthop Res
loproteinase-disintegrin family protein with thrombospondin motifs 22:11931200
as an inflammation associated gene. J Biol Chem 272:556562 MacLean JJ, Lee CR, Alini M, Iatridis JC (2005) The effects of short-
Kutz WE, Wang LW, Bader HL, Majors AK, Iwata K, Traboulsi EI, term load duration on anabolic and catabolic gene expression in the
Sakai LY, Keene DR, Apte SS (2011) ADAMTS10 protein interacts rat tail intervertebral disc. J Orthop Res 23:11201127
with fibrillin-1 and promotes its deposition in extracellular matrix of Malfait A-M, Arner EC, Song R-H, Alston JT, Markosyan S, Staten N,
cultured fibroblasts. J Biol Chem 286:1715617167 Yang Z, Griggs DW, Tortorella MD (2008) Proprotein convertase
Lapire CM, Nusgens BV (1993) Ehlers-Danlos type VII-C, or human activation of aggrecanases in cartilage in situ. Arch Biochem
dermatosparaxis. The offspring of a union between basic and clini- Biophys 478:4351
cal research. Arch Dermatol 129:13161319 McCulloch DR, Nelson CM, Dixon LJ, Silver DL, Wylie JD, Lindner V,
Le Goff C, Somerville RPT, Kesteloot F, Powell K, Birk DE, Colige AC, Sasaki T, Cooley MA, Argraves WS, Apte SS (2009) ADAMTS
Apte SS (2006) Regulation of procollagen amino-propeptide pro- metalloproteases generate active versican fragments that regulate
cessing during mouse embryogenesis by specialization of homolo- interdigital web regression. Dev Cell 17:687698
gous ADAMTS proteases: insights on collagen biosynthesis and Melrose J, Shu C, Young C, Ho R, Smith M, Young A, Smith S,
dermatosparaxis. Development 133:15871596 Gooden B, Dart A, Podadera J, Appleyard R, Little C (2012)
Le Goff C, Morice-Picard F, Dagoneau N, Wang LW, Perrot C, Crow YJ, Mechanical destabilisation induced by controlled annular incision
Bauer F, Flori E, Prost-Squarcioni C, Krakow D, Ge G, Greenspan of the intervertebral disc dysregulates metalloproteinase expression
DS, Bonnet D, Le Merrer M, Munnich A, Apte SS, Cormier-Daire and induces disc degeneration. Spine (Phila Pa 1976) 37:1825
V (2008) ADAMTSL2 mutations in geleophysic dysplasia demon- Mittaz L, Russell DL, Wilson T, Brasted M, Tkalcevic J, Salamonsen
strate a role for ADAMTS-like proteins in TGF-beta bioavailability LA, Hertzog PJ, Pritchard MA (2004) Adamts-1 is essential for the
regulation. Nat Genet 40:11191123 development and function of the urogenital system. Biol Reprod
Le Goff C, Mahaut C, Wang LW, Allali S, Abhyankar A, Jensen S, 70:10961105
Zylberberg L, Collod-Beroud G, Bonnet D, Alanay Y, Brady AF, Pasternak B, Aspenberg P (2009) Metalloproteinases and their inhibi-
Cordier M-P, Devriendt K, Genevieve D, Kiper PS, Kitoh H, tors-diagnostic and therapeutic opportunities in orthopedics. Acta
Krakow D, Lynch SA, Le Merrer M, Mgarbane A, Mortier G, Orthop 80:693703
Odent S, Polak M, Rohrbach M, Sillence D, Stolte-Dijkstra I, Patel KP, Sandy JD, Akeda K, Miyamoto K, Chujo T, An HS, Masuda K
Superti-Furga A, Rimoin DL, Topouchian V, Unger S, Zabel B, (2007) Aggrecanases and aggrecanase-generated fragments in the
Bole-Feysot C, Nitschke P, Handford P, Casanova J-L, Boileau C, human intervertebral disc at early and advanced stages of disc
Apte SS, Munnich A, Cormier-Daire V (2011) Mutations in the degeneration. Spine 32:25962603
TGFb binding-protein-like domain 5 of FBN1 are responsible for Pockert AJ, Richardson SM, Le Maitre CL, Lyon M, Deakin JA,
acromicric and geleophysic dysplasias. Am J Hum Genet 89:714 Buttle DJ, Freemont AJ, Hoyland JA (2009) Modified expression of
Le Maitre CL, Freemont AJ, Hoyland JA (2004) Localization of degra- the ADAMTS enzymes and tissue inhibitor of metalloproteinases 3
dative enzymes and their inhibitors in the degenerate human inter- during human intervertebral disc degeneration. Arthritis Rheum
vertebral disc. J Pathol 204:4754 60:482491
Le Maitre CL, Freemont AJ, Hoyland JA (2005) The role of interleu- Rodrguez-Manzaneque JC, Carpizo D, Plaza-Calonge Mdel C,
kin-1 in the pathogenesis of human intervertebral disc degeneration. Torres-Collado AX, Thai SN-M, Simons M, Horowitz A, Iruela-
Arthritis Res Ther 7:R732R745 Arispe ML (2009) Cleavage of syndecan-4 by ADAMTS1 provokes
Le Maitre CL, Pockert A, Buttle DJ, Freemont AJ, Hoyland JA (2007) defects in adhesion. Int J Biochem Cell Biol 41:800810
Matrix synthesis and degradation in human intervertebral disc Roughley PJ (2004) Biology of intervertebral disc aging and degenera-
degeneration. Biochem Soc Trans 35:652655 tion: involvement of the extracellular matrix. Spine 29:26912699
Lee NV, Sato M, Annis DS, Loo JA, Wu L, Mosher DF, Iruela-Arispe Saftig P, Reiss K (2011) The A Disintegrin and Metalloproteases
ML (2006) ADAMTS1 mediates the release of antiangiogenic poly- ADAM10 and ADAM17: novel drug targets with therapeutic poten-
peptides from TSP1 and 2. EMBO J 25:52705283 tial? Eur J Cell Biol 90:527535
Liang Q-Q, Zhou Q, Zhang M, Hou W, Cui X-J, Li C-G, Li T-F, Shi Q, Sahebjam S, Khokha R, Mort JS (2007) Increased collagen and aggre-
Wang Y-J (2008) Prolonged upright posture induces degenerative can degradation with age in the joints of Timp3(/) mice. Arthritis
changes in intervertebral discs in rat lumbar spine. Spine Rheum 56:905909
33:20522058 Saito M, Kurokawa M, Oda M, Oshima M, Tsutsui K, Kosaka K, Nakao
Liu C-J (2009) The role of ADAMTS-7 and ADAMTS-12 in the patho- K, Ogawa M, Manabe R, Suda N, Ganjargal G, Hada Y, Noguchi T,
genesis of arthritis. Nat Clin Pract Rheumatol 5:3845 Teranaka T, Sekiguchi K, Yoneda T, Tsuji T (2011) ADAMTSL6b
protein rescues fibrillin-1 microfibril disorder in a Marfan syndrome (2005) ADAMTS5 is the major aggrecanase in mouse cartilage
mouse model through the promotion of fibrillin-1 assembly. J Biol in vivo and in vitro. Nature 434:648652
Chem 286:3860238613 Tortorella MD, Arner EC, Hills R, Easton A, Korte-Sarfaty J, Fok K,
Sandy JD, Westling J, Kenagy RD, Iruela-Arispe ML, Verscharen C, Wittwer AJ, Liu R-Q, Malfait A-M (2004) Alpha2-macroglobulin is
Rodriguez-Mazaneque JC, Zimmermann DR, Lemire JM, Fischer a novel substrate for ADAMTS-4 and ADAMTS-5 and represents
JW, Wight TN, Clowes AW (2001) Versican V1 proteolysis in an endogenous inhibitor of these enzymes. J Biol Chem 279:
human aorta in vivo occurs at the Glu441-Ala442 bond, a site that is 1755417561
cleaved by recombinant ADAMTS-1 and ADAMTS-4. J Biol Chem Virtanen IM, Noponen N, Barral S, Karppinen J, Li H, Vuoristo M,
276:1337213378 Niinimki J, Ott J, Ala-Kokko L, Mnnikk M (2007) Putative sus-
Sguin CA, Pilliar RM, Roughley PJ, Kandel RA (2005) Tumor necro- ceptibility locus on chromosome 21q for lumbar disc disease (LDD)
sis factor-alpha modulates matrix production and catabolism in in the Finnish population. J Bone Miner Res 22:701707
nucleus pulposus tissue. Spine 30:19401948 Wang J, Markova D, Anderson DG, Zheng Z, Shapiro IM, Risbud MV
Sengle G, Tsutsui K, Keene DR, Tufa SF, Carlson EJ, Charbonneau NL, (2011) TNF-a and IL-1b promote a disintegrin-like and metallopro-
Ono RN, Sasaki T, Wirtz MK, Samples JR, Fessler LI, Fessler JH, tease with thrombospondin type I motif-5-mediated aggrecan degra-
Sekiguchi K, Hayflick SJ, Sakai LY (2012) Microenvironmental dation through syndecan-4 in intervertebral disc. J Biol Chem 286:
regulation by fibrillin-1. PLoS Genet 8:e1002425 3973839749
Shi S, Ciurli C, Cartman A, Pidoux I, Poole AR, Zhang Y (2003) Wylie JD, Ho JC, Singh S, McCulloch DR, Apte SS (2012) Adamts5
Experimental immunity to the G1 domain of the proteoglycan ver- (aggrecanase-2) is widely expressed in the mouse musculoskeletal
sican induces spondylitis and sacroiliitis, of a kind seen in human system and is induced in specific regions of knee joint explants by
spondylarthropathies. Arthritis Rheum 48:29032915 inflammatory cytokines. J Orthop Res 30:226233
Sohn P, Cox M, Chen D, Serra R (2010) Molecular profiling of the Yu H, Zhu Y (2012) Expression of ADAMTS-7 and ADAMTS-12 in
developing mouse axial skeleton: a role for Tgfbr2 in the develop- the nucleus pulposus during degeneration of rat caudal interverte-
ment of the intervertebral disc. BMC Dev Biol 10:29 bral disc. J Vet Med Sci 74:915
Somerville RPT, Longpr J-M, Apel ED, Lewis RM, Wang LW, Sanes Yurube T, Takada T, Suzuki T, Kakutani K, Maeno K, Doita M,
JR, Leduc R, Apte SS (2004) ADAMTS7B, the full-length product Kurosaka M, Nishida K (2012) Rat tail static compression model
of the ADAMTS7 gene, is a chondroitin sulfate proteoglycan con- mimics extracellular matrix metabolic imbalances of matrix metal-
taining a mucin domain. J Biol Chem 279:3515935175 loproteinases, aggrecanases, and tissue inhibitors of metalloprotei-
Stankunas K, Hang CT, Tsun Z-Y, Chen H, Lee NV, Wu JI, Shang C, nases in intervertebral disc degeneration. Arthritis Res Ther
Bayle JH, Shou W, Iruela-Arispe ML, Chang C-P (2008) Endocardial 14:R51
Brg1 represses ADAMTS1 to maintain the microenvironment for Zhao C-Q, Zhang Y-H, Jiang S-D, Li H, Jiang L-S, Dai L-Y (2011)
myocardial morphogenesis. Dev Cell 14:298311 ADAMTS-5 and intervertebral disc degeneration: the results of tis-
Stanton H, Rogerson FM, East CJ, Golub SB, Lawlor KE, Meeker CT, sue immunohistochemistry and in vitro cell culture. J Orthop Res
Little CB, Last K, Farmer PJ, Campbell IK, Fourie AM, Fosang AJ 29:718725
Part II
Intervertebral Disc Disease: Pathogenesis
and Current Treatment Modalities
Epidemiology of Lumbar
Disc Degeneration 9
Yue Wang and Michele C. Batti

9.1 Overview ........................................................................ 139
The World Health Organization defines epidemiology as the
9.2 Historical Perspective ................................................... 139
study of the distribution and determinants of health-related
9.2.1 Disc Degeneration ........................................................... 139
9.2.2 Degenerative Disc Disease .............................................. 140 states or events (WHO 2012). This chapter on the epidemiol-
ogy of disc degeneration begins with a brief overview of the
9.3 Case Definition: Measuring Disc Degeneration.......... 142
early observations of disc degeneration and growing interest
9.4 Prevalence of Disc Degeneration.................................. 143 in the phenomenon, including the introduction and evolution
9.5 Etiological Factors (Environmental, of the currently ambiguous term degenerative disc disease.
Behavioral, and Constitutional) ................................... 145 Attention is given to case definition, as it is a core concept in
9.5.1 Age .................................................................................. 146 epidemiology and the study of occurrence rates. While
9.5.2 Occupational and Leisure Physical Demands
and Spinal Loading ......................................................... 147 definitions used for disc degeneration vary and depend
9.5.3 Driving and Whole-Body Vibration ................................ 148 greatly on the methods used to study or image the disc, the
9.5.4 Trauma ............................................................................ 149 occurrence rates of degenerative findings reported in this
9.5.5 Smoking .......................................................................... 149 chapter focus on population-based studies using MRI, which
9.5.6 Vertebral Endplate Morphology and Status .................... 149
9.5.7 Genetic Influences ........................................................... 150 may provide a reference for clinical observations. Finally,
we briefly discuss environmental, behavioral, and constitu-
9.6 Summary of Critical Concepts
Discussed in the Chapter .............................................. 153
tional factors associated with accelerated disc degeneration,
where there has been a recent dramatic shift in views.
References .................................................................................... 154
9.2 Historical Perspective
9.2.1 Disc Degeneration
The description of degenerative changes can be traced back

to 1824, when Wenzel first recorded pathological changes in
discs of the lumbar spine (Wenzel 1824). Yet, the interverte-
bral disc received little attention over the following 100
years. During this time, degenerative findings of the disc
were mentioned only sporadically by a few pathologists,
Y. Wang (*) including Luschka (Schmorl and Junghanns 1971) and
Department of Orthopedic Surgery, Rokitanski (1855). In the 1920s, however, spine pathology
The First Affiliated Hospital, College of Medicine, Zhejiang and disc degeneration research substantially advanced, as
University, Hangzhou, Peoples Republic of China
did research in other medical disciplines.
e-mail: wangyuespine@gmail.com
Among the pioneers exploring the spine during this
M.C. Batti
period, one of the most outstanding was Christian Georg
Faculty of Rehabilitation Medicine, University of Alberta,
8205 114St, Edmonton, AB T6G 2G4, Canada Schmorl (18611932). The academic career and pioneering
e-mail: mc.battie@ualberta.ca contributions of this legendary spine explorer were well

140 Y. Wang and M.C. Batti
Box 9.1 Herbert Junghanns

Herbert Junghanns (19021986)
Herbert Junghanns was a German surgeon and a respected
researcher who focused on the biomechanics of the spine.
He was the head of the Institute for Spinal Column
Research in Frankfurt, a president of the German Spine
Society, and the founder of the Association of the
Scientific Medical Societies in Germany, which consisted
of 16 scientific medical professional societies and is now
an umbrella organization for more than 150 German med-
ical societies. He also served as the editor of the German
book series The Spine in Research and Practice.
As one of Christian Georg Schmorls outstanding stu-
dents, Junghanns was responsible for bringing much of
Schmorls pioneering work on spinal pathology to light. It
is this contribution for which he is perhaps best known.
Junghanns produced a book compiling much of Schmorls
work on the spine, The Human Spine in Health and
Disease, that was first published in 1937. After Schmorls printed in North America in 1959, it became widely
tremendous body of research on the spine contained in the known around the world and regarded as a landmark of
4th edition of the book was translated into English and modern spine research.
described previously (Vernon-Roberts 1994) and are only 1998). With the accumulation of pathological knowledge
briefly noted here. Schmorl was a versatile German patholo- and clinical observations of the lumbar spine in the 1920s,
gist who creatively advanced a number of pathological surgeons began to suspect that a diseased intervertebral disc
techniques, including tissue processing and staining, photo- may cause sciatica. In 1929, Walter Dandy (18861946), a
micrography, and barium sulfate discography. While he neurosurgeon from The Johns Hopkins Hospital, accurately
substantially contributed to knowledge of the pathology described disc herniation in detail, including related surgical,
associated with many nonskeletal diseases, in 1926 he pathological, and clinical findings (Dandy 1929; Weinstein
focused his research on the spine and devoted the last 6 years and Burchiel 2009). Dandys report on two such cases was a
of his career exclusively to spine pathology. By advancing historical first. However, William Mixters (18801958)
pathological techniques used in spine autopsy, Schmorls 1934 paper on the relationship between disc herniation and
extensive work on spine pathology constituted the founda- sciatica (Mixter and Barr 1934) gained much more attention
tional knowledge of a number of spine conditions, including and is often credited with providing the foundation for the
disc degeneration (Vernon-Roberts 1994). For example, contemporary understanding of disc herniation and clinical
based on autopsy examinations of over 4,000 spines, Schmorl symptoms. His work is also frequently credited or admon-
systematically described the normal structure of the verte- ished for ushering in the so-called dynasty of the disc
brae and discs, including typical morphology of age-related (Parisien and Ball 1998; Hadler et al. 2007). The interverte-
disc degeneration, various types of annular tears, as well as bral disc was thus pushed to the forefront of spine practice
posterior and vertical disc protrusions. The latter pathology and research and remains there today as a primary target of
was named after him as Schmorls nodes (Schmorl and spine-related activities.
Junghanns 1971). Unfortunately, Schmorls work was
entirely written in German and did not obtain wide recogni-
tion until 1959 when his classic book was translated into 9.2.2 Degenerative Disc Disease
English and published in North America as The Human
Spine in Health and Disease. Degenerative disc disease as a concept and a term has
Schmorls work on the intervertebral disc basically com- also flourished over recent decades. The word disease
prised postmortem autopsy descriptions and did not link originated from desaise in the early fourteenth century, in
pathological findings of the disc to clinical symptoms. In which des means without, away and aise means ease.
fact, Schmorl believed that such degenerative findings were Clearly, disease is related to symptoms. On the other hand,
too common to be clinically important (Parisien and Ball degeneration basically means the process of declining
9 Epidemiology of Lumbar Disc Degeneration 141
Box 9.2 Alf Nachemson

Alf Nachemson (19312006)
Alf Nachemson has been described as the most influential
spinal researcher of the twentieth century. During more
than 45 years as a clinician (orthopaedic surgeon)-scien-
tist, his research on the spine had a broad span, ranging
from basic science to population health. Before his work
on macroscopic grading of disc degeneration and the
discs nutrient supply, noted in this chapter, he began his
research career studying biomechanics. It was this early
work in the 1950s that led him to intradiscal pressure
measurements, which were the first in vivo measurements
to directly determine loading conditions of the spine, the
research for which he may be best known.
Above all, Nachemson was a passionate and outspoken
advocate for research excellence and improved patient
care, particularly in his discipline of orthopaedic surgery.
He held innumerable related leadership positions in
Sweden and internationally. Among them was founding (now ISSLS Prize) in 1979, which is arguably the most
member of the International Society for the Study of the well-recognized award in spine research. His genuine
Lumbar Spine (ISSLS), which became the premier dedication to excellence, enthusiasm and humor, were an
research society on common spinal disorders. Through inspiration to all who had the good fortune to work with
the ISSLS, he initiated the prestigious Volvo Award him.
from a former state, especially through loss of structure and who were discharged from the army for back pain (Fletcher
function. Therefore, it is generally accepted that disc degen- 1947). As is the case today, the pathologies underlying most
eration is one of the many progressive changes of the human spine disorders remained unclear. Based on radiographs,
body primarily attributable to natural aging (Adams and Captain Fletcher used signs of advanced disc degeneration of
Roughley 2006), which surely is not a disease. The process the lumbar spine, such as disc narrowing, sclerosis, hypertro-
of disc degeneration, however, may be dramatically acceler- phic changes (likely osteophytes), and subluxation of the
ated by other etiological factors, such as severe trauma. This facet joints, to diagnose degenerative disc disease. Later he
view, if generalized to lesser physical insults and more com- concluded that the degenerative findings were highly related
mon loading of the spine through occupational and leisure to backward displacement of the lumbar vertebra. In retro-
activities, can lead in the notion that all disc degeneration is spect, Captain Fletcher was describing radiographic
a result of injury or pathology and thus a disease. While manifestations of degenerative retrolisthesis.
current research does not support the repetitive or cumulative Over the next two to three decades, the term degenera-
trauma model of disc degeneration, this was once the domi- tive disc disease was only occasionally used (Friedenberg
nant paradigm (Battie et al. 2004). and Miller 1963). When it was used, it was typically restricted
Also, as most degenerated discs are not symptomatic, the to patients with symptomatic spinal disorders for which the
use of degenerative disc disease or DDD as a synonym disc was suspected as important, such as in nerve root impair-
of disc degeneration may be misleading. Such use may have ment (Weiner and Macnab 1970), back and leg pain (Dilke
contributed to the occasional clinical practice of labeling et al. 1973; Macnab 1973), and disc herniation (Gertzbein
idiopathic low back pain, in the presence of disc degenera- et al. 1975). With growing interest in disc degeneration and
tion, but no other clear pathology, as degenerative disc dis- back pain problems, use of the term degenerative disc dis-
ease. Thus, the acronym of DDD has been translated ease also increased substantially in the 1970s. After 1975,
humorously as diagnostic deficiency disease. however, use of the term was no longer reserved for
To better understand the origin and evolution of the use of symptomatic disc-related disorders and began to be used as a
the term degenerative disc disease, we traced it back to the synonym of disc degeneration, as well. Among the dozen or
early period of spine research in the 1940s and Captain so early papers we came across referring to findings of disc
Gilbert Fletcher, a military physician in Pennsylvania. In degeneration as degenerative disc disease, a typical one
1947, he studied a cohort of 600 veterans of World War II was the well-cited paper published in the Journal of Bone
and Joint Surgery (Am) (Torgerson and Dotter 1976). In this degeneration. According to a review (Kettler and Wilke
paper, disc degeneration was judged as disc space narrowing 2006), there are at least 22 grading systems available to rate
on lateral radiographs; it was later called degenerative disc the degree of lumbar disc degeneration. These scales were
disease, a suggested cause of low back pain. classified into five groups based on materials and modalities
The term degenerative disc disease was widely used in used, including macroscopic anatomy, histology, plain radi-
medical and scientific literature in the 1980s, and its use has ography, discography, and magnetic resonance (MR) imag-
increased exponentially over subsequent years. Sometimes, ing. Due to technical limitations in detecting changes within
the term refers to painful disorders that are suspected or the intervertebral discs, no grading system used computer-
known to be disc related. In fact, currently, degenerative ized tomography.
disc disease is the most common reason for lumbar fusion Early descriptions of disc degeneration were based on
in the USA (Rajaee et al. 2012). Yet, gradually degenera- autopsy examinations of cadaveric spines. Findings from
tive disc disease has also become a commonly used syn- macroscopic and histological observations primarily included
onym of disc degeneration, and unfortunately, its specificity desiccation of the nucleus pulposus, fibrous fraying, clefts,
for painful disc-related disorders has been lost. Adding to and a loss of disc height. Although degenerative changes of
the confusion, this use of the term degenerative disc disease the lumbar intervertebral discs were described in detail by a
can also be seen in large-scale population-based epidemio- number of scholars from the 1930s to the 1950s (Coventry
logic studies of disc degeneration (Mok et al. 2010; et al. 1945b; Friberg and Hirsch 1948; Schmorl and Junghanns
Sambrook et al. 1999). 1971; Virgin 1951), no grading system was proposed. In a
Imprecise case definitions and the interchangeable use of 1960 study investigating the effects of disc degeneration on
distinctly different terms, such as disc degeneration, intradiscal pressure, Nachemson developed the first reliable
degenerative disc disease, and back pain, so common in the macroscopic grading system to rate disc degeneration
epidemiological literature, continue to cloud the interpreta- (Nachemson 1960). Grading disc degeneration using histo-
tion of available research (Battie et al. 2007a; Videman and logical examination was introduced as late as the 1990s
Battie 2012). Clearer definitions and more uniform use of (Gunzburg et al. 1992). However, such macroscopic and his-
terms would help facilitate accurate communication in medi- tological measurements of the disc cannot be applied to non-
cine and research, avoid unnecessary confusion, and allow surgical patients and, thus, are of limited clinical value.
clearer comparison of different studies. As the phenomenon Plain radiography was extensively used in imaging the
of disc degeneration is also lacking a standard definition, the spine by Schmorl in the 1920s, in an attempt to correlate
need to clarify concepts and terminology is of particular radiological features to pathological findings (Schmorl and
significance. Junghanns 1971). However, Schmorl realized that radio-
graphs showed only bony signs that are secondary to disc
degeneration, such as cortification of trabecular bone
9.3 Case Denition: Measuring (Schmorls nodes), endplate sclerosis, and narrowing of the
Disc Degeneration intervertebral disc space. Schmorl clearly pointed out that
only with considerable degrees of degeneration of the disc
No other human tissue or organ undergoes a process of such tissue, a decrease in the height of the disc space takes place.
profound degeneration or starts so early (usually in the sec- Thus, he concluded that a loss of disc height cannot detect
ond decade of age) as the intervertebral disc (Schmorl and disc degeneration in the early stage. Yet, after 1950, the
Junghanns 1971). Despite substantial clinical interest and increasing clinical needs of investigating back pain
extensive research, disc degeneration lacks a clear, standard demanded a noninvasive, feasible, and inexpensive modality
definition. This may be due, in part, to the wide variation in to evaluate the lumbar spine and disc. Consequently, mea-
methodologies used to interrogate the disc by researchers suring disc degeneration by judging disc height and verte-
and clinicians from many different disciplines. Thus, one bral endplate sclerosis on plain radiographs of the spine was
standard definition to address the pathogenesis and charac- introduced in 1952 (Kellgren and Lawrence 1952). Since
teristics of disc degeneration is likely unrealistic. Other then, plain radiography has served as a primary approach to
chapters in this book highlight what can be learned about the assess disc degeneration. Later, in the 1990s, osteophytes
disc and its degeneration from histological, biochemical, were added as another indicator of disc degeneration using
biomechanical, radiological, and other perspectives, includ- the radiographic grading system (Mimura et al. 1994; Lane
ing clinical practice. In this chapter, taking an epidemiologi- et al. 1993). Despite reasonable reliability (Kellgren and
cal perspective, we will focus on disc degeneration as defined Lawrence 1952), measuring disc degeneration from plain
primarily from imaging and, in particular, magnetic reso- radiography has inherent limitations, as previously
nance imaging. mentioned. In addition to depicting primarily nonspecific
Measurement and case definition are fundamental to any bony degenerative findings of the lumbar spine and the
study of occurrence, etiology, or clinical relevance of disc inability to detect early degenerative changes, measuring
disc height and surrounding osteophytes together may be develop summary scores of overall disc degeneration, while
questionable, as the measurement of one may bias the others use individual findings (e.g., signal or bulging) to
assessment of the other. answer specific questions. In addition to the findings
Another approach that gained popularity and was once correlating to one another, the qualitative assessment of one
regarded as a gold standard for evaluating disc degeneration finding may influence the assessment of another, such that
is discography. Different from plain radiography, discogra- summing of findings into a global score may exaggerate the
phy reflects the degree of morphological changes (annular degree of disc degeneration.
disruption) within the disc by injecting opaque contrast into MR imaging of the disc depends largely on changes of
the disc and observing its distribution. The first discography water content (Modic et al. 1984), which is the first step in
using barium sulfate was performed in cadaveric spines by the process of disc degeneration. Therefore, MRI is able to
Schmorl in the 1920s to demonstrate various types of annu- detect disc degeneration in very early stages, even before
lar tears (Vernon-Roberts 1994). In 1948, Lindblom and morphological degenerative changes appear. Correspondingly,
Hirsch applied discography clinically to patients with back decreased signal intensity on T2-weighted MR imaging
pain to image herniated lumbar discs (Hirsch 1948; Lindblom demonstrates the most marked change related to disc degen-
1948). However, what impressed them most was that discog- eration. There is evidence that digital signal measurements
raphy could provoke or replicate patients back pain or sci- may provide the best MR depiction of degeneration available
atica. The enthusiasm for discography quickly grew, although from standard imaging, as validated through association with
it was later found to have a strikingly high false-positive rate age (Videman et al. 2008). Yet, these measurements require
(Guyer and Ohnmeiss 2003). For a long period, discography specialized software and do not have immediate clinical use.
was used in clinics to diagnose so-called discogenic back Thus, disc signal changes continue to be based primarily on
pain, rather than to detect morphological changes of the visual impressions.
disc. There are a number of visual rating systems used for disc
The distribution of contrast in discography was used to degeneration (Kettler and Wilke 2006). The Pfirrmann
assess disc degeneration as late as the 1980s (Adams et al. classification of disc degeneration on MRI is one of the more
1986; Videman et al. 1987, 1990). Before MR imaging was commonly used grading protocols, with supporting evidence
generally used in clinical practice, traditional discography of reliability and validity (Pfirrmann et al. 2001). Based on
was considered the best approach to evaluate disc degenera- the structure of the disc, distinction of the nucleus and annu-
tion. It can differentiate successive morphological changes lus, signal intensity, and disc height, this system rates sever-
inside the disc (Adams et al. 1986), beyond that possible ity of disc degeneration using a five-grade ordinal scale. Due
with MRI (Schneiderman et al. 1987). Although discography in part to its simplicity, the grading scheme is currently pop-
may not be able to detect all annular lesions (e.g., those that ular in both spine practice and research. However, the grad-
do not connect to the nucleus), it was reported to be more ing is more or less based on a subjective judgment of many
sensitive in detecting annular tears than using MRI (Gunzburg perspectives of the disc, where it sometimes is difficult to
et al. 1992). Yet, due to the invasive nature and the possibility distinguish one stage of disc degeneration from another
of accelerating the progression of disc degeneration (Carragee (Haughton 2006). In addition, the ordinal nature of the scale
et al. 2009), the application of discography clinically has also limits its sensitivity to detecting changes in disc degen-
fallen out of favor and may eventually fade from surgical eration in longitudinal studies. Yet, its current widespread
practice. use will likely aid comparisons between studies, which is
The advent of MRI, and its widespread use since the needed to advance the field.
beginning of the 1990s, was a major advance for clinical and
research investigation of the disc. Capable of depicting disc
morphology in multiple planes noninvasively and without 9.4 Prevalence of Disc Degeneration
ionizing radiation, it quickly became the clinical imaging
modality of choice for the disc. MRI also allowed large-scale Due to the wide variations in definitions, methodologies,
epidemiological studies of general population samples to and samples used, the prevalence rates of disc degenera-
investigate the prevalence, etiological factors, and clinical tion reported in the scientific literature vary substantially
relevance of disc degeneration. (Battie et al. 2004). Imaging approaches, such as
Traditionally, findings of disc degeneration on MR radiography and discography, are currently less commonly
images have included decreased signal intensity (desicca- used to evaluate disc degeneration and thus will not be
tion), disc bulging (anterior and posterior), disc height fully discussed in this section. Instead, the focus will be
narrowing, annulus fissures, osteophytes, and endplate on studies using MRI, the preferred clinical imaging
irregularities. Which one should be used to indicate the modality, as results from such studies may provide a
severity of disc degeneration remains a matter of controversy. reference for clinical observations. Furthermore, while
Some investigators combine some or all of these findings and many studies of disc degeneration using MRI investigate
various highly selected populations, such as patients with and age groups. The most commonly studied degenerative
back or leg pain or workers of certain occupations, we finding of the lumbar disc, with the most consistent findings
restrict this chapter to studies of general population-based between studies, is reduced signal intensity, which report-
samples. In addition, in previous studies, some bony edly is already present to some degree in approximately half
findings, such as osteophytes, Schmorls nodes, and end- of adolescents (Table 9.1). Perhaps the findings that vary
plate irregularities, were included in the evaluation of disc most in prevalence between adolescence and later adulthood
degeneration. Although more or less associated with disc are annular tears and high-intensity zones, which are visual-
degeneration, they are not specific findings of the disc and, ized in less than 10 % of adolescents and young adults and
therefore, are not presented in this section. Instead, we which increases to 3050 % by middle adulthood. The prev-
will further discuss epidemiological studies of endplate alence of disc narrowing is relatively high in the young with
findings in a later section and consider their relationship nearly 40 % of adolescents having some degree of narrow-
to disc degeneration. ing, but there is only one study on which to base this obser-
A review of population-based studies of disc degeneration vation (Table 9.1).
using MRI reveals that the prevalence rates vary consider- As the prevalence of degenerative findings at different
ably between different degenerative findings, spinal levels, disc levels of the lumbar spine varies considerably, ideally
Table 9.1 Prevalence of disc degeneration findings on MR images in lumbar spines of general populations
Reduced Reduced
Sample signal disc Annular High-intensity
Author and year size Race Age Gender intensity height Bulging tears zone
Videman (1995)a 232 Finnish 49.3 (3569) 100 % M L1/2 41.6 % 9.3 % 58.6 % 11.6 %
L2/3 48.7 % 13.6 % 74.7 % 17.5 %
L3/4 57.7 % 24.1 % 81.9 % 27.4 %
L4/5 80.4 % 51.3 % 92.8 % 53.1 %
L5/S1 86.0 % 55.6 % 78.5 % 49.8 %
Kjaer (2005b)a 439 Danish 13.1 (1214) 46.7 % M L1/2 12 % 3% 0% 1% 0%
L2/3 8% 1% 0.3 % 1% 0%
L3/4 13 % 6% 3% 2% 0%
L4/5 31 % 31 % 10 % 2% 1%
L5/S1 50 % 17 % 12 % 3% 4%
57.6 % of spine 38 % 16.1 % 7.3 % 5%
Kjaer (2005a)a 412 Danish 40 48.2 % M 45.2 % 50.2 % 52.4 % 39.3 % 40.8 %
Takatalo (2009)b 558 Finnish 21 (2022) 58 % M L1/2 2%
L2/3 3%
L3/4 5%
L4/5 22 %
L5/S1 35 %
47.2 % of spine 24.9 % 9.1 % 6.8 %
Cheung (2009)c 1,043 Chinese (1855)
1829 42 % 27.3 % 3.4 %
3039 48 % 20.6 % 8.2 %
4049 70 % 27.3 % 16.1 %
50 88 % 43.1 % 29.0 %
Mok (2010)c 2,449 Chinese 40.4 10.9 40 % M L1/2 9.5 %
(9.788.4) L2/3 15.7 %
L3/4 27.2 %
L4/5 45.8 %
L5/S1 51.0 %
Samartzis (2011)c 83 Chinese 18.3 2.1 46 % M 34.9 % 22.9 % 3.6 %
(1320)
a
Signal intensity was judged using a 4-point scale, with 0 representing normal and 13 progressive signal loss
b
A modified Pfirrmann scale was used to evaluate disc degeneration, which takes both signal and disc height into consideration. Discs with a
Pfirrmann grade of 1 or 2 were classified as normal and discs with grades 3, 4, or 5 were defined as degenerated
c
In these three studies, disc degeneration was rated using a 4-point Schneiderman scale, which assesses signal intensity and disc narrowing
together
Fig. 9.1 Mean scores for specific Signal intensity Disc height narrowing
manifestations of disc
degeneration by disc level
T12/L1 T12/L1
(Adapted from the paper by
Batti et al. (2004), with L1/2 L1/2
permission)
L2/3 L2/3
L3/4 L3/4
L4/5 L4/5
L5/SI L5/SI
0 0.5 1 1.5 2 2.5 0 0.2 0.4 0.6 0.8 1

Mean score Mean score
Disc bulging Herniations
T12/L1 T12/L1
L1/2 L1/2
L2/3 L2/3
L3/4 L3/4
L4/5 L4/5
L5/SI L5/SI
0 0.5 1 1.5 0 0.05 0.1

Mean score Mean score
prevalence would be presented level by level. Most disc- associated tissue insults and injuries. Correspondingly,
specific degenerative findings on MRI, such as bulging, her- behavioral and environmental factors, such as occupa-
niation, disc space narrowing, and annular tears, are more tional materials handling, were viewed as the main causes
common and more severe at the L4/5 and L5/S1 discs than of disc degeneration. Not surprisingly, these influences
the upper lumbar discs (Fig. 9.1) (Videman et al. 1995). This were the focus of research and prevention strategies.
finding is also consistent between MR evaluation protocols Based on knowledge of that time, Frymoyer summarized
of disc degeneration. In general, degenerative findings in the the findings of epidemiological studies on degenerative
L1/2, L2/3, and L3/4 discs are more similar to each other, as disc disease by stating Among the factors associated
compared to those in the L4/5 and L5/S1 discs (Table 9.1). with its occurrence are age, gender, occupation, cigarette
Therefore, it may be advisable to, at least, divide the five smoking, and exposure to vehicular vibration. The contri-
lumbar intervertebral discs into upper (L1/2, L2/3, and L3/4 bution of other factors such as height, weight, and genet-
discs) and lower lumbar regions (L4/5 and L5/S1 discs) when ics is less certain (Frymoyer 1992). A decade later, a
investigating disc degeneration. There also appears to be a major shift in views was underway. Taking into account
clear difference in genetic influences between upper and new knowledge gained from the intervening decade, in
lower lumbar regions (Battie et al. 2008), further strengthen- 2002 Ala-Kokko concluded that Even though several
ing the case against using a summary score for degeneration environmental and constitutional risk factors have been
across the entire lumbar spine. implicated in this disease, their effects are relatively
minor, and recent family and twin studies have suggested
that sciatica, disc herniation and disc degeneration may be
9.5 Etiological Factors (Environmental, explained to a large degree by genetic factors (Ala-Kokko
Behavioral, and Constitutional) 2002). Disc degeneration is now considered a condition
that is largely genetically influenced, with environmental
Before the mid- to late 1990s, the dominant model of the factors, although elusive, also playing an important role
etiology of disc degeneration was one of repetitive load- (Battie et al. 2009). An overview of the main factors which
ing or wear and tear. This paradigm generally viewed disc have been of interest in the etiology of disc degeneration
degeneration as a consequence of repeated loading and follows.
Box 9.3 Denitions the prevalence of the state in the general population.
Variance is a measure of the variability in the data or mea- While observations of familial aggregation often lead to
surements of interest. Central tendency and variability are suspicions of genetic influences, familial aggregation can
two important statistical features for the distribution of also be due to shared cultural and environmental influences
observations. For a set of continuous scores (X) measured among family members.
from N samples, mean (x ) usually is used to indicate its Heritability is the proportion of population variance in a
central tendency and variance (S2) or standard deviation trait due to interindividual genetic variation or, in other
(SD) is used to indicate its variability. Variance is defined words, the proportion of phenotypic (observable trait)
as the average of the squared deviations from the mean, variance attributable to genetic effects.
which is calculated as:
Association refers to an occurrence relation, such that
2 two traits or states occur together either more or less
S2 =
(X - X) often than expected by chance. However, an observed,
n -1 statistically significant association between a risk indica-
tor and a disease does not necessarily infer a causal
Prevalence refers to the existence of a particular state relationship.
(e.g., disease) among members of a population, typically Causation infers a causal connection between a
at a given point in time, whereas prevalence rate is the determinant and disease or outcome phenomenon. To
proportion of the population in that state. judge whether an observed association represents a
Familial aggregation is the clustering of a trait or state in cause-effect relationship between exposure and disease,
family members beyond that expected by chance given comprehensive inferences or criteria are needed.
9.5.1 Age of age in disc degeneration. For example, mucoid degenera-

tion and mild clefts of the nucleus pulposus may present
With the possible exception of genetics, which will be before the age of two (Boos et al. 2002). In the second
discussed later, age is perhaps the factor most strongly asso- decade of life, nucleus pulposus clefts and radial tears are
ciated with disc degeneration. The clear relationship between often observed in the disc center, as are cartilage cracks
age and disc degeneration has been observed since the time (Boos et al. 2002). From 20 years old on, disc degenerative
of the pioneering pathologists using autopsy studies in the findings have been found to progressively increase in both
early twentieth century. Some classic autopsy studies using frequency and severity. However, there is considerable vari-
large samples clearly documented that the occurrence of disc ability in degenerative disc findings within age groups.
degeneration increased linearly with age. For example, in In general, use of MRI has provided a good opportunity to
a study from 1926 based on 1,000 consecutive autopsies, investigate the association between age and disc degenera-
Heine reported the prevalence rate of macroscopic disc tion. Cerebrospinal fluid-adjusted disc signal intensity mea-
degeneration (or so-called spondylitis deformans) increased surements, reflecting the water content of the nucleus
from 0 to 72 % from the age of 3970 years (Heine 1926). pulposus, have been found to decrease from early childhood
Schmorl and Coventry further confirmed that the occurrence through late adulthood (Videman et al. 1994). In a study of
rates of histological findings of disc degeneration, such as 116 men ranging in age from 35 to 70 years, the associations
disc cell death, nucleus clefts, annular tears, and cartilage of different degenerative findings on MRI were plotted in
endplate fissures, increase consistently with greater age relation to age. Disc signal intensity measurement was more
(Schmorl and Junghanns 1971; Coventry et al. 1945a). More highly associated with age than other MR findings (Fig. 9.2).
recently, Miller reported that the prevalence of macroscopic A further study reported that age could explain up to 31.7 %
disc degeneration increased from 16 % at 20 years of age to of the variation of signal-based digital measurements of disc
98 % at 70 years (Miller et al. 1988). Later, in a systematic degeneration, such as cerebrospinal fluid-adjusted disc sig-
investigation of age-related changes in lumbar intervertebral nal, in the upper lumbar intervertebral discs and 11.5 % in
discs, Boos et al. clearly depicted the association of increased the lower lumbar intervertebral discs (Videman et al. 2008).
prevalence rates of almost all histological degenerative Another advantage of signal intensity measurements acquired
findings in the disc with greater age in a sample spanning through standard MRI, as opposed to qualitative assess-
from fetal to 88 years (Boos et al. 2002). ments, is that they are continuous in nature, providing greater
On the other hand, some degenerative changes seen from sensitivity to change and facilitating statistical analyses.
histology can occur in early childhood, complicating the role However, such quantitative measurements can be influenced
Fig. 9.2 Mean scores for specific Signal intensity Disc height narrowing
manifestations of disc 3
3
degeneration by age
(Adapted from the paper by

2
2
Batti el al. (2004), with
Score

Score
permission)

1
1

0 0
35 45 55 65 35 45 55 65
Age
Age
Disc bulging Herniations

3 1

2

Score
Score

.5

1

0 0
35 45 55 65 35 45 55 65
Age
Age
by scanner variations and are currently only used for research or wear-and-tear model was that heavy work and greater
purposes. physical demands would lead to greater disc degeneration.
In epidemiological studies age is a more complicated Yet, epidemiological evidence of this has always been
factor than it may appear (Videman et al. 2008). Age is an conflicted.
intrinsic indicator of the natural process of aging, with a In early studies using radiography, it was observed that
dose-response effect of greater age associated with more disc degeneration, as judged from disc space narrowing and
degenerative changes. It reflects, however, more than natural endplate sclerosis, was more common in miners and manual
aging and can be a complex risk factor with respect to disc workers than in non-laborers (Lawrence 1955) and that the
degeneration and other outcomes, as it represents not only onset of disc degeneration was on average 10 years earlier in
natural aging but also associated factors. For example, age workers with heavy physical demands (Hult 1954).
also reflects cumulative exposure to a variety of known or Conversely, in a study of 15,160 back trouble patients, disc
unknown etiological factors that generally increase with the degeneration, as judged from disc space narrowing and
passage of time. In fact, many factors suspected of accelerat- osteophytes, was not found to be associated with heavy
ing disc degeneration, such as excessive mechanical loading physical work (Friberg and Hirsch 1949) nor was occupa-
or trauma to the spine, cannot be accurately measured, par- tional lifting as reported by others (Frymoyer et al. 1984).
ticularly when lifetime exposures are of interest. However, More recent studies have continued to produce mixed results,
the exposure to such factors, more or less, is associated with with some suggesting a negative effect of greater physical
age, with greater age generally associated with greater expo- demands (Seidler et al. 2009), while others fail to observe
sure to all risk factors. detrimental effects on the disc (Porter et al. 1989; Videman
et al. 2007). In fact, Porter put forward the hypothesis that
physical activity strengthens both the vertebrae and discs
9.5.2 Occupational and Leisure Physical (Porter et al. 1989). Results of a recent study by Videman
Demands and Spinal Loading et al. suggest that greater loading of the discs during routine
activities may even have some positive effects on the disc, as
Before genetic factors began to assume a leading role in seen through higher disc signal (hydration) (Videman et al.
disc degeneration in the middle to late 1990s (Battie et al. 2010). Also, although occupational loading history, with a
1995a, b; Sambrook et al. 1999), a wear-and-tear model of dose-response effect, was greater in patients with pain
disc degeneration prevailed. Various forms of physical attributed to disc pathology than in controls (Seidler et al.
demands and spinal loading conditions, particularly heavy 2009), most studies of the association between mechanical
material handling, were regarded as the main causes of disc loading and disc degeneration have failed to uncover a clear
degeneration. The natural assumption under such an injury relationship (Caplan et al. 1966; Riihimaki et al. 1990;
Sairanen et al. 1981), further questioning whether there is a associated with slightly less upper lumbar disc desiccation,
strong causal link between physical loading and disc suggesting that greater routine loading of the lumbar spine is
degeneration. not detrimental to the disc (Videman et al. 2010). The key,
Among the problems underlying the inconsistencies are perhaps, is the amount of loading and the manner that load-
measurement limitations of exposures and outcomes. ing is imposed on the spine in relation to tissue strength,
Epidemiological studies on mechanical loading and disc which likely varies substantially between individuals and
degeneration are particularly challenging, as lifetime physi- over the lifespan.
cal activities and associated loading cannot be measured It has been quite common for epidemiological studies of
accurately, which invariably has the effect of diluting the physical loading and disc degeneration to focus on occupa-
appearance of true associations. Moreover, standard tional exposures and ignore physical activities outside of
definitions of disc degeneration are also lacking. While most work. Considering that activities outside of paid employment
degenerative findings (e.g., signal loss, bulging, narrowing) usually occupy more of a persons time than work hours and
are correlated to some degree, they do not necessarily repre- that combined work and leisure physical loading may give a
sent the same phenomenon and important information may very different picture than occupational loading alone, this is
be lost by aggregating distinct findings into summary scores, no small oversight.
which is a common practice. In one of the more comprehensive epidemiological stud-
Uncontrolled confounding factors are another major limi- ies to date investigating the effects of lifetime exposures to
tation of most epidemiological studies. Exposure-discordant suspected risk factors, physical loading exposures explained
twin studies, which may provide for the best control of only 7 % of the variance in disc degeneration in the upper
known and unknown confounding factors while minimizing lumbar levels and 2 % in the lower lumbar levels, suggesting
extraneous variability related to the substantial genetic a modest role of commonly experienced occupational and
influences, provide a particularly strong research design to leisure activities in disc degeneration (Battie et al. 1995b).
investigate the effects of environmental exposures. Such Interestingly, it was exposures reported during leisure activi-
studies have been employed to help clarify the effects of a ties that accounted for the variance in lower lumbar disc
number of suspected risk factors in disc degeneration, which degeneration explained by physical loading exposures.
are highlighted below.
It is well known that musculoskeletal tissues, such as bone,
ligaments, cartilages, and muscles, are able to increase their 9.5.3 Driving and Whole-Body Vibration
physiological capacity and mechanical strength in response
to repeated physical activities and greater physical loading. Whole-body vibration (WBV) resulting from operating or
The adaptation to a changing mechanical environment is a riding in motorized vehicles is a special form of physical
rule in sport science and is typically the purpose of exercise loading that was long thought to be deleterious to the lumbar
and training. Yet, this rule is curiously often viewed as not spine and disc, in particular. According to an extensive
applicable to the intervertebral disc. Instead, mechanical review conducted in the early 1990s during the height of
loading or repeated material handling, either from occupa- research activity and interest in whole-body vibration and
tion or exercise, is viewed as a hazard that will accelerate lumbar disc degeneration, degenerative findings tended to be
disc degeneration. Herein is one of the great paradoxes in the more common in subjects with greater exposure (Kjellberg
relationship between physical loading and the lumbar spine et al. 1994). However, the authors of the review also noted
as compared to other musculoskeletal structures. uncontrolled confounding factors may have affected the
Although doubts about the cumulative loading or results in all studies, and the conclusions about the causal
wear-and-tear model of disc degeneration have long existed, role of WBV for the observed injuries and/or disorders there-
substantial evidence to challenge this paradigm was forth- fore become uncertain.
coming from exposure-discordant twin and heritability Since that time, findings have been conflicting and effects
studies. Despite extreme exposure, discordance in monozy- uncertain. An exposure-discordant twin study, perhaps the
gotic co-twins, very little variance in disc degeneration out- most well-controlled epidemiological study of the associa-
comes was explained by substantial, long-term occupational tion of lifetime exposure to motorized vehicles and related
or sport physical demands and lumbar loading conditions WBV, investigated 45 pairs of monozygotic twin siblings
(Batti et al. 2009). highly discordant for lifetime vehicular driving. No significant
Also, using monozygotic twins with similar lifestyle but differences, nor trends, of greater disc degeneration (e.g.,
substantially different body weights (average body weight disc signal loss, bulging, herniation, narrowing) were found
difference of 30 lb), long-term physical loading in the form in drivers as compared to their lesser exposed twin brothers
of additional body weight was not associated with increased (Battie et al. 2002).
disc degeneration. Instead, greater body weight within the Whether or not WBV related to motorized vehicles exac-
range studied, which did not include extreme obesity, was erbates symptoms from lumbar degenerative conditions or
affects other structures is unclear (Lings and Leboeuf-Yde may be one mechanism through which trauma or injury
2000), but these are different issues. Interestingly, vibration eventually leads to disc degeneration. Unfortunately, current
has also been pursued more recently as a treatment for low clinical imaging modalities are unlikely to be capable of
back pain, using various frequencies (del Pozo-Cruz et al. detecting endplate and annular lesions to the same degree,
2011; Rittweger et al. 2002). With respect to deleterious although the sensitivity of MRI to disc disruption may be
effects of exposures to motorized vehicles and related WBV improving.
during work and leisure on disc degeneration in humans, Another study of recalled back injuries related to sport,
findings have not been persuasive and, in any event, do not leisure, or work activities, and episodes of sudden onset low
suggest a major effect. Yet, after more than 50 years of back pain associated with an exceptional physical activity,
research on the topic, pleas for continued in-depth biome- used an exposure-discordant twin model. The study revealed
chanical studies of the effects of vibration on the disc, both that back injury based on recollection of such traumatic
good and bad, continue (Hill et al. 2009). This may signify incidents was not associated with a subsequent increase in
the desire and need for rigorous scientific evaluation to clar- disc degeneration, as judged from disc height narrowing or
ify effects, better taking into account frequency, amplitude, reduced signal intensity (Hancock et al. 2010). The inconsis-
and other aspects of vibration. On the other hand, it may also tencies between studies could be explained by high variabil-
reflect the absence of clear environmental risk factors with ity in the degree of trauma experienced and reported and that
substantial effects on which to focus etiological and inter- pain was the indicator of injury, which may not have been
vention research. associated with substantial or consequential tissue change to
the disc. Also, some painful injuries to the back may have
been forgotten, resulting in recall bias.
9.5.4 Trauma
Evidence from both experimental and clinical observations 9.5.5 Smoking

support that clear trauma is a risk factor for disc degenera-
tion. One animal model of disc degeneration involves surgi- Cigarette smoking is the only chemical exposure known to
cally producing a peripheral annular lesion, which quickly associate with lumbar disc degeneration. While it is com-
leads to degenerative changes in the disc (Osti et al. 1990). A monly accepted that smoking has deleterious effects on the
recent clinical study observed a similar, although less dra- disc and research has turned to understanding the mecha-
matic, causal association between injury to the annulus nisms involved (Vo et al. 2011), it is important to keep in
fibrosus and disc degeneration related to needle puncture in mind that the effects are likely to be very modest. Using
discography. Greater disc degeneration was observed in the identical twins grossly discordant for smoking exposure
non-degenerated or healthy discs following discography over (over 32 pack-years), the overall disc degeneration across the
a 10-year follow-up (Carragee et al. 2009). Direct trauma or entire lumbar spine was greater in smokers as compared to
injury to the disc may destroy the homeostasis of the disc, their nonsmoker siblings. Yet, the effect size was very small,
influence the metabolism of the disc cell, and alter biome- explaining less than 2 % of the variance in degeneration
chanics of the disc, leading to an accelerated degenerative scores (Battie et al. 1991).
process.
Trauma to tissues adjacent to the disc, including the ver-
tebral body and endplate, may also accelerate disc degenera- 9.5.6 Vertebral Endplate Morphology
tion. In animal models, endplate trauma directly induces disc and Status
cell apoptosis and promotes disc degeneration (Haschtmann
et al. 2008). Clinical long-term follow-up of young people Although substantial progress has been made in understand-
with previous vertebral compression fractures revealed ing the pathogenesis of disc degeneration, the pathological
greater amounts of disc degeneration than in healthy con- mechanisms or pathways leading to degeneration are far
trols, also supporting an association of endplate trauma with from explicit. Recent research suggests that the endplate, an
accelerated disc degeneration (Kerttula et al. 2000). interface tissue between the vertebral body and the
The question is: how much trauma or force is needed to intervertebral disc, may be the structure through which a
damage the human disc in vivo and accelerate disc degenera- number of etiological factors lead to disc degeneration
tion? A recent autopsy study found that a history of lumbar (Wang 2011).
injury, which referred to sudden onset back pain associated The endplate is essential to maintaining the structural
with a specific accident or unusual activity, was associated integrity and physiological function of the intervertebral disc
with the presence of lumbar endplate lesions, which were (Moore 2006). It is the physical shield preventing the nucleus
closely associated with more severe discographic disc degen- pulposus from escaping (Roberts et al. 1996) and the
eration (Wang et al. 2012a; Wang 2011) Endplate lesions mechanical interface facilitating load distribution in the
vertebra-disc complex (Brinckmann et al. 1983; Setton et al. suggested that various types of endplate lesions may have
1993; Ferguson and Steffen 2003). In addition, it is the gate- different pathological origins and varying degrees of associ-
way for metabolite transport between the vertebral marrow ation with adjacent disc degeneration (Wang et al. 2012a).
and intervertebral disc (Benneker et al. 2005; Nachemson In summary, the endplate is important to the intervertebral
et al. 1970; Urban et al. 2004). disc. While the influence of endplate morphometrics on disc
Given the multiple functions of the endplate, its morpho- degeneration may be modest, endplate lesions may play an
metrics are generally thought to associate with disc degen- important role in the pathogenesis of disc degeneration.
eration. Yet, due to variations in study materials and Endplate lesions are common findings in the spines of mid-
limitations of inadequate measurements, the association dle-aged men and appear to be grossly underestimated with
between endplate morphometrics and disc degeneration standard clinical imaging approaches. In addition to
remains largely unclear and controversial. In a series of stud- Schmorls nodes, there are various types of endplate lesions,
ies to investigate the role of endplate morphometrics in disc which may have different effects on adjacent discs. Thus, it
degeneration (Wang 2011), a large sample of lumbar end- will be important to differentiate other types of endplate
plates was extensively measured in vitro using accurate tech- lesions from Schmorls nodes to further understanding of
nologies, such as laser scanning and micro-CT. Among the their etiology and role in disc degeneration. Such research
endplate morphometrics studied, including thickness, con- would greatly benefit from advances in imaging technologies
cavity, circularity, size, and bone mineral density, only end- for use in vivo.
plate thickness (Wang et al. 2011) and size were found to
associate with the degenerative disc (Wang 2011). However,
the observed associations were relatively weak, suggesting 9.5.7 Genetic Inuences
their contribution may be relatively small.
A more interesting area in endplate research is endplate This section will not deal with the specific genes associated
pathology or endplate lesions, which drew substantial atten- with disc degeneration and the mechanisms through
tion in early spine research. Schmorls nodes, one type of which they act, as this is the subject of another chapter
endplate lesion in which the nucleus pulposus protrudes (see Chap. 10). Instead, the focus here will be on the broad
through the endplate into the vertebral body, have been stud- concept of genetic versus environmental influences on disc
ied for over a century. To date the understanding of Schmorls degeneration.
nodes remains limited. The reported prevalence of Schmorls
nodes varies dramatically from 9 to 75 % (Hamanishi et al. 9.5.7.1 Familial Aggregation
1994; Hilton et al. 1976; Saluja et al. 1986; Williams et al. Studies of genetic influences typically begin by determining
2007), and the association between Schmorls nodes and disc whether familial aggregation of a disease or condition exists.
degeneration remains controversial (Hilton et al. 1976; If the condition is found to aggregate in family members
Pfirrmann and Resnick 2001; Mok et al. 2010). beyond what would be expected by chance given its preva-
Using an archive of over 150 lumbar spines, endplate lence in the population, then distinguishing between genetic
pathologies were systematically investigated to determine and other sources of familial similarity becomes of interest
the prevalence rate and association with disc degeneration in better understanding the etiology of the condition.
(Wang et al. 2012b). Strikingly, the study revealed that nearly There have been a number of studies of familial aggrega-
half of the lumbar vertebral endplates examined had some tion of disc pathology associated with painful conditions,
sort of lesion, suggesting that the prevalence of endplate particularly disc herniation with radiculopathy (Batti and
lesions has been substantially underestimated in clinical Videman 2004). Yet, persons identified as having disc her-
studies using MRI. Moreover, based on morphological char- niation are typically those who access and receive spine sur-
acteristics, four types of endplate lesions were further gery for associated pain, and significant regional variations
identified, including Schmorls nodes, fracture, erosion, and in rates of spine surgery demonstrate that this measure is
calcification. The various types of endplate lesions have likely to be significantly influenced by other factors than
different distribution patterns. For example, Schmorls nodes purely disc pathology (Cherkin et al. 1994).
usually involved the endplate center and were more common There have been a number of case examples of strikingly
in the upper lumbar region, while calcification lesions tended similar histories within family members, both related to
to affect the whole endplate and were mainly located in the juvenile (Matsui et al. 1990; Varlotta et al. 1991) and adult
lower lumbar region (Wang 2011). Increasing age was found disc herniation (Scapinelli 1993; Varughese and Quartey
to associate with not only the presence of endplate lesions 1979; Scapinelli 1993). Such case reports demonstrate that
but also greater number and size of lesions, suggesting that familial aggregation occurs, but clarifying whether or not it
age and other associated factors may play a critical role in occurs more often than might be expected by chance requires
the pathogenesis of endplate lesions. A further analysis comparison to a control or reference group. When such
studies were conducted, they, too, revealed a higher inci- influences suggested the possibility of a strong genetic
dence of symptomatic disc herniation in the families of cases influence on disc degeneration. Yet, the finding that
of juvenile disc herniation than in those without (Matsui significantly less variance in degeneration was explained in
et al. 1990; Varlotta et al. 1991), with one study providing a the lower lumbar region, as compared to the upper lumbar
typical age-adjusted relative risk of herniation of 4.5 in fam- region, was puzzling. It was speculated that such a differ-
ily members of young patients who had undergone surgery ence could be due to variations in spinal anthropometrics
for disc herniation as compared to control family members between upper and lower lumbar regions interacting with
(Varlotta et al. 1991). Furthermore, when considering those 9 environmental conditions resulting in a disproportional
through 15, 16 through 19, and 20 through 25 years of age, effect at the lower lumbar levels (Battie et al. 1995b). Since
younger patients who underwent discectomy were found to that time, further study has suggested that while some
be significantly more likely to have a family history of back genetic effects appear to act on all lumbar discs, there are
disorders than those of older age groups (Nelson et al. 1972). also substantial genetic influences that are distinct for the
Such a finding is consistent with genetic epidemiological lit- upper and lower lumbar regions (Battie et al. 2007b).
erature of stronger genetic effects associated with earlier
onset. Case-control studies have also revealed familial aggre- 9.5.7.2 Heritability
gation of symptomatic lumbar disc herniation in adults Once familial aggregation in a trait is established, interest
(Matsui et al. 1998; Postacchini et al. 1988; Richardson et al. shifts to disentangling genetic and shared environmental
1997; Simmons et al. 1996). Collectively, these studies make influences. Classic twin studies comparing concordance of
a convincing case that disc herniation for which care is findings in monozygotic and dizygotic twin pairs provide a
sought in juveniles and adults is influenced by familial methodology for doing this. Classic twin studies are based,
factors. in part, on the assumption that monozygotic (identical) twin
Evidence of substantial familial aggregation of type, siblings have 100 % of their genes in common, whereas
extent, and location of degenerative changes was found in dizygotic (nonidentical) same-sex twins share 50 % of their
two earlier studies of monozygotic twins published in 1995 genes, on average, while co-twins of either zygosity share
(Battie et al. 1995a, b). The first study of 20 pairs of adult the same environment similarly. Such studies can provide
male monozygotic (identical) twins assessed the degree of overall estimates of genetic versus environmental influences.
similarities in degenerative findings of qualitatively assessed Specifically, heritability, which is the proportion of popula-
disc desiccation, narrowing, and bulging or herniation in the tion variance in a trait, in this case disc degeneration, that is
upper and lower lumbar regions (Battie et al. 1995a). due to interindividual genetic variation, can be estimated.
Concordance in findings between twins was compared to that Following the previous studies suggesting the strong pos-
which would have been expected by chance based on the sibility of substantial genetic influences, Sambrook et al.
prevalence of the findings among all 40 subjects. Depending conducted a classic twin study to examine heritability of disc
on the specific disc degeneration phenotype and lumbar level, degeneration using spine MRI for 86 pairs of monozygotic
015 % of the variance in disc degeneration was explained and 77 dizygotic pairs, primarily women from Australia and
by age and smoking, but the explained variance rose to the UK (Sambrook et al. 1999). A substantial genetic
2672 % when familial aggregation (the twin siblings influence was found. For an overall disc degeneration score
findings) was also considered. Less variance in disc degen- comprised of qualitative ratings of disc height, signal inten-
eration was explained at the lower than upper lumbar levels. sity, bulging and anterior osteophyte formation, heritability
Subsequently, an investigation of the lumbar MRI of 115 estimates were 74 % (95 % CI, 6481 %) for the lumbar
pairs of adult male identical twins revealed even more sub- spine and 73 % (95 % CI, 6480 %) for the cervical spine.
stantial familial aggregation in terms of signal intensity, Interestingly, when examining the components of the sum-
bulging, and height narrowing (Fig. 9.3) (Battie et al. 1995b). mary score, they reported that a genetic influence was not
In the multivariable analysis of the T12L4 region, history of apparent for signal intensity, which is perhaps the most fun-
physical loading explained 7 % of the variance in disc degen- damental sign of disc degeneration on MRI, and that most
eration scores among the 230 subjects, which rose to 16 % highly associated with aging.
with the addition of age and to 77 % with the addition of a A later classic twin study of 600 Finnish men (300 twin
variable representing familial aggregation. In the L45 and pairs) further confirmed a substantial genetic influence on
L5S1 region, history of physical loading explained only disc degeneration (Battie et al. 2008). In this case, the herita-
2 % of the variance in disc degeneration summary scores, bility estimates, ranging from 29 to 54 % (depending on the
which rose to 9 % with the addition of age and to 43 % with particular phenotype and lumbar level), were not as high as
consideration of familial aggregation. those reported in the earlier classic twin study of Australian
These studies revealing modest effects of suspected and UK women. The substantial but more moderate genetic
environmental risk factors and substantial familial influences found for disc signal, height narrowing, and
a b
c d
Fig. 9.3 Examples of monozygotic twin sibling MRIs demonstrating (b), 49 (c) and 61 years of age (d) (From the paper by Batti el al.
similarities in degenerative findings despite substantial lifelong differ- (2004), with permission)
ences in occupational physical loading. The twin siblings are 44 (a), 48
bulging in the study of Finnish twins were in line with what of distinct disc degeneration phenotypes, the investigation of
could have been expected from the earlier study of familial genetic influences by spinal level, and the use of multivariate
aggregation among male monozygotic twins. Among the genetic analyses to investigate shared genetic and environ-
strengths of the study of Finnish men were the examination mental pathways between phenotypes.
The variance in heritability estimates between the classic As might be expected from the strong genetic influence
twin studies may reflect differences in definitions of the on disc degeneration based on cross-sectional data, there is
phenotype of disc degeneration, sample characteristics, or also evidence of a strong familial and genetic influence
true population differences. The disc degeneration pheno- on the rate of progression of disc degeneration in adult-
type used in the study by Sambrook et al. was a combination hood (Videman et al. 2006; Williams et al. 2011). It was
of individual qualitative ratings of disc narrowing, signal, suggested by one longitudinal study, however, that the heri-
bulging, and osteophytes, which were summed across all lev- tability of progression of disc degeneration may be most
els of the lumbar spine. Yet, as disc narrowing and bulging apparent, at least in women, under the age of 50 years
were credited with primarily being responsible for the herita- (Williams et al. 2011).
bility estimates, it seems unlikely that the difference in the Several mechanisms have been suggested through which
estimates between studies would be mainly due to pheno- hereditary factors could influence disc degeneration and her-
types used. The one dramatic exception, however, relates to niation. Turnover of the discs biochemical and structural
the measurement of disc signal and related heritability esti- constituents could be genetically predetermined, in part,
mates between studies. Heritability estimates for disc signal leading to varying susceptibility to degenerative changes in
in Finnish men ranged from 30 to 54 %, depending on spinal some persons relative to others. This has attracted the most
level, whereas the study findings from the UK and Australia attention in gene studies. Genes may also act through deter-
suggested that disc signal was not heritable. We suspect that mining the size and shape of spinal structures, affecting the
this difference is largely due to the quality of the measures of spines mechanical properties and thus its vulnerability to
signal used. A qualitative score of 03 was used in the study internal and external forces. Given the far-reaching effects of
primarily of female twins, whereas the study of male twins genes, the heritability estimates of disc degeneration may
used a quantitative measure with greater precision, based on come from a myriad of relevant, genetically influenced struc-
the MRI digital data. tural, functional, and behavioral factors.
The most obvious difference in the classic twin studies In summary, the results of classic twin studies of disc
samples was that one was comprised primarily of middle- degeneration reveal a substantial role for genetics but also
aged women, and the other of men. Since heritability esti- emphasize that environmental factors are influential. Yet,
mates are dependent on genotype and exposure to influential the major suspected environmental risk factors of materials
environmental factors, it may be that the adult men had handling and physical demands, as indicated by occupation
greater exposure to environmental risk factors on average or regularly performed leisure and sports activities,
than the sample of women, such that the genetic contribution explain little of the variation seen in disc degeneration.
to total variance in disc degeneration was greater in women. Progress in identifying important environmental influences
Also, theoretically, sex-linked genetic differences may be will likely require new ideas and reconceptualizing
influencing disc degeneration. influential environmental factors, as well as advances in
One potentially important finding from the study of their measurement.
Finnish twins was the surprisingly modest shared genetic
and environmental influences between upper and lower lum-
bar regions. Although some genetic influences were shared, 9.6 Summary of Critical Concepts
a large portion appeared to be unique to either the upper or Discussed in the Chapter
lower lumbar discs. This finding has important implications
for studying the effects of genes, environmental exposures, Although degenerative disc disease and disc degenera-
and biomechanical factors and strongly suggests that the tion have been used as synonyms in the scientific litera-
upper and lower lumbar regions should be considered sepa- ture by many, disc degeneration does not equate to disc
rately. It was speculated that in addition to genetic effects on disease or necessarily result in back pain.
the discs structural components and their integrity, genes Currently there is no universal case definition for disc
may also influence lumbar morphology, neuromuscular degeneration. The presence and severity of disc degenera-
control, biomechanics, and other factors, which may have tion are typically judged from MRI based on decreased
different or disproportionate effects on degeneration in upper signal intensity, disc bulging, disc height narrowing,
and lower lumbar levels (Battie et al. 2008). Conversely, annular fissures, and osteophytes, either separately or
there was a high degree of shared genetic influences between combined.
the disc degeneration phenotypes of disc signal intensity, Disc degeneration begins in youth and is ubiquitous in
height narrowing, and bulging, suggesting that these pheno- adulthood, with a higher prevalence and severity in the
types may have a common etiopathogenesis and represent lower than upper lumbar region.
different aspects or stages of the same degenerative Previously, disc degeneration was thought to be a conse-
process. quence of mechanical wear and tear. Current research,
however, supports the view that disc degeneration is pri- bral discs: 2002 Volvo Award in basic science. Spine (Phila Pa
1976) 27:26312644
marily a result of genetic influences, with environmental Brinckmann P, Frobin W, Hierholzer E, Horst M (1983) Deformation of
factors also being important. the vertebral end-plate under axial loading of the spine. Spine (Phila
Age and age-associated factors also play a critical role in Pa 1976) 8:851856
the pathogenesis of disc degeneration. Caplan PS, Freedman LM, Connelly TP (1966) Degenerative joint
disease of the lumbar spine in coal minersa clinical and x-ray
The endplate may be the structure through which a num- study. Arthritis Rheum 9:693702
ber of etiological factors lead to disc degeneration. Carragee EJ, Don AS, Hurwitz EL, Cuellar JM, Carrino JA, Herzog R
(2009) 2009 ISSLS Prize Winner: does discography cause acceler-
Acknowledgements M.C. Batti receives support from the Canada ated progression of degeneration changes in the lumbar disc: a ten-
Research Chairs Program. year matched cohort study. Spine (Phila Pa 1976) 34:23382345
Cherkin DC, Deyo RA, Loeser JD, Bush T, Waddell G (1994) An inter-
national comparison of back surgery rates. Spine (Phila Pa 1976)
19:12011206
Cheung KM, Karppinen J, Chan D, Ho DW, Song YQ, Sham P,
References Cheah KS, Leong JC, Luk KD (2009) Prevalence and pattern of
lumbar magnetic resonance imaging changes in a population study
Adams MA, Roughley PJ (2006) What is intervertebral disc degenera- of one thousand forty-three individuals. Spine (Phila Pa 1976)
tion, and what causes it? Spine 31:21512161 34:934940
Adams MA, Dolan P, Hutton WC (1986) The stages of disc degenera- Coventry M, Ghormley R, Kernohan JW (1945a) The intervertebral
tion as revealed by discograms. J Bone Joint Surg Br 68:3641 disc: its microscopic anatomy and pathology. Part II. Changes in the
Ala-Kokko L (2002) Genetic risk factors for lumbar disc disease. Ann intervertebral disc concomitant with age. J Bone Joint Surg Am
Med 34:4247 27:233247
Batti M, Videman T (2004) Genetic transmission of common spinal Coventry M, Ghormley R, Kernohan JW (1945b) The intervertebral
disorders. In: Herkowitz H (ed) The lumbar spine. Lippincott disc: its microscopic anatomy and pathology. Part III, Pathological
Williams &Wilkins, Philadelphia, pp 98106 changes in the intervertebral disc. J Bone Joint Surg Am
Battie MC, Videman T, Gill K, Moneta GB, Nyman R, Kaprio J, 27:460474
Koskenvuo M (1991) 1991 Volvo Award in clinical sciences. Dandy W (1929) Loose cartilage from intervertebral disc simulating
Smoking and lumbar intervertebral disc degeneration: an MRI study tumor of the spinal cord. Arch Surg 660672
of identical twins. Spine (Phila Pa 1976) 16:10151021 del Pozo-Cruz B, Hernandez Mocholi MA, Adsuar JC, Parraca JA,
Battie MC, Haynor DR, Fisher LD, Gill K, Gibbons LE, Videman T Muro I, Gusi N (2011) Effects of whole body vibration therapy on
(1995a) Similarities in degenerative findings on magnetic resonance main outcome measures for chronic non-specific low back pain: a
images of the lumbar spines of identical twins. J Bone Joint Surg single-blind randomized controlled trial. J Rehabil Med 43:
Am 77:16621670 689694
Battie MC, Videman T, Gibbons LE, Fisher LD, Manninen H, Dilke TF, Burry HC, Grahame R (1973) Extradural corticosteroid
Gill K (1995b) 1995 Volvo Award in clinical sciences. Determinants injection in management of lumbar nerve root compression. Br Med
of lumbar disc degeneration. A study relating lifetime exposures J 2:635637
and magnetic resonance imaging findings in identical twins. Spine Ferguson SJ, Steffen T (2003) Biomechanics of the aging spine. Eur
(Phila Pa 1976) 20:26012612 Spine J 12:S97S103
Battie MC, Videman T, Gibbons LE, Manninen H, Gill K, Pope M, Fletcher GH (1947) Backward displacement of fifth lumbar vertebra in
Kaprio J (2002) Occupational driving and lumbar disc degenera- degenerative disc disease; the significance of the difference in
tion: a case-control study. Lancet 360:13691374 anteroposterior diameters of the fifth lumbar and first sacral
Battie MC, Videman T, Parent E (2004) Lumbar disc degeneration: epi- vertebrae. J Bone Joint Surg Am 29:10191026
demiology and genetic influences. Spine (Phila Pa 1976) 29: Friberg S, Hirsch C (1948) Anatomical studies on lumbar disc degen-
26792690 eration. Acta Orthop Scand 17:224230
Battie MC, Videman T, Carragee EJ (2007a) Re: Virtanen IM, Friberg S, Hirsch C (1949) Anatomical and clinical studies on lumbar
Karppinen J, Taimela S, et al. Occupational and genetic risk factors disc degeneration. Acta Orthop Scand 19:222242
associated with intervertebral disc disease. Spine 2007;32:112934. Friedenberg Z, Miller W (1963) Degenerative disc disease of the cervi-
Spine (Phila Pa 1976) 32:29262927 cal spine. J Bone Joint Surg Am 45:11711178
Battie MC, Videman T, Levalahti E, Gill K, Kaprio J (2007b) Heritability Frymoyer JW (1992) Lumbar disk disease: epidemiology. Instr Course
of low back pain and the role of disc degeneration. Pain Lect 41:217223
131:272280 Frymoyer JW, Newberg A, Pope MH, Wilder DG, Clements J,
Battie MC, Videman T, Levalahti E, Gill K, Kaprio J (2008) Genetic MacPherson B (1984) Spine radiographs in patients with low-back
and environmental effects on disc degeneration by phenotype and pain. An epidemiological study in men. J Bone Joint Surg Am
spinal level: a multivariate twin study. Spine (Phila Pa 1976) 66:10481055
33:28012808 Gertzbein SD, Tile M, Gross A, Falk R (1975) Autoimmunity in degen-
Batti MC, Videman T, Kaprio J, Gibbons LE, Gill K, Manninen H, erative disc disease of the lumbar spine. Orthop Clin North Am
Saarela J, Peltonen L (2009) The Twin Spine Study: contributions to 6:6773
a changing view of disc degeneration. Spine J 9:4759 Gunzburg R, Parkinson R, Moore R, Cantraine F, Hutton W,
Benneker LM, Heini PF, Alini M, Anderson SE, Ito K (2005) 2004 Vernon-Roberts B, Fraser R (1992) A cadaveric study comparing
Young Investigator Award Winner: vertebral endplate marrow con- discography, magnetic resonance imaging, histology, and mechani-
tact channel occlusions and intervertebral disc degeneration. Spine cal behavior of the human lumbar disc. Spine (Phila Pa 1976)
(Phila Pa 1976) 30:167173 17:417426
Boos N, Weissbach S, Rohrbach H, Weiler C, Spratt KF, Nerlich AG Guyer RD, Ohnmeiss DD (2003) Lumbar discography. Spine J 3:
(2002) Classification of age-related changes in lumbar interverte- 11S27S
Hadler NM, Tait RC, Chibnall JT (2007) Back pain in the workplace. Miller JA, Schmatz C, Schultz AB (1988) Lumbar disc degeneration:
JAMA 297:15941596 correlation with age, sex, and spine level in 600 autopsy specimens.
Hamanishi C, Kawabata T, Yosii T, Tanaka S (1994) Schmorls nodes Spine (Phila Pa 1976) 13:173178
on magnetic resonance imaging. Their incidence and clinical rele- Mimura M, Panjabi MM, Oxland TR, Crisco JJ, Yamamoto I, Vasavada
vance. Spine (Phila Pa 1976) 19:450453 A (1994) Disc degeneration affects the multidirectional flexibility of
Hancock MJ, Battie MC, Videman T, Gibbons L (2010) The role of the lumbar spine. Spine (Phila Pa 1976) 19:13711380
back injury or trauma in lumbar disc degeneration: an exposure- Mixter W, Barr J (1934) Rupture of the intervertebral disc with involve-
discordant twin study. Spine (Phila Pa 1976) 35:19251929 ment of the spinal canal. N Engl J Med 210215
Haschtmann D, Stoyanov JV, Gedet P, Ferguson SJ (2008) Vertebral Modic MT, Pavlicek W, Weinstein MA, Boumphrey F, Ngo F, Hardy R,
endplate trauma induces disc cell apoptosis and promotes organ Duchesneau PM (1984) Magnetic resonance imaging of interverte-
degeneration in vitro. Eur Spine J 17:289299 bral disk disease. Clinical and pulse sequence considerations.
Haughton V (2006) Imaging intervertebral disc degeneration. J Bone Radiology 152:103111
Joint Surg Am 88(Suppl 2):1520 Mok FP, Samartzis D, Karppinen J, Luk KD, Fong DY, Cheung KM
Heine J (1926) ber die Arthritis deformans. Virch Arch Pathol Anat (2010) ISSLS prize winner: Prevalence, determinants, and associa-
521663 tion of Schmorl nodes of the lumbar spine with disc degeneration: a
Hill TE, Desmoulin GT, Hunter CJ (2009) Is vibration truly an injuri- population-based study of 2449 individuals. Spine (Phila Pa 1976)
ous stimulus in the human spine? J Biomech 42:26312635 35:19441952
Hilton RC, Ball J, Benn RT (1976) Vertebral end-plate lesions Moore RJ (2006) The vertebral endplate: disc degeneration, disc
(Schmorls nodes) in the dorsolumbar spine. Ann Rheum Dis regeneration. Eur Spine J 15:S333S337
35:127132 Nachemson A (1960) Lumbar intradiscal pressure. Experimental stud-
Hirsch C (1948) On lumbar facetectomies. Acta Orthop Scand 17: ies on post-mortem material. Acta Orthop Scand Suppl 43:1104
240252 Nachemson A, Lewin T, Maroudas A, Freeman MA (1970) In vitro dif-
Hult L (1954) The Munkfors investigation; a study of the frequency and fusion of dye through the end-plates and the annulus fibrosus of
causes of the stiff neck-brachialgia and lumbago-sciatica syn- human lumbar inter-vertebral discs. Acta Orthop Scand 41:
dromes, as well as observations on certain signs and symptoms from 589607
the dorsal spine and the joints of the extremities in industrial and Nelson CL, Janecki CJ, Gildenberg PL, Sava G (1972) Disk protrusions
forest workers. Acta Orthop Scand Suppl 16:176 in the young. Clin Orthop Relat Res 88:142150
Kellgren JH, Lawrence JS (1952) Rheumatism in miners. II. X-ray Osti OL, Vernon-Roberts B, Fraser RD (1990) 1990 Volvo Award in
study. Br J Ind Med 9:197207 experimental studies. Anulus tears and intervertebral disc degenera-
Kerttula LI, Serlo WS, Tervonen OA, Paakko EL, Vanharanta HV tion. An experimental study using an animal model. Spine (Phila Pa
(2000) Post-traumatic findings of the spine after earlier vertebral 1976) 15:762767
fracture in young patients: clinical and MRI study. Spine (Phila Pa Parisien RC, Ball PA (1998) William Jason Mixter (1880-1958).
1976) 25:11041108 Ushering in the dynasty of the disc. Spine (Phila Pa 1976) 23:
Kettler A, Wilke HJ (2006) Review of existing grading systems for 23632366
cervical or lumbar disc and facet joint degeneration. Eur Spine Pfirrmann CW, Resnick D (2001) Schmorl nodes of the thoracic and
J 15:705718 lumbar spine: radiographic-pathologic study of prevalence, charac-
Kjaer P, Leboeuf-Yde C, Korsholm L, Sorensen JS, Bendix T (2005a) terization, and correlation with degenerative changes of 1,650 spinal
Magnetic resonance imaging and low back pain in adults: a diagnos- levels in 100 cadavers. Radiology 219:368374
tic imaging study of 40-year-old men and women. Spine 30: Pfirrmann CW, Metzdorf A, Zanetti M, Hodler J, Boos N (2001)
11731180 Magnetic resonance classification of lumbar intervertebral disc
Kjaer P, Leboeuf-Yde C, Sorensen JS, Bendix T (2005b) An epidemio- degeneration. Spine (Phila Pa 1976) 26:18731878
logic study of MRI and low back pain in 13-year-old children. Spine Porter RW, Adams MA, Hutton WC (1989) Physical activity and the
(Phila Pa 1976) 30:798806 strength of the lumbar spine. Spine (Phila Pa 1976) 14:201203
Kjellberg A, Wikstrm B-Q, Landstrm U (1994) Injuries and other Postacchini F, Lami R, Pugliese O (1988) Familial predisposition to
adverse effects of occupational exposure to whole-body vibration: a discogenic low-back pain. An epidemiologic and immunogenetic
review for criteria documentation. Arbete Och Hlsa vetenskaplig study. Spine (Phila Pa 1976) 13:14031406
skriftserie 41:174 Rajaee SS, Bae HW, Kanim LE, Delamarter RB (2012) Spinal fusion in
Lane NE, Nevitt MC, Genant HK, Hochberg MC (1993) Reliability of the United States: analysis of trends from 1998 to 2008. Spine (Phila
new indices of radiographic osteoarthritis of the hand and hip and Pa 1976) 37:6776
lumbar disc degeneration. J Rheumatol 20:19111918 Richardson JK, Chung T, Schultz JS, Hurvitz E (1997) A familial
Lawrence JS (1955) Rheumatism in coal miners. III. Occupational predisposition toward lumbar disc injury. Spine (Phila Pa 1976)
factors. Br J Ind Med 12:249261 22:14871492
Lindblom K (1948) Diagnostic puncture of intervertebral disks in Riihimaki H, Mattsson T, Zitting A, Wickstrom G, Hanninen K, Waris P
sciatica. Acta Orthop Scand 17:231239 (1990) Radiographically detectable degenerative changes of the
Lings S, Leboeuf-Yde C (2000) Whole-body vibration and low back lumbar spine among concrete reinforcement workers and house
pain: a systematic, critical review of the epidemiological literature painters. Spine (Phila Pa 1976) 15:114119
1992-1999. Int Arch Occup Environ Health 73:290297 Rittweger J, Just K, Kautzsch K, Reeg P, Felsenberg D (2002) Treatment
Macnab I (1973) Chapter 14. Pain and disability in degenerative disc of chronic lower back pain with lumbar extension and whole-body
disease. Clin Neurosurg 20:193196 vibration exercise: a randomized controlled trial. Spine (Phila Pa
Matsui H, Tsuji H, Terahata N (1990) Juvenile lumbar herniated 1976) 27:18291834
nucleus pulposus in monozygotic twins. Spine (Phila Pa 1976) Roberts S, Urban JP, Evans H, Eisenstein SM (1996) Transport proper-
15:12281230 ties of the human cartilage endplate in relation to its composition
Matsui H, Kanamori M, Ishihara H, Yudoh K, Naruse Y, Tsuji H (1998) and calcification. Spine (Phila Pa 1976) 21:415420
Familial predisposition for lumbar degenerative disc disease. Rokitansky C (1855) Lehrbuch der patholoqische Anatomie. Braumller,
A case-control study. Spine (Phila Pa 1976) 23:10291034 Wien
Sairanen E, Brushaber L, Kaskinen M (1981) Felling work, low-back relation to history of back pain, occupation, and physical loading.
pain and osteoarthritis. Scand J Work Environ Health 7:1830 Spine 15:728740
Saluja G, Fitzpatrick K, Bruce M, Cross J (1986) Schmorls nodes Videman T, Nummi P, Battie MC, Gill K (1994) Digital assessment of
(intravertebral herniations of intervertebral disc tissue) in two MRI for lumbar disc desiccation. A comparison of digital versus
historic British populations. J Anat 145:8796 subjective assessments and digital intensity profiles versus disco-
Samartzis D, Karppinen J, Mok F, Fong DY, Luk KD, Cheung KM gram and macroanatomic findings. Spine (Phila Pa 1976)
(2011) A population-based study of juvenile disc degeneration and 19:192198
its association with overweight and obesity, low back pain, and Videman T, Battie MC, Gill K, Manninen H, Gibbons LE, Fisher LD
diminished functional status. J Bone Joint Surg Am 93:662670 (1995) Magnetic resonance imaging findings and their relationships
Sambrook PN, MacGregor AJ, Spector TD (1999) Genetic influences in the thoracic and lumbar spine. Insights into the etiopathogenesis
on cervical and lumbar disc degeneration: a magnetic resonance of spinal degeneration. Spine (Phila Pa 1976) 20:928935
imaging study in twins. Arthritis Rheum 42:366372 Videman T, Battie MC, Ripatti S, Gill K, Manninen H, Kaprio J (2006)
Scapinelli R (1993) Lumbar disc herniation in eight siblings with a posi- Determinants of the progression in lumbar degeneration: a 5-year
tive family history for disc disease. Acta Orthop Belg 59:371376 follow-up study of adult male monozygotic twins. Spine (Phila Pa
Schmorl G, Junghanns H (1971) The human spine in health and disease. 1976) 31:671678
Grune and Stratton, New York Videman T, Levalahti E, Battie MC (2007) The effects of anthropomet-
Schneiderman G, Flannigan B, Kingston S, Thomas J, Dillin WH, rics, lifting strength, and physical activities in disc degeneration.
Watkins RG (1987) Magnetic resonance imaging in the diagnosis of Spine 32:14061413
disc degeneration: correlation with discography. Spine (Phila Pa Videman T, Gibbons LE, Battie MC (2008) Age- and pathology-specific
1976) 12:276281 measures of disc degeneration. Spine (Phila Pa 1976) 33:
Seidler A, Bergmann A, Jager M, Ellegast R, Ditchen D, Elsner G, 27812788
Grifka J, Haerting J, Hofmann F, Linhardt O, Luttmann A, Michaelis Videman T, Gibbons LE, Kaprio J, Battie MC (2010) Challenging the
M, Petereit-Haack G, Schumann B, Bolm-Audorff U (2009) cumulative injury model: positive effects of greater body mass on
Cumulative occupational lumbar load and lumbar disc disease disc degeneration. Spine J 10:2631
results of a German multi-center case-control study (EPILIFT). Virgin W (1951) Experimental investigations into the physical
BMC Musculoskelet Disord 10:48 properties of the intervertebral disc. J Bone Joint Surg Br 33-B:
Setton LA, Zhu W, Weidenbaum M, Ratcliffe A, Mow VC (1993) 607611
Compressive properties of the cartilaginous end-plate of the baboon Vo N, Wang D, Sowa G, Witt W, Ngo K, Coelho P, Bedison R, Byer B,
lumbar spine. J Orthop Res 11:228239 Studer R, Lee J, Di YP, Kang J (2011) Differential effects of nico-
Simmons ED Jr, Guntupalli M, Kowalski JM, Braun F, Seidel T (1996) tine and tobacco smoke condensate on human annulus fibrosus cell
Familial predisposition for degenerative disc disease. A case-control metabolism. J Orthop Res 29:15851591
study. Spine (Phila Pa 1976) 21:15271529 Wang Y (2011) The roles of vertebra and vertebral endplate in lumbar
Takatalo J, Karppinen J, Niinimaki J, Taimela S, Nayha S, Jarvelin MR, disc degeneration. Doctoral Thesis, University of Alberta
Kyllonen E, Tervonen O (2009) Prevalence of degenerative imaging Wang Y, Battie MC, Boyd SK, Videman T (2011) The osseous end-
findings in lumbar magnetic resonance imaging among young plates in lumbar vertebrae: thickness, bone mineral density and their
adults. Spine (Phila Pa 1976) 34:17161721 associations with age and disk degeneration. Bone 48:804809
Torgerson WR, Dotter WE (1976) Comparative roentgenographic study Wang Y, Videman T, Battie M (2012a) ISSLS prize winner: Lumbar
of the asymptomatic and symptomatic lumbar spine. J Bone Joint vertebral endplate lesions: associations with disc degeneration and
Surg Am 58:850853 back pain history. Spine (Phila Pa 1976) 37:14901496
Urban JP, Smith S, Fairbank JC (2004) Nutrition of the intervertebral Wang Y, Videman T, Battie MC (2012b) Lumbar vertebral endplate
disc. Spine (Phila Pa 1976) 29:27002709 lesions: prevalence, classification and association with age. Spine
Varlotta GP, Brown MD, Kelsey JL, Golden AL (1991) Familial predis- (Phila Pa 1976) 37:14321439
position for herniation of a lumbar disc in patients who are less than Weiner DS, Macnab I (1970) The use of chymopapain in degenerative
twenty-one years old. J Bone Joint Surg Am 73:124128 disc disease: a preliminary report. Can Med Assoc J 102:12521256
Varughese G, Quartey GR (1979) Familial lumbar spinal stenosis with Weinstein JS, Burchiel KJ (2009) DandyS disc. Neurosurgery 65:
acute disc herniations. Case reports of four brothers. J Neurosurg 201205
51:234236 Wenzel C (1824) Uber die Krankbeiten am Ruckgrate. Wesche,
Vernon-Roberts B (1994) Christian Georg Schmorl. Pioneer of spinal Bamberg
pathology and radiology. Spine (Phila Pa 1976) 19:27242727 WHO (2012) World Health Organization website. www.who.int/topics/
Videman T, Battie MC (2012) Commentary: back pain epidemiology- epidemiology/en/. Access date Feb 2012
the challenge of case definition and developing new ideas. Spine J Williams FM, Manek NJ, Sambrook PN, Spector TD, MacGregor AJ
12:7172 (2007) Schmorls nodes: common, highly heritable, and related to
Videman T, Malmivaara A, Mooney V (1987) The value of the axial lumbar disc disease. Arthritis Rheum 57:855860
view in assessing discograms. An experimental study with cadavers. Williams FM, Popham M, Sambrook PN, Jones AF, Spector TD,
Spine 12:299304 MacGregor AJ (2011) Progression of lumbar disc degeneration over
Videman T, Nurminen M, Troup JD (1990) 1990 Volvo Award in a decade: a heritability study. Ann Rheum Dis 70:12031207
clinical sciences. Lumbar spinal pathology in cadaveric material in
Genetic Basis of Intervertebral Disc
Degeneration 10
Anita Yee and Danny Chan
Contents 10.7 Technologies in Genetic Studies and Intervertebral

Disc Degeneration ........................................................... 168
10.1 Understanding Disease Mechanism
Through Genetic Studies ................................................ 157 10.8 The Moving Goals in Modern-Day
Genetic Studies ................................................................ 169
10.2 Applying Genetic Studies to Intervertebral
Disc Degeneration ........................................................... 158 10.9 Genetic Studies of Intervertebral Disc Degeneration:
10.2.1 Classical Approaches in Human Genetic Studies ............. 158 The Way Forward ........................................................... 171
10.2.2 Intervertebral Disc Degeneration Is
a Complex Trait................................................................. 159 10.10 Summary of Critical Concepts Discussed
10.2.3 Genetics as a Major Contributing Factor to in the Chapter .................................................................. 173
Intervertebral Disc Degeneration ...................................... 160 References ...................................................................................... 173
10.3 Phenotypic Parameters for Assessing the Genetics
of Intervertebral Disc Degeneration .............................. 160
10.3.1 What Defines Disc Degeneration? .................................... 160
10.3.2 Parameter Currently Used to Define
Disc Degeneration ............................................................. 160
10.3.3 Aging and Intervertebral Disc Degeneration .................... 161
10.1 Understanding Disease Mechanism
Through Genetic Studies
10.4 Moving from Phenotype Information to Identification
of Genes Causing or Contributing to Disease ............... 162
10.4.1 Candidate Genes ............................................................... 162 Genetics is an area of study that connects genes with func-
10.4.2 Family Linkage Analysis .................................................. 162 tions, variations with phenotypes, and, of most interest to
10.4.3 Case-Control Association Studies..................................... 162 scientists and clinicians, mutations with diseases.
10.5 Genes Associated with Intervertebral According to the Online Mendelian Inheritance in Man
Disc Degeneration ........................................................... 162 (OMIM), an online catalog for human genes and Mendelian
10.5.1 Extracellular Matrix Proteins ............................................ 163 disorders, over 3,000 known genotype-phenotype pairs
10.5.2 Matrix Metalloproteinases and Other Proteases ............... 166
10.5.3 Proinflammatory Cytokines .............................................. 166
have been recorded. A direct application of such knowl-
10.5.4 Genes Affecting Cell Function and Survival .................... 167 edge is diagnosis, of which prenatal testing and newborn
screening of rare conditions are common nowadays; ide-
10.6 Making Sense of Risk Factors and Intervertebral
Disc Degeneration ........................................................... 167 ally, this would be extended in future for disease preven-
tion or the development of personalized medicine. It is
clear that our understanding of disease mechanism of rare
skeletal disorders such as osteogenesis imperfecta (OI),
spondyloepiphyseal dysplasia (SED), achondroplasia, and
Ehlers-Danlos syndrome (EDS) has come directly from
knowing the genes involved, allowing detailed in vitro
and in vivo functional studies. While more difficult for
common disorders such as intervertebral disc degenera-
A. Yee D. Chan (*) tion, the same principle can be applied, bearing in mind
Department of Biochemistry, the effect size, the number of genes involved, and contri-
The University of Hong Kong,
butions from environment factors. These are the issues
3/F Laboratory Block, LKS Faculty of Medicine,
21 Sassoon Road, Pokfulam, Hong Kong that this chapter will address systematically, namely, to
e-mail: chand@hku.hk provide the basic concepts of what genetic studies can

158 A. Yee and D. Chan
achieve and with the benefit of hindsight from studies of

similar disorders such as osteoarthritis (OA) to generate a Dominance: The relationship of the two alleles at a
road map for studies of intervertebral disc degeneration. certain locus where one allele can mask the effect of
the other to demonstrate the phenotype.
Recessive: The relationship of the two alleles at a cer-
10.2 Applying Genetic Studies tain locus where the effect of one allele is masked by
to Intervertebral Disc Degeneration the other and thus two copies of such allele are required
to demonstrate the phenotype.
10.2.1 Classical Approaches in Human Linkage disequilibrium (LD): A situation when geno-
Genetic Studies types at two or more loci are not independent of each
other and that there would be a difference between the
Gregor Mendel, regarded as the Father of Genetics, laid the
observed and expected frequencies of certain
foundation of modern genetics with his work in the nineteenth
combinations of alleles (haplotype) in a population.
century, Versuche ber Pflanzen-Hybriden (Experiments in
Plant Hybridization). Restriction fragment length polymorphism (RFLP):
Genetic variation leading to the creation or destruction
10.2.1.1 Mendelian Genetics: Single Gene Defects of restriction enzyme site and therefore generating dif-
The genetic materials in human, except those in sex chromo- ferent DNA fragments after enzymatic digestion. The
somes, exist in pairs which are called alleles (Box 10.1). The variation can be identified simply by detecting the
two alleles at a certain position can be the same or different. In resultant fragment lengths using gel electrophoresis.
a simple scenario of genetic diseases, they represent normal and This is an inexpensive technique for genotyping.
disease-causing alleles. Disease-causing alleles (often regarded Variable number of tandem repeat (VNTR): A type of
as mutations) connect to phenotypes; however, whether or not genetic variation in which a short nucleotide sequence
one would develop the phenotype depends on the number of repeats consecutively. The number of repeats can be
disease-causing allele being carried and whether the allele exerts variable, and thus in a population, more than two
a dominant or recessive trait. For dominant alleles, a single copy alleles can be present for a particular locus.
is sufficient to show an effect or phenotype, while for recessive Single nucleotide polymorphism (SNP): The most
alleles, two copies are required. To pass on the alleles to next common type of genetic variation which involves only
generation, Mendelian genetics state that the parental alleles a single nucleotide difference. It occurs throughout the
would segregate, and an individual would obtain one allele from whole genome. Unlike VNTR, each SNP typically
each parent randomly. In dominant inheritance, one parent contains only two alleles.
would be carrying a disease-causing allele, and the offspring Genome-wide association study (GWAS): Examination
who receive this allele will have the phenotype (Fig. 10.1a). For of a huge number of genetic variants that are distributed
recessive inheritance, both parents must carry at least one copy all over the genome in multiple individuals to identify
of the disease-causing allele, and only the offspring who obtain whether any of them are associated with a trait. The
both copies would show the phenotype (Fig. 10.1b). focuses are generally placed on the common SNPs and
traits such as common diseases.
Box 10.1: Glossary Epigenome-wide association study (EWAS): A new
Allele: An alternative form of the genetic material at a concept similar to GWAS examination of a large
certain position (locus), which may or may not lead to number of epigenetic variations, such as DNA methy-
a difference in observable traits (phenotype). lation, over the genome in multiple individuals to iden-
Mutation: A change in the genetic material which can tify their associations with a trait.
involve a single nucleotide (point mutation) or a seg-
ment of genomic sequence (insertion, deletion, dupli-
cation, inversion, and translocation). The outcome of a There are a number of skeletal abnormalities that follow
mutation can be harmful leading to diseases, while Mendelian inheritance. Classic examples included osteogenesis
neutral or beneficial effects are also possible. imperfecta (OI), a fragile bone condition due to mutations in
Phenotype: An observable trait as a result of the geno- one of the two a-chains of the collagen I genes (COL1A1 and
typic composition but at the same time can be affected COL1A2), and spondyloepiphyseal dysplasia (SED), which is
by environmental factors and their interactions with characterized by dwarfism and shortened limbs, caused by
the genetic components. mutations in the gene encoding collagen II (COL2A1), affecting
the spine and epiphyses of long bones. These are rare disorders
10 Genetic Basis of Intervertebral Disc Degeneration 159
a
Gene A Environmental
factor 1
Disease
Dd dd
Gene B Environmental
factor 2
Dd Dd dd dd
Gene C
b
Fig. 10.2 Concept of complex diseases. Multiple genetic and environ-
mental factors can be involved in a complex disease. They can act inde-
pendently (e.g., gene A and environmental factor 1). Genetic factors can
interact (e.g., gene B and gene C), or they can interact with environmen-
Rr Rr tal factor (e.g., environmental factor 2 and gene C) contributing to the
disease
factor could modulate the genetic effect on the disease

RR Rr Rr rr (Mitchell and Yerges-Armstrong 2011). Similarly for OA,
multiple genes including COL2A1, SMAD3, growth dif-
ferentiation factor 5 (GDF5), and type II iodothyronine
Fig. 10.1 Concepts of Mendelian inheritance. (a) In dominant
deiodinase (DIO2) also showed association with age, gen-
inheritance, one parent carries the disease-causing allele D, with one
copy sufficient to show the phenotype. Offspring who inherited the D der, and body mass index as known risk factors (Valdes
allele from this parent would show the phenotype. Based on Mendelian and Spector 2011). In these cases, the allelic variations
genetics, the ratio of inheriting to not inheriting the D allele would be 1:1. associated with the diseases are common, and individuals
(b) In recessive inheritance, both parents have to be carriers of the dis-
with these alleles may only increase their risk of develop-
ease-causing allele r. Only the offspring who inherited two r alleles from
both parents would show the phenotype. The Mendelian ratio of non- ing the diseases. Therefore, in complex diseases, multiple
carrier to carrier to diseased offspring in this example would be 1:2:1 genetic susceptibility and environmental factors contribute
to disease outcome; they may act independently, or they
resulting from a single rare mutation which seriously affects the modify the effect of each other through gene-gene interac-
proper functioning of the gene products. On the other hand, tion and/or gene-environment interaction (Fig. 10.2). Such
common genetic variants in these genes can give rise to milder effects can be additive, synergistic, or suppressive. Thus,
conditions. For example, genetic variations in COL1A1 are complex diseases no longer present as simple Mendelian
associated with reduced bone density and osteoporosis (Grant inheritance, and intervertebral disc degeneration is another
et al. 1996), while the COL2A1 gene is linked to osteoarthritis such an example.
(OA) (Vikkula et al. 1993). These conditions, in general, are
caused by multiple factors rather than a single genetic mutation,
leading to the concept of complex diseases. 10.2.2 Intervertebral Disc Degeneration
Is a Complex Trait
10.2.1.2 Complex Diseases
Complex diseases are dependent on multiple factors, which Prior to the involvement of genetic studies, age, gender,
can be due to genes alone, or genes in combination with cigarette smoking, and mechanical loading related to occu-
environmental factors. Taking osteoporosis as an example, pation and sport activities have been reported to influence
apart from COL1A1, genes associated with bone loss include disc degeneration (this topic is discussed in more detail in
apolipoprotein E (APOE), vitamin D receptor (VDR), and Chap. 9). These findings place disc degeneration within
interleukin 6 (IL6); identified environmental risk factor are the category of a complex trait. However, there is no con-
age, gender, body mass index, smoking, and medications. sensus on the contribution of these environment factors to
Moreover, the association with APOE and COL1A1 being the degenerative process and hence this begs the question:
restricted to a subgroup of postmenopausal women not using what is the contribution of genetics to the intervertebral
hormone replacement therapy suggests that environmental disc disease?
10.2.3 Genetics as a Major Contributing Factor pairs of dizygotic twins from Australia and Britain for the con-
to Intervertebral Disc Degeneration tributions of genetic and environmental effects on disc degen-
eration (Sambrook et al. 1999). Using an overall degenerative
To decipher the involvement of genetics in intervertebral score (summing the grades of disc height, bulge, osteophytosis,
disc degeneration, a classical heritability test (Box 10.2) was and signal intensity), heritability was estimated to be 74 % at the
undertaken to identify patterns of familial aggregation. Two lumbar spine, after adjusting for age, weight, height, smoking,
such analyses were conducted in 1995. In one study, 20 pairs occupation, and exercise (Sambrook et al. 1999). This finding
of Finnish male identical twins were assessed based on mag- indicated a high degree of genetic involvement in intervertebral
netic resonance imaging (MRI) for the degree of similari- disc degeneration that lead to the first publication in 1997 on
ties in degenerative findings, including disc desiccation, disc an associated gene, vitamin D receptor (VDR) (Videman et al.
height narrowing, and disc bulging or herniation. The authors 1998), and subsequently other genes.
concluded that while smoking and age accounted for at most
15 % of the variability in the degenerative findings, 2672 %
of the variability was explained when the co-twin status was 10.3 Phenotypic Parameters for Assessing
included (Battie et al. 1995a). In another study, 115 pairs of the Genetics of Intervertebral Disc
male identical twins with discordant exposures to suspected Degeneration
environmental risks were assessed for degenerative changes of
the spine and symptomology. For changes in the upper lumbar 10.3.1 What Denes Disc Degeneration?
region, occupational physical loading explained only 7 % of the
variability in the summary score for degeneration; this increased Knowing that genetic components are involved in interverte-
to 16 % when age was included and to 77 % when twin status bral disc degeneration, it raises the question what genes or
was added. In the lower lumbar region, leisure physical loading genetic variations contribute to the disease? To address this
explained only 2 % of the variability in the degeneration scores; question, it is first critical to define disc degeneration. Having
this increased to 9 % with the addition of age as a factor and to a precise phenotype definition is essential for genetic studies
43 % with the addition of the twin status (Battie et al. 1995b). in that the phenotype should be a distinguishable trait and
These studies provide quantitative evidence for the existence of preferably quantifiable (Wagner and Zhang 2011). Disc
familial aggregation and potential genetic influences in interver- degeneration is a continuous process throughout life: predict-
tebral disc degeneration. These findings were later consolidated able macroscopic changes include disruption of the highly
in a study comparing 86 pairs of monozygotic twins and 77 organized lamellae structure, formation of tears and fissures
in the annulus fibrosus, leakage of the nucleus pulposus
through the fissures leading to disc bulge and herniation,
damage of the end plate, an overall reduction in disc height,
Box 10.2: Heritability Test dehydration, increased cell death, and cell cluster formation.
Heritability refers to the proportion of phenotypic vari- From a biochemical perspective, the major biochemical alter-
ance attributed to genetic variance. It involves observations include the loss and breakdown of collagens and aggre-
ing and statistically analyzing the patterns of can, resulting in reduced tensile strength and impaired
phenotypes with varying levels of genetic or environ- hydration, respectively. Thus, while ample observations have
mental background in close kin, such as parent- been made to assess the alterations in disc degeneration, its
offspring, siblings, and twins (Visscher et al. 2008). definition remains indistinct, and there is a poor understand-
More often, heritability can be estimated from identi- ing of the relationship between these changes and their repre-
cal twins grown up in separate environment (adoption sentation in the degenerative process.
studies). Under this situation, the genotype would be
identical but the environmental factors vary, and thus
the effects of the two factors can be separated. However, 10.3.2 Parameter Currently Used
such twins are not easy to gather, and the age when to Dene Disc Degeneration
they are separated may affect the findings. Another
design would be to compare monozygotic and dizy- Theoretically, all the features mentioned above, and in other
gotic twin pairs. In this scenario, the twin pairs would chapters of this book, could be used as measurable parameters
experience similar environmental factors, and thus for the study of disc degeneration; however, only a few can be
comparing the phenotypic concordance of these two assessed in vivo. Current assessments of disc degeneration
types of twins allows an estimation of the impact of rely on radiography and MRI. Radiographs display the density
genetic factors and composition of an object based on the proportion of X-rays
being absorbed, providing information such as disc height and
osteophyte formation. In contrast, MRI detects the rotating may display a different stage of degeneration. This raises the
magnetic field of photons, an obvious advantage for the inter- question, should the grades of various levels be combined and
vertebral disc as the nucleus pulposus is a hydrated tissue. a summation score obtained representative of the degree of
Therefore, MRI can provide information on the hydration sta- proneness of an individual to disc degeneration? Alternatively,
tus, bulging and herniation, as well as irregularities of the end should each level be treated separately and reported as multi-
plates. In particular, the MRI images of a disc which is bright level disc degeneration? Moreover, should other parameters
and bloated represent a highly hydrated tissue, and with the such as disc bulging, herniation, and end-plate irregularities
progressive loss of water, the image becomes dark. As indi- from the MRI images be treated independently or considered
cated in Chap. 12, MRI can also assess proteoglycan contents as a part of the degenerative process? This is still uncertain,
in the disc, although improvements in accuracy are still again due to the limited understanding of disc degeneration.
required (Benneker et al. 2005; Marinelli et al. 2009). The variability of phenotypic parameters that define the
There are multiple ways to evaluate the degenerative degenerative status of the intervertebral disc confounds
changes in the intervertebral disc. The first one would be to genetic studies. This is especially true for many situations,
provide a definitive notation on the presence or absence of where replication and meta-analysis studies are required to
disc degeneration. This method is simple but the shortcom- substantiate the research findings, particularly, when the
ing is loss of information on the progressive changes during effect size is small. Replication provides credibility to initial
the degenerative process. Another method is to classify the findings of association, while meta-analysis increases the
severity of degeneration based on a scaling system. This is statistical power to detect associated genes. Both require
the most widely used method, and a number of classification comparable phenotypes among studies to produce meaning-
systems have been developed. For example, Kellgren scale ful results (Chanock et al. 2007; Nakaoka and Inoue 2009).
(Kellgren et al. 1963) combines the features of osteophytes Therefore, a unified phenotypic definition for disc degenera-
and joint space narrowing based on radiographs and sum- tion is an absolute requirement for the successful identification
marizes using a four-point score, ranging from score of 1 of the involved genetic components.
indicating no or very small osteophytes to score of 4 rep-
resenting large osteophytes and pronounced disc space nar-
rowing; Schneidermans grading (Schneiderman et al. 1987) 10.3.3 Aging and Intervertebral
focuses on the signal intensity of the nucleus pulposus from Disc Degeneration
MRI images and classifies them into four grades, with the
lowest grade indicating hyperintense signal to the highest Disc degeneration is part of the normal aging process. With
grade illustrating hypointense signal with disc space narrow- age, an increasing prevalence and severity of disc degenera-
ing; Pfirrmanns classification (Pfirrmann et al. 2001) also tion have been observed (Cheung et al. 2009). However, an
utilizes the MRI images to evaluate the homogeneity of disc understanding of what constitutes normal progression
structure, signal intensity, distinction of nucleus pulposus remains unclear, as many factors can participate and modify
and annulus fibrosus, as well as disc height. The information the degenerative process. Genetics is one of the components
is converted into five grades, the lowest grade being homo- that can alter this normality by accelerating or decelerating
geneous disc structure, hyperintense bright signal intensity, the degenerative process. This change is reviewed elsewhere
and normal disc height and the highest grade being inho- in the book and includes changes in cell function such as
mogeneous disc structure, hypointense black signal, loss of expression levels of genes, the stability of mRNA transcripts
distinction between nucleus pulposus and annulus fibrosus, or proteins, or the binding affinity of proteins interacting
as well as collapsed disc space. While these grading systems partners, caused by the genetic variations in or near partici-
maintain information regarding the severity of disc degener- pating genes. In establishing a cohort for genetic study of
ation and provide semiquantitative evaluation of the degen- disc degeneration, the effect of age must be taken into con-
erative status, interpretation of the MRI images is subjective sideration in terms of subject recruitment and data analyses.
and thus requires multiple experienced observers to perform Statistically, if sufficient population information is available,
the grading. The third method for assessing disc degenera- then appropriate adjustments can be established. An example
tion is by computational evaluation of the absolute signal would be a sliding window method, in which the degenera-
intensity values of the MRI images (Videman et al. 1994). tive score was logarithmic transformed to reduce skewness
This approach can circumvent the potential bias arising from and standardized to a mean of 0 and a variance of 1 in each
observers judgment, but the data generated would likely be decade of age of the samples (Virtanen et al. 2007). The idea
composed of a spectrum of values complicating subsequent behind such adjustment is to identify the normality within
data analyses. the age band of a certain cohort, such that samples showing
The methods mentioned above evaluate the status of a accelerated or retarded degeneration can be highlighted
single disc, whereas there are multiple disc levels and each during analysis.
10.4 Moving from Phenotype Information 10.4.2 Family Linkage Analysis

to Identication of Genes Causing
or Contributing to Disease Unlike the candidate gene approach, classical linkage analy-
sis relies less on biological knowledge of the disease being
10.4.1 Candidate Genes studied; rather, it utilizes the principle of co-segregation
where the disease-causing alleles, together with nearby mark-
This approach utilizes the biological knowledge and etiol- ers, are passed on to the next generation within a family.
ogy of the disease to identify genes of interest and to These markers are linked to the disease because recombina-
determine correlations between variants within the genes tion events are infrequent within a short stretch of DNA. They
and the phenotype. The phenotype can provide clues to can be in the form of restriction fragment length polymor-
potential candidate genes, while information gathered phisms (RFLP) of a candidate gene, where specific patterns
from expression studies and animal models can enhance of DNA fragments are linked to disease. If there is no clue to
the selection process (Tabor et al. 2002). This method has any candidate genes, whole-genome scans using microsatel-
been successful in identifying the genetic components of lite markers can be used to locate disease susceptible loci,
several skeletal diseases. For example, although over 90 % while further genotyping or sequencing can identify the caus-
of the osteogenesis imperfecta cases are due to mutations ative variants. More recently, initial mapping has used high-
in COL1A1 and COL1A2 genes, the causes of the remain- density whole-genome SNP arrays and was successful in
ing cases remained unknown. Until recently, candidate locating several new OI loci for downstream investigation
gene approach helped identify mutations in genes involved (Alanay et al. 2010; Lapunzina et al. 2010; Martinez-Glez
in posttranslational modifications required for collagen et al. 2011). In general, detection is most successful with large
folding and stability. One of the modifying complexes, families with multiple generations of affected members evi-
prolyl 3-hydroxylase 1 (LEPRE1) coupled with cartilage- dencing a clearly defined phenotype. As such, the traditional
associated protein (CRTAP) and cyclophilin B (PPIB), family linkage approach is more commonly used to identify
converts specific proline to 3-hydroxyproline, which is affected genetic regions of rare disorders.
important in the formation of collagen triple helix. Also,
since Crtap-null mice demonstrated skeletal abnormali-
ties resembling a subtype of OI (Morello et al. 2006), 10.4.3 Case-Control Association Studies
mutation analysis of CRTAP in this subtype of OI (Barnes
et al. 2006) identified mutations associated with the dis- This is the method of choice for common diseases with
ease. Similarly, mutations in LEPRE1 and PPIB were also a complex trait. It does not rely on family data, but rather on
detected in OI patients (Cabral et al. 2007; van Dijk et al. a large cohort selected from the general population or recruit-
2009). ment of patients with a defined phenotype. The allele or
Apart from rare diseases, candidate gene approach also genotype frequencies between the case and control groups
aids the identification of risk factors in common skeletal dis- are compared, with statistical tests being applied to objec-
eases, with osteoarthritis being an excellent example. There tively discern if there are differences (Daly 2009). Chi-square
have been many genes reported to be associated with OA test is commonly used for this purpose, while Fishers exact
(Valdes and Spector 2011), among which GDF5 is one of the test should be used when sample sizes are small. On the other
promising candidates as it survived testing in multiple popu- hand, regression analysis can be used if it is presumed that
lations and showed high significance in a meta-analysis there is varying degree of effects, such as an additive influence
(Miyamoto et al. 2007). It was selected as a candidate since between different genotypes. A disadvantage of this approach
it is involved in joint formation (Francis-West et al. 1999), is that if the effect size is small, statistical significance is usu-
and thus, a common variant in GDF5 could play a role in the ally achieved only with very large cohorts (in the thousands)
disease (Miyamoto et al. 2007). Similarly, genes related to or if meta-analysis is performed using different cohort col-
cartilage homeostasis have been studied as candidates, lected from multiple centers, regional and international.
including extracellular matrix components, matrix degrading
enzymes, and genes involved in TGF-b and Wnt signaling
pathways, inflammation, and apoptosis. A careful selection 10.5 Genes Associated with Intervertebral
of candidates with high biological relevance to the diseases Disc Degeneration
of interest is often the key to success. Moreover, the results
and downstream functional studies can provide new under- While different strategies can be used to find genetic risk fac-
standing in connecting molecular mechanisms with the dis- tors for intervertebral disc degeneration, the predominant
ease state. approach is through case-control studies and the selection of
Decorin Collagen II/XI
a b
COMP Impaired cell
Syndecan Collagen IX maintenance
Proinflammatory
cytokines
Matrillin
Collagen VI
Integrin
Fibronectin
MMPs, aggrecanases...
Aggrecan
Hyaluronan
Fig. 10.3 Current concepts of candidate gene selection. Candidate genetic candidates. (b) During degeneration, increased matrix degrada-
gene selection is closely related to the biology of normal disc and tion is caused by enzymes such as MMPs and aggrecanases, while the
involvement in disc degeneration. (a) The extracellular matrix mole- presence of proinflammatory cytokines can accelerate such a process;
cules form an integral part of the disc, providing mechanical strength, moreover, cell maintenance is affected. These are the current under-
water absorption properties, as well as proper environment for cells. standing of the degeneration mechanisms, in which the involved mole-
Their functional importance and high abundance make them the major cules serve as important candidates to be studied
appropriate candidate genes. These genes are usually chosen cells that are responsible for producing and maintaining the
based on our current knowledge of the biology of the disc in extracellular matrix; thus, genes that affect cell function and
health and disease, and the logic behind the selection is cen- survival have also been studied for their associations with
tered on the integrity of the disc tissue (Fig. 10.3). Collagens disc degeneration. In this section, specific genes will be high-
and aggrecan, together with other structural proteins, form lighted, enhancing our understanding of their functions in the
the basis of the extracellular matrix which is an integral part disc as well as their roles in the degenerative process.
of the disc. Not only are they the most abundant molecules,
their role in providing tensile strength and osmotic pressure is
essential for proper disc function. Therefore, it is not surpris- 10.5.1 Extracellular Matrix Proteins
ing that the extracellular matrix genes were prime candidates
for study. In addition, tissue homeostasis a balance between Aggrecan is the major proteoglycan in the disc, responsible
synthesis and degradation would be equally important, and for water absorption and retention through its highly negative
there are a number of studies focusing on catabolic genes, charged chondroitin sulfate (CS) side chains. A variable
namely, the aggrecanases and matrix metalloproteinases number of tandem repeat (VNTR), located at exon 12
(MMPs). In fact, a number of these genes are found to have encoding the CS attachment domain, have been identified,
elevated expression levels as well as increased enzymatic which dictate the length of the aggrecan core protein as well
activity during disc degeneration (Roberts et al. 2000; Le as the potential number of attached CS chains (Doege et al.
Maitre et al. 2004). Such changes can be triggered by the 1997). The first study showed an association between this
proinflammatory cytokines, such as interleukins (Millward- VNTR of aggrecan with lumbar disc degeneration in a group
Sadler et al. 2009), which are also expressed at elevated levels of 64 young Japanese women aged 2029 (Kawaguchi et al.
in the degenerative process (Le Maitre et al. 2005, 2007), 1999). There was an overrepresentation of the shorter alleles
resulting in overall accelerated degradation and posing an of 18 and 21 repeats in multilevel disc degeneration as well as
adverse effect on matrix integrity. Ultimately, it is the disc an association with the severity of disc degeneration
(Kawaguchi et al. 1999). The relevance of 21 repeat alleles in a small study of 40 young Greek army recruits. Here, the
with multilevel disc degeneration was replicated in a study of TT genotype was not found in any of the controls, but 33.3 %
Koreans; the age of the subjects analyzed was limited to the among those with disc degeneration (Tilkeridis et al. 2005).
fourth decade or younger (Kim et al. 2011). On the other Association studies also showed that the Sp1 polymorphism
hand, a Turkish study showed that the shorter alleles (1325 was involved in hip osteoarthritis (Lian et al. 2005), myocar-
repeats) were overrepresented in young individuals with disc dial infarction (Speer et al. 2006), cruciate ligament ruptures
degeneration (Eser et al. 2011). The shorter alleles were also (Khoschnau et al. 2008), and stress urinary incontinence
associated with disc herniation in Han Chinese (21 and 25 (Skorupski et al. 2006). It remains uncertain how a single
repeats) (Cong et al. 2010a, b) and Turkish populations (13 polymorphism can participate in all of these dissimilar con-
25 repeats) (Eser et al. 2011). Together, these studies indicate ditions, but it is generally agreed that increased binding
that individuals carrying the shorter alleles are more suscep- affinity of Sp1 for the T allele leads to an increase in mRNA
tible to the severe forms of disc degeneration. Interestingly, and protein levels. This in turn changes the ratio of a1(I) to
the study in Han Chinese also showed a 4.5-fold increase in a2(I) chains, resulting in an altered biomechanical proper-
risk for symptomatic disc degeneration with smoking, ties (Mann et al. 2001).
suggesting an interaction between this polymorphism and Collagen IX is minor collagen coating the surface of col-
smoking (Cong et al. 2010a, b). It is possible that smoking or lagen II/XI fibrils and is thought to interact with other matrix
nicotine metabolites can alter the metabolism of aggrecan, molecules to maintain matrix integrity (see Chap. 5). Its
with a greater effect on the shorter forms of this molecule. importance has been demonstrated in mice carrying truncated
However, similar association tests have not been conducted form of a1(IX) (Kimura et al. 1996) or inactivated Col9a1
with other populations, and the author also pointed out the gene (Boyd et al. 2008), both of which showed accelerated
study was limited by the small sample size (132 cases) (Cong disc degeneration when compared with their normal counter-
et al. 2010a, b). While most studies support the influence of parts. It is a heterotrimer encoded by three different genes,
the shorter alleles, 25 repeats or less, with disc degeneration, namely, COL9A1, COL9A2, and COL9A3. Analysis of the
a study conducted with a Finnish cohort of 132 males, 4045 COL9A2 gene identified two consecutive SNP polymorphisms
years of age, showed that only the allele with 26 repeats was (rs12077871 and rs2228564) in exon 19, leading to a substitu-
significantly associated among individuals with a dark nucleus tion of tryptophan for either glutamine or arginine at residue
pulposus (Solovieva et al. 2007). The differences could be 326 (Annunen et al. 1999). Interestingly, this tryptophan allele
due to ethnic variations, as well as the relatively small sample was present in 6 out of 157 Finnish individuals with disc
size. Nevertheless, as these studies provided supportive evi- degeneration and associated sciatica, but none among the 174
dence for aggrecan as a genetic risk factor for disc degenera- controls (Annunen et al. 1999). Since this polymorphism
tion, it would be worthwhile to perform a meta-analysis. involves a tryptophan (Trp) substitution in the a2(IX) chain, it
Collagen I is the predominant collagen in the annulus is named the Trp2 allele. An age-stratified analysis of a group
fibrosus and responsible for the highly organized lamellae of 804 Southern Chinese (4049 years) showed a 2.4-fold
structure that provides the tissue with its tensile strength. It is increase in risk of developing disc degeneration and end-plate
encoded by two genes, COL1A1 and COL1A2. An SNP herniation in those carrying the Trp2 allele (Jim et al. 2005).
located at the binding site of the transcription factor Sp1 Moreover, affected Trp2 individuals were found to have more
(rs1800012) in the first intron of COL1A1 was initially severe disc degeneration (Jim et al. 2005). This was confirmed
identified and found to be associated with reduced bone min- in a study of 84 Japanese patients (under 40 years) undergoing
eral density and an increased risk of fracture and turnover lumbar disc nucleotomy (Higashino et al. 2007). However,
(Grant et al. 1996; Garnero et al. 1998; Uitterlinden et al. another larger-scale Japanese study of 658 controls and 470
1998). Since it was suggested that there was an inverse rela- cases could not replicate the findings (Seki et al. 2006).
tionship between osteoporosis and disc degeneration, the In addition to the Trp2 allele, a similar arginine to tryp-
Sp1 binding site polymorphism in the COL1A1 gene was tophan substitution at residue 103 was detected in exon 5
investigated in a Dutch cohort of 517 individuals who are at of the COL9A3 gene (rs61734651) (Paassilta et al. 2001).
least 65 years of age (Pluijm et al. 2004). Individuals with This Trp3 allele was found in a Finnish cohort of patients at
the TT genotype were shown to have a 3.6 times higher risk a significantly higher frequency of 12.2 % among 171 cases
of disc degeneration when compared with those having GT when compared to 4.7 % among the 321 controls, with a three-
or GG genotype, after adjusting for age, sex, and body weight fold increase in the risk of disc degeneration (Paassilta et al.
(Pluijm et al. 2004). It should be noted that disc degeneration 2001). A higher proportion of the Trp3 allele was also detected
in this study was defined by the presence of osteophytes and among people with disc degeneration than controls in a Greek
articular joint space narrowing based on radiographs, as study, but the difference did not reach statistical significance
opposed to the reduced signal intensity seen on MRI images. (Kales et al. 2004). There was also the possibility that the
Despite this limitation, the Sp1 polymorphism was replicated Trp3 allele might act synergistically with persistent obesity, an
interaction that would serve to increase the risk of disc degen- Japanese population (Seki et al. 2005). This SNP is non-syn-
eration (Solovieva et al. 2002). In addition, interaction of Trp3 onymous, resulting in an amino acid substitution of isoleu-
with another polymorphism, interleukin-1b (C(3954)-T), was cine at residue 395 to threonine. The authors demonstrated
examined. It was noted that carrying this allele in the absence in vitro that CILP could inhibit TGF-b-induced transcrip-
of the interleukin-1b 3954T allele resulted in an increase in the tion, and the effect of inhibition was increased in the pres-
risk of a dark nucleus pulposus (Solovieva et al. 2006). These ence of C allelic product (Seki et al. 2005). This SNP might
results indicated the potential of the Trp3 allele interacting with also exhibit a differential gender effect since a small study in
environmental and genetic factors modifying its effect. There Japanese collegiate athletes showed an association in male,
were other studies testing the association of Trp2 and Trp3 but not female athletes (Min et al. 2010). On the other hand,
alleles, but the Trp2 allele was absent in Greek (Kales et al. a recent study in a Finnish cohort found an association only
2004) and only present at a low frequency of 1.2 % in German in females (Kelempisioti et al. 2011), while replication in
(Wrocklage et al. 2000), while Trp3 was absent in Southern Chinese population cohort failed (Virtanen et al. 2007), sug-
Chinese (Jim et al. 2005) and Japanese (Higashino et al. 2007), gesting that other factors such as ethnicity, sex, and environ-
suggesting substantial ethnic variations. ment may be modulating the effect of this polymorphism.
An association for the COL9A1 gene with disc degenera- Asporin (ASPN) belongs to the family of small leucine-
tion was also suggested in a study of 25 selected candidate rich proteoglycans (SLRP), members of which include deco-
genes in a cohort of 588 Finnish male monozygotic and dizy- rin and biglycan (see Chap. 4) (Lorenzo et al. 2001). It
gotic twins (Videman et al. 2009). A particular SNP contains a stretch of aspartic acid residues at the amino-
(rs696990) located at the 5 of the gene was associated with terminal, which are variable in number (Lorenzo et al. 2001).
the disc signal intensity, which survived an empirical thresh- While a repeat of 13 aspartic acid residues (D13) was the
old value for global significance. most common variant, a repeat of 14 residues (D14) was
The Trp2 and Trp3 allelic products are incorporated into the overrepresented among patients with osteoarthritis in a
cross-linked fibrillar network of developing human cartilage Japanese population (Kizawa et al. 2005). Since both osteoar-
(Matsui et al. 2003). Thus, any pathological consequences are thritis and disc degeneration are degenerative cartilage dis-
likely to be long term and cause alterations in the tissues eases, ASPN was hypothesized as a candidate gene for disc
mechanical properties (Matsui et al. 2003). Indeed, among degeneration (Song et al. 2008a). The aspartic acid repeats
human disc samples carrying the Trp2 allele, altered or com- were tested in Chinese and Japanese cohorts of 1,055 and
prised swelling pressure and compressive modulus was detected 1,353 individuals, respectively, and the D14 allele was over-
(Aladin et al. 2007). Although, the precise mechanism is still represented in the case groups. Meta-analysis showed that
unclear, a hypothesis is the bulk side chain of Trp residue may individuals carrying this allele were at higher risk with an
interfere with the interaction of collagen IX with other matrix overall odds ratio of 1.7 (Song et al. 2008a). Increased
molecules including collagen II. This would destabilize the asporin expression was detected among degenerated discs
matrix and thus affect the biomechanical properties of the disc. (Song et al. 2008a; Gruber et al. 2009); in addition, the D14
Type XI collagen forms the core of collagen II/XI/XI allelic product showed a greater suppression of TGF-b-
fibrils and functions to control the diameter of the fibril. It is mediated transcription than that of the D13 allelic product
composed of three a-chains encoded by the COL11A1, in vitro (Kizawa et al. 2005). As discussed elsewhere in this
COL11A2, and COL2A1 genes (see Chap. 5). An initial book, TGF-b signaling regulates the expression of key matrix
screening of these genes identified an SNP c.4603C T proteins such as collagen II and aggrecan. Indeed, in vitro
(rs1676486) in the coding region of COL11A1 to be associ- studies support the notion that the risk allele would have a
ated with lumbar disc herniation among Japanese (Mio et al. negative effect on the synthesis of matrix molecules (Kizawa
2007). The association was confirmed by testing all the et al. 2005).
sequence variations in COL11A1, among which this SNP Matrilins are a four-member family of multi-subunit
remained the most significant, as well as by increasing the extracellular matrix proteins (see Chap. 5). They function as
cohort size to a total of 823 cases and 838 controls. It was adaptors in the assembly of various matrix molecules includ-
suggested that this SNP affected mRNA stability since the ing aggrecan, collagen type II, SLRPs, and cartilage oligo-
expression level of the T allele is significantly lower than that meric matrix protein (COMP) (Klatt et al. 2011). In a
of the C allele (Mio et al. 2007). Rotterdam study, it was found that a non-synonymous SNP
Cartilage intermediate layer protein (CILP) is a non-col- (rs28939676) of matrilin-3, in which threonine at position
lagenous matrix component initially found in the middle 303 was substituted by methionine, was associated with disc
zone of human articular cartilage (Lorenzo et al. 1998). An degeneration at two or more levels based on radiographs,
SNP 1184T C (rs2073711) in the encoded region of the leading to an increased risk of 2.9 among individuals carry-
gene was identified to be significantly associated with disc ing the T allele (Min et al. 2006). On the other hand, this
degeneration in a cohort of 467 cases and 654 controls of a association was not confirmed in another cohort study of
sibling pairs of Dutch origin (Min et al. 2006). While the found to disrupt an Sp1 binding site that can lead to a reduction
effect of this polymorphism during disc degeneration in transcriptional activity (Price et al. 2001).
remained unknown, it was postulated that the polymorphism A polymorphic site for either six continuous adenosines
weakened the role of matrilin-3 in stabilizing the extracellu- (6A) or five adenosines (5A) in the promoter region of MMP3
lar matrix molecules (Min et al. 2006). Related to this finding, was assessed for an association with disc degeneration in 49
recent studies demonstrated that in primary human chondro- elderly Japanese (Takahashi et al. 2001). When compared with
cytes, the presence of matrilin-3 can induce the expression a individuals having only the 6A allele, the 5A allele was associ-
number of proinflammatory cytokines including IL6, IL8, ated with disc degeneration as well as severity of degeneration;
and TNFa, as well as degradative enzymes MMP1, MMP3, however, this association was not detected in a group of 54
and MMP13 (Klatt et al. 2009). These molecules are known young subjects (Takahashi et al. 2001). For MMP9, in addition
to be triggered during disc degeneration, suggesting a possi- to the interacting THBS2 SNP described earlier, an SNP at the
ble relationship between matrilin-3 and inflammation. promoter region, 1562C T, was shown to be associated
Thrombospondin-2 (THBS2) belongs to a family of extra- with disc degeneration in a Northern Chinese cohort of 408
cellular matrix proteins, thrombospondins, with multi-subunit. cases and 451 controls. Those individuals carrying the TT or
The protein is thought to be involved in cell-matrix interac- CT genotype had an increased risk of developing and having
tion, antiangiogenesis, regulation of collagen fibrillogenesis, more severe grades of disc degeneration (Sun et al. 2009). One
and the effective levels of MMP2 and MMP9 (Bornstein explanation for this observation was that the T allele may have
et al. 2004). An intronic SNP, IVS10-8C T (rs9406328) in a higher transcriptional activity than the C allele (Sun et al.
THBS2 was shown to be significantly associated with lumbar 2009). These studies, while limited, have focused on polymor-
disc herniation in two independent Japanese populations, phic sites within the promoter region of MMP genes that func-
composed of 847 cases and 896 controls (Hirose et al. 2008). tionally influence transcriptional activity, thus providing a
The TT genotype caused a significantly higher rate of exon functional support for the genetic findings. It is likely that vari-
11 skipping when compared to the CC genotype, with a ations at promoters or cis-regulatory elements of genes could
reduction of MMP2 and MMP9 binding. These data sug- have a more significant role in disc degeneration, and hence, a
gested that THBS2 could be involved in regulating MMP more thorough investigation of the genome is warranted.
expression in the disc, which in turn participate in the patho-
genesis of disc herniation. Moreover, the authors also
identified a combinatorial effect with a non-synonymous 10.5.3 Proinammatory Cytokines
SNP (rs17576) in MMP9, with an odds ratio of 3.03, indicat-
ing a potential gene-gene interaction (Hirose et al. 2008). Genes within the interleukin 1 (IL1) cluster are among the
proinflammatory cytokines that have been associated with
disc degeneration; from a functional viewpoint, elevated
10.5.2 Matrix Metalloproteinases expression could enhance the expression of MMPs, an initia-
and Other Proteases tion factor in the degenerative process. A number of common
variants in the IL1 gene cluster were evaluated in a group of
Matrix metalloproteinases (MMPs) are a large protein family 133 Finnish males, and IL1a 889C T and IL1b 3954C T
with a wide spectrum of substrates including extracellular were found to be associated with disc bulges, with odds ratio
matrix components. Based on their specificity, they can be of 2.4 and 1.9 for individuals carrying the respective T alleles
broadly categorized into subgroups such as collagenases (Solovieva et al. 2004). Interestingly, a genetic interaction
(MMP1, MMP8, and MMP13), gelatinases (MMP2 and between the IL1b 3954T and the COL9A3 Trp3 allele was
MMP9), and stromelysins (MMP3) (Goupille et al. 1998). identified; its implication however is not clear.
Details of these enzymes are presented in Chap. 8. A polymor- Two SNPs at the promoter region, 592C A and
phism for G insertion/deletion (G/D) at the 1607 promoter 1082G A, of interleukin 10 (IL10) were studied for an
region of MMP1 was assessed in a Southern Chinese cohort of association in a Chinese cohort of 320 cases and 269 controls
691 individuals. The identified deletion (D) allele was found to (Lin et al. 2011). The AA genotypes for both SNPs were
be significantly associated with disc degeneration; this was par- found to be more frequent in the cases. An effect on tran-
ticularly evident among individuals aged 40 or above (Song scriptional level is suggested as lower IL10 mRNA levels
et al. 2008b). In another Chinese cohort of 162 cases and 318 were detected in disc samples from individuals with the AA
controls, an SNP in the promoter region of MMP2, 1306C T, genotypes, when compared to the CC genotype at 592 and
was shown to be associated with disc degeneration, with the GG genotype at 1082 (Lin et al. 2011).
CC genotype being more prevalent in cases of severe degenera- A number of other specific polymorphic sites in genes of the
tion (Dong et al. 2007). This polymorphism was previously inflammatory pathways have been studied in relationship to disc
degeneration. These included a GGG haplotype from three development and maintenance of the intervertebral disc, it
respective SNPs rs1800797, rs1800796, and rs1800795 of inter- thus served as a candidate gene. In a study of two indepen-
leukin 6 (IL6) (Kelempisioti et al. 2011), the SNP rs1420100 in dent Japanese populations with a total of 862 cases and 896
intron 2 and rs917997 downstream of interleukin 18 receptor controls, out of the 68 selected tag SNPs in SKT, two SNPs
accessory protein (IL18RAP) (Videman et al. 2009), and a syn- located at intron 2 (rs16924573 and 2285592) were found to
onymous substitution of Val at position 102 (rs5277) in cycloox- be significantly associated with disc herniation. Moreover,
ygenase 2 (COX2) (Valdes et al. 2005). How these variations rs16924573 was further replicated in Finnish cohorts
may affect the respective genes in disc degeneration is not clear. (Karasugi et al. 2009; Kelempisioti et al. 2011). The function
of SKT is unknown and further studies are needed to evalu-
ate the effect of this SNP on disc herniation.
10.5.4 Genes Affecting Cell Function As mentioned earlier, growth and differentiation factor 5
and Survival (GDF5) is required for joint formation (Francis-West et al.
1999) (see Chap. 3). An SNP (rs143383) located at the 5
Vitamin D receptor (VDR) is an intracellular receptor for untranslated region of the gene was identified as a key risk
1,25-dihydroxyvitamin D3, the metabolite involved in min- factor for OA, and in vitro studies suggested that the risk allele
eral metabolism. VDR was the first gene to be found T led to a reduced promoter activity (Miyamoto et al. 2007). In
associated with disc degeneration in a Finnish population five independent European cohorts with a total sample size of
(Videman et al. 1998). Two variants of VDR, namely, the 5,259, this SNP was recently tested for association with disc
FokI polymorphism at exon 2 and the TaqI polymorphism at degeneration. Meta-analysis showed significant association
exon 9, were initially studied in 85 pairs of monozygotic between the SNP and the combination of disc space narrowing
twins. The TaqI tt genotype (with restriction enzyme site) plus osteophytes in women (Williams et al. 2011). The result
was found to be associated with individuals with significantly suggested that in addition to ASPN, GDF5 is another shared
lower disc MRI signal intensities when compared with the genetic risk factor for both disc degeneration and OA, further
other two genotypes; similar findings were noted when FokI implicating similarities in disease mechanisms.
polymorphism was evaluated (Videman et al. 1998). The
association of the TaqI polymorphism with disc degeneration
was replicated in a Japanese cohort of 205 individuals 10.6 Making Sense of Risk Factors
(Kawaguchi et al. 2002). The tt genotype was absent, while and Intervertebral Disc Degeneration
the Tt genotype was associated with multilevel and severe
disc degeneration, as well as disc herniation (Kawaguchi At present, over 20 genes have been reported to be associated
et al. 2002). In a study of a Southern Chinese cohort, indi- with intervertebral disc degeneration. Since these are selected
viduals carrying the t allele were 2.6 times more likely to candidates, they are not difficult to assign an individual bio-
develop disc degeneration and disc bulge. Moreover, the logical function while grouping of these genes can be easily
association was highly significant for a subgroup of individ- linked to molecular processes leading to disc degeneration.
uals age 40 or below (Cheung et al. 2006). The TaqI poly- However, many of the findings need to be critically scrutinized
morphism represents a synonymous substitution, which is for the quality of the genetic data: interim guidelines pro-
believed, in conjunction with other nearby polymorphisms, posed by the HuGENet working group in 2008 provide an
to affect mRNA stability (Uitterlinden et al. 2004). excellent assessment based on three criteria: amount of evi-
Insulin-like growth factor 1 receptor (IGF1R) acts as a dence, replication, and protection from bias (Ioannidis et al.
signal transducer for insulin-like growth factor 1 (IGF1), an 2008). Under each criterion, a classification of strong (A),
anabolic protein that stimulates matrix synthesis and cell moderate (B), or weak (C) can be assigned to the gene study
proliferation in the disc. An intronic SNP, IVS1 + 14488C G or studies. The degree of credibility can then be estimated
(rs11247361) in IGF1R was found to be associated with from standardized 3 3 tables as illustrated in Fig. 10.4: a
radiographic disc narrowing in a cohort of 434 postmeno- score of AAA represents the highest credibility and CCC is
pausal Japanese women (Urano et al. 2008). However, the the lowest. Additional validation is provided by biological
effect of this polymorphism remained unknown. data and functional studies. Leaving the complex issue of the
Sickle tail (SKT) is a gene recently identified through phenotype aside, none of the studies of disc degeneration
gene-trap mutagenesis in mice. SKT-null mice showed devel- genes reach the level of strong credibility, and few are within
opmental spinal abnormalities including dislocation and the level of moderate support. Reasons for this include rela-
defects of the nucleus pulposus at E17.5, leading to a kinky tively small cohort size and lack of replication.
tail phenotype in the adult (Semba et al. 2006). As these VDR represents the best replicated gene linked to disc
observations suggested that SKT has an important role in the degeneration across three different populations. In contrast,
AAA ABA ACA 10.7 Technologies in Genetic Studies

First letter = amount
Second letter = replication
and Intervertebral Disc Degeneration
AAB ABB ACB
Third letter = protection from bias
Advances in both the knowledge of genetic variants and in
AAC ABC ACC
genotyping technologies provide a useful platform for under-
BAA BBA BCA standing how genes influence disc degeneration. The
International HapMap Project aims to capture common vari-
BAB BBB BCB
ants across the genome, as well as frequencies and correla-
BAC BBC BCC tions in different populations (International HapMap
Consortium 2003; Altshuler et al. 2010) (Box 10.3). SNPs
CAA CBA CCA
were the initial focus of study as they represented major
Strong evidence CAB CBB CCB types of variations in populations. From the data generated,
Moderate evidence
the variants and their frequencies provide useful reference
CAC CBC CCC information and allow identification of regions of SNPs with
Weak evidence
strong associations where alleles are co-inherited or in link-
age disequilibrium (LD). An LD map can be produced allow-
Fig. 10.4 Categories for the credibility of cumulative epidemiology
evidence. The three letters, A, B, and C correspond (in order) to the
ing the visualization of the relative co-inheritance of the
amount of evidence, replication, and protection from bias. Evidence SNPs. A practical application of this information is that by
is categorized as strong when there is A for all three items and is catego- careful selection, genotyping of only a few SNPs (tag SNPs)
rized as weak when there is a C for any of the three items. All other can be informative as they cover a large region of interest.
combinations are categorized as moderate (Permission obtained from
Springer for the reproduction of this figure from Ioannidis et al. 2008;
Information concerning the remaining common SNPs in that
License number 2938180337505) region can be predicted, thus significantly reducing genotyp-
ing costs. More recently, the 1000 Genomes Project has been
launched, which is aimed at providing a more profound char-
ASPN (OR 1.70 [95 % CI 1.352.20], P = 0.000013), CILP acterization of genetic variations in multiple populations
(OR 1.61 [95 % CI 1.311.98], P = 0.0000068), COL11A1 (1000 Genomes Project Consortium 2010). Through high-
(OR 1.42 [95 % CI 1.231.65], P = 0.0000033), GDF5 (OR throughput sequencing technologies, its target is to identify
1.72 [95 % CI 1.152.57], P = 0.008), SKT (OR 1.34 [95 % over 95 % of variants, with the ability to detect low-frequency
CI 1.141.58], P = 0.0004), THBS2 (OR 1.43 [95 % CI 1.20 variants down to 1 %. The types of variants being detected
1.70], P = 0.00004), and MMP9 (OR 1.29 [95 % CI 1.12 are also broadened to cover short insertions and deletions
1.48], P = 0.00049) could be considered as potential risk (indels) and structural variations.
factors with moderate genetic evidence. Based on this lim-
ited set of genes, it is clear that the effect size of these genes
is very modest with odd ratios between 1.3 and 1.7. With the
estimated heritability of 74 %, there is likely to be many Box 10.3: The International HapMap Project
more genes associated with disc degeneration yet to be dis- The International HapMap Project is a large-scale col-
covered. Whether there are genes with strong effects remains laborative program involving multiple research centers
to be seen and perhaps such genes will be identified from a from Japan, UK, Canada, China, USA, and Nigeria,
more unbiased genome-wide approach. aiming to determine the common genetic variants in
In general, increasing emphasis should be placed on rep- human genome, their frequencies and correlations
lication of association studies in cohorts with comparable between them, in populations with ancestry from
phenotypes; this is important when evaluating the initial Africa, Asia, and Europe (International HapMap
findings and minimizing false positives (Neale and Sham Consortium 2003). The idea for initiation of this proj-
2004; Chanock et al. 2007). False positives may easily arise ect is as follows. Common diseases are believed to be
when the sample size is small, leading to insufficient power caused by common variants, each having a modest
and imprecise statistical estimation. It can also occur when effect. Although one can genotype all the variants for
the experimental design is inappropriate. Heterogeneity in testing their association with the traits of interest, it is
the genetic background, environmental exposures, and popu- impractical due to the cost. An alternative method is to
lation stratification between cases and controls will result in identify a subset of these variants that can serve as
bias. Technical and genotyping artifacts can also be another genetic markers to refine the regions for association,
source of false positives. Ideally, functional studies should be which can then be further analyzed in details. However,
performed to provide an understanding of the biological rel- how can the suitable markers be selected?
evance of the associated polymorphisms with the disease.
sequencing (NGS), an emerging technology which provides

The probability of recombination events reduces the most comprehensive genetic information with low fre-
with genetic distance, and therefore, alleles in close quency and simultaneously identifies rare SNPs, indels, and
proximity are likely to be co-inherited together. Such structural variants (Metzker 2010). Its application is not only
group of alleles is known as a haplotype. A stronger restricted to genotyping but also includes genome-wide
association among the nearby variants represents a marker and variant discovery (Davey et al. 2011). Despite a
higher level of linkage disequilibrium (LD) and that continuous reduction in NGS costs and improvement in
there would only be a few haplotypes in that region throughput, sequencing the whole genome of every individual
accounting for most of the variations in a population. in a large cohort would appear to be ideal, but not practical.
By choosing and genotyping a few key variants within Depending on the purpose of the study, various methods for
the haplotype (tags), the identity of remaining vari- restricting the specified region and reducing the number of
ants can be predicted. Unfortunately, the extent of samples to be sequenced have been proposed so as to maxi-
association of nearby variants varies across the genome. mize the data yield at reasonable price (Davey et al. 2011).
Neither a random nor an evenly spaced selection of The dramatic increase in genetic information and improve-
variants is ideal for recognizing good tags. The ments in genotyping technology over this past decade has
International HapMap Project assists this important changed the usual practice of studying the genetics of human
step in association studies by developing a haplotype disease. The comprehensive resources of the International
map (HapMap) of the human genome which con- HapMap Project can be used to select variants for associa-
tains comprehensive information of the common SNPs tion studies, while the 1000 Genomes Project has the poten-
and the association between them. Tag SNPs that tial of providing information on the relationship between
best represent the regions of interest can be generated low-frequency variants and diseases. Various genotyping
based on this information. platforms provide opportunities for study of large number of
variants or samples in a short period of time. At present,
genetic studies of disc degeneration are based on the candi-
For genotyping platforms, there are now more options date gene approach, analyzing specific variants within the
than the traditional, labor-intensive approach of RFLP. For genes. It is clear that GWAS analysis is the direction that the
SNP genotyping, the multiplexing Sequenom MassARRAY spine community needs to consider as it is lagging behind
system can process up to 40 SNPs with a maximum of 384 genetic studies of other common disorders such as diabetes,
DNA samples in a single assay. Illumina VeraCode technol- neural degenerative disorder, cardiovascular diseases, and
ogy is similar multiplexing system. They are efficient and osteoarthritis (Bertram et al. 2010; Loughlin 2011; Visscher
cost-effective when there are well-selected candidate genes et al. 2012; Zeller et al. 2012). However, as we enter into yet
or specific regions for fine-mapping with a considerable another new era of genetic data acquisition and building on
number of SNPs to be genotyped in a large cohort. On the important lessons learnt from other GWAS, we should be
other hand, the whole-genome approach can also be used to able to custom design a new approach to evaluate genes asso-
identify new candidates. Predesigned genotyping arrays of ciated with disc degeneration.
genetic markers of SNPs and copy number variations are
commercially available. The Illumina Omni array is able to
detect more than 4.3 million markers, while the Affymetrix 10.8 The Moving Goals in Modern-Day
SNP array can assess 1.8 million markers. The objective of Genetic Studies
both systems is to cover variants with at least 1% of fre-
quency. Huge amount of data can thus be generated using Large-scale genetic studies of a plethora of many common
these arrays and tested directly for genotype-phenotype asso- diseases provide an excellent baseline on which to refine
ciation; this type of whole-genome association is termed as study designs and anticipate potential problems that may
genome-wide association study (GWAS). arise. Indeed, over 1,100 GWAS for various human traits and
As common diseases usually require large sample size for diseases have been published, with the identification of over
statistical power, a major drawback is cost. To address this 2,700 SNP associations (P < 108) (Hindorff et al. 2012).
problem, studies can be designed in which smaller represen- Large-scale GWAS has also been conducted for osteoarthri-
tative sets of cases and controls are used for GWAS. The tis in populations that include Dutch (Rotterdam study)
results can then be validated in larger cohorts or other GWAS (Kerkhof et al. 2010), Japanese (Nakajima et al. 2010), and
in order to achieve desirable sample size for meta-analysis. British (arcOGEN) (Panoutsopoulou et al. 2011), which
Another issue is the limited flexibility of predesigned arrays, sums to a total sample size of over 17,000. Despite regions
which are restricted to relatively common variants and eth- such as chromosome 7q22, human leukocyte antigen (HLA)
nicity. These limitations can be overcome by next-generation locus on chromosome 6p, and other SNPs being identified
that influence osteoarthritis susceptibility, the effect sizes are variants or on a small number of selected individuals with
small and appear to explain only a small portion of the genetic extreme phenotypes to identify shared rare variants. The
variance of the disease. Since this is similar to other diseases resultant candidates can then be genotyped in a large cohort
that include type 2 diabetes, coronary artery disease, and for validation (Cirulli and Goldstein 2010). Apart from the
schizophrenia (So et al. 2011), this begs the question: what sequencing study design, other challenges for studying rare
contributes to the missing heritability? There are intense dis- variants that need to be overcome include selection of appro-
cussions on this topic and a number of possibilities have been priate controls and performing proper statistical analysis.
raised (Manolio et al. 2009; Eichler et al. 2010). GWAS focuses So far the discussion has focused on the presence and
on mainly common SNPs, whereas the human genome con- identification of genetic variants leading to disease. However,
tains many more low-frequency SNPs and structural variants, there are numerous potential modifiers that can alter the
including deletions, duplication, and inversion. These variants, impact on the penetrance of genetic susceptibility, and epigenet-
although individually rare, are collectively common, and their ics is one good example. Epigenetics refers to the heritable
impact on common diseases has not been widely studied. changes of gene expression that are due to mechanisms other
Moreover, interactions among genes and to what extent they than the underlying DNA sequences (Box 10.4). These mecha-
shape phenotypic, epigenetic, and transgenerational genetic nisms can be DNA methylation and histone modification, which
effects may introduce another level of complexity. Genetic remodel chromatin and modify the accessibility of transcription
variants may also influence noncoding microRNA, involving factors to gene promoters. The involvement of epigenetic fac-
translational regulation and therefore, in turn, influencing tors in common diseases has been demonstrated in a recent
mRNA and protein expression and interactions. The contribu- study of schizophrenia and bipolar disorder, in which genome-
tion of these variables may explain missing heritability and, wide DNA methylation patterns between monozygotic twins
more importantly, shed light on future research strategies. discordant for the phenotypes were analyzed. Substantial differ-
Rare variants are generally not in linkage disequilibrium ences were identified, indicating that epigenetic variations con-
with common variants, as they are likely to occur in more tribute to phenotypic differences (Dempster et al. 2011). These
recent generations of the population, and therefore may not observations have culminated in a new hypothesis, common
be detectable in GWAS (Bodmer and Bonilla 2008). However, disease genetic and epigenetic (CDGE), which argues that
collectively, these variants can be numerous. There is increas- epigenetic variations can interplay with genetic variations,
ing evidences that rare variants do play a role in common thus serving as a potential heritable determinant and mod-
diseases, contributing to intermediate effect sizes. Examples ulating the outcome of a disease (Bjornsson et al. 2004;
include three rare deletions associated with schizophrenia,
with one demonstrating an odds ratio of 14.8 (Stefansson
et al. 2008); four rare SNPs in IFIH1 independently reduced Box 10.4: Epigenetics
the risk of type 1 diabetes, with one of the SNPs having an Epigenetics refers to the heritable changes of gene
odds ratio of 0.5 (Nejentsev et al. 2009); as well as 36 very expression that are due to mechanisms other than the
rare non-synonymous variants associated with type 2 diabetes underlying DNA sequences, such as DNA methylation
with odds ratio of 3.3 (Bonnefond et al. 2012). These findings and histone modifications. DNA methylation generally
suggest that rare variants contribution to susceptibility of occurs at GC-rich regions; however, in certain special
common diseases is not an unusual event. Even though the areas named CpG islands, the proportion of methyla-
1000 Genomes Project can serve as a reference panel for the tion is much lower. CpG islands are mainly located
less common variants, it has a limitation in terms of identify- near the promoter regions of human genes. Their meth-
ing variants with frequency of at least 1 %. Genetic variants ylation patterns, which may change during develop-
with lower frequencies are unlikely to be revealed in this ment, are believed to affect the transcriptional activity
database. Moreover, rare variants are generally population of the associated genes. Histones are molecules that
specific, and that information generated from a certain popu- allow DNA to wrap around to form a highly ordered
lation may not be applicable to another. To uncover the rare chromatin structure. Various types of modifications
variants in a specific cohort study, sequencing is currently the such as methylation, acetylation, and phosphorylation
best, but expensive, method. Strategies to reduce costs have can occur on the histone molecules. The chromatin
been proposed; whole-exome sequencing is useful when the structure would be opened up or tightened depending
variants contributing to the diseases are expected to be located on the positions and types of modification acquires,
in exons; or sequencing the regions where potential associa- thus, changing the accessibility of transcription factors
tions have been indicated by GWAS, or with relevance to dis- to the DNA and, in turn, the expression level of corre-
ease etiology. Alternatively, sequencing can be performed on sponding genes.
co-affected members in families to determine co-segregated
differentially regulate the mRNA allelic products (Sethupathy

Epigenetics can be stable as well as plastic (Petronis et al. 2007; Brest et al. 2011); moreover, variance can also
2010). It can be transmitted to the next generation exist at the level of the microRNA transcripts themselves
(transgenerational epigenetic heritability) while at the (Ripke et al. 2011), potentially affecting multiple mRNAs.
same time reprogrammed during fertilization and Currently, there is no gold standard for identification of
gametogenesis. This can be used to interpret the situa- genetic components causing or promoting disc degeneration.
tions of sporadic and familial diseases, when patho- Based on what was discussed earlier in this and other chapters
logical epigenetic marks are reprogrammed or failed to of the book, it is unlikely that its genetic architecture would be
reset, respectively. Epigenetic changes can be induced too different from the many other common diseases. The
by environmental stimuli, but these events can be sto- ample examples of large-scale GWAS would suggest that a
chastic as well. The fidelity of epigenetic pattern trans- more comprehensive whole-genome approach should be used
mission during mitosis is much lower than that of DNA to identify new susceptible loci that may not be obvious from
replication, leading to high variability which can the current understanding of the biology of disc degeneration.
account for the differences observed in monozygotic Successful studies of rare variants, epigenetics, and noncoding
twins. All these properties of epigenetics fit in with the RNAs are important references to consider, thus broadening
understanding of complex diseases, and therefore, the scope of investigation. Given that there are many advances
there are increasing emphasis of investigating the roles in DNA sequencing technology, we should not be confined to
of epigenetics in the etiology of complex diseases. candidate gene case-control analysis. Instead, every possibil-
ity should be explored to achieve the best understanding of the
genetic components of disc degeneration.
Feinberg 2007). Attempt has been made to integrate genetic
and epigenetic data, in which increased DNA methylation was
noted in the FTO locus, a region previously shown to be sus- 10.9 Genetic Studies of Intervertebral Disc
ceptible to type 2 diabetes and obesity. While the methylation Degeneration: The Way Forward
difference was driven by several SNPs creating CpG sites, the
resultant effects on the diseases still require further elucidation It is difficult to recommend a specific direction for future
(Bell et al. 2010). This study provides a hint concerning the investigations of the genetics of disc degeneration.
potential genetic and epigenetic interactions in shaping com- Nevertheless, from the authors experience, there are sev-
mon diseases; from this perspective there is a need to further eral points that hopefully can make the most of the studies
global investigation of epigenetics. Epigenome-wide associa- already performed and maximize the chance of identifying
tion studies (EWAS), similar to the notion of GWAS, is a new new risk factors. A unified phenotypic definition is required
concept directed at studying epigenetic variations in relation- such that there is a consensus of what constitutes the dis-
ship to disease in a genome-wide manner (Rakyan et al. 2011b). ease status, so that the same group of underlying genetic
This has been made possible by array-based profiling or whole- factors can be identified. Careful selection of cases and
genome bisulphite sequencing with appropriate experimental controls can avoid biases and hidden population
design and statistical analysis. Despite these approaches, cur- stratifications leading to false positive results. Larger sam-
rently the only target to detect is DNA methylation. An initial ple sizes always provide a more accurate evaluation of vari-
EWAS study has already been conducted in type 1 diabetes ants under study. All these criteria form the important basis
(Rakyan et al. 2011a), which has provided insights on its appli- of genetic studies; when fulfilled, they can make each indi-
cation to common human diseases. vidual study more reliable and relevant. Moreover, the new
Due to the ability to regulate gene expression, microR- information can be easily integrated for replication in other
NAs, a class of small noncoding RNAs of approximately 22 populations and for confirmation through meta-analysis.
nucleotides, also contribute to genetic outcomes. Depending Regions previously showing (marginal) significance should
on the degree of complementarity between microRNAs and be considered for reanalysis. For example, partially linked
mRNAs, they repress translation processes through mRNA common variants with weak associations may actually
degradation, or by inhibiting translation initiation. Differential reflect causal and neighboring rare alleles. As we now
expression of microRNAs has been reported for many com- know, since rare variants can affect common diseases with
mon diseases, such as osteoarthritis (Jones et al. 2009), moderate effect sizes, revisiting these studies can help to
schizophrenia (Perkins et al. 2007), and bipolar disorder provide a more comprehensive evaluation of these regions.
(Moreau et al. 2011). Another level of regulation complexity In parallel, functional analyses of the associated variants
arises when the genetic variants occur at microRNA binding are required as they provide biological insights while at the
sites of mRNA. This can alter the binding affinity and same time further consolidating the initial findings.
Common allele
GWAS
Using HapMap and 1000
Genomes as reference
Higher density arrays

Allele frequency
Candidate
gene
Family
Sequencing linkage
Rare allele
Single gene Whole-genome

Proportion of genome
Fig. 10.5 Summary on the approaches for genetic studies. Candidate alleles, but now, lower frequency alleles are also incorporated in the
gene approach is a primarily gene-by-gene investigation. Common higher density arrays as well. Family linkage is also a whole-genome
alleles (or lower frequency alleles) can be studied easily using data- approach. Its detection of susceptible loci relies on large families with
bases such as HapMap and 1000 Genomes as reference. On the other multiple affected members and, more importantly, variants with high
hand, if rare alleles are of interest, one would need to adopt the sequenc- penetrance and large effect sizes. Therefore, its application in detecting
ing method. GWAS approach, as its name suggested, looks into the common disease variants is limited
whole genome. In general, it is designed to accommodate common
Disc degeneration
Genetics
Non coding Environment

RNA
Epigenetics
Fig. 10.6 Current and future genetics studies of disc degeneration. Based epigenetic patterns may be modified by genetic variants. Environmental
on the studies of other common diseases, we understand that genetics, factors can interact with genes and epigenetics, and at the same time, they
while by itself is one of the major contributor to disc degeneration, can also can directly contribute to disc degeneration. Last but not least, gene-gene
interact with a number of other factors leading to altered effects. Both non- interactions also exist, which leads to variable consequences. Together,
coding RNA and epigenetics can modulate the expression outcome. On the these represent a complicated network that needs to be investigated for a
other hand, noncoding RNAs can be affected by the genetic variants at their better understanding of disc degeneration and that genetics is not indepen-
expression levels, binding sites, or even their own sequence, while dent but should be studied in conjunction with other factors
Technology for providing genetic information is rapidly References

developing. Its success with other common diseases leads
to the following questions: is it necessary to catch up with 1000 Genomes Project Consortium GP (2010) A map of human genome
variation from population-scale sequencing. Nature 467(7319):
the latest technology in studying the genetics of disc
10611073
degeneration, and if so, how should that be achieved? Aladin DM, Cheung KM et al (2007) Expression of the Trp2 allele of
Obviously, each of the current technologies has its own COL9A2 is associated with alterations in the mechanical properties
strength: GWAS allows SNP analysis on the whole genome of human intervertebral discs. Spine 32(25):28202826
Alanay Y, Avaygan H et al (2010) Mutations in the gene encoding the
and is advantageous for identifying new susceptible loci,
RER protein FKBP65 cause autosomal-recessive osteogenesis
while sequencing is favorable for discovering new variants imperfecta. Am J Hum Genet 86(4):551559
(Fig. 10.5). Both are promising tools; however, instead of Altshuler DM, Gibbs RA et al (2010) Integrating common and rare genetic
blindly chasing after a new technology, it is more impor- variation in diverse human populations. Nature 467(7311):5258
Annunen S, Paassilta P et al (1999) An allele of COL9A2 associated
tant to consider carefully which aspects of genetics is
with intervertebral disc disease. Science 285(5426):409412
being studied and the type of information desired. Based Barnes AM, Chang W et al (2006) Deficiency of cartilage-associated
on these considerations, the optimum technology can be protein in recessive lethal osteogenesis imperfecta. N Engl J Med
chosen or various strategies can be combined to maximize 355(26):27572764
Battie MC, Haynor DR et al (1995a) Similarities in degenerative
data generation while minimizing costs. Factors modify-
findings on magnetic resonance images of the lumbar spines of
ing expression outcome such as epigenetics and noncoding identical twins. J Bone Joint Surg Am 77(11):16621670
RNA and interactions between genes, genes and proteins, Battie MC, Videman T et al (1995b) 1995 Volvo Award in clinical sci-
as well as genes and environment should not be neglected ences. Determinants of lumbar disc degeneration. A study relating
lifetime exposures and magnetic resonance imaging findings in
(Fig. 10.6). The overall information can then be fed back
identical twins. Spine 20(24):26012612
to the investigators to provide a more complete under- Bell CG, Finer S et al (2010) Integrated genetic and epigenetic analysis
standing of the biology of the intervertebral disc and identifies haplotype-specific methylation in the FTO type 2 diabetes
mechanisms of degeneration. and obesity susceptibility locus. PLoS One 5(11):e14040
Benneker LM, Heini PF et al (2005) Correlation of radiographic and
MRI parameters to morphological and biochemical assessment of
intervertebral disc degeneration. Eur Spine J 14(1):2735
10.10 Summary of Critical Concepts Bertram L, Lill CM et al (2010) The genetics of Alzheimer disease:
Discussed in the Chapter back to the future. Neuron 68(2):270281
Bjornsson HT, Fallin MD et al (2004) An integrated epigenetic and
genetic approach to common human disease. Trends Genet 20(8):
Intervertebral disc degeneration is a complex trait contrib- 350358
uted by genetics and multiple environmental factors such Bodmer W, Bonilla C (2008) Common and rare variants in multifacto-
as age, gender, smoking, and mechanical loading. rial susceptibility to common diseases. Nat Genet 40(6):695701
Bonnefond A, Clement N et al (2012) Rare MTNR1B variants impair-
Heritability for disc degeneration at the lumbar region is
ing melatonin receptor 1B function contribute to type 2 diabetes.
approximately 70 %, suggesting a high degree of genetic Nat Genet 44(3):297301
involvement. Bornstein P, Agah A et al (2004) The role of thrombospondins 1 and 2
Over 20 genes are reported to be associated with in the regulation of cell-matrix interactions, collagen fibril forma-
tion, and the response to injury. Int J Biochem Cell Biol 36(6):
disc degeneration by case-control studies of selected
11151125
candidates. Boyd LM, Richardson WJ et al (2008) Early-onset degeneration of the
VDR is the best replicated gene, while ASPN, CILP, intervertebral disc and vertebral end plate in mice deficient in type
COL11A1, GDF5, SKT, THBS2, and MMP9 can be con- IX collagen. Arthritis Rheum 58(1):164171
Brest P, Lapaquette P et al (2011) A synonymous variant in IRGM
sidered as potential risk factors with moderate genetic
alters a binding site for miR-196 and causes deregulation of IRGM-
evidence based on interim guidelines propose by HuGENet dependent xenophagy in Crohns disease. Nat Genet 43(3):
to assess the quality of genetic data. 242245
The International HapMap Project and 1000 Genomes Cabral WA, Chang W et al (2007) Prolyl 3-hydroxylase 1 deficiency
causes a recessive metabolic bone disorder resembling lethal/severe
Project, as well as the latest high-throughput genotyping
osteogenesis imperfecta. Nat Genet 39(3):359365
technologies, are valuable resources that can be integrated Chanock SJ, Manolio T et al (2007) Replicating genotype-phenotype
with current approaches to push forward genetics studies associations. Nature 447(7145):655660
of disc degeneration. Cheung KM, Chan D et al (2006) Association of the Taq I allele in
vitamin D receptor with degenerative disc disease and disc bulge in
Factors such as epigenetics and noncoding RNA and
a Chinese population. Spine 31(10):11431148
interactions between genes, genes and proteins, as well as Cheung KM, Karppinen J et al (2009) Prevalence and pattern of lumbar
genes and environment should not be neglected as they MRI changes in a population study of 1043 individuals. Spine 34(9):
may affect the final outcome of the genetic components. 934940
Cirulli ET, Goldstein DB (2010) Uncovering the roles of rare variants
in common disease through whole-genome sequencing. Nat Rev
Acknowledgements This work is supported by the University Genet 11(6):415425
Grants Committee of Hong Kong (AoE/M-04/04) and AOSPINE Cong L, Pang H et al (2010a) The interaction between aggrecan gene
(AOSBRC-07-2). VNTR polymorphism and cigarette smoking in predicting incident
symptomatic intervertebral disc degeneration. Connect Tissue Res Karasugi T, Semba K et al (2009) Association of the tag SNPs in the
51(5):397403 human SKT gene (KIAA1217) with lumbar disc herniation. J Bone
Cong L, Pang H et al (2010b) Association between the expression of Miner Res 24(9):15371543
aggrecan and the distribution of aggrecan gene variable number of Kawaguchi Y, Osada R et al (1999) Association between an aggrecan
tandem repeats with symptomatic lumbar disc herniation in Chinese gene polymorphism and lumbar disc degeneration. Spine
Han of Northern China. Spine 35(14):13711376 24(23):24562460
Daly MJ (2009) Assessing significance in genetic association studies. Kawaguchi Y, Kanamori M et al (2002) The association of lumbar disc
Cold Spring Harb Protoc 2009(8):pdb top58 disease with vitamin-D receptor gene polymorphism. J Bone Joint
Davey JW, Hohenlohe PA et al (2011) Genome-wide genetic marker Surg Am 84-A(11):20222028
discovery and genotyping using next-generation sequencing. Nat Kelempisioti A, Eskola PJ et al (2011) Genetic susceptibility of inter-
Rev Genet 12(7):499510 vertebral disc degeneration among young Finnish adults. BMC Med
Dempster EL, Pidsley R et al (2011) Disease-associated epigenetic Genet 12(1):153
changes in monozygotic twins discordant for schizophrenia and Kellgren JH, Jeffrey MR et al (1963) The epidemiology of chronic rheu-
bipolar disorder. Hum Mol Genet 20(24):47864796 matism: atlas of standard radiographs. Blackwell Scientific, Oxford
Doege KJ, Coulter SN et al (1997) A human-specific polymorphism in Kerkhof HJ, Lories RJ et al (2010) A genome-wide association study
the coding region of the aggrecan gene. Variable number of tandem identifies an osteoarthritis susceptibility locus on chromosome
repeats produce a range of core protein sizes in the general popula- 7q22. Arthritis Rheum 62(2):499510
tion. J Biol Chem 272(21):1397413979 Khoschnau S, Melhus H et al (2008) Type I collagen alpha1 Sp1 poly-
Dong DM, Yao M et al (2007) Association between the -1306C/T polymorphism and the risk of cruciate ligament ruptures or shoulder
morphism of matrix metalloproteinase-2 gene and lumbar disc dis- dislocations. Am J Sports Med 36(12):24322436
ease in Chinese young adults. Eur Spine J 16(11):19581961 Kim NK, Shin DA et al (2011) The association of aggrecan gene poly-
Eichler EE, Flint J et al (2010) Missing heritability and strategies for morphism with the risk of intervertebral disc degeneration. Acta
finding the underlying causes of complex disease. Nat Rev Genet Neurochir (Wien) 153(1):129133
11(6):446450 Kimura T, Nakata K et al (1996) Progressive degeneration of articular
Eser O, Eser B et al (2011) Short aggrecan gene repetitive alleles asso- cartilage and intervertebral discs. An experimental study in trans-
ciated with lumbar degenerative disc disease in Turkish patients. genic mice bearing a type IX collagen mutation. Int Orthop
Genet Mol Res 10(3):19231930 20(3):177181
Feinberg AP (2007) Phenotypic plasticity and the epigenetics of human Kizawa H, Kou I et al (2005) An aspartic acid repeat polymorphism in
disease. Nature 447(7143):433440 asporin inhibits chondrogenesis and increases susceptibility to
Francis-West PH, Abdelfattah A et al (1999) Mechanisms of GDF-5 osteoarthritis. Nat Genet 37(2):138144
action during skeletal development. Development 126(6):13051315 Klatt AR, Klinger G et al (2009) Matrilin-3 activates the expression of
Garnero P, Borel O et al (1998) Collagen Ialpha1 Sp1 polymorphism, osteoarthritis-associated genes in primary human chondrocytes.
bone mass, and bone turnover in healthy French premenopausal FEBS Lett 583(22):36113617
women: the OFELY study. J Bone Miner Res 13(5):813817 Klatt AR, Becker AK et al (2011) The matrilins: modulators of
Goupille P, Jayson MI et al (1998) Matrix metalloproteinases: the clue extracellular matrix assembly. Int J Biochem Cell Biol 43(3):
to intervertebral disc degeneration? Spine 23(14):16121626 320330
Grant SF, Reid DM et al (1996) Reduced bone density and osteoporosis Lapunzina P, Aglan M et al (2010) Identification of a frameshift muta-
associated with a polymorphic Sp1 binding site in the collagen type tion in Osterix in a patient with recessive osteogenesis imperfecta.
I alpha 1 gene. Nat Genet 14(2):203205 Am J Hum Genet 87(1):110114
Gruber HE, Ingram JA et al (2009) Asporin, a susceptibility gene in Le Maitre CL, Freemont AJ et al (2004) Localization of degradative
osteoarthritis, is expressed at higher levels in the more degenerate enzymes and their inhibitors in the degenerate human intervertebral
human intervertebral disc. Arthritis Res Ther 11(2):R47 disc. J Pathol 204(1):4754
Higashino K, Matsui Y et al (2007) The alpha2 type IX collagen tryp- Le Maitre CL, Freemont AJ et al (2005) The role of interleukin-1 in the
tophan polymorphism is associated with the severity of disc degen- pathogenesis of human intervertebral disc degeneration. Arthritis
eration in younger patients with herniated nucleus pulposus of the Res Ther 7(4):R732R745
lumbar spine. Int Orthop 31(1):107111 Le Maitre CL, Hoyland JA et al (2007) Catabolic cytokine expression
Hindorff LA, MacArthur J et al (2012) A catalog of published genome- in degenerate and herniated human intervertebral discs: IL-1beta
wide association studies. Available at www.genome.gov/gwastudies. and TNFalpha expression profile. Arthritis Res Ther 9(4):R77
Accessed 24 Feb 2012 Lian K, Zmuda JM et al (2005) Type I collagen alpha1 Sp1 transcrip-
Hirose Y, Chiba K et al (2008) A functional polymorphism in THBS2 tion factor binding site polymorphism is associated with reduced
that affects alternative splicing and MMP binding is associated with risk of hip osteoarthritis defined by severe joint space narrowing in
lumbar-disc herniation. Am J Hum Genet 82(5):11221129 elderly women. Arthritis Rheum 52(5):14311436
Internation HapMap Consortium (2003) The International HapMap Lin WP, Lin JH et al (2011) Interleukin-10 promoter polymorphisms
Project. Nature 426(6968):789796 associated with susceptibility to lumbar disc degeneration in a
Ioannidis JP, Boffetta P et al (2008) Assessment of cumulative evidence Chinese cohort. Genet Mol Res 10(3):17191727
on genetic associations: interim guidelines. Int J Epidemiol 37(1): Lorenzo P, Bayliss MT et al (1998) A novel cartilage protein (CILP)
120132 present in the mid-zone of human articular cartilage increases with
Jim JJ, Noponen-Hietala N et al (2005) The TRP2 allele of COL9A2 is age. J Biol Chem 273(36):2346323468
an age-dependent risk factor for the development and severity of Lorenzo P, Aspberg A et al (2001) Identification and characterization of
intervertebral disc degeneration. Spine 30(24):27352742 asporin. A novel member of the leucine-rich repeat protein family
Jones SW, Watkins G et al (2009) The identification of differentially closely related to decorin and biglycan. J Biol Chem 276(15):
expressed microRNA in osteoarthritic tissue that modulate the pro- 1220112211
duction of TNF-alpha and MMP13. Osteoarthritis Cartilage 17(4): Loughlin J (2011) Osteoarthritis year 2010 in review: genetics.
464472 Osteoarthritis Cartilage 19(4):342345
Kales SN, Linos A et al (2004) The role of collagen IX tryptophan poly- Mann V, Hobson EE et al (2001) A COL1A1 Sp1 binding site polymor-
morphisms in symptomatic intervertebral disc disease in Southern phism predisposes to osteoporotic fracture by affecting bone density
European patients. Spine 29(11):12661270 and quality. J Clin Invest 107(7):899907
Manolio TA, Collins FS et al (2009) Finding the missing heritability of Price SJ, Greaves DR et al (2001) Identification of novel, functional
complex diseases. Nature 461(7265):747753 genetic variants in the human matrix metalloproteinase-2 gene: role
Marinelli NL, Haughton VM et al (2009) T2 relaxation times of inter- of Sp1 in allele-specific transcriptional regulation. J Biol Chem
vertebral disc tissue correlated with water content and proteoglycan 276(10):75497558
content. Spine 34(5):520524 Rakyan VK, Beyan H et al (2011a) Identification of type 1 diabetes-
Martinez-Glez V, Valencia M et al (2011) Identification of a mutation associated DNA methylation variable positions that precede disease
causing deficient BMP1/mTLD proteolytic activity in autosomal diagnosis. PLoS Genet 7(9):e1002300
recessive osteogenesis imperfecta. Hum Mutat 33(2):343350 Rakyan VK, Down TA et al (2011b) Epigenome-wide association stud-
Matsui Y, Wu JJ et al (2003) Matrix deposition of tryptophan-contain- ies for common human diseases. Nat Rev Genet 12(8):529541
ing allelic variants of type IX collagen in developing human carti- Ripke S, Sanders AR et al (2011) Genome-wide association study
lage. Matrix Biol 22(2):123129 identifies five new schizophrenia loci. Nat Genet 43(10):
Metzker ML (2010) Sequencing technologies the next generation. 969976
Nat Rev Genet 11(1):3146 Roberts S, Caterson B et al (2000) Matrix metalloproteinases and
Millward-Sadler SJ, Costello PW et al (2009) Regulation of catabolic aggrecanase: their role in disorders of the human intervertebral disc.
gene expression in normal and degenerate human intervertebral disc Spine 25(23):30053013
cells: implications for the pathogenesis of intervertebral disc degen- Sambrook PN, MacGregor AJ et al (1999) Genetic influences on cervi-
eration. Arthritis Res Ther 11(3):R65 cal and lumbar disc degeneration: a magnetic resonance imaging
Min JL, Meulenbelt I et al (2006) Association of matrilin-3 polymor- study in twins. Arthritis Rheum 42(2):366372
phisms with spinal disc degeneration and osteoarthritis of the first car- Schneiderman G, Flannigan B et al (1987) Magnetic resonance imaging
pometacarpal joint of the hand. Ann Rheum Dis 65(8):10601066 in the diagnosis of disc degeneration: correlation with discography.
Min SK, Nakazato K et al (2010) Cartilage intermediate layer protein Spine 12(3):276281
gene is associated with lumbar disc degeneration in male, but not Seki S, Kawaguchi Y et al (2005) A functional SNP in CILP, encoding
female, collegiate athletes. Am J Sports Med 38(12):25522557 cartilage intermediate layer protein, is associated with susceptibility
Mio F, Chiba K et al (2007) A functional polymorphism in COL11A1, to lumbar disc disease. Nat Genet 37(6):607612
which encodes the alpha 1 chain of type XI collagen, is associated Seki S, Kawaguchi Y et al (2006) Association study of COL9A2 with
with susceptibility to lumbar disc herniation. Am J Hum Genet lumbar disc disease in the Japanese population. J Hum Genet 51(12):
81(6):12711277 10631067
Mitchell BD, Yerges-Armstrong LM (2011) The genetics of bone loss: Semba K, Araki K et al (2006) A novel murine gene, Sickle tail, linked
challenges and prospects. J Clin Endocrinol Metab 96(5):12581268 to the Danforths short tail locus, is required for normal develop-
Miyamoto Y, Mabuchi A et al (2007) A functional polymorphism in the ment of the intervertebral disc. Genetics 172(1):445456
5 UTR of GDF5 is associated with susceptibility to osteoarthritis. Sethupathy P, Borel C et al (2007) Human microRNA-155 on chromo-
Nat Genet 39(4):529533 some 21 differentially interacts with its polymorphic target in the
Moreau MP, Bruse SE et al (2011) Altered microRNA expression AGTR1 3 untranslated region: a mechanism for functional single-
profiles in postmortem brain samples from individuals with schizo- nucleotide polymorphisms related to phenotypes. Am J Hum Genet
phrenia and bipolar disorder. Biol Psychiatry 69(2):188193 81(2):405413
Morello R, Bertin TK et al (2006) CRTAP is required for prolyl 3- Skorupski P, Krol J et al (2006) An alpha-1 chain of type I collagen
hydroxylation and mutations cause recessive osteogenesis imper- Sp1-binding site polymorphism in women suffering from stress uri-
fecta. Cell 127(2):291304 nary incontinence. Am J Obstet Gynecol 194(2):346350
Nakajima M, Takahashi A et al (2010) New sequence variants in HLA So HC, Gui AH et al (2011) Evaluating the heritability explained by
class II/III region associated with susceptibility to knee osteoarthri- known susceptibility variants: a survey of ten complex diseases.
tis identified by genome-wide association study. PLoS One 5(3): Genet Epidemiol 35(5):310317
e9723 Solovieva S, Lohiniva J et al (2002) COL9A3 gene polymorphism and
Nakaoka H, Inoue I (2009) Meta-analysis of genetic association stud- obesity in intervertebral disc degeneration of the lumbar spine: evi-
ies: methodologies, between-study heterogeneity and winners dence of gene-environment interaction. Spine 27(23):26912696
curse. J Hum Genet 54(11):615623 Solovieva S, Kouhia S et al (2004) Interleukin 1 polymorphisms and
Neale BM, Sham PC (2004) The future of association studies: gene- intervertebral disc degeneration. Epidemiology 15(5):626633
based analysis and replication. Am J Hum Genet 75(3):353362 Solovieva S, Lohiniva J et al (2006) Intervertebral disc degeneration in
Nejentsev S, Walker N et al (2009) Rare variants of IFIH1, a gene relation to the COL9A3 and the IL-1ss gene polymorphisms. Eur
implicated in antiviral responses, protect against type 1 diabetes. Spine J 15(5):613619
Science 324(5925):387389 Solovieva S, Noponen N et al (2007) Association between the aggrecan
Paassilta P, Lohiniva J et al (2001) Identification of a novel common gene variable number of tandem repeats polymorphism and inter-
genetic risk factor for lumbar disk disease. JAMA 285(14): vertebral disc degeneration. Spine 32(16):17001705
18431849 Song YQ, Cheung KM et al (2008a) Association of the asporin D14
Panoutsopoulou K, Southam L et al (2011) Insights into the genetic allele with lumbar-disc degeneration in Asians. Am J Hum Genet
architecture of osteoarthritis from stage 1 of the arcOGEN study. 82(3):744747
Ann Rheum Dis 70(5):864867 Song YQ, Ho DW et al (2008b) Association between promoter -1607
Perkins DO, Jeffries CD et al (2007) microRNA expression in the pre- polymorphism of MMP1 and lumbar disc disease in Southern
frontal cortex of individuals with schizophrenia and schizoaffective Chinese. BMC Med Genet 9:38
disorder. Genome Biol 8(2):R27 Speer G, Szenthe P et al (2006) Myocardial infarction is associated with
Petronis A (2010) Epigenetics as a unifying principle in the aetiology of Spl binding site polymorphism of collagen type 1A1 gene. Acta
complex traits and diseases. Nature 465(7299):721727 Cardiol 61(3):321325
Pfirrmann CW, Metzdorf A et al (2001) Magnetic resonance Stefansson H, Rujescu D et al (2008) Large recurrent microdeletions
classification of lumbar intervertebral disc degeneration. Spine associated with schizophrenia. Nature 455(7210):232236
26(17):18731878 Sun ZM, Miao L et al (2009) Association between the -1562 C/T poly-
Pluijm SM, van Essen HW et al (2004) Collagen type I alpha1 Sp1 morphism of matrix metalloproteinase-9 gene and lumbar disc dis-
polymorphism, osteoporosis, and intervertebral disc degeneration in ease in the young adult population in North China. Connect Tissue
older men and women. Ann Rheum Dis 63(1):7177 Res 50(3):181185
Tabor HK, Risch NJ et al (2002) Candidate-gene approaches for study- Videman T, Leppavuori J et al (1998) Intragenic polymorphisms of the
ing complex genetic traits: practical considerations. Nat Rev Genet vitamin D receptor gene associated with intervertebral disc degen-
3(5):391397 eration. Spine 23(23):24772485
Takahashi M, Haro H et al (2001) The association of degeneration of Videman T, Saarela J et al (2009) Associations of 25 structural, degrada-
the intervertebral disc with 5a/6a polymorphism in the promoter of tive, and inflammatory candidate genes with lumbar disc desiccation,
the human matrix metalloproteinase-3 gene. J Bone Joint Surg Br bulging, and height narrowing. Arthritis Rheum 60(2):470481
83(4):491495 Vikkula M, Palotie A et al (1993) Early-onset osteoarthritis linked to
Tilkeridis C, Bei T et al (2005) Association of a COL1A1 polymor- the type II procollagen gene. Detailed clinical phenotype and further
phism with lumbar disc disease in young military recruits. J Med analyses of the gene. Arthritis Rheum 36(3):401409
Genet 42(7):e44 Virtanen IM, Song YQ et al (2007) Phenotypic and population differ-
Uitterlinden AG, Burger H et al (1998) Relation of alleles of the colla- ences in the association between CILP and lumbar disc disease.
gen type Ialpha1 gene to bone density and the risk of osteoporotic J Med Genet 44(4):285288
fractures in postmenopausal women. N Engl J Med 338(15): Visscher PM, Hill WG et al (2008) Heritability in the genomics
10161021 era concepts and misconceptions. Nat Rev Genet 9(4):255266
Uitterlinden AG, Fang Y et al (2004) Genetics and biology of vitamin Visscher PM, Brown MA et al (2012) Five years of GWAS discovery.
D receptor polymorphisms. Gene 338(2):143156 Am J Hum Genet 90(1):724
Urano T, Narusawa K et al (2008) Association of a single nucleotide Wagner GP, Zhang J (2011) The pleiotropic structure of the genotype-
polymorphism in the insulin-like growth factor-1 receptor gene with phenotype map: the evolvability of complex organisms. Nat Rev
spinal disc degeneration in postmenopausal Japanese women. Spine Genet 12(3):204213
33(11):12561261 Williams FM, Popham M et al (2011) GDF5 single-nucleotide
Valdes AM, Spector TD (2011) Genetic epidemiology of hip and knee polymorphism rs143383 is associated with lumbar disc degen-
osteoarthritis. Nat Rev Rheumatol 7(1):2332 eration in Northern European women. Arthritis Rheum 63(3):
Valdes AM, Hassett G et al (2005) Radiographic progression of lumbar 708712
spine disc degeneration is influenced by variation at inflammatory Wrocklage C, Wassmann H et al (2000) COL9A2 allelotypes in
genes: a candidate SNP association study in the Chingford cohort. intervertebral disc disease. Biochem Biophys Res Commun
Spine 30(21):24452451 279(2):398400
van Dijk FS, Nesbitt IM et al (2009) PPIB mutations cause severe Zeller T, Blankenberg S et al (2012) Genomewide association
osteogenesis imperfecta. Am J Hum Genet 85(4):521527 studies in cardiovascular disease an update 2011. Clin Chem
Videman T, Nummi P et al (1994) Digital assessment of MRI for lum- 58(1):92103
bar disc desiccation. A comparison of digital versus subjective
assessments and digital intensity profiles versus discogram and
macroanatomic findings. Spine 19(2):192198
Pathogenesis of Intervertebral
Disc Degeneration 11
Stephen M. Richardson, Anthony J. Freemont,
and Judith A. Hoyland

11.1 Introduction ..................................................................... 177
11.2 The Cell Biology of the Normal Human It is estimated that as much as 84 % of the population will
Intervertebral Disc .......................................................... 178 suffer from low back pain (LBP) at some point in their life-
11.3 Intervertebral Disc Degeneration .................................. 179 time (Walker 2000), with around 10 % of sufferers being
11.3.1 Morphological Features of Degeneration.......................... 179 chronically disabled. As such LBP is one of the most preva-
11.3.2 Genetic Influences ............................................................. 179 lent musculoskeletal conditions affecting Western society
11.3.3 Alterations in Extracellular Matrix Composition.............. 180
11.3.4 Matrix Degradation ........................................................... 181
(Stewart et al. 2003), and its prevalence has increased over
recent decades (Harkness et al. 2005). The socio-economic
11.4 Vascular and Nerve Ingrowth ........................................ 182
cost of LBP is also huge, with associated costs, in terms of
11.5 Alterations in Disc Cell Biology in Degeneration ......... 183
11.5.1 Soluble Regulators of Cellular Function ........................... 183
lost productivity, disability benefits and direct and indirect
health-care costs, estimated in the UK to be around 12 bil-
11.6 Changes in Disc Nutrition and Oxygen Tension........... 186
lion annually (Maniadakis and Gray 2000) and in the USA to
11.7 Cell Ageing and Death .................................................... 187
be over $85 billion per annum (Martin et al. 2008).
11.8 Response to Mechanical Load........................................ 190 Importantly, increases in both the size and average age of the
11.9 Clinical Implications: Relevance of Understanding population both suggest that the prevalence and costs associ-
the Cell Biology and Pathogenesis of Intervertebral ated with LBP will continue to rise over future decades,
Disc Degeneration for Development of Novel
Therapeutic Agents ......................................................... 191 unless novel therapies can be developed to alleviate pain and
restore long-term function and mobility to the spine.
in the Chapter .................................................................. 192 However, in order to develop such therapies, a more thor-
References ..................................................................................... 193 ough understanding of the underlying aetiology is required.
While it is acknowledged that LBP is a multifactorial con-
dition, a strong correlation with degeneration of the interver-
tebral disc has been shown in 40 % of cases (Cheung et al.
2009). However, until recently, the pathogenesis of disc
degeneration and its role in LBP were poorly understood.
A.J. Freemont
Centre for Regenerative Medicine,
S.M. Richardson J.A. Hoyland (*) Institute of Inflammation and Repair,
Centre for Regenerative Medicine, Faculty of Medical and Human Sciences,
Institute of Inflammation and Repair, The University of Manchester,
Faculty of Medical and Human Sciences, Stopford Building Oxford Road,
The University of Manchester, Manchester M13 9PT, UK
Stopford Building Oxford Road, School of Medicine, The University of Manchester,
Manchester M13 9PT, UK Stopford Building Oxford Road,
e-mail: s.ricahrdson@manchester.ac.uk; Manchester M13 9PT, UK
judith.hoyland@manchester.ac.uk e-mail: tony.freemont@manchester.ac.uk

178 S.M. Richardson et al.
This chapter will describe the normal intervertebral disc and The nucleus pulposus is constrained circumferentially by
the changes which occur during degeneration at a molecular, the outer region of the disc, the annulus fibrosus, a fibrous
cellular and tissue level and review the clinical implications ring of tissue with highly ordered collagen I fibrils orientated
of these pathological changes in the context of LBP. in 60 oblique lamellae (Marchand and Ahmed 1990).
Collagens II and III are also present in the annulus fibrosus
and the total collagen content is around 80 %, compared to
11.2 The Cell Biology of the Normal Human only around 20 % in the nucleus pulposus (Roughley 2004;
Intervertebral Disc Le Maitre et al. 2007d). While the extracellular matrix of the
annulus fibrosus does contain proteoglycans, predominantly
The intervertebral disc is located between the vertebrae of versican, they are mainly located between the lamellae, along
the spine and is comprised of three morphologically distinct with elastin fibres which are thought to allow flexion or
regions. The central core of the disc, the nucleus pulposus, is extension during movement (Yu et al. 2002; Melrose et al.
a highly hydrated, gelatinous tissue containing small rounded 2008; Smith et al. 2009). Morphologically and phenotypi-
nucleus pulposus cells embedded within a dense extracellu- cally, fibroblastic annulus fibrosus cells appear to orientate
lar matrix. These cells are routinely described as being chon- with the collagen fibres in each lamella ring. Although the
drocyte-like (Sive et al. 2002), due in part to similarities in cell and matrix biology differs between the nucleus pulposus
both their morphology and composition of the matrix that and annulus fibrosus, there is no distinct demarcation between
they synthesise and secrete. Noteworthy, recent microarray the tissues. Instead, the inner annulus fibrosus, sometimes
studies have shed new light on the exact phenotype of nucleus referred to as a transition zone, demonstrates a mix of cell
pulposus and annulus fibrosus cells, illustrating distinct dif- types with both rounded nucleus pulposus and flattened
ferences in gene expression profiles between these cells and annulus fibrosus cells present. The matrix changes are also
articular chondrocytes (Lee et al. 2007; Sakai et al. 2009; gradual, with collagens I and II contents being inversely cor-
Minogue et al. 2010a, b). The extracellular matrix of the related (Eyre and Muir 1976). This integration of tissues
nucleus pulposus is rich in proteoglycans, particularly the allows the disc to bulge in a constrained manner under load-
large aggregating proteoglycans aggrecan and versican ing, allowing distribution and dissipation of the mechanical
which possess a large number of negatively charged gly- forces, including flexion, tension, compression and torsion,
cosaminoglycan (GAG) side chains. The GAGs attract posi- experienced during everyday motion. These concepts are
tively charged ions, which gives the nucleus pulposus a high further developed in Chap. 7.
osmotic potential, which in turn acts to draw in water result- The inferior and superior faces on the disc, where they
ing in a tissue with a water content between 70 and 90 % meet adjacent vertebral bodies, are covered by a thin layer of
(Antoniou et al. 1996). A range of other smaller PGs are also hyaline cartilage, the cartilaginous end plates. This is a thin
present, including biglycan, decorin and fibromodulin, which layer of hyaline cartilage containing a population of chon-
are thought to play a number of structural and physiological drocytic cells. Collagen fibres from the annulus fibrosus
roles, including growth factor binding and mediation of sig- embed directly into the vertebral bodies and into the carti-
nalling between cells and the extracellular matrix (Roughley laginous end plates, which prevent the nucleus pulposus
2004; Feng et al. 2006). Further details of the proteoglycans bulging into the vertebral bodies (Humzah and Soames
of the disc are presented in Chap. 4. The nucleus pulposus 1988). The cartilaginous end plates are also thought to play a
also contains a range of collagens, predominantly collagen crucial role in regulating disc nutrition (Nachemson et al.
II, although III, V, VI, IX and XI have also been described 1970; Roberts et al. 1996).
(Nerlich et al. 1998; Roughley 2004). The collagen II fibrils The adult human intervertebral disc is both avascular and
appear randomly distributed within the matrix, meaning the aneural, with blood vessels and associated nerves found only
extracellular matrix lacks the structural architecture noted in in the very outer regions of the annulus fibrosus and in the
similar tissues such as articular cartilage (see Chap. 5). While vertebral bodies adjacent to the cartilaginous end plates
the extracellular matrix possesses many of the same compo- (Yasuma et al. 1993; Repanti et al. 1998; Roughley 2004). It
nent molecules as articular cartilage, the ratio of PGs/colla- is these capillaries that provide nutrients to cells within the
gens in the nucleus pulposus is substantially higher, at around disc, through a process of diffusion facilitated by fluid trans-
27:1, compared to only around 2:1 in cartilage (Mwale et al. port that occurs during normal movement. Under compres-
2004). The elevated proteoglycans content of the nucleus sive loading, water is extruded from the nucleus pulposus,
pulposus, in conjunction with the associated high water con- taking metabolic waste products such as lactic acid away
tent, results in a high swelling pressure giving it the ability to from cells towards the blood vessels. As load is briefly alle-
effectively act as a shock absorber and withstand the high viated, the osmotic potential of the nucleus draws water back
compressive forces experienced within the spine (see in, supplying the cells with nutrients such as glucose and
Chap. 2). oxygen. However, as cells in the core of the nucleus pulposus
11 Pathogenesis of Intervertebral Disc Degeneration 179
can be as much as 8 mm from the nearest blood vessel, this Histologically, degeneration can be characterised using a
results in a tissue which is both nutrient and oxygen poor and range of features (Sive et al. 2002). There is progressive loss
with a relatively low pH, primarily due to an accumulation of of demarcation between the nucleus pulposus and annulus
lactic acid (Urban et al. 1982; Katz et al. 1986). This hostile fibrosus with loss of the transition zone. This is due, in part,
environment is reflected in the cell densities of the tissue to a change in collagen synthesis by the nucleus cells from
which are considerably less than in other cartilaginous tis- collagen II to collagen I; a loss of proteoglycan, which results
sues. In early life, cellularity decreases to approximately in dehydration of the nucleus pulposus; presence and extent
4,000 cells/mm3 in the normal adult nucleus pulposus and of fissuring within the nucleus pulposus, which radiate even-
9,000 cells/mm3 in the annulus fibrosus by the time skeletal tually into the annulus fibrosus; and formation of cell clus-
maturity is achieved (Maroudas et al. 1975). There is also a ters due to abnormal cell turnover. Changes also occur in the
reduction in the proportion of large, vacuolated, morphologi- annulus fibrosus, with disruption of the collagen lamellae as
cally distinct notochordal cells within the nucleus pulposus fissures extend, and there is a change in collagen fibre organ-
and an increase in or transition to smaller mature nucleus isation with fibres bifurcating and interdigitating (Lyons
cells, which are thought to have a lower metabolic activity et al. 1981). The poor repair capacity of the disc means that
than notochordal cells (Guehring et al. 2008). Cell metabo- in the late stages of degeneration, the nucleus pulposus is
lism within the disc is also thought to be relatively low, due replaced by disorganised scar and granulation tissue, and
in part to the low pH and oxygen concentration, with cells repair of the annulus fibrosus results in scaring and neovas-
generating ATP predominantly through glycolysis (Urban cularisation (Peng et al. 2006). Vascular ingrowth extends
et al. 2004). eventually into the nucleus pulposus and is associated with
However, while cell number and metabolic activity are innervation into the disc causing discogenic pain (Freemont
both low, the resident cells are responsible for homeostatic et al. 1997, 2002). The decrease in water content and increas-
turnover of the extracellular matrix, producing catabolic fac- ing fibrous nature of the nucleus pulposus cause the disc nar-
tors and degradative enzymes, as well as anabolic growth rowing observed radiographically. There is also an increase
factors and new matrix proteins. This process is tightly con- in collagen cross-linking with tissue sugars, making the disc
trolled and any imbalance in degradative and synthetic pro- stiffer, more difficult to repair and more easily injured
cesses can lead to a matrix breakdown and loss of tissue (Hormel and Eyre 1991; Duance et al. 1998; Pokharna and
integrity. While the exact reasons for this imbalance are not Phillips 1998; Wagner et al. 2006). Importantly, this reduced
fully understood, research over recent years has shed light on disc height significantly alters the biomechanics of the spinal
the processes which contribute to, or are potentially respon- motion segment, with decompression of the nucleus pulpo-
sible for, the tissue breakdown observed during interverte- sus and removal of stress within fibres of the annulus fibrosus
bral disc degeneration. that ultimately causes spinal instability and leaves the disc
less able to resist forces experienced during motions such as
bending (Zhao et al. 2005; Adams and Roughley 2006).
11.3 Intervertebral Disc Degeneration Increased pressure is also placed on the neural arch and this
can result in non-discogenic nerve pain during movement
Disc degeneration is characterised by an overall breakdown (Pollintine et al. 2004).
of extracellular matrix, combined with altered matrix synthe- Although the anatomical and morphological features of
sis and changes in resident cell number, cell phenotype and degeneration have been well documented, the underlying
behaviour. While many of these features are evident during cellular and pathophysiological changes occurring during
normal ageing, changes are accelerated in degeneration and degeneration have not been thoroughly described. However,
are associated with discogenic pain or pain caused by spinal in an attempt to identify novel therapies, recent widespread
instability and impingement of nerve roots in the spine. interest in the elucidating mechanisms underlying degenera-
tion has resulted in a more thorough understanding of the
pathogenesis of degenerative disc disease.
11.3.1 Morphological Features of Degeneration
Degeneration is routinely characterised radiographically 11.3.2 Genetic Inuences

(Antoniou et al. 1998; Pfirrmann et al. 2001), with the
Thompson grading system used to classify gross morpho- Age and environmental factors, such as smoking, vibration,
logical changes (Thompson et al. 1990). This system excessive heavy loading and localised injury, have all been
describes the decreasing water content within the nucleus proposed as risk factors (Holm and Nachemson 1988; Hirano
pulposus, combined with disc narrowing and bulging and et al. 1988; Deyo and Bass 1989; Wilder et al. 1996; Adams
eventually osteophyte formation and end-plate sclerosis. et al. 1999, 2000). Genetic and hereditary factors are
considered to play a dominant role in predisposing individu- The induction of collagen X during late-stage degeneration is
als to disc degeneration and back pain. A familial survey by suggested to be a cellular response to enhance oxidative stress
Postacchini et al. found that in the group of individuals with and is thought to signify nucleus pulposus cell hypertrophy as
discogenic LBP, 35 % had at least 1 family member with a it is often accompanied by increased expression of Runt-
history of discogenic LBP and 5 % had one or two members related transcription factor 2 (Runx2), osteoprotegerin and
who had undergone disc surgery (Postacchini et al. 1988). alkaline phosphatase in areas of calcification (Boos et al. 1997;
This compared to 12 and 1 % respectively in the asymptom- Nerlich et al. 1997; Rutges et al. 2010).
atic cohort. Further studies, including identical and noniden- Collagen cross-linking also changes during degeneration,
tical twin studies, have also shown strong familial links in with a decrease in pyridinoline cross-links which give stabil-
discogenic LBP predisposition (Richardson et al. 1997; ity to collagen fibrils, especially in the nucleus pulposus
MacGregor et al. 2004; Frino et al. 2006). where the collagen fibres are less densely packed than the
In addition to familial studies, a growing body of research annulus fibrosus. There is also an increase in nonenzymatic
has investigated genetic associations with degeneration. The glycosylation, which causes cross-linking of matrix proteins,
genes for collagen I (COL1A1), IX (COL9A2 and COL9A3), with an increase in pentosidine during degeneration being
XI (COL11A2), aggrecan, MMP-3, IL-1, IL-6, vitamin D one marker of this process. This increase in advanced glyca-
receptor (VDR), cartilage intermediate layer protein (CILP) tion end products (AGEs) within the disc during both natural
and hyaluronan and proteoglycan link protein 1 (HAPLN1) ageing and degeneration has been demonstrated to cause tis-
have all been associated with disc degeneration (Videman sue stiffness, particularly in the annulus fibrosus, and may
et al. 1998; Annunen et al. 1999; Takahashi et al. 2001; make the disc more susceptible to mechanical damage dur-
Kawaguchi et al. 2002; Pluijm et al. 2004; Solovieva et al. ing degeneration (Hormel and Eyre 1991; Duance et al.
2004, 2006; Seki et al. 2005; Kawakami et al. 2005; Roughley 1998; Pokharna and Phillips 1998; Wagner et al. 2006;
et al. 2006). However, to date, in different ethnic groups, Adams et al. 2010).
only COL1A1, COL9A2, MMP-3 and VDR polymorphisms Combined with the changes in collagen expression, there
have been shown reproducibly to be disease associated. This is a decrease in the proteoglycans content of the disc (Pearce
topic is developed further in Chap. 10. In order to prove links et al. 1987; Inkinen et al. 1998; Cs-Szabo et al. 2002;
between gene polymorphisms and disc degeneration, then Sztrolovics et al. 2002). While degenerate nucleus pulposus
more detailed and large-scale linkage studies on families cells are capable of synthesising aggrecan, versican synthe-
with members who are predisposed to early onset degenera- sis increases, as does production of biglycan and decorin
tion are required. If genetic predisposition and association (Lyons et al. 1981; Buckwalter 1995; Inkinen et al. 1998; Le
with single gene polymorphisms can be established, this may Maitre et al. 2007d). In contrast, there is increased degrada-
lead to the development of diagnostic tools to screen disc tion of aggrecan and versican fragmentation thereby reduc-
degeneration predisposition. However, these studies may ing disc overall proteoglycans content. Several versican
also reveal that disc degeneration is a complex, multifacto- isoforms have been identified in the disc with varying molec-
rial, oliogenic disorder for which a clear predisposition is ular weights (Sztrolovics et al. 2002). However, one of the
difficult to detect. key features of all isoforms is that they contain fewer chon-
droitin sulphate side chains than aggrecan and hence have a
lower negative charge which reduces the osmotic potential of
11.3.3 Alterations in Extracellular Matrix versican-containing aggregates. Therefore, the overall loss
Composition of aggrecan, combined with the shift to versican production,
reduces the water content in the disc, and the shift in colla-
While many of the matrix changes evident during degenera- gen production to collagen I results in a more fibrous tissue,
tion are due to increased matrix catabolism by degradative less capable of withstanding load.
enzymes (as will be discussed in the next section), there are There are other changes evident in the disc, caused either
also alterations in the synthesis and distribution of matrix as a result of degeneration or as an attempt at repair. One
components. In the early stages of degeneration, there is an example is the increase in fibronectin and importantly
increase in expression of collagen II, thought to be an attempted fibronectin fragments during degeneration (Oegema et al.
repair mechanism (Takaishi et al. 1997). However, with 2000). Fibronectin is a large extracellular glycoprotein that
advancing degeneration, there is a general decrease in colla- contains binding sites for several cell membrane and matrix
gen II synthesis and a shift to collagen I production by nucleus proteins, including integrins and collagens, respectively, and
pulposus and inner annulus fibrosus cells (Buckwalter 1995; is thought to play a role in extracellular matrix organisation.
Schollmeier et al. 2000; Le Maitre et al. 2007d). Collagen X Its expression increases in degeneration, although there is
has been identified in the disc during advanced degeneration, also an increase in fibronectin fragments which have been
particularly around clefts and cell clusters (Boos et al. 1997). demonstrated in vitro to stimulate matrix metalloprotein
(MMP) production and suppress aggrecan synthesis aged and diseased discs, with a correlation being shown
(Anderson et al. 2005; Aota et al. 2005) and in vivo to stimu- between increasing grade of degeneration and an increase in
late disc degeneration (Greg et al. 2003). One mechanism for the presence of aggrecanase-generated fragments (Sztrolovics
this change may be the stimulation of catabolic cytokine et al. 1997; Roberts et al. 2000). Studies initially identified
expression, which has been demonstrated to occur following ADAMTS 4 (aggrecanase 1) in the disc (Le Maitre et al.
the addition of fibronectin fragments to cartilage explants 2004; Hatano et al. 2006), with a correlation noted between
in vitro (Homandberg et al. 1997). ADAMTS 4, but not ADAMTS 5 (aggrecanase 2), expres-
sion with degeneration (Patel et al. 2007). Subsequently, we
have shown that the expression of ADAMTSs 1, 4, 5, 9 and
11.3.4 Matrix Degradation 15 are all increased with intervertebral disc degeneration
(Pockert et al. 2009), with their expression potentially being
A range of proteolytic enzymes are responsible for break- regulated by IL-1b (Demircan et al. 2005). Building on this
down of the extracellular matrix, including members of the finding, in vitro stimulation experiments demonstrated
MMP and a disintegrin and metalloproteinase with throm- increased expression of both ADAMTSs 4 and 5 following
bospondin motifs (ADAMTS) families (for details of these IL-1b stimulation, with nitric oxide being the mediating fac-
enzymes, see Chap. 8). MMPs are capable of cleaving the tor (Le Maitre et al. 2005a; Zhao et al. 2011). More recently,
majority of constituents of the disc extracellular matrix. Most ADAMTSs 7 and 12, enzymes which are capable of degrad-
notably, MMPs 1, 8 and 13 degrade intact triple-helical col- ing cartilage oligomeric matrix protein (COMP), were both
lagens including collagens I and II, while 2 and 9 are gelati- shown to be upregulated in a rat model of degeneration; how-
nases cleaving partially degraded triple-helical domains ever, their presence in human discs or role in degeneration is
(Nagase and Woessner 1999). yet to be elucidated (Yu and Zhu 2012).
MMPs 1, 2, 3, 7, 8, 9, 10, 13, 19 and 28 have all been Those molecules discussed above, along with a broad
identified within the disc, with levels of many increasing spectrum of other proteolytic enzymes including cathepsins
during degeneration (Roberts et al. 2000; Weiler et al. 2002; D, G, K and L (Konttinen et al. 1999; Ariga et al. 2001), are
Le Maitre et al. 2004, 2006b; Gruber et al. 2005; Richardson responsible for the homeostatic turnover of disc extracellular
et al. 2009; Bachmeier et al. 2009; Klawitter et al. 2011). In matrix. Their expression and activity are closely controlled
particular, the number of cells immunopositive for MMPs 1, by soluble mediators, such as catabolic (pro-inflammatory)
3, 7 and 13 was shown to be increased in degeneration (Le cytokines and anabolic growth factors, and through blocking
Maitre et al. 2004, 2006b). We have also demonstrated a cognate inhibitors, the tissue inhibitors of metalloproteinases
significant increase in the expression of MMP-10 in symp- (TIMPs). The TIMP family consists of four members (TIMPs
tomatic (painful) degenerate discs and shown a correlation 1, 2, 3 and 4) which play various roles, including MMP acti-
between expression of MMP-10 and IL-1 and NGF, but not vation and inhibition and induction of angiogenesis (Brew
TNF-a, suggesting a possible role for MMP-10 in the initia- et al. 2000). However, their main role is the inhibition of
tion of nociception during degeneration (Richardson et al. MMPs and ADAMTSs, which is achieved primarily through
2009). MMP-10 is also capable of activating proMMPs, irreversible non-covalent coupling to active MMPs in a 1:1
including MMPs 1, 7, 8, 9 and 13, and has been shown to be stoichiometric fashion (Cooper et al. 1985; Stetler-Stevenson
capable of superactivating proMMPs 1, 8 and 13, giving et al. 1989). TIMPs 1, 2 and 3 have been identified in the
them a higher-than-normal specific activity and potentially disc, with TIMPs 1 and 2 exhibiting broad specificity of inhi-
shifting the homeostatic balance of activity towards catabo- bition of members of the MMP family, while TIMP 3 appears
lism (Barksby et al. 2006). to selectively inhibit aggrecanases (Kashiwagi et al. 2001).
In addition to their activity against collagens, members of Both TIMP 1 and 2 are upregulated in degeneration (Le
the MMP family also have the ability to degrade aggrecan at Maitre et al. 2004), although they have higher specificity for
discrete sites within the G1-G2 and G2-G3 interglobular certain MMPs than others, suggesting those MMPs (e.g.
domains, although their activity against this substrate is MMP 7) that are resistant to TIMP inhibition may play a
significantly lower than that of the aggrecanase members of greater role in degeneration (Le Maitre et al. 2006b).
the ADAMTS family. The ADAMTS family includes the Conversely, while TIMP 3 expression levels in nondegener-
aggrecanases ADAMTSs 1, 4, 5, 8, 9 and 15, which are all ate human nucleus pulposus cells have been shown to be
capable of degrading aggrecan at sites distinct from those of higher than that of any of the ADAMTSs (Pockert et al.
the MMPs and at an activity substantially higher than that of 2009), its expression does not change in degeneration (Le
the MMPs (Tortorella et al. 1999; Abbaszade et al. 1999; Cal Maitre et al. 2004; Pockert et al. 2009), suggesting a poten-
et al. 2002; Nagase and Kashiwagi 2003; Somerville et al. tial imbalance between active ADAMTSs and TIMP 3 that
2003; Collins-Racie et al. 2004). An increase in aggrecanase- could lead to the matrix, particularly aggrecan, degradation,
generated aggrecan fragments has been identified in both a characteristic of the degenerate disc.
11.4 Vascular and Nerve Ingrowth NGF. Since nerves possess the high-affinity NGF receptor
TrkA, it is likely that there is a vasoregulatory role for the
At birth, both the end plates and annulus fibrosus possess nerve fibres (Freemont et al. 2002). These nerve fibres have
blood vessels, although these soon recede, meaning that, been shown to be positive for protein gene product 9.5
with the exception of the external lamellae of the annulus, (PGP9.5), acetylcholinesterase, neurofilament protein (NFP),
the normal adult intervertebral disc is both avascular and substance P (SP) and calcitonin gene-related peptide (CGRP),
aneural (Yasuma et al. 1993; Repanti et al. 1998; Roughley amongst other proteins. On this basis, it has been suggested
2004; Roberts et al. 2006b). However, during degeneration, that these nerve fibres originate from the dorsal root ganglion
both neovascularisation and innervation occur, with blood (Ashton et al. 1994; Brown et al. 1997; Ohtori et al. 2002;
vessels and nerve fibres infiltrating the annulus fibrosus and Takahashi et al. 2009; Garcia-Cosamalon et al. 2010) and are
eventually the nucleus pulposus (Freemont et al. 1997, 2002; nociceptive. Studies on degenerate intervertebral disc tissues
Coppes et al. 1997; Nerlich et al. 2007). Matrix alterations, have demonstrated a similar protein expression profile for
particularly vascular ingrowth, and increasing angiogenesis nerves infiltrating the inner annulus fibrosus and nucleus pul-
have been correlated with decreasing proteoglycan content posus, suggesting a quantitative increase in nociceptive neu-
in an ovine annular lesion model (Melrose et al. 2002a). rite number, rather than a change in type of neurite (Ashton
Indeed, aggrecan has been shown to be inhibitory to endothe- et al. 1994; Freemont et al. 1997; Brown et al. 1997; Johnson
lial cell adhesion and migration in a concentration-dependant et al. 2002; Melrose et al. 2002a; Takahashi et al. 2009). On
manner (Johnson et al. 2005), a finding that provides a pos- the other hand, there is evidence that suggests there is an
sible mechanism for accelerating the degenerative process. increase in sympathetic afferents in degenerate tissue, which
Extensive capillary networks have also been found to be are hypothesised to play a significant role in low back pain
associated with annular clefts and tears (Nerlich et al. 2007), (Takebayashi et al. 2006). In vitro studies demonstrated a
suggesting the breakdown of the normal disc extracellular concentration-dependant inhibition in nerve fibre outgrowth
matrix is permissive for neovascularisation. However, the by human aggrecan, suggesting the decreases in proteogly-
mechanisms or the factors underlying initiation of angiogen- cans content in the disc during degeneration may permit neu-
esis have not yet been clearly elucidated. Pleiotrophin, a ral ingrowth (Johnson et al. 2002). The same study also
growth factor reported to be involved in cell migration and indicates that deglycosylation of aggrecan may also be impor-
differentiation in various cellular processes, has been impli- tant as enzymatic removal of keratan and chondroitin sulphate
cated as an angiogenic factor in the disc as it has been shown from aggrecan abrogated the inhibitory effect of intact aggre-
that the frequency of pleiotrophin-positive disc cells was can. As the effect was greater following chondroitinase ABC
significantly correlated with the amount of vascularisation than keratanase treatment, it was inferred that there was a
(Johnson et al. 2007). Additionally, other angiogenic factors greater role for chondroitin sulphate than keratan sulphate in
have also been implicated including vascular endothelial inhibiting nerve outgrowth. Since there is an increase in the
growth factor (VEGF) (Ohba et al. 2009), basic fibroblast ratio of keratin sulphate/chondroitin sulphate, the authors also
growth factor (FGF-2), TGF-b and osteonectin (Melrose hypothesised that this change may also be important in terms
et al. 2002b). More recently, it has also been suggested that of allowing nerve infiltration into the disc.
IL-1b is capable of inducing angiogenesis through stimula- Importantly, however, a complex interplay between cata-
tion of the growth factors VEGF, NGF and BDNF, although bolic cytokines, neurotrophins, neurotrophin receptors and
these results were based on immunohistochemical correla- chemorepellant molecules may be responsible for guiding
tion studies rather than direct stimulation with IL-1b (Lee nerve ingrowth during degeneration, in particular sema-
et al. 2011). While the mechanisms underlying vascular phorins, a large family of secreted and membrane-bound
ingrowth are still unclear, neovascularisation is thought to axonal guidance molecules (Kolodkin et al. 1993). Gene and
provide a route for various cytokines and growth factors to protein expression studies of the class 3 semaphorin family
reach the inner disc regions at an accelerated rate than member Sema3A, which can cause axonal collapse when
through the usual route of diffusion (Nerlich et al. 2007), found in high concentrations, identified high levels of cell
which may be one of the driving forces behind the increased immunopositivity in the outer annulus fibrosus of normal tis-
nucleus pulposus cell-derived degradative enzyme produc- sue (Tolofari et al. 2010). The percentage of semaphorin-
tion and accelerated proteoglycan loss in the inner annulus positive cells was found to decrease with increasing grades
fibrosus and nucleus pulposus. of degeneration, particularly in individuals with symptom-
During degeneration, there is also an increase in nerve atic (painful) degeneration, suggesting an important role for
fibres both physically associated with, and distant from, Sema3A in inhibiting nerve ingrowth into normal disc
infiltrating blood vessels. In the majority of these cases, nerve tissue.
fibres are found alongside blood vessels, and during angio- Both nucleus pulposus and annulus fibrosus cells from
genesis, endothelial cells from infiltrating vessels secrete normal discs have been shown to express low levels of the
neurotrophin nerve growth factor (NGF) and brain-derived 11.5 Alterations in Disc Cell Biology
neurotrophic factor (BDNF), and expression levels increase in Degeneration
in degeneration, particularly in individuals with symptom-
atic (painful) disc degeneration (Abe et al. 2007; Purmessur A wide range of factors are thought to be involved in the
et al. 2008; Gruber et al. 2008). Interestingly, these disc initiation and progression of degeneration (Fig. 11.1). While
cells also express the high-affinity NGF receptor TrkA, the individuals may experience discogenic back pain for differ-
high-affinity BDNF receptor TrkB and the low-affinity ent reasons, correlation studies and detailed molecular and
NGF/BDNF receptor p75NTR, as well as SP, suggesting pos- cellular biology studies suggest that there are four main cat-
sible autocrine signalling by the disc cells themselves egories of factors influencing cell function and hence drive
(Purmessur et al. 2008). However, the predominant role for intervertebral disc degeneration. These include soluble regu-
NGF and BDNF may be to act in a paracrine manner on lators of disc cell function (mainly cytokines and growth fac-
dorsal root ganglion neurons to stimulate nerve ingrowth. tors), nutritional status, cell ageing and death and response to
We have recently demonstrated that coculture of human mechanical load.
nucleus pulposus cells, derived from the degenerate inter-
vertebral disc, with the neural cell line SH-SY5Y cells
caused an increase in both percentage of neurite-expressing 11.5.1 Soluble Regulators of Cellular Function
cells and mean neurite length (Richardson et al. 2011). This
finding supports earlier work by Johnson and colleagues During degeneration, a range of pro-inflammatory cytokines
who showed that normal inhibition of neurite outgrowth by and inflammatory mediators are increased. These include
aggrecan could be prevented by cells derived from degener- members of the interleukin family, including IL-1, IL-2,
ate disc, suggesting that such cells release neurotrophins IL-6, IL-12 and IL-17, as well as interferon gamma (IFN-g),
(Johnson et al. 2006). Indeed, our own studies show that TNF-a and the inflammatory mediators prostaglandin E2
these increases in neurite-expressing cells and neurite (PGE2) and nitric oxide (NOx) (Kang et al. 1996; Olmarker
length could be inhibited by the addition of anti-BDNF and Larsson 1998; Le Maitre et al. 2005a, 2007b; Bachmeier
antibodies. In contrast, when inhibition was activated by et al. 2007; Akyol et al. 2010; Gabr et al. 2011; Studer et al.
anti-NGF antibodies, there was only a decrease in the per- 2011). The pro-inflammatory cytokines are all thought to
centage number of neurite-expressing cells (Richardson play independent roles in matrix catabolism, although inter-
et al. 2011). plays between the molecules have been identified. For exam-
Interestingly, both NGF and BDNF expression can be ple, IL-6 is thought to potentiate the response of nucleus
stimulated by addition of recombinant IL-1b and TNF-a pulposus cells to both IL-1 and TNF-a (Studer et al. 2011),
(cytokines shown to be increased in intervertebral disc while IL-17 synergises with both TNF-a and IFN-g, increas-
degeneration) to cultured human nucleus pulposus cells ing the catabolic activities of human nucleus pulposus and
in vitro, while TNF-a stimulation also induces expression of annulus fibrosus cells and possibly serving as a key regulator
substance P (Purmessur et al. 2008). Such in vitro results of inflammation in the degenerating disc (Gabr et al. 2011).
suggest that these pro-inflammatory cytokines stimulate the However, although there is evidence for the involvement
production of neurotrophins which promotes the growth of of multiple pro-inflammatory cytokines in the pathogenesis
sensory nerve fibres into the intervertebral disc and induce of disc degeneration, the predominant catabolic cytokines
substance P related with pain transmission. Noteworthy, sig- appear to be interleukin-1 (IL-1) and tumour necrosis factor-
nalling of NGF and BDNF through their receptors initiates alpha (TNF-a). While there is no clear consensus on which
activation of a number of pathways, including the NF-kB molecule mediates degeneration, research suggests both are
pathway; activation induces a range of pro-inflammatory fundamentally important in controlling the observed cellular
cytokines which may then perpetuate the cycle leading to and matrix changes.
innervation (Wallach et al. 2002). The expression of NGF
also correlates with expression of specific MMPs (Richardson 11.5.1.1 Interleukin-1
et al. 2009), suggesting a potential role for neurotrophins in Both isoforms of IL-1 (IL-1a and IL-1b) have been identified
driving matrix catabolism, possibly to ease nerve ingrowth within the disc, along with their receptor (IL-1R1), the
through the disc. Importantly, this interaction between cytok- exported decoy receptor (IL-1RII) and their natural inhibitor
ines and neurotrophins is complex and requires further exten- (IL-1 receptor antagonist or IL-1Ra) (Le Maitre et al. 2005a).
sive study before a clear pathway can be elucidated. During degeneration, expression of IL-1a and b and IL1RI
Unfortunately, however, this is hindered by the inability to increases significantly in both the nucleus pulposus and inner
study nerve ingrowth in humans and the potential differences annulus fibrosus. However, IL-1Ra expression does not
between humans and the model animal systems routinely increase and this imbalance leads to an excess of IL-1 iso-
used in disc degeneration research. forms in degenerate tissues. Importantly, in vitro studies have
Genetic Reduced
factors? nutrition?
Environmental Other
factors? factors?
Altered matrix synthesis

type II collagen & aggrecan
type I & X collagens in NP
Degradative enzyme synthesis

Increased
MMPs 1, 3, 7, 10 & 13
IL1
ADAMTs 1, 4, 5, 9 & 15
TNF, TGF, NGF,
BDNF, VEGF, CTGF Cell proliferation,
apoptosis and
senescence
Neovascularisation and innervation

Matrix degradation
Degeneration
Back pain
Fig. 11.1 A schematic overview of the pathogenesis of intervertebral disc degeneration demonstrating the involvement of IL-1 in driving the
aberrant cell biology and processes involved in matrix catabolism and generation of back pain
shown that IL-1 induces a number of cellular and molecular the addition of exogenous IL-1Ra, and application of IL-1Ra
changes associated with disc degeneration. Stimulation of has been proposed as a potential therapeutic intervention to
human nucleus pulposus cells with recombinant IL-1 has inhibit intervertebral disc degeneration (Le Maitre et al.
been shown to induce an upregulation of both MMPs, includ- 2006a, 2007c; Box 11.1).
ing MMPs 3 and 13, and ADAMTSs, including ADAMTS 4,
a shift in collagen expression from II to I and reduction in
aggrecan expression (Le Maitre et al. 2005a). There appear Box 11.1: Interleukin-1 as the Driving Force Behind the
to be differences in the responses of normal and degenerate Pathogenesis of Disc Degeneration
disc cells to IL-1 stimulation, with a more catabolic response While for some TNF-a has been the focus of inves-
in degenerate nucleus pulposus cells compared to normal. tigation as the molecular regulator of disc degenera-
IL-1 stimulation also resulted in significant increases in both tion, our research has focussed on the involvement
IL-1 isoforms by degenerate nucleus pulposus cells and a of IL-1. These studies have demonstrated an increase
decrease in expression by normal nucleus pulposus cells, in the expression of both isoforms (a and b) of IL-1,
suggesting a homeostatic response in nondegenerate cells along with its receptor (IL-1RI) during degeneration.
and an aberrant catabolic response once degeneration has However, no such increase was demonstrated for its
been activated. IL-1 has also been shown to induce both natural inhibitor, IL-1Ra, suggesting an imbalance
angiogenesis (by inducing expression of VEGF) and neuro- that may be responsible for driving the cellular and
nogenesis (via the stimulation of neurotrophic factors) into matrix changes evident during degeneration. These
disc tissue (Lee et al. 2011) and stimulation of apoptosis (Cui roles include inducing expression of both MMPs
et al. 2007; Zhao et al. 2007a). Conversely an inhibition or and ADAMTSs, which are known to catabolise the
reversal of these processes has been demonstrated through
superfamily, are known to be present in the normal interver-

extracellular matrix; reducing matrix component tebral disc, while evidence for their involvement in disease
molecule expression, most notably aggrecan; induc- comes from both in vitro cell stimulation studies and studies
ing apoptosis and senescence of disc cells; and induc- of in vivo models of disc degeneration. Many workers have
ing both angiogenesis and innervation into the disc. focussed on the ability of growth factors to stimulate proteo-
However, the key question at present is: what initiates glycans synthesis by nucleus pulposus and annulus fibrosus
the upregulation of IL-1 that then induces the degen- cells and shown positive results following stimulation with
erative cascade? Elucidation of this mechanism may transforming growth factor-b (TGF-b), epidermal growth
lead to the development of novel therapies or allow factor (EGF), insulin-like growth factor-1 (IGF-1), connec-
disc degeneration to be prevented. tive tissue growth factor (CTGF) and the bone morphoge-
netic proteins (BMPs) 2, 7 (also known as osteogenic
protein-1 or OP-1), 12, 13 (also known as growth and dif-
ferentiation factor 6 (GDF-6) or cartilage-derived morphoge-
11.5.1.2 TNF-a netic protein-2 (CDMP-2)) and 14 (also known as growth
TNF-a, like IL-1, has been shown to be capable of inducing and differentiation factor 5 (GDF-5) or cartilage-derived
neural ingrowths into the degenerate intervertebral disc. morphogenetic protein-1 (CDMP-1)) either alone or in com-
TNF-a has also been implicated in causing nerve root dam- bination (Thompson et al. 1991; Osada et al. 1996; Gruber
age and sciatic pain, with blocking studies supporting this et al. 1997; Masuda et al. 2003; Tim et al. 2003; Imai et al.
theory (Igarashi et al. 2000; Olmarker and Rydevik 2001). 2007a; Gilbertson et al. 2008; Le Maitre et al. 2009). Cell
While evidence for a role for TNF-a in nerve ingrowth is proliferation has also been demonstrated in vitro following
compelling, evidence supporting its role in driving matrix stimulation with TGF-b and IGF-1, while the same growth
catabolism during degeneration is less clear. Although expres- factors and platelet-derived growth factor (PDGF) are all
sion of TNF-a is increased in degenerate tissues (Weiler et al. capable of reducing disc cell apoptosis and secretion of cata-
2005; Bachmeier et al. 2007), other studies have either failed bolic cytokines (Gruber et al. 1997, 2000).
to identify TNF receptor I in degenerate samples or shown In vivo studies also show positive effects of TGF-b, IGF-
that there is no increase in the receptor gene expression in 1, BMPs 2 and 5, fibroblast growth factor-2 (FGF-2) and
degenerate samples. The result suggests that native disc cells OP-1 on cell proliferation, matrix synthesis and restoration
in vivo may not be able to respond to TNF-a (Le Maitre et al. of disc height in models of disc degeneration (Walsh et al.
2007b). However, in studies where recombinant TNF-a was 2004; An et al. 2005; Masuda et al. 2006; Miyamoto et al.
used to stimulate cultured nucleus pulposus cells, there were 2006). Of these, the most widely studied is OP-1, with posi-
increases in expression of MMPs 1, 3, 9 and 13, as well as tive effects noted both in vitro and in vivo. Indeed, OP-1 was
ADAMTSs 4 and 5, although induction of expression of a capable of stimulating proteoglycans and collagen synthesis
number of these enzymes was greater following stimulation by human nucleus pulposus cells following IL-1 and chon-
with IL-1 than with TNF-a (Hoyland et al. 2008). Conversely, droitinase ABC treatment in vitro (Takegami et al. 2005;
in situ zymography studies of normal and degenerate human Imai et al. 2007a) and restoring disc height and matrix deg-
nucleus pulposus tissue treated with either IL-1 or TNF-a radation caused by annular needle injury or chemonucleoly-
indicated there was only an increase in enzyme activity in the sis with chondroitinase ABC (Miyamoto et al. 2006; Imai
IL-1 group and not the TNF-a group. In this study, addition et al. 2007b). OP-1 also reduced the production of aggreca-
of IL-1Ra caused a decrease in enzyme activity that was not nase, MMP-13, substance P, TNF-a and IL-1b, suggesting
evident following addition of anti-TNF. While demonstrating that it may have both anabolic and anti-catabolic effects
an increase in TNF-a in degeneration, the documented low (Chubinskaya et al. 2007). The effect of OP-1 on substance
expression of its receptor on disc cells together with differ- P expression, combined with the interplay between catabolic
ences in findings from the other studies begs the question: cytokines and pain markers or pain modulators, has led to the
what is the target of nucleus pulposus cell secreted TNF-a? hypothesis that application of recombinant growth factors
Does it play a more fundamental role in innervation and may have beneficial effects in terms of pain reduction, as
development of discogenic pain, than in matrix catabolism well as matrix restoration during degeneration (Kawakami
and tissue breakdown? et al. 2005). However, the clinical translation of growth fac-
tor therapies is complicated by their diverse and sometimes
11.5.1.3 Anabolic Growth Factors biphasic roles. While growth factors such as TGF-b and
Growth factors have a number of effects on cells, most nota- CTGF are known to promote extracellular matrix synthesis,
bly the shift in metabolic balance towards anabolism, promo- studies on clinical human samples have linked their over-
tion of cell proliferation and prevention of cell death. A range expression to fibrosis and angiogenesis (Ali et al. 2008; Peng
of growth factors, including many members of the TGF-b et al. 2009). The application of growth factor therapies is
further complicated by possible changes in receptor distribu- distribution throughout the disc: cell number in the outer
tion. While data in this area is limited, studies suggest that annulus (which is closer to capillaries surrounding the tis-
there is no change in the expression levels of major growth sue) is substantially higher than that of the nucleus which
factor receptors, including TGFbRII, BMPRII, FGFR3 and can be as far as 8 mm from the nearest capillary.
IGFRI (Le Maitre et al. 2005b; Peng et al. 2006). However, With age, the nutrient supply to the disc is reduced in part
expression of growth factor receptors, including TGFRII, due to reductions in both the density and integrity of capillar-
FGFR3, IGFRI and VEGF receptors I and II, on ingrowing ies in the vertebral bodies and in part due to calcification of
blood vessels (Haro et al. 2002; Le Maitre et al. 2005b) and the cartilaginous end plates (Bernick and Cailliet 1982;
in granulation tissue in painful degenerate discs (Peng et al. Roberts et al. 1996). While it is currently unclear whether
2006) suggests that the use of growth factors should care- cartilaginous end-plate calcification is causative of, or the
fully be controlled to avoid stimulation of unwanted events result of, disc degeneration, it is thought to play an important
such as angiogenesis. Therefore, a more detailed understand- role in disease progression by posing a significant barrier to
ing of their expression profiles and roles in both disc degen- diffusion of solutes into and out of the disc. However, recent
eration and repair are currently required. mCT studies on graded normal and degenerate human sam-
Another area where growth factors show potential is in ples have suggested that, contrary to popular belief, porosity
the stimulation of adult mesenchymal stem cell (MSC) dif- in vertebral end plates increases in degeneration by as much
ferentiation. Previous studies have shown that TGF-b, along as 130 %, while trabecular thickness decreases by as much as
with members of the BMP family, most notably BMPs 2 and 50 % (Rodriguez et al. 2011, 2012). One result of these
14, can stimulate MSC differentiation towards nucleus pul- changes appeared to be an increase in cell proliferation and
posus-like cells in vitro (Stoyanov et al. 2011; McCanless decrease in proteoglycan content in the nucleus pulposus.
et al. 2011). Given the increasing information on the nucleus While solute transport was not assessed in this study, the
pulposus phenotype, further studies will no doubt shed light authors proposed that ischemic cell changes in degeneration
on the role that growth factors play in MSC differentiation may reflect capillary transport activity rather than a decrease
and may lead to combined regenerative cell/growth factor in end-plate permeability.
therapies for treatment of disc degeneration. Whatever its cause, a reduction in essential nutrients is
thought to drive the progression of degeneration. Given that
the main energy-generating pathway in disc cells, even in the
11.6 Changes in Disc Nutrition and Oxygen presence of oxygen, is glycolysis (Holm et al. 1981; Ishihara
Tension and Urban 1999), which requires principally glucose and
produces lactic acid, free diffusion of solutes to and from the
As mentioned previously, nutrient supply to the disc is pre- cells is essential. Evidence suggests that reduction in glucose
dominantly from blood vessels in the vertebral bodies and concentration below 0.5 mmol/L, even for a relatively short
occurs via diffusion through the cartilaginous end plates. period, can cause cell death. Likewise, reduction in the pH to
This theory has been well studied and confirmed using a below 6.4 can also promote death (Horner and Urban 2001;
range of tracer diffusion experiments, including MRI con- Bibby and Urban 2004), while less severe reductions in pH
trast media, fluorescent and radioactive tracers and gaseous can impact cell metabolism (Ohshima and Urban 1992).
tracers such as nitrous oxide (Brodin 1955; Holm and Evidence from a range of assays, including biochemical
Nachemson 1982, 1983; Adams and Hutton 1986; Urban assays and microelectrode measurements, shows that this
et al. 2001). Exposure to cigarette smoke has also been shown drop in glucose concentration and pH is similar to that
to inhibit transport of oxygen into, and lactic acid out of, the observed in degenerate discs. Therefore, while cell activity
disc in animal experiments due to constriction of the micro- and even viability may be impaired by a decrease in nutrient
vasculature in the vertebrae (Holm and Nachemson 1988); supply, enzyme activity is not reduced. In this case, there
this finding supports epidemiological evidence linking smok- would be an imbalance between matrix anabolism and catab-
ing to disc degeneration in humans. The size and charge of olism which may contribute to an elevation in matrix degra-
solutes also affect their penetration into the disc, with anions dation in degenerate discs.
showing lower rates of diffusion than cations due to the poly- The other key metabolite in the disc is oxygen, although
anionic nature of the disc (Urban et al. 2004), while larger its role is less clear than that of glucose. Oxygen levels vary
molecules such as albumin are effectively prevented from widely in human discs, with no clear correlation between a
diffusing through the cartilaginous end plate (Urban et al. change in oxygen tension and disc degeneration. However,
2004). Compared to uni- or bivalent electrolytes, and the fact there exists a steep oxygen gradient within discs studies in
that glucose is a relatively large molecule, Urban et al. opined dogs demonstrating a decrease from 8 to 10 % O2 at the
that diffusion into and through the disc may be slow (Urban disc-vertebral body interface to 0.30.5 % in the centre of the
et al. 2004). This lack of nutrient supply to the core of the nucleus pulposus (Holm et al. 1981). Human discs show
nucleus pulposus is reflected in the non-uniform cell similar trends, with readings as low as 0.7 % O2 in the core
of the degenerate human nucleus (Bartels et al. 1998). While 10
Relative gene expression normalised

oxygen is consumed by disc cells, relatively little CO2 is pro-
duced and disc cells can survive for at least 2 weeks without
oxygen (Horner and Urban 2001), suggesting oxidative 1
phosphorylation is not the primary mechanism for energy
to GAPDH
metabolism. Further studies on both canine and bovine disc
cells demonstrated that as O2 concentration decreases from *
*
21 to 1 %, there was a decrease in oxygen consumption of 0.1
around 75 % and a sharp increase in lactate production
(Ishihara and Urban 1999). This suggests a positive Pasteur
effect since glycolysis is stimulated under hypoxic condi-
tions. However, studies on the effect of hypoxia on disc cells 0.01
CAIX CAXII
have also demonstrated cellular inactivity (Horner and Urban
Mild Moderate Severe
2001) and a loss of matrix synthesis (Ishihara and Urban
1999) below an oxygen concentration of around 5 %. As the Fig. 11.2 Histogram illustrating real-time quantitative PCR data
increase in lactate production under hypoxic conditions is demonstrating a decrease in expression of carbonic anhydrase 9 (CAIX)
likely to reduce the pH, this in turn would reduce matrix syn- and 12 (CAXII) in nucleus pulposus cells during disc degeneration. The
expression of CAIX is significantly decreased with disease progression,
thesis (Ohshima and Urban 1992) and eventually cause cell while a decrease is also seen for CAXII. The carbonic anhydrases play
death (Horner and Urban 2001; Bibby and Urban 2004). a role in regulating intracellular pH, and a downregulation in their
Hence, the effects of low oxygen and low pH are likely to be expression may result in cells being incapable of withstanding the harsh
cumulative and a destructive influence of disc cell survival. physicochemical environment of the degenerate intervertebral disc
The mechanism by which cells sense oxygen and accommo-
date to the oxemic state is discussed in exhaustive detail in the introduction of too many cells, or cells which are unable
Chap. 6. to withstand the conditions of the microenvironmental niche,
Confounding the problem of evaluating the consequences may exacerbate the degenerative problem rather than provide
of oxygen and nutrient limitation is the fact that human tis- a cure.
sues are different from those of animal models. Therefore,
investigators have increasingly turned to mathematical finite
element modelling to elucidate the effect of nutrient limita- 11.7 Cell Ageing and Death
tion on the disc microenvironment and cellular metabolism.
However, since all of the variables involved in disc nutrition During development, the nucleus pulposus is populated by
are poorly understood, the studies published in this area are large, vacuolated, metabolically active and morphologically
relatively simple, dealing with only one or two aspects of distinct notochordal cells which produce high levels of pro-
nutrient supply. Nevertheless, they all suggest that limited teoglycans (Boos et al. 2002; Cappello et al. 2006). However,
nutrient supply affects disc cell viability and metabolic activ- by the age of 10, this population has been replaced by smaller,
ity (Selard et al. 2003; Yao and Gu 2006; Mokhbi et al. 2009; chondrocyte-like nucleus pulposus cells which are less meta-
Malandrino et al. 2011; Jackson et al. 2011). bolically active (Wolfe et al. 1965; Pazzaglia et al. 1989;
As with findings from finite element modelling, in vitro Boos et al. 2002; Guehring et al. 2008). The period between
studies appear to confirm that during degeneration, glucose 3 and 10 years of age, during which identifiable notochordal
limitation and decrease in pH are the predominant factors cells disappear, is also the time during which there is a high
affecting cell metabolism. While nucleus pulposus cells have level of cell death (Boos et al. 2002). These changes appear
mechanisms to regulate intracellular pH, such as the expres- to signal the initiation of a transition from a highly hydrated,
sion of the carbonic anhydrases 9 and 12 (Minogue et al. gelatinous extracellular matrix to a more fibrous, cartilagi-
2010a), our preliminary data indicates that the expression of nous nucleus pulposus seen in adults. It also coincides with
these molecules decreases with degeneration, suggesting that the earliest identifiable signs of degeneration seen in MRI
cells may not be capable of withstanding the low pH over studies. In addition, histological studies have identified
extended periods (Fig. 11.2). Likewise, changes in the matrix changes in this age group, which progress throughout
nucleus pulposus membrane glucose transporters (GLUTs), adult life (Boos et al. 2002). Although these changes may be
present in degenerative tissues, suggest that there are molec- considered to be part of the normal ageing process, as dis-
ular adaptations to compensate for the reduced glucose con- cussed previously, the accelerated matrix degradation sug-
centrations (Richardson et al. 2008b). However, the harsh gests that in disc degeneration, there is premature ageing.
environment of the degenerate disc is likely to be detrimental Several studies have identified necrotic cells within the
not just to resident cells but also to cells that are introduced disc which increase both with age and degeneration;
into the disc. Accordingly, for future cell-based therapies, however, more recently, apoptosis has been identified as the
principle mechanism of cell death. Evidence for this has 1998; Ha et al. 2006), although this would be an overesti-
come from a number of studies, using a diverse range of mate since it would result in the de-cellularisation of the disc
markers including transferase-mediated dUTP nick-end within days (Alvarez and Ortiz 1999). Conversely, studies
labelling (TUNEL) staining (Gruber and Hanley 1998; Lotz using fluorescent cell viability assays on fresh disc tissue
and Chin 2000; Rannou et al. 2004; Kim et al. 2005; Risbud have repeatedly demonstrated 60 % to over 90 % cell viabil-
et al. 2005; Heyde et al. 2006; Park et al. 2006; Loreto et al. ity even in degenerate and scoliotic discs, further suggesting
2011), annexinV-propidium iodide flow cytometry (Rannou that the apoptosis measurements are an overestimation
et al. 2004; Risbud et al. 2005; Park et al. 2006), caspase (Bibby et al. 2002; Johnson and Roberts 2007). However,
activity analysis (Rannou et al. 2004; Heyde et al. 2006; Park while no causative link has been found between apoptosis
et al. 2006; Tschoeke et al. 2008) and gene expression stud- and increased extracellular matrix degradation, it is probable
ies of apoptosis-related markers such as Bax and Bcl-2 that programmed cell death plays an essential role in the
(Heyde et al. 2006; Tschoeke et al. 2008; Loreto et al. 2011). pathogenesis of disc disease.
The method employed to establish the incidence of apoptosis Autophagy, a pathway that may lead to programmed cell
appears to influence findings both in vivo and in vitro. Serum death, has recently been identified in the rat nucleus pulpo-
deprivation resulted in only 1 % cell apoptosis when studied sus (Ye et al. 2011) and annulus fibrosus (Shen et al. 2011).
using TUNEL staining (Gruber et al. 2000); in contrast, sim- We have also demonstrated evidence of autophagy in tis-
ilar conditions increased in the incidence of apoptosis in rat sues of the degenerate human intervertebral disc, through
annulus fibrosus cells to 56 % when annexinV-propidium immunostaining for the key markers LC-3 and beclin-1
iodide flow cytometry was used (Risbud et al. 2005). Other (Fig. 11.3). While autophagy can lead to cell death, it can
studies have also suggested that incidence of apoptosis in also maintain cell viability during periods of environmental
disc cells can be as high as 7374 % (Gruber and Hanley or nutritional stress by catabolism of intracellular
a
100
Non-degenerate
80 Mild
Percentage cell positivity
Moderate
60
Severe
40
Fig. 11.3 Evidence for the
presence of autophagy markers in
normal and degenerate human 20
IVD cells. Immunohistochemistry
for LC-3 (ac) and beclin-1 (df)
demonstrated staining for both
markers in normal (b, e) and 0
degenerate (c, f) nucleus pulposus NP IAF OAF
cells, with strong immunopositiv-
ity in degenerate cell clusters.
Semi-quantitative analysis of cell
b c
positivity demonstrated significant
increases in both LC-3 and
beclin-1 in the nucleus pulposus
with progression of degeneration
(a, d) and in the inner annulus
fibrosus (IAF), but not outer
annulus fibrosus (OAF), for
beclin-1. Although preliminary
data, these findings suggest a role
for autophagy in disc cells and that
there may be an increase in the
number of autophagic cells in
intervertebral disc degeneration
Fig.11.3 (continued) d 100
Non-degenerate
80 Mild
Moderate
Percentage cell positivity

Severe
60

40

20
0
NP IAF OAF
e f
organelles and unfolded or damaged proteins. Therefore, tion in the inner annulus fibrosus in association with regional
more work is required to establish the role of autophagy in neovascularisation (Beard et al. 1981). Although signs of
disc degeneration. cell death are seldom seen, markers of cell proliferation,
Studies on intervertebral disc cell number suggest that most notably proliferating cell nuclear antigen (PCNA) and
rather than a steady decline in cell numbers over time, there the proliferation-associated Ki-67, have been identified
is a cyclical pattern of cell death followed by a period of within these cell clusters (Johnson et al. 2001; Zhao et al.
proliferation. Indeed between 11 and 16 years, following the 2007b).
extensive cell death noted between 3 and 10 years, there is A number of potential theories have been proposed to
notable cell proliferation. The authors hypothesise that this is explain this finding. The increased local nutrition may enable
a compensatory mechanism in response to the early matrix cells within clusters to resist apoptosis, or possibly dead cells
changes seen during this time period (Boos et al. 2002). Cell may be cleared more quickly from clusters than from the rest
clustering and proliferation, a characteristic phenomenon of of the disc. In this respect, other workers have identified
degenerating tissue, is seen in the nucleus pulposus; both morphologically nucleus pulposus-like but CD68-positive
parameters increase both with age and with stage of degen- cells within cells cluster in the nucleus pulposus of degener-
eration (Boos et al. 2002). Both histological parameters are ate discs (Nerlich et al. 2002). These cells have the ability to
routinely identified adjacent to clefts and tears within the phagocytose apoptotic bodies and are thought to be trans-
tissue, possibly due to a localised increase in the local nutri- formed nucleus pulposus cells, rather than infiltrating cells.
ent supply in these regions (compared to an overall decrease In vitro studies have also demonstrated the ability of bovine
in nutrients within the disc) rather than as a reparative mech- nucleus pulposus cells to phagocytose apoptotic cells, sug-
anism (Beard et al. 1981; Boos et al. 2002; Zhao et al. 2007b). gesting that cell clearance from the disc may be undertaken
This hypothesis is further supported by evidence obtained by resident disc cells rather than infiltrating macrophages or
from scoliotic discs which exhibit an increase in cell popula- monocytes (Jones et al. 2008).
Within clusters, increased proliferation predisposes the

resident cells to replicative senescence. Several studies have Box 11.2: Elucidation of Cellular Phenotype Is Central to
identified markers of cellular senescence in degenerate discs Improving Understanding of the Degenerative Process
(Roberts et al. 2006a; Gruber et al. 2007), and in 2007, a link and Development of Novel Therapies
between accelerated cell senescence and degeneration was The adult nucleus pulposus is populated by cells rou-
identified. Through both gene and protein expression stud- tinely described as chondrocyte-like based on their
ies, Le Maitre et al. (2007a) showed that there was an increase rounded morphology and expression of SOX-9, type II
in expression of p16INK4A, a cell cycle inhibitor which is collagen and aggrecan, although controversy has long
upregulated in senescence, a decrease in both mean telomere surrounded their origin and exact phenotype. However,
length and replicative potential and an increase in senes- recent microarray studies from our group and others
cence-associated b-gal staining. While similar trends were have begun to elucidate the true phenotype of these
identified with ageing, these senescence changes were posi- cells. They reveal interesting gene signatures, the
tively related to stage of degeneration, irrespective of age, significance of which are yet to be fully understood.
clearly demonstrating a role for senescence in degeneration. However, the expression of some of these genes (such
Such findings are supported by those of Kim and colleagues as FOXF1) lends weight to the growing body of evi-
who also demonstrated an accumulation of senescent cells in dence from developmental biology studies and other
degenerate discs. These workers reported that the telomere- sources that the adult human nucleus pulposus is popu-
based p53-p21-pRB pathway plays an important role in lated, at least in part, by notochordally derived cells.
inducing senescence in nucleus pulposus cells (Kim et al. Notochordal cells produce higher levels of proteogly-
2009). In addition to replicative senescence, caveolin-1, a cans than mature disc cells and animals which retain
marker of stress-induced premature senescence (SIPS), has their notochordal cells have a gelatinous nucleus pul-
also been identified in the disc. Its expression is correlated posus which does not show signs of degeneration. How
with increases in p16INK4A, but not with age, suggesting a the novel marker genes change during degeneration
potential role for SIPS in degeneration (Heathfield et al. and their potential role in the pathogenesis of disease
2008). Interestingly, various stressors such as reactive oxy- has yet to be fully elucidated. However, the elucidation
gen species (Homma et al. 1994; Chen et al. 1995), mechani- of the nucleus pulposus phenotype has important
cal loading (Martin et al. 2004) and the presence of cytokines implications for the development of novel stem cell-
such as IL-1 (Dai et al. 2006), all thought to play a role in the based regenerative medicine therapies as it allows
processes leading to disc degeneration, have been reported to researchers to understand the end point of differentia-
induce SIPS, suggesting that this type of senescence may tion and avoid generation of chondrocyte-like cells
contribute to disc degeneration. However, more work is which may not produce a correctly functioning extra-
required to confirm these links. cellular matrix.
Expression of cell senescence markers in the study by Le
Maitre et al. was also shown to be correlated with changes in
expression of two proteolytic enzymes, MMP 13 and
ADAMTS 5, which are known to be upregulated in degener- 11.8 Response to Mechanical Load
ate tissue. This finding indicated that a link may exist between
senescence and induction of matrix catabolism (Le Maitre The human intervertebral disc is exposed to a number of
et al. 2007a). As well as inhibiting proliferation, studies in physical stresses, including compressive loading, which
other tissues show that senescent cells adopt an altered phe- predominantly affects the nucleus pulposus, and stretch,
notype, described as the senescence-associated secretory phe- shear and torsion which mainly affect the annulus fibrosus.
notype (SASP) (Freund et al. 2011). Such cells secrete a range Using pressure-sensitive needles inserted into the nucleus
of pro-ageing and catabolic factors, most notably IL-1, which pulposus, Nachemson et al. demonstrated that loads expe-
are found to be elevated in the degenerate intervertebral disc. rienced within the human disc ranged from 250N when
Studies on senescent chondrocytes show similar increases in lying down to 1900N when lifting a 10 kg weight with a
cytokines, as well as MMPs (including MMP 13) and other bent spine (Nachemson 1981). However, further analysis
proteolytic enzymes. Likewise, these cells exhibit a decreased of this data, taking into consideration the tensile forces
response to anabolic stimuli that is characteristic of cells from exerted by muscles in the back, suggests that loading may
degenerate discs. While more work is clearly needed to eluci- be as high as 9000N when lifting. It has been suggested
date a potential causative link between senescence and disc that this combination of compression and flexion is respon-
degeneration, it is clear that the increase in senescence in sible for disc prolapse, with over-flexion combined with
degeneration and the phenotype adopted by the senescent cell moderate load being more detrimental than excessive load
suggest that this process may be important in disease with moderate flexion (Hutton and Adams 1982). Following
progression. on from Nachemsons research, Wilke et al. demonstrated a
load of between 0.1 MPa when lying prone and 2.3 MPa tion. For a further discussion of the effects of force on cells,
when lifting a 20 kg weight with a flexed spine (Wilke et al. see Chap.7.
1999); however, this study again did not take into consider-
ation the tensile forces exerted by the musculature in
the back. 11.9 Clinical Implications: Relevance of
The investigations mentioned above indicate that cells in Understanding the Cell Biology and
the disc experience substantial loads, and predictably, these Pathogenesis of Intervertebral Disc
loads are thought to have a profound effect on cell behaviour. Degeneration for Development of Novel
Indeed, mechanical-loading studies using in vivo animal Therapeutic Agents
models and in vitro cell culture techniques have demon-
strated that the type, magnitude, frequency and duration of Current clinical interventions for back pain are predomi-
loading are paramount in determining cell response (MacLean nantly aimed at relieving symptoms rather than treating the
et al. 2004, 2005; Wang et al. 2007; Wuertz et al. 2009; underlying disorder. In many cases, this involves regular
Korecki et al. 2009; Sowa et al. 2011). In particular, while administration of pain-relieving pharmaceuticals, e.g. non-
moderate loads and low-frequency loading both promote steroidal anti-inflammatories, or application of more novel
anabolic responses, high-magnitude, high-frequency as well therapies, such as transcutaneous electrical nerve stimula-
as sustained static loads all elicit a catabolic or anti-anabolic tion, physical manipulation, exercise therapy or behavioural
response in disc cells. Using human intervertebral disc cells, therapies (Bogduk 2004). However, despite widespread use,
Neidlinger-Wilke et al. reported increased matrix protein the efficacy of these interventions is still questionable (van
expression (type I collagen and aggrecan), but no change in der Roer et al. 2005). In patients who are unresponsive to
the expression of matrix-degrading enzymes (MMPs 1, 2, 3 conservative therapies, but have identifiable imaging deficits
and 13) after low-magnitude compressive loading; high- and clinical symptoms of back pain, fusion surgery is the
magnitude loads led to decreased matrix protein expression ultimate end point (Errico 2005). This approach removes the
with increased matrix-degrading enzyme expression (mainly source of pain, but due to alterations in spinal biomechanics,
MMP 3) (Neidlinger-Wilke et al. 2006). Similarly, Handa it reduces mobility and can cause problems at adjacent
et al. found that load influenced proteoglycan synthesis and motion segments (Hilibrand and Robbins 2004). One alter-
MMP gene expression in human nucleus explants, with low native to spinal fusion is whole intervertebral disc or nucleus
loads promoting matrix anabolism and high loads leading to pulposus transplantation, using either autologous or alloge-
matrix catabolism (Handa et al. 1997). neic tissues (Katsuura and Hukuda 1994; Luk et al. 1997).
While the effect of mechanical load on nondegenerate These procedures have been successfully performed, but
disc cells has been well documented, studies into the response require complicated surgery, and issues have arisen regard-
of degenerate nucleus pulposus cells remain limited. To ing loss of tissue integrity, tissue instability and immunoge-
address this need, recent investigations have compared load- nicity (Alini et al. 2002). While tissue transplantation does
ing responses of nondegenerate and degenerate human not seem a feasible alternative, an increasing range of disc
nucleus pulposus. With physiological loads, Le Maitre et al. replacement devices are currently being investigated. These
showed that nondegenerate nucleus pulposus cells produced include devices such as the prosthetic disc nucleus (PDN)
an anabolic response, while degenerate cells remained unre- device or whole-disc replacements such as Charite and
sponsive (Le Maitre et al. 2008). This finding suggests an ProDisc (Jin et al. 2003; Guyer et al. 2009; Delamarter et al.
alteration in mechanotransduction pathways between normal 2011). Large-scale trials are ongoing with these devices, and
and degenerate cells. In support of this notion, Gilbert et al. with the whole-disc replacement, a significant reduction in
showed differences in response to cyclic tensile strain pain score has been demonstrated. However, complications
between nondegenerate and degenerate human annulus including device migration, extrusion and failure are all
fibrosus cells (Gilbert et al. 2010, 2011). Accordingly, while issues and studies have so far failed to show improved out-
cyclic tensile strain applied at 1 Hz to nondegenerate annulus comes compared to fusion (Errico 2005; Di et al. 2005;
fibrosus cells resulted in a decrease in catabolic gene expres- Lindley et al. 2010). For a more detailed discussion of these
sion, the same strain caused a decrease in anabolic gene devices, see Chaps. 13 and 14.
expression by degenerate annulus fibrosus cells. In the latter This lack of clinically successful long-term treatment for
cells, there was evidence of an altered mechanotransduction discogenic back pain has led researchers to investigate both
pathway which appeared to be independent of cytokine biological modulators of disc cell function and novel cell-
involvement (Gilbert et al. 2010, 2011). While the implica- based tissue engineering and regenerative medicine
tions of these changes require further elucidation, the pro- therapies.
found effects of mechanical forces on cell behaviour cannot In line with the increase in knowledge surrounding the
be ignored, and the possibility exists that force plays a funda- control of disc matrix anabolism and tissue degradation
mental role in the initiation or progression of disc degenera- during degeneration, the utility of a number of biologically
active agents has been evaluated. These modulators include To aid cell survival and function following implantation,
cytokine inhibitors, such as IL-1Ra, to inhibit matrix deg- advanced biomaterials are required which can be implanted
radation (Le Maitre et al. 2006a) and growth factors, such using minimally invasive procedures. These materials must
as OP-1, to promote matrix restoration (Masuda et al. be deformable and able to withstand the loads experienced
2006). Given the growing understanding of nerve ingrowth by the spine; they must be able to support or promote cell
into the degenerate disc, it should also be possible to use survival, differentiation and matrix formation; and finally,
biological modulators to prevent or inhibit migration of they can biodegrade over a suitable timescale to non-toxic
nerves into the disc and thereby block transmission of dis- by-products. At present, no ideal biomaterials exist, but the
cogenic pain. Of course, while biological agents provide a field is developing rapidly and cell-based regenerative medi-
mechanism to inhibit the progression of early-stage degen- cine therapies appear likely to revolutionise the treatment of
eration, they may not be sufficient to regenerate tissue at discogenic back pain over the coming decades.
later stages of degeneration. At this late stage, pain is the
driver behind an individual seeking clinical help, and thus,
identification of a suitable cohort of early-stage patients 11.10 Summary of Critical Concepts Discussed
may be difficult. However, a clearer understanding of in the Chapter
genetic predisposition may enable individuals to be
screened and interventions targeted prior to the develop- Disc degeneration is a complex, multifactorial process, in
ment of symptomatic back pain. which disc cells themselves play a fundamental role.
The other area where biological modulators may be While there are strong genetic predeterminants, a clear
beneficial is in combination with cell-based tissue engineer- predisposition is difficult to detect.
ing therapies, where they may be used to stimulate cell dif- During degeneration, there is a cell-driven loss of proteo-
ferentiation or matrix formation. For intervertebral disc glycans from the extracellular matrix, which results in
tissue engineering, the use of nucleus pulposus cells would gross morphological, biological and biomechanical
initially appear to be the obvious choice, with disc cell reim- changes within the spine and the development of clinical
plantation showing promising results in both animal models back pain.
and small-scale human safety trials (Meisel et al. 2006). Members of the MMP and ADAMTS families are respon-
However, the alterations in disc cell phenotype and function sible for breakdown of the extracellular matrix, including
during degeneration raises questions about the applicability an imbalance between ADAMTSs and TIMP that could
of using autologous cells, isolated from degenerate discs, for lead to aggrecan degradation.
such therapies. Problems also surround the acquisition of In the disease state, the loss of aggrecan, with the shift to
autologous cells from discs adjacent to the degenerate level, versican production, reduces the disc water content and
as the local damage caused by removing tissue from those the shift to collagen type I production results in a more
regions has been shown to lead to degeneration at an acceler- fibrous tissue, less capable of withstanding load.
ated rate (Nomura et al. 2001). The use of allogeneic cells Expression levels of NGF and BDNF are increased in
from young, healthy donors also poses immunogenicity individuals with symptomatic disc degeneration. The disc
risks. For these reasons, the focus has shifted to the use of cells express the high-affinity NGF and BDNF receptors
autologous adult stem cells, derived from either bone mar- and the low-affinity NGF/BDNF receptor p75NTR, as well
row or adipose tissue, which have been shown to be capable as SP, suggesting autocrine signalling.
of differentiating into nucleus pulposus-like cells and pro- Interplay between catabolic cytokines and neurotrophins,
ducing an nucleus pulposus-like extracellular matrix both neurotrophin receptors and chemorepellant molecules
in vitro and in vivo (Sakai et al. 2005; Richardson et al. may guide nerve ingrowth during degeneration.
2006a, b, 2008a; Box 11.2). The potential value of these The predominant catabolic cytokines are IL-1 and TNF-a.
therapies is that they may be able to regenerate disc tissue During degeneration, interleukin family members, inter-
and restore long-term functionality. However, for the therapy feron gamma (IFN-g), TNF-a, PGE2 and NOx are
to be successful, it is important to take into account the envi- increased.
ronment into which these cells will be implanted. Thus, these TGF-b, along with BMPs 2 and 14, can stimulate MSC
studies must be conducted under conditions which closely differentiation towards nucleus pulposus-like cells
mimic the complex microenvironment of the degenerate in vitro.
disc. This microenvironment includes increased levels of A reduction in essential nutrients drives degeneration.
catabolic cytokines, with low levels of nutrients and a low With tissue hypoxia, an increase in lactate production
pH, all factors which are known to affect cell function and reduces the pH, which in turn reduces matrix synthesis
may have a profound effect on the ability of stem cells to and can cause cell death. The effects of low oxygen and
survive, differentiate or secrete matrix. low pH may be cumulative and promote disc cell death.
During 3 and 10 years of age, identifiable notochordal Alini M, Roughley PJ, Antoniou J, Stoll T, Aebi M (2002) A biological
cells disappear, and the highly hydrated, gelatinous extra- approach to treating disc degeneration: not for today, but maybe for
tomorrow. Eur Spine J 11(Suppl 2):S215S220
cellular matrix converts to a more fibrous, cartilaginous Alvarez L, Ortiz A (1999) The study of apoptosis in spine pathology.
nucleus pulposus. Spine (Phila Pa 1976) 24:500
It is probable that apoptosis plays a critical role in the An HS, Takegami K, Kamada H, Nguyen CM, Thonar EJ, Singh K,
pathogenesis of disc disease. Autophagic changes have Andersson GB, Masuda K (2005) Intradiscal administration of
osteogenic protein-1 increases intervertebral disc height and proteo-
also been identified in the degenerate disc. While glycan content in the nucleus pulposus in normal adolescent rabbits.
autophagy can lead to cell death, it can also maintain cell Spine (Phila Pa 1976) 30:2531
viability during periods of stress. Anderson DG, Li X, Balian G (2005) A fibronectin fragment alters the
Senescent cells are present in the disc, adopt a senescence- metabolism by rabbit intervertebral disc cells in vitro. Spine (Phila
Pa 1976) 30:12421246
associated secretory phenotype and secrete pro-ageing Annunen S, Paassilta P, Lohiniva J, Perala M, Pihlajamaa T, Karppinen
and catabolic factors. J, Tervonen O, Kroger H, Lahde S, Vanharanta H, Ryhanen L,
Disc loading has a profound effect on cell behaviour. Goring HH, Ott J, Prockop DJ, la-Kokko L (1999) An allele of
Moderate loads and low-frequency loading promote ana- COL9A2 associated with intervertebral disc disease. Science 285:
409412
bolic responses; high-magnitude, high-frequency and Antoniou J, Steffen T, Nelson F, Winterbottom N, Hollander AP, Poole
static loads elicit a catabolic response. RA, Aebi M, Alini M (1996) The human lumbar intervertebral disc:
The increasing understanding of the processes involved in evidence for changes in the biosynthesis and denaturation of the
degeneration is leading to the development of novel strat- extracellular matrix with growth, maturation, ageing, and degenera-
tion. J Clin Invest 98:9961003
egies which have the potential to revolutionise medical Antoniou J, Pike GB, Steffen T, Baramki H, Poole AR, Aebi M, Alini
intervention in discogenic back pain. M (1998) Quantitative magnetic resonance imaging in the assess-
ment of degenerative disc disease. Magn Reson Med 40:900907
Acknowledgments Supported by grants from the Biotechnology and Aota Y, An HS, Homandberg G, Thonar EJ, Andersson GB, Pichika R,
Biological Sciences Research Council (BBSRC), Arthritis Research Masuda K (2005) Differential effects of fibronectin fragment on
UK (ARUK) and Research Councils UK (RCUK). proteoglycan metabolism by intervertebral disc cells: a comparison
with articular chondrocytes. Spine (Phila Pa 1976) 30:722728
Ariga K, Yonenobu K, Nakase T, Kaneko M, Okuda S, Uchiyama Y,
Yoshikawa H (2001) Localization of cathepsins D, K, and L in
degenerated human intervertebral discs. Spine (Phila Pa 1976) 26:
References 26662672
Ashton IK, Roberts S, Jaffray DC, Polak JM, Eisenstein SM (1994)
Abbaszade I, Liu RQ, Yang F, Rosenfeld SA, Ross OH, Link JR, Ellis Neuropeptides in the human intervertebral disc. J Orthop Res 12:
DM, Tortorella MD, Pratta MA, Hollis JM, Wynn R, Duke JL, 186192
George HJ, Hillman MC Jr, Murphy K, Wiswall BH, Copeland RA, Bachmeier BE, Nerlich AG, Weiler C, Paesold G, Jochum M, Boos
Decicco CP, Bruckner R, Nagase H, Itoh Y, Newton RC, Magolda N (2007) Analysis of tissue distribution of TNF-alpha, TNF-alpha-
RL, Trzaskos JM, Burn TC et al (1999) Cloning and characteriza- receptors, and the activating TNF-alpha-converting enzyme sug-
tion of ADAMTS11, an aggrecanase from the ADAMTS family. gests activation of the TNF-alpha system in the aging intervertebral
J Biol Chem 274:2344323450 disc. Ann N Y Acad Sci 1096:4454
Abe Y, Akeda K, An HS, Aoki Y, Pichika R, Muehleman C, Kimura T, Bachmeier BE, Nerlich A, Mittermaier N, Weiler C, Lumenta C, Wuertz
Masuda K (2007) Proinflammatory cytokines stimulate the expres- K, Boos N (2009) Matrix metalloproteinase expression levels sug-
sion of nerve growth factor by human intervertebral disc cells. Spine gest distinct enzyme roles during lumbar disc herniation and degen-
(Phila Pa 1976) 32:635642 eration. Eur Spine J 18:15731586
Adams MA, Hutton WC (1986) The effect of posture on diffusion into Barksby HE, Milner JM, Patterson AM, Peake NJ, Hui W, Robson T,
lumbar intervertebral discs. J Anat 147:121134 Lakey R, Middleton J, Cawston TE, Richards CD, Rowan AD
Adams MA, Roughley PJ (2006) What is intervertebral disc degenera- (2006) Matrix metalloproteinase 10 promotion of collagenolysis via
tion, and what causes it? Spine (Phila Pa 1976) 31:21512161 procollagenase activation: implications for cartilage degradation in
Adams MA, Mannion AF, Dolan P (1999) Personal risk factors for first- arthritis. Arthritis Rheum 54:32443253
time low back pain. Spine (Phila Pa 1976) 24:24972505 Bartels EM, Fairbank JC, Winlove CP, Urban JP (1998) Oxygen and
Adams MA, Freeman BJ, Morrison HP, Nelson IW, Dolan P (2000) lactate concentrations measured in vivo in the intervertebral discs of
Mechanical initiation of intervertebral disc degeneration. Spine patients with scoliosis and back pain. Spine (Phila Pa 1976) 23:17
(Phila Pa 1976) 25:16251636 Beard HK, Roberts S, OBrien JP (1981) Immunofluorescent staining
Adams MA, Stefanakis M, Dolan P (2010) Healing of a painful inter- for collagen and proteoglycan in normal and scoliotic intervertebral
vertebral disc should not be confused with reversing disc degenera- discs. J Bone Joint Surg Br 63B:529534
tion: implications for physical therapies for discogenic back pain. Bernick S, Cailliet R (1982) Vertebral end-plate changes with aging of
Clin Biomech (Bristol, Avon) 25:961971 human vertebrae. Spine (Phila Pa 1976) 7:97102
Akyol S, Eraslan BS, Etyemez H, Tanriverdi T, Hanci M (2010) Bibby SR, Urban JP (2004) Effect of nutrient deprivation on the viabil-
Catabolic cytokine expressions in patients with degenerative disc ity of intervertebral disc cells. Eur Spine J 13:695701
disease. Turk Neurosurg 20:492499 Bibby SR, Fairbank JC, Urban MR, Urban JP (2002) Cell viability in
Ali R, Le Maitre CL, Richardson SM, Hoyland JA, Freemont AJ (2008) scoliotic discs in relation to disc deformity and nutrient levels. Spine
Connective tissue growth factor expression in human intervertebral (Phila Pa 1976) 27:22202228
disc: implications for angiogenesis in intervertebral disc degenera- Bogduk N (2004) Management of chronic low back pain. Med J Aust
tion. Biotech Histochem 83:239245 180:7983
Boos N, Nerlich AG, Wiest I, von der Mark K, Aebi M (1997) arthrodesis for the treatment of two-level lumbar degenerative disc
Immunolocalization of type X collagen in human lumbar interverte- disease: results at twenty-four months. J Bone Joint Surg Am 93:
bral discs during ageing and degeneration. Histochem Cell Biol 705715
108:471480 Demircan K, Hirohata S, Nishida K, Hatipoglu OF, Oohashi T,
Boos N, Weissbach S, Rohrbach H, Weiler C, Spratt KF, Nerlich AG Yonezawa T, Apte SS, Ninomiya Y (2005) ADAMTS-9 is synergis-
(2002) Classification of age-related changes in lumbar interverte- tically induced by interleukin-1beta and tumor necrosis factor alpha
bral discs: 2002 Volvo Award in basic science. Spine (Phila Pa in OUMS-27 chondrosarcoma cells and in human chondrocytes.
1976) 27:26312644 Arthritis Rheum 52:14511460
Brew K, Dinakarpandian D, Nagase H (2000) Tissue inhibitors of met- Deyo RA, Bass JE (1989) Lifestyle and low-back pain. The influence of
alloproteinases: evolution, structure and function. Biochim Biophys smoking and obesity. Spine (Phila Pa 1976) 14:501506
Acta 1477:267283 Di MA, Vaccaro AR, Lee JY, Denaro V, Lim MR (2005) Nucleus pul-
Brodin H (1955) Paths of nutrition in articular cartilage and interverte- posus replacement: basic science and indications for clinical use.
bral discs. Acta Orthop Scand 24:177183 Spine 30:S16S22
Brown MF, Hukkanen MV, McCarthy ID, Redfern DR, Batten JJ, Crock Duance VC, Crean JK, Sims TJ, Avery N, Smith S, Menage J, Eisenstein
HV, Hughes SP, Polak JM (1997) Sensory and sympathetic innerva- SM, Roberts S (1998) Changes in collagen cross-linking in degen-
tion of the vertebral endplate in patients with degenerative disc dis- erative disc disease and scoliosis. Spine (Phila Pa 1976)
ease. J Bone Joint Surg Br 79:147153 23:25452551
Buckwalter JA (1995) Aging and degeneration of the human interverte- Errico TJ (2005) Lumbar disc arthroplasty. Clin Orthop Relat Res
bral disc. Spine (Phila Pa 1976) 20:13071314 435:106117
Cal S, Obaya AJ, Llamazares M, Garabaya C, Quesada V, Lopez-Otin Eyre DR, Muir H (1976) Types I and II collagens in intervertebral disc.
C (2002) Cloning, expression analysis, and structural characteriza- Interchanging radial distributions in annulus fibrosus. Biochem
tion of seven novel human ADAMTSs, a family of metalloprotei- J 157:267270
nases with disintegrin and thrombospondin-1 domains. Gene 283: Feng H, Danfelter M, Stromqvist B, Heinegard D (2006) Extracellular
4962 matrix in disc degeneration. J Bone Joint Surg Am 88(Suppl
Cappello R, Bird JL, Pfeiffer D, Bayliss MT, Dudhia J (2006) 2):2529
Notochordal cell produce and assemble extracellular matrix in a dis- Freemont AJ, Peacock TE, Goupille P, Hoyland JA, OBrien J, Jayson
tinct manner, which may be responsible for the maintenance of MI (1997) Nerve ingrowth into diseased intervertebral disc in
healthy nucleus pulposus. Spine (Phila Pa 1976) 31:873882 chronic back pain. Lancet 350:178181
Chen Q, Fischer A, Reagan JD, Yan LJ, Ames BN (1995) Oxidative Freemont AJ, Watkins A, Le MC, Baird P, Jeziorska M, Knight MT,
DNA damage and senescence of human diploid fibroblast cells. Ross ER, OBrien JP, Hoyland JA (2002) Nerve growth factor
Proc Natl Acad Sci U S A 92:43374341 expression and innervation of the painful intervertebral disc. J Pathol
Cheung KM, Karppinen J, Chan D, Ho DW, Song YQ, Sham P, Cheah 197:286292
KS, Leong JC, Luk KD (2009) Prevalence and pattern of lumbar Freund A, Patil CK, Campisi J (2011) p38MAPK is a novel DNA dam-
magnetic resonance imaging changes in a population study of one age response-independent regulator of the senescence-associated
thousand forty-three individuals. Spine (Phila Pa 1976) 34: secretory phenotype. EMBO J 30:15361548
934940 Frino J, McCarthy RE, Sparks CY, McCullough FL (2006) Trends
Chubinskaya S, Kawakami M, Rappoport L, Matsumoto T, Migita N, in adolescent lumbar disk herniation. J Pediatr Orthop 26:
Rueger DC (2007) Anti-catabolic effect of OP-1 in chronically 579581
compressed intervertebral discs. J Orthop Res 25:517530 Gabr MA, Jing L, Helbling AR, Sinclair SM, Allen KD, Shamji MF,
Collins-Racie LA, Flannery CR, Zeng W, Corcoran C, Annis-Freeman Richardson WJ, Fitch RD, Setton LA, Chen J (2011) Interleukin-17
B, Agostino MJ, Arai M, DiBlasio-Smith E, Dorner AJ, Georgiadis synergizes with IFNgamma or TNFalpha to promote inflammatory
KE, Jin M, Tan XY, Morris EA, LaVallie ER (2004) ADAMTS-8 mediator release and intercellular adhesion molecule-1 (ICAM-1)
exhibits aggrecanase activity and is expressed in human articular expression in human intervertebral disc cells. J Orthop Res
cartilage. Matrix Biol 23:219230 29:17
Cooper TW, Eisen AZ, Stricklin GP, Welgus HG (1985) Platelet-derived Garcia-Cosamalon J, del Valle ME, Calavia MG, Garcia-Suarez O,
collagenase inhibitor: characterization and subcellular localization. Lopez-Muniz A, Otero J, Vega JA (2010) Intervertebral disc, sensory
Proc Natl Acad Sci U S A 82:27792783 nerves and neurotrophins: who is who in discogenic pain? J Anat
Coppes MH, Marani E, Thomeer RT, Groen GJ (1997) Innervation of 217:115
painful lumbar discs. Spine (Phila Pa 1976) 22:23422349 Gilbert HT, Hoyland JA, Millward-Sadler SJ (2010) The response of
Cs-Szabo G, Ragasa-San JD, Turumella V, Masuda K, Thonar EJ, An human anulus fibrosus cells to cyclic tensile strain is frequency-
HS (2002) Changes in mRNA and protein levels of proteoglycans of dependent and altered with disc degeneration. Arthritis Rheum
the anulus fibrosus and nucleus pulposus during intervertebral disc 62:33853394
degeneration. Spine (Phila Pa 1976) 27:22122219 Gilbert HT, Hoyland JA, Freemont AJ, Millward-Sadler SJ (2011) The
Cui LY, Liu SL, Ding Y, Huang DS, Ma RF, Huang WG, Hu BS, Pan QH involvement of interleukin-1 and interleukin-4 in the response of
(2007) IL-1beta sensitizes rat intervertebral disc cells to Fas ligand human annulus fibrosus cells to cyclic tensile strain: an altered
mediated apoptosis in vitro. Acta Pharmacol Sin 28:16711676 mechanotransduction pathway with degeneration. Arthritis Res
Dai SM, Shan ZZ, Nakamura H, Masuko-Hongo K, Kato T, Nishioka Ther 13:R8
K, Yudoh K (2006) Catabolic stress induces features of chondrocyte Gilbertson L, Ahn SH, Teng PN, Studer RK, Niyibizi C, Kang JD
senescence through overexpression of caveolin 1: possible involve- (2008) The effects of recombinant human bone morphogenetic pro-
ment of caveolin 1-induced down-regulation of articular chondro- tein-2, recombinant human bone morphogenetic protein-12, and
cytes in the pathogenesis of osteoarthritis. Arthritis Rheum adenoviral bone morphogenetic protein-12 on matrix synthesis in
54:818831 human annulus fibrosis and nucleus pulposus cells. Spine J 8:
Delamarter R, Zigler JE, Balderston RA, Cammisa FP, Goldstein JA, 449456
Spivak JM (2011) Prospective, randomized, multicenter Food and Greg AD, Li X, Tannoury T, Beck G, Balian G (2003) A fibronectin
Drug Administration investigational device exemption study of the fragment stimulates intervertebral disc degeneration in vivo. Spine
ProDisc-L total disc replacement compared with circumferential (Phila Pa 1976) 28:23382345
Gruber HE, Hanley EN Jr (1998) Analysis of aging and degeneration of Holm S, Nachemson A (1983) Variations in the nutrition of the canine
the human intervertebral disc. Comparison of surgical specimens intervertebral disc induced by motion. Spine (Phila Pa 1976)
with normal controls. Spine (Phila Pa 1976) 23:751757 8:866874
Gruber HE, Fisher EC Jr, Desai B, Stasky AA, Hoelscher G, Hanley EN Holm S, Nachemson A (1988) Nutrition of the intervertebral disc: acute
Jr (1997) Human intervertebral disc cells from the annulus: three- effects of cigarette smoking. An experimental animal study. Ups J
dimensional culture in agarose or alginate and responsiveness to Med Sci 93:9199
TGF-beta1. Exp Cell Res 235:1321 Holm S, Maroudas A, Urban JP, Selstam G, Nachemson A (1981)
Gruber HE, Norton HJ, Hanley EN Jr (2000) Anti-apoptotic effects of Nutrition of the intervertebral disc: solute transport and metabolism.
IGF-1 and PDGF on human intervertebral disc cells in vitro. Spine Connect Tissue Res 8:101119
(Phila Pa 1976) 25:21532157 Homandberg GA, Hui F, Wen C, Purple C, Bewsey K, Koepp H, Huch
Gruber HE, Ingram JA, Hanley EN Jr (2005) Immunolocalization of K, Harris A (1997) Fibronectin-fragment-induced cartilage chon-
MMP-19 in the human intervertebral disc: implications for disc drolysis is associated with release of catabolic cytokines. Biochem
aging and degeneration. Biotech Histochem 80:157162 J 321(Pt 3):751757
Gruber HE, Ingram JA, Norton HJ, Hanley EN Jr (2007) Senescence in Homma Y, Tsunoda M, Kasai H (1994) Evidence for the accumulation
cells of the aging and degenerating intervertebral disc: immunolo- of oxidative stress during cellular ageing of human diploid
calization of senescence-associated beta-galactosidase in human fibroblasts. Biochem Biophys Res Commun 203:10631068
and sand rat discs. Spine (Phila Pa 1976) 32:321327 Hormel SE, Eyre DR (1991) Collagen in the ageing human interverte-
Gruber HE, Ingram JA, Hoelscher G, Zinchenko N, Norton HJ, Hanley bral disc: an increase in covalently bound fluorophores and chro-
EN Jr (2008) Brain-derived neurotrophic factor and its receptor in mophores. Biochim Biophys Acta 1078:243250
the human and the sand rat intervertebral disc. Arthritis Res Ther Horner HA, Urban JP (2001) 2001 Volvo Award Winner in Basic
10:R82 Science Studies: effect of nutrient supply on the viability of cells
Guehring T, Urban JP, Cui Z, Tirlapur UK (2008) Noninvasive 3D vital from the nucleus pulposus of the intervertebral disc. Spine (Phila Pa
imaging and characterization of notochordal cells of the intervertebral 1976) 26:25432549
disc by femtosecond near-infrared two-photon laser scanning micros- Hoyland JA, Le MC, Freemont AJ (2008) Investigation of the role of
copy and spatial-volume rendering. Microsc Res Tech 71:298304 IL-1 and TNF in matrix degradation in the intervertebral disc.
Guyer RD, McAfee PC, Banco RJ, Bitan FD, Cappuccino A, Geisler Rheumatology (Oxford) 47:809814
FH, Hochschuler SH, Holt RT, Jenis LG, Majd ME, Regan JJ, Humzah MD, Soames RW (1988) Human intervertebral disc: structure
Tromanhauser SG, Wong DC, Blumenthal SL (2009) Prospective, and function. Anat Rec 220:337356
randomized, multicenter Food and Drug Administration investiga- Hutton WC, Adams MA (1982) Can the lumbar spine be crushed in
tional device exemption study of lumbar total disc replacement with heavy lifting? Spine (Phila Pa 1976) 7:586590
the CHARITE artificial disc versus lumbar fusion: five-year follow- Igarashi T, Kikuchi S, Shubayev V, Myers RR (2000) 2000 Volvo
up. Spine J 9:374386 Award winner in basic science studies: exogenous tumor necrosis
Ha KY, Koh IJ, Kirpalani PA, Kim YY, Cho YK, Khang GS, Han CW factor-alpha mimics nucleus pulposus-induced neuropathology.
(2006) The expression of hypoxia inducible factor-1alpha and apop- Molecular, histologic, and behavioral comparisons in rats. Spine
tosis in herniated discs. Spine (Phila Pa 1976) 31:13091313 (Phila Pa 1976) 25:29752980
Handa T, Ishihara H, Ohshima H, Osada R, Tsuji H, Obata K (1997) Imai Y, Miyamoto K, An HS, Thonar EJ, Andersson GB, Masuda K
Effects of hydrostatic pressure on matrix synthesis and matrix met- (2007a) Recombinant human osteogenic protein-1 upregulates pro-
alloproteinase production in the human lumbar intervertebral disc. teoglycan metabolism of human anulus fibrosus and nucleus pulpo-
Spine (Phila Pa 1976) 22:10851091 sus cells. Spine (Phila Pa 1976) 32:13031309
Harkness EF, Macfarlane GJ, Silman AJ, McBeth J (2005) Is musculo- Imai Y, Okuma M, An HS, Nakagawa K, Yamada M, Muehleman C,
skeletal pain more common now than 40 years ago?: Two popula- Thonar E, Masuda K (2007b) Restoration of disc height loss by
tion-based cross-sectional studies. Rheumatology (Oxford) recombinant human osteogenic protein-1 injection into interverte-
44:890895 bral discs undergoing degeneration induced by an intradiscal
Haro H, Kato T, Komori H, Osada M, Shinomiya K (2002) Vascular injection of chondroitinase ABC. Spine (Phila Pa 1976) 32:
endothelial growth factor (VEGF)-induced angiogenesis in herni- 11971205
ated disc resorption. J Orthop Res 20:409415 Inkinen RI, Lammi MJ, Lehmonen S, Puustjarvi K, Kaapa E, Tammi
Hatano E, Fujita T, Ueda Y, Okuda T, Katsuda S, Okada Y, Matsumoto MI (1998) Relative increase of biglycan and decorin and altered
T (2006) Expression of ADAMTS-4 (aggrecanase-1) and possible chondroitin sulfate epitopes in the degenerating human interverte-
involvement in regression of lumbar disc herniation. Spine (Phila Pa bral disc. J Rheumatol 25:506514
1976) 31:14261432 Ishihara H, Urban JP (1999) Effects of low oxygen concentrations and
Heathfield SK, Le Maitre CL, Hoyland JA (2008) Caveolin-1 expres- metabolic inhibitors on proteoglycan and protein synthesis rates in
sion and stress-induced premature senescence in human interverte- the intervertebral disc. J Orthop Res 17:829835
bral disc degeneration. Arthritis Res Ther 10:R87 Jackson AR, Huang CY, Gu WY (2011) Effect of endplate calcification
Heyde CE, Tschoeke SK, Hellmuth M, Hostmann A, Ertel W, and mechanical deformation on the distribution of glucose in inter-
Oberholzer A (2006) Trauma induces apoptosis in human thora- vertebral disc: a 3D finite element study. Comput Methods Biomech
columbar intervertebral discs. BMC Clin Pathol 6:5 Biomed Engin 14:195204
Hilibrand AS, Robbins M (2004) Adjacent segment degeneration and Jin D, Qu D, Zhao L, Chen J, Jiang J (2003) Prosthetic disc nucleus
adjacent segment disease: the consequences of spinal fusion? Spine (PDN) replacement for lumbar disc herniation: preliminary report
J 4:190S194S with six months follow-up. J Spinal Disord Tech 16:331337
Hirano N, Tsuji H, Ohshima H, Kitano S, Itoh T, Sano A (1988) Johnson WE, Roberts S (2007) Rumours of my death may have been
Analysis of rabbit intervertebral disc physiology based on water greatly exaggerated: a brief review of cell death in human interver-
metabolism. II. Changes in normal intervertebral discs under axial tebral disc disease and implications for cell transplantation therapy.
vibratory load. Spine (Phila Pa 1976) 13:12971302 Biochem Soc Trans 35:680682
Holm S, Nachemson A (1982) Nutritional changes in the canine inter- Johnson WE, Eisenstein SM, Roberts S (2001) Cell cluster formation in
vertebral disc after spinal fusion. Clin Orthop Relat Res 169: degenerate lumbar intervertebral discs is associated with increased
243258 disc cell proliferation. Connect Tissue Res 42:197207
Johnson WE, Caterson B, Eisenstein SM, Hynds DL, Snow DM, factors in human intervertebral disc: implications for repair and
Roberts S (2002) Human intervertebral disc aggrecan inhibits nerve regeneration of the disc. J Pathol 207:445452
growth in vitro. Arthritis Rheum 46:26582664 Le Maitre CL, Freemont AJ, Hoyland JA (2006a) A preliminary in vitro
Johnson WE, Caterson B, Eisenstein SM, Roberts S (2005) Human study into the use of IL-1Ra gene therapy for the inhibition of inter-
intervertebral disc aggrecan inhibits endothelial cell adhesion and vertebral disc degeneration. Int J Exp Pathol 87:1728
cell migration in vitro. Spine (Phila Pa 1976) 30:11391147 Le Maitre CL, Freemont AJ, Hoyland JA (2006b) Human disc degen-
Johnson WE, Sivan S, Wright KT, Eisenstein SM, Maroudas A, Roberts eration is associated with increased MMP 7 expression. Biotech
S (2006) Human intervertebral disc cells promote nerve growth over Histochem 81:125131
substrata of human intervertebral disc aggrecan. Spine (Phila Pa Le Maitre CL, Freemont AJ, Hoyland JA (2007a) Accelerated cellular
1976) 31:11871193 senescence in degenerate intervertebral discs: a possible role in the
Johnson WE, Patterson AM, Eisenstein SM, Roberts S (2007) The pres- pathogenesis of intervertebral disc degeneration. Arthritis Res Ther
ence of pleiotrophin in the human intervertebral disc is associated 9:R45
with increased vascularization: an immunohistologic study. Spine Le Maitre CL, Hoyland JA, Freemont AJ (2007b) Catabolic cytokine
(Phila Pa 1976) 32:12951302 expression in degenerate and herniated human intervertebral discs:
Jones P, Gardner L, Menage J, Williams GT, Roberts S (2008) IL-1beta and TNFalpha expression profile. Arthritis Res Ther
Intervertebral disc cells as competent phagocytes in vitro: implica- 9:R77
tions for cell death in disc degeneration. Arthritis Res Ther 10:R86 Le Maitre CL, Hoyland JA, Freemont AJ (2007c) Interleukin-1 receptor
Kang JD, Georgescu HI, Intyre-Larkin L, Stefanovic-Racic M, antagonist delivered directly and by gene therapy inhibits matrix
Donaldson WF III, Evans CH (1996) Herniated lumbar interverte- degradation in the intact degenerate human intervertebral disc: an in
bral discs spontaneously produce matrix metalloproteinases, nitric situ zymographic and gene therapy study. Arthritis Res Ther 9:R83
oxide, interleukin-6, and prostaglandin E2. Spine (Phila Pa 1976) Le Maitre CL, Pockert A, Buttle DJ, Freemont AJ, Hoyland JA (2007d)
21:271277 Matrix synthesis and degradation in human intervertebral disc
Kashiwagi M, Tortorella M, Nagase H, Brew K (2001) TIMP-3 is a degeneration. Biochem Soc Trans 35:652655
potent inhibitor of aggrecanase 1 (ADAM-TS4) and aggrecanase 2 Le Maitre CL, Frain J, Fotheringham AP, Freemont AJ, Hoyland JA
(ADAM-TS5). J Biol Chem 276:1250112504 (2008) Human cells derived from degenerate intervertebral discs
Katsuura A, Hukuda S (1994) Experimental study of intervertebral disc respond differently to those derived from non-degenerate interverte-
allografting in the dog. Spine 19:24262432 bral discs following application of dynamic hydrostatic pressure.
Katz MM, Hargens AR, Garfin SR (1986) Intervertebral disc nutrition. Biorheology 45:563575
Diffusion versus convection. Clin Orthop Relat Res (210):243245 Le Maitre CL, Freemont AJ, Hoyland JA (2009) Expression of carti-
Kawaguchi Y, Kanamori M, Ishihara H, Ohmori K, Matsui H, Kimura T lage-derived morphogenetic protein in human intervertebral discs
(2002) The association of lumbar disc disease with vitamin-D recep- and its effect on matrix synthesis in degenerate human nucleus pul-
tor gene polymorphism. J Bone Joint Surg Am 84-A:20222028 posus cells. Arthritis Res Ther 11:R137
Kawakami M, Matsumoto T, Hashizume H, Kuribayashi K, Chubinskaya Lee CR, Sakai D, Nakai T, Toyama K, Mochida J, Alini M, Grad S
S, Yoshida M (2005) Osteogenic protein-1 (osteogenic protein-1/ (2007) A phenotypic comparison of intervertebral disc and articular
bone morphogenetic protein-7) inhibits degeneration and pain- cartilage cells in the rat. Eur Spine J 16:21742185
related behavior induced by chronically compressed nucleus pulpo- Lee JM, Song JY, Baek M, Jung HY, Kang H, Han IB, Kwon YD, Shin
sus in the rat. Spine (Phila Pa 1976) 30:19331939 DE (2011) Interleukin-1beta induces angiogenesis and innervation in
Kim KW, Kim YS, Ha KY, Woo YK, Park JB, Park WS, An HS (2005) human intervertebral disc degeneration. J Orthop Res 29:265269
An autocrine or paracrine Fas-mediated counterattack: a potential Lindley EM, Jaafar S, Noshchenko A, Baldini T, Nair DP, Shandas R,
mechanism for apoptosis of notochordal cells in intact rat nucleus Burger EL, Patel VV (2010) Nucleus replacement device failure: a
pulposus. Spine (Phila Pa 1976) 30:12471251 case report and biomechanical study. Spine (Phila Pa 1976)
Kim KW, Chung HN, Ha KY, Lee JS, Kim YY (2009) Senescence 35:E1241E1247
mechanisms of nucleus pulposus chondrocytes in human interverte- Loreto C, Musumeci G, Castorina A, Loreto C, Martinez G (2011)
bral discs. Spine J 9:658666 Degenerative disc disease of herniated intervertebral discs is associ-
Klawitter M, Quero L, Bertolo A, Mehr M, Stoyanov J, Nerlich AG, ated with extracellular matrix remodeling, vimentin-positive cells
Klasen J, Aebli N, Boos N, Wuertz K (2011) Human MMP28 and cell death. Ann Anat 193:156162
expression is unresponsive to inflammatory stimuli and does not Lotz JC, Chin JR (2000) Intervertebral disc cell death is dependent on
correlate to the grade of intervertebral disc degeneration. J Negat the magnitude and duration of spinal loading. Spine (Phila Pa 1976)
Results Biomed 10:9 25:14771483
Kolodkin AL, Matthes DJ, Goodman CS (1993) The semaphorin genes Luk KD, Ruan DK, Chow DH, Leong JC (1997) Intervertebral disc
encode a family of transmembrane and secreted growth cone guid- autografting in a bipedal animal model. Clin Orthop Relat Res
ance molecules. Cell 75:13891399 337:1326
Konttinen YT, Kaapa E, Hukkanen M, Gu XH, Takagi M, Santavirta S, Lyons G, Eisenstein SM, Sweet MB (1981) Biochemical changes in inter-
Alaranta H, Li TF, Suda A (1999) Cathepsin G in degenerating and vertebral disc degeneration. Biochim Biophys Acta 673:443453
healthy discal tissue. Clin Exp Rheumatol 17:197204 MacGregor AJ, Andrew T, Sambrook PN, Spector TD (2004) Structural,
Korecki CL, Kuo CK, Tuan RS, Iatridis JC (2009) Intervertebral disc psychological, and genetic influences on low back and neck pain: a
cell response to dynamic compression is age and frequency depen- study of adult female twins. Arthritis Rheum 51:160167
dent. J Orthop Res 27:800806 MacLean JJ, Lee CR, Alini M, Iatridis JC (2004) Anabolic and cata-
Le Maitre CL, Freemont AJ, Hoyland JA (2004) Localization of degra- bolic mRNA levels of the intervertebral disc vary with the magni-
dative enzymes and their inhibitors in the degenerate human inter- tude and frequency of in vivo dynamic compression. J Orthop Res
vertebral disc. J Pathol 204:4754 22:11931200
Le Maitre CL, Freemont AJ, Hoyland JA (2005a) The role of interleu- MacLean JJ, Lee CR, Alini M, Iatridis JC (2005) The effects of short-
kin-1 in the pathogenesis of human intervertebral disc degeneration. term load duration on anabolic and catabolic gene expression in the
Arthritis Res Ther 7:R732R745 rat tail intervertebral disc. J Orthop Res 23:11201127
Le Maitre CL, Richardson SM, Baird P, Freemont AJ, Hoyland JA Malandrino A, Noailly J, Lacroix D (2011) The effect of sustained
(2005b) Expression of receptors for putative anabolic growth compression on oxygen metabolic transport in the intervertebral
disc decreases with degenerative changes. PLoS Comput Biol Nachemson A, Lewin T, Maroudas A, Freeman MA (1970) In vitro dif-
7:e1002112 fusion of dye through the end-plates and the annulus fibrosus of
Maniadakis N, Gray A (2000) The economic burden of back pain in the human lumbar inter-vertebral discs. Acta Orthop Scand 41:589607
UK. Pain 84:95103 Nagase H, Kashiwagi M (2003) Aggrecanases and cartilage matrix deg-
Marchand F, Ahmed AM (1990) Investigation of the laminate structure radation. Arthritis Res Ther 5:94103
of lumbar disc anulus fibrosus. Spine 15:402410 Nagase H, Woessner JF Jr (1999) Matrix metalloproteinases. J Biol
Maroudas A, Stockwell RA, Nachemson A, Urban J (1975) Factors Chem 274:2149121494
involved in the nutrition of the human lumbar intervertebral disc: Neidlinger-Wilke C, Wurtz K, Urban JP, Borm W, Arand M, Ignatius A,
cellularity and diffusion of glucose in vitro. J Anat 120:113130 Wilke HJ, Claes LE (2006) Regulation of gene expression in inter-
Martin JA, Brown TD, Heiner AD, Buckwalter JA (2004) Chondrocyte vertebral disc cells by low and high hydrostatic pressure. Eur Spine
senescence, joint loading and osteoarthritis. Clin Orthop Relat Res J 15(Suppl 3):S372S378
(427 Suppl):S96103 Nerlich AG, Schleicher ED, Boos N (1997) 1997 Volvo Award winner
Martin BI, Deyo RA, Mirza SK, Turner JA, Comstock BA, Hollingworth in basic science studies. Immunohistologic markers for age-related
W, Sullivan SD (2008) Expenditures and health status among adults changes of human lumbar intervertebral discs. Spine (Phila Pa 1976)
with back and neck problems. JAMA 299:656664 22:27812795
Masuda K, Takegami K, An H, Kumano F, Chiba K, Andersson GB, Nerlich AG, Boos N, Wiest I, Aebi M (1998) Immunolocalization of
Schmid T, Thonar E (2003) Recombinant osteogenic protein-1 major interstitial collagen types in human lumbar intervertebral
upregulates extracellular matrix metabolism by rabbit annulus discs of various ages. Virchows Arch 432:6776
fibrosus and nucleus pulposus cells cultured in alginate beads. Nerlich AG, Weiler C, Zipperer J, Narozny M, Boos N (2002)
J Orthop Res 21:922930 Immunolocalization of phagocytic cells in normal and degenerated
Masuda K, Imai Y, Okuma M, Muehleman C, Nakagawa K, Akeda K, intervertebral discs. Spine (Phila Pa 1976) 27:24842490
Thonar E, Andersson G, An HS (2006) Osteogenic protein-1 injec- Nerlich AG, Schaaf R, Walchli B, Boos N (2007) Temporo-spatial dis-
tion into a degenerated disc induces the restoration of disc height tribution of blood vessels in human lumbar intervertebral discs. Eur
and structural changes in the rabbit anular puncture model. Spine Spine J 16:547555
(Phila Pa 1976) 31:742754 Nomura T, Mochida J, Okuma M, Nishimura K, Sakabe K (2001)
McCanless JD, Cole JA, Slack SM, Bumgardner JD, Zamora PO, Nucleus pulposus allograft retards intervertebral disc degeneration.
Haggard WO (2011) Modeling nucleus pulposus regeneration Clin Orthop Relat Res 389:94101
in vitro: mesenchymal stem cells, alginate beads, hypoxia, BMP-2, Oegema TR Jr, Johnson SL, Aguiar DJ, Ogilvie JW (2000) Fibronectin
and synthetic peptide B2A. Spine (Phila Pa 1976) 36:22752285 and its fragments increase with degeneration in the human interver-
Meisel HJ, Ganey T, Hutton WC, Libera J, Minkus Y, Alasevic O tebral disc. Spine (Phila Pa 1976) 25:27422747
(2006) Clinical experience in cell-based therapeutics: intervention Ohba T, Haro H, Ando T, Wako M, Suenaga F, Aso Y, Koyama K,
and outcome. Eur Spine J 15(Suppl 3):S397S405 Hamada Y, Nakao A (2009) TNF-alpha-induced NF-kappaB signal-
Melrose J, Roberts S, Smith S, Menage J, Ghosh P (2002a) Increased ing reverses age-related declines in VEGF induction and angiogenic
nerve and blood vessel ingrowth associated with proteoglycan activity in intervertebral disc tissues. J Orthop Res 27:229235
depletion in an ovine anular lesion model of experimental disc Ohshima H, Urban JP (1992) The effect of lactate and pH on proteogly-
degeneration. Spine (Phila Pa 1976) 27:12781285 can and protein synthesis rates in the intervertebral disc. Spine
Melrose J, Smith S, Little CB, Kitson J, Hwa SY, Ghosh P (2002b) (Phila Pa 1976) 17:10791082
Spatial and temporal localization of transforming growth factor- Ohtori S, Takahashi K, Chiba T, Yamagata M, Sameda H, Moriya H
beta, fibroblast growth factor-2, and osteonectin, and identification (2002) Substance P and calcitonin gene-related peptide immunore-
of cells expressing alpha-smooth muscle actin in the injured anulus active sensory DRG neurons innervating the lumbar intervertebral
fibrosus: implications for extracellular matrix repair. Spine (Phila Pa discs in rats. Ann Anat 184:235240
1976) 27:17561764 Olmarker K, Larsson K (1998) Tumor necrosis factor alpha and nucleus-
Melrose J, Smith SM, Appleyard RC, Little CB (2008) Aggrecan, ver- pulposus-induced nerve root injury. Spine (Phila Pa 1976) 23:
sican and type VI collagen are components of annular translamellar 25382544
crossbridges in the intervertebral disc. Eur Spine J 17:314324 Olmarker K, Rydevik B (2001) Selective inhibition of tumor necrosis
Minogue BM, Richardson SM, Zeef LA, Freemont AJ, Hoyland JA factor-alpha prevents nucleus pulposus-induced thrombus forma-
(2010a) Characterization of the human nucleus pulposus cell pheno- tion, intraneural edema, and reduction of nerve conduction velocity:
type and evaluation of novel marker gene expression to define adult possible implications for future pharmacologic treatment strategies
stem cell differentiation. Arthritis Rheum 62:36953705 of sciatica. Spine (Phila Pa 1976) 26:863869
Minogue BM, Richardson SM, Zeef LA, Freemont AJ, Hoyland JA Osada R, Ohshima H, Ishihara H, Yudoh K, Sakai K, Matsui H, Tsuji H
(2010b) Transcriptional profiling of bovine intervertebral disc cells: (1996) Autocrine/paracrine mechanism of insulin-like growth fac-
implications for identification of normal and degenerate human tor-1 secretion, and the effect of insulin-like growth factor-1 on pro-
intervertebral disc cell phenotypes. Arthritis Res Ther 12:R22 teoglycan synthesis in bovine intervertebral discs. J Orthop Res
Miyamoto K, Masuda K, Kim JG, Inoue N, Akeda K, Andersson GB, 14:690699
An HS (2006) Intradiscal injections of osteogenic protein-1 restore Park JB, Park IC, Park SJ, Jin HO, Lee JK, Riew KD (2006) Anti-
the viscoelastic properties of degenerated intervertebral discs. Spine apoptotic effects of caspase inhibitors on rat intervertebral disc
J 6:692703 cells. J Bone Joint Surg Am 88:771779
Mokhbi SD, Shirazi-Adl A, Urban JP (2009) Investigation of solute Patel KP, Sandy JD, Akeda K, Miyamoto K, Chujo T, An HS, Masuda
concentrations in a 3D model of intervertebral disc. Eur Spine J K (2007) Aggrecanases and aggrecanase-generated fragments in the
18:254262 human intervertebral disc at early and advanced stages of disc
Mwale F, Roughley P, Antoniou J (2004) Distinction between the extra- degeneration. Spine (Phila Pa 1976) 32:25962603
cellular matrix of the nucleus pulposus and hyaline cartilage: a req- Pazzaglia UE, Salisbury JR, Byers PD (1989) Development and involu-
uisite for tissue engineering of intervertebral disc. Eur Cell Mater tion of the notochord in the human spine. J R Soc Med 82:413415
8:5863 Pearce RH, Grimmer BJ, Adams ME (1987) Degeneration and the
Nachemson AL (1981) Disc pressure measurements. Spine (Phila Pa chemical composition of the human lumbar intervertebral disc.
1976) 6:9397 J Orthop Res 5:198205
Peng B, Hao J, Hou S, Wu W, Jiang D, Fu X, Yang Y (2006) Possible Roberts S, Urban JP, Evans H, Eisenstein SM (1996) Transport proper-
pathogenesis of painful intervertebral disc degeneration. Spine ties of the human cartilage endplate in relation to its composition
(Phila Pa 1976) 31:560566 and calcification. Spine (Phila Pa 1976) 21:415420
Peng B, Chen J, Kuang Z, Li D, Pang X, Zhang X (2009) Expression Roberts S, Caterson B, Menage J, Evans EH, Jaffray DC, Eisenstein
and role of connective tissue growth factor in painful disc fibrosis SM (2000) Matrix metalloproteinases and aggrecanase: their role in
and degeneration. Spine (Phila Pa 1976) 34:E178E182 disorders of the human intervertebral disc. Spine (Phila Pa 1976)
Pfirrmann CW, Metzdorf A, Zanetti M, Hodler J, Boos N (2001) 25:30053013
Magnetic resonance classification of lumbar intervertebral disc Roberts S, Evans EH, Kletsas D, Jaffray DC, Eisenstein SM (2006a)
degeneration. Spine (Phila Pa 1976) 26:18731878 Senescence in human intervertebral discs. Eur Spine J 15(Suppl 3):
Pluijm SM, van Essen HW, Bravenboer N, Uitterlinden AG, Smit JH, S312S316
Pols HA, Lips P (2004) Collagen type I alpha1 Sp1 polymorphism, Roberts S, Evans H, Trivedi J, Menage J (2006b) Histology and pathol-
osteoporosis, and intervertebral disc degeneration in older men and ogy of the human intervertebral disc. J Bone Joint Surg Am 88(Suppl
women. Ann Rheum Dis 63:7177 2):1014
Pockert AJ, Richardson SM, Le Maitre CL, Lyon M, Deakin JA, Buttle Rodriguez AG, Slichter CK, Acosta FL, Rodriguez-Soto AE, Burghardt
DJ, Freemont AJ, Hoyland JA (2009) Modified expression of the AJ, Majumdar S, Lotz JC (2011) Human disc nucleus properties
ADAMTS enzymes and tissue inhibitor of metalloproteinases 3 and vertebral endplate permeability. Spine (Phila Pa 1976) 36:
during human intervertebral disc degeneration. Arthritis Rheum 60: 512520
482491 Rodriguez AG, Rodriguez-Soto AE, Burghardt AJ, Berven S, Majumdar
Pokharna HK, Phillips FM (1998) Collagen crosslinks in human S, Lotz JC (2012) Morphology of the human vertebral endplate.
lumbar intervertebral disc aging. Spine (Phila Pa 1976) 23: J Orthop Res 30:280287
16451648 Roughley PJ (2004) Biology of intervertebral disc aging and degen-
Pollintine P, Przybyla AS, Dolan P, Adams MA (2004) Neural arch eration: involvement of the extracellular matrix. Spine 29:
load-bearing in old and degenerated spines. J Biomech 37:197204 26912699
Postacchini F, Lami R, Pugliese O (1988) Familial predisposition to Roughley P, Martens D, Rantakokko J, Alini M, Mwale F, Antoniou
discogenic low-back pain. An epidemiologic and immunogenetic J (2006) The involvement of aggrecan polymorphism in degenera-
study. Spine (Phila Pa 1976) 13:14031406 tion of human intervertebral disc and articular cartilage. Eur Cell
Purmessur D, Freemont AJ, Hoyland JA (2008) Expression and regula- Mater 11:17
tion of neurotrophins in the nondegenerate and degenerate human Rutges JP, Duit RA, Kummer JA, Oner FC, van Rijen MH, Verbout AJ,
intervertebral disc. Arthritis Res Ther 10:R99 Castelein RM, Dhert WJ, Creemers LB (2010) Hypertrophic dif-
Rannou F, Lee TS, Zhou RH, Chin J, Lotz JC, Mayoux-Benhamou MA, ferentiation and calcification during intervertebral disc degenera-
Barbet JP, Chevrot A, Shyy JY (2004) Intervertebral disc degeneration. Osteoarthritis Cartilage 18:14871495
tion: the role of the mitochondrial pathway in annulus fibrosus cell Sakai D, Mochida J, Iwashina T, Watanabe T, Nakai T, Ando K, Hotta
apoptosis induced by overload. Am J Pathol 164:915924 T (2005) Differentiation of mesenchymal stem cells transplanted to
Repanti M, Korovessis PG, Stamatakis MV, Spastris P, Kosti P (1998) a rabbit degenerative disc model: potential and limitations for stem
Evolution of disc degeneration in lumbar spine: a comparative his- cell therapy in disc regeneration. Spine (Phila Pa 1976) 30:
tological study between herniated and postmortem retrieved disc 23792387
specimens. J Spinal Disord 11:4145 Sakai D, Nakai T, Mochida J, Alini M, Grad S (2009) Differential phe-
Richardson JK, Chung T, Schultz JS, Hurvitz E (1997) A familial pre- notype of intervertebral disc cells: microarray and immunohis-
disposition toward lumbar disc injury. Spine (Phila Pa 1976) tochemical analysis of canine nucleus pulposus and anulus fibrosus.
22:14871492 Spine (Phila Pa 1976) 34:14481456
Richardson SM, Curran JM, Chen R, Vaughan-Thomas A, Hunt JA, Schollmeier G, Lahr-Eigen R, Lewandrowski KU (2000) Observations
Freemont AJ, Hoyland JA (2006a) The differentiation of bone mar- on fiber-forming collagens in the anulus fibrosus. Spine (Phila Pa
row mesenchymal stem cells into chondrocyte-like cells on poly-L- 1976) 25:27362741
lactic acid (PLLA) scaffolds. Biomaterials 27:40694078 Seki S, Kawaguchi Y, Chiba K, Mikami Y, Kizawa H, Oya T, Mio F,
Richardson SM, Walker RV, Parker S, Rhodes NP, Hunt JA, Freemont Mori M, Miyamoto Y, Masuda I, Tsunoda T, Kamata M, Kubo T,
AJ, Hoyland JA (2006b) Intervertebral disc cell-mediated mesen- Toyama Y, Kimura T, Nakamura Y, Ikegawa S (2005) A functional
chymal stem cell differentiation. Stem Cells 24:707716 SNP in CILP, encoding cartilage intermediate layer protein, is asso-
Richardson SM, Hughes N, Hunt JA, Freemont AJ, Hoyland JA (2008a) ciated with susceptibility to lumbar disc disease. Nat Genet
Human mesenchymal stem cell differentiation to NP-like cells in 37:607612
chitosan-glycerophosphate hydrogels. Biomaterials 29:8593 Selard E, Shirazi-Adl A, Urban JP (2003) Finite element study of nutri-
Richardson SM, Knowles R, Tyler J, Mobasheri A, Hoyland JA (2008b) ent diffusion in the human intervertebral disc. Spine (Phila Pa 1976)
Expression of glucose transporters GLUT-1, GLUT-3, GLUT-9 and 28:19451953
HIF-1alpha in normal and degenerate human intervertebral disc. Shen C, Yan J, Jiang LS, Dai LY (2011) Autophagy in rat annulus
Histochem Cell Biol 129:503511 fibrosus cells: evidence and possible implications. Arthritis Res
Richardson SM, Doyle P, Minogue BM, Gnanalingham K, Hoyland JA Ther 13:R132
(2009) Increased expression of matrix metalloproteinase-10, nerve Sive JI, Baird P, Jeziorsk M, Watkins A, Hoyland JA, Freemont AJ
growth factor and substance P in the painful degenerate interverte- (2002) Expression of chondrocyte markers by cells of normal and
bral disc. Arthritis Res Ther 11:R126 degenerate intervertebral discs. Mol Pathol 55:9197
Richardson SM, Purmessur D, Baird P, Probyn B, Freemont AJ, Smith SM, Whitelock JM, Iozzo RV, Little CB, Melrose J (2009)
Hoyland JA (2012) Degenerate human nucleus pulposus cells pro- Topographical variation in the distributions of versican, aggrecan
mote neurite outgrowth in neural cells. PLoS One 7(10):e47735. and perlecan in the foetal human spine reflects their diverse func-
PubMed PMID: 23091643 tional roles in spinal development. Histochem Cell Biol 132:
Risbud MV, Fertala J, Vresilovic EJ, Albert TJ, Shapiro IM (2005) 491503
Nucleus pulposus cells upregulate PI3K/Akt and MEK/ERK signal- Solovieva S, Kouhia S, Leino-Arjas P, la-Kokko L, Luoma K, Raininko
ing pathways under hypoxic conditions and resist apoptosis induced R, Saarela J, Riihimaki H (2004) Interleukin 1 polymorphisms and
by serum withdrawal. Spine (Phila Pa 1976) 30:882889 intervertebral disc degeneration. Epidemiology 15:626633
Solovieva S, Lohiniva J, Leino-Arjas P, Raininko R, Luoma K, la- vertebral disc: potential role in regulating neural ingrowth in the
Kokko L, Riihimaki H (2006) Intervertebral disc degeneration in degenerate intervertebral disc. Arthritis Res Ther 12:R1
relation to the COL9A3 and the IL-1ss gene polymorphisms. Eur Tortorella MD, Burn TC, Pratta MA, Abbaszade I, Hollis JM, Liu R,
Spine J 15:613619 Rosenfeld SA, Copeland RA, Decicco CP, Wynn R, Rockwell A,
Somerville RP, Longpre JM, Jungers KA, Engle JM, Ross M, Evanko S, Yang F, Duke JL, Solomon K, George H, Bruckner R, Nagase H,
Wight TN, Leduc R, Apte SS (2003) Characterization of ADAMTS-9 Itoh Y, Ellis DM, Ross H, Wiswall BH, Murphy K, Hillman MC Jr,
and ADAMTS-20 as a distinct ADAMTS subfamily related to Hollis GF, Newton RC, Magolda RL, Trzaskos JM, Arner EC (1999)
Caenorhabditis elegans GON-1. J Biol Chem 278:95039513 Purification and cloning of aggrecanase-1: a member of the
Sowa GA, Coelho JP, Bell KM, Zorn AS, Vo NV, Smolinski P, ADAMTS family of proteins. Science 284:16641666
Niyonkuru C, Hartman R, Studer RK, Kang JD (2011) Alterations Tschoeke SK, Hellmuth M, Hostmann A, Robinson Y, Ertel W,
in gene expression in response to compression of nucleus pulposus Oberholzer A, Heyde CE (2008) Apoptosis of human intervertebral
cells. Spine J 11:3643 discs after trauma compares to degenerated discs involving both
Stetler-Stevenson WG, Krutzsch HC, Liotta LA (1989) Tissue inhibitor receptor-mediated and mitochondrial-dependent pathways. J Orthop
of metalloproteinase (TIMP-2). A new member of the metallopro- Res 26:9991006
teinase inhibitor family. J Biol Chem 264:1737417378 Urban JP, Holm S, Maroudas A, Nachemson A (1982) Nutrition of the
Stewart WF, Ricci JA, Chee E, Morganstein D, Lipton R (2003) Lost intervertebral disc: effect of fluid flow on solute transport. Clin
productive time and cost due to common pain conditions in the US Orthop Relat Res (170):296302
workforce. JAMA 290:24432454 Urban MR, Fairbank JC, Etherington PJ, Loh FL, Winlove CP, Urban
Stoyanov JV, Gantenbein-Ritter B, Bertolo A, Aebli N, Baur M, Alini JP (2001) Electrochemical measurement of transport into scoliotic
M, Grad S (2011) Role of hypoxia and growth and differentiation intervertebral discs in vivo using nitrous oxide as a tracer. Spine
factor-5 on differentiation of human mesenchymal stem cells (Phila Pa 1976) 26:984990
towards intervertebral nucleus pulposus-like cells. Eur Cell Mater Urban JP, Smith S, Fairbank JC (2004) Nutrition of the intervertebral
21:533547 disc. Spine (Phila Pa 1976) 29:27002709
Studer RK, Vo N, Sowa G, Ondeck C, Kang J (2011) Human nucleus van der Roer N, Goossens ME, Evers SM, van Tulder MW (2005)
pulposus cells react to IL-6: independent actions and amplification What is the most cost-effective treatment for patients with low back
of response to IL-1 and TNF-alpha. Spine (Phila Pa 1976) pain? A systematic review. Best Pract Res Clin Rheumatol
36:593599 19:671684
Sztrolovics R, Alini M, Roughley PJ, Mort JS (1997) Aggrecan degra- Videman T, Leppavuori J, Kaprio J, Battie MC, Gibbons LE, Peltonen
dation in human intervertebral disc and articular cartilage. Biochem L, Koskenvuo M (1998) Intragenic polymorphisms of the vitamin D
J 326(Pt 1):235241 receptor gene associated with intervertebral disc degeneration.
Sztrolovics R, Grover J, Cs-Szabo G, Shi SL, Zhang Y, Mort JS, Spine (Phila Pa 1976) 23:24772485
Roughley PJ (2002) The characterization of versican and its mes- Wagner DR, Reiser KM, Lotz JC (2006) Glycation increases human
sage in human articular cartilage and intervertebral disc. J Orthop annulus fibrosus stiffness in both experimental measurements and
Res 20:257266 theoretical predictions. J Biomech 39:10211029
Takahashi M, Haro H, Wakabayashi Y, Kawa-uchi T, Komori H, Walker BF (2000) The prevalence of low back pain: a systematic review
Shinomiya K (2001) The association of degeneration of the inter- of the literature from 1966 to 1998. J Spinal Disord 13:205217
vertebral disc with 5a/6a polymorphism in the promoter of the Wallach D, Arumugam TU, Boldin MP, Cantarella G, Ganesh KA,
human matrix metalloproteinase-3 gene. J Bone Joint Surg Br Goltsev Y, Goncharov TM, Kovalenko AV, Rajput A, Varfolomeev
83:491495 EE, Zhang SQ (2002) How are the regulators regulated? The search
Takahashi Y, Ohtori S, Takahashi K (2009) Peripheral nerve pathways for mechanisms that impose specificity on induction of cell death
of afferent fibers innervating the lumbar spine in rats. J Pain and NF-kappaB activation by members of the TNF/NGF receptor
10:416425 family. Arthritis Res 4(Suppl 3):S189S196
Takaishi H, Nemoto O, Shiota M, Kikuchi T, Yamada H, Yamagishi M, Walsh AJ, Bradford DS, Lotz JC (2004) In vivo growth factor treatment of
Yabe Y (1997) Type-II collagen gene expression is transiently degenerated intervertebral discs. Spine (Phila Pa 1976) 29:156163
upregulated in experimentally induced degeneration of rabbit inter- Wang DL, Jiang SD, Dai LY (2007) Biologic response of the interver-
vertebral disc. J Orthop Res 15:528538 tebral disc to static and dynamic compression in vitro. Spine (Phila
Takebayashi T, Cavanaugh JM, Kallakuri S, Chen C, Yamashita T Pa 1976) 32:25212528
(2006) Sympathetic afferent units from lumbar intervertebral discs. Weiler C, Nerlich AG, Zipperer J, Bachmeier BE, Boos N (2002) 2002
J Bone Joint Surg Br 88:554557 SSE Award Competition in Basic Science: expression of major
Takegami K, An HS, Kumano F, Chiba K, Thonar EJ, Singh K, Masuda matrix metalloproteinases is associated with intervertebral disc deg-
K (2005) Osteogenic protein-1 is most effective in stimulating radation and resorption. Eur Spine J 11:308320
nucleus pulposus and annulus fibrosus cells to repair their matrix Weiler C, Nerlich AG, Bachmeier BE, Boos N (2005) Expression and
after chondroitinase ABC-induced in vitro chemonucleolysis. Spine distribution of tumor necrosis factor alpha in human lumbar inter-
J 5:231238 vertebral discs: a study in surgical specimen and autopsy controls.
Thompson JP, Pearce RH, Schechter MT, Adams ME, Tsang IK, Bishop Spine (Phila Pa 1976) 30:4453
PB (1990) Preliminary evaluation of a scheme for grading the gross Wilder DG, Pope MH, Magnusson M (1996) Mechanical stress reduc-
morphology of the human intervertebral disc. Spine (Phila Pa 1976) tion during seated jolt/vibration exposure. Semin Perinatol
15:411415 20:5460
Thompson JP, Oegema TR Jr, Bradford DS (1991) Stimulation of Wilke HJ, Neef P, Caimi M, Hoogland T, Claes LE (1999) New in vivo
mature canine intervertebral disc by growth factors. Spine (Phila Pa measurements of pressures in the intervertebral disc in daily life.
1976) 16:253260 Spine (Phila Pa 1976) 24:755762
Tim YS, Su KK, Li J, Soo PJ, Akamaru T, Elmer WA, Hutton WC Wolfe HJ, Putschar WG, Vickery AL (1965) Role of the notochord in
(2003) The effect of bone morphogenetic protein-2 on rat interver- human intervetebral disk. I. Fetus and infant. Clin Orthop Relat Res
tebral disc cells in vitro. Spine (Phila Pa 1976) 28:17731780 39:205212
Tolofari SK, Richardson SM, Freemont AJ, Hoyland JA (2010) Wuertz K, Godburn K, MacLean JJ, Barbir A, Donnelly JS, Roughley
Expression of semaphorin 3A and its receptors in the human inter- PJ, Alini M, Iatridis JC (2009) In vivo remodeling of intervertebral
discs in response to short- and long-term dynamic compression. Yu J, Winlove PC, Roberts S, Urban JP (2002) Elastic fibre organiza-
J Orthop Res 27:12351242 tion in the intervertebral discs of the bovine tail. J Anat 201:
Yao H, Gu WY (2006) Physical signals and solute transport in human 465475
intervertebral disc during compressive stress relaxation: 3D finite Zhao F, Pollintine P, Hole BD, Dolan P, Adams MA (2005) Discogenic
element analysis. Biorheology 43:323335 origins of spinal instability. Spine (Phila Pa 1976) 30:26212630
Yasuma T, Arai K, Yamauchi Y (1993) The histology of lumbar inter- Zhao CQ, Liu D, Li H, Jiang LS, Dai LY (2007a) Interleukin-1beta
vertebral disc herniation. The significance of small blood vessels in enhances the effect of serum deprivation on rat annular cell apopto-
the extruded tissue. Spine (Phila Pa 1976) 18:17611765 sis. Apoptosis 12:21552161
Ye W, Xu K, Huang D, Liang A, Peng Y, Zhu W, Li C (2011) Age- Zhao CQ, Wang LM, Jiang LS, Dai LY (2007b) The cell biology of
related increases of macroautophagy and chaperone-mediated intervertebral disc aging and degeneration. Ageing Res Rev 6:
autophagy in rat nucleus pulposus. Connect Tissue Res 52: 247261
472478 Zhao CQ, Zhang YH, Jiang SD, Li H, Jiang LS, Dai LY (2011)
Yu H, Zhu Y (2012) Expression of ADAMTS-7 and ADAMTS-12 in ADAMTS-5 and intervertebral disc degeneration: the results of tis-
nucleus pulposus during degeneration of rat caudal intervetebral sue immunohistochemistry and in vitro cell culture. J Orthop Res
disc. J Vet Med Sci 74:915 29:718725
Imaging Modalities
for Studying Disc Pathology 12
Ari Borthakur and Ravinder Reddy

12.1 Introduction ................................................................... 201
Low back pain is the second most frequent reason for a
12.2 Targeting Degenerative Disc Disease ........................... 202
physician visit and permanently disables more than 5 million
12.3 Radiographic Imaging .................................................. 202 Americans with annual costs of $100 billion in the USA
12.4 Computed Tomography ................................................ 202 (Sheehan 2010; Chou et al. 2007). While not well under-
stood, it is widely accepted that degenerative disc disease of
12.5 MRI Methods to Investigate
Spinal Morphology ........................................................ 203 the intervertebral disc contributes directly to axial back pain
(Deyo 2002). With degeneration, alterations in matrix com-
12.6 Biochemistry-Based MRI Techniques ......................... 204
12.6.1 T1R MRI ........................................................................... 204 position, structure, and mechanical loading occur in a pro-
12.6.2 Magnetization Transfer (MT) MRI ................................. 205 gressive cascade that leads to tissue breakdown and pain. It
12.6.3 Chemical Exchange Saturation Transfer would not be unreasonable to surmise that (1) aging alone
(CEST) MRI.................................................................... 206 does not account for degenerative disc disease since that
12.6.4 Ultrashort Echo Time (UTE) MRI .................................. 207
12.6.5 Sodium MRI.................................................................... 207 would imply that all older people have painful discs and (2)
success of treatment strategies aimed at halting the progres-
12.7 Path to Clinical Utility .................................................. 209
sion of degenerative disc disease-related chronic back pain
12.8 Summary of Critical Concepts Discussed will benefit from noninvasive methods to detect early changes
in the Chapter ................................................................ 209
in discs undergoing transition down the cascade.
References .................................................................................... 210 New imaging technologies (Sheehan 2010) have proven
useful for surgical planning and to categorize patients into
those with radiculopathy (nerve root pain) and stenosis (spi-
nal pathology). However, the majority of chronic low back
pain patients fall into a third category, nonspecific low back
pain, where no specific pathology can be identified, and con-
sequently, the use of advanced imaging has not been shown
to improve outcomes (Chou et al. 2007). In some cases,
imaging can increase the identification of incidental findings
that can trigger a cascade of diagnostic procedures or treat-
ments that can be costly and may be riskier to the patient
(Deyo 2002; Lurie et al. 2000). Consequently, current con-
sensus recommends against the routine use of advanced
imaging except in cases of suspected serious pathology and
A. Borthakur, PhD (*) R. Reddy, PhD
when conservative care is unsuccessful (Rubinstein and van
Department of Radiology,
Center for Magnetic Resonance and Optical Imaging, Tulder 2008; Koes et al. 2010).
Perelman School of Medicine at the Mixed pathologies are common in degenerative disc
University of Pennsylvania, disease, challenging the clinicians ability to identify the
B1 Stellar-Chance Laboratories,
single or dominant pathology responsible for the patients
422 Curie Boulevard, Philadelphia, PA 19104-6100, USA
e-mail: borthaku@upenn.edu; URL: http://med.cmroi.upenn.edu, pain. Hence, reducing the chances for unnecessary
krr@upenn.edu and unsuccessful treatment will require a shift from the

202 A. Borthakur and R. Reddy
traditional approach of detecting and diagnosing specific age and degeneration, the granular changes, mucoid
types of late-stage disc pathology based on the clinical phe- degeneration, clefts and tears, and decaying cells are seen
notype of back pain to one that includes the detection of the with increasing frequency throughout the disc (Boos et al.
responsible pathology as part of the diagnostic inclusion cri- 2002). Mechanical function of degenerated motion segments
teria. To this end, noninvasive imaging methods have the is compromised in all loading categories (Goel 1996;
potential to increase diagnostic certainty, particularly during Pope 1992). However, predicting mechanical changes with
the initial stage of disc degeneration, and to provide informa- degeneration is complicated because motion segment
tion about disease activity that can be used to measure treat- stiffness tends to decrease with moderate degeneration and
ment efficacy in clinical trials. increase with advanced degeneration under all loading
What follows are highlights of novel MRI-based imaging conditions (Berkson et al. 1979; Fujiwara et al. 2000;
tools that may localize functional attributes specific to patho- Haughton et al. 2000; Reuber et al. 1982).
logical disc degeneration rather than simply anatomical fea- Pathologic, painful degeneration can occur when matrix
tures readily visualized in conventional MRI, X-ray, or CT damage at disc margins exceeds the bodys ability to heal,
imaging. The novel approaches not only would detect pain- thereby inciting a wound reaction. Increased production of
ful spinal levels but also could sensitively monitor new and growth factors such as fibroblast growth factor, insulin-like
experimental therapies aimed at preventing or repairing dis- growth factor, and platelet-derived growth factor within
eased disc tissue. We begin with a description of the degen- annular fissures can encourage granulation tissue formation
erative changes amenable to targeting with imaging (Pratsinis and Kletsas 2008). Inflammatory stimulation of
biomarkers. disc cells causes secretion of neurotrophic factors, such as
nerve growth factor and brain-derived neurotrophic factor
that, along with decreases in proteoglycan, encourage neo-
12.2 Targeting Degenerative Disc Disease innervation and neovascularization at the vertebral endplate
and peripheral annulus (Purmessur et al. 2008). Inflammatory
The early stage of degenerative disc disease is characterized and neurotrophic factors may diffuse into the adjacent verte-
by loss of proteoglycan in the nucleus pulposus resulting in bra through endplate fissures leading to bone marrow edema
reduced capacity to bind water and a loss of hydration and (Ulrich et al. 2007; Crock 1986; Brown et al. 1997; Ohtori
pressure in the disc (Antoniou et al. 1996; Sieber and Kostuik et al. 2006). Endplate and annular nociceptors can be further
2004). In its late stages, degenerative disc disease is charac- sensitized by release of pro-inflammatory cell products, such
terized by a loss of disc height, annular tears and rim lesions, as TNF-alpha or lactic acid (Olmarker et al. 1995).
and osteophyte formation (Andersson 1998). It has been
implicated as a potential source of low back pain (Erkintalo
et al. 1995; Luoma et al. 2000; Urban and Maroudas 1979). 12.3 Radiographic Imaging
Marked compositional changes also occur with degenera-
tion: water and proteoglycan content decrease, proteoglycan The main objective of imaging the intervertebral disc is to
distribution changes, total collagen content increases, and provide the physician with a classification scheme that pro-
the distribution of collagen types changes (Eyre 1979; vides information concerning treatment options. The ideal
Roberts et al. 1991; Antoniou et al. 1996; Nerlich et al. classification system for disc degeneration is quantitative,
1998). The earliest structural degenerative changes occur in permits region-specific evaluation within the disc substruc-
the nucleus pulposus and the endplate (Buckwalter et al. tures, avoids observer bias, can detect early subtle changes,
2000; Boos et al. 2002). As early as the teenage years, the and correlates with clinical symptoms. X-ray radiographic
nucleus pulposus exhibits granular changes, clefts, tears, and detection of disc degeneration via the Thompson grading
cell death (Boos et al. 2002). The moderately fibrous nucleus scale (Thompson et al. 1990) is the most common clinical
pulposus during this time period is intact with no tears (Yu method in use but can only detect gross morphological defor-
et al. 1989). At this same early age, proteoglycan of the mities. Moreover, it does not depict soft tissues clearly and
nucleus pulposus begins to form clusters of short aggregated requires the injection of contrast agents to detect subtle disc
and non-aggregated molecules, the glycosaminoglycan con- abnormalities such as fissures.
tent decreases, and the water content falls (Antoniou et al.
1996; Buckwalter et al. 2000). Mechanically, the nucleus
pulposus is the first substructure to exhibit degenerative 12.4 Computed Tomography
changes. Its increased modulus and decreased hydrostatic
pressure result in a phase change from a fluidlike gel to a Computed tomography (CT) discography has been used to
more solid-like material (Urban and McMullin 1988; Iatridis improve surgical outcome (Tehranzadeh 1998). However, it
et al. 1997; Johannessen and Elliott 2005). With increasing is an invasive technique due to its use of radiation and some
12 Imaging Modalities for Studying Disc Pathology 203
cases require the placement of needle and injection of a these classification systems are qualitative, are susceptible to
contrast agent through the annular fibers; a further drawback observer bias, and are not specific for disc substructures. As
to its use is the high incidence of false positives (An and there is little correlation with presence or severity of clinical
Haughton 1993). While CT can accurately visualize disc symptoms, its usefulness in clinical decision-making is
morphology, it fails to distinguish findings that are dependent on correlating Pfirrmann grades with other clini-
symptomatic from those that are incidental. Related to these cal and radiographic findings.
problems, CT fails to demonstrate the cause of pain in To improve the capability of MRI-based methods to
patients without obvious nerve root compression. Dynamic objectively quantify disc degeneration, several studies have
CT can be employed to accurately measure ranges of motion measured T1 and T2 relaxation times and the diffusion
of vertebrae when a load or torque is applied, in vivo. This coefficient of water. In ex vivo disc specimens (Chatani et al.
technique is made possible due to the advent of high-speed 1993; Weidenbaum et al. 1992; Chiu et al. 2001), correla-
capability and image processing tools in CT scanners tions were observed between T2 and water content and
(Haughton 2004). significant differences noted in relaxation rates between the
human nucleus and annulus fibrosus (Marinelli et al. 2009).
And although an in vivo study (Jenkins et al. 1985) found no
12.5 MRI Methods to Investigate correlation between proton density and age, there were
Spinal Morphology significant differences in T1 and T2 between normal and
degenerated discs. However, a later study (Boos et al. 1994)
MRI with its exquisite soft tissue contrast has tremendous showed that the differences were only 196 ms for T1 and
potential for characterizing disc quality through morphologi- 15 ms for T2 and reproducibility for the measurements were
cal and functional parameters. In conventional (T2-weighted) low, at 16.4 and 13.4 %, respectively. Diurnal variations in T2
MR images, the nucleus pulposus appears bright and the due to changing water content from morning to evening in
annulus fibrosus is invisible due to its short T2 in the normal the same individual would further confound the T2 measure-
disc. Degeneration can be qualitatively graded by quantify- ments. However, quantitative T2 measurements may have a
ing the reduction in signal intensity of the nucleus pulposus clinical relevance, for example, if diurnal differences were
(Modic et al. 1984), while in advance stages of degeneration, present between cohorts of patients with back pain and
there is no clear distinction between nucleus and annulus. asymptomatic controls (Roberts et al. 1998). Indeed,
The widely used classification system of Pfirrmann (Pfirrmann significant decreases in these parameters were observed in
et al. 2001) is based on T2-weighted MR images. An integer both human nucleus and annulus fibrosus compartments with
grade (between I and V) is assigned to the disc, based on increase in Thompson grade and loading in disc specimens
structural morphology (e.g., homogeneity within the nucleus (Chiu et al. 2001). A decrease in the diffusion of water in the
pulposus, distinction between the nucleus and annulus, sig- nucleus was detected by diffusion-weighted MRI with reduc-
nal intensity, and disc height). Although this classification tion of proteoglycan ex vivo (Antoniou et al. 2004) and with
system is among the most widely accepted and used (Kettler degeneration in vivo compared to healthy controls (Kerttula
and Wilke 2005) and provides excellent detection of advanced et al. 2001; Kealey et al. 2005). However, diffusion-weighted
stage degeneration, integer-based classification systems can- MR images suffer from low SNR and resolution and are sus-
not discriminate among early degenerative changes ceptible to motion artifacts and difficult to reproduce mea-
(Bertagnoli and Kumar 2002; Luoma et al. 2001). Moreover, surements in vivo.
Box 12.1 What Is MRI? gradient fields so that the subsequently detected MRI sig-
Magnetic resonance imaging is a diagnostic imaging nal can be spatially encoded. The MR image is simply
method that uses radiofrequency (RF) waves to excite and a computer-processed map of hydrogen nuclei in tis-
detect a magnetic signal from the human body. The MRI sues. The 2003 Nobel Prize in physiology or medicine
system consists of a scanner that produces a strong mag- was awarded to Paul Lauterbur, of the University of
netic field in order to magnetize hydrogen nuclei of water Illinois, and Sir Peter Mansfield, of the University of
in tissues. The MRI scanner is typically cylindrical shaped Nottingham, for their invention of MRI.
and also contains magnetic gradient coils that can be elec- Unlike other imaging modalities like X-rays, com-
tronically controlled to alter the main magnetic field. puted tomography (CT), and positron emission tomogra-
A computer-controlled console has several pulse phy (PET), MRI uses no ionizing radiation and, hence, is
sequence software to dynamically manipulate the RF and ideally suited for clinical research.
MRI scanners differ in field strength, shape, dimensions, (4.7T and 9.4T) and are a more cost-effective option than
and bore size. Clinical scanners found in hospitals are clinical scanners.
typically 1.5T or 3T, in units of magnetic field strength Variations in T1, T2, T2*, and T1r relaxation time con-
(tesla), while a few research centers now have 7T whole- stants produce different MRI signal decay and recovery
body scanners. The higher field strength produces more characteristics in different tissues. A MRI technician will
magnetization and could potentially generate higher qual- utilize predefined protocols to generate several images
ity images. There are open-bore scanners that are more during a single scan session. The protocols manipulate
comfortable for some patients but come at the cost of parameters in the MRI pulse sequence in order to create
lower field strength. Smaller bore scanners used for small images with different contrasts to fulfill the clinical diag-
animal imaging typically attain higher field strengths nostic or research study requirements.
51,7, P 250
69,21, N 86 ms
95,19, N
62,17, N 93 ms
42,17, P
52,23, N 82 ms
45, 7, P 88,17, N
120 ms
0 ms
T2 T1 T2 T1 T2 T1
a b c
Fig. 12.1 Representative proton T2 MRI (grayscale) and correspond- sure (in psi) and whether discs were painful (P) or non-painful (N), both
ing T1r maps (in color overlaid on grayscale T1r-weighted image) of determined by discography are indicated in the lower back pain patients
the lumbar discs from a 52-year-old female (a) and a 35-year-old male (Borthakur et al. 2011). This data demonstrates the potential for nonin-
(b) patients diagnosed with low back pain and from an asymptomatic vasive T1r measurements as a surrogate measure of disc pressure in the
38-year-old male (c). Average T1r (in ms) was measured in the disc clinics
nucleus and is displayed below each disc, followed by the opening pres-
12.6 Biochemistry-Based MRI Techniques 12.6.1 T1R MRI
A major drawback of the MRI methods discussed above is Tumors, muscle, myocardium, blood flow, and cartilage
that while they effectively detect morphological changes in have been imaged using T1r(T-1-rho) (Santyr et al. 1989;
the disc, they lack sensitivity at the early stages of disc Lamminen et al. 1993; Dixon et al. 1996; Markkola et al.
degeneration. Since changes in nucleus pulposus and annu- 1997; Charagundla et al. 1998; Mlynarik et al. 1999; Grohn
lus fibrosus composition and structure are the earliest changes et al. 2000; Poptani et al. 2001; Duvvuri et al. 2001;
in degenerative disc disease, the newer MRI-based methods Borthakur et al. 2004; Wheaton et al. 2004; Hulvershorn
have approached the diagnostic process by attempting to et al. 2005). T1r MRI is an alternative to conventional T1 and
detect and quantify the biochemical composition of the T2 MRI (Borthakur et al. 2006) in which a long-duration,
extracellular matrix of the disc. The following techniques low-power radiofrequency (RF) referred to as spin-lock
have evolved over the last few years by several research (SL) pulse is applied to the magnetization in the transverse
groups for eventual clinical application. plane (Fig. 12.1). The spin-locked magnetization undergoes
T2 MRI T1 @ 50 Hz T1 @ 100 Hz T1 @ 200 Hz T1 @ 300 Hz
T1 @ 400 Hz T1 @ 500 Hz T1 @ 1,000 Hz T1 @ 1,500 Hz T1 @ 2,000 Hz
Fig. 12.2 Increased MRI signal was observed in the annulus fibrosus cially in the annulus fibrosus, at the standard operating frequency of
region in bovine disc specimens compared to T2 MRI (image in the top 500 Hz for clinical T1r MRI protocol is clearly visible. T1r MRI pro-
left) by increasing the spin-lock frequency of the T1r MRI pulse vides a twofold higher signal in the nucleus pulposus and fourfold
sequence. The qualitative difference between T2 and T1r images, espe- higher in the annulus fibrosus compared to T2 MRI
relaxation in the presence of a RF field (B1) in the rotating orientation (when the surface normal was 54.7 with respect
frame, a situation similar to that of the longitudinal magne- to B0), neither T2 nor T1r images demonstrated laminae.
tization in the main magnetic (B0) field. The spin-locked However, T1r values were greater than T2 at both orientations
magnetization will relax with a time constant T1r, called the throughout the cartilage layers.
spinlattice relaxation in the rotating frame, during the B1 In experiments on a bovine disc specimen (Fig. 12.2), we
field which attenuates the effect of signal loss mechanisms noted that the annulus signal could be enhanced by simply
(i.e., dipolar relaxation, static dipolar coupling, chemical increasing the spin-lock frequency of the T1r MRI pulse
exchange, and background gradients) on the MRI signal sequence compared to T2 MRI (image in the top left). The
(Borthakur et al. 2006). For this reason, T1r is always greater qualitative difference between T2 and T1r images, especially
than T2. In a typical T1r mapping experiment, the duration of in the annulus fibrosus, at the standard operating frequency
the SL pulse is incremented while the amplitude of SL pulse of 500 Hz for clinical T1r MRI protocol is obvious. T1r MRI
(gB1 ~ 0.1-few kHz) is fixed. T1r MRI has recently been used provides 2-fold higher signal in the nucleus and 4-fold higher
as a biomarker for degenerative disc disease with low values in the annulus compared to T2 MRI. This approach serves as
correlating with higher degeneration, low proteoglycan con- a method to overcome MRI signal loss mechanisms. The
tent, and reduced swelling pressure in the nucleus pulposus measurement of T1r as a function of the B1 amplitude is
(Johannessen et al. 2006; Auerbach et al. 2006; Nguyen known as T1r-dispersion and contrast is governed by the
et al. 2008; Wang et al. 2010b; Borthakur et al. 2011). spectral density components of the sample that are in the
Since annular fissures can be innervated, degeneration of neighborhood of gB1. While the literature on the contributors
the annulus fibrosus may be the source of back pain in to T1r dispersion in cartilage is sparse, the data in Fig. 12.2
patients with degenerative disc disease (Peng et al. 2006). clearly supports the notion that the non-averaged dipolar
Unfortunately, due to the non-averaged dipolar interaction of interaction between water protons associated with collagen
water bound to highly oriented collagen fibers, which shorten is the predominant contributor.
T2, the annulus appears dark in conventional MRI. The effect
of spin-locking on the laminar appearance of cartilage in
MRI, using orientation-dependent studies, indicated that 12.6.2 Magnetization Transfer (MT) MRI
when the normal to the surface of cartilage was parallel to
B0, a typical laminar appearance was present in T2-weighted The MT pulse sequence employs a pulse to selectively
images but was absent in T1r-weighted images of the saturate magnetization, and hence the MRI signal,
same specimen (Akella et al. 2004). At the magic angle from water bound to macromolecules such as collagen
c 0.4
a b
30-year-old
0.3
69-year-old 0.2
0.1
0
30-year-old 69-year-old 30-year-old 69-year-old MTR
Fig. 12.3 (a) Proton-density MRI of a 30-year-old healthy and a 69-year-old asymptomatic subject. (b) MTR map (in color), (c) MTR maps of
the L3/L2 disc (Wang et al. 2010c)
(Wolff et al. 1991). MT ratio (MTR) maps can be generated The asymmetric distribution of labile protons around the
from two images that are collected, with MT preparation central water peak in cartilage is readily detected in the dif-
(Ms) and without (M0). These maps are generated by setting ference image obtained after saturating symmetrically on
the off-resonance saturation pulse at 6.4 kHz down field of either side of the water. A very homogeneous B0 is neces-
the free water proton resonance frequency (Fig. 12.3). sary for a robust estimation of the difference signal.
Differences in lumbar discs can be seen in the proton-density Furthermore, care must also be taken to avoid the direct
MRI of a 30-year-old healthy individual and a 69-year-old saturation of the water peak due to the close resonances of
asymptomatic subject who exhibits decreased disc height, labile species of interest.
loss of nucleus pulposus hydration, as well as herniation in In spite of these stringent requirements, solutes in the
the lower lumbar discs. The corresponding MTR maps (in nanomolar to millimolar range have been reliably detected
color) showed increased values in the nucleus pulposus of using this approach in vitro (Aime et al. 1988; Zhang et al.
the 69-year-old subject, while there is a clear delineation of 2001; Goffeney et al. 2001; Gilad et al. 2007) and in vivo
the annulus fibrosus-nucleus pulposus boundary in the (Zhou et al. 2003; van Zijl et al. 2007). The feasibility of
30-year-old subject. High MTR values (in white) indicate CEST MRI to reliably detect glycosaminoglycan (GAG)
intact collagen in the annulus fibrosus of the younger in vivo was recently demonstrated in the cartilage of the knee
individual. A closer look at the MTR maps of the L3/L2 disc, joint (Schmitt et al. 2011) by comparing it with sodium MRI
for example, demonstrates a distinct boundary between MTR of the same joint and in the disc by Kim et al. (Fig. 12.4)
values within the nucleus pulposus and annulus fibrosus of utilizing a new water frequency mapping approach called
the 30-year-old subject, while it is diffuse in the 69-year-old water saturation shift referencing (WASSR), which pro-
subject, indicating a loss of collagen architecture in the vides a more accurate quantification of CEST effects (Kim
annulus and deposition of collagen into the nucleus et al. 2011). However, the gagCEST approach (Ling et al.
(Wang et al. 2010c). 2008) should be limited to high-field scanners such as 7T
since OH protons in cartilage are only 1 ppm downfield
from the water resonance and exchange at a rate of ~1,000 s1
12.6.3 Chemical Exchange Saturation (Schiller et al. 2001). Hence, the OH chemical shift is only
Transfer (CEST) MRI 129 Hz at 3FT and CEST MRI would introduce a substantial
direct water saturation. Further, the slow to intermediate
Chemical exchange saturation transfer (CEST) is a method exchange limit (i.e., chemical shift between exchanging site
to directly detect exchangeable solute protons in tissues by and water, Dw > exchange rate, k), required for optimal CEST
constant irradiation and saturation of their chemically effect, is not fulfilled in the case of OH protons at 3T.
shifted magnetization (Ward et al. 2000; Ward and Balaban However, at 7T and higher fields, the larger frequency sepa-
2000). The CEST signal is detected as a reduction in the ration of OH resonance from water fulfills both the (Dw > k)
water signal after the saturated spins undergo chemical condition and also reduces the direct water saturation effect
exchange with water (Forsen and Hoffman 1963), provided enabling a more reliable quantification of GAG at 7T using
that the exchange is more rapid than the T1 of either site. the gagCEST approach (Schmitt et al. 2011).
Fig. 12.4 Representative water 0.25 0.50 0.75

%
saturation shift referencing 8
a
(WASSR)-corrected
glycosaminoglycan chemical
exchange saturation transfer 6
(gagCEST) maps of intervertebral
disc (IVD) (zoomed) in two 4
subjects: (a) 25-year-old woman
at L5/S1 and (b) 54-year-old man
at L5/S1. The gagCEST maps are 2
shown at 0.25, 0.5, 0.75, 1, 1.25,
and 1.5 ppm. The broken circle 0
inside the map at 1.5 ppm shows 1.00 1.25 1.50 [ppm]
an approximation of the nucleus
pulposus (NP) and annulus 0.25 0.50 0.75
fibrosus (AF) regions. In both b %
subjects, the gagCEST effect is 8
highest in the 0.751 ppm
frequency range (Kim et al. 2011)
6
0
1.00 1.25 1.50 [ppm]
12.6.4 Ultrashort Echo Time (UTE) MRI method has been employed to visualize the collagen-rich
annulus fibrosus in human discs (Hall-Craggs et al. 2004) in
The vertebral endplate is composed of an inner bony and which patients with the most severe degeneration show irreg-
outer cartilaginous endplates. This endplate is the route for ular signal from the endplates compared to normal; however,
nutrient and metabolite supply to the avascular disc and is the reason for this observation is not clear (Fig. 12.5).
involved in metabolism, exchange of waste products, and
biomechanics of the disc (Urban and Winlove 2007). It has
previously been hypothesized that changes in disc mechanics 12.6.5 Sodium MRI
may be initiated by damage to the endplate (Adams and
Roughley 2006). Hence, visualizing morphological defects Based on the fact that Donnan equilibrium holds for cartilage
in the endplate could facilitate diagnosis of the cause of disc equilibrated in very dilute solutions, Maroudas et al. have
degeneration. shown that the fixed charge density of cartilage is correlated
However, it appears dark on T2, endplate integrity is not with the proteoglycan content of the extracellular matrix of
discernable by conventional T2 MRI. This is due to a strong cartilage (Maroudas et al. 1969). In addition, Gu et al.
dipolar interaction experienced by protons in water bound to reported that there was a reduction in the streaming potential
collagen resulting in short T2s (~110 ms). The UTE MRI (a measurement directly related to the fixed charge density)
pulse sequence overcomes this impediment by one of the two with intervertebral disc degeneration in tissue specimens (Gu
methods, selecting the short T2 components with RF pulses et al. 1999a, b). As the charge density is counterbalanced by
or combining images acquired at different echo times effec- Na+ activity, the loss of the negatively charged PG due to
tively producing a MRI of only the short T2 species (Robson degeneration lowers the fixed charge density in the tissue.
et al. 2003). There are several methods to perform UTE MRI Hence, there is a reduction in the concentration of positively
on clinical scanners. Some employ a narrow bandwidth exci- charged sodium ions and a lowering tissue osmotic
tation RF pulse to selectively nutate long T2 components into pressure.
the transverse plane, where they are nominally de-phased by Sodium (23Na) is one of the most NMR-visible nuclei in
gradients, leaving the short T2 components largely unaffected living systems. As a spin-3/2 nucleus with a nonzero quadru-
(Sussman et al. 1998; Larson et al. 2006). Other methods polar moment, 23Na exhibits bi-exponential relaxation in tis-
combine pairs of images such as those obtained by a half RF sues, i.e., fast and slow components of T2 and T2*. While
excitation, with and without gradient reversal, followed by nonquantitative sodium MRI of the spine has been performed
radial imaging (Gatehouse and Bydder 2003). The later (Insko et al. 2002), Wang et al. (2010a) recently imaged
Fig. 12.5 (a) Normal spine in a

33-year-old volunteer. Sagittal fat
a b
and long T2-suppressed ultrashort
echo time MRI of the lumbar and
lower thoracic discs display a faint
high signal in the region of the
annulus fibrosus and endplates and
low signal in the nucleus pulposus.
(b) Male aged 29 years with
b-thalassemia major-induced
degenerative disc disease.
High-signal bands are observed
parallel to the endplates at all
levels as well as more centrally
in the discs in some levels
(Hall-Craggs et al. 2004)
a 300 b
325
250 300 R = 0.71
y-intercept = 111.54 mM
Slice1 Slice2 Slice3 200 275
250
[Na] (mM)
150
225
100 200
Slice4 Slice5 Slice6 175
50
150
0 125
[Na] 100
Slice7 Slice8 Slice9 (mM) 0 25 50 75 100 125 150
[PG] (ug/mL/ww)
Fig. 12.6 Nine consecutive axial slices of a bovine disc specimen with [Na] ranging from 150 to 300 mM as calculated from sodium MRI (a).
A plot of [Na] from the nucleus pulposus (NP) correlated well (r = 0.7) with PG measured by DMMB blue assay (b) (Wang et al. 2010a)
sodium in the bovine intervertebral disc and noted a and L4L5 discs in the patient with back pain indicated a
significant correlation with proteoglycans measured by loss of proteoglycan, perhaps an early sign of disc disease.
DMMB assay (Fig. 12.6). Subsequent imaging of the human Due to its low gyromagnetic ratio (g) and concentrations
disc in vivo revealed differences in the sodium MRI between in tissues, sodium MRI requires field strengths 3T to
an asymptomatic subject and a subject with lower back pain obtain quality images to enable accurate quantification
(Fig. 12.7). The lower sodium concentration in both L3L4 of fixed charge density and proteoglycan. The short T2 of
300 300
Subject A Subject B
250 250
200 200
150 150
L4/L3 L4/L3 100 L4/L3 L4/L3 100
50 50
L5/L4 L5/L4 0 L5/L4 L5/L4 0

[Na] [Na]
(mM) (mM)
LBP patient healthy
Fig. 12.7 Representative proton T2 MRI (grayscale) and corresponding [Na] maps (color) obtained in vivo on a 22-year-old male with lower
lumbar trauma that resulted in chronic lower back pain and a 26-year-old healthy male (Wang et al. 2010a)
23
Na (~few ms) indicates that sodium MRI has to be performed approach gives the most timely, reliable, and accurate mea-
with MRI pulse sequences with ultrashort echo times. Since surement of the pathological changes due to degenerative
23
Na g is 1/4 of that of proton, sodium MRI requires 4 times disc disease?
stronger gradients to obtain images with identical resolution The experimental MRI CEST, MT, and T1r pulse sequences
to that of proton MRI. Consequently, sodium MRI pulse represent small modification to routine clinical imaging
sequences are forced to image at a long echo time (TE) of sequences and carry a potential risk of local tissue heating
2 ms, but again because of the short T2, substantial MRI from the RF pulses. A previously validated method has been
signal is lost before acquisition. Recent advances in the used to determine that the heat generated both theoretically
development of high-field 7-tesla MRI scanners and gradient and experimentally (Borthakur et al. 2003) by these pulse
technology (with a gradient strength of >4 G/cm) should sequences is not in violation of FDA guidelines (Borthakur
allow an ultrashort TE (<200 ms) that can significantly et al. 2004). Additional safeguards are also built into the
improve resolution and signal and provide hope for clinical scanner software, such that during MRI, the scanner continu-
sodium MRI. Radiofrequency coil technology (multiple ously monitors the RF power automatically stopping the scan
channel capability) and parallel imaging approaches such as when FDA-allowed limits are exceeded. This safeguard
SENSE (Pruessmann et al. 1999) and SMASH (Sodickson eliminates the risk associated with any inadvertent errors
and Manning 1997) and tuned preamplifiers could further during the scan. In addition, MRI near metallic implants
contribute to make clinical sodium MRI feasible. remains a challenge because of severe image artifacts mainly
stemming from large metal-induced field inhomogeneities
causing local gradient-induced eddy currents (Guermazi
12.7 Path to Clinical Utility et al. 2003). This is an impediment to routine MRI in back
pain patients who have implants, such as pedicle screws,
The biochemical-based MRI biomarker technologies dis- from prior surgery. To meet this challenge, new methods of
cussed here detect subtle molecular events that occur with MRI artifact reduction (Glover 1999; Lu et al. 2009) may
disc degeneration, and therefore, they are of considerable need to be incorporated into existing protocols in order to
clinical value. However, there remain some conflicts con- advance these MRI-based biomarker technologies into rou-
cerning their use (Lurie et al. 2000), while the natural his- tine clinical use.
tory of the subtle findings revealed by the new technologies
needs careful consideration and study (Deyo 2010). Further,
image biomarkers are considerably more complex than bio- 12.8 Summary of Critical Concepts
chemical biomarkers, and even if a composite biomarker can Discussed in the Chapter
be delineated in the future, it is not clear how this would
impact disease diagnosis. One potential answer is to utilize Noninvasive imaging methods have the potential to
statistical and machine-learning algorithms to combine bio- increase diagnostic certainty and to provide information
markers to build a classifier suitable for diagnosis (Breiman about disease activity that can be used to measure
1996; Guyon et al. 2002) to answer the question: which treatment efficacy in clinical trials.
Current imaging methods (X-ray, computed tomography, Borthakur A, Wheaton A, Charagundla SR, Shapiro EM, Regatte RR,
and MRI) can detect morphological changes in the disc, Akella SV et al (2003) Three-dimensional T1rho-weighted MRI at
1.5 Tesla. J Magn Reson Imaging 17(6):730736
but they lack sensitivity at the early stages of disc Borthakur A, Wheaton AJ, Gougoutas AJ, Akella SV, Regatte RR,
degeneration. Charagundla SR et al (2004) In vivo measurement of T1rho disper-
Novel MRI techniques such as T1r, magnetization transfer sion in the human brain at 1.5 tesla. J Magn Reson Imaging 19(4):
(MT), chemical exchange saturation transfer (CEST), 403409
Borthakur A, Mellon E, Niyogi S, Witschey W, Kneeland JB, Reddy R
ultrashort echo time (UTE) MRI, and sodium MRI are (2006) Sodium and T1rho MRI for molecular and diagnostic imag-
sensitive to matrix macromolecules such as proteoglycan ing of articular cartilage. NMR Biomed 19(7):781821
and collagen to varying degrees. Borthakur A, Maurer PM, Fenty M, Wang C, Berger R, Yoder J et al
These MRI-based biomarkers of IVD degeneration (2011) T1r MRI and discography pressure as novel biomarkers for
disc degeneration and low back pain. Spine 36:21902196
should be incorporated into existing research imaging Breiman L (1996) Heuristics of instability and stabilization in model
protocols in order to advance these technologies into selection. Ann Statist 24(6):23502383
routine clinical use. Brown MF, Hukkanen MV, McCarthy ID, Redfern DR, Batten JJ,
Crock HV et al (1997) Sensory and sympathetic innervation of the
vertebral endplate in patients with degenerative disc disease. J Bone
Acknowledgements This work was performed at an NCRR-supported Joint Surg Br 79(1):147153
Biomedical Technology Research Center (NIH RR02305) with Buckwalter JA, Boden SD, Eyre DR, Mow VC, Weidenbaum M (2000)
additional funding from an AOSpine research grant (AOSBRC-07-05). Intervertebral disk aging, degeneration, and herniation. In:
Buckwalter JA, Einhorn TA, Simon SR (eds) Orthopaedic basic
science. American Academy of Orthopaedic Surgeons, Rosemont
Charagundla SR, Stolpen AH, Leigh JS, Reddy R (1998) Off-resonance
References proton T1r dispersion imaging of 17O-enriched tissue phantoms.
Magn Reson Med 39(4):588595
Adams MA, Roughley PJ (2006) What is intervertebral disc degenera- Chatani K, Kusaka Y, Mifune T, Nishikawa H (1993) Topographic
tion, and what causes it? Spine 31:21512161 differences of 1H-NMR relaxation times (T1, T2) in the normal
Aime S, Nano R, Grandi M (1988) A new class of contrast agents for intervertebral disc and its relationship to water content. Spine
magnetic resonance imaging based on selective reduction of water- 18(15):22712275
T2 by chemical exchange. Invest Radiol 23(Suppl 1):S267S270 Chiu EJ, Newitt DC, Segal MR, Hu SS, Lotz JC, Majumdar S (2001)
Akella SV, Regatte RR, Wheaton AJ, Borthakur A, Reddy R (2004) Magnetic resonance imaging measurement of relaxation and water
Reduction of residual dipolar interaction in cartilage by spin-lock diffusion in the human lumbar intervertebral disc under compres-
technique. Magn Reson Med 52(5):11031109 sion in vitro. Spine 26(19):E437E444
An HS, Haughton VM (1993) Nondiscogenic lumbar radiculopathy: Chou R, Qaseem A, Snow V, Casey D, Cross JT, Shekelle P, Clinical
imaging considerations. Semin Ultrasound CT MR 14(6):414424 Efficacy Assessment Subcommittee of the American College of
Andersson GB (1998) What are the age-related changes in the spine? Physicians, American College of Physicians, and American Pain
Baillieres Clin Rheumatol 12(1):161173 Society Low Back Pain Guidelines Panel (2007) Diagnosis and
Antoniou J, Goudsouzian NM, Heathfield TF, Winterbottom N, treatment of low back pain: a joint clinical practice guideline from
Steffen T, Poole AR et al (1996) The human lumbar endplate. the American College of Physicians and the American Pain Society.
Evidence of changes in biosynthesis and denaturation of the extra- Ann Intern Med 147(7):478491
cellular matrix with growth, maturation, aging, and degeneration. Crock HV (1986) Internal disc disruption. A challenge to disc prolapse
Spine 21(10):11531161 fifty years on. Spine 11(6):650653
Antoniou J, Demers CN, Beaudoin G, Goswami T, Mwale F, Aebi M Deyo RA (2002) Cascade effects of medical technology. Annu Rev
et al (2004) Apparent diffusion coefficient of intervertebral discs Public Health 23:2344
related to matrix composition and integrity. Magn Reson Imaging Deyo RA (2010) Treatment of lumbar spinal stenosis: a balancing act.
22(7):963972 Spine J 10(7):625627
Auerbach JD, Johannessen W, Borthakur A, Wheaton AJ, Dolinskas CA, Dixon WT, Oshinski JN, Trudeau JD, Arnold BC, Pettigrew RI (1996)
Balderston RA et al (2006) In vivo quantification of human lumbar Myocardial suppression in vivo by spin locking with composite
disc degeneration using T(1rho)-weighted magnetic resonance pulses. Magn Reson Med 36(1):9094
imaging. Eur Spine J 15(Suppl 3):S338S344 Duvvuri U, Charagundla SR, Kudchodkar SB, Kaufman JH, Kneeland JB,
Berkson MH, Nachemson A, Schultz AB (1979) Mechanical properties Rizi R et al (2001) Human knee: in vivo T1(rho)-weighted MR imag-
of human lumbar spine motion segments part II: responses in ing at 1.5 Tpreliminary experience. Radiology 220(3):822826
compression and shear; influence of gross morphology. J Biomech Erkintalo MO, Salminen JJ, Alanen AM, Paajanen HE, Kormano MJ
Eng 101:5357 (1995) Development of degenerative changes in the lumbar inter-
Bertagnoli R, Kumar S (2002) Indications for full prosthetic disc vertebral disk: results of a prospective MR imaging study in adoles-
arthroplasty: a correlation of clinical outcome against a variety of cents with and without low-back pain. Radiology 196(2):529533
indications. Eur Spine J 11(Suppl 2):S131S136 Eyre DR (1979) Biochemistry of the intervertebral disc. Int Rev Connect
Boos N, Wallin A, Schmucker T, Aebi M, Boesch C (1994) Quantitative Tissue Res 8:227291
MR imaging of lumbar intervertebral disc and vertebral bodies: Forsen S, Hoffman RA (1963) Study of moderately rapid chemical
methodology, reproducibility, and preliminary results. Magn Reson exchange reactions by means of nuclear magnetic double resonance.
Imaging 12(4):577587 J Chem Phys 39:28922901
Boos N, Weissbach S, Rohrbach H, Weiler C, Spratt KF, Nerlich AG Fujiwara A, Lim TH, An HS, Tanaka N, Jeon CH, Andersson GB et al
(2002) Classification of age-related changes in lumbar interverte- (2000) The effect of disc degeneration and facet joint osteoarthritis
bral discs (2002 Volvo Award in basic science). Spine 27(23): on the segmental flexibility of the lumbar spine. Spine 25(23):
26312644 30363044
Gatehouse PD, Bydder GM (2003) Magnetic resonance imaging of relaxation time measurements to evaluate disc degeneration.
short T2 components in tissue. Clin Radiol 58(1):119 A quantitative MR study of young patients with previous vertebral
Gilad AA, McMahon MT, Walczak P, Winnard PT, Raman V, fracture. Acta Radiol 42(6):585591
van Laarhoven HW et al (2007) Artificial reporter gene providing Kettler A, Wilke HJ (2005) Review of existing grading systems for
MRI contrast based on proton exchange. Nat Biotechnol cervical or lumbar disc and facet joint degeneration. Eur Spine J
25(2):217219 15:705718
Glover GH (1999) 3D z-shim method for reduction of susceptibility Kim M, Chan Q, Anthony MP, Cheung KM, Samartzis D, Khong PL
effects in BOLD fMRI. Magn Reson Med 42(2):290299 (2011) Assessment of glycosaminoglycan distribution in human
Goel VK (1996) Biomechanics of disk herniation and related issues. In: lumbar intervertebral discs using chemical exchange saturation
Weinstein JN, Gordon SL (eds) Low back pain: a scientific and transfer at 3 T: feasibility and initial experience. NMR Biomed
clinical overview. American Academy of Orthopaedic Surgeons, 24:11371144
Rosemont Koes BW, van Tulder M, Lin CW, Macedo LG, McAuley J, Maher C
Goffeney N, Bulte JW, Duyn J, Bryant LH, van Zijl PC (2001) Sensitive (2010) An updated overview of clinical guidelines for the manage-
NMR detection of cationic-polymer-based gene delivery systems ment of non-specific low back pain in primary care. Eur Spine J
using saturation transfer via proton exchange. J Am Chem Soc 19(12):20752094
123(35):86288629 Lamminen AE, Tanttu JI, Sepponen RE, Pihko H, Korhola OA (1993)
Grohn OHJ, Kettunen MI, Makela HI, Penttonen M, Pitkanen A et al T1rho dispersion imaging of diseased muscle tissue. Br J Radiol
(2000) Early detection of irreversible cerebral ischemia in the rat 66(789):783787
using dispersion of the magnetic resonance imaging relaxation time, Larson PE, Gurney PT, Nayak K, Gold GE, Pauly JM, Nishimura DG
T1rho. J Cereb Blood Flow Metab 20(10):14571466 (2006) Designing long-T2 suppression pulses for ultrashort echo
Gu WY, Mao XG, Foster RJ, Weidenbaum M, Mow VC, Rawlins BA time imaging. Magn Reson Med 56(1):94103
(1999a) The anisotropic hydraulic permeability of human lumbar Ling W, Regatte RR, Navon G, Jerschow A (2008) Assessment of
anulus fibrosus. Influence of age, degeneration, direction, and water glycosaminoglycan concentration in vivo by chemical exchange-
content. Spine 24(23):24492455 dependent saturation transfer (gagCEST). Proc Natl Acad Sci U S A
Gu WY, Mao XG, Rawlins BA, Iatridis JC, Foster RJ, Sun DN et al 105(7):22662270
(1999b) Streaming potential of human lumbar anulus fibrosus is Lu W, Pauly KB, Gold GE, Pauly JM, Hargreaves BA (2009) SEMAC:
anisotropic and affected by disc degeneration. J Biomech 32(11): slice encoding for metal artifact correction in MRI. Magn Reson
11771182 Med 62(1):6676
Guermazi A, Miaux Y, Zaim S, Peterfy CG, White D, Genant HK Luoma K, Riihimaki H, Luukkonen R, Raininko R, Viikari-Juntura E,
(2003) Metallic artefacts in MR imaging: effects of main field orien- Lamminen A (2000) Low back pain in relation to lumbar disc
tation and strength. Clin Radiol 58(4):322328 degeneration. Spine 25(4):487492
Guyon I, Weston J, Barnhill S, Vapnik V (2002) Gene selection for Luoma K, Vehmas T, Riihimaki H, Raininko R (2001) Disc height and
cancer classification using support vector machines. Mach Learn signal intensity of the nucleus pulposus on magnetic resonance
46(1):389422 imaging as indicators of lumbar disc degeneration. Spine 26(6):
Hall-Craggs MA, Porter J, Gatehouse PD, Bydder GM (2004) Ultrashort 680686
echo time (UTE) MRI of the spine in thalassaemia. Br J Radiol Lurie JD, Gerber PD, Sox HC (2000) Clinical problem-solving. A pain
77(914):104110 in the back. N Engl J Med 343(10):723726
Haughton V (2004) Medical imaging of intervertebral disc degeneration: Marinelli NL, Haughton VM, Muoz A, Anderson PA (2009) T2 relax-
current status of imaging. Spine (Phila Pa 1976) 29(23):27512756 ation times of intervertebral disc tissue correlated with water content
Haughton VM, Schmidt TA, Keele K, An HS, Lim TH (2000) Flexibility and proteoglycan content. Spine (Phila Pa 1976) 34(5):520524
of lumbar spinal motion segments correlated to type of tears in the Markkola AT, Aronen HJ, Paavonen T, Hopsu E, Sipila LM, Tanttu JI,
annulus fibrosus. J Neurosurg 92(1 Suppl):8186 Sepponen RE (1997) T1rho dispersion imaging of head and neck
Hulvershorn J, Borthakur A, Bloy L, Gualtieri EE, Reddy R, Leigh JS, tumors: a comparison to spin lock and magnetization transfer
Elliott MA (2005) T1rho contrast in functional magnetic resonance techniques. J Magn Reson Imaging 7(5):873879
imaging. Magn Reson Med 54(5):11551162 Maroudas A, Muir H, Wingham J (1969) The correlation of fixed
Iatridis JC, Setton LA, Weidenbaum M, Mow VC (1997) Alterations in negative charge with glycosaminoglycan content of human articular
the mechanical behavior of the human lumbar nucleus pulposus cartilage. Biochim Biophys Acta 177(3):492500
with degeneration and aging. J Orthop Res 15:318322 Mlynarik V, Trattnig S, Huber M, Zembsch A, Imhof H (1999) The role
Insko EK, Clayton DB, Elliott MA (2002) In vivo sodium MR imaging of relaxation times in monitoring proteoglycan depletion in articular
of the intervertebral disk at 4 T. Acad Radiol 9(7):800804 cartilage. J Magn Reson Imaging 10(4):497502
Jenkins JP, Hickey DS, Zhu XP, Machin M, Isherwood I (1985) MR Modic MT, Pavlicek W, Weinstein MA, Boumphrey F, Ngo F, Hardy R
imaging of the intervertebral disc: a quantitative study. Br J Radiol et al (1984) Magnetic resonance imaging of intervertebral disk
58(692):705709 disease. Clinical and pulse sequence considerations. Radiology
Johannessen W, Elliott DM (2005) Effects of degeneration on the 152(1):103111
biphasic material properties of human nucleus pulposus in confined Nerlich AG, Boos N, Wiest I, Aebi M (1998) Immunolocalization of
compression. Spine 30(24):E724E729 major interstitial collagen types in human lumbar intervertebral
Johannessen W, Auerbach JD, Wheaton AJ, Kurji A, Borthakur A, discs of various ages. Virchows Arch 432(1):6776
Reddy R et al (2006) Assessment of human disc degeneration and Nguyen AM, Johannessen W, Yoder JH, Wheaton AJ, Vresilovic EJ,
proteoglycan content using T1rho-weighted magnetic resonance Borthakur A et al (2008) Noninvasive quantification of human
imaging. Spine 31(11):12531257 nucleus pulposus pressure with use of T1rho-weighted magnetic
Kealey SM, Aho T, Delong D, Barboriak DP, Provenzale JM, Eastwood resonance imaging. J Bone Joint Surg Am 90(4):796802
JD (2005) Assessment of apparent diffusion coefficient in normal Ohtori S, Inoue G, Ito T, Koshi T, Ozawa T, Doya H et al (2006) Tumor
and degenerated intervertebral lumbar disks: initial experience. necrosis factor-immunoreactive cells and PGP 9.5-immunoreactive
Radiology 235(2):569574 nerve fibers in vertebral endplates of patients with discogenic low
Kerttula L, Kurunlahti M, Jauhiainen J, Koivula A, Oikarinen J, back Pain and Modic Type 1 or Type 2 changes on MRI. Spine
Tervonen O (2001) Apparent diffusion coefficients and T2 31(9):10261031
Olmarker K, Blomquist J, Strmberg J, Nannmark U, Thomsen P, Sussman MS, Pauly JM, Wright GA (1998) Design of practical
Rydevik B (1995) Inflammatogenic properties of nucleus pulposus. T2-selective RF excitation (TELEX) pulses. Magn Reson Med
Spine 20(6):665669 40(6):890899
Peng B, Hao J, Hou S, Wu W, Jiang D, Fu X et al (2006) Possible Tehranzadeh J (1998) Discography 2000. Radiol Clin North Am
pathogenesis of painful intervertebral disc degeneration. Spine 36(3):463495
31(5):560566 Thompson JP, Pearce RH, Schechter MT, Adams ME, Tsang IK, Bishop PB
Pfirrmann CW, Metzdorf A, Zanetti M, Hodler J, Boos N (2001) (1990) Preliminary evaluation of a scheme for grading the gross mor-
Magnetic resonance classification of lumbar intervertebral disc phology of the human intervertebral disc. Spine 15(5):411415
degeneration. Spine 26(17):18731878 Ulrich JA, Liebenberg EC, Thuillier DU, Lotz JC (2007) ISSLS prize
Pope MH (1992) Intervertebral disk: mechanical changes that occur winner: repeated disc injury causes persistent inflammation. Spine
with age. In: Buckwalter JA, Goldberg VM, Woo SLY (eds) 32(25):28122819
Musculoskeletal soft-tissue aging: impact on mobility. American Urban JPG, Maroudas A (1979) The measurement of fixed charged
Academy of Orthopaedic Surgeons, Rosemont density in the intervertebral disc. Biochimica et Biophysica Acta
Poptani H, Duvvuri U, Miller CG, Mancuso A, Charagundla S, Fraser (BBA) General Subjects 586(1):166178
NW et al (2001) T1rho imaging of murine brain tumors at 4 T. Acad Urban JP, McMullin JF (1988) Swelling pressure of the lumbar inter-
Radiol 8(1):4247 vertebral discs: influence of age, spinal level, composition, and
Pratsinis H, Kletsas D (2008) Growth factors in intervertebral disc degeneration. Spine 13(2):179187
homeostasis. Connect Tissue Res 49(3):273276 Urban JP, Winlove CP (2007) Pathophysiology of the intervertebral disc
Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P (1999) and the challenges for MRI. J Magn Reson Imaging 25(2):419432
SENSE: sensitivity encoding for fast MRI. Magn Reson Med van Zijl PC, Jones CK, Ren J, Malloy CR, Sherry AD (2007) MRI
42(5):952962 detection of glycogen in vivo by using chemical exchange satura-
Purmessur D, Freemont AJ, Hoyland JA (2008) Expression and regulation transfer imaging (glycoCEST). Proc Natl Acad Sci U S A
tion of neurotrophins in the nondegenerate and degenerate human 104(11):43594364
intervertebral disc. Arthritis Res Ther 10(4):R99 Wang C, McArdle E, Fenty M, Witschey W, Elliott M, Sochor M,
Reuber M, Schultz A, Denis F, Spencer D (1982) Bulging of lumbar Reddy R, Borthakur A (2010a) Validation of sodium magnetic reso-
intervertebral discs. J Biomech Eng 104:187192 nance imaging of intervertebral disc. Spine 35(5):505510
Roberts S, Menage J, Duance V, Wotton S, Ayad S (1991) Collagen Wang C, Witschey W, Elliott MA, Borthakur A, Reddy R (2010b)
types around the cells of the intervertebral disc and cartilage end Measurement of intervertebral disc pressure with T 1r MRI. Magn
plate: an immunolocalization study (1991 Volvo Award in basic Reson Med 64(6):17211727
sciences). Spine 16(9):10301038 Wang C, Witschey W, Goldberg A, Elliott M, Borthakur A, Reddy R
Roberts N, Hogg D, Whitehouse GH, Dangerfield P (1998) Quantitative (2010c) Magnetization transfer ratio mapping of intervertebral disc
analysis of diurnal variation in volume and water content of lumbar degeneration. Magn Reson Med 64(5):15201528
intervertebral discs. Clin Anat 11(1):18 Ward KM, Balaban RS (2000) Determination of pH using water
Robson MD, Gatehouse PD, Bydder M, Bydder GM (2003) Magnetic protons and chemical exchange dependent saturation transfer
resonance: an introduction to ultrashort TE (UTE) imaging. J (CEST). Magn Reson Med 44(5):799802
Comput Assist Tomogr 27(6):825846 Ward KM, Aletras AH, Balaban RS (2000) A new class of contrast
Rubinstein SM, van Tulder M (2008) A best-evidence review of diag- agents for MRI based on proton chemical exchange dependent
nostic procedures for neck and low-back pain. Best Pract Res Clin saturation transfer (CEST). J Magn Reson 143(1):7987
Rheumatol 22(3):471482 Weidenbaum M, Foster RJ, Best BA, Saed-Nejad F, Nickoloff E,
Santyr GE, Henkelman RM, Bronskill MJ (1989) Spin locking for Newhouse J et al (1992) Correlating magnetic resonance imaging
magnetic resonance imaging with application to human breast. with the biochemical content of the normal human intervertebral
Magn Reson Med 12(1):2537 disc. J Orthop Res 10(4):552561
Schiller J, Naji L, Huster D, Kaufmann J, Arnold K (2001) 1H and 13C Wheaton AJ, Casey FL, Gougoutas AJ, Dodge GR, Borthakur A,
HR-MAS NMR investigations on native and enzymatically digested Lonner JH et al (2004) Correlation of T1rho with fixed charge
bovine nasal cartilage. MAGMA 13(1):1927 density in cartilage. J Magn Reson Imaging 20(3):519525
Schmitt B, Zbyn S, Stelzeneder D, Jellus V, Paul D, Lauer L et al (2011) Wolff SD, Chesnick S, Frank JA, Lim KO, Balaban RS (1991)
Cartilage quality assessment by using glycosaminoglycan chemical Magnetization transfer contrast: MR imaging of the knee. Radiology
exchange saturation transfer and Na MR imaging at 7 T. Radiology 179(3):623628
260:257264 Yu S, Haughton VM, Sether LA, Ho KC, Wagner M (1989) Criteria for
Sheehan NJ (2010) Magnetic resonance imaging for low back pain: classifying normal and degenerated lumbar intervertebral disks.
indications and limitations. Ann Rheum Dis 69(1):711 Radiology 170(2):523526
Sieber AN, Kostuik JP (2004) Concepts in nuclear replacement. Spine Zhang S, Winter P, Wu K, Sherry AD (2001) A novel europium(III)-
J 4(6 Suppl):322S324S based MRI contrast agent. J Am Chem Soc 123(7):15171518
Sodickson DK, Manning WJ (1997) Simultaneous acquisition of spatial Zhou J, Lal B, Wilson DA, Laterra J, van Zijl PC (2003) Amide proton
harmonics (SMASH): fast imaging with radiofrequency coil arrays. transfer (APT) contrast for imaging of brain tumors. Magn Reson
Magn Reson Med 38(4):591603 Med 50(6):11201126
Surgical Indications for Lumbar
Degenerative Disease 13
Ravi R. Patel, Jeffrey A. Rihn, Ravi K. Ponnoppan,
and Todd J. Albert
Contents 13.6 Concluding Comments .................................................. 222
13.1 Introduction .................................................................... 213 13.7 Summary of Critical Concepts Discussed

in the Chapter ................................................................. 223
13.2 Discogenic Low Back Pain............................................. 214
13.2.1 Epidemiology of Low Back Pain ..................................... 214 References ..................................................................................... 223
13.2.2 Pathogenesis of Disc Disease ........................................... 214
13.2.3 Clinical Presentation of Discogenic Disease ................... 214
13.2.4 Evaluation of the Spine by Imaging Technologies .......... 215 13.1 Introduction
13.2.5 Treatment of Discogenic Pain .......................................... 215
13.3 Intervertebral Disc Herniation ..................................... 216 Lumbar degenerative disease is a common and debilitat-
13.3.1 Epidemiology of Disc Herniation .................................... 216
ing condition that describes the age-dependent pathologic
13.3.2 Pathoetiology of Disc Herniation ..................................... 217
13.3.3 Clinical Presentation of Herniation .................................. 217 changes that occur in the spine. Degenerative disease of the
13.3.4 Imaging of Herniation ...................................................... 217 lumbar spine accounts for a relatively large proportion of the
13.3.5 Herniation Treatment ....................................................... 218 annual health-care expenditures in the USA and is respon-
13.4 Lumbar Stenosis............................................................. 219 sible for considerable indirect losses due to time-off of work
13.4.1 Epidemiology of Lumbar Stenosis................................... 219 (Luo et al. 2008). A study by Hanson et al. assessed the health
13.4.2 Etiology of Lumbar Stenosis............................................ 219 status of patients with musculoskeletal pathology requiring
13.4.3 Clinical Presentation of Lumbar Stenosis ........................ 219
13.4.4 Imaging of Lumbar Stenosis ............................................ 219 surgical treatment and found that patients with lumbar spinal
13.4.5 Treatment of Lumbar Stenosis ......................................... 220 conditions (i.e., chronic low back pain, stenosis, and spon-
dylolisthesis) suffered from considerable disability when
13.5 Degenerative Spondylolisthesis ..................................... 220
13.5.1 Epidemiology of Spondylolisthesis ................................. 220 compared to patients with other orthopedic conditions, such
13.5.2 Etiology of Spondylolisthesis .......................................... 221 as hip and knee arthritis (Hansson et al. 2008).
13.5.3 Clinical Presentation of Spondylolisthesis....................... 221 Degenerative changes that occur in the lumbar spine with
13.5.4 Imaging of Spondylolisthesis ........................................... 221
aging include intervertebral disc degeneration, facet arthrop-
13.5.5 Treatment of Spondylolisthesis ........................................ 221
athy and hypertrophy, and ligamentum flavum hypertrophy.
This process may manifest as discogenic low back pain, disc
herniation, spinal stenosis, and/or spondylolisthesis. It can
cause a variety of symptoms including low back pain, neuro-
R.R. Patel, MD genic claudication, or radicular symptoms, including pain,
numbness, tingling, and weakness in the affected nerve root
Emory University School of Medicine,
49 Jesse Hill Jr. Drive, #307, Atlanta, GA 30303, USA distribution. The pathophysiology producing these symp-
e-mail: ravi.r.patel@gmail.com toms is multifactorial and often a specific pain generator can-
J.A. Rihn, MD (*) T.J. Albert, MD not be identified. Each component of the functional spinal
Department of Orthopaedic Surgery, unit undergoes changes with aging and degeneration. It is not
Thomas Jefferson University Hospital, The Rothman Institute, clear, however, why some people become symptomatic and
925 Chestnut Street, 5th Floor, Philadelphia, PA 19107, USA
others do not. Boden et al. reviewed the lumbar spine MRI
e-mail: jrihno16@yahoo.com; tjsurg@aol.com
studies in patients who had never experienced back pain or
R.K. Ponnoppan, MD
sciatica and noted that in patients under age 60, 22 % had a
Private Orthopaedic Practice, Jersey Spine Associates,
110 Roosevelt Blvd #2w, Marmora, NJ 08223, USA disc herniation, 54 % had a disc bulge, and 46 % had disc
e-mail:rkp500@gmail.com degeneration. In patients over age 60, the percentages

214 R.R. Patel et al.
increased to 36, 79 and 93 %, respectively (Boden et al. Yong-Hing and Kirkaldy-Willis 1983). These stages were
1999a, b). The purpose of this chapter is to provide an over- dysfunction, instability, and stabilization (Yong-Hing and
view of the most common manifestations of lumbar degen- Kirkaldy-Willis 1983). However, it is important to note that
eration and the indications for surgical management of these this cascade of individual motion segment degeneration is a
conditions. continuum rather than as three definable and separate stages.
Most commonly, the L45 and L5S1 discs typically are the
first two lumbar levels to undergo degenerative changes. The
13.2 Discogenic Low Back Pain initial change that occurs during disc degeneration is circum-
ferential fissuring of the outer annulus fibrosus. This is
13.2.1 Epidemiology of Low Back Pain thought to be the result of repetitive microtrauma and subse-
quent disruption in the intervertebral vascular supply and
Disc degeneration and low back pain is a difficult condition impairment of the normal disc metabolism (Yong-Hing and
to evaluate. Chronic low back pain is the most common cause Kirkaldy-Willis 1983). Over time, there is delamination of
of morbidity and chronic pain in the USA (Deyo et al. 2006). the annular layers and the circumferential tears coalesce to
It is estimated that up to 80 % of individuals will experience form larger radial tears. There is a decrease in the amount
low back pain at one time in their lives (Walker 2000). For a and organization of proteoglycans in the nucleus pulposus
detailed discussion of the epidemiology of back pain, see (Yong-Hing and Kirkaldy-Willis 1983). Vertebral segment
Chaps. 9 and 16. The majority of mechanical nonspecific low instability occurs, and this results in a decline in the amount
back pain requires little or no formal medical treatment and of nuclear proteoglycan composition with a resulting fall in
should be an infrequent indication for surgical intervention. osmotic pressure and of water content (Yong-Hing and
Surgery for chronic low back pain in the absence of defor- Kirkaldy-Willis 1983). Over time, there are continued end-
mity or instability is generally reserved for when a degener- stage tissue damage and unsuccessful attempts at repair. The
ated disc is thought to be the primary generator of end result is intervertebral disc space narrowing, fibrosis,
non-radicular discogenic pain. end plate irregularities, and osteophyte formation. The
pathophysiology of the degenerative process is described in
considerable detail in Chap. 11.
13.2.2 Pathogenesis of Disc Disease The changes that occur with intervertebral disc degenera-
tion have a major effect on disc function and its load-bearing
Discogenic low back pain is the result of a complex cas- capacity. With matrix disorganization and the fragmentation
cade of degenerative changes. The intervertebral discs lie and loss of proteoglycan, the osmotic pressure of the disc is
between each vertebral segment and they provide motion reduced and it is subsequently less able to maintain hydra-
and flexibility to the spine. They are approximately 710 mm tion under load (Yong-Hing and Kirkaldy-Willis 1983; Lyons
thick and 4 cm in diameter in the lumbar section (Twomey et al. 1981). Inappropriate stress develops along the end plate
and Taylor 1987; Roberts et al. 1989). The disc consists of and in the annulus.
two distinct anatomic regions: the outer annulus fibrosus and The relationship between disc degeneration and low back
the inner nucleus pulposus. The annulus fibrosus in the lumbar pain is incompletely understood. Some patients with pro-
spine has upwards of 25 layers containing mostly collagen I nounced degenerative changes will have low back pain while
arranged in a dense parallel pattern (Roberts 2002). The intri- others with a similar degree of pathology are pain-free.
cate cross-linked configuration of the collagen fibrils allows Indeed, approximately one third of asymptomatic individu-
the intervertebral disc to resist the tensile forces that occur in als will have degenerative changes apparent on a lumbar
the lumbar spine during bending and torsional movements. MRI scan (Boden et al. 1999a, b). Factors that may contrib-
The central nucleus pulposus consists of predominantly colla- ute to the generation of pain include altered spine biome-
gen II fibrils within a rich proteoglycan matrix. This composi- chanics, neural hypersensitivity, and neurovascular ingrowth
tion produces a highly viscoelastic core that allows resistance into the disc.
to axial loads (Buckwalter et al. 2002). A third morphologi-
cally distinct region that is often described is the cartilaginous
end plate. The end plate is a thin horizontal layer of hyaline 13.2.3 Clinical Presentation
cartilage measuring less than 1 mm in thickness. Its collagen of Discogenic Disease
fibers interface the disc and the vertebral body (Johnson et al.
2001). For more detailed information on the structure of the Discogenic disease of the lumbar spine may produce low
intervertebral disc, see Chaps. 4 and 5. back pain. Acute low back pain is defined as pain lasting less
Kirkaldy-Willis described a widely accepted pathway that than 3 months and chronic pain as greater than 3 months.
divided lumbar disc degeneration into three stages based on Patient complaints are often of nonspecific back symptoms.
the amount of damage that has occurred (Johnson et al. 2001; Typically, discogenic pain is associated with activities that
13 Surgical Indications for Lumbar Degenerative Disease 215
increase the pressure within the intervertebral disc such as

sitting and trunk flexion. Most episodes of acute low back
pain are self-limiting. It is important for the clinician to
maintain an awareness of the so-called red flags that may
signify a serious underlying pathology such as fracture,
infection, or malignancy.
13.2.4 Evaluation of the Spine

by Imaging Technologies
AP and lateral radiographs should be the first study in the

evaluation of the lumbar spine. These views should be
reserved for patients who have had low back pain for 6 weeks.
However, earlier imaging is warranted if there is a concern
for serious pathology (red flags). MRI should be performed
if back pain is unresponsive to conservative management for
3 months. Plain radiographs, magnetic resonance imaging
(MRI), and computed tomography (CT) are sensitive to
degenerative changes but clinically are nonspecific (Boden
et al. 1999a, b; Wiesel et al. 1984), as indicated by the rela-
tively high incidence of degenerative changes noted by the
MRI study by Boden et al. in asymptomatic patients (Boden
et al. 1999a, b) (Fig. 13.1).
Discography can be a useful tool to differentiate the
source in discogenic low back pain. The test involves pres- Fig. 13.1 Sagittal T2-weighted lumbar MRI image demonstrating
surizing the disc with contrast dye in an attempt to stimulate degeneration of the L45 intervertebral disc (white arrow). Water has
nerve endings in injured discs. Discography can provide high signal intensity on T2-weighted MRI imaging. As the disc degen-
erates, it loses its water content and, as such, loses its high signal inten-
information regarding the morphology of the degenerated
sity compared to the healthy disc. The degenerative disc becomes grey
disc and can assist with identifying the pain generator or black and is often referred to as a dark disc. Note that the L34 and
(Fig. 13.2). Precise reproduction of the patients presenting L5S1 discs above and below the L45 level still have high signal
pain symptoms, or concordance, makes the test clinically intensity
useful. Some studies have identified a high level of sensitiv-
ity and specificity, whereas others have disputed this claim back pain, systematic review has shown no or insufficient evi-
(Mooney et al. 1988; Walsh et al. 1990; Simmons et al. 1988; dence to support the use of botulinum toxin injections, local
Carragee et al. 2000; Guyer and Ohnmeiss 1995). Carragee injections, prolotherapy, intraspinal steroid injections, epi-
et al. showed discography to have a best-case positive pre- dural injections, facet joint or intradiscal steroid injections,
dictive value of 5060 % (Carragee et al. 2006). The subjec- sacroiliac injections or medial branch blocks, intradiscal
tive nature of the test can never completely be overcome, but electrothermic therapy, radiofrequency denervation, or spinal
it is arguably the only study whereby a painful degenerated cord stimulation (Chou et al. 2007, 2009). There are weak
disc can potentially be identified. However, the use of dis- recommendations from the American Pain Society for exer-
cography should be carefully considered in terms of risks cise, spinal manipulation, and interdisciplinary rehabilitation
and benefits because accelerated disc degeneration, disc her- for chronic LBP or subacute LBP, as well as acupuncture,
niation, and loss of disc height have been reported in com- massage therapy, and yoga (Chou et al. 2007, 2009).
parison to matched controls at a 710-year follow-up Despite these limitations, the initial treatment of disco-
(Carragee et al. 2009). genic LBP is nonsurgical and includes medication and physi-
cal therapy. Acetaminophen is effective for pain relief but has
minimal anti-inflammatory effects. Hepatotoxicity may be
13.2.5 Treatment of Discogenic Pain encountered at high doses. Nonsteroidal anti-inflammatory
drugs (NSAIDs) and COX-2 inhibitors are a mainstay in
The initial treatment of discogenic pain is nonsurgical. It treatment as well. Opiate analgesics are more useful in treat-
is important to note, however, that the evidence available ing chronic nociceptive pain and are rarely necessary in the
supporting various nonsurgical treatment options is limited treatment of acute pain. Muscle relaxants and antidepres-
(Ostelo et al. 2009; Thomas et al. 2006). For non-radicular low sants, particularly those that block norepinephrine, may
structured 6-month regimen of physical therapy, NSAIDs,

and activity modification has failed to improve the patients
symptoms. In addition, it is important to rule out serious
underlying pathology as the source of pain it is necessary
to screen for secondary pain, localize the pain to a specific
region, and characterize the type of pain as mechanical versus
myofascial. Brox et al. performed a single-blind randomized
study to compare the effectiveness of lumbar instrumented
fusion with cognitive intervention and exercises in patients
with chronic low back pain and disc degeneration (Brox et al.
2003). At 1-year follow-up, there was equal improvement in
patients with chronic low back pain and disc degeneration
randomized to cognitive intervention and exercises, or lum-
bar fusion (Brox et al. 2003). Fairbanks et al. performed a
randomized control trial to assess the effectiveness of spinal
fusion versus intensive rehabilitation in patients with chronic
low back pain. Over the 2-year follow-up, both groups
reported statistically similar reductions in disability (Fairbank
et al. 2005). A randomized controlled study was carried out
to determine whether lumbar fusion could reduce pain and
diminish disability more effectively than nonsurgical treat-
ment in patients with severe chronic low back pain (Fritzell
et al. 2001). Patients in the nonsurgical group were treated
with different kinds of physical therapy. At 2 years, back
pain was reduced in the surgical group by 33 % compared
with 7 % (63 to 58) in the nonsurgical group (Fritzell et al.
Fig. 13.2 A lateral lumbar fluoroscopic image obtained during a dis- 2001).
cogram. Needles are inserted into the L34, L45, and L5S1 discs (in Surgical options are largely limited to lumbar arthrod-
this specific case), and contrast dye is inserted under pressure into each
esis through an anterior, posterior, or combined approach.
of the discs. The L45 and L5S1 levels (solid black arrows) demon-
strate posterior leakage of the dye, indicating the presence of an annular More recent options that are still under investigation
tear at each of these levels. There is no dye leakage at L34 level (bro- include dynamic stabilization and total disc arthroplasty.
ken black arrow) as it all remains contained within the disc space, thus Surgery is considered if the patient has failed a prolonged
indicating the absence of a significant annular tear. If, while injecting
course of conservative management and there exists a
the dye into a certain disc, the patient complains of pain similar to that
which he/she experiences clinically, the disc that triggers such pain is strong correlation between clinical presentation and radio-
said to cause concordant pain. This study is largely used in patients who graphic findings, and possibly discography. Posterolateral
are considering a surgical fusion for low back pain and is used to help fusion has been shown to successfully manage chronic dis-
decide which levels need to be fused, i.e., those levels which produce
cogenic back pain in a highly selected subset of patients
concordant pain on discography
receiving workers compensation and those chronically
disabled (Parker et al. 1996).
provide analgesia. Physical therapy relies on mobilization,
stretching, conditioning, and aerobic training. For discogenic
low back pain, McKenzie extension-based exercise programs 13.3 Intervertebral Disc Herniation
are the most beneficial. It is believed that these exercises will
unload the disc by restoring lumbar lordosis and decreasing 13.3.1 Epidemiology of Disc Herniation
mechanical strain. Other recommendations include smoking
cessation, weight loss, and activity alteration. The topic of Disc herniation is one of the most common intervertebral
nonsurgical treatment of low back pain is discussed in detail disc pathologies presented to spinal surgeons. Lumbar disc
in Chap. 15. herniation has a peak incidence in the fourth decade of life,
Surgical interventions for discogenic disc disease are con- and men are more likely to be affected than women (Weinstein
troversial and the results of high-level randomized studies et al. 2008a, b). However, it is reported that only 46 % of
are conflicting (Ostelo et al. 2009; Thomas et al. 2006; Chou lumbar herniations will become symptomatic (Weinstein
et al. 2007, 2009). Surgery should not be considered until a et al. 2008a, b).
13.3.2 Pathoetiology of Disc Herniation pressure but also causes the involved nerve to be stretched
over the herniated disc. Often, paracentral and foraminal
A cascade of degenerative changes occurs in the lumbar herniations will present with a predominance of leg pain,
spine with age. The nucleus pulposus desiccates and loses its whereas central herniations often present with a predomi-
ability to recover from compressive deformation. In the nance of back pain, as the central herniation, unless very
annulus fibrosis, there are fissures that form between the col- large, does not directly compress the exiting or traversing
lagen fibrils of the lamellae. With continued torsional, axial, nerve root.
and flexion strains across the disc, the nucleus pulposus can A thorough evaluation and examination is needed for
herniate through the fissures in the weakened outer annulus. a patient with suspected disc disease. A detailed medical
The loss of nucleus pulposus containment alters the biome- history includes investigation of possible serious underly-
chanics of the entire disc. The annular fibrils become exposed ing pathology, such as tumor, infection, fracture, or critical
to higher forces, which may lead to chondro-osseous changes neurologic compromise. Red flags such as fever, chills,
at the discvertebral body junction. Details of these changes night pain, unrelenting pain, unexplained weight loss,
are presented in greater detail in Chaps. 7 and 19. A herni- progressive lower extremity weakness, and bowel or blad-
ated nucleus pulposus can cause mechanical and/or chemical der dysfunction are indicative of a more severe and poten-
irritation of the affected nerve root and lead to severe radicu- tially emergent condition. These symptoms should prompt
lar symptoms. an expedited workup, including advanced imaging, such
Disc herniation may have one of three morphological as an MRI. Family history may be informative as well as
characteristics. A disc protrusion is defined as an eccen- reports exist of a familiar and genetic predisposition to
tric bulging of the nucleus pulposus into an intact but lumbar disc herniation. A full neurologic evaluation of
thinned annulus. Disc extrusion occurs when disc material muscle strength, sensation, proprioception, vibration, and
crosses through an annular defect but remains continuous deep tendon reflexes should be performed. Examination
with the disc space. A sequestered disc refers to one that of patient standing position and gait may also provide use-
is not continuous with the disc space and is a free frag- ful information. A variety of maneuvers exist, such as the
ment. Disc herniation can also be classified according to straight leg raise, to help diagnose a herniated disc.
the location of the disc herniation in relation to the spinal
canal, i.e., central, paracentral, foraminal, or far lateral.
The most common location of disc herniation is paracen- 13.3.4 Imaging of Herniation
tral, in which the herniated disc compresses the traversing
nerve root (e.g., the L5 nerve root at the L45 disc level). Imaging studies are useful to confirm clinical diagnosis.
A foraminal or far lateral disc will compress the exiting Plain lumbar radiographs are of little benefit in the direct
nerve root of the involved level (e.g., the L4 nerve root at evaluation of a disc herniation, as they do not show soft tis-
the L45 disc level). sue. However, they are indicated in patients with longer than
6 weeks of low back pain and in patients with medical history
of serious underlying pathology, such as cancer or infection.
13.3.3 Clinical Presentation of Herniation Furthermore, radiographs (anteroposterior, lateral, flexion,
and extension views) are helpful in evaluating a patient with
Most disc herniations are characterized by back and leg pain lumbar degenerative disease for evidence of instability, disc
of varying severity and duration. Onset may be insidious or collapse, and deformity (i.e., scoliosis). MRI is the diagnos-
proceeded by a traumatic event. The L45 and L5S1 discs tic test of choice for evaluating a patient for a disc hernia-
are most commonly affected. Leg pain commonly occurs in tion (Fig. 13.3a, b). However, relatively high rates of lumbar
the dermatome supplied by the compressed nerve root. Low disc herniations are found on MRI in asymptomatic patients
back pain may subside after herniation because of the depres- (Boden et al. 1999a, b). Therefore, clinical correlation is of
surization of the intervertebral space and relief of annular great importance. Dynamic MRI performed with patients in
tension. Radiculopathy can present as pain, paresthesias, a variety of patients has begun to gain popularity. CT scan
motor deficits, sensory deficits, and/or depressed reflexes. may be useful in specific situations, such as to assess bony
The distribution of symptoms (i.e., specific dermatome and anatomy in a patient who has had prior fusion surgery. CT
myotome) often provides information as to the level of disc myelogram is the study of choice in patients who cannot
herniation than can then be correlated with the findings on have an MRI. It is important to keep in mind when treating
MRI. Symptoms from a disc herniation are typically worse a patient with a symptomatic lumbar disc herniation that up
with sitting and improved with standing and walking. Sitting to 90 % of patients will have gradual resolution of symptoms
with the waist flexed not only causes an increase in the disc within 3 months of onset. For this reason, unless red flags
a b
Fig. 13.3 Sagittal (a) and axial (b) T2-weighted lumbar MRI images represents the bone response to abnormal load distribution that results
demonstrating a left-sided L5S1 paracentral disc herniation. On the from the degenerative disc. The axial image (b) demonstrates the loca-
sagittal image (a) the disc herniation is demonstrated by the open white tion of the disc in the left paracentral region, extending somewhat into
arrow. The herniated disc can be seen protruding posteriorly into the the left L5S1 intervertebral foramen (open white arrow). The right
spinal canal. The L5S1 disc itself is considerably degenerated, and exiting L5 nerve root (black arrow) and right traversing S1 nerve root
bone marrow edema (black arrow) is noted within this L5 vertebral (closed white arrow) are demonstrated. The exiting L5 and traversing
body, adjacent to the inferior end plate, and, to a lesser extent, in the S1 S1 nerve roots on the left side are significantly compressed by the disc
vertebral body, adjacent to the superior end plate. This edema likely herniation
exist, imaging is not indicated for at least 6 weeks following improve with conservative treatment, recurrent symptoms
the onset of radiculopathy and low back pain. following a successful trial of nonsurgical care, and significant
motor deficit. Absolute indications include cauda equina dis-
ease or a progressive neurologic deficit.
13.3.5 Herniation Treatment The Spine Patient Outcomes Research Trial (SPORT)
prospectively evaluated the outcomes of surgical versus
Initial treatment for patients presenting with acute radicular nonsurgical intervention in patients with symptomatic lum-
symptoms is conservative. It involves a short period of rest, bar disc herniation. Candidates with image-confirmed her-
activity modification, and anti-inflammatory analgesia. niation meeting eligibility criteria were enrolled and
Muscle relaxants and a short course of oral steroids have also randomized to standard open discectomy versus nonopera-
proven beneficial. Physical therapy and spinal manipulation tive care. At 4 years, patients who underwent surgery
are used frequently as well. Fluoroscopic epidural steroid achieved greater improvement than nonoperatively treated
injections can be administered to reduce nerve root patients (Weinstein et al. 2008a, b). Furthermore, it has also
inflammation and significantly improve symptoms. Generally, been demonstrated that with appropriate patient selection
conservative management is performed for at least 6 weeks. (i.e., symptoms and physical examination correlate to the
Radicular symptoms not relieved by nonoperative meth- findings on MRI), surgery provides faster resolution of
ods are often treated by surgery to remove the herniated disc. symptoms and perceived quicker recovery than nonopera-
Relative indications for surgery include intractable radicular tive management at 12 years (Weinstein et al. 2006; Peul
pain, pseudoclaudication, neurologic deficit that does not et al. 2007).
Microdiscectomy is the gold standard surgical procedure spondylolisthesis) or scoliosis. The shift forward, backward,
for lumbar disc herniation. This procedure can be performed or laterally can further reduce space in the spinal column
through a small incision in an outpatient setting. Complications centrally and/or foraminal spaces. Less common causes of
of surgical intervention include inadvertent durotomy, nerve lumbar spinal stenosis include epidural lipomatosis, tumors,
root injury, and infection. Later complications can include infections, and metabolic bone disorders such as Pagets dis-
postoperative instability and recurrent disc herniation, which ease. The main anatomic regions of lumbar stenosis include
can occur up to 7 % of cases (Weinstein et al. 2008a, b). It is the central canal, lateral recess (subarticular), foraminal, and
often difficult to discern recurrent disc herniation from epi- extraforaminal (Arbit and Pannullo 2001).
dural fibrosis on standard MRI imaging. Therefore, when
evaluating for recurrent disc herniation, gadolinium-enhanced
MRI should be used to differentiate between disc herniation 13.4.3 Clinical Presentation of Lumbar Stenosis
and fibrosis.
Lumbar spinal stenosis symptoms are often reflective of the
severity and the anatomic region of the narrowed spaces.
13.4 Lumbar Stenosis Often insidious due to the gradual narrowing over many
years, spinal stenosis can be asymptomatic for long periods
13.4.1 Epidemiology of Lumbar Stenosis of time (Arbit and Pannullo 2001). Early symptoms may
include intermittent low back pain without radiculopathy,
Lumbar stenosis refers to the narrowing of the neural canal signaling the onset of early disc degeneration. As the nar-
and neuroforaminal spaces of the lumbar spine. It is the most rowing of the spaces continues to compress nerves and limit
common end-stage consequence of disc degeneration often normal nerve gliding and excursion, radiculopathy or claudi-
resulting from adaptive changes of the facet joints and end cation symptoms may become more apparent. Patients often
plates. Most patients with symptomatic lumbar stenosis pres- complain of leg pain that is activity related, associated with
ent between the ages of 60 and 80 (Johnson et al. 1992; Arbit upright posture and/or prolonged walking (Paine 1976;
and Pannullo 2001). Women are slightly more likely to have Wilson 1969). The symptoms are often distributed along der-
lumbar stenosis than men, and the majority of patients with matomes or myotomes corresponding to the areas of affected
lumbar stenosis are of Caucasian descent (Johnson et al. neural compression. Pain symptoms may also be associated
1992; Arbit and Pannullo 2001). Degenerative spinal steno- with weakness, paresthesias, or numbness.
sis most commonly affects the L34 and L45 (Arbit and The classic clinical presentation of lumbar stenosis is
Pannullo 2001). Clinically, adults with radiographic evidence neurogenic claudication (Paine 1976; Wilson 1969).
of lumbar stenosis may be asymptomatic; however, as age Claudication refers to the gradual increase in symptoms with
increases, a higher percentage becomes symptomatic prolonged or repetitive activity (i.e., walking), which often is
(Johnson et al. 1992; Arbit and Pannullo 2001). associated with upright posture. The patients often relieve
this pain by changing posture such as sitting or leaning for-
ward (lumbar flexion) which results in neuroforaminal and
13.4.2 Etiology of Lumbar Stenosis spinal canal opening. Patients presenting with spinal stenosis
may note symptoms gradually worsening over time or acute
The most common cause of lumbar stenosis is age-related onset radiculopathy. Aggravation of severe stenosis can
degenerative changes of the spine (Arbit and Pannullo 2001). sometimes lead to cauda equina syndrome (loss of bowel or
End-stage lumbar disc degeneration alters facet and disc bio- bladder control). There are no specific physical exam tests or
mechanics and may induce hypertrophy and overgrowth of signs associated with lumbar stenosis. Patients often have
the facet joints and vertebral end plates, respectively. The normal strength and sensation. Confirmation of posture-
resultant encroachment and the narrowing of the neurofo- related improvement of pain (i.e., sitting or lumbar flexion)
ramina from osteophyte formation, loss of disc space height, may further assist in clinical diagnosis. Nerve tension signs
and hypertrophy of the ligamentum flavum can cause marked may be present in patients with severe radiculopathy.
reduction in space available for the spinal nerve roots.
Another cause of spinal stenosis is congenital lumbar spi-
nal stenosis, which is associated with shorter pedicle lengths 13.4.4 Imaging of Lumbar Stenosis
and innately narrowed spaces for neural elements in patients
from birth. Instability associated with spinal stenosis most Lumbar stenosis is a radiological finding as well as a clini-
commonly occurs due to lumbar disc and/or facet degen- cal diagnosis. Imaging studies are often used to correlate
eration and translation (spondylolisthesis) of the vertebra clinical symptoms and quantify severity. Initial evaluation is
relative to each other. Instability can also result from the performed using plain lumbar radiographs to assess spinal
structural incompetence of the pars interarticularis (isthmic alignment and rule out instability or metabolic bone disease.
CT scans are useful to view bony anatomy in detail and rule severe pain in others (Sirvanci et al. 2008). In patients whose
out fractures or detect other bone defects. Plain CT is not radiological findings confirm their clinical symptoms, appro-
often employed for the diagnosis of lumbar stenosis. CT scan priate treatment plans can be employed and predictable out-
combined with myelography is extremely useful to visual- comes can be anticipated.
ize neural element compression and confirm areas of bony
impingement and is often used to evaluate patients who can-
not undergo MRI. The most common test used to confirm 13.4.5 Treatment of Lumbar Stenosis
lumbar stenosis is MRI, which can show disc degeneration,
end plate morphology, facet and ligamentum hypertrophy, The first-line treatment in lumbar spinal stenosis is nonop-
neural elements, and available neural spaces (Fig. 13.4). erative. These measures may include NSAIDs, analgesics,
Analysis of sequential images of the lumbar spine in sagittal and activity modification. Oral steroids may also be given to
and axial planes can allow localization of specific areas of help reduce acute inflammatory symptoms and minimize
stenosis in all four anatomic regions (central, lateral recess, radicular pain. Physical therapy can also be attempted, with
foraminal, and extraforaminal). Relative stenosis is defined the main exercises being low impact, core strengthening,
as a central canal diameter of 1013 mm, whereas absolute aerobic conditioning, and stretching. Alternative exercise
spinal stenosis describes a canal diameter of less than 10 mm modalities include Tai chi, water therapy, chiropractic care,
(Verbiest 1979). and stationary biking.
It is important to note that there is very little correlation Epidural steroid injections can reduce inflammation
between radiological findings and patient symptoms (Sirvanci around the nerve roots and may help treat radiculopathy.
et al. 2008). Severe stenosis radiologically can be asymp- Selective nerve root block can be very effective for pain
tomatic in some patients, while low-grade stenosis may cause relief in the short term and can also be useful as a diagnostic
tool. Recent level 1 evidence confirms their effectiveness for
short-term relief of radicular pain (Weinstein et al. 2008a, b).
Unlike injections for radiculopathy associated with a herni-
ated disc where the natural history is spontaneous resolution,
injections for stenosis do not resolve the underlying com-
pression of the nerve root(s) associated with bony stenosis,
and symptoms may recur.
Definitive management of symptomatic lumbar stenosis
involves surgery. Surgery is indicated for patients who fail
nonoperative management. The standard surgical procedure
for degenerative lumbar stenosis is a decompressive laminec-
tomy. Neural decompression is achieved by removing the cen-
tral bony elements of the posterior spine (spinous process and
lamina). This allows the restoration of adequate central canal
space, whereas medial facetectomy and foraminotomy allow
for lateral recess and foraminal decompression, respectively.
Surgical outcomes for decompressive laminectomy are cost
effective, reliable, and superior to nonoperative management
in functional restoration (Weinstein et al. 2008a, b; Tosteson
et al. 2008). Complications of lumbar laminectomy include
incidental dural tears, postoperative instability, persistent back
pain, persistent radiculopathy, and recurrence of stenosis.
13.5 Degenerative Spondylolisthesis
13.5.1 Epidemiology of Spondylolisthesis
Degenerative spondylolisthesis refers to spinal instability

Fig. 13.4 Sagittal T2-weighted lumbar MRI image demonstrating
and translation resulting from lumbar degenerative condi-
moderate L23 and severe L34 and L45 stenosis. In addition, there is
a slight degenerative spondylolisthesis at L45, where the L4 vertebral tions. Most commonly, patients initially present after 50 years
body is located slightly anterior to the L5 vertebral body (black arrow) of age; however, more severe symptoms are often noted with
increasing age and in patients over 65 years of age (Valkenburg wall translation (Meyerding Grading System) (Fig. 13.5).
and Haanen 1982; Frymoyer 1994). Females are affected Oblique radiographs can be used to diagnose pars defects.
more, with a 3:1 ratio over men (Frymoyer 1994). Body mass Advanced imaging such as MRI or CT myelography can fur-
index, age, and angle of lordosis were significantly associ- ther delineate areas of neural compression and severity of
ated with degenerative spondylolisthesis in women. In men, stenosis (Fig. 13.6). Evaluation of fluid in the facet joints on
no individual risk factors for degenerative spondylolisthesis axial cuts may also indicate instability or spondylolisthesis
were found, save increased age (Jacobsen et al. 2007). when x-rays are unavailable (Rihn et al. 2007).
13.5.2 Etiology of Spondylolisthesis 13.5.5 Treatment of Spondylolisthesis
Degenerative spondylolisthesis is most commonly brought Conservative treatment should be employed initially. This
on by facet joint degeneration and intervertebral disc degen- may include activity modification, NSAIDs, and low-impact
eration. This occurs most commonly at level L45 (Frymoyer exercise programs. A weight loss regimen can aid in the alle-
1994). Degenerative changes of the facet joints result in associ- viation of symptoms, but this will not cure or reverse the
ated morphological changes that allow vertebral body transla- problem. Increased activity can cause aggravation of pain
tion and secondary neuroforaminal compression. Loss of disc
height contributes to the magnitude of translation and subse-
quent neural compression. In degenerative spondylolisthesis,
slippage of the vertebral body is typically less than 50 % of the
anteriorposterior vertebral body diameter (Frymoyer 1994).
Facet joints tend to enlarge in degenerative spondylolisthesis,
which further encroaches into the spinal canal and can cause
lateral recess stenosis and radicular symptoms.
13.5.3 Clinical Presentation of Spondylolisthesis
It should be noted that the degree of slippage is not always

directly correlated with the severity of clinical findings or
pain. Symptomatic patients most often present with aching
back pain that travels into the buttocks and back of the thighs
(radiculopathy). Concomitant neurogenic claudication may
also be noted. Neuroforaminal stenosis is often associated
with radicular pain along the affected nerve root.
Physical exam findings are similar to lumbar spinal stenosis
and may include point tenderness at the lumbosacral junction
and a restricted trunk range of motion. Neurological exami-
nation is usually normal in terms of strength and sensation;
however, paresthesias and weakness may be noted in patients
with a significant radiculopathy component. Mechanical
back pain may be present more in patients with spinal steno-
sis associated with spondylolisthesis. Patients often report
more pain with extremes of flexion and extension.
13.5.4 Imaging of Spondylolisthesis
Standing orthogonal radiographs with dynamic views are

very useful to diagnose disc degeneration and instability. The
Fig. 13.5 Lateral lumbar plain radiograph demonstrating a degenera-
lateral views will show slippage of one vertebra in relation to
tive spondylolisthesis at L45, where the L4 vertebral body is located
the vertebra below it, and grading is often done based on the slightly anterior to the L5 vertebral body. There is also multilevel lum-
percentage of slip as measured by the posterior vertebral body bar disc collapse, most significant at L45
symptoms; however, physical therapy may help improve

mechanical back pain associated with a slipping vertebral
body, but may not improve nerve compression and neuro-
genic symptoms. Focal epidural steroid injections may help
reduce inflammation and alleviate radicular pain.
If nonoperative measures fail, surgical options should be
considered. Indicated procedures should include neural
decompression with laminectomy and foraminotomy. Often
times, fusion is also performed to reduce risk of slippage
progression and recurrence of symptoms and optimize long-
term outcomes. Decompressive laminectomy with success-
ful fusion has been shown to result in improved outcomes
when compared to decompression alone (Herkowitz and
Kurz 1991; Fischgrund et al. 1997; Kornblum et al. 2004)
(Fig. 13.7a, b). Clinical outcomes of lumbar decompression
and fusion for degenerative spondylolisthesis indicate high
success rates and better outcomes than nonoperative man-
agement alone (Weinstein et al. 2009). Complications unique
to spinal fusion include failure to achieve solid fusion, stiff-
ness, and/or persistent back pain.
13.6 Concluding Comments
Degenerative changes of lumbar spine produce a spectrum

Fig. 13.6 Sagittal T2-weighted lumbar MRI image demonstrating
degenerative spondylolisthesis and severe stenosis at the L45 level of potentially disabling conditions that affect patients
a b
Fig. 13.7 Postoperative anteroposterior (a) and lateral (b) lumbar fusion with the use of a posterior pedicle screw-rod construct and an
plain radiographs in the patient from Fig. 13.6 who underwent a lumbar interbody cage with bone grafting
laminectomy and a posterior and transforaminal lumbar interbody
across a wide age range, including discogenic low back References

pain, lumbar disc herniation, lumbar stenosis, and degen-
erative spondylolisthesis. In the absence of red flags Arbit E, Pannullo S (2001) Lumbar stenosis: a clinical review. Clin
Orthop Relat Res 384:137143
such as progressive neurological deficits, loss of bowel
Boden SD, Davis DO, Dina TS, Patronas NJ, Wiesel SW (1999a)
and/or bladder function, and constitutional symptoms, Abnormal magnetic-resonance scans of the lumbar spine in asymp-
these conditions should initially be managed conserva- tomatic subjects: a prospective investigation. J Bone Joint Surg Am
tively with a combination of medication, physical therapy, 72(3):403408
Boden SD, McCowin PR, Davis DO, Dina TS, Mark AS, Wiesel S
patient education, and possibly epidural injections. In
(1999b) Abnormal magnetic-resonance scans of the cervical spine
patients who have persistent, disabling symptoms despite a in asymptomatic patients: a prospective investigation. J Bone Joint
dedicated conservative treatment program, surgery may Surg Am 72:11781184
offer significant benefit over a continued nonoperative Brox JI, Sorenson R, Friis A et al (2003) Randomized clinical trial of
lumbar instrumented fusion and cognitive intervention and exer-
approach.
cises in patients with chronic low back pain and disc degeneration.
Spine 28:19131921
Buckwalter JA, Mow VC, Boden SD, Eyre DR, Weidenbaum M (2002)
13.7 Summary of Critical Concepts Intervertebral disc structure, composition, and mechanical function. In:
Buckwalter JA, Einhorn TA, Simon SR (eds) Orthopaedic basic science-
Discussed in the Chapter biology and biomechanics for the musculoskeletal system, 2nd edn.
American Academy of Orthopaedic Surgeons, Rosemont, pp 548555
Degenerative disease of the lumbar spine accounts for a Carragee EJ, Chen Y, Tanner CM, Hayward C, Rossi M, Hagle C (2000)
relatively large proportion of the annual health-care Can discography cause long-term back symptoms in previously
asymptomatic subjects? Spine 25:18031808
expenditures in the USA and is responsible for consider-
Carragee EJ, Lincoln T, Parmar VS, Alamin T (2006) A gold standard
able indirect losses due to time-off of work. evaluation of the discogenic pain diagnosis as determined by
Chronic low back pain is the most common cause of mor- provocative discography. Spine 31(18):21152123
bidity and chronic pain in the USA. It is estimated that up Carragee EJ, Don AS, Hurwitz EL, Cuellar JM, Carrino J, Herzog R
(2009) Does discography cause accelerated progression of degen-
to 80 % of individuals will experience low back pain at
eration changes in the lumbar disc: a ten-year matched cohort study.
one time in their lives. Spine 34(21):23382345
Kirkaldy-Willis described a widely accepted pathway that Chou R, Qaseem A, Snow V et al (2007) Diagnosis and treatment of
divided lumbar disc degeneration into three stages based low back pain: a joint clinical practice guideline from the American
College of Physicians and the American Pain Society. Ann Intern
on the amount of damage that has occurred; these stages
Med 147:478491
are dysfunction, instability, and stabilization. Chou R, Loeser JD, Owens DK et al (2009) Interventional therapies,
The relationship between disc degeneration and low back surgery, and interdisciplinary rehabilitation for low back pain. An
pain is incompletely understood. evidence-based clinical practice guideline from the American Pain
Society. Spine 34:10661077
Most episodes of acute low back pain are self-limiting.
Deyo RA, Mirza SK, Martin BL (2006) Back pain prevalence and visit
It is important for the clinician to maintain an awareness rates: estimates from U.S. national surveys 2002. Spine 31(23):
of the so-called red flags that may signify a serious 27242727
underlying pathology such as fracture, infection, or Fairbank J, Frost H, Wilson-MacDonald J, Yu LM, Barker K, Collins R
(2005) Randomized controlled trial to compare surgical stabiliza-
malignancy.
tion of the lumbar spine with an intensive rehabilitation programme
Surgical intervention for discogenic disc disease is confor patients with chronic low back pain: the MRC spine stabilization
troversial, and the results of high-level randomized stud- trial. BMJ 330:1233
ies are conflicting. Fischgrund JS, Mackay M et al (1997) 1997 Volvo Award winner in
clinical studies. Degenerative lumbar spondylolisthesis with spinal
Patients who underwent surgery for disc herniation
stenosis: a prospective, randomized study comparing decompres-
achieved greater improvement than nonoperatively treated sive laminectomy and arthrodesis with and without spinal instru-
patients at 4 years. Furthermore, it has been demonstrated mentation. Spine 22(24):28072812
that with appropriate patient selection (i.e., symptoms and Fritzell P, Hagg O, Wessberg P, Nordwall A, Swedish Lumbar Spine
Study Group (2001) 2001 Volvo Award winner in clinical studies:
physical examination correlate to the findings on MRI),
lumbar fusion versus nonsurgical treatment for chronic low back
surgery provides faster resolution of symptoms and per- pain: a multicenter randomized controlled trial from the Swedish
ceived quicker recovery than nonoperative management Lumbar Spine Study Group. Spine 26:25212532
at one to two years. Frymoyer J (1994) Degenerative spondylolisthesis: diagnosis and treat-
ment. J Am Acad Orthop Surg 2:915
There may be poor correlation between radiological
Guyer RD, Ohnmeiss DD (1995) Lumbar discography: position state-
findings in patients with lumbar stenosis and their ment from the North American Spine Society Diagnostic and
symptoms. Therapeutic Committee. Spine 20:20482059
For degenerative lumbar spondylolisthesis, clinical out- Hansson T, Hansson E, Malchau H (2008) Utility of spine surgery: a
comparison of common elective orthopaedic surgical procedures.
comes of decompression and fusion indicate high success
Spine 33(25):28192830
rates and better outcomes than nonoperative management Herkowitz HN, Kurz LT (1991) Degenerative lumbar spondylolisthesis
alone. with spinal stenosis. A prospective study comparing decompression
with decompression and intertransverse process arthrodesis. J Bone stenosis: correlation with Oswestry Disability Index and MR imag-
Joint Surg Am 73(6):802808 ing. Spine J 17:679685
Jacobsen S, Sonne-Holm S, Rovsing H, Monrad H, Gebuhr P (2007) Thomas KJ, MacPherson H, Thorpe L et al (2006) Randomised controlled
Degenerative lumbar spondylolisthesis: an epidemiological perspec- trial of a short course of traditional acupuncture compared with usual
tive: the Copenhagen Osteoarthritis Study. Spine 1(32):120125 care for persistent non-specific low back pain. BMJ 333:623
Johnson KE, Rosean I, Udean A (1992) The natural course of lumbar Tosteson AN, Lurie JD et al (2008) Surgical treatment of spinal stenosis
spinal stenosis. Clin Orthop Relat Res 279:8286 with and without degenerative spondylolisthesis: cost-effectiveness
Johnson WE, Evans H, Menage J, Eisenstein SM, El Haj A, Roberts S after 2 years. Ann Intern Med 149(12):845853
(2001) Immunohistochemical detection of Schwann cells in innervated Twomey LT, Taylor JR (1987) Age changes in lumbar vertebrae and
and vascularized human intervertebral discs. Spine 26:25502557 intervertebral discs. Clin Orthop Relat Res 224:97104
Kornblum MB, Fischgrund JS et al (2004) Degenerative lumbar spon- Valkenburg HA, Haanen HCM (1982) The epidemiology of low back
dylolisthesis with spinal stenosis: a prospective long-term study pain. In: White AA III, Gordon SL (eds) AAOS symposium on idio-
comparing fusion and pseudarthrosis. Spine 29:726733; discus- pathic low back pain. CV Mosby, St. Louis, pp 922
sion 733724 Verbiest H (1979) The significance and principles of computed axial
Luo X, Pietrobon R, Sun SX, Liu GG, Hey L (2004) Estimates and pat- tomography in idiopathic developmental stenosis of the bony lum-
terns of direct health care expenditures among individuals with back bar vertebral canal. Spine 4:369378
pain in the United States. Spine 29(1):7986 Walker BF (2000) The prevalence of low back pain: a systematic review
Lyons G, Eisenstein SM, Sweet MB (1981) Biochemical changes in inter- of the literature from 1966 to 1998. J Spinal Disord 13(3):205217
vertebral disc degeneration. Biochim Biophys Acta 673:443453 Walsh TR, Weinstein JN, Spratt KF, Lehmann TR, April C, Sayre H
Mooney V, Haldeman S, Nasca RJ et al (1988) Position statement on (1990) Lumbar discography in normal subjects: a controlled, pro-
discography: the executive committee of the North American Spine spective study. J Bone Joint Surg Am 72:10811088
Society. Spine 13:1343 Weinstein JN, Lurie JD, Tosteson TD, Skinner JS, Hanscom B, Tosteson
Ostelo RWJG, van Tulder MW, Vlaeyen JWS, Linton SJ, Morley S, AN, Herkowitz H, Fischgrund J, Cammisa FB, Albert T, Devo RA
Assendelft WJJ (2009) Behavioral treatment for chronic low-back (2006) Surgical vs nonoperative treatment for lumbar disk hernia-
pain (Review). Cochrane Collab 2:149 tion: the Spine Patient Outcomes Research Trial (SPORT) observa-
Paine KWE (1976) Clinical features of lumbar spinal stenosis. Clin tional cohort. JAMA 296(20):24412459
Orthop Relat Res 115:7782 Weinstein JN, Tosteson TD et al (2008a) Surgical versus nonsurgical
Parker LM, Murrell SE, Boden S, Horton WC (1996) The outcome of therapy for lumbar spinal stenosis. N Engl J Med 358(8):794810
posterolateral fusion in highly selected patients with discogenic low Weinstein JN, Lurie JD, Tosteson TD et al (2008b) Surgical versus non-
back pain. Spine 21(16):19091916 operative treatment for lumbar disk herniation: four-year results for
Peul WC, van Houwelingen HC, van den Hout WB et al (2007) Surgery the Spine Patient Outcomes Research Trial (SPORT). Spine 33(25):
versus prolonged conservative treatment for sciatica. N Engl J Med 27892800
356:22452256 Weinstein JN, Lurie JD et al (2009) Surgical compared with nonoperative
Rihn JA et al (2007) Does lumbar facet fluid detected on magnetic reso- treatment for lumbar degenerative spondylolisthesis. Four-year results
nance imaging correlate with radiographic instability in patients in the Spine Patient Outcomes Research Trial (SPORT) randomized
with degenerative lumbar disease? Spine 32(14):15551560 and observational cohorts. J Bone Joint Surg Am 91:12951304
Roberts S (2002) Disc morphology in health and disease. Biochem Soc Wiesel SW, Tsoursmas N, Feffer HL, Citrin CM, Patronas N (1984) A
Trans 30:864869 study of computer-assisted tomography. Part I: the incidence of positive
Roberts S, Menage J, Urban JPG (1989) Biochemical and structural CAT scans in an asymptomatic group of patients. Spine 9:549551
properties of the cartilage end-plate and its relation to the interver- Wilson CB (1969) Significance of the small lumbar spinal canal: cauda
tebral disc. Spine 14:166174 equina compression syndromes due to spondylosis, 3: intermittent
Simmons JW, Aprill CN, Dwyer AP, Brodsky AE (1988) A reassess- claudication. J Neurosurg 31:499506. http://www.ncbi.nlm.nih.
ment of Holts data on: the question of lumbar discography. Clin gov/pubmed/5351760
Orthop Relat Res 237:120124 Yong-Hing K, Kirkaldy-Willis WH (1983) The pathophysiology of
Sirvanci M, Bhatia M, Ganiyusufoglu KA, Duran C, Tezer M, Ozturk degenerative disease of the lumbar spine. Orthop Clin North Am
C, Aydogan M, Hamzaoglu A (2008) Degenerative lumbar spinal 14:491504
Spinal Motion Restoration
Devices for the Degenerative Disc 14
Daniel G. Kang, Melvin D. Helgeson,
and Alexander R. Vaccaro
Contents 14.9 Other Spinal Motion-Preserving Devices.................... 243

14.9.1 Nuclear Replacements..................................................... 243
14.1 Introduction ................................................................... 225 14.9.2 Lumbar Dynamic Posterior
14.2 Adjacent Segment Disease: Does There Stabilization/Facet Replacement ..................................... 243
Need to Be an Alternative to Spinal Fusion? .............. 226 14.10 Summary of Critical Concepts Discussed
14.3 History and Evolution of Disc in the Chapter ................................................................ 243
Arthroplasty Devices ..................................................... 227 References .................................................................................... 244
14.4 Total Disc Arthroplasty Design:
Concepts and Biomechanics ......................................... 228
14.4.1 Biomechanics of the Functional Spinal Unit .................. 228
14.4.2 Biomechanical Objectives of Disc Arthroplasty ............. 229 14.1 Introduction
14.4.3 Disc Arthroplasty: Material Considerations
and Bearing Surfaces ...................................................... 229 Cervical and lumbar degenerative disc disease is common in
14.4.4 Semiconstrained Versus Unconstrained Devices ............ 231
14.4.5 Implant Fixation .............................................................. 231 todays society and can be associated with significant pain
and disability (Taksali et al. 2004). Spinal fusion continues
14.5 Comparison of Total Disc Arthroplasty Devices ........ 232
14.5.1 Lumbar TDA Devices ..................................................... 232
to be the most common surgical treatment for degenerative
14.5.2 Cervical TDA Devices .................................................... 233 conditions in the neck and low back (Davis 1994; Lee and
Langrana 2004). While fusion has been similarly used for
14.6 Operative Technique ..................................................... 238
the treatment of degenerative arthritis of the hip and knee
14.7 Deciding When to Use Total Disc Arthroplasty.......... 240 joints, it has been replaced by revolutionary joint arthro-
14.7.1 Indications for Lumbar Total Disc Arthroplasty ............. 240
14.7.2 Indications for Cervical Total Disc Arthroplasty ............ 240
plasty techniques with excellent outcomes in terms of reliev-
ing pain and restoring function (Santos et al. 2004). In
14.8 Total Disc Arthroplasty Revision Strategies ............... 241
contrast, disc arthroplasty has only recently been considered
an alternative to spinal arthrodesis and has not replaced
fusion as the gold standard treatment. Studies regarding
spinal fusion for degenerative disc disease have demon-
The opinions or assertions contained herein are the private views of strated inconsistent clinical results even in properly selected
the authors and are not to be construed as official or as reflecting the patients, and despite advances in spinal fusion technique and
views of the United States Army or the Department of Defense. instrumentation, patient outcomes have not been significantly
Two authors are employees of the United States government. This work
altered (Barrick et al. 2000; Kleeman et al. 2001; Madan and
was prepared as part of their official duties and as such, there is no
copyright to be transferred. Boeree 2003; Bono and Lee 2004; Lee and Langrana 2004;
Santos et al. 2004).
D.G. Kang, MD (*) M.D. Helgeson, MD
Department of Orthopaedic Surgery and Rehabilitation, A particular concern following spinal arthrodesis has been
Walter Reed National Military Medical Center, the failure to restore normal physiologic motion, possibly
8901 Wisconsin Avenue, Bethesda, MD 20889, USA leading to an increase in adjacent segment degeneration and
e-mail: daniel.g.kang@gmail.com; mel_helgeson@yahoo.com
disease (Lehmann et al. 1987; Lee 1988; Eck et al. 1999;
A.R. Vaccaro, MD, PhD Akamaru et al. 2003; Gillet 2003; Lee and Langrana 2004;
Santos et al. 2004). Alterations in spine biomechanics
Rothman Institute/Thomas Jefferson University,
925 Chestnut Street, 5th Floor, Philadelphia, PA 19107, USA following arthrodesis include loss of motion and shock
e-mail: alexvaccaro3@aol.com absorption; these changes have been shown to cause a

226 D.G. Kang et al.
compensatory increase in motion, abnormal stress flow, and to pelvis (Hilibrand and Robbins 2004). Spinal segment
elevated intradiscal pressure at adjacent segments (Lee and arthrodesis was first described in 1911 by Hibbs (1911) for
Langrana 1984, 2004; Shono et al. 1998; Akamaru et al. treatment of spinal deformity and Albee (1911) for the treat-
2003; Chang et al. 2007; Gao et al. 2011). Although adjacent ment of Potts disease. While spinal fusion has also been
segment degeneration and disease may be due to natural dis- used for other spinal conditions such as spinal trauma, spon-
ease progression, rather than a consequence of spinal fusion, dylolisthesis, spinal stenosis, or tumors, it is most commonly
there has been continued enthusiasm and advancement of used for the treatment of symptomatic degenerative disc dis-
motion restoration surgery for the degenerative disc (Santos ease causing neck or low back pain (Davis 1994; Hilibrand
et al. 2004). With the basic premise that motion preservation and Robbins 2004). Treatment goals, not unlike those for
at the treated symptomatic degenerative disc may slow pro- fusion of arthritic appendicular joints, are to eliminate pain,
gression or even prevent symptomatic degeneration at adja- maintain or restore stability, and correct deformity/height
cent segments, the change represents a potential paradigm loss (Orr et al. 2007).
shift in approach: from motion-elimination surgery to motion- Chandler first reported the use of spinal fusion for the
preserving surgery (Santos et al. 2004; Madigan et al. 2009). treatment of low back pain in 1929 (Chandler 1929), while
Spinal motion restoration devices encompass various new Robinson and Smith described anterior cervical discectomy
technologies and surgical procedures, including total disc and fusion (ACDF) in 1955 (Robinson and Smith 1955) for
arthroplasty (TDA), posterior soft tissue stabilization, poste- treatment of degenerative spondylotic conditions (Lee and
rior dynamic screw and rod systems, nucleus pulposus Langrana 2004). In fact, ACDF is considered by many to be
replacement, and even injection of the degenerative disc with one of the most successful spine procedures performed in
polymeric agents (Madan and Boeree 2003). However, most the last several decades (Murrey et al. 2009). Following
of these spinal motion restoration devices remain experimen- ACDF, patients have reported greater than 90 % rate of
tal with limited clinical use. However, TDA has been rou- relief of radicular complaints and stabilization/improve-
tinely performed in Europe since the early 1990s, although a ment of myelopathic symptoms (Hilibrand and Robbins
few devices have been recently approved in the USA by the 2004; Riew et al. 2008; Murrey et al. 2009). On the other
Food and Drug Administration (FDA) following prospec- hand, use of spinal fusion for treatment of discogenic low
tive, randomized clinical trials (Santos et al. 2004; back pain has been a controversial subject, with often
Mummaneni et al. 2007; Sasso et al. 2007; Zigler et al. 2007; conflicting findings (Frymoyer et al. 1978; Lee and
Guyer et al. 2009; Murrey et al. 2009; Burkus et al. 2010; Langrana 2004). In a landmark study, Fritzell et al. (2003)
Garrido et al. 2010; Delamarter et al. 2011). While total disc found that patients with discogenic low back pain have bet-
arthroplasty continues to gain popularity as an alternative to ter outcomes after successful spinal fusion compared with
spinal fusion, the ideal device design, as well as the appropri- continued conservative treatment (McAfee 2004). The
ate indications for arthroplasty in the cervical spine and lum- study was a randomized controlled trial of 294 patients
bar spine, remains unknown. Further experience and with a 2-year follow-up by an independent observer. It was
long-term follow-up studies will determine if TDA provides reported that disability assessed by the Oswestry Disability
acceptable clinical outcomes, as well as influencing the rate Index (ODI) was reduced to 25 % in the surgical group
of adjacent segment disease (Hilibrand and Robbins 2004). compared with 6 % in the nonsurgical group (Fritzell et al.
Clinical success of total disc arthroplasty will require a com- 2003). Despite these findings, because of the published
bination of appropriate patient selection, good implant mate- variable success rates ranging from 32 to 99 % and because
rials and design, and proper implantation technique. The of the wide variety of study methods, it is difficult to draw
following discussion will address the evolving history and definitive conclusions concerning the indications for fusion
principles of TDA; describe the biomechanical, biomaterial, and surgical techniques (Jackson et al. 1985; Lehmann
and design concepts behind TDA; and then detail the surgi- et al. 1987; OBeirne et al. 1992; Bono and Lee 2004; Lee
cal indications and techniques for TDA in both the cervical and Langrana 2004; Lin and Wang 2006).
and lumbar spines. In addition, other emerging spinal motion Although symptomatic degenerative disc disease contin-
restoration devices for the lumbar spine will be briefly ues to be treated surgically, there is debate concerning the
considered. long-term consequences of spinal fusion. Of particular con-
cern has been adjacent segment degeneration and adjacent
segment disease. The term adjacent segment degenera-
14.2 Adjacent Segment Disease: tion is used to describe radiographic changes without
Does There Need to Be an symptoms evident at a level adjacent to a previous fusion,
Alternative to Spinal Fusion? while adjacent segment disease refers to radiographic
findings that correlate with new clinical symptoms
Spinal fusion has evolved over the past 100 years for (Hilibrand et al. 1999; Hilibrand and Robbins 2004; Lee
treatment of numerous pathologic conditions, throughout and Langrana 2004). The current scientific literature fails
the human spine, involving spinal segments from the occiput to clearly demonstrate if adjacent segment disease is the
14 Spinal Motion Restoration Devices for the Degenerative Disc 227
result of altered biomechanics due to an iatrogenic rigid notion that cervical and lumbar degenerative disc disease
motion segment, a consequence of surgical technique, pro- is exceedingly complex, and a complete understanding of
gression of the natural history of degenerative disease, or its etiology continues to be elusive. Even in carefully
more likely a combination of all of these factors (Hilibrand selected patient populations, a successful clinical result
et al. 1999; Hilibrand and Robbins 2004; Ishihara et al. can be difficult to achieve even when radiographic fusion
2004; Robertson et al. 2005; Yue et al. 2005; Lin and Wang is successful. Since pain is a critical measure of success,
2006). However, biomechanical studies have shown fusion may fail due to generation of pain from the degen-
increased motion, strain, and intradiscal pressures at a spine erative nucleus pulposus, nociceptive nerve endings within
segment adjacent to a spinal fusion, which may explain the the annulus fibrosus, dorsal root ganglion, facet joints, and
occurrence of adjacent segment disease (Lee and Langrana joint capsules, or even the surrounding ligamentous and
1984; Weinhoffer et al. 1995; Matsunaga et al. 1999; muscular structures at a single or multiple levels (Bono
Reitman et al. 2004; Murrey et al. 2009). These conse- and Garfin 2004). Incomplete pain relief due to failure to
quences have been inconsistently experienced in the clini- address a specific pain generator may lead to treatment
cal setting, and there have been conflicting findings in the failure and may be further impacted by factors such as
cervical and lumbar spines due to dissimilar biomechanical psychosocial dysfunction, pseudarthrosis, adjacent seg-
environments. ment disease, as well as morbidity arising from surgery
Following cervical spine fusion, the prevalence of adja- (Kumar et al. 2001; Lin and Wang 2006; Orr et al. 2007).
cent segment disease requiring additional surgery ranges Given the uncertainties and potential complications fol-
from 9 to 17 %, with an annual incidence of 1.5 to 4 % lowing spinal fusion for treatment of degenerative disc
(Williams et al. 1968; Gore and Sepic 1984; Bohlman et al. disease, there has been increasing enthusiasm for total disc
1993; Hilibrand and Robbins 2004). In a landmark study, arthroplasty.
Hilibrand et al. (1999) reported that following anterior cer-
vical fusion, the annual incidence of adjacent segment dis-
ease was approximately 3 % and predicted prevalence of 14.3 History and Evolution
25.6 % at 10 years. Interestingly, it was noted that when of Disc Arthroplasty Devices
anterior cervical fusion was performed at more than one
level, adjacent segment disease was at a significantly lower Disc arthroplasty devices are an emerging technology for
rate than when performed at a single level. The authors the treatment of spinal disc degeneration. With recent
therefore concluded that adjacent segment disease is a com- advances in our understanding of this technology, an appre-
mon problem following anterior cervical fusions but may ciation of its origins, history, and evolution provides a use-
also be related to the natural history of cervical spondylosis ful perspective on its role in the treatment of spinal
(Hilibrand et al. 1999). The argument for natural disease disorders (Bono and Garfin 2004). The earliest disc replace-
progression is also supported by Herkowitz et al. (1990). ment procedure was reported in the late 1950s by Fernstrom,
Thus, following posterior foraminotomy without fusion, at approximately the time of Charnleys initial reports on
41 % of patients randomized to anterior cervical fusion total hip arthroplasty (Fernstrom 1966; Bono and Garfin
developed adjacent segment degeneration compared to 2004). Following nucleus pulposus removal, Fernstrom
50 % of patients. implanted a rudimentary intervertebral disc (a metallic
In the lumbar spine, studies of adjacent segment disease ball) prosthesis into the annulus fibrosus, in both the cervi-
following spinal fusion have also elicited conflicting cal and lumbar spines of human patients (Fernstrom 1966).
results. Although one-third to one-half of patients develop Although simplistic in design, Fernstroms metallic ball
adjacent segment degeneration following lumbar fusion, achieved the intended goal: pain relief through removal of
the radiographic findings did not correlate with clinical the painful nucleus, along with maintaining intervertebral
symptoms (Lehmann et al. 1987; Luk et al. 1987; Hilibrand height and motion. Despite good short-term results, there
and Robbins 2004). Regarding the argument for natural was long-term failure from subsidence of the implant
disease progression, Penta et al. (1995) compared patients through the vertebral end plates, resulting in loss of inter-
with greater than 10-year follow-up treated nonoperatively vertebral height in about 88 % of cases at 4- to 7-year fol-
with those treated by anterior lumbar interbody fusion. low-up (Fernstrom 1966; Bono and Garfin 2004). In
These workers found no difference in the rate of adjacent hindsight and with continued understanding of total disc
segment degeneration (approximately one-third of patients) requirements, this failure was predictable given the very
between the two groups. However, a study by Ghiselli et al. limited contact of the steel ball with the flat vertebral end
(2004) found a significantly higher incidence (27.4 %) of plate. In addition to the problem of excessive stress con-
adjacent segment disease requiring further surgery at 6.7 centration, there was also a mismatch in the modulus of
years. elasticity of the steel ball and bone, especially as the device
The inconsistent clinical results and conflicting opin- was placed in the soft, central portion of the vertebral end
ions on the efficacy of spinal fusion lend strength to the plate (Bono and Garfin 2004). Despite the shortcomings of
the technique, the surgeons were thoughtful in their 14.4 Total Disc Arthroplasty Design:
approach: similar to contemporary disc arthroplasty Concepts and Biomechanics
designs, the steel ball was positioned along the spinal
motion segments sagittal arc of angular rotation at the Since their inception, the primary clinical objectives of con-
junction of the middle and posterior thirds of the disc space temporary disc arthroplasty devices have been pain relief and
(Bono and Garfin 2004). successful functional recovery (Bono and Garfin 2004).
To address these biologic failures, Fassio developed a Despite continued advancements in understanding, much work
disc arthroplasty device consisting of a central silastic com- needs to be done to optimize disc arthroplasty design and
pressible ball with intrinsic shock-absorbing properties and material properties so as to replicate the functional spinal unit.
with a larger noncompressible footprint (Fassio and
Ginestie 1978). The authors implanted the device into three
patients; however, the overall contact surface area of the 14.4.1 Biomechanics of the Functional
implant failed to prevent subsidence. In all patients, at Spinal Unit
4-year follow-up, there was implant migration and subsid-
ence into the vertebral body (Fassio and Ginestie 1978). Consisting of three components the intervertebral disc and
Kostuik also developed a device with intrinsic shock- two facet joints the functional spinal unit, also termed the
absorption properties; however, it was never implanted in spinal motion segment, is a complex articulation, which is
humans and is thus of only historical importance (Kostuik significantly different from most other joints (Lee and Goel
1997). The device consisted of an articulating hinge in the 2004; Lee and Langrana 2004). As has been discussed in
posterior third of the disc space, with a spring interposed other chapters of the book, the intervertebral disc is also
between the two metallic end plates. Despite promising made of three tissues, including the nucleus pulposus, annu-
biomechanical cyclical testing, the device failed when lus fibrosus, and the vertebral end plates. The three compo-
implanted in animals and was never made available for nents of the functional spinal unit together with the three
clinical use (Kostuik 1997). tissues of the intervertebral disc are interdependent in terms
An important design concept was the development of the of their contributions to spinal motion and function. As was
SB Charit (DePuy Spine, Raynham, MA) lumbar disc discussed in Chaps. 2 and 8, the healthy functional spinal
arthroplasty device by Bttner-Janz and Schellnack in the unit plays a significant role in resisting or transmitting loads
1980s. The initial first-generation design consisted of a small across the intervertebral disc, maintaining disc height, and
bottle-cap-like end-plate interface with a polyethylene core segmental stability (Lee and Goel 2004; Lee and Langrana
and used a synthetic-on-synthetic articulating surface (Link 2004). From a functional viewpoint, the spine cycles between
2002). However, this design also exhibited insufficient con- 100,000 and 1 million times per year and experiences load
tact area and high stress concentrations, leading to subsid- up to three times its body weight (Silva et al. 2002; Polly
ence into the vertebral bodies (Lin and Wang 2006). In an 2003; Santos et al. 2004). Under axial compression, with a
attempt to address this complication, the second-generation small contribution by the facet joints, the intervertebral disc
device used thin lateral extensions to augment the surface supports most of the load (Lee and Goel 2004; Lee and
area; however, failure occurred with fatigue fracture of the Langrana 2004). However, at higher axial compression loads
lateral extensions. This led to a third-generation device, engi- and greater extension, there may be an increased biomechan-
neered with a broad end-plate interface and manufactured ical role of the facet joints through contact of the inferior
from cobalt-chromium-molybdenum (CoCrMo): (Lin and facets with the lamina of the vertebra below (Lee and
Wang 2006). Langrana 1984, 2004; Luk et al. 1987; Weinhoffer et al.
Important lessons were learned from these early design 1995). This distribution of load is also reflected in intradiscal
concepts which have provided a base for development of pressure, which is proportional to the external load under
current arthroplasty systems. The most common impedi- axial compression; it is significantly increased with flexion
ment of the early designs was implant subsidence, and sev- and minimal with pure extension and torsion (Weinhoffer
eral general principles were adopted to prevent this et al. 1995; Lee and Goel 2004; Lee and Langrana 2004). In
complication including maximizing vertebral end-plate the degenerative disc, this complex biomechanical environ-
contact area as well as using a synthetic-on-synthetic artic- ment is altered, placing an extra burden on the facet joints,
ulating surface to avoid direct articulation with the bone. ligamentous structures, and adjacent motion segments (Lee
From these principles, there have been continued efforts to and Langrana 1984, 2004; Lee and Goel 2004).
develop and improve total disc arthroplasty devices, Total disc arthroplasty is intended to replace the dysfunc-
although none have fully replicated the healthy interverte- tional nucleus pulposus and annulus fibrosus and restore the
bral disc. biomechanical environment of a healthy functional spinal
Table 14.1 Properties of total disc arthroplasty prostheses

Implant
(year of FDA approval) Material Bearing surface Articulations Constraint COR Fixation
Cervical
ProDisc-C (2007) CoCrMo Metal on polymer 1 Semiconstrained Fixed Midline keel bone ingrowth
UHMWPE
Prestige (2007) Stainless steel Metal on metal 1 Semiconstrained Mobile Anterior screws
Bryan (2009) Titanium Metal on polymer 2 Unconstrained Mobile Milled cavities bone
Polyurethane ingrowth
Lumbar
SB Charit III (2004) CoCrMo Metal on polymer 2 Unconstrained Mobile Small fins/teeth bone
UHMWPE ingrowth
ProDisc-L (2006) CoCrMo Metal on polymer 1 Semiconstrained Fixed Midline keel bone ingrowth
UHMWPE
Adapted from Lin and Wang (2006)
COR center of rotation, CoCrMo cobalt-chromium-molybdenum alloy, UHMWPE ultrahigh molecular weight polyethylene
unit. However, as with disc degeneration, any significant many different arthroplasty devices; however, each has
deviation in implant design or placement can cause abnormal specific differences with respect to material, bearing surface,
and detrimental effects on the facet joints within the same number of articulations, constraint, mobility of the center of
segment and on adjacent levels (Lee and Goel 2004; Lee rotation, and fixation to bone (Table 14.1). The biomechani-
and Langrana 2004). Biomechanical cadaveric studies indi- cal significance of each of these design variations on the sur-
cate that a properly placed total disc arthroplasty device rounding structures such as the facet joints of the same
maintains motion within physiologic range at the treated segment and on adjacent segments has received limited
level and decreases stresses and intradiscal pressure on study, and different arthroplasty devices have not been
adjacent segments (Cunningham et al. 2003b; DiAngelo compared to each other in clinical trials. Notwithstanding, it
et al. 2004; Puttlitz et al. 2004; Dmitriev et al. 2005). is important to recognize that lumbar disc arthroplasty
devices prompted advancements in cervical disc devices.
The design concepts and biomechanical objectives were
14.4.2 Biomechanical Objectives similar; however, specific differences exist which will be
of Disc Arthroplasty further discussed.
As our understanding of cervical and lumbar spine function

has continued to increase, several biomechanical objectives 14.4.3 Disc Arthroplasty: Material
of total disc arthroplasty have been defined: preservation of Considerations and Bearing Surfaces
motion, restoration of intervertebral height, maintenance of
stability, and conversion of shock-absorption properties. The development of a successful total disc arthroplasty
Preservation of motion involves coupled motion patterns in device with acceptable long-term survival has relied on the
compression-bending and compression-torsion, as well as immense clinical and technical experience of surgeons and
the instantaneous axis of rotation (Lee and Goel 2004; Lee engineers. The lessons learned from this field of study have
and Langrana 2004). Restoration of intervertebral height is helped to better understand and improve motion-preserving
required for indirect neuroforaminal decompression, to ade- devices in the spine while avoiding some of the intrinsic pit-
quately restore spinal alignment, and to unload abnormal falls and complications especially in relation to the proper-
stresses on the facet joints. Immediate implant stability pre- ties of materials and the mechanics of the system (Santos
vents displacement or migration, maintains the stability of et al. 2004). The material properties necessary for any pros-
the spinal motion segment, and prevents abnormal motion thesis must be determined based on the requirements
and wear events which can lead to early device failure (Lee expected of the implant (Taksali et al. 2004). For example,
and Goel 2004; Lee and Langrana 2004). Finally, shock the lumbar spine sustains significantly greater loads than the
absorption incorporates load transmission with shock attenu- cervical spine, whereas the cervical spine has a different
ation and ultimately prevents abnormal stress concentration kinematic pattern and greater range of motion. From this
on surrounding structures and adjacent motion segments. perspective, materials that perform well in the cervical spine
These biomechanical objectives have been used to develop may perform poorly in the lumbar spine.
Modern disc arthroplasty devices have a prosthesis-bone stiffness, energy absorption, and viscous damping of metal-
interface consisting of a broad rigid end plate made of metal; on-polyurethane cervical disc arthroplasty with both a
most devices are flat or slightly convex (dome) shaped. Not fusion construct and an intact intervertebral disc. The inves-
surprisingly, implant geometry is an important factor in disc tigators found the dynamic stiffness of metal-on-polyure-
arthroplasty, the goal being to maximize prosthesis-bone thane was similar to that of the intact disc, and energy
contact for bony ingrowths and to prevent subsidence. The absorption and viscous damping exceeded both the intact
primary alloys used in disc arthroplasty devices are stainless disc and fusion construct. In another study, Dahl et al.
steel, cobalt chrome and titanium alloys. Cobalt chrome (2011) compared the shock-absorbing properties of poly-
alloys have the greatest wear resistance, whereas titanium urethane, polyethylene, and titanium alloy bearing surface
alloys have poor wear characteristics and surface hardness, materials. These workers demonstrated that polyurethane
making them unacceptable as an articulating surface. provides significantly lower stiffness and greater energy
Titanium alloys demonstrate excellent biocompatibility with absorption and damping characteristics than polyethylene
less susceptibility to bacterial surface colonization. Also, and titanium alloy. Of note, titanium alloy is currently not
they generate a significantly less artifact during computed available as a bearing surface due to its poor wear charac-
tomography or magnetic resonance imaging (Arens et al. teristics. A concern with the use of polyurethane is the bio-
1996; Hallab et al. 2003a; Santos et al. 2004). They are most logic durability or wear resistance. However, polyurethane
commonly used as a coating of the end-plate interface to has compared favorably to polyethylene and has been found
facilitate bony ingrowth (Santos et al. 2004). to have excellent wear properties, with wear particles that
In terms of a bearing surface, the basic requirements is do not incite a significant inflammatory response (Anderson
that the articular surface must allow for mobility, load distri- et al. 2003, 2004; Taksali et al. 2004; Pitzen et al. 2007).
bution, low friction, and high wear resistance, as well as While these reports acknowledged that shock absorption
being biologically compatible with adequate longevity remains a theoretical benefit of disc arthroplasty devices,
(Taksali et al. 2004; Lin and Wang 2006). Like total joint no clinical study has compared outcomes of shock- versus
arthroplasty, the principal bearing surfaces of disc arthro- non-shock-absorbing devices (LeHuec et al. 2003).
plasty devices are metal-on-polymer or metal-on-metal artic- Also of importance is the polymer core design, which
ulations. Although ceramic-bearing surfaces have gained predominately exists as either a single-gliding surface with
some popularity due to excellent wear characteristic, the a securely fixed polymer core or one with a metal end plate
brittle material properties and concerns for catastrophic fail- (ProDisc; Synthes Spine, Paoli, PA) or a double-gliding
ure in proximity to neural elements have precluded significant surface with a mobile polymer core sandwiched between
development (Garino 2000; Santos et al. 2004). In fact, a two metal end plates (Bryan Cervical Disc; Medtronic
recent case report has demonstrated catastrophic failure of an Sofamor Danek, Memphis, TN). Concerns with the fixed
experimental ceramic-bearing surface disc device implanted polymer core are the presence of another point of implant
in a human patient (Nguyen et al. 2011). failure due to constant micromotion, stress concentration,
The two primary polymers used for arthroplasty devices and a large difference in elastic properties at the metal-
have been polyethylene and polyurethane. Ultrahigh molec- polymer interface. Problems associated with a mobile poly-
ular weight polyethylene has a good track record for sev- mer core include concerns for polymer extrusion as well as
eral decades in extremity arthroplasty and has avoided increased debris from wear of the two articulating surfaces.
complications associated with poor wear properties due to Not surprisingly, there are continued research efforts to
polyethylene sterilization techniques (Kurtz et al. 1999a, b; maximize the design properties and wear characteristics of
Hallab et al. 2003a; Taksali et al. 2004). Polyurethane has metal-on-polymer bearing surfaces. Questions remain
been used for decades for cardiovascular devices but has regarding wear debris and subsequent osteolysis with late
had limited use in spine surgery. Recently, the importance implant failure, although the amount of debris from disc
of its shock-absorbing properties has been recognized and arthroplasty devices may be insignificant when compared
incorporated into a disc arthroplasty device, in particular to hip and knee joints due to the low range of motion and
the Bryan cervical disc replacement (Medtronic Sofamor cyclic rate (Santos et al. 2004).
Danek, Memphis, TN). However, there are few studies Despite concerns for systemic metal deposition (Wagner
evaluating axial motion and load transfer properties. This is and Wagner 2000; Bisseling et al. 2011), adverse soft tissue
unfortunate because one of the functions of the normal, reactions (i.e., pseudotumor) (Williams et al. 2011), and
healthy intervertebral disc is to provide shock absorption, higher rates of early failure for some total hip replacement
and if absent may produce abnormal stress concentrations implants, metal-on-metal articulations have been used in
on surrounding structures within the segment and at the total hip arthroplasty devices (Bernthal et al. 2012; Langton
adjacent segment (Dahl et al. 2006). Dahl evaluated axial et al. 2011). The advantage of a metal-on-metal articulation
compared to metal-on-polymer bearing surface is the Unlike total joint arthroplasty which has successfully used
significantly lower (approximately 10 times) wear rate cemented knee arthroplasty implants and cemented femoral
(Goldsmith et al. 2000; Santos et al. 2004; Taksali et al. stem implants as originally advocated by Charnley (Schulte
2004). However, there have been reports of elevated sys- et al. 1993), disc arthroplasty has not been performed with
temic metal ion levels in patients following lumbar metal- cement fixation because of the proximity to neural elements
on-metal total disc replacement (Wagner and Wagner 2000; (Santos et al. 2004). Another reason cement fixation may be
Bisseling et al. 2011). A recent review of complications in inappropriate for disc arthroplasty is because of the younger
metal-on-metal disc arthroplasty devices reported abnor- patient population compared to extremity arthroplasty, which
mal inflammatory reactions and soft tissue masses with may potentially increase the incidence of cement fatigue and
lymphocyte or macrophage infiltration, similar to those aseptic loosening due to greater activity levels and physical
found in patients following metal-on-metal bearing surface demands (MacWilliam et al. 1996; McLaughlin and Lee
total hip arthroplasty (Golish and Anderson 2012). Because 2000; Santos et al. 2004).
of this paucity of information, the surgical community Therefore, immediate stability is obtained by screw
awaits information on the long-term outcomes and fixation of metal end plates to the anterior vertebral body in
complication profile concerning the use of metal-on-metal some devices, whereas others have end-plate designs with a
disc arthroplasty devices. midline fin/keel, or spikes that project perpendicular to the
end plate (Lee and Goel 2004). There have been some con-
cerns regarding screw fixation, particularly in the lumbar
14.4.4 Semiconstrained Versus spine which presents a greater risk for major vascular com-
Unconstrained Devices plications. In the cervical spine, due to the increased anterior
profile, there is the possibility of causing dysphagia; due to
Similar to total joint arthroplasty implants, the amount of fixation on the anterior cervical body, difficulties may be
constraint is an important factor in disc arthroplasty design. experienced in revision surgery in the setting of adjacent seg-
With increasing constraint in total joint arthroplasty, the ment disease (Kulkarni et al. 2003; van Ooij et al. 2003;
trade-off is greater stability for greater stress on the implant- Bertagnoli et al. 2005b; Patel et al. 2008).
bone interface. For disc arthroplasty, there are both semicon- Long-term stability requires osseointegration achieved
strained and unconstrained designs, with each having through porous ingrowth surfaces or on-growth surface
advantages and disadvantages (Huang et al. 2003; Santos coatings (Taksali et al. 2004). The basic requirements of
et al. 2004). The unconstrained design provides a mobile successful bone ingrowth include implant stability, optimal
instantaneous axis of rotation and is more representative of pore size, and optimal surface geometry/surface area
the physiologic axis, which was demonstrated by Gertzbein (Kienapfel et al. 1999; Santos et al. 2004). Surface coat-
et al. (1986) to be an ellipse rather than a single point. ings to improve bone on-growth include roughened tita-
Theoretically, an unconstrained device would provide a nium, titanium wire mesh, plasma-sprayed titanium, and
greater range of motion and may be more tolerant of small bioactive materials such as hydroxyapatite and calcium
errors in implant placement (Hallab et al. 2003a; Huang et al. phosphate (Taksali et al. 2004). Animal studies have shown
2003). However, an unconstrained articulation subjects the successful bone integration with titanium- or hydroxyapa-
facets and posterior ligaments to increased shear and tor- tite-coated surfaces (Cunningham et al. 2002, 2003a;
sional loads, whereas semiconstrained devices unload the Santos et al. 2004). In fact, Cunningham et al. (2002)
facet joints and ligaments (Cunningham et al. 2003a; Huang reported that a lumbar disc arthroplasty device with porous-
et al. 2003; Polly 2003; Santos et al. 2004). Semiconstrained coated titanium end plates in a nonhuman primate model
disc arthroplasty devices have increased stability, but similar evidenced bone ingrowth into 56 % of the end-plate sur-
to joint arthroplasty there is higher load transfer to the face at 12 months. This is comparable to total joint replace-
implant-bone interface. Currently, the ideal amount of con- ment components, which have found cementless femoral
straint remains unknown and will eventually be determined and acetabular implant ingrowth of 9.733 % (Sumner
by long-term clinical studies. et al. 1990; Harvey et al. 1999) and 12 % (Pidhorz et al.
1993), respectively. However, these coatings should con-
tinue to be carefully evaluated to ensure acceptable tensile
14.4.5 Implant Fixation strength, shear strength, and fatigue strength. The potential
concern is that inadequate surface coating integrity may
Fixation of the disc arthroplasty device to host bone is cause migration of coating material debris in between the
another important factor to consider, taking into account both articular surface, leading to accelerated third-body wear
initial and long-term implant stability (Santos et al. 2004). (Taksali et al. 2004).
14.5 Comparison of Total

Disc Arthroplasty Devices suggests the device will be effective with no significant
safety concerns.
Cervical and lumbar disc replacements have been catego- There have been only a few total disc arthroplasty
rized as significant risk devices by the United States Food devices which have received FDA premarket approval,
and Drug Administration (FDA) and require both preclinical which is a rigorous process of scientific and regula-
and clinical evaluations to determine their safety and effec- tory review to evaluate the safety and effectiveness of
tiveness (Orr et al. 2007). As noted previously, many of these Class III medical devices. The completion of a pre-
devices have been used in Europe for decades, and only market approval for a device usually requires the
recent interest in the United States has resulted in several manufacturer to dedicate a significant amount of time,
ongoing FDA-sponsored clinical trials, comparing the out- energy, and funding. However, the FDA allows certain
comes and safety of disc arthroplasty devices, using fusion modified devices to undergo the 510(k) notification
procedures as the control. Although there have been over 100 process which relies on comparisons with an already-
disc arthroplasty designs and patents, to date there have been approved and similar device and provides an expe-
two lumbar disc replacement devices (SB Charit III, DePuy dited review, usually within 90 days. The 510(k)
Spine, Raynham, MA; and ProDisc-L, Synthes Spine, Paoli, notification does not require the modified device to
PA) and three cervical disc replacement devices cleared for undergo extensive clinical testing, thereby avoiding
marketing (Prestige, Medtronic Sofamor Danek, Memphis, the rigorous premarket approval process. Many man-
TN; ProDisc-C, Synthes Spine, Paoli, PA; Bryan, Medtronic ufacturers attempt 510(k) notifications for their
Sofamor Danek, Memphis, TN) by the FDA (Orr et al. 2007). devices because of the ability to rapidly market their
All devices have demonstrated that human use for one-level product; however, this process may permit a modified
degenerative disc disease is non-inferior to the fusion con- device to be used in a widespread fashion without a
trol group (Orr et al. 2007). We will further discuss the full understanding of the risks, failure rate, and
specific design considerations for each FDA-approved complications.
device.
14.5.1 Lumbar TDA Devices

Box 14.1 FDA Investigational Device Exemption (IDE),
Premarket Approval, and 510(k) Notication The SB Charit III (DePuy Spine, Raynham, MA)
Cervical and lumbar disc replacements are categorized (Fig. 14.1) was the first total disc arthroplasty device to be
as significant risk devices (Class III) by the US Food implanted in the USA in 2000 as part of a controlled ran-
and Drug Administration (FDA) and have required domized study and was subsequently the first FDA-approved
both preclinical and clinical evaluations to determine disc arthroplasty device in 2004. SB Charit has evolved
their safety and effectiveness prior to marketing. The from its initial design (SB Charit I and II) designed by
FDA authorizes an investigational device exemption Shellnack and Bttner-Janz in the early 1980s, which con-
(IDE), which allows a device to be used for the pur- sisted of small, shell-like end plates made of stainless steel,
pose of a clinical study to collect safety and effective- which resulted in implant subsidence and migration
ness data to support a premarket approval or premarket (Buttner-Janz et al. 2002; Link 2002). The Charit II
notification 510(k). When under IDE status, the device attempted to address this problem by adding thin lateral
remains unapproved by the FDA and typically can only wings to increase the surface area of the end plates; how-
be used on human subjects when it is under clinical ever, these wings developed early fatigue fractures (Lin and
investigation and used by investigators participating in Wang 2006). The device is now in its third and current
the clinical trial. However, the FDA also allows the use version and was developed in 1987 with broad flat end
of an unapproved device before premarket approval for plates manufactured from cobalt-chromium-molybdenum
emergency use to save the life of a patient or prevent (CoCrMo) alloy with small teeth projecting into the verte-
irreversible morbidity, for compassionate use to help bral end plates (Lin and Wang 2006). There is also the new
a patient suffering from a serious disease or condition InMotion design, which retains the essential characteristics
for which there exists no other alternative therapy, and of the Charit III, with the modifications involving the teeth
after an IDE clinical trial for continued access if orientation and the addition of a central rail portion allow-
there is a public health need, and preliminary evidence ing the prosthesis to glide onto a ramp inserter (Serhan
et al. 2011). The device end plates are porous-coated with
a b
Fig. 14.1 (a, b) SB Charit III lumbar disc arthroplasty device (Images provided by DePuy Spine, Raynham, MA)
plasma-sprayed titanium and calcium phosphate (TiCaP) to six small teeth of the SB Charit III. The semiconstrained
assist with bony ingrowth. The polyethylene core is a interface between the polyethylene core fixed to the inferior
sliding, unconstrained biconvex design to allow for an end plate articulating with a polished superior metallic end
instantaneous axis of rotation (IAR) that permits anterior plate is thought to decrease the risk of polymer extrusion.
and posterior movement to the midpoint of the disc during However, the rotational axis lies within the anterosuperior
flexion and extension, respectively, which is believed to aspect of the lower vertebral body, and the fixed axis of
more closely replicate normal segmental motion (Lin and rotation does not allow for an anatomic coupled anteropos-
Wang 2006). However, the IAR of the arthroplasty device terior translation with flexion and extension (Lin and Wang
has been found to be more anterior than normal, in both 2006). In patients with a degenerative spinal motion seg-
flexion and extension, and may detract from its presumed ment with abnormally increased range of motion, the semi-
advantage (Bono and Garfin 2004). Cunningham showed constrained design is thought to be beneficial by allowing
that the motion of the Charit III prosthesis closely resem- stability through a more controlled arc of motion and pro-
bles normal motion in cadaveric testing, with disc motion tecting the facet joints from shear forces. However, some
similar to the intact segment in flexion-extension and lat- authorities believe that increased constraint may cause
eral bending; however, normal limits of axial rotation were abnormal force transfer to the bone-end plate interface
exceeded (Cunningham 2004). Another disadvantage is resulting in premature loosening, abnormal forces within
that there are two articulating surfaces which theoretically the facet joints, and anteroposterior dimensional changes of
may increase polyethylene wear and debris formation, the neuroforamina during motion (Huang et al. 2003; Bono
compared to only one gliding surface. Also, the polyethyl- and Garfin 2004).
ene core is unconstrained with the potentially catastrophic
complication of core extrusion.
ProDisc-L (Synthes Spine, Paoli, PA) (Fig. 14.2) was 14.5.2 Cervical TDA Devices
designed and developed by Marnay in the 1980s and
first implanted in France in 1990. The device received ProDisc-C (Synthes Spine, Paoli, PA) (Figs. 14.3 and 14.4)
FDA approval in 2006 after undergoing several design is similar in design to its lumbar counterpart and received
modifications, including a change of end-plate material from FDA approval in 2007. The device is also made up of three
titanium to CoCrMo and the addition of ultrahigh molecu- components, which include an inferior and superior
lar weight polyethylene (UHMWPE) bearing surface as a CoCrMo alloy end plate with a midline keel-oriented
separate modular piece which is snap-locked to the inferior anteroposterior anchoring into the end plate of the respec-
end plate (Lin and Wang 2006). A single, titanium plasma tive vertebral body. In addition there is a highly polished
spray-coated midline sagittal fin is used to improve imme- concave bearing surface from the superior alloy end plate
diate and long-term end-plate fixation, as opposed to the which articulates with a convex (spherical dome) UHMWPE
b c
Fig. 14.2 (ac) ProDisc-L lumbar disc arthroplasty device (Images provided by Synthes Spine, Paoli, PA)
insert that is preassembled and snap-locked into the inferior The Prestige ST (Medtronic Sofamor Danek, Memphis,
alloy end plate (Lin and Wang 2006). Similar concerns TN) (Figs. 14.5 and 14.6) was developed by Gill and col-
exist regarding the semiconstrained articulation and may be leagues in 2002 and was approved by the FDA in 2007. The
of greater concern in the cervical spine due to the inherent Prestige ST is a modification of the original Prestige I (1989)
greater range of motion; however, this has yet to be and II (1999) initially designed and developed by Bristol-
evaluated. Cummins at Frenchay Hospital in 1989 (Traynelis 2004).
a
c
Fig. 14.3 (a-c) ProDisc-C cervical disc arthroplasty device (Images provided by Synthes Spine, Paoli, PA)
The key improvements between Prestige I and Prestige II inferior, anterior vertebral bodies (Bono and Garfin 2004).
were a more anatomic end-plate design, which was rough- The ball-in-trough design provides semiconstrained motion;
ened/grit-blasted to promote bony ingrowth; an improvement however, unlike the ProDisc ball-in-socket design, there is
from Prestige II to Prestige ST was a 2-mm reduction in the some coupled translation with flexion-extension. Again, the
height of each anterior flange. The device is a semicon- consequences of this added translation are unknown, and
strained metal-on-metal prosthesis (stainless steels), with a further long-term follow-up is necessary to determine if there
ball-in-trough design which is postulated to replicate physi- are detrimental effects on the facets joints from shear
ologic segmental motion (Traynelis 2004). To provide forces.
immediate stability, and unique to other designs, screws are The Bryan Cervical Disc (Medtronic Sofamor Danek,
placed through plate-like extensions on the superior and Memphis, TN) (Figs. 14.7 and 14.8) was developed in the
a b
Fig. 14.4 (a, b) Radiographs of patient with single-level ProDisc-C device
late 1990s and received FDA approval in 2009. The device is shock-absorption and load-dampening properties compared
an unconstrained, biarticulating metal-on-polyurethane to polyethylene and metal bearing surfaces; however, the
prosthesis. It is composed of two titanium alloy shells, a clinical benefits have not been established.
polycarbonate polyurethane nucleus, and a polyether poly-
urethane sheath with titanium-retaining wires. The polyure-
thane sheath spans between the metal end plates and Box 14.2 Spinal Device Registry
surrounds the nucleus, forming a cavity that is filled with The development of a spinal device registry within
saline, acting as synovial fluid or lubricant (Bono and Garfin the USA remains a necessity, particularly with the
2004). This is thought to keep potential wear debris within real and potential benefits demonstrated by registries
the cavity while also preventing soft tissue ingrowth for other medical devices. Registries for hip and knee
between the articulating surfaces (Lin and Wang 2006). The replacement devices have become a worldwide real-
device also has a unique method of fixation to the bone, ity with preeminent registries in Sweden, Finland,
sitting in a pocket milled into the vertebral end plate; Norway, Australia, Denmark, and New Zealand,
long-term stability is provided by bone incorporation with a approaching 15 years of experience. A device regis-
beaded titanium coating on the device end plates it is not try allows a real-time assessment of the current clini-
screwed or secured by teeth to the vertebra. As previously cal practice and associated outcomes, providing
discussed, the polyurethane nucleus provides greater
ments), and analysis of registry data generated

sufficient evidence to validate continued federal health
insurance reimbursement. The success of the Swiss
spine registry should serve as an example for the
development of a US spinal device registry and
collection of observational data in a nationwide
framework to enhance the quality of patient care and
tracking the performance of spinal implants.
Box 14.3 Evolution of Total Disc Arthroplasty Devices:

Timeline
1950s Fernstrom ball, considered the first rudimen-
tary intervertebral disc prosthesis
1978 Fassio develops disc arthroplasty device with
a central silastic compressible ball
1997 Kostuik develops a device with an articulating
hinge and springs between two metallic end plates;
however, it was never implanted in humans
SB Charite Lumbar Disc Arthroplasty
Early 1980s SB Charit I lumbar disc arthroplasty
device developed by Bttner-Janz and Schellnack
1987 SB Charit III, third generation and current
version was developed
2000 SB Charit III, first total disc arthroplasty
Fig. 14.5 Prestige ST cervical disc arthroplasty device (Images pro-
device implanted in the USA
vided by Medtronic Sofamor Danek USA, Inc.) 2004 SB Charit III, approved by US FDA for
marketing
2011 SB Charit III, removed from market by
manufacturer
ProDisc-L and ProDisc-C Disc Arthroplasty
1980s ProDisc-L, developed by Marnay
timely feedback, and permits healthcare organiza- 1990 ProDisc-L, first implanted in France
tions to significantly influence physician behavior. 2006 ProDisc-L, approved by US FDA for marketing
Most importantly, to avoid unnecessary complica- 2007 ProDisc-C, approved by US FDA for marketing
tions for patients, device registries provide an early Prestige Cervical Disc Arthroplasty
warning system for early implant failure. Device reg- 1989 Prestige I, developed by Bristol-Cummins at
istries also validate the value and cost-effectiveness Frenchay Hospital
for the use of healthcare resources and promote 1999 Prestige II, modified with a more anatomic
improved clarity and evidence in the development of end-plate design
clinical practice guidelines. 2002 Prestige ST, developed by Gill and colleagues,
The Swiss Spine Registry is the first national spine modified with a 2-mm reduction in the height of
device registry started in March 2005 to gather data on each anterior flange
patient outcomes, cost, and performance information 2007 Prestige ST, approved by US FDA for
for cervical and lumbar total disc arthroplasty and bal- marketing
loon kyphoplasty procedures. A 3-year pilot study with Bryan Cervical Disc Arthroplasty
135 participating Swiss surgeons demonstrated an Late 1990s Bryan Disc, developed by Medtronic
80 % data capture rate for total disc arthroplasty cases 2009 Bryan Disc, approved by US FDA for
performed (925 cervical and 497 lumbar disc replace- marketing
a b
Fig. 14.6 (a, b) Radiographs of patient with single-level Prestige ST cervical disc device
14.6 Operative Technique
Careful preoperative planning is required for each patient,

and although there is currently no method to template for
implant size/positioning, the surgeon should be fully famil-
iar with implant-specific instructions, instrumentation, and
sizing options. Additional equipment and implants for a
fusion procedure should also be readily available in the rare
instance the arthroplasty procedure must be aborted. The
operative techniques for both lumbar and cervical disc
replacements use an anterior approach. In the lumbar spine,
an anterior retroperitoneal dissection is performed to access
the desired disc space, and assistance may be provided by a
general or vascular surgeon. In the cervical spine, a standard
Smith-Robinson anterior approach is used to access the dis-
eased level. Once adequate exposure is obtained, the dis-
eased intervertebral disc is removed, and the end plates of
Fig. 14.7 Bryan cervical disc arthroplasty device (Images provided by the vertebral bodies are prepared based on device-specific
Medtronic Sofamor Danek USA, Inc.) requirements. Some require arduous end-plate preparation,
a b
c d
Fig. 14.8 (ad) Radiographs of patient with single-level Bryan cervical disc arthroplasty device (Images provided by Medtronic Sofamor Danek
USA, Inc.)
whereas others recommend limited end-plate disruption to enrollment criteria of FDA IDE trials (Mummaneni et al.
reduce the risk of implant subsidence. An important step for 2007; Sasso et al. 2007; Zigler et al. 2007; Guyer et al. 2009;
most devices is to adequately release the posterior longitudi- Murrey et al. 2009). The primary indication for lumbar TDA
nal ligament to allow correct positioning and function of the is isolated mechanical discogenic back pain without radicul-
prosthesis. At this step, each arthroplasty device varies in opathy or instability at L3L4, L4L5, or L5S1 interverte-
specific technique, instrumentation, and implant position- bral levels. Degenerative disc disease as the primary symptom
ing/sizing, and for lumbar TDA the lordotic angle should be source is corroborated on CT or MRI studies with one or
verified using biplanar fluoroscopic imaging (Lin and Wang more of the following findings: vacuum disc sign, contained
2006). In positioning most prostheses, the axis of rotation nucleus pulposus, absence of lateral recess stenosis, paucity
should be posteriorly located in the disc space, but caution of facet joint degeneration changes, decrease of interverte-
should be taken due to the risk of encroachment of the neural bral disc height of greater than 4 mm, scarring/thickening of
elements (Santos et al. 2004). Appropriately sizing the the annulus fibrosus, formation of degenerative cyst, or mar-
implant is also important to restore intervertebral disc space ginal vertebral body osteophytes. Controversy continues to
height and provide indirect decompression, and to allow exist regarding the utility and validity of discography as an
normal soft tissue and ligamentous tension, thereby provid- investigative tool (Sandhu et al. 2000; Carragee and Alamin
ing additional stability to the implant. Also, a prosthesis 2001). However, a provocative discogram can be used to fur-
covering a large surface area is desired to reduce the risk of ther delineate the symptomatic levels, with a positive result
subsidence as well as allowing greater range of motion demonstrating concordant pain reproduction and at least one
(Cinotti et al. 1996; Santos et al. 2004). control level that is not painful and does not reproduce the
patients symptoms. Otherwise, TDA is contraindicated in
patients with obesity (>1 standard deviation above normal
14.7 Deciding When to Use Total Disc body mass index), osteopenia, chronic steroid use, insulin-
Arthroplasty requiring diabetes mellitus, pregnancy, previous lumbar
fusion, objective evidence of nerve root compression, spinal
While ideal implant design and surgical technique are impor- fracture, spondylolysis, spondylolisthesis, scoliosis, spinal
tant, more critical are appropriate patient selection and diag- tumor, stenosis, or severe facet joint arthrosis (McAfee 2004;
nosis. The importance of determining the true source of Madigan et al. 2009). The only exception for treatment of
symptoms cannot be overemphasized, as failure to recognize radicular pain using a lumbar TDA may be carefully selected
and treat all sources of pain will result in a suboptimal out- cases of neuroforaminal stenosis that can be corrected by
come. History and physical examination are vital in deter- restoring intervertebral disc and neuroforaminal height
mining the true source of symptoms and may be complemented through TDA placement (Table 14.2).
with radiologic investigations such as plain radiographs,
computed tomography (CT), myelography, and magnetic
resonance imaging (MRI), when indicated. For degenerative 14.7.2 Indications for Cervical Total Disc
disc disease, the ideal surgical candidate is one who has Arthroplasty
severe, functionally debilitating symptoms and has exhausted
all conservative treatment modalities for a minimum of 6 The primary indication for cervical TDA is radiculopathy
months in the lumbar spine and 6 weeks in the cervical spine and/or myelopathy due to disc herniation or spondylosis
(McAfee 2004; Santos et al. 2004). Conservative manage- without instability, requiring discectomy/decompression at
ment may include physical therapy, facet joint injections, intervertebral levels between C3 and T1 (Orr et al. 2007).
epidural steroids, acupuncture, back school, behavior The patient must have an abnormal neurologic examination
modification, ultrasound, anti-inflammatory medications, indicative of radiculopathy or myelopathy, which may
analgesic medications, muscle relaxants, lumbosacral stabi- include abnormal reflexes or decrease in sensation or motor
lization therapy, and orthotic management (McAfee 2004). strength in a correlating dermatome/myotome. Also, a focal
compressive lesion must be observed by CT, myelography,
or MRI (McAfee 2004). Cervical TDA should be avoided in
14.7.1 Indications for Lumbar Total Disc patients with ankylosing spondylitis, rheumatoid arthritis,
Arthroplasty ossification of the posterior longitudinal ligament, or diffuse
idiopathic skeletal hyperostosis, as well as a relative con-
Appropriate patient selection requires an understanding of traindication in patients with obesity (>1 standard deviation
the indications and contraindications for lumbar total above normal body mass index), osteopenia, chronic steroid
disc arthroplasty, which have been established from the use, insulin-requiring diabetes mellitus, pregnancy, previous
Table 14.2 Current indications for lumbar disc arthroplasty Table 14.3 Current indications for cervical disc arthroplasty
Indications Relative contraindication Indications Relative contraindication
Symptomatic 1- to 2-level Central or lateral recess Radiculopathy and/or Isolated axial neck pain
discogenic back pain without stenosis myelopathy due to disc
radiculopathy or instability at herniation or spondylosis
L3L4, LL5, or L5S1 levels without instability, requiring
Concordant degenerative disc Facet arthroplasty discectomy, or decompression
disease on CT or MRI studies at intervertebral levels between
or discograms C3 to T1, with 1- to 3-level
Failure of >6 months Obesity (>1 standard deviation disc disease
of conservative treatment above normal body mass Concordant focal compressive Ankylosing spondylitis
index) lesion on CT, myelography, or
Specific radiographic findings Osteopenia MRI
include vacuum disc sign, Failure >6 weeks of conserva- Rheumatoid arthritis
contained nucleus pulposus, tive treatment Ossification of the posterior
absence of lateral recess stenosis, longitudinal ligament
paucity of facet joint degeneration Diffuse idiopathic skeletal
changes, decrease of intervertebral hyperostosis
disc height of greater than 4 mm, Cervical instability
scarring/thickening of the annulus
Previous spinal fusion/infection/
fibrosus, formation of degenerative
fracture
cyst, or marginal vertebral body
osteophytes Obesity (>1 standard deviation
above normal body mass index)
Treatment of radicular pain may be Chronic steroid use
performed in carefully selected Osteopenia
Insulin-requiring diabetes
cases of neuroforaminal stenosis mellitus Chronic steroid use
Pregnancy Insulin-requiring diabetes mellitus
Previous lumbar fusion/ Pregnancy
infection/fracture Spinal tumor
Objective evidence of nerve Severe facet joint arthrosis
root compression
Spondylolysis/
spondylolisthesis
Scoliosis
14.8 Total Disc Arthroplasty
Spinal tumor Revision Strategies
Spinal stenosis
Total disc arthroplasty failure and complications may require
revision or removal of the prosthesis; however, the long-term
cervical spinal infection, spinal fracture, or severe facet joint failure rate and all potential complications remain unknown.
arthrosis. Axial neck pain as a solitary symptom is also a Moreover, there are currently no known specific consider-
contraindication for cervical TDA, which is in contrast to ations or protocols for revision or replacement surgery (Lee
lumbar TDA where an ideal candidate has isolated mechani- and Goel 2004). Failure of disc arthroplasty can be due to
cal back pain with no radicular symptoms (McAfee 2004). A many different factors, including suboptimal surgical tech-
potential complication of cervical arthroplasty is the poten- nique, implant malpositioning, and poor patient selection;
tial for recurrent radiculopathy resulting from spondylotic others include mechanical implant and biologic failure
progression and/or ossification (Albert and Eichenbaum (Kostuik 2004; Bertagnoli et al. 2005b; Patel et al. 2008). In
2004). Because motion preservation may lead to recurrence fact, the Charit FDA IDE study evaluating 304 patients
of spondylosis when compared to fusion procedures, patients found 17 % had suboptimal or poor placement of the pros-
with spondylotic radiculopathy or myelopathy may require thesis, which was significantly correlated with ODI and VAS
wider uncinate resection and decompression. Although not scores (McAfee et al. 2005). Van Ooji et al. (2003) reported
studied in the FDA IDE trials, there are reports of implanting on 27 patients with failed SB Charit devices, with all index
cervical TDA for adjacent-level disease next to an estab- procedures performed at other institutions. The failures
lished fusion (Kim et al. 2003); these have been performed in occurred 53 months (range 11127 months) after initial sur-
multilevel TDA constructs (Cardoso and Rosner 2010) and gery. All replacements were either at L4L5 or at L5S1
in hybrid constructs with TDA above a primary fusion levels, and the most common causes of failure were adja-
(Cardoso et al. 2011). There have been few reported compli- cent-level spinal disease, subsidence, and facet joint arthro-
cations and adverse implant-related events (Table 14.3). sis. Two patients experienced anterior dislocation of the
implant, and 11 patients required additional salvage surgery helpful starting point. Also, bacterial culture after aspiration
(Kostuik 2004). Other authors of ongoing clinical trials with of periprosthetic fluid, with saline lavage, can provide
short-term follow-up have reported a low incidence of infec- additional information regarding the presence of infection,
tion, vertebral body fracture, implant malposition, subsid- as well as sensitivity analysis for antibiotic treatment
ence, mechanical failure, and paravertebral heterotopic (Kostuik 2004). Evidence of bone resorption on CT scan
ossification (Delamarter et al. 2003; Lin and Wang 2006; may be another indicator of infection, but it is not specifically
Guyer et al. 2009). diagnostic (Kostuik 2004). However, in the presence of a
Indications for revision surgery of a failed total disc metal and/or plastic prosthesis, nonsurgical management
arthroplasty may include, but are not limited to, implant with targeted antibiotics may be insufficient to eradicate
loosening, malposition, displacement, early wear, and infec- infection. Use of long-term suppressive antibiotics may be
tion (Kostuik 2004). There is currently limited understand- indicated if the patient is unable to undergo revision sur-
ing of the effects of wear debris and osteolysis on spinal gery. Otherwise, the surgical technique for management of
arthroplasty devices, and the long-term effects, such as asep- infection would consist of implant removal, meticulous
tic loosening seen in extremity joint replacements, remain debridement, and irrigation, followed by fusion and an
unknown. With metal-on-metal disc arthroplasty, patients extended course of targeted antibiotics. Fusion may be best
experience an inflammatory response similar to that seen in achieved in the setting of infection, with structural autograft
total joint arthroplasty (Hallab et al. 2003b; Golish and rather than allograft, and consideration should be given to a
Anderson 2012). staged posterior fusion procedure with instrumentation
A comprehensive strategy for treating failed disc arthro- (Kostuik 2004).
plasty begins by defining patient symptoms and the radio- Surgical options for disc arthroplasty failure include
graphic status of the arthroplasty device (Patel et al. 2008). revision arthroplasty, anterior interbody fusion, or instru-
Asymptomatic patients with implant subsidence or migra- mented posterior fusion. A particular problem with removal
tion without extrusion can be treated nonsurgically with or revision arthroplasty for devices in current use is the
frequent serial examinations. Symptomatic patients may surgery through previous scar tissue. This may be difficult
present with (1) continued pain secondary to implant fail- for revision cervical arthroplasty, but for lumbar exposure,
ure, (2) pain due to symptomatic disease at an adjacent this is a challenge even for a technically advanced vascular
level, (3) pain due to late surgical site infection, or may be surgeon (McAfee 2004; Patel et al. 2008). For lumbar revi-
categorized as (4) continued pain of unknown etiology sion surgery through an anterior approach, preventative mea-
(Kostuik 2004). Although differentiating between these sures can reduce the incidence of life-threatening adverse
states may be difficult, it is important to determine whether events. The ureter must be carefully identified and ureteral
the new pain resembles or is significantly different from the stents may be placed before reexposure to prevent iatrogenic
pain/symptoms originally treated by the disc replacement. injury (Wagner et al. 2006; Patel et al. 2008). Also, angiog-
A diagnostic algorithm for patients with new pain after a raphy and venography can define the location and distortions
period of relief or continued pain of unknown etiology may of the great vessels due to scar tissue, and catheterization of
include careful attention to history and physical examina- the iliac vessels with inflatable balloon catheters can prevent
tion, the use of radiographs with flexion-extension views, catastrophic blood loss (Patel et al. 2008).
and a CT scan to ensure adequate prosthesis placement, Noteworthy, removal of the arthroplasty device may
stability, and incorporation. If imaging fails to reveal an create a significant bone loss that would preclude revision
underlying pathology, further investigations may include arthroplasty (Kostuik 2004). If there is infection and prosthe-
injection of local anesthetic and radiosensitive dye into the sis failure due to displacement, loosening, or malposition,
periprosthetic area, a facet block at the level of arthroplasty, anterior interbody fusion will be required. In the absence of
or discography at adjacent levels (Kostuik 2004). These infection, after removal of the failed arthroplasty device and
modalities can provide additional information and identify preparation of the interbody space, the use of allograft should
a potential source of pain which may have not been recog- enhance fusion; however, some cases may warrant same-day
nized prior to the arthroplasty procedure. If the pain is posterior fusion and/or instrumentation at the level of the
relieved after periprosthetic injection, the assumed diagno- revision surgery (Kostuik 2004). The use of instrumented
sis would be implant loosening, malposition, or displace- posterior fusion or dynamic stabilization alone at the level of
ment (Kostuik 2004). the failed arthroplasty has been suggested, but its role has not
Periprosthetic loosening from infection may be more been defined and may be insufficient to alleviate symptoms.
difficult to differentiate, and laboratory tests associated with It should be stated that because there are no significant
infection such as leukocytosis, elevated erythrocyte sedi- reports or experience with failed arthroplasty devices,
mentation rate, and high C-reactive protein levels may be a protocols and strategies for revision arthroplasty remain
speculative and unclear. Continued long-term follow-up, of to nuclear replacement may include advanced disc space
at least 510 years, will likely further elucidate the best man- collapse (<5 mm of residual disc height), end-plate defects,
agement strategies for patients with continued pain after disc and obesity (BMI >30) (Ray 2002).
arthroplasty surgery (Kostuik 2004).
14.9.2 Lumbar Dynamic Posterior Stabilization/

14.9 Other Spinal Motion-Preserving Devices Facet Replacement
14.9.1 Nuclear Replacements Other motion-preserving devices for the lumbar spine include
posterior dynamic devices and facet replacements. The only
There have been continued efforts to develop a prosthetic approved interspinous device is the X-stop (Kyphon,
nuclear replacement device (Bono and Garfin 2004). In Sunnyvale, CA), achieving FDA approval in 2005 (Zucherman
essence, the previously discussed Fernstrom ball is the pre- et al. 2005). The device is made of titanium and PEEK com-
decessor for modern nuclear replacement devices. It differs ponents with side wings surrounding the lateral aspects of
from the current total disc arthroplasty system in that it the spinous processes. It is inserted between the spinous pro-
lacks an end-plate component (Bono and Garfin 2004). The cesses of a spinal segment and holds the spine in a position
prosthesis replaces the nucleus pulposus and serves to of slight flexion to decompress the spinal cord or spinal nerve
restore load transfer through the spinal segment, particu- roots. The rationale for the use of the interspinous device for
larly the annulus fibrosus and facet joints. Therefore, the spinal stenosis is reasonable; however, their role in the treat-
prerequisite for nuclear replacement is an intact or mini- ment of the degenerative disc remains unknown (Anderson
mally disrupted annulus structure and vertebral end plates et al. 2006).
(Lee and Goel 2004). There are currently four nucleus pros- The role of facet replacement continues to grow, and it
thesis designs which aim to reproduce the biomechanical has been proposed as an adjunct total disc arthroplasty: as a
environment of the intact intervertebral disc. The first method of reconstruction after laminectomy and for treat-
design is an impermeable balloon or bladder, filled with ment of facet joint pain. As previously mentioned, as disc
gas, fluid, gel, oil, or a soft polymer. The second is a solid degeneration progresses, facet loads increase significantly
body such as a metal ball or spacer that is placed in the (Yang and King 1984) and may lead to additional pain due to
nuclear cavity (Fernstrom 1966). The third approach is to facet arthrosis. In fact, as continued motion across an arthritic
implant a dehydrated or partially hydrated hydrophilic and painful facet joint is thought to be a cause of arthroplasty
polymer material into a permeable cavity or fibrous jacket, failure, a common contraindication for total disc arthroplasty
which becomes hydrated within the nuclear cavity (Ray is facet arthrosis (Wong et al. 2007). To date, there are no
2002). The final approach is nucleus augmentation, which FDA-approved devices, and a substantial amount of clinical
involves injection of a biomaterial into the nuclear cavity data will be needed before facet replacements can be consid-
for in situ polymerization. ered as a viable treatment option.
Nuclear replacements remain in the experimental phase,
mostly evaluated in vitro and by animal implant studies.
Limited clinical trials in humans have reported clinical 14.10 Summary of Critical Concepts
improvement and restoration of disc function. Other pre- Discussed in the Chapter
liminary studies indicate problems with migration, extru-
sion, vertebral end-plate changes, or subsidence (Bertagnoli Despite the enthusiasm and excitement surrounding total
and Schonmayr 2002; Klara and Ray 2002; Ray 2002; Lee disc arthroplasty, the long-term clinical results, durability,
and Goel 2004; Bertagnoli et al. 2005b; Ahrens et al. 2009). and complications remain unknown.
Design criteria which may predispose to these complications No studies have conclusively demonstrated any difference
include a small contact surface area at the interface between in patient outcomes when compared to fusion in both the
the nucleus replacement device and the vertebral end plates, lumbar and cervical spine.
causing abnormal stress concentration and subsidence. Also, Short-term clinical results through FDA IDE trials appear
abnormal movement of the implanted prosthesis within the to be promising, with acceptable complication rates, and
disc during motion could cause harmful effects on the annu- all demonstrating non-inferior results compared to fusion.
lus. To overcome this problem, some devices allow fit and/ However, many questions remain regarding the true place
or interlocking at the interface (Lee and Goel 2004). At this of total disc arthroplasty in spine surgery, and only well-
time, it is unclear which patients would benefit from total disc performed, prospective studies with long-term outcomes
arthroplasty versus nuclear replacement. Contraindications data and cost analysis will provide the answer.
The indications and role of other motion-preserving from a prospective randomized controlled clinical trial. J Neurosurg
devices such as posterior stabilization and facet replace- Spine 13(3):308318
Buttner-Janz K, Hahn S et al (2002) Basic principles of successful
ments remain unproven. implantation of the SB Charit model LINK intervertebral disk
Efforts must continue to understand the origins of pain endoprosthesis. Orthopade 31(5):441453
within the lumbar and cervical spine, as surgical treat- Cardoso MJ, Rosner MK (2010) Multilevel cervical arthroplasty with
ment of neck or back pain in the absence of radiculopathy artificial disc replacement. Neurosurg Focus 28(5):E19
Cardoso MJ, Mendelsohn A et al (2011) Cervical hybrid arthroplasty
or myelopathy has demonstrated inconsistent results with with 2 unique fusion techniques. J Neurosurg Spine 15(1):4854
the potential for failure. Carragee EJ, Alamin TF (2001) Discography. A review. Spine J 1(5):
364372
Chandler FA (1929) Spinal fusion operations in the treatment of low
back and sciatic pain. JAMA 93:1447
References Chang UK, Kim DH et al (2007) Changes in adjacent-level disc pres-
sure and facet joint force after cervical arthroplasty compared with
Ahrens M, Tsantrizos A et al (2009) Nucleus replacement with the cervical discectomy and fusion. J Neurosurg Spine 7(1):3339
DASCOR disc arthroplasty device: interim two-year efficacy and Cinotti G, David T et al (1996) Results of disc prosthesis after a mini-
safety results from two prospective, non-randomized multicenter mum follow-up period of 2 years. Spine (Phila Pa 1976) 21(8):
European studies. Spine (Phila Pa 1976) 34(13):13761384 9951000
Akamaru T, Kawahara N et al (2003) Adjacent segment motion after a Cunningham BW (2004) Basic scientific considerations in total disc
simulated lumbar fusion in different sagittal alignments: a biome- arthroplasty. Spine J 4(6 Suppl):219S230S
chanical analysis. Spine (Phila Pa 1976) 28(14):15601566 Cunningham BW, Lowery GL et al (2002) Total disc replacement
Albee FH (1911) Transplantation of a portion of the tibia into the spine arthroplasty using the AcroFlex lumbar disc: a non-human primate
for Potts disease. JAMA 57:885 model. Eur Spine J 11(Suppl 2):S115S123
Albert TJ, Eichenbaum MD (2004) Goals of cervical disc replacement. Cunningham BW, Dmitriev AE et al (2003a) General principles of total
Spine J 4(6 Suppl):292S293S disc replacement arthroplasty: seventeen cases in a nonhuman pri-
Anderson PA, Rouleau JP et al (2003) Wear analysis of the Bryan mate model. Spine (Phila Pa 1976) 28(20):S118S124
Cervical Disc prosthesis. Spine (Phila Pa 1976) 28(20):S186S194 Cunningham BW, Gordon JD et al (2003b) Biomechanical evaluation
Anderson PA, Sasso RC et al (2004) The Bryan Cervical Disc: wear of total disc replacement arthroplasty: an in vitro human cadaveric
properties and early clinical results. Spine J 4(6 Suppl):303S309S model. Spine (Phila Pa 1976) 28(20):S110S117
Anderson PA, Tribus CB et al (2006) Treatment of neurogenic claudica- Dahl MC, Rouleau JP et al (2006) Dynamic characteristics of the intact,
tion by interspinous decompression: application of the X STOP fused, and prosthetic-replaced cervical disk. J Biomech Eng 128(6):
device in patients with lumbar degenerative spondylolisthesis. 809814
J Neurosurg Spine 4(6):463471 Dahl MC, Jacobsen S, Metcalf N, Sasso R, Ching RP (2011) A com-
Arens S, Schlegel U et al (1996) Influence of materials for fixation parison of the shock-absorbing properties of cervical disc prosthesis
implants on local infection. An experimental study of steel versus bearing materials. SAS J 5:4854
titanium DCP in rabbits. J Bone Joint Surg Br 78(4):647651 Davis H (1994) Increasing rates of cervical and lumbar spine surgery
Barrick WT, Schofferman JA et al (2000) Anterior lumbar fusion in the United States, 19791990. Spine (Phila Pa 1976) 19(10):
improves discogenic pain at levels of prior posterolateral fusion. 11171123; discussion 11231114
Spine (Phila Pa 1976) 25(7):853857 Delamarter RB, Fribourg DM et al (2003) ProDisc artificial total lum-
Bernthal NM, Celestre PC et al (2012) Disappointing short-term results bar disc replacement: introduction and early results from the United
with the DePuy ASR XL metal-on-metal total hip arthroplasty. States clinical trial. Spine (Phila Pa 1976) 28(20):S167S175
J Arthroplasty 27:539544 Delamarter R, Zigler JE et al (2011) Prospective, randomized, multi-
Bertagnoli R, Schonmayr R (2002) Surgical and clinical results with the center Food and Drug Administration investigational device exemp-
PDN prosthetic disc-nucleus device. Eur Spine J 11(Suppl 2): tion study of the ProDisc-L total disc replacement compared with
S143S148 circumferential arthrodesis for the treatment of two-level lumbar
Bertagnoli R, Sabatino CT et al (2005a) Mechanical testing of a novel degenerative disc disease: results at twenty-four months. J Bone
hydrogel nucleus replacement implant. Spine J 5(6):672681 Joint Surg Am 93(8):705715
Bertagnoli R, Zigler J et al (2005b) Complications and strategies for DiAngelo DJ, Foley KT et al (2004) In vitro biomechanics of cervical
revision surgery in total disc replacement. Orthop Clin North Am disc arthroplasty with the ProDisc-C total disc implant. Neurosurg
36(3):389395 Focus 17(3):E7
Bisseling P, Zeilstra DJ et al (2011) Metal ion levels in patients with a Dmitriev AE, Cunningham BW et al (2005) Adjacent level intradiscal
lumbar metal-on-metal total disc replacement: should we be con- pressure and segmental kinematics following a cervical total disc
cerned? J Bone Joint Surg Br 93(7):949954 arthroplasty: an in vitro human cadaveric model. Spine (Phila Pa
Bohlman HH, Emery SE et al (1993) Robinson anterior cervical discec- 1976) 30(10):11651172
tomy and arthrodesis for cervical radiculopathy. Long-term follow- Eck JC, Humphreys SC et al (1999) Adjacent-segment degeneration
up of one hundred and twenty-two patients. J Bone Joint Surg Am after lumbar fusion: a review of clinical, biomechanical, and radio-
75(9):12981307 logic studies. Am J Orthop (Belle Mead NJ) 28(6):336340
Bono CM, Garfin SR (2004) History and evolution of disc replacement. Fassio B, Ginestie JF (1978) Discal prosthesis made of silicone: experi-
Spine J 4(6 Suppl):145S150S mental study and 1st clinical cases. Nouv Presse Med 7(3):207
Bono CM, Lee CK (2004) Critical analysis of trends in fusion for Fernstrom U (1966) Arthroplasty with intercorporal endoprosthesis in her-
degenerative disc disease over the past 20 years: influence of tech- niated disc and in painful disc. Acta Chir Scand Suppl 357:154159
nique on fusion rate and clinical outcome. Spine (Phila Pa 1976) Fritzell P, Hagg O et al (2003) Complications in lumbar fusion surgery
29(4):455463; discussion Z455 for chronic low back pain: comparison of three surgical techniques
Burkus JK, Haid RW et al (2010) Long-term clinical and radiographic used in a prospective randomized study. A report from the Swedish
outcomes of cervical disc replacement with the Prestige disc: results Lumbar Spine Study Group. Eur Spine J 12(2):178189
Frymoyer JW, Hanley E et al (1978) Disc excision and spine fusion in Kleeman TJ, Ahn UM et al (2001) Laparoscopic anterior lumbar
the management of lumbar disc disease. A minimum ten-year interbody fusion with rhBMP-2: a prospective study of clinical and
followup. Spine (Phila Pa 1976) 3(1):16 radiographic outcomes. Spine (Phila Pa 1976) 26(24):27512756
Gao SG, Lei GH et al (2011) Biomechanical comparison of lumbar Kostuik JP (1997) Intervertebral disc replacement. Experimental study.
total disc arthroplasty, discectomy, and fusion: effect on adjacent- Clin Orthop Relat Res (337):2741
level disc pressure and facet joint force. J Neurosurg Spine 15(5): Kostuik JP (2004) Complications and surgical revision for failed disc
507514 arthroplasty. Spine J 4(6 Suppl):289S291S
Garino JP (2000) Modern ceramic-on-ceramic total hip systems in the Kulkarni SS, Lowery GL et al (2003) Arterial complications following
United States: early results. Clin Orthop Relat Res (379):4147 anterior lumbar interbody fusion: report of eight cases. Eur Spine J
Garrido BJ, Taha TA et al (2010) Clinical outcomes of Bryan cervical 12(1):4854
disc arthroplasty a prospective, randomized, controlled, single site Kumar MN, Jacquot F et al (2001) Long-term follow-up of functional
trial with 48-month follow-up. J Spinal Disord Tech 23(6):367371 outcomes and radiographic changes at adjacent levels following
Gertzbein SD, Seligman J et al (1986) Centrode characteristics of the lumbar spine fusion for degenerative disc disease. Eur Spine J
lumbar spine as a function of segmental instability. Clin Orthop 10(4):309313
Relat Res (208):4851 Kurtz SM, Muratoglu OK et al (1999a) Advances in the processing,
Ghiselli G, Wang JC et al (2004) Adjacent segment degeneration in the sterilization, and crosslinking of ultra-high molecular weight
lumbar spine. J Bone Joint Surg Am 86-A(7):14971503 polyethylene for total joint arthroplasty. Biomaterials 20(18):
Gillet P (2003) The fate of the adjacent motion segments after lumbar 16591688
fusion. J Spinal Disord Tech 16(4):338345 Kurtz SM, Pruitt LA et al (1999b) Radiation and chemical crosslinking
Goldsmith AA, Dowson D et al (2000) A comparative joint simulator promote strain hardening behavior and molecular alignment in ultra
study of the wear of metal-on-metal and alternative material combi- high molecular weight polyethylene during multi-axial loading con-
nations in hip replacements. Proc Inst Mech Eng H 214(1):3947 ditions. Biomaterials 20(16):14491462
Golish SR, Anderson PA (2012) Bearing surfaces for total disc Langton DJ, Jameson SS et al (2011) Accelerating failure rate of the
arthroplasty: metal-on-metal versus metal-on-polyethylene and ASR total hip replacement. J Bone Joint Surg Br 93(8):10111016
other biomaterials. Spine J 12:693701 Lee CK (1988) Accelerated degeneration of the segment adjacent to a
Gore DR, Sepic SB (1984) Anterior cervical fusion for degenerated or lumbar fusion. Spine (Phila Pa 1976) 13(3):375377
protruded discs. A review of one hundred forty-six patients. Spine Lee CK, Goel VK (2004) Artificial disc prosthesis: design concepts and
(Phila Pa 1976) 9(7):667671 criteria. Spine J 4(6 Suppl):209S218S
Guyer RD, McAfee PC et al (2009) Prospective, randomized, Lee CK, Langrana NA (1984) Lumbosacral spinal fusion. A biome-
multicenter Food and Drug Administration investigational device chanical study. Spine (Phila Pa 1976) 9(6):574581
exemption study of lumbar total disc replacement with the CHARITE Lee CK, Langrana NA (2004) A review of spinal fusion for degenera-
artificial disc versus lumbar fusion: five-year follow-up. Spine J tive disc disease: need for alternative treatment approach of disc
9(5):374386 arthroplasty? Spine J 4(6 Suppl):173S176S
Hallab N, Link HD et al (2003a) Biomaterial optimization in total disc Lehmann TR, Spratt KF et al (1987) Long-term follow-up of lower
arthroplasty. Spine (Phila Pa 1976) 28(20):S139S152 lumbar fusion patients. Spine (Phila Pa 1976) 12(2):97104
Hallab NJ, Cunningham BW et al (2003b) Spinal implant debris- LeHuec JC, Kiaer T et al (2003) Shock absorption in lumbar disc pros-
induced osteolysis. Spine (Phila Pa 1976) 28(20):S125S138 thesis: a preliminary mechanical study. J Spinal Disord Tech 16(4):
Harvey EJ, Bobyn JD et al (1999) Effect of flexibility of the femoral 346351
stem on bone-remodeling and fixation of the stem in a canine total Lin EL, Wang JC (2006) Total disk arthroplasty. J Am Acad Orthop
hip arthroplasty model without cement. J Bone Joint Surg Am 81(1): Surg 14(13):705714
93107 Link HD (2002) History, design and biomechanics of the LINK SB
Herkowitz HN, Kurz LT et al (1990) Surgical management of cervical Charit artificial disc. Eur Spine J 11(Suppl 2):S98S105
soft disc herniation. A comparison between the anterior and poste- Luk KD, Lee FB et al (1987) The effect on the lumbosacral spine of
rior approach. Spine (Phila Pa 1976) 15(10):10261030 long spinal fusion for idiopathic scoliosis. A minimum 10-year
Hibbs RA (1911) An operation for progressive spinal deformities. NY follow-up. Spine (Phila Pa 1976) 12(10):9961000
Med J 93:1013 MacWilliam CH, Yood MU et al (1996) Patient-related risk factors that
Hilibrand AS, Robbins M (2004) Adjacent segment degeneration and predict poor outcome after total hip replacement. Health Serv Res
adjacent segment disease: the consequences of spinal fusion? Spine 31(5):623638
J 4(6 Suppl):190S194S Madan SS, Boeree NR (2003) Comparison of instrumented anterior
Hilibrand AS, Carlson GD et al (1999) Radiculopathy and myelopathy interbody fusion with instrumented circumferential lumbar fusion.
at segments adjacent to the site of a previous anterior cervical arthr- Eur Spine J 12(6):567575
odesis. J Bone Joint Surg Am 81(4):519528 Madigan L, Vaccaro AR et al (2009) Management of symptomatic
Huang RC, Girardi FP et al (2003) The implications of constraint in lumbar degenerative disk disease. J Am Acad Orthop Surg 17(2):
lumbar total disc replacement. J Spinal Disord Tech 16(4):412417 102111
Ishihara H, Kanamori M et al (2004) Adjacent segment disease after Matsunaga S, Kabayama S et al (1999) Strain on intervertebral discs
anterior cervical interbody fusion. Spine J 4(6):624628 after anterior cervical decompression and fusion. Spine (Phila Pa
Jackson RK, Boston DA et al (1985) Lateral mass fusion. A prospective 1976) 24(7):670675
study of a consecutive series with long-term follow-up. Spine (Phila McAfee PC (2004) The indications for lumbar and cervical disc
Pa 1976) 10(9):828832 replacement. Spine J 4(6 Suppl):177S181S
Kienapfel H, Sprey C et al (1999) Implant fixation by bone ingrowth. McAfee PC, Cunningham B et al (2005) A prospective, randomized,
J Arthroplasty 14(3):355368 multicenter Food and Drug Administration investigational device
Kim WJ, Lee SH et al (2003) Treatment of juxtafusional degeneration exemption study of lumbar total disc replacement with the CHARITE
with artificial disc replacement (ADR): preliminary results of an artificial disc versus lumbar fusion: part II: evaluation of radio-
ongoing prospective study. J Spinal Disord Tech 16(4):390397 graphic outcomes and correlation of surgical technique accuracy
Klara PM, Ray CD (2002) Artificial nucleus replacement: clinical with clinical outcomes. Spine (Phila Pa 1976) 30(14):15761583;
experience. Spine (Phila Pa 1976) 27(12):13741377 discussion E1388E1590
McLaughlin JR, Lee KR (2000) Total hip arthroplasty in young patients. Schulte KR, Callaghan JJ et al (1993) The outcome of Charnley total
8- to 13-year results using an uncemented stem. Clin Orthop Relat hip arthroplasty with cement after a minimum twenty-year follow-
Res (373):153163 up. The results of one surgeon. J Bone Joint Surg Am 75(7):
Mummaneni PV, Burkus JK et al (2007) Clinical and radiographic 961975
analysis of cervical disc arthroplasty compared with allograft fusion: Serhan H, Mhatre D, Defossez H, Bono CM (2011) Motion-preserving
a randomized controlled clinical trial. J Neurosurg Spine 6(3): technologies for degenerative lumbar spine: the past, present, and
198209 future horizons. SAS J 5:7589
Murrey D, Janssen M et al (2009) Results of the prospective, random- Shono Y, Kaneda K et al (1998) Stability of posterior spinal instrumen-
ized, controlled multicenter Food and Drug Administration investi- tation and its effects on adjacent motion segments in the lumbosacral
gational device exemption study of the ProDisc-C total disc spine. Spine (Phila Pa 1976) 23(14):15501558
replacement versus anterior discectomy and fusion for the treatment Silva M, Shepherd EF et al (2002) Average patient walking activity
of 1-level symptomatic cervical disc disease. Spine J 9(4):275286 approaches 2 million cycles per year: pedometers under-record
Nguyen NQ, Kafle D et al (2011) Ceramic fracture following cervical walking activity. J Arthroplasty 17(6):693697
disc arthroplasty: a case report. J Bone Joint Surg Am 93(22): Sumner DR, Bryan JM et al (1990) Measuring the volume fraction of
e1321e1324 bone ingrowth: a comparison of three techniques. J Orthop Res
OBeirne J, ONeill D et al (1992) Spinal fusion for back pain: a clinical 8(3):448452
and radiological review. J Spinal Disord 5(1):3238 Taksali S, Grauer JN et al (2004) Material considerations for interverte-
Orr RD, Postak PD et al (2007) The current state of cervical and lumbar bral disc replacement implants. Spine J 4(6 Suppl):231S238S
spinal disc arthroplasty. J Bone Joint Surg Am 89(Suppl 3):7075 Traynelis VC (2004) The Prestige cervical disc replacement. Spine J
Patel AA, Brodke DS et al (2008) Revision strategies in lumbar total 4(6 Suppl):310S314S
disc arthroplasty. Spine (Phila Pa 1976) 33(11):12761283 van Ooij A, Oner FC et al (2003) Complications of artificial disc
Penta M, Sandhu A et al (1995) Magnetic resonance imaging assess- replacement: a report of 27 patients with the SB Charit disc.
ment of disc degeneration 10 years after anterior lumbar interbody J Spinal Disord Tech 16(4):369383
fusion. Spine (Phila Pa 1976) 20(6):743747 Wagner M, Wagner H (2000) Medium-term results of a modern metal-
Pidhorz LE, Urban RM et al (1993) A quantitative study of bone and on-metal system in total hip replacement. Clin Orthop Relat Res
soft tissues in cementless porous-coated acetabular components (379):123133
retrieved at autopsy. J Arthroplasty 8(2):213225 Wagner WH, Regan JJ et al (2006) Access strategies for revision or
Pitzen T, Kettler A et al (2007) Cervical spine disc prosthesis: radio- explantation of the Charite lumbar artificial disc replacement. J Vasc
graphic, biomechanical and morphological post mortal findings 12 Surg 44(6):12661272
weeks after implantation. A retrieval example. Eur Spine J 16(7): Weinhoffer SL, Guyer RD et al (1995) Intradiscal pressure measure-
10151020 ments above an instrumented fusion. A cadaveric study. Spine (Phila
Polly DW Jr (2003) Adapting innovative motion-preserving technology Pa 1976) 20(5):526531
to spinal surgical practice: what should we expect to happen? Spine Williams JL, Allen MB Jr et al (1968) Late results of cervical discec-
(Phila Pa 1976) 28(20):S104S109 tomy and interbody fusion: some factors influencing the results.
Puttlitz CM, Rousseau MA et al (2004) Intervertebral disc replacement J Bone Joint Surg Am 50(2):277286
maintains cervical spine kinetics. Spine (Phila Pa 1976) 29(24): Williams DH, Greidanus NV et al (2011) Prevalence of pseudotumor in
28092814 asymptomatic patients after metal-on-metal hip arthroplasty. J Bone
Ray CD (2002) The PDN prosthetic disc-nucleus device. Eur Spine J Joint Surg Am 93(23):21642171
11(Suppl 2):S137S142 Wong DA, Annesser B et al (2007) Incidence of contraindications to
Reitman CA, Hipp JA et al (2004) Changes in segmental intervertebral total disc arthroplasty: a retrospective review of 100 consecutive
motion adjacent to cervical arthrodesis: a prospective study. Spine fusion patients with a specific analysis of facet arthrosis. Spine J
(Phila Pa 1976) 29(11):E221E226 7(1):511
Riew KD, Buchowski JM et al (2008) Cervical disc arthroplasty com- Yang KH, King AI (1984) Mechanism of facet load transmission as
pared with arthrodesis for the treatment of myelopathy. J Bone Joint a hypothesis for low-back pain. Spine (Phila Pa 1976) 9(6):
Surg Am 90(11):23542364 557565
Robertson JT, Papadopoulos SM et al (2005) Assessment of adjacent- Yue WM, Brodner W et al (2005) Long-term results after anterior
segment disease in patients treated with cervical fusion or arthro- cervical discectomy and fusion with allograft and plating: a 5- to
plasty: a prospective 2-year study. J Neurosurg Spine 3(6):417423 11-year radiologic and clinical follow-up study. Spine (Phila Pa
Robinson RA, Smith GW (1955) Anterolateral cervical disc removal 1976) 30(19):21382144
and interbody fusion for cervical disc syndrome. Bull Johns Hopkins Zigler J, Delamarter R et al (2007) Results of the prospective,
Hosp 96:223224 randomized, multicenter Food and Drug Administration investiga-
Sandhu HS, Sanchez-Caso LP et al (2000) Association between findings tional device exemption study of the ProDisc-L total disc replace-
of provocative discography and vertebral endplate signal changes as ment versus circumferential fusion for the treatment of 1-level
seen on MRI. J Spinal Disord 13(5):438443 degenerative disc disease. Spine (Phila Pa 1976) 32(11):11551162;
Santos EG, Polly DW Jr et al (2004) Disc arthroplasty: lessons learned discussion 1163
from total joint arthroplasty. Spine J 4(6 Suppl):182S189S Zucherman JF, Hsu KY et al (2005) A multicenter, prospective,
Sasso RC, Smucker JD et al (2007) Clinical outcomes of BRYAN cervi- randomized trial evaluating the X STOP interspinous process
cal disc arthroplasty: a prospective, randomized, controlled, multi- decompression system for the treatment of neurogenic intermittent
center trial with 24-month follow-up. J Spinal Disord Tech 20(7): claudication: two-year follow-up results. Spine (Phila Pa 1976)
481491 30(12):13511358
The Nonsurgical Treatment of Back Pain
15
Isaac L. Moss, Howard S. An, Francis H. Shen, Zemin Li,
Gunnar B.J. Andersson, and Yejia Zhang
The single most striking chaos in the whole field of medicine in a disease that is the most common
Kochs (1925)
Contents 15.1 Etiology of Back Pain

15.1 Etiology of Back Pain .................................................... 247
Each spinal motion segment is a three-joint complex com-
15.2 Epidemiology and Natural History ............................. 248
prised of the intervertebral disc, two facet joints, and a host
15.3 Nonoperative Treatment ............................................... 248 of ligamentous and muscular attachments. This complex
15.3.1 Education, Activity Modification, Behavioral allows for multiaxial movement and loading of the spine
Therapy, and Exercise Therapy ....................................... 248
15.3.2 Medications ..................................................................... 250 while maintaining an upright posture and protection of the
15.3.3 Injection Therapy ............................................................ 254 neural elements. The anatomic complexity of this articula-
15.3.4 Physical Modalities ......................................................... 255 tion has complicated the search for a specific pathoanatomic
15.4 Future Directions .......................................................... 256 cause for low back pain. The most common etiologies can be
broadly categorized as neural, muscular, osseous, disc-
in the Chapter ................................................................ 256 related, and facet-related and have been the subject of much
debate (Lutz et al. 2003). In the early decades of the twenti-
References .................................................................................... 257
eth century, nerve dysfunction, including neuritis and neural-
gia, as well as muscular dysfunction were proposed as the
leading causes of back pain. As the use of radiographs
became more widespread, the classic bony changes associ-
ated with disc degeneration and spondylosis were found in
many patients complaining of low back pain; hence, osseous
I.L. Moss, MD
Comprehensive Spine Center, etiologies became a popular theory. Initially, an inflammatory
New England Musculoskeletal Institute, etiology for these bony changes was assumed; however,
University of Connecticut Health Center, when no consistent marker of inflammation was found, it
263 Farmington Avenue, Farmington, CT 06030-5353, USA
was then seen as a degenerative disorder. In the late 1930s
e-mail: imoss@uchc.edu
and early 1940s, intervertebral disc pathology began to be
H.S. An, MD G.B.J. Andersson, MD, PhD
recognized as a major low back pain generator (Barr 1938;
Y. Zhang, MD, PhD (*)
Department of Orthopaedic Surgery, Key 1945) and became the dominant theory for several
Rush University Medical Center, decades. This resulted in an increase in the number and vari-
1735 W. Harrison St., Chicago, IL 60612, USA ety of surgical procedures directed at the intervertebral disc.
e-mail: howard.an@rushortho.com;
By the end of the twentieth century, evidence emerged that
gunnar_andersson@rush.edu; yejia_zhang@rush.edu
anatomic abnormalities, as visualized by diagnostic imaging
F.H. Shen, MD
tests, often did not correlate with clinical symptoms (Boden
University of Virginia, et al. 1990). As a result, most modern treatments for low
Charlottesville, VA, USA back pain are pragmatic in approach rather than searching
e-mail: fhs2g@hscmail.mcc.virginia.edu for a specific anatomic directed cause (Lutz et al. 2003), with
Z. Li, MD nonsurgical intervention taking center stage.
Department of Spine Surgery, However, several indications for prompt surgery exist.
The First Affiliated Hospital of Sun Yat-sen University,
While they only account for a small percentage of all patients
58 Zhongshan 2nd Road, #74, Guangzhou
510080, Peoples Republic of China receiving surgery, unless they are diagnosed and treated, they
e-mail: sdlizemin@gmail.com can lead to significant morbidity or mortality. These

248 I.L. Moss et al.
conditions are flagged from the patients history and physical will return within 12 weeks (Andersson et al. 1983).
examination and require further investigation. The presence Unfortunately, recurrence of back pain is common affecting
of neurologic deficits, including saddle anesthesia, urinary 2072 % of individuals (Andersson 1999).
retention or incontinence, and progressive lower extremity Chronic low back pain has been variably defined as back
weakness, should prompt rapid evaluation for cauda equina pain which lasts longer than 3 months, reoccurs frequently,
syndrome. A history of significant trauma in a young indi- or lasts longer than the expected healing time for this type of
vidual or minor trauma in an individual with osteoporosis malady. The course of chronic back pain can be variable and
should prompt the clinician to rule out a fracture of the spinal unpredictable (Von Korff and Saunders 1996). In many cases,
column. Nonmechanical back pain, which wakes patients mild pain may persist for long periods of time with little
from their sleep, can be associated with spinal neoplasm or impact on overall function. Therefore, when evaluating treat-
infection, especially if associated with constitutional symp- ment strategies, activity limitation may be a better measure
toms, history of cancer, or disseminated infection. of outcome than level or presence of pain.
It is important to recognize nonorganic etiologies of low
back pain as treatment can be difficult, requiring specialized
expertise (Waddell et al. 1980). Symptoms include pain, 15.3 Nonoperative Treatment
numbness, or weakness in a nonanatomic distribution.
Common physical examination findings are nonspecific and Despite the extremely high prevalence of back pain in soci-
include widespread tenderness, lumbar pain with axial load- ety, a streamlined and successful treatment strategy still
ing, improvement in straight leg raise with distraction, and eludes physicians. While rates of operative intervention have
inconsistent motor or sensory exam. Evaluation of the psy- steadily and dramatically increased over the past 20 years
chosocial context of the pain with the rehabilitation team that (Deyo et al. 2009), treatment outcomes for back pain without
includes a psychologist and a social worker is often radicular symptoms remain unpredictable with success rates
necessary. in the 5070 % range depending on the measures used (Chou
et al. 2009b; Mirza and Deyo 2007). Given the limited
benefit, high cost, and significant risk, surgery is reserved as
15.2 Epidemiology and Natural History a last resort in a carefully selected population. Nonsurgical
measure is, therefore, the mainstay of treatment for both
Back pain is an extremely common public health problem acute and chronic back pain (Table 15.1).
with significant social and economic ramifications. It is esti-
mated that 7085 % of people in developed countries will
experience back pain at some point in their life (Andersson 15.3.1 Education, Activity Modication,
1999). In a US national survey, 26.4 % of the population had Behavioral Therapy, and Exercise Therapy
back pain lasting at least 1 day in the past 3 months (Deyo
et al. 2006). Prevalence was found to be highest in Native 15.3.1.1 Patient Education
Americans and Alaska Natives and lowest in Asian Patient education, including information about correct spine
Americans. Back pain was more common in adults over the biomechanics during regular activity and posture, and simple
age of 45 and was slightly more prevalent in women as com- methods for reducing symptom are key elements in the man-
pared to men. The annual prevalence of significant back pain agement of both acute and chronic back pain. Ensuring that
is approximately 15 % (Andersson 1999). These statistics patients understand the favorable natural history of the disor-
have remained relatively stable over the past 30 years. The der can empower them to take an active role in the treatment.
health-care resource utilization and cost of care for patients As past incidences are the strongest predictor for future epi-
with back pain are extraordinarily high. Back pain is respon- sodes of back pain, the importance of a lifelong commitment
sible for approximately 1215 % of all physician visits (Deyo to active treatment must be conveyed to the symptomatic
et al. 2006). In 2005, the treatment of back and neck disor- individuals. Several studies have demonstrated that brief
ders accounted for $86 billion in health-care expenditure education can be more effective than conventional care on
(Martin et al. 2008). It affects approximately 2 % of the reducing sick leave and disability (Brox et al. 2008).
workforce, and back pain is the largest single cause of More formal educational interventions have been the sub-
absence from work (12.5 % of all sick days) (Andersson ject of investigation since the introduction of the Swedish
1999). For more details of the epidemiology of low back back school in 1980 (Forssell 1980). The original program
pain, see Chaps. 9 and 16. was designed to teach patients how to protect the spine dur-
Most acute back pain is self-limited, with the majority of ing daily activities and involved four educational sessions on
patients recovering quickly with no residual loss of function. spine anatomy, biomechanics, ergonomics, optimal posture,
Of individuals with symptoms severe enough to miss work, and back exercises in a group setting. The specific format
6070 % will return to work within 6 weeks and 8090 % and content of back schools have varied over the years;
15 The Nonsurgical Treatment of Back Pain 249
Table 15.1 Nonsurgical management options for back pain and rec- There is, however, strong evidence against the historically
ommended stages for interventions based on currently available common recommendation for bed rest as a treatment for
evidence
acute low back pain, which can negatively affect outcomes
Treatment Therapy Stage of symptoms (Atlas and Volinn 1997; Waddell et al. 1997). Results of
Nonsurgical Patient education Acute investigations support the view that after the acute symptoms
therapies Activity modification Acute subside, timely return to modified activities avoids the dele-
Behavioral therapy Acute and chronic
terious effects associated with prolonged immobilization and
Exercise therapy Subacute and chronic
bed rest. Patients who continue moderate levels of activities
Oral medication Non-narcotic Acute
analgesic during episodes of acute back pain generally have a more
Narcotic analgesic Acute rapid recovery, quicker return to work, and a lower risk of
Nonsteroidal Acute and chronic chronic disability (Mkel et al. 2011; Waddell et al. 1997).
anti-inflammatory
drugs 15.3.1.3 Behavioral Therapy
Muscle relaxants Acute The concept of fear avoidance was introduced as a model of
Oral corticosteroids Acute, only with exaggerated pain perception in 1983 (Lethem et al. 1983)
radiculopathy
and has subsequently been applied to chronic back and mus-
Antidepressants Chronic
Topical treatment Acute and chronic
culoskeletal pain (Vlaeyen and Linton 2000; Waddell et al.
Injection therapy Epidural Acute, only with 1993). Central to this model is the concept that the fear of
corticosteroids radiculopathy increased pain, as a result of a movement or activity, may
Soft tissue Subacute and chronic lead to a phobic state and result in inferior physical perfor-
Facet joint Subacute and chronic mance and increased disability. Conversely, confrontation of
Sacroiliac joint Chronic the fear of pain via exposure to an activity or movement can
Modalities Manipulation Acute and chronic lead overtime to a reduction in fear. The application of this
Traction Unknown concept, by including fear-reducing techniques, in the treat-
Acupuncture Chronic ment of low back pain in the primary care setting has been
Transcutaneous Unknown shown to produce an increase in activity, but not impact
electrical nerve
stimulation return to employment (Von Korff et al. 2005). Furthermore,
Orthoses Not recommended when this type of cognitive intervention is combined with an
exercise program, the effect on back pain, disability, and sick
leave has been shown to be equivalent to that of spinal fusion
however, the general concept has been the subject of several (Brox et al. 2003; Brox et al. 2006).
investigations and systematic reviews (Airaksinen et al.
2006; Cohen et al. 1994; Koes et al. 1994; Maier-Riehle and 15.3.1.4 Exercise Therapy
Hrter 2001; Tveito et al. 2004). In comparison to other Exercise therapy, including trunk or core stabilization, restor-
interventions or no treatment at all, the results of these stud- ing normal lumbosacral motion, and low-impact aerobic
ies conflict with the claims for success of back schools in activity, is among the most commonly prescribed noninva-
reducing pain, improving function, and accelerating return to sive interventions for patients with back pain. Based on the
work. The most recent systematic reviews (Airaksinen et al. current literature, exercise therapy is no more effective for
2006; Brox et al. 2008; Heymans et al. 2005) conclude that pain relief or functional improvement when compared to no
forms of group back education can be effective for short- treatment or other nonoperative interventions for acute low
term improvements in pain and disability, especially as part back pain (Chou et al. 2007b; Hayden et al. 2005). There is,
of a multidisciplinary program. In addition, within an occu- however, a general acceptance that low-impact cardiovascu-
pational setting, back school can have a positive effect on lar and aerobic fitness programs are beneficial in that they
return to work and function. can reduce fatigue, improve mood, and prevent general
deconditioning (Anshel and Russell 1994; Casazza et al.
15.3.1.2 Activity Modication 1998). Trunk stabilization and muscle strengthening exer-
Most patients with acute back pain naturally modify their cises are not tolerated by patients; thus, they are not recom-
activities to avoid exacerbation of symptoms. This subcon- mended during acute episodes of back pain. Low-impact
scious protective mechanism is likely helpful when bounded aerobic exercise can be commenced as early as tolerated,
with appropriate education. It is generally accepted that, often by 2 weeks after the onset of acute low back pain, and
while activity may worsen symptoms, it is unlikely to cause activity can progress in a graded fashion.
physical injury to the spine or the surrounding soft tissues In contrast to the acute situation, for chronic pain, low back
(Indahl et al. 1995). Therefore, the common recommenda- pain exercise therapy has been shown to have a beneficial effect
tion is to limit activity for a short period of time (23 days). (Chou et al. 2007b). When compared to no treatment, usual
care, or other noninvasive treatments, exercise therapy has been

associated with a small but significant improvement in pain Box 15.1 Robin McKenzie, CNZM, O.B.E., FCSP (Hon),
and function (Hayden et al. 2005; Team 2004). Exercise ther- FNZSP (Hon), NZCP (HLM), Dip. MT, Dip.MDT
apy has also been linked with reduced sick leave and a higher
rate of return to work within 1 year of treatment in patients with
subacute (<90 days of sick leave) low back pain who were not
already severely disabled (Oesch et al. 2010). Core strengthen-
ing exercises are routinely recommended to improve perfor-
mance and prevent future injuries (McGill 2010).
15.3.2 Medications
Most patients with both acute and chronic low back pain will
include medication in the management of their condition.
While the oral route of administration is most common,
injections also play a role. The most common classes of oral
drugs include analgesics, nonsteroidal anti-inflammatory
drugs (NSAIDs), muscle relaxants, and antidepressants.
Injection therapy employs the use of corticosteroids and
often local anesthetics. As the choices are numerous, it is Robin McKenzie is a New Zealand-born physical thera-
important to distinguish specific indications, doses, dura- pist that revolutionized the nonoperative treatment of back
tions, and potential side effects. Patient education with regard pain. As is the case with many important advancements in
to use of safe and effective medication so as to avoid depen- science and medicine, McKenzie discovered the central-
dence, particularly when prescribing narcotic analgesics, is ization phenomenon by accident in 1956 when he asked
an important consideration and should be included in the a patient to lie down in a prone position in a treatment
treatment goals. room while finishing up with another patient. The end of
the treatment table had been elevated, forcing the patients
15.3.2.1 Analgesics lumbar spine into a hyperextension, a position previously
Analgesic medications can be divided into narcotic and non- thought to be harmful. When McKenzie returned to the
narcotic. Few patients require narcotics and most have ade- room and saw the patients position, he was surprised to
quate pain relief from over-the-counter analgesics. find that the patients leg pain had completely resolved,
that residual pain was mild and centralized to the middle
Non-narcotic Analgesics of the low back, and that after getting up, the pain relief
Acetaminophen is effective for mild to moderate pain. was sustained. As he worked with subsequent patients,
Although prolonged use of high-dose acetaminophen is con- McKenzie discovered that based on an assessment of a
traindicated and may result in hepatotoxicity, acetaminophen patient in various positions, an effective patient-specific
is generally safe, affordable, and available over the counter exercise treatment program could be developed. This has
thus making it a common choice for most patients with acute come to be known as Mechanical Diagnosis and Therapy
low back pain (Malanga and Nadler 1999). Acetaminophen and is the centerpiece of the McKenzie Method which is
use in patients with chronic low back pain and with known practiced worldwide. Once the appropriate exercises are
liver sensitivities due to disease or alcohol abuse is gener- found, patients are instructed to perform the program on
ally not recommended. In patients with renal impairment, their own and thus can take ownership over their disorder.
acetaminophen is recommended over NSAIDs as the risk of McKenzie spent much of his life disseminating his meth-
renal toxicity due to acetaminophen is low. Acetaminophen ods to practitioners throughout the world and has pub-
does not have any muscle-relaxing or anti-inflammatory lished many articles and books on the subject. He is now
properties. retired and lives in New Zealand.
In patients with more severe pain, the centrally acting
non-opiate analgesic tramadol is an attractive alternative as it
has a more favorable side effect profile and lower potential
for abuse than narcotics. Based on a meta-analysis, which
included 908 patients being treated for chronic lower back and should therefore be used with caution in patients on
pain, tramadol was shown to be superior to placebo in reduc- monoamine oxidase inhibitors. Dosage should be reduced in
ing pain and improving function (Deshpande et al. 2007). patients over the age of 75 years old or in those with renal or
Tramadol inhibits uptake of serotonin and norepinephrine hepatic function impairment.
Box 15.2 Prostaglandins (PG) Biosynthesis released by phospholipase A2. PG biosynthesis has two
The starting molecule for PG biosynthesis is the fatty acid control points. The first control point is the release of the
of the phospholipid phosphatidylinositol. PG and related fatty acid from the phospholipid. The second is prosta-
molecules are eicosanoids, the term derived from eicosa glandin synthase, also known as cyclooxygenase (COX).
meaning twenty, referring to the 20 carbons of the fatty The eicosanoids generally act locally due to their short
acid. Most PGs are synthesized from arachidonic acid, half-lives.
COOH
Arachidonic acid
Cox-1/Cox-2
O
COOH
PGG2
OOH
Cox-1/Cox-2
O
COOH
PGH2
OH
Specific
synthases mPGES1 mPGES2
O cPGES
PGI2 COOH
TXA2 PGE2
Prostaglandins (PG) biosynthesis. Cox PGF2
cyclooxygenase, PG prostaglandin, TX PGD2 HO
thromboxane OH
Box 15.3 Cyclooxygenase (COX) Inhibitors the safety profile of these important drugs, primarily by
The mechanism of action of NSAIDs is based on their limiting gastrointestinal side effects. The use of computer-
ability to block the synthesis of prostaglandin (PG) by aided drug design, where a computer modeling is used to
inhibiting COX, an enzyme responsible for catalyzing the synthesize a drug based on the structure of a particular
conversion of arachidonic acid to prostaglandin, PGH2. In target, was instrumental in the development of this new
1991, Daniel Simmons of Brigham Young University dis- drug class. The introduction of COX-2 inhibitors was met
covered a second isoform of the COX enzymes, now with significant enthusiasm, and they became among the
known as COX-2 (Xie et al. 1991). Research has since most widely prescribed medications. This enthusiasm
clarified that the COX-1 isoform is a constitutive enzyme was tempered when a somewhat controversial study dem-
responsible for the maintenance of renal and gastric onstrated a fourfold increased risk of myocardial infarc-
functions. The COX-2 isoform, however, is an inducible tion in patients taking rofecoxib (Vioxx), one of the more
enzyme which drives the inflammatory process. The clas- commonly prescribed COX-2 inhibitors, when compared
sical NSAIDs (e.g., aspirin, ibuprofen, naproxen) to patients taking naproxen, a classical NSAID. Rofecoxib
nonspecifically inhibit both COX isoforms and are thus was voluntarily removed from the market by Merck, the
associated with side effects, mostly notable a risk of gas- drugs manufacturer, in 2004. Since 2005, no new COX-2
tric injury and bleeding. Therefore, since its discovery, inhibitors have been approved for use in the USA. Other
there has been a push to develop medications which selec- COX-2 inhibitors remain on the market and are still
tively inhibit the COX-2 isoform and potentially increase widely prescribed including celecoxib (Celebrex) and
Chemical structure of common cycloox- O OH

ygenase (COX) inhibitors
OH
O
O
O
O
Aspirin Naproxen
O O
H2N
S S
O O
N
N CF3
O
Celecoxib Rofecoxib
O
H
N O
O + O
O N O
HO O
Acetaminophen Naproxcinod (NO-NSAID)
parecoxib (Dynastat, only available in Europe). Current family of molecules involved in the inflammatory path-
research is exploring new application for this class of way. LOX/COX inhibitors have been shown to be effec-
drugs including prevention or treatment of neuroblas- tive and have a tolerable safety profile in pre-market
toma, colon cancer, and neuropsychiatric disorders (Lau studies (Alvaro-Gracia 2004).
et al. 2007).
A third isozyme, COX-3, was discovered in 2002; it is References
thought to be a splice variant of COX-1. Comparison of Alvaro-Gracia JM (2004) Licofelone clinical update on a
canine COX-3 activity with murine COX-1 and COX-2 novel LOX/COX inhibitor for the treatment of osteoar-
demonstrated that this enzyme is selectively inhibited by thritis. Rheumatology (Oxford) 43(Suppl 1):i21i25
analgesic/antipyretic drugs such as acetaminophen and is Keeble JE, Moore PK (2002) Pharmacology and potential
potently inhibited by some nonsteroidal anti-inflammatory therapeutic applications of nitric oxide-releasing non-
drugs. steroidal anti-inflammatory and related nitric oxide-
COX-inhibiting nitric oxide (NO) donors (aka donating drugs. Br J Pharmacol 137(3):295310
NO-NSAIDs) are another novel class of drugs developed Lau L, Hansford LM, Cheng LS, Hang M, Baruchel S,
to further improve the safety profile of the traditional Kaplan DR, Irwin MS (2007) Cyclooxygenase inhibi-
NSAIDs by taking advantage of some of the known effects tors modulate the p53/HDM2 pathway and enhance
of NO (Wallace et al. 1994). These are produced by chem- chemotherapy-induced apoptosis in neuroblastoma.
ically fusing existing NSAIDs to a nitric oxide-donating Oncogene 26(13):19201931
moiety and are intended to provide the COX-inhibiting Wallace JL, Reuter B, Cicala C, McKnight W, Grisham MB,
benefits in addition to vasorelaxation, and inhibition of Cirino G (1994) Novel nonsteroidal anti-inflammatory
white blood cell adhesion and caspase activity (Keeble drug derivatives with markedly reduced ulcerogenic
and Moore 2002). No NO-NSAIDs have been approved properties in the rat. Gastroenterology 107(1):173179
for use at this time; however, several are in clinical trials. Xie WL, Chipman JG, Robertson DL, Erikson RL,
Lipoxygenase (LOX)/COX inhibitors are a final novel Simmons DL (1991) Expression of a mitogen-respon-
class of NSAIDs. These drugs inhibit not only COX and sive gene encoding prostaglandin synthase is regulated
prostaglandin formation but also inhibition of 5-lipoxyge- by mRNA splicing. Proc Natl Acad Sci U S A
nase (LOX) and prevent formation of leukotrienes, another 88(7):26922696
Narcotic Analgesics Nonetheless, both benzodiazepine (e.g., diazepam) and

Despite ongoing controversies concerning the use of nar- non-benzodiazepine (e.g., cyclobenzaprine, methocarba-
cotic (also known as opioid) analgesics for the treatment of mol) muscle relaxants are often prescribed as part of the
chronic low back pain, there has been a steady rise in pretreatment regimen. Systematic reviews of clinical trials
scription rates of both long- and short-acting varieties of have found that muscle relaxants are superior to placebo for
these medications (Deyo et al. 2011). Concerns with opioid the treatment of acute and chronic low back pain (Chou
analgesics center around the high risk of dependency and the et al. 2007a; Van Tulder et al. 2003). There are no well-
complications related to overdose, which have risen in paral- controlled studies that directly compare the efficacy of
lel to prescription rates (Edlund et al. 2010). The use of nar- muscle relaxants to that of NSAIDs. Muscle relaxants are
cotics to treat chronic low back pain has been associated likely to be most beneficial as an adjunct to pharmacologic
with age, psychiatric and personality disorders, and sub- therapies with other drug classes, specifically NSAIDs and
stance abuse as opposed to severity of the underlying pathol- analgesics (Chou et al. 2007a; Van Tulder et al. 2003).
ogy (Breckenridge and Clark 2003). With regard to efficacy, There is little evidence to aid with the selection of one mus-
there is some evidence to suggest that short-term use of nar- cle relaxant over the others. Carisoprodol was found to be
cotic analgesics may be efficacious; however, long-term use superior to diazepam in one study (Van Tulder et al. 2003).
has little or no benefit in reduction of pain (Martell et al. However, due to issues with dependency and abuse when
2007) or improvement of function (Deshpande et al. 2007). combined with narcotics, the use of carisoprodol is gener-
When chronic narcotics are prescribed, aberrant medication- ally limited. Muscle relaxants have been shown to lead to a
taking behavior can be found in as high as 24 % of patients number of central nervous system adverse effects, includ-
(Martell et al. 2007). Therefore, use of these medications ing somnolence and dizziness. These are increased when
requires regular follow-up to evaluate efficacy, overuse, and used in conjunction with other medications and should be
complications. closely monitored by the treating physician.
15.3.2.2 Nonsteroidal Anti-inammatory 15.3.2.4 Oral Corticosteroids

Drugs (NSAIDs) The systemic administration of corticosteroids can be an
NSAIDs are among the most widely prescribed groups of effective treatment for patients with acute radiculopathy.
medications for the treatment of back pain. Their mecha- However, when back pain is not accompanied by radiculopa-
nism of actions involves inhibition of cyclooxygenase thy, studies have shown no clinically significant benefit over
(COX), an essential enzyme for the synthesis of the pro- placebo (Chou et al. 2007a). Systemic corticosteroids can
inflammatory prostaglandins. Two types of COX enzymes have a significant side effect when administered over the
exist, with COX-1 implicated in the protection of the gastric long term or in high doses in the short term. Thus, this class
and intestinal lining and COX-2 involved in the pathways of medications is only recommended for the treatment of low
that produce pain, fever, and inflammation. The first- back pain with radiculopathy.
generation NSAIDs (e.g., ibuprofen, naproxen)
nonspecifically inhibit both COX enzymes. The second- 15.3.2.5 Antidepressants
generation NSAIDs (e.g., celecoxib, meloxicam) are more There is an intimate association between pain and mood,
selectively directed at COX-2 inhibition. In an effort to especially in chronic pain disorders (Fishbain et al. 1997).
reduce gastrointestinal side effects associated with the first- While the mechanism is unknown, there is some evidence
generation NSAID, several randomized controlled studies that tricyclic antidepressants (TCAs) can effectively allevi-
have demonstrated a significant effect on the improvement ate neuropathic pain, independent of mood or depression
of pain and possibly function for the treatment of both acute status (McQuay et al. 1996). Serotonin reuptake inhibitors
and chronic back pain (Chou et al. 2007a; White et al. 2011). (SSRIs) are less effective than TCAs; however, they are pre-
COX-2-selective NSAIDs are probably as effective as non- scribed for neuropathic pain (Jung et al. 1997). The utility of
selective NSAIDs. Remarkably, there have been few studies antidepressants in the treatment of low back pain is not as
that directly compare the two drugs (Pohjolainen et al. clear. The efficacy of both tricyclic antidepressants and
2000). However, both prescription and over-the-counter selective SSRIs has been the subject of several studies and
NSAIDs are associated with adverse effect, including gas- recent systematic reviews. There is no generally accepted
trointestinal upset, bleeding, and exacerbation of preexist- role for antidepressants in the treatment of acute back pain.
ing renal dysfunction. Patients should therefore be closely For chronic back pain, TCAs have been demonstrated to be
monitored, especially with long-term usage. moderately effective in pain reduction when compared to
placebo (Chou et al. 2007a; Staiger et al. 2003; White et al.
15.3.2.3 Muscle Relaxants 2011). As with neuropathic pain, SSRIs are not as effective
Acute back pain is often accompanied by muscle spasm, as TCAs and offer no benefit over placebo. The effect of
but the association with pain is not well understood. antidepressants on function is not clear (Staiger et al. 2003).
Antidepressants can have significant side effects, including Lee et al. 1998). While specialized training is required for all
drowsiness, dry mouth, dizziness, and constipation. Some epidural injections, caudal injection is the least technically
of these adverse events can be mitigated with low starting demanding and has a lower risk of dural puncture when com-
doses that are slowly titrated up for efficacy. Due to the sed- pared to the other approaches. Transforaminal injections must
ative properties of TCAs, these drugs should be adminis- be directed at specific pathologies, while interlaminar injec-
tered at night and may in fact improve the sleep disturbance tions can result in more broad distribution of medication.
often associated with chronic back pain (Harman et al. Fluoroscopy is routinely used to improve the accuracy of
2002). injection needle placement, although an improvement in
efficacy has not been proven (Chou et al. 2009a).
15.3.2.6 Topical Treatments The efficacy of epidural corticosteroid injections has been
Both topical NSAIDs and local anesthetics in the form of the subject of multiple studies and reviews. In patients with
patches, creams, or gels are used to treat low back pain. back pain accompanied by radiculopathy, there is some,
Topical application of NSAIDs is attractive as it can theoreti- albeit weak, evidence to suggest the epidural corticosteroid
cally reduce the adverse events associated with their systemic injection provides short-term (up to 6 weeks) pain relief
administration. It should be noted, however, that with all (Carette et al. 1997; Karppinen et al. 2001; Ng et al. 2005).
topical formulations, systemic NSAID absorption occurs to There is, however, no evidence that epidural corticosteroid
variable degrees and adverse effects have been documented injections are effective in patients with back pain without
(Zimmerman et al. 1995). There is evidence to suggest that radicular symptoms and they are, therefore, not a recom-
topical NSAIDs are effective for the treatment of musculosk- mended treatment option (Chou et al. 2009a, b). When epi-
eletal pain, although there is no evidence with regard to their dural injection therapy is pursued, a series of two to three
efficacy specifically for acute or chronic back pain injections is often recommended. Generally, no more than
(Haroutiunian et al. 2010). Adhesive local anesthetic lido- three injections are administered over a 612-month period,
caine patches are also often used to treat back pain. Again, and if there is no response to the first injection, there is some
their efficacy is unknown. Thus, topical treatments should be evidence to suggest that further injections within the acute
used with discretion and caution as adjuvants to other treat- period are unlikely to be of significant benefit (Arden et al.
ment modalities. 2005). Complications from epidural injections are rare; how-
ever, dural puncture, epidural hematoma, spinal cord injury,
infection, and nerve damage have been reported (Chou et al.
15.3.3 Injection Therapy 2009a).
Therapeutic injections are often incorporated into back pain 15.3.3.2 Soft Tissue Injections
treatment regimens, particularly after less invasive methods, Injections for back pain outside of the spine, the most com-
including exercise and oral medications, have failed. mon being trigger point injections, are targeted at soft tissue
Injections can be directed at anatomic location both within structures believed to be significant pain generators. These
and around the axial skeleton. Injection therapy should only treatments involve the injection of local anesthetic and/or
be considered when a reasonable etiologic diagnosis has corticosteroid into specific myofascial trigger points, which
been made and should not be used for nonspecific low back are thought to result from irritable foci of taut muscle bands.
pain. Focal pressure on these points should produce a local twitch
response with distally referred pain (Kraus and Fischer
15.3.3.1 Epidural Corticosteroid Injections 1991). This so-called myofascial syndrome generally
and Medial Branch Blocks responds to a regimen of exercise or manual therapy, with
Injection of corticosteroid, often with local anesthetic, is com- injections considered as an adjuvant. The evidence to sup-
monly administered for treatment of acute and chronic spine port local trigger point injection therapy alone is weak,
pathology. The medication can be delivered via an interlami- showing short-term relief for subacute or chronic back pain
nar, caudal, or transforaminal approach, depending on the (Chou et al. 2009a, b). However, some studies showed no
pathoanatomy and specific patient symptoms. The potent difference when compared to placebo, and the addition of
anti-inflammatory effects of the corticosteroids coupled with corticosteroid to local anesthetic does not appear to exert a
the analgesic effects of local anesthetics are thought to inter- significant effect. Therefore, the number of intramuscular
rupt the pain and spasm cycle as well as nociceptive transmis- injections should be limited, as there is concern for develop-
sion. Preclinical experiments suggest that corticosteroids can ment of muscle damage, scar tissue, and altered function
reduce cell membrane permeability, diminish neural peptide after multiple injections.
synthesis and neuronal discharge, and moderate sensitization Injection of botulinum toxin A has been studied to treat
of dorsal horn neurons (Byrd et al. 2000; Devor et al. 1985; chronic low back pain (Foster et al. 2001). Results demonstrated
reduction in pain and patient reported disability in the short of spinal manipulation is unclear, with contradictory
term, with cessation of benefits after 34 months in 60 % of findings in two recent systematic reviews (Assendelft et al.
patients. 2004; Bronfort et al. 2004). There is some evidence to sug-
Prolotherapy utilizes injection of sclerosing agents into gest that spinal manipulation can provide short-term relief
the back and pelvic ligaments. Studies on this controversial of acute low back pain. In cases of chronic low back pain,
intervention have not shown any consistent benefit in the spinal manipulation has an effect comparable to that of
improvement of pain or disability (Chou et al. 2009a, b). NSAIDs and superior to placebo in the short term and an
effect comparable to that of physical therapy in the long
15.3.3.3 Facet Joint Injections term. Efficacy can likely be improved when spinal manipu-
The facet joints are richly innervated synovial articulations lation is combined with other nonoperative treatment
that frequently develop osteoarthritic degenerative changes. modalities including an exercise program, medication, and
These joints are thus implicated as significant pain generator lifestyle changes, but this has not been rigorously
in patients with low back pain, especially when symptoms documented.
are exacerbated by lumbar extension maneuvers. Intra- If patients choose to pursue spinal manipulation therapy,
articular facet injections as well as medial branch blocks and defined treatment goals should be established. If symptoms
ablation are often incorporated into a back pain treatment do abate, there is no evidence to suggest a need for ongoing
strategy. However, despite their common use, there is no maintenance therapy. However, if back pain persists or if
clear evidence as to their efficacy. One systematic review radicular symptoms develop, treatment should be discontin-
found facet joint injections were associated with short-term ued and the patient should be reassessed. The risk of
improvement (Boswell et al. 2007); conversely, other reviews significant injury with manipulation in the alert patient with-
found no benefit when compared to placebo injection espe- out significant spinal stenosis is considered to be low.
cially over the long term (Slipman et al. 2003; Staal et al. However, manipulation under general anesthesia is thought
2009). Thus, facet joint-directed therapy is rarely indicated to carry significant risk of injury (Koes et al. 1996).
in cases of acute back pain. Manipulation is contraindicated in patients with progressive
The literature, with regard to medial branch block and neurologic deficits.
ablations (also known as rhizotomy), is scant, although there
may be some degree of short-term pain relief when treating 15.3.4.2 Traction
chronic back pain (Leclaire et al. 2001; Niemist et al. 2003; Another physical treatment modality often sought by patients
Slipman et al. 2003). for relief of both radicular and non-radicular back pain is
spinal traction. Traction is thought to distract the disc space
15.3.3.4 Sacroiliac Joint Injections and widen the neural foramen. In order to achieve distrac-
The sacroiliac is generally not considered a primary pain tion, the traction force must be sufficient to overcome muscle
generator in patients with low back pain; however, it can be contraction, ligamentous resistance, and friction from the
a common area of referred pain (Fortin et al. 1994). Based on table surface and machinery and is estimated to be 3550 %
history and physical examination, difficulties are experi- of total body weight (Beurskens et al. 1997). Based on sev-
enced in diagnosing sacroiliac joint dysfunction (Dreyfuss eral published reviews of the literature, there is currently no
et al. 1996), and it is thus regarded as a diagnosis of exclu- evidence that traction is an effective treatment for low back
sion after other sources of pain have been ruled out. When pain (Beurskens et al. 1997; Borman et al. 2003; Malanga
dysfunction is evident, diagnostic and therapeutic injections and Nadler 1999; van der Heijden et al. 1995; van Middelkoop
of corticosteroids and local anesthetics can be considered; et al. 2011).
however, there is no convicting evidence with regard to
efficacy (Chou et al. 2009a, b). 15.3.4.3 Acupuncture
Acupuncture is a traditional Chinese medicine technique,
which has been practiced for over 2,000 years and has gained
15.3.4 Physical Modalities popularity in the Western world. Acupuncture is used to
treat a wide variety of ailments and involves insertion and
15.3.4.1 Manipulation manipulation of thin, solid metal needles at specific points
Spinal manipulation is among the most popular non-phar- throughout the body. It has been difficult to produce high-
maceutical alternative treatments for low back pain (Carey quality evidence with regard to the efficacy of acupuncture
et al. 1995). Chiropractors, osteopaths, physical therapists, as the treatment involves close patient-practitioner interac-
and other practitioners perform various types of manual tion, which may have a positive effect in and of itself, and
therapy, which usually include a combination of massage the historical lack of an appropriate sham procedure to serve
and joint mobilization. Evidence with regard to the efficacy as a control (Madsen et al. 2009). There is currently no
strong evidence to support the use of acupuncture for acute capacity as proprioceptive reminders to use correct spine
low back pain. For chronic low back pain, there is evidence mechanics during lifting and bending activities (Reyna et al.
from a systematic review of the literature to suggest that it 1995; Woodhouse et al. 1995).
can provide some degree of pain relief and improvement in
function in the short term (Furlan et al. 2005). In the same
review, it is suggested that acupuncture has a small effect in 15.4 Future Directions
improving outcomes as an adjunct to conventional therapies.
Acupuncture is therefore not generally recommended as a The pathologic changes associated with intervertebral disc
first-line treatment for back pain, but is often considered as degeneration are well described. With aging, there is a reduc-
a part of a comprehensive chronic pain management tion in disc cell concentration and an imbalance in extracel-
program. lular matrix homeostasis resulting in molecular and
biomechanical alteration in intervertebral disc structure and
15.3.4.4 Transcutaneous Electrical function. The precise mechanism linking these changes to
symptomatic back pain remains elusive. Present research
Nerve Stimulation
Transcutaneous electrical nerve stimulation therapy hopes to better delineate the link between pathology and
applies electrical stimulation to the skin in an effort to symptoms, thus leading to more accurate diagnoses and
achieve pain relief. The common high-frequency (>50 Hz) treatments. The eventual goal of this research is to allow for
low-intensity stimulation results in sensory stimulation a paradigm shift to more biological and mechanistic treat-
without motor contraction and is thought to disturb neural ment strategies for discogenic pain that are less invasive, yet
pathway conduction and thus modulate pain. There is no more effective than current regimens. Biological treatment
convincing evidence that transcutaneous electrical nerve strategies involve the application of three principal compo-
stimulation therapy provides significant relief of acute or nents: application of therapeutic molecules, delivery of cells
chronic low back pain; however, there is a lack of high- to repopulate the disc, and supplementation of the matrix
quality investigations on the subject (van Middelkoop (Yoon 2005).
et al. 2011).
15.3.4.5 Other Modalities in the Chapter
Cold packs, superficial heat, short-wave diathermy, massage,
and ultrasound are often part of physical therapy and chiro-
The most common etiologies can be broadly categorized
practic treatment. Further, many patients use cold or heat to
as neural, muscular, osseous, facet related, and disc related
relieve symptoms. The choice between the two depends on
and have been the subject of much debate.
the stage of injury. Cold provides pain relief and reduces the
Back pain is an extremely common public health prob-
inflammatory response following an acute injury by vaso-
lem, with a relatively benign natural history.
constriction. Heat relaxes muscles and improves tolerance to
After a thorough physical examination and imaging stud-
exercise, and may be a reasonable modality when the acute
ies, education, activity modification, behavioral therapy,
phase is over. Apart from the short-term relief, there is no
and exercise therapy may help the patients recover.
documented value to the use of these modalities. Medications include analgesics, anti-inflammatories, mus-
Magnets have been used for centuries to cure a variety cle relaxants, antidepressants, and topical remedies.
of ailments, including back pain. Magnets sold for pain are
Steroids injected into the epidural space, facet joints, soft
weak and have no effect on circulation or tissue temperature.
tissues, and sacroiliac joints may help to alleviate symp-
Controlled trials have found no benefit of magnet therapy for
toms and promote recovery.
chronic low back pain (Collacott et al. 2000).
Physical modalities that are commonly used such as
manipulations, traction, orthoses, acupuncture, and trans-
15.3.4.6 Orthoses cutaneous electrical nerve stimulation (TENS) have lim-
There is no evidence to support the effectiveness of orthoses
ited value in treating back pain.
in the treatment of acute and chronic low back pain (Million
Investments in research may result in a new generation of
et al. 1981). There is some, but quite weak, evidence that
biological nonoperative therapies to provide reliable, safe,
they may decrease absence in the workplace (van Poppel
and efficacious relief of spinal pain.
et al. 1997, 1998; Walsh and Schwartz 1990). The mecha-
nism of action of orthoses is debated because they do not
Acknowledgment Yejia Zhang, MD, PhD, is supported by the Eunice
improve lumbosacral biomechanics or enhance dynamic Kennedy Shriver National Institute of Child Health and Human
lifting capacity. Their effectiveness may be attributed to their Development (NICHD, 1K08 HD049598-01).
References in spinal nerve roots induced by epidural nucleus pulposus applica-

tion. J Orthop Res 18(6):983987
Carette S, Leclaire R, Marcoux S, Morin F, Blaise GA, St-Pierre A,
Airaksinen O, Brox J, Cedraschi C, Hildebrandt J, Klaber-Moffett Truchon R, Parent F, Levsque J, Bergeron V, Montminy P,
J, Kovacs F, Mannion A, Reis S, Staal J, Ursin H, Zanoli G Blanchette C (1997) Epidural corticosteroid injections for sciatica
(2006) Chapter 4 European guidelines for the management of due to herniated nucleus pulposus. N Engl J Med 336(23):
chronic nonspecific low back pain. Eur Spine J 15(S2): 16341640
s192s300 Carey TS, Evans A, Hadler N, Kalsbeek W, McLaughlin C, Fryer J
Andersson GB (1999) Epidemiological features of chronic low-back (1995) Care-seeking among individuals with chronic low back pain.
pain. Lancet 354(9178):581585 Spine 20(3):312317
Andersson GB, Svensson HO, Odn A (1983) The intensity of work Casazza BA, Young JL, Herring SA (1998) The role of exercise in the
recovery in low back pain. Spine 8(8):880884 prevention and management of acute low back pain. Occup Med
Anshel MH, Russell KG (1994) Effect of aerobic and strength 13(1):4760
training on pain tolerance, pain appraisal and mood of unfit Chou R, Huffman LH, American Pain Society; American College of
males as a function of pain location. J Sports Sci 12(6): Physicians (2007a) Medications for acute and chronic low back
535547 pain: a review of the evidence for an American Pain Society/
Arden NK, Price C, Reading I, Stubbing J, Hazelgrove J, Dunne C, American College of Physicians clinical practice guideline. Ann
Michel M, Rogers P, Cooper C (2005) A multicentre randomized Intern Med 147(7):505514
controlled trial of epidural corticosteroid injections for sciatica: the Chou R, Huffman LH, American Pain Society; American College of
WEST study. Rheumatology (Oxford) 44(11):13991406 Physicians (2007b) Nonpharmacologic therapies for acute and
Assendelft WJ, Morton SC, Yu EI, Suttorp MJ, Shekelle PG (2004) chronic low back pain: a review of the evidence for an American
Spinal manipulative therapy for low back pain. Cochrane Database Pain Society/American College of Physicians clinical practice
Syst Rev (1):CD000447 guideline. Ann Intern Med 147(7):492504
Atlas SJ, Volinn E (1997) Classics from the spine literature revisited: a Chou R, Atlas SJ, Stanos SP, Rosenquist RW (2009a) Nonsurgical
randomized trial of 2 versus 7 days of recommended bed rest for interventional therapies for low back pain: a review of the evidence
acute low back pain. Spine 22(20):23312337 for an American Pain Society clinical practice guideline. Spine
Barr JS (1938) Intervertebral disk lesions as cause of sciatica. Br Med J 34(10):10781093
2(4067):12471251, 1274-11274-2 Chou R, Loeser JD, Owens DK, Rosenquist RW, Atlas SJ, Baisden J,
Beurskens AJ, de Vet HC, Kke AJ, Regtop W, van der Heijden GJ, Carragee EJ, Grabois M, Murphy DR, Resnick DK, Stanos SP,
Lindeman E, Knipschild PG (1997) Efficacy of traction for Shaffer WO, Wall EM, American Pain Society Low Back Pain
nonspecific low back pain. 12-week and 6-month results of a ran- Guideline Panel (2009b) Interventional therapies, surgery, and inter-
domized clinical trial. Spine 22(23):27562762 disciplinary rehabilitation for low back pain: an evidence-based
Boden SD, Davis DO, Dina TS, Patronas NJ, Wiesel SW (1990) clinical practice guideline from the American Pain Society. Spine
Abnormal magnetic-resonance scans of the lumbar spine in asymp- 34(10):10661077
tomatic subjects. A prospective investigation. J Bone Joint Surg Am Cohen JE, Goel V, Frank JW, Bombardier C, Peloso P, Guillemin F
72(3):403408 (1994) Group education interventions for people with low back
Borman P, Keskin D, Bodur H (2003) The efficacy of lumbar traction in pain. An overview of the literature. Spine 19(11):12141222
the management of patients with low back pain. Rheumatol Int Collacott EA, Zimmerman JT, White DW, Rindone JP (2000) Bipolar
23(2):8286 permanent magnets for the treatment of chronic low back pain: a
Boswell MV, Colson JD, Sehgal N, Dunbar EE, Epter R (2007) A sys- pilot study. JAMA 283(10):13221325
tematic review of therapeutic facet joint interventions in chronic Deshpande A, Furlan A, Mailis-Gagnon A, Atlas S, Turk D (2007)
spinal pain. Pain Physician 10(1):229253 Opioids for chronic low-back pain. Cochrane Database Syst Rev
Breckenridge J, Clark JD (2003) Patient characteristics associated with (3):CD004959
opioid versus nonsteroidal anti-inflammatory drug management of Devor M, Govrin-Lippmann R, Raber P (1985) Corticosteroids sup-
chronic low back pain. J Pain 4(6):344350 press ectopic neural discharge originating in experimental neuro-
Bronfort G, Haas M, Evans RL, Bouter LM (2004) Efficacy of spinal mas. Pain 22(2):127137
manipulation and mobilization for low back pain and neck pain: a Deyo RA, Mirza SK, Martin BI (2006) Back pain prevalence and visit
systematic review and best evidence synthesis. Spine J 4(3): rates: estimates from U.S. national surveys, 2002. Spine
335356 31(23):27242727
Brox JI, Srensen R, Friis A, Nygaard , Indahl A, Keller A, Deyo RA, Mirza SK, Turner JA, Martin BI (2009) Overtreating chronic
Ingebrigtsen T, Eriksen HR, Holm I, Koller AK, Riise R, Reikers back pain: time to back off? J Am Board Fam Med 22(1):6268
O (2003) Randomized clinical trial of lumbar instrumented fusion Deyo RA, Smith DH, Johnson ES, Donovan M, Tillotson CJ, Yang X,
and cognitive intervention and exercises in patients with chronic Petrik AF, Dobscha SK (2011) Opioids for back pain patients: pri-
low back pain and disc degeneration. Spine 28(17):19131921 mary care prescribing patterns and use of services. J Am Board Fam
Brox JI, Reikers O, Nygaard , Srensen R, Indahl A, Holm I, Keller Med 24(6):717727
A, Ingebrigtsen T, Grundnes O, Lange JE, Friis A (2006) Lumbar Dreyfuss P, Michaelsen M, Pauza K, McLarty J, Bogduk N (1996) The
instrumented fusion compared with cognitive intervention and exer- value of medical history and physical examination in diagnosing
cises in patients with chronic back pain after previous surgery for sacroiliac joint pain. Spine 21(22):25942602
disc herniation: a prospective randomized controlled study. Pain Edlund MJ, Martin BC, Fan MY, Devries A, Braden JB, Sullivan MD
122(12):145155 (2010) Risks for opioid abuse and dependence among recipients of
Brox JI, Storheim K, Grotle M, Tveito TH, Indahl A, Eriksen HR (2008) chronic opioid therapy: results from the TROUP study. Drug
Systematic review of back schools, brief education, and fear-avoid- Alcohol Depend 112(12):9098
ance training for chronic low back pain. Spine J 8(6):948958 Fishbain DA, Cutler R, Rosomoff HL, Rosomoff RS (1997) Chronic
Byrd G, Otani K, Brisby H, Rydevik B, Olmarker K (2000) pain-associated depression: antecedent or consequence of chronic
Methylprednisolone reduces the early vascular permeability increase pain? A review. Clin J Pain 13(2):116137
Forssell MZ (1980) The Swedish back school. Physiotherapy Maier-Riehle B, Hrter M (2001) The effects of back schools a meta-
66(4):112114 analysis. Int J Rehabil Res 24(3):199206
Fortin JD, Dwyer AP, West S, Pier J (1994) Sacroiliac joint: pain refer- Mkel KT, Peltola M, Sund R, Malmivaara A, Hkkinen U, Remes V
ral maps upon applying a new injection/arthrography technique. (2011) Regional and hospital variance in performance of total hip
Part I: asymptomatic volunteers. Spine 19(13):14751482 and knee replacements: a national population-based study. Ann
Foster L, Clapp L, Erickson M, Jabbari B (2001) Botulinum toxin A Med 43(Suppl 1):S31S38
and chronic low back pain: a randomized, double-blind study. Malanga GA, Nadler SF (1999) Nonoperative treatment of low back
Neurology 56(10):12901293 pain. Mayo Clin Proc 74(11):11351148
Furlan AD, van Tulder MW, Cherkin DC, Tsukayama H, Lao L, Koes Martell BA, OConnor PG, Kerns RD, Becker WC, Morales KH,
BW, Berman BM (2005) Acupuncture and dry-needling for low Kosten TR, Fiellin DA (2007) Systematic review: opioid treatment
back pain. Cochrane Database Syst Rev (1):CD001351 for chronic back pain: prevalence, efficacy, and association with
Harman K, Pivik RT, DEon JL, Wilson KG, Swenson JR, Matsunaga addiction. Ann Intern Med 146(2):116127
L (2002) Sleep in depressed and nondepressed participants with Martin BI, Deyo RA, Mirza SK, Turner JA, Comstock BA, Hollingworth
chronic low back pain: electroencephalographic and behaviour W, Sullivan SD (2008) Expenditures and health status among adults
findings. Sleep 25(7):775783 with back and neck problems. JAMA 299(6):656664
Haroutiunian S, Drennan DA, Lipman AG (2010) Topical NSAID ther- McGill SM (2010) Core training: evidence translating to better perfor-
apy for musculoskeletal pain. Pain Med 11(4):535549 mance and injury prevention. Strength Cond J 32(3):3346
Hayden JA, van Tulder MW, Malmivaara A, Koes BW (2005) Exercise McQuay HJ, Tramr M, Nye BA, Carroll D, Wiffen PJ, Moore RA
therapy for treatment of non-specific low back pain. Cochrane (1996) A systematic review of antidepressants in neuropathic pain.
Database Syst Rev (3):CD000335 Pain 68(23):217227
Heymans MW, van Tulder MW, Esmail R, Bombardier C, Koes BW Million R, Nilsen KH, Jayson MI, Baker RD (1981) Evaluation of low
(2005) Back schools for nonspecific low back pain: a systematic back pain and assessment of lumbar corsets with and without back
review within the framework of the Cochrane Collaboration Back supports. Ann Rheum Dis 40(5):449454
Review Group. Spine 30(19):21532163 Mirza SK, Deyo RA (2007) Systematic review of randomized trials
Indahl A, Velund L, Reikeraas O (1995) Good prognosis for low back comparing lumbar fusion surgery to nonoperative care for treatment
pain when left untampered. A randomized clinical trial. Spine of chronic back pain. Spine 32(7):816823
20(4):473477 Ng L, Chaudhary N, Sell P (2005) The efficacy of corticosteroids in
Jung AC, Staiger T, Sullivan M (1997) The efficacy of selective sero- periradicular infiltration for chronic radicular pain: a randomized,
tonin reuptake inhibitors for the management of chronic pain. J Gen double-blind, controlled trial. Spine 30(8):857862
Intern Med 12(6):384389 Niemist L, Kalso E, Malmivaara A, Seitsalo S, Hurri H, Cochrane
Karppinen J, Malmivaara A, Kurunlahti M, Kyllnen E, Pienimki T, Collaboration Back Review Group (2003) Radiofrequency denerva-
Nieminen P, Ohinmaa A, Tervonen O, Vanharanta H (2001) tion for neck and back pain: a systematic review within the frame-
Periradicular infiltration for sciatica: a randomized controlled trial. work of the cochrane collaboration back review group. Spine
Spine 26(9):10591067 28(16):18771888
Key JA (1945) Intervertebral disk lesions are the most common cause Oesch P, Kool J, Hagen KB, Bachmann S (2010) Effectiveness of exer-
of low back pain with or without sciatica. Ann Surg 121(4): cise on work disability in patients with non-acute non-specific low
534539 back pain: systematic review and meta-analysis of randomised con-
Kochs J (1925) Uber den objektiven tastbefund bei der myalgia. Dtsch trolled trials. J Rehabil Med 42(3):193205
Med Wochenschr 15:621 Pohjolainen T, Jekunen A, Autio L, Vuorela H (2000) Treatment of
Koes BW, van Tulder MW, van der Windt WM, Bouter LM (1994) The acute low back pain with the COX-2-selective anti-inflammatory
efficacy of back schools: a review of randomized clinical trials. drug nimesulide: results of a randomized, double-blind comparative
J Clin Epidemiol 47(8):851862 trial versus ibuprofen. Spine 25(12):15791585
Koes BW, Assendelft WJ, van der Heijden GJ, Bouter LM (1996) Reyna JR, Leggett SH, Kenney K, Holmes B, Mooney V (1995) The
Spinal manipulation for low back pain. An updated systematic effect of lumbar belts on isolated lumbar muscle. Strength and
review of randomized clinical trials. Spine 21(24):28602871; dis- dynamic capacity. Spine 20(1):6873
cussion 28722873 Slipman CW, Bhat AL, Gilchrist RV, Issac Z, Chou L, Lenrow DA
Kraus H, Fischer AA (1991) Diagnosis and treatment of myofascial (2003) A critical review of the evidence for the use of zygapophy-
pain. Mt Sinai J Med 58(3):235239 sial injections and radiofrequency denervation in the treatment of
Leclaire R, Fortin L, Lambert R, Bergeron YM, Rossignol M (2001) low back pain. Spine J 3(4):310316
Radiofrequency facet joint denervation in the treatment of low back Staal JB, de Bie RA, de Vet HC, Hildebrandt J, Nelemans P (2009)
pain: a placebo-controlled clinical trial to assess efficacy. Spine Injection therapy for subacute and chronic low back pain: an updated
26(13):14111416; discussion 1417 Cochrane review. Spine 34(1):4959
Lee HM, Weinstein JN, Meller ST, Hayashi N, Spratt KF, Gebhart Staiger TO, Gaster B, Sullivan MD, Deyo RA (2003) Systematic review
GF (1998) The role of steroids and their effects on phospholipase of antidepressants in the treatment of chronic low back pain. Spine
A2. An animal model of radiculopathy. Spine 23(11): (Phila Pa 1976) 28(22):25402545
11911196 Tveito TH, Hysing M, Eriksen HR (2004) Low back pain interventions
Lethem J, Slade PD, Troup JD, Bentley G (1983) Outline of a Fear- at the workplace: a systematic literature review. Occup Med (Lond)
Avoidance Model of exaggerated pain perceptionI. Behav Res 54(1):313
Ther 21(4):401408 UK BEAMT Team (2004) United Kingdom back pain exercise and
Lutz GK, Butzlaff M, Schultz-Venrath U (2003) Looking back on back manipulation (UK BEAM) randomised trial: effectiveness of phys-
pain: trial and error of diagnoses in the 20th century. Spine ical treatments for back pain in primary care. BMJ 329(7479):
28(16):18991905 1377
Madsen MV, Gtzsche PC, Hrbjartsson A (2009) Acupuncture treat- van der Heijden GJ, Beurskens AJ, Koes BW, Assendelft WJ, de Vet
ment for pain: systematic review of randomised clinical trials with HC, Bouter LM (1995) The efficacy of traction for back and neck
acupuncture, placebo acupuncture, and no acupuncture groups. pain: a systematic, blinded review of randomized clinical trial meth-
BMJ 338:a3115 ods. Phys Ther 75(2):93104
van Middelkoop M, Rubinstein SM, Kuijpers T, Verhagen AP, Ostelo Waddell G, McCulloch JA, Kummel E, Venner RM (1980) Nonorganic
R, Koes BW, Van Tulder M (2011) A systematic review on the physical signs in low-back pain. Spine 5(2):117125
effectiveness of physical and rehabilitation interventions for chronic Waddell G, Newton M, Henderson I, Somerville D, Main CJ (1993) A
non-specific low back pain. Eur Spine J 20(1):1939 fear-avoidance beliefs questionnaire (FABQ) and the role of fear-
van Poppel MN, Koes BW, Smid T, Bouter LM (1997) A systematic avoidance beliefs in chronic low back pain and disability. Pain
review of controlled clinical trials on the prevention of back pain in 52(2):157168
industry. Occup Environ Med 54(12):841847 Waddell G, Feder G, Lewis M (1997) Systematic reviews of bed rest
van Poppel MN, Koes BW, van der Ploeg T, Smid T, Bouter LM (1998) and advice to stay active for acute low back pain. Br J Gen Pract
Lumbar supports and education for the prevention of low back pain in 47(423):647652
industry: a randomized controlled trial. JAMA 279(22):17891794 Walsh NE, Schwartz RK (1990) The influence of prophylactic orthoses
Van Tulder M, Touray T, Furlan AD, Solway S, Bouter LM, Cochrane on abdominal strength and low back injury in the workplace. Am J
Back Review Group (2003) Muscle relaxants for nonspecific low Phys Med Rehabil 69(5):245250
back pain: a systematic review within the framework of the cochrane White AP, Arnold PM, Norvell DC, Ecker E, Fehlings M (2011)
collaboration. Spine 28(17):19781992 Pharmacologic management of chronic low back pain: synthesis of
Vlaeyen JW, Linton SJ (2000) Fear-avoidance and its consequences in the evidence. Spine 36(21 Suppl):S131S143
chronic musculoskeletal pain: a state of the art. Pain 85(3): Woodhouse ML, McCoy RW, Redondo DR, Shall LM (1995) Effects of
317332 back support on intra-abdominal pressure and lumbar kinetics dur-
Von Korff M, Saunders K (1996) The course of back pain in primary ing heavy lifting. Hum Factors 37(3):582590
care. Spine 21(24):28332837; discussion 28382839 Yoon ST (2005) Molecular therapy of the intervertebral disc. Spine J
Von Korff M, Balderson BH, Saunders K, Miglioretti DL, Lin EH, 5(6 Suppl):280S286S
Berry S, Moore JE, Turner JA (2005) A trial of an activating inter- Zimmerman J, Siguencia J, Tsvang E (1995) Upper gastrointestinal
vention for chronic back pain in primary care and physical therapy hemorrhage associated with cutaneous application of diclofenac
settings. Pain 113(3):323330 gel. Am J Gastroenterol 90(11):20322034
Back Pain and Disc Degeneration:
Are They Really Linked? 16
Kjell Olmarker
Contents 16.1 Introduction and Overview

16.1 Introduction and Overview ............................................ 261
Back pain or lower back pain (LBP) still remains a contro-
16.2 Definitions ........................................................................ 262
versy, with an apparent lack of understanding of the basic
16.3 Controversies ................................................................... 262 genetic and pathophysiologic mechanisms that predispose
16.4 The Basis of Pain: Innervation patients to this common disabling condition. It has been esti-
of the Intervertebral Disc ............................................... 262 mated that LBP affects up to about 80 % of all people during
16.5 Degeneration of the Intervertebral Disc: their lifetime (Andersson 1995, 1999). At a given moment,
A Normal Aging Process?............................................... 265 approximately 30 % of the American population suffer from
16.6 LBP and Disc Degeneration in Clinical Studies ........... 265 LBP (Frank et al. 1996). LBP thus affects many individuals
and is a major societal cost in terms of inability to work,
16.7 Secondary Changes of Disc Degeneration
That May Induce Pain .................................................... 266
medical treatment, and rehabilitation. In the USA, the total
cost was estimated to $50 billion in 1991 (Frymoyer and
16.8 Therapeutic Strategies: Should the Target
Cats-Baril 1991) and in Sweden in 1995, with a population
Be Nucleus Pulposus Degeneration? ............................. 268
of eight million people, to SEK 29.4 billion (approx. US $5
16.9 Disc Degeneration and Its Relationship to LBP: billion) (Nachemson and Jonsson 2000). Low back pain usu-
Conclusions ...................................................................... 269
ally has a good prognosis, and most people are able to return
16.10 Summary of Critical Concepts Discussed to work within a limited period of time, although about 20 %
in the Chapter .................................................................. 269
may have recurring problems within 6 months (Cassidy et al.
References ...................................................................................... 270 2005). Also, a considerable number of individuals may suf-
fer from long-lasting low back pain. Because of the lack of
knowledge of the underlying causes for back pain and sciat-
ica, it is difficult to define specific treatment modalities.
Based on these facts, numerous studies and theories have
been presented regarding a possible cause of LBP. One point
that is often overlooked is that LBP is not a specific medical
condition with a precise pathogenesis. Instead, it should be
looked upon as a mere symptom, like headache of fever, with
a variety of potential causes. However, one specific mecha-
nism that has attracted much attention in recent years has
been linked to so-called disc degeneration.
Degeneration of the intervertebral as a clinical entity was
first described in the 1940s (Epps 1942; Friberg 1948; Friberg
K. Olmarker, MD, PhD and Hirsch 1949; Olsen 1950; Alvik 1950), and at the same
Musculoskeletal Research, time suggestions were made that the intervertebral disc could
Department of Medical Chemistry and Cell biology, be a source of back pain (Lindblom 1948). Since then, con-
Institute of Biomedicine, Sahlgrenska Academy,
siderable effort and time has been devoted to establishing
University of Gothenburg, 440, Medicinaregatan 9A,
SE-405 30 Gothenburg, Sweden specific pathophysiological mechanisms, which could then
e-mail: kjell.olmarker@gu.se be used to treat disc degeneration. This has, unfortunately,

262 K. Olmarker
not been entirely successful. The purpose of this chapter is to Nevertheless, irrespective of the cause, the term degenera-
try to review the rationale and mechanism by which disc tion will be used throughout this chapter regarding changes
degeneration might produce LBP and to briefly discuss treat- in the intervertebral discs.
ment strategies that might be used to address this problem.
16.4 The Basis of Pain: Innervation

16.2 Denitions of the Intervertebral Disc
To form a rational basis for the ensuing discussion, the fol- The innervation of the lumbar intervertebral discs has been
lowing definitions will be used: studied in detail for many years. The consensus of studies
Pain is derived from the Latin word poena meaning fine dating back to the early 1930s has been that there are free
or penalty. It is defined as a more or less localized sen- nerve endings and mechanoreceptors in the superficial layers
sation of discomfort, distress, or agony, resulting from the of the annulus fibrosus and in the posterior longitudinal liga-
stimulation of specialized nerve endings (Newman ment (Jung and Brunschwig 1932; Roofe 1940; Ehrenhaft
Dorland 2007). In other words, to experience pain, there 1943; Bogduk et al. 1981; Bogduk 1983; Malinsky 1959;
is a need for transmission of a stimulus from nerve end- Roberts et al. 1995; Cavanaugh et al. 1997; Palmgren et al.
ings or receptors; alternatively, pain is set up by pathophys- 1999; Fagan et al. 2003) (Fig. 16.1). There are two conjoin-
iologic processes in the axons that are then interpreted as ing nerve plexuses that innervate the anterior and the poste-
pain stimulus by the central nervous system (neuropathic rior aspect of the intervertebral disc, respectively, in an
pain). overlapping manner (Pedersen et al. 1956; Edgar and
Degeneration in turn is derived from the Latin word Ghadially 1976; Bogduk 1983; Weinstein et al. 1988; Groen
degeneratio which implies change from a higher form to et al. 1988). Innervation to the posterior aspect of the disc is
a lower form; especially change of a tissue to a less func- mainly provided by the sinuvertebral nerve also known as
tionally active form (Newman Dorland 2007). This may the nerve of Luschka; this nerve penetrates the neurofora-
be interpreted as the result of injury or disease but may men and supplies the epidural tissues (Luschka 1850). The
also apply to the normal aging process. sinuvertebral nerve was initially thought to innervate only
the posterior annulus of the intervertebral disc and the poste-
rior longitudinal ligament, but later studies have demon-
16.3 Controversies strated that there are multiple branches, both ascending and
descending, that innervate other structures such as the facet
The first controversy relates to the disc itself: can the disc per joints (Bogduk and Long 1979; Giles and Taylor 1987;
se elicit pain? According to the definition of pain, since McLain 1993; Masini et al. 2005; Takahashi et al. 2010) and
nerves must mediate the pain, a tissue producing pain should the vertebral end plate (Brown et al. 1997; Fagan et al. 2003;
be innervated. As will be discussed here and in other chap- Ohtori et al. 2006; Bailey et al. 2011; Bogduk et al. 1981)
ters of the book, the innervation of the intervertebral disc is and that they overlap within the epidural space (Pedersen
sparse and normally restricted to the most superficial layers et al. 1956; Edgar and Ghadially 1976; Bogduk et al. 1981;
of the annulus fibrosus. Groen et al. 1988). Also, in addition to a pure somatosensory
The second controversy concerns what is generally called innervation, the intervertebral disc has been found to be
disc degeneration it begs the question: is it really a innervated by autonomic nerve fibers (Nakamura et al. 1996;
pathophysiologic process? Since the intervertebral discs of Yamada et al. 2001; van Roy et al. 2001; Raoul et al. 2003;
most individuals exhibit various degrees of degeneration Takebayashi et al. 2006; Garcia-Cosamalon et al. 2010).
and not all experience LBP, it is questioned whether the label The overall conflict with the hypothesis that changes in
degeneration is consistent with its pathophysiologic status. the intervertebral disc would transmit pain to the central ner-
It seems even less appropriate to term such changes in the vous system is the observation that the innervation is located
structure of the intervertebral discs as degenerative disc dis- only in the superficial layers of the annulus fibrosus (Malinsky
ease, which undoubtedly lends a pathophysiological status to 1959; Hirsch et al. 1963; Groen et al. 1988; Ashton et al.
something that might simply reflect normal aging. It must, 1994; Roberts et al. 1995; Palmgren et al. 1999; Fagan et al.
however, be noted that for 95 % of LBP cases, no specific 2003). This is the case in normal nondegenerated discs. On
cause can be found other than exhibiting a radiologic image the other hand, degenerated disc specimens exhibit ingrowths
that suggests rightfully or not that the pain is associated with of newly sprouted nerves into the deeper portions of the disc.
degenerative disc disease. Probably, with age, the degenera- However, this study was performed mainly on the anterior
tive changes are such that it might possibly be possible to portion of the intervertebral disc, and ingrowths were only
distinguish normal from those associated with pathology. demonstrated in the inner part of the annulus fibrosus and, in
16 Back Pain and Disc Degeneration: Are They Really Linked? 263
Normal intervertebral disc
NP
III
A, C
II
Degenerated intervertebral disc
NP
III
A, C
II
Non-peptidergic small sized neurons / Ret / GFR1 / GFR3 / IB 4 / VR1 /P2X3 /TMP
Peptidergic small sized neurons / TrkA + / TrkB + / GFR 3 / CGRP / SP / TRPV1
Intermediate sized neurons / TrkB / Ret / GFR 1, DEG, / EnaCs TRP ion channels
Large sized neurons / TrK C= / Ca2+ binding proteins / EnaCs TRP ion channels
Fig. 16.1 Schematic representation of the innervation of normal (top) they mediate sensations of touch, pressure, and vibration. Most of the
and degenerated (bottom) intervertebral discs (IVDs), as well as the sensory nerve fibers innervating the IVD are Ad or C fibers. They origi-
origin of sensory nerve fibers that innervate them. In the normal IVD, nate from small peptidergic neurons expressing TrkATrkB (the recep-
innervation is restricted to the outer layers of the annulus fibrosus (AF) tor for nerve growth factor/brain-derived neurotrophic factor, red) or
and consists of small nerve fibers (red and green) and some large fibers non-peptidergic neurons expressing the common signaling receptor for
forming mechanoreceptors (brown). In the degenerated IVD, nerve glial cell-derived neurotrophic factor family of neurotrophic factors
fibers are increased in number, and they enter the inner layers of the AF (Ret) (red). Neurons in DRGs can be differentiated based on their pat-
and even the nucleus pulposus (NP). Furthermore, in these conditions, tern of expression of receptors for neurotrophic factors, pattern of
the density of mechanoreceptors in the superficial layers of IVDs is expression of different ion channels primarily of the degenerinepithelial
increased. Dorsal root ganglions (DRGs) contain different types of sen- sodium channels (DEGENaCs) (ENaCa, ENaCb, and ENaCc), acid-
sory neurons that project to the IVD and to the dorsal horn of the spinal sensing ion channel (ASIC) (ASIC1, ASIC2, and ASIC3) and transient
cord (DH of SC). Thin myelinated Ad fibers and unmyelinated C fibers receptor potential (TRP) (TRPA1, TRPC1, TRPC6, and TRPV14)
arise from small neurons (red and green), which, in the spinal cord, families, and peptide content. CGRP calcitonin gene-related peptide,
synapse in laminas I and II and mediate nociception. The myelinated GFRa1 and GFRa3 glial cell line-derived neurotrophic receptor sub-
Ab fibers (brown) arise from intermediate neurons; at the periphery, types a1 and a3, P2X3 ATP-gated ion channel subtype P2X3, SP sub-
they form slowly and rapidly adapting low-threshold mechanoreceptors stance P, TMP thiamine monophosphatase, VR1 vanilloid receptor
and synapse in laminas III and IV in the dorsal horn of the spinal cord; subtype 1 (Modified from Garcia-Cosamalon et al. (2010))
264 K. Olmarker
some cases, possibly extending into the nucleus pulposus innervation of the disc comprises a mixture of fibers that are
(Yoshizawa et al. 1980; Ashton et al. 1994; Coppes et al. nociceptive, related to mechanoreceptors, and sympathetic
1997; Freemont et al. 1997; Johnson et al. 2001). A later in nature (reviewed in Garcia-Cosamaln) (Garcia-
study acknowledged that the posterior region of the disc Cosamalon et al. 2010). In this regard, the disc nerve fibers
might be more relevant, and indeed an ingrowth of nerves are positive for PGP 9.5, substance P, calcitonin gene-related
positive for substance P and VIP was noted extending to the peptide (CGRP), acetylcholinesterase, vasoactive intestinal
outer part of the nucleus pulposus (Peng et al. 2005). Since peptide (VIP), neuropeptide Y, C-flanking peptide, and syn-
this ingrowth was always present when there was granulation aptophysin (Garcia-Cosamalon et al. 2010). Interestingly,
tissue and annular fissures, the authors concluded that the following degenerate disc injury and repair, the nerve fibers
zone of granulation may be responsible for discogenic pain that penetrate deep into the annulus fibrosus, and later even
due to neo-innervation of the disc and possibly a cause of further, appear to exhibit a similar distribution of fiber types
pain from discography (Peng et al. 2005). It was also recently (Takahashi et al. 2009). As neo-innervation and the neovas-
demonstrated that when autologous nucleus pulposus tissue cularization occur in parallel, it has been suggested that
fragments were placed subcutaneously in pigs, there is an newly formed nerve fibers are mainly vasoregulatory in func-
ingrowth of newly formed nerves and blood vessels. In con- tion (Freemont et al. 1997, 2002; Johnson et al. 2001, 2007).
trast, similar ingrowth were not seen when disc tissue was However, nerve endings not related anatomically to newly
placed in retroperitoneal fat. It was also observed that formed vessels have also been identified (Ashton et al. 1994).
cytokine inhibitors reduced this ingrowth (Olmarker 2005). In addition, since the peptide content of the nerve fibers
It has also been suggested that bFGF and TGF-b1 as well as found in degenerate intervertebral discs includes those
macrophages and mast cells are involved in the ingrowth of related to pain transmission and since the nerves related to
nerves and blood vessels into the intervertebral disc and that blood vessels seem to be of sympathetic origin (Yamada
this phenomenon is related to the injury repair of the annulus et al. 2001), the possibility exists that there are nerve fibers
fibrosus (Peng et al. 2006a). that may be involved in nociception.
Semaphorin is an axonal guidance molecule that serves to Based on what was considered above, it begs the ques-
repel axonal growth (Rohm et al. 2000; Nakamura et al. tion: are nerve fibers with peptides related to nociception
2000; Liu and Strittmatter 2001). Recently, it was found that activated in the center of the intervertebral disc, or is their
this molecule is present in the healthy intervertebral disc, presence merely accidental and that they are dormant? Most
mainly in the outer annulus (Tolofari et al. 2010), where it innervation studies of intervertebral discs have reported the
presumably could prevent axonal ingrowth. Even more inter- presence of nerve endings or axons (McCarthy et al. 1991;
esting is that the same authors found that the levels of sema- Palmgren et al. 1999; Fagan et al. 2003; Aoki et al. 2004),
phorin are lowered in degenerate discs, which would then and the only receptor types are mechanoreceptors (Roberts
enhance neo-innervation (Tolofari et al. 2010). Recently, it et al. 1995; Dimitroulias et al. 2010; Cavanaugh et al. 1995).
was suggested that neurotrophins family members were The receptors are always found in the superficial layers of
involved in the ingrowth of nerves into the central part of the the annulus (Roberts et al. 1995; Cavanaugh et al. 1995).
intervertebral discs (Purmessur et al. 2008; Garcia-Cosamalon However, it is questionable if the mechanoreceptors could
et al. 2010). In the human, this family comprises NGF, produce pain; thus, induced pain is probably related to direct
BDNF, NT-3, and NT-4/5 (Ebendal 1992; Barbacid 1995; effects on axons or so-called free nerve endings in the inter-
Lessmann 1998). The neurotrophins regulate cell prolifera- vertebral disc.
tion and differentiation via the Trk family of receptor tyrosine It is known that during degeneration of the nucleus pulpo-
kinase and via p75-NTR (Dechant and Barde 1997; Lu et al. sus there are a number of biochemical changes. One such
2005; Skaper 2008). Since the neurotrophins seem to be change is the accumulation of metabolites such as lactate and
expressed at higher levels in degenerate than in nondegener- the concomitant pH decrease (Diamant et al. 1968;
ate discs, it may be assumed that they may facilitate ingrowth Nachemson 1969; Buckwalter 1995; Bartels et al. 1998;
of new nerve fibers into the deeper parts of the intervertebral Keshari et al. 2008; Rajasekaran et al. 2010). Accordingly,
disc during the degeneration process (Freemont et al. 2002). pain could be due to chemical excitation of adjacent nerve
In summary, based on existing knowledge, there is indeed fibers. Similarly, the increased expression of various bioac-
reason to assume that there is neo-innervation and neovascu- tive substances, such as members of the neurotrophin family,
larization of the annulus fibrosus and nucleus pulposus and may activate pain receptors (Freemont et al. 2002; Purmessur
that such ingrowth relates to the presence of bioactive sub- et al. 2008; Sugiura et al. 2008; Garcia-Cosamalon et al.
stances generated during repair following annular fissures. 2010; Orita et al. 2011). Biomechanical instability of the
Although there may be neo-innervation of injured inter- motion segment may lead to excess mobility of the degener-
vertebral discs, the function and properties of such newly ated intervertebral disc. Likewise, it has been suggested that
formed nerves are not known (Fig. 16.1). The normal such mobility may actively induce signaling in newly formed
nerve fibers in the deeper parts of the intervertebral disc, components through fissures in the weakened annulus
another plausible cause for disc-triggered pain (Morgan fibrosus into the spinal canal. Similar leakage may also occur
and King 1957; Kirkaldy-Willis and Farfan 1982; Pope and on the ventral aspect of the disc and also through the verte-
Panjabi 1985; Bradford 1994; Kim et al. 2005). Finally, bral end plates into the adjacent vertebral bodies (nodes of
injured axons are known to be stimulated by the nerve sig- Schmorl) (Schmorl 1929). What is not known is whether
nals from adjacent axons, so-called cross-excitation these variants of leakage/herniation produce clinical symp-
(Rasminsky 1987; Lisney and Devor 1987; Devor and Wall toms (Sward et al. 1990; Takahashi et al. 1995; Zhang et al.
1990; Amir and Devor 1992). Injury can cause a myelin 2010). More importantly, the question remains, should these
defect due to the loss of its electro-isolating properties: an changes in disc structure be considered as normal aging, or
electrical phenomenon in which nerve impulses in one axons should they be described as a degenerative process?
cause an impulse in adjacent axons. Such artificial ectopic As discussed previously, the term degeneration may be
impulses may be interpreted as pain by the central nervous related to the aging process, and hence, the pathophysiologi-
system, regardless of the axon that originally transmits prop- cal connotation cannot be ignored. Indeed, if the aging pro-
rioceptive, temperature, or pressure information (Burchiel cesses are shown to induce pain, it might rightfully be termed
1984; Zimmermann 1984; Devor 1991, 2006; Han et al. degeneration. However, signs of spinal pathology, including
2000; Costigan et al. 2009). Whether this occurs in the inter- so-called disc degeneration, are often encountered in asymp-
vertebral disc seems unlikely due to its sparse innervation. tomatic individuals (Boden et al. 1990; Jensen et al. 1994;
In summary, in recent years, knowledge of intervertebral Boos et al. 1995; Stadnik et al. 1998). In this regard, it might
discs innervation has increased tremendously, and the possi- be more appropriate to talk about normal and pathological/
bility exists that there is disc-triggered or discogenic symptomatic disc aging instead of degeneration. It is thus
pain. However, pain mechanisms still need to be demon- still not evident how to apply a proper definition to this state.
strated convincingly before any conclusions can be drawn as However, according to this author, the term degeneration
to its clinical relevance. may be unfortunate and misleading, particularly in persons
without any symptoms.
16.5 Degeneration of the Intervertebral

Disc: A Normal Aging Process? 16.6 LBP and Disc Degeneration
in Clinical Studies
The intervertebral discs are usually referred to as the largest
avascular structures of the human body (Holm et al. 1982; The literature suggests that there is a strong correlation
Horner and Urban 2001; Roberts 2002; Grunhagen et al. between degenerative disc changes and LBP (Lindblom
2011). The nutrition of the cells of the discs, mainly located 1948; Kelsey and White 1980; Lutz et al. 2003; Chou et al.
in the nucleus pulposus, is mainly provided by diffusion 2011). However, since no causal link has been confirmed and
from the vertebral end plates. Accordingly, if the nutritional since a majority of non-symptomatic patients have degenera-
supply becomes insufficient, the nucleus pulposus may tive disc changes, a causal relationship may be unlikely. This
undergo changes that may be referred to as age-related or is analogous to the recent discussion on the relationship
degenerative. This process may be initiated as early as the between Modic-type changes in the vertebrae and LBP.
second decade of life, and its frequency will increase with Similar to disc degeneration, there is a correlation between
increasing age (Takatalo et al. 2009; Samartzis et al. 2011). Modic changes and low back clinical pain (Braithwaite et al.
There are reports that 40 % of individuals aged 30 years, 1998; Kjaer et al. 2005; Jensen et al. 2008; Thompson et al.
53 in the age group 3055 years, and 90 % aged 5055 years 2009). Again, there is no established causal pathophysiologi-
exhibit degeneration of one or more discs (Kanayama et al. cal link. One must therefore consider that there may be a
2009; Cheung et al. 2009). Degeneration would be expected transferred statistical significance, i.e., the Modic changes
to lead to changes in the biochemical composition of the disc and low back pain may coexist and have a common patho-
as well as changes in its mechanical properties. The nucleus genic source but no peer relationship. One such pathogenetic
pulposus will be more liquefied, and there may also be a factor might be disc injury. For example, it is known that
weakening of the containing structure, the annulus fibrosus. osteoarthritis in the knee joint may result in changes, fairly
Eventually, the disc will form a more solid and organized similar to the Modic changes, in the femur and that these
structure, which is also usually associated with a reduction changes/erosions are considered to be strong indicators of
of the disc height (Burton et al. 1996; Benneker et al. 2005; knee cartilage injury (Alexander 1960; Rose and Cockshott
Inoue and Espinoza Orias 2011). During the transition from 1982; Pattrick et al. 1993). As will be discussed later, disc
a young healthy disc to the more solid state, there is risk of injury with annular tears and leakage of liquefied nucleus
leakage of the liquefied nucleus pulposus with its bioactive pulposus tissue to the superficial and innervated part of the
266 K. Olmarker
annulus fibrosus could be one cause of low back pain. In this

regard, disc injury would be the common denominator, and a systematic literature review of prevalence and
an apparent statistical relationship would exist although association with non-specific low back pain. Eur
Modic changes and low back pain merely coexist with no Spine J 17:14071422
pathophysiologic link. The same phenomenon may be Kjaer P, Leboeuf-Yde C, Korsholm L et al (2005)
applied to the relationship between disc degeneration and Magnetic resonance imaging and low back pain in
LBP. Since no causal link has been established, there may be adults: a diagnostic imaging study of 40-year-old men
another common denominator perhaps in this case, sec- and women. Spine (Phila Pa 1976) 30:11731180
ondary to changes in the intervertebral disc. Modic MT, Steinberg PM, Ross JS et al (1998)
Degenerative disk disease: assessment of changes
in vertebral body marrow with MR imaging.
Box 16.1 Modic Changes Radiology 166:193199
The so-called Modic changes were first described by Thompson KJ, Dagher AP, Eckel TS et al (2009)
Michael Modic and collaborators in 1988 (Modic et al. Modic changes on MR images as studied with pro-
1998). Magnetic resonance images of vertebral body vocative diskography: clinical relevance a retro-
changes were reviewed of 474 patients. It was found spective study of 2457 disks. Radiology
that 4 % of the patients had decreased signal intensity 250:849855
on T1-weighted spin-echo images and increased signal
intensity on T2-weighted images, a finding that was
given the grade 1. In 16 % of the patients, there was
increased signal intensity on T1-weighted images and 16.7 Secondary Changes of Disc
isointense or slightly increased signal intensity on Degeneration That May Induce Pain
T2-weighted images, which was graded as 2. A third
grade was added later and is defined as bone scar tis- About 80 % of the population worldwide will experience
sue. The authors concluded, These signal intensity LBP sometime during their lifetime (Andersson 1995, 1999).
changes appear to reflect a spectrum of vertebral body However, in only a minority of cases, a definitive clinical
marrow changes associated with degenerative disk dis- diagnosis such as vertebral fracture, infection, or tumor will
ease. Today it is known that these changes closely be established (Staiger et al. 1999; Goupille et al. 2000;
correlates to low back pain (Braithwaite et al. 1998; Della-Giustina and Kilcline 2000; Henschke et al. 2009).
Kjaer et al. 2005; Jensen et al. 2008; Thompson et al. The remaining 95 % will therefore be termed idiopathic.
2009). However, no causal pathophysiological expla- Most likely, there may be one or more specific mechanisms
nation has been found. It is therefore a possibility that that may be hidden within these 95 % that have yet to be
there might be a transferred statistical significance, discovered. Although events occurring within the disc might
i.e., the Modic changes and low back pain may coexist theoretically induce LBP (discogenic LBP), the degenera-
but have a common pathogenic source and no peer tion of the intervertebral disc may also cause secondary
relationship. One such likely source might be disc changes that could be overlooked.
injury. It is known that there is an increased incidence One such effect is the subsequent reduction of the disc
of Modic changes in patients with disc herniations height during the degeneration process. A consequence of this
(Albert and Manniche 2007). If the Modic changes lowered height is that there may be malpositioning of the facet
might be responsible for low back pain or not is thus joints (Ghomley 1993; Eisenstein and Parry 1987; Mooney and
debatable, but it seems plausible that these changes Robertson 1976). Theoretically such a change could lead to
may serve as an indicator of disc injury. cartilage injury and damage to the capsule facet joint (Fujiwara
et al. 2000a, b; Kong et al. 2009). It is known that the facet
References joints are richly innervated (Bogduk and Long 1979; Bogduk
Albert HB, Manniche C (2007) Modic changes follow- 1983), and attempts have been made to treat facet-joint-related
ing lumbar disc herniation. Eur Spine J 16:977982 problems with local injections of anesthetics (Carrera 1980;
Braithwaite I, White J, Saifuddin A et al (1998) Manchikanti et al. 2007, 2008). However, it is likely that
Vertebral end-plate (Modic) changes on lumbar osteoarthritis of these joints and the ensuing production of
spine MRI: correlation with pain reproduction at inflammatory agents that can leak through the injured joint
lumbar discography. Eur Spine J 7:363368 capsule could serve to irritate and stimulate intraspinal nervous
Jensen TS, Karppinen J, Sorensen JS et al (2008) structures including the nerve root and nerve endings on the
Vertebral endplate signal changes (Modic change): posterior aspect of the intervertebral disc (Hasue 1993;
Willburger and Wittenberg 1994; Igarashi et al. 2004).
Another striking feature of the degenerative changes of

Nerve root: buttock and leg pain
the intervertebral discs is the formation of annular tears or
fissures (Hilton et al. 1980; Videman and Nurminen 2004;
Ross et al. 1989; Osti et al. 1992). Such fissures may allow
leakage of the degenerated and partly liquefied nucleus Facet capsule:
pulposus (Stadnik et al. 1998; Saifuddin et al. 1999; Derby occasional cause of
et al. 2005; Peng et al. 2006b). Biologic product from the back pain
degenerative nucleus pulposus could diffuse to the outer
annulus fibrosus or even the spinal canal. MRI analysis has Outer annulus:
shown that a zone of increased inflammatory activity, the the site of back pain
high-intensity zone (HIZ), exists in the superficial regions of
the annulus (Schellhas et al. 1996; Peng et al. 2006b; Carragee
et al. 2000). This zone is associated with LBP although it
may also be found in asymptomatic patients (Weishaupt Fig. 16.2 Observations at probing in conscious patients under local
anesthesia during laminectomy. Probing of the nerve root adjacent to a
et al. 1998; Stadnik et al. 1998; Carragee et al. 2000; Wang
disc herniation reproduces buttock and leg pain. Probing of the fact
et al. 2008). capsule occasionally reproduces back pain, whereas probing of the
It is known that nerve endings and mechanoreceptors outer annulus always reproduces back pain and is also concluded to be
exist on the posterior aspect of the annulus fibrosus; while the site of back pain (Republished from Kuslich et al. (1991))
these are normally dormant, they can be activated by
inflammatory stimuli (Rang et al. 1991; Dray 1995; Our group studied whether application of nucleus pul-
Ozaktay et al. 1994; Cavanaugh 1995; Coutaux et al. posus samples to the superficial annulus fibrosus might
2005). It is therefore reasonable to assume that biologi- induce LBP in rodents. As discussed in Chap. 18, assess-
cally active substances released from the degenerating ment of LBP in an animal model is a significant challenge.
nucleus pulposus may activate cognate receptors by diffus- While alterations in specific nerves activity may be
ing into the spinal canal through the annular fissures. One assessed by neurophysiologic recordings, pain itself is
diagnostic approach for assessing symptomatic interverte- dependent on its interpretation by the central nervous sys-
bral discs is through discography, a procedure in which a tem. In animals, there is a limitation in available experi-
fluid is injected under pressure into the center of the disc to mental methodologies to assess whether the transmitted
provoke pain (Moneta et al. 1994; Schellhas et al. 1996; signals are perceived as pain. Methods that can be used
Carragee 2000; Stout 2010). Possibly, the pain is not for assessing pain is through functional MRI (Hsu et al.
related to activation of newly formed nerves within 1998; Weber et al. 2006; Adamczak et al. 2010) (see
the intervertebral disc, but to a washout of bioactive sub- Chap. 12); another approach is to study behavioral changes
stances into the spinal canal and the innervated structures (Kawakami et al. 1994; Abbott et al. 1995; Olmarker and
including the disc, the longitudinal ligament, and facet- Myers 1998).
joint capsules. In rodents, we have assessed spontaneous behavioral
In a pioneering study, in patients undergoing surgery for changes following experimental disc herniation (Olmarker
disc herniation under local, progressive anesthesia, Kuslich et al. 2002). In contrast to most other assessment modalities,
and colleagues stimulated various parts of the spinal com- spontaneous behavior analysis, which has been used to eval-
plex (Kuslich et al. 1991). They found that stimulation of a uate psychological parameters (Wuttke and Hoffmeister
nerve root did not generate pain, whereas mechanical stimu- 1968; Monti and Carlini 1975; Rodriquez-Enchandia et al.
lation of the nerve root adjacent to the herniated disc pro- 1986; Garcia-Cabrera and Berge 1990), examines an ani-
duced a radiating pain out into the lower limb, similar to mals instinctive involuntary activities. Despite the high vari-
sciatic pain. The most interesting observation, however, was ation in behavior between animals, changes in spontaneous
that stimulation of the posterior part of the annulus fibrosus behavior may reveal discomfort or irritation that is not easily
produced pain in the lower back, similar to LBP. The authors obtained by other behavior modalities. In collaboration with
concluded that the posterior annulus fibrosus was the site of the Department of Physiology, University of Bergen, Norway,
back pain (Kuslich et al. 1991). There is thus compelling spontaneous behavior assessment was adapted to study
evidence that activation of nociceptive nerve endings or changes following experimental disc herniation. Procedures
receptors in the posterior aspect of the annulus fibrosus used included disc puncture with transfer of the nucleus pul-
(Weber et al. 2006) may produce pain in the lower back; this posus tissue to, and a slight mechanical deformation of, the
finding may be due to the release of agents from the nucleus adjacent nerve root (Olmarker and Myers 1998; Olmarker
pulposus during discography and also possibly at non- et al. 1998, 2003). The study showed that experimental disc
provoked sites (Fig. 16.2). herniation induced an increased rotation of the head towards
268 K. Olmarker
the hind paw on the operated side and an elevation in paw experimental and clinical evidence that leakage of disc
lifting (Olmarker and Myers 1998). These two behaviors material onto the posterior annulus fibrosus may be an actual
were most pronounced the day after surgery and then gradu- cause of LBP possibly mediated through TNF.
ally declined during the following 14 days. At day 21, the
rats displayed increased immobility and reduced locomo-
tion. The behavior changes during the first 14 days sug- Box 16.2 Molecular Events That Are Characteristic
gested that following surgery, animals experienced focal pain of the Aging Process
or irritation located in the hind paw on the operated side. There is limited information on factors that character-
During days 1421, a more chronic non-focal pain compo- ize the normal aging process in the intervertebral disc.
nent was present and seen as reduced mobility. Using this However, since the aging disc shares some genetic as
assessment tool, the authors studied changes induced by disc well as molecular characteristics of cartilage, changes
puncture, thus simulating a leakage of nucleus pulposus onto that are evident in cartilage are listed below:
the posterior aspect of the annulus fibrosus, but with no con- Mild fibrillation of and softening of the articular
tact with the adjacent nerve root. The assumption was that surface
the rats would display reduced motion, analogous to LBP, Osteophyte formation
similar to that reported at 21 days in the radiculopathy model. Loss of matrix tensile strength and stiffness
While the rats did not exhibit reduced mobility, they evi- Responsiveness to inflammatory stimuli
denced increased grooming and an uncharacteristic shak- Changes in metabolism
ing of the head and upper body: both behaviors were Accumulation of ROS and glycation products
displayed significantly more often in disc punctured rats than Changes in IGF, insulin, and growth hormone levels
in rats with non-punctured discs (Olmarker 2008). Increased Progressive chondrocyte senescence
grooming has been observed in other studies and has been Erosion in of chondrocyte telomere length
suggested to reflect anxiety, stress, and chronic neuropathic Changes in phenotype
pain (Millan and Emrich 1981; Crawley and Moody 1983; Decrease in average size of proteoglycan monomers
Vos et al. 1994; Deseure and Adriaensen 2002; Eriksson Decrease in the aggregation capacity of the proteogly-
et al. 2005). The observed shaking is commonly seen in psy- can monomers
chopharmacological studies and termed wet-dog shakes Expression of the senescence-associated enzyme beta-
(WDS): it is a typical sign of withdrawal from opiates, ben- galactosidase
zodiazepines, and barbiturates (Colasanti and Khazan 1975; Mitochondrial degeneration
Baldino et al. 1979; Horowitz and Allan 1982; Martin et al.
1982). The shakes have also been considered to relate to
stress (Treptow et al. 1986; Deschamps and Couture 2005;
Brotto et al. 1999) and pain (Papir-Kricheli et al. 1987;
Kitamura et al. 2007). It was concluded that rats with punc- 16.8 Therapeutic Strategies: Should
tured discs experienced some kind of discomfort. However, the Target Be Nucleus Pulposus
it was not possible to determine if this discomfort was Degeneration?
analogous to LBP.
In a follow-up study, we evaluated if the observed behav- Inevitably, a chapter on disc degeneration and LBP must
ior changes were induced by disc injury or the presence of provide insights into possible therapeutic strategies.
nucleus pulposus tissue on the posterior aspect of the annu- Although it is not fully understood if the disc tissue per se
lus fibrosus. It was found that disc puncture and application can produce clinical pain, the target of much of the work is
of nucleus pulposus induced similar behavioral changes. In the degeneration process itself and the disc as a pain genera-
contrast, the same behavioral changes were not evident fol- tor. Surgically, this has been approached by fusion of the
lowing ventral disc puncture and a superficial disc injury potentially painful disc function unit, a procedure that is fre-
without penetration of the annulus fibrosus, with no leakage quently performed with varying degrees of success (Gibson
of nucleus pulposus tissue (Olmarker 2011). It was also and Waddell 2005; Carreon et al. 2008; Glassman et al.
observed that inhibition of TNF also markedly reduced the 2009). There are also promising attempts to revitalize the
behavioral effects (Nakamae et al. 2011). Interestingly, a disc by injection of stem cells or biological components
recent randomized clinical trial focused on evaluating the (Wehling et al. 1997; Nishida et al. 2000; Yoon et al. 2004;
effects of epidural TNF inhibition on sciatic pain also Sakai et al. 2003; Brisby et al. 2004; Risbud et al. 2004) (see
revealed that in addition to the inhibitors beneficial effects Chaps. 23 and 24). However, although these procedures
on the sciatic pain, it also induced a marked reduction of might be successful in the future, they do raise a set of new
LBP (Cohen et al. 2009). In summary, there is emerging questions. Based on current knowledge, it will be difficult to
assess when, how often, and in which discs should such

Discogenic pain
interventions be performed. We know that there are dan-
gerous years: from the initiation of degeneration/aging,
Pain from the disc per se
when the disc may leak, herniate, and produce pain, until Less likely
the end stage when the disc has been transformed to a more
solid connective tissue, with less likelihood to produce pain. Disc
A provocative but appropriate question would be: would it degeneration
not be a better strategy to increase the degeneration/aging
Secondary changes
process so as to shorten the duration of these dangerous
years? This notion was suggested 50 years ago by Hirsch Disc height reduction
who recommended that injection of a chondrolytic enzyme Less likely
into the disc would increase the degeneration process and
Facet joint involvement
convert the disc to a solid, asymptomatic structure Likely
(Hirsch 1959). This original idea was later modified by
Smith who indicated that chymopapain would disintegrate NP leakage
Very likely
the inner structure of the disc and thus be of clinical use for
the treatment of sciatica; this procedure was later known as
chemonucleolysis (Smith 1964). Bearing this in mind, it Fig. 16.3 The authors view of pain mechanisms and their likelihood
might be prudent to consider the apparent risk of delaying as the result of disc degeneration
the degeneration/aging process of the disc by biological
means, in the sense that instead of having mostly asymp-
tomatic discs at the age of 6070, there is the possibility 16.9 Disc Degeneration and Its Relationship
that symptomatic discs may be present at much higher to LBP: Conclusions
ages than would be expected if the disc would be allowed to
age normally. The author opines that this potential problem, Although emerging evidence indicates that degeneration may
a reduction in the degeneration process, has not being fully induce innervation of deeper parts of the intervertebral disc, it
considered and needs to be further explored. seems unlikely that the degenerative process produces pain
An alternate strategy to that of modifying the degenera- per se. More likely, it is the result of secondary changes asso-
tion/aging process per se is to control pain pathways. Since ciated with degeneration (Fig. 16.3). As challenging as ever
the source of the pain has not been identified, this form of are questions that include whether future treatment strategies
symptomatic treatment would be analogous to aiming should be directed at the degeneration process itself, what is
water at flames and not at the source of the fire. In fact, the importance of controlling secondary changes, and should
since it is not evident that the disc generates the pain, sec- the disc be rejuvenated and when and how should this be
ondary changes such as loss of disc height might eventually accomplished. While this chapter is both speculative and
prove to be involved (Twomey and Taylor 1985; Berlemann somewhat provocative, it is merely the result of the apparent
et al. 1998; Shao et al. 2002). For example, the loss of disc lack of understanding of the pathophysiologic background of
height could influence facet-joint function. Misalignment LBP, and it is the hope that this contribution may stimulate
of the articular surfaces could induce osteoarthritic changes continued discussion and research in this complex field both
and the release of bioactive molecules. Diffusion of these from a basic scientific and a clinical perspective.
agents to the intraspinal nervous structures could then
induce pain by similar mechanisms to those discussed pre-
viously (Hasue 1993; Willburger and Wittenberg 1994; 16.10 Summary of Critical Concepts
Igarashi et al. 2004). Discussed in the Chapter
Another inevitable consequence of degeneration is disar-
rangement of the fibrils of the annulus fibrosus and leakage The origin of disc pain is still not understood. Pain most
of the partly liquefied degenerate disc. To the author, this likely originates from receptors on nerve fibers on the sur-
seems to be a very likely scenario and could well be the cause face of the posterior aspect of the intervertebral disc.
of LBP of idiopathic origin. Perhaps, in the future, LBP Pain activation may be due to leakage of disc-derived bio-
treatment may comprise a dual strategy: epidural administra- logically active molecules which diffuse through annular
tion of agents that inhibit key biological processes and a tears of the annulus fibrosus.
sealant of annular tears, possibly in combination with sur- The paucity of innervating nerve fibers would indicate
gical immobilization of the spinal segment and the preserva- that pain is unlikely to originate from the center of the
tion of disc height. intervertebral disc.
270 K. Olmarker
Disc degeneration and degenerative disc disease are inap- Bogduk N (1983) The innervation of the lumbar spine. Spine (Phila Pa
propriate terms for changes in the intervertebral disc that 1976) 8(3):286293
Bogduk N, Long DM (1979) The anatomy of the so-called articular
may well be merely a stage in the normal aging process. nerves and their relationship to facet denervation in the treatment
Efforts to rejuvenate the intervertebral disc by injection of of low-back pain. J Neurosurg 51(2):172177. doi:10.3171/jns.
cells or biologic substances may be contraindicative, since 1979.51.2.0172
it might prolong the degeneration process. Bogduk N, Tynan W, Wilson AS (1981) The nerve supply to the human
lumbar intervertebral discs. J Anat 132(Pt 1):3956
Boos N, Rieder R, Schade V, Spratt KF, Semmer N, Aebi M (1995)
Acknowledgements AFA Insurance, Stiftelsen Olle Engkvist 1995 Volvo Award in clinical sciences. The diagnostic accuracy of
Byggmstare, and the Spine Society of Europe Task Force on magnetic resonance imaging, work perception, and psychosocial
Research factors in identifying symptomatic disc herniations. Spine
20(24):26132625
Bradford DS (1994) Surgical treatment of low back pain in spine insta-
bility. Chir Organi Mov 79(1):6368
References Braithwaite I, White J, Saifuddin A, Renton P, Taylor BA (1998)
Vertebral end-plate (Modic) changes on lumbar spine MRI: correla-
Abbott FV, Franklin KB, Westbrook RF (1995) The formalin test: scor- tion with pain reproduction at lumbar discography. Eur Spine J
ing properties of the first and second phases of the pain response in 7(5):363368
rats. Pain 60(1):91102 Brisby H, Tao H, Ma DD, Diwan AD (2004) Cell therapy for disc
Adamczak JM, Farr TD, Seehafer JU, Kalthoff D, Hoehn M (2010) High degenerationpotentials and pitfalls. Orthop Clin North Am
field BOLD response to forepaw stimulation in the mouse. 35(1):8593. doi:10.1016/s0030-5898(03)00104-4
Neuroimage 51(2):704712. doi:10.1016/j.neuroimage.2010.02.083 Brotto LA, Hanson LA, Gorzalka BB (1999) Nefazodone attenuates the
Alexander C (1960) Erosion of the femoral shaft due to patellofemoral stress-induced facilitation of wet dog shaking behaviour but not the
osteoarthritis. Clin Radiol 11:110113 facilitation of sexual behaviour in female rats. Eur J Pharmacol
Alvik I (1950) Degeneration of the lumbar intervertebral disks. Nord 381(23):101104
Med 44(32):12711274 Brown MF, Hukkanen MV, McCarthy ID, Redfern DR, Batten JJ, Crock
Amir R, Devor M (1992) Axonal cross-excitation in nerve-end neuromas: HV, Hughes SP, Polak JM (1997) Sensory and sympathetic innerva-
comparison of A- and C-fibers. J Neurophysiol 68(4):11601166 tion of the vertebral endplate in patients with degenerative disc
Andersson GBJ (1995) Intervertebral disc herniation: epidemiology disease. J Bone Joint Surg Br 79(1):147153
and natural history. In: Weinstein JN, Gordon SL (eds) Low back Buckwalter JA (1995) Aging and degeneration of the human interverte-
pain. A scientific and clinical overview. American Academy of bral disc. Spine (Phila Pa 1976) 20(11):13071314
Orthopaedic Surgeons, Rosemont, pp 721 Burchiel KJ (1984) Spontaneous impulse generation in normal and
Andersson GB (1999) Epidemiological features of chronic low-back pain. denervated dorsal root ganglia: sensitivity to alpha-adrenergic stim-
Lancet 354(9178):581585. doi:10.1016/s0140-6736(99)01312-4 ulation and hypoxia. Exp Neurol 85(2):257272
Aoki Y, Ohtori S, Takahashi K, Ino H, Takahashi Y, Chiba T, Moriya H Burton AK, Battie MC, Gibbons L, Videman T, Tillotson KM (1996)
(2004) Innervation of the lumbar intervertebral disc by nerve growth Lumbar disc degeneration and sagittal flexibility. J Spinal Disord
factor-dependent neurons related to inflammatory pain. Spine 9(5):418424
29(10):10771081 Carragee EJ (2000) Is lumbar discography a determinate of discogenic
Ashton IK, Roberts S, Jaffray DC, Polak JM, Eisenstein SM (1994) low back pain: provocative discography reconsidered. Curr Rev
Neuropeptides in the human intervertebral disc. J Orthop Res Pain 4(4):301308
12(2):186192. doi:10.1002/jor.1100120206 Carragee EJ, Paragioudakis SJ, Khurana S (2000) 2000 Volvo Award win-
Bailey JF, Liebenberg E, Degmetich S, Lotz JC (2011) Innervation pat- ner in clinical studies: lumbar high-intensity zone and discography in
terns of PGP 9.5-positive nerve fibers within the human lumbar verte- subjects without low back problems. Spine 25(23):29872992
bra. J Anat 218(3):263270. doi:10.1111/j.1469-7580.2010.01332.x Carreon LY, Glassman SD, Howard J (2008) Fusion and nonsurgical
Baldino F Jr, Cowan A, Geller EB, Adler MW (1979) Effects of antip- treatment for symptomatic lumbar degenerative disease: a system-
sychotic and antianxiety drugs on the morphine abstinence syn- atic review of Oswestry Disability Index and MOS Short Form-36
drome in rats. J Pharmacol Exp Ther 208(1):6366 outcomes. Spine J 8(5):747755. doi:10.1016/j.spinee.2007.06.013
Barbacid M (1995) Structural and functional properties of the TRK Carrera GF (1980) Lumbar facet joint injection in low back pain and
family of neurotrophin receptors. Ann N Y Acad Sci 766: sciatica: preliminary results. Radiology 137(3):665667
442458 Cassidy JD, Cote P, Carroll LJ, Kristman V (2005) Incidence and course
Bartels EM, Fairbank JC, Winlove CP, Urban JP (1998) Oxygen and of low back pain episodes in the general population. Spine (Phila Pa
lactate concentrations measured in vivo in the intervertebral discs of 1976) 30(24):28172823
patients with scoliosis and back pain. Spine (Phila Pa 1976) 23(1): Cavanaugh JM (1995) Neural mechanisms of lumbar pain. Spine
17; discussion 8 20(16):18041809
Benneker LM, Heini PF, Anderson SE, Alini M, Ito K (2005) Correlation Cavanaugh JM, Kallakuri S, Ozaktay AC (1995) Innervation of the rab-
of radiographic and MRI parameters to morphological and bio- bit lumbar intervertebral disc and posterior longitudinal ligament.
chemical assessment of intervertebral disc degeneration. Eur Spine Spine (Phila Pa 1976) 20(19):20802085
J 14(1):2735. doi:10.1007/s00586-004-0759-4 Cavanaugh JM, Ozaktay AC, Yamashita T, Avramov A, Getchell TV,
Berlemann U, Gries NC, Moore RJ (1998) The relationship between King AI (1997) Mechanisms of low back pain: a neurophysiologic
height, shape and histological changes in early degeneration of the and neuroanatomic study. Clin Orthop Relat Res 335:166180
lower lumbar discs. Eur Spine J 7(3):212217 Cheung KM, Karppinen J, Chan D, Ho DW, Song YQ, Sham P, Cheah
Boden SD, Davis DO, Dina TS, Patronas NJ, Wiesel SW (1990) KS, Leong JC, Luk KD (2009) Prevalence and pattern of lumbar
Abnormal magnetic-resonance scans of the lumbar spine in asymp- magnetic resonance imaging changes in a population study of one
tomatic subjects. A prospective investigation. J Bone Joint Surg Am thousand forty-three individuals. Spine (Phila Pa 1976) 34(9):
72(3):403408 934940. doi:10.1097/BRS.0b013e3181a01b3f
Chou D, Samartzis D, Bellabarba C, Patel A, Luk KD, Kisser JM, Epps PG (1942) A case of degeneration of the intervertebral disc fol-
Skelly AC (2011) Degenerative magnetic resonance imaging lowing lumbar puncture: (section of orthopaedics). Proc R Soc Med
changes in patients with chronic low back pain: a systematic review. 35(3):220221
Spine (Phila Pa 1976) 36(21 Suppl):S43S53. doi:10.1097/ Eriksson J, Jablonski A, Persson AK, Hao JX, Kouya PF, Wiesenfeld-
BRS.0b013e31822ef700 Hallin Z, Xu XJ, Fried K (2005) Behavioral changes and trigeminal
Cohen SP, Bogduk N, Dragovich A, Buckenmaier CC 3rd, Griffith S, ganglion sodium channel regulation in an orofacial neuropathic pain
Kurihara C, Raymond J, Richter PJ, Williams N, Yaksh TL (2009) model. Pain 119(13):8294
Randomized, double-blind, placebo-controlled, doseresponse, and Fagan A, Moore R, Vernon Roberts B, Blumbergs P, Fraser R (2003)
preclinical safety study of transforaminal epidural etanercept for the ISSLS prize winner: the innervation of the intervertebral disc:
treatment of sciatica. Anesthesiology 110(5):11161126. doi:10.1097/ a quantitative analysis. Spine (Phila Pa 1976) 28(23):25702576.
ALN.0b013e3181a05aa0 doi:10.1097/01.brs.0000096942.29660.b1
Colasanti B, Khazan N (1975) Imipramine-induced changes in the rapid Frank JW, Kerr MS, Brooker AS, DeMaio SE, Maetzel A, Shannon HS,
eye movement sleep rebound and wet dog shakes of morphine- Sullivan TJ, Norman RW, Wells RP (1996) Disability resulting from
abstinent rats. Neuropharmacology 14(56):361367 occupational low back pain. Part I: what do we know about primary
Coppes MH, Marani E, Thomeer RT, Groen GJ (1997) Innervation of prevention? A review of the scientific evidence on prevention before
painful lumbar discs. Spine 22(20):23422349; discussion disability begins. Spine (Phila Pa 1976) 21(24):29082917
23492350 Freemont AJ, Peacock TE, Goupille P, Hoyland JA, OBrien J, Jayson
Costigan M, Scholz J, Woolf CJ (2009) Neuropathic pain: a maladap- MI (1997) Nerve ingrowth into diseased intervertebral disc in
tive response of the nervous system to damage. Annu Rev Neurosci chronic back pain. Lancet 350(9072):178181
32:132. doi:10.1146/annurev.neuro.051508.135531 Freemont AJ, Watkins A, Le Maitre C, Baird P, Jeziorska M, Knight
Coutaux A, Adam F, Willer JC, Le Bars D (2005) Hyperalgesia and MT, Ross ER, OBrien JP, Hoyland JA (2002) Nerve growth factor
allodynia: peripheral mechanisms. Joint Bone Spine 72(5):359371. expression and innervation of the painful intervertebral disc.
doi:10.1016/j.jbspin.2004.01.010 J Pathol 197(3):286292. doi:10.1002/path.1108
Crawley JN, Moody TW (1983) Anxiolytics block excessive grooming Friberg S (1948) Anatomical studies on lumbar disc degeneration. Acta
behavior induced by ACTH1-24 and bombesin. Brain Res Bull Orthop Scand 17(34):224230
10(3):399401 Friberg S, Hirsch C (1949) Anatomical and clinical studies on lumbar
Dechant G, Barde YA (1997) Signalling through the neurotrophin disc degeneration. Acta Orthop Scand 19(2):222242, illust
receptor p75NTR. Curr Opin Neurobiol 7(3):413418 Frymoyer JW, Cats-Baril WL (1991) An overview of the incidences
Della-Giustina D, Kilcline BA (2000) Acute low back pain: a compre- and costs of low back pain. Orthop Clin North Am 22(2):263271
hensive review. Compr Ther 26(3):153159 Fujiwara A, Lim TH, An HS, Tanaka N, Jeon CH, Andersson GB,
Derby R, Kim BJ, Lee SH, Chen Y, Seo KS, Aprill C (2005) Comparison Haughton VM (2000a) The effect of disc degeneration and facet
of discographic findings in asymptomatic subject discs and the neg- joint osteoarthritis on the segmental flexibility of the lumbar spine.
ative discs of chronic LBP patients: can discography distinguish Spine (Phila Pa 1976) 25(23):30363044
asymptomatic discs among morphologically abnormal discs? Spine Fujiwara A, Tamai K, An HS, Kurihashi T, Lim TH, Yoshida H,
J 5(4):389394 Saotome K (2000b) The relationship between disc degeneration,
Deschamps K, Couture R (2005) The ventral tegmental area as a putative facet joint osteoarthritis, and stability of the degenerative lumbar
target for tachykinins in cardiovascular regulation. Br J Pharmacol spine. J Spinal Disord 13(5):444450
145(6):712727 Garcia-Cabrera I, Berge OG (1990) Interaction of high pressure and a
Deseure KR, Adriaensen HF (2002) Comparison between two types of narcotic dose of ethanol on spontaneous behavior in rats. Pharmacol
behavioral variables of non-evoked facial pain after chronic constric- Biochem Behav 37(3):577581
tion injury to the rat infraorbital nerve. Comp Med 52(1):4449 Garcia-Cosamalon J, del Valle ME, Calavia MG, Garcia-Suarez O,
Devor M (1991) Chronic pain in the aged: possible relation between Lopez-Muniz A, Otero J, Vega JA (2010) Intervertebral disc, sen-
neurogenesis, involution and pathophysiology in adult sensory gan- sory nerves and neurotrophins: who is who in discogenic pain?
glia. J Basic Clin Physiol Pharmacol 2(12):115 J Anat 217(1):115. doi:10.1111/j.1469-7580.2010.01227.x
Devor M (2006) Sodium channels and mechanisms of neuropathic pain. Ghomley RK (1993) Low back pain with special reference to the articu-
J Pain 7(1 Suppl 1):S3S12. doi:10.1016/j.jpain.2005.09.006 lar facets with presentation of an operative procedure. JAMA
Devor M, Wall PD (1990) Cross-excitation in dorsal root ganglia of 101:17731777
nerve-injured and intact rats. J Neurophysiol 64(6):17331746 Gibson JN, Waddell G (2005) Surgery for degenerative lumbar spondylo-
Diamant B, Karlsson J, Nachemson A (1968) Correlation between lac- sis.CochraneDatabaseSystRev(4):CD001352.doi:10.1002/14651858.
tate levels and pH in discs of patients with lumbar rhizopathies. CD001352.pub3
Experientia 24(12):11951196 Giles LG, Taylor JR (1987) Innervation of lumbar zygapophyseal joint
Dimitroulias A, Tsonidis C, Natsis K, Venizelos I, Djau SN, Tsitsopoulos synovial folds. Acta Orthop Scand 58(1):4346
P (2010) An immunohistochemical study of mechanoreceptors in Glassman SD, Carreon LY, Djurasovic M, Dimar JR, Johnson JR, Puno
lumbar spine intervertebral discs. J Clin Neurosci 17(6):742745. RM, Campbell MJ (2009) Lumbar fusion outcomes stratified by
doi:10.1016/j.jocn.2009.09.032 specific diagnostic indication. Spine J 9(1):1321. doi:10.1016/j.
Dray A (1995) Inflammatory mediators of pain. Br J Anaesth 75(2): spinee.2008.08.011
125131 Goupille P, Avimadje AM, Zerkak D, Valat JP (2000) Clinical elements
Ebendal T (1992) Function and evolution in the NGF family and its recep- of the diagnostic guidelines for low back pain. Rev Prat
tors. J Neurosci Res 32(4):461470. doi:10.1002/jnr.490320402 50(16):17601764
Edgar MA, Ghadially JA (1976) Innervation of the lumbar spine. Clin Groen GJ, Baljet B, Drukker J (1988) The innervation of the spinal dura
Orthop Relat Res 115:3541 mater: anatomy and clinical implications. Acta Neurochir (Wien)
Ehrenhaft JL (1943) Development of the vertebral column as related to cer- 92(14):3946
tain congenital and pathological changes. Surg, Gyn Obstr 76:282292 Grunhagen T, Shirazi-Adl A, Fairbank JC, Urban JP (2011) Intervertebral
Eisenstein SM, Parry CR (1987) The lumbar facet arthrosis syndrome. disk nutrition: a review of factors influencing concentrations of
Clinical presentation and articular surface changes. J Bone Joint nutrients and metabolites. Orthop Clin North Am 42(4):465477.
Surg Br 69(1):37 doi:10.1016/j.ocl.2011.07.010
272 K. Olmarker
Han HC, Lee DH, Chung JM (2000) Characteristics of ectopic dis- Keshari KR, Lotz JC, Link TM, Hu S, Majumdar S, Kurhanewicz J
charges in a rat neuropathic pain model. Pain 84(23):253261 (2008) Lactic acid and proteoglycans as metabolic markers for dis-
Hasue M (1993) Pain and the nerve root. An interdisciplinary approach. cogenic back pain. Spine (Phila Pa 1976) 33(3):312317.
Spine (Phila Pa 1976) 18(14):20532058 doi:10.1097/BRS.0b013e31816201c3
Henschke N, Maher CG, Refshauge KM, Herbert RD, Cumming RG, Kim CW, Perry A, Garfin SR (2005) Spinal instability: the orthope-
Bleasel J, York J, Das A, McAuley JH (2009) Prevalence of and dic approach. Semin Musculoskelet Radiol 9(1):7787.
screening for serious spinal pathology in patients presenting to pri- doi:10.1055/s-2005-867098
mary care settings with acute low back pain. Arthritis Rheum Kirkaldy-Willis WH, Farfan HF (1982) Instability of the lumbar spine.
60(10):30723080. doi:10.1002/art.24853 Clin Orthop Relat Res 165:110123
Hilton RC, Ball J, Benn RT (1980) Annular tears in the dorsolumbar Kitamura Y, Kitagawa K, Fujitani Y, Shibata K, Araki H, Sendou T,
spine. Ann Rheum Dis 39(6):533538 Gomita Y (2007) The 5-HT(1A) receptor full agonist, 8-OH-DPAT
Hirsch C (1959) Studies on the pathology of low back pain. J Bone inhibits ACTH-induced 5-HT(2A) receptor hyperfunction in rats:
Joint Surg Br 41-B(2):237243 involvement of 5-HT(1A) receptors in the DOI-induced wet-dog
Hirsch C, Ingelmark BE, Miller M (1963) The anatomical basis for low shakes in ACTH-treated rats. Biol Pharm Bull 30(1):117120
back pain. Studies on the presence of sensory nerve endings in liga- Kjaer P, Leboeuf-Yde C, Korsholm L, Sorensen JS, Bendix T (2005)
mentous, capsular and intervertebral disc structures in the human Magnetic resonance imaging and low back pain in adults: a diagnos-
lumbar spine. Acta Orthop Scand 33:117 tic imaging study of 40-year-old men and women. Spine (Phila Pa
Holm S, Selstam G, Nachemson A (1982) Carbohydrate metabolism 1976) 30(10):11731180
and concentration profiles of solutes in the canine lumbar interver- Kong MH, Morishita Y, He W, Miyazaki M, Zhang H, Wu G, Hymanson
tebral disc. Acta Physiol Scand 115(1):147156 HJ, Wang JC (2009) Lumbar segmental mobility according to the
Horner HA, Urban JP (2001) 2001 Volvo Award Winner in Basic grade of the disc, the facet joint, the muscle, and the ligament pathol-
Science Studies: effect of nutrient supply on the viability of cells ogy by using kinetic magnetic resonance imaging. Spine (Phila Pa
from the nucleus pulposus of the intervertebral disc. Spine (Phila Pa 1976) 34(23):25372544. doi:10.1097/BRS.0b013e3181b353ea
1976) 26(23):25432549 Kuslich SD, Ulstrom CL, Michael CJ (1991) The tissue origin of low
Horowitz GP, Allan AM (1982) Morphine withdrawal in mice selec- back pain and sciatica: a report of pain response to tissue stimula-
tively bred for differential sensitivity to ethanol. Pharmacol Biochem tion during operations on the lumbar spine using local anesthesia.
Behav 16(1):3539 Orthop Clin North Am 22(2):181187
Hsu EW, Hedlund LW, MacFall JR (1998) Functional MRI of the rat Lessmann V (1998) Neurotrophin-dependent modulation of gluta-
somatosensory cortex: effects of hyperventilation. Magn Reson matergic synaptic transmission in the mammalian CNS. Gen
Med 40(3):421426 Pharmacol 31(5):667674
Igarashi A, Kikuchi S, Konno S, Olmarker K (2004) Inflammatory Lindblom K (1948) Diagnostic puncture of intervertebral disks in sci-
cytokines released from the facet joint tissue in degenerative lumbar atica. Acta Orthop Scand 17(34):231239
spinal disorders. Spine (Phila Pa 1976) 29(19):20912095 Lisney SJ, Devor M (1987) Afterdischarge and interactions among
Inoue N, Espinoza Orias AA (2011) Biomechanics of intervertebral fibers in damaged peripheral nerve in the rat. Brain Res 415(1):
disk degeneration. Orthop Clin North Am 42(4):487499. 122136
doi:10.1016/j.ocl.2011.07.001 Liu BP, Strittmatter SM (2001) Semaphorin-mediated axonal guidance
Jensen MC, Brant-Zawadzki MN, Obuchowski N, Modic MT, via Rho-related G proteins. Curr Opin Cell Biol 13(5):619626
Malkasian D, Ross JS (1994) Magnetic resonance imaging of the Lu B, Pang PT, Woo NH (2005) The yin and yang of neurotrophin
lumbar spine in people without back pain. N Engl J Med 331(2):69 action. Nat Rev Neurosci 6(8):603614. doi:10.1038/nrn1726
73. doi:10.1056/nejm199407143310201 Luschka H (1850) Die Nerven des menschlichen Wirbelkanales. Laupp
Jensen TS, Karppinen J, Sorensen JS, Niinimaki J, Leboeuf-Yde C und Siebeck, Tbingen
(2008) Vertebral endplate signal changes (Modic change): a system- Lutz GK, Butzlaff M, Schultz-Venrath U (2003) Looking back on back
atic literature review of prevalence and association with non-specific pain: trial and error of diagnoses in the 20th century. Spine (Phila Pa
low back pain. Eur Spine J 17(11):14071422. doi:10.1007/s00586- 1976) 28(16):18991905. doi:10.1097/01.brs.0000083365.41261.cf
008-0770-2 Malinsky J (1959) The ontogenetic development of nerve terminations
Johnson WE, Evans H, Menage J, Eisenstein SM, El Haj A, Roberts S in the intervertebral discs of man. (Histology of intervertebral discs,
(2001) Immunohistochemical detection of Schwann cells in inner- 11th communication). Acta Anat (Basel) 38:96113
vated and vascularized human intervertebral discs. Spine (Phila Pa Manchikanti L, Manchikanti KN, Manchukonda R, Cash KA,
1976) 26(23):25502557 Damron KS, Pampati V, McManus CD (2007) Evaluation of
Johnson WE, Patterson AM, Eisenstein SM, Roberts S (2007) The pres- lumbar facet joint nerve blocks in the management of chronic
ence of pleiotrophin in the human intervertebral disc is associated low back pain: preliminary report of a randomized, double-blind
with increased vascularization: an immunohistologic study. Spine controlled trial: clinical trial NCT00355914. Pain Physician
32(12):12951302 10(3):425440
Jung A, Brunschwig A (1932) Rescherches histologique des articula- Manchikanti L, Singh V, Falco FJ, Cash KA, Pampati V (2008) Lumbar
tions des corps vertebeaux. La presse Medicale 40:316317 facet joint nerve blocks in managing chronic facet joint pain: one-
Kanayama M, Togawa D, Takahashi C, Terai T, Hashimoto T (2009) year follow-up of a randomized, double-blind controlled trial: clini-
Cross-sectional magnetic resonance imaging study of lumbar disc cal trial NCT00355914. Pain Physician 11(2):121132
degeneration in 200 healthy individuals. J Neurosurg Spine Martin WR, McNicholas LF, Cherian S (1982) Diazepam and pentobar-
11(4):501507. doi:10.3171/2009.5.spine08675 bital dependence in the rat. Life Sci 31(8):721730
Kawakami M, Weinstein JN, Chatani K, Spratt KF, Meller ST, Gebhart Masini M, Paiva WS, Araujo AS Jr (2005) Anatomical description
GF (1994) Experimental lumbar radiculopathy. Behavioral and his- of the facet joint innervation and its implication in the treat-
tologic changes in a model of radicular pain after spinal nerve root ment of recurrent back pain. J Neurosurg Sci 49(4):143146;
irritation with chromic gut ligatures in the rat. Spine 19(16): discussion 146
17951802 McCarthy PW, Carruthers B, Martin D, Petts P (1991) Immunohistochemical
Kelsey JL, White AA 3rd (1980) Epidemiology and impact of low-back demonstration of sensory nerve fibers and endings in lumbar interver-
pain. Spine (Phila Pa 1976) 5(2):133142 tebral discs of the rat. Spine 16(6):653655
McLain RF (1993) Mechanoreceptor endings in human cervical facet Olsen AK (1950) Disc degeneration: a distinct clinical entity. J Nerv
joints. Iowa Orthop J 13:149154 Ment Dis 112(3):262263
Millan MJ, Emrich HM (1981) Endorphinergic systems and the Orita S, Eguchi Y, Kamoda H, Arai G, Ishikawa T, Miyagi M, Inoue G,
response to stress. Psychother Psychosom 36(1):4356 Suzuki M, Toyone T, Aoki Y, Takahashi K, Ohtori S (2011) Brain-
Moneta GB, Videman T, Kaivanto K, Aprill C, Spivey M, Vanharanta derived neurotrophic factor inhibition at the punctured interverte-
H, Sachs BL, Guyer RD, Hochschuler SH, Raschbaum RF et al bral disc downregulates the production of calcitonin gene-related
(1994) Reported pain during lumbar discography as a function of peptide in dorsal root ganglia in rats. Spine (Phila Pa 1976)
anular ruptures and disc degeneration. A re-analysis of 833 disco- 36(21):17371743. doi:10.1097/BRS.0b013e31821d7b9f
grams. Spine (Phila Pa 1976) 19(17):19681974 Osti OL, Vernon-Roberts B, Moore R, Fraser RD (1992) Annular tears
Monti JM, Carlini EA (1975) Spontaneous behavior and sleep- and disc degeneration in the lumbar spine. A post-mortem study of
wakefulness cycle in isolated and paired REM sleep deprived- 135 discs. J Bone Joint Surg Br 74(5):678682
marihuana treated rats. Pharmacol Biochem Behav 3(6):10251029 Ozaktay AC, Cavanaugh JM, Blagoev DC, Getchell TV, King AI (1994)
Mooney V, Robertson J (1976) The facet syndrome. Clin Orthop Relat Effects of a carrageenan-induced inflammation in rabbit lumbar facet
Res 115:149156 joint capsule and adjacent tissues. Neurosci Res 20(4):355364
Morgan FP, King T (1957) Primary instability of lumbar vertebrae as a Palmgren T, Gronblad M, Virri J, Kaapa E, Karaharju E (1999) An
common cause of low back pain. J Bone Joint Surg Br 39-B(1):622 immunohistochemical study of nerve structures in the anulus
Nachemson A (1969) Intradiscal measurements of pH in patients with fibrosus of human normal lumbar intervertebral discs. Spine
lumbar rhizopathies. Acta Orthop Scand 40(1):2342 24(20):20752079
Nachemson A, Jonsson E (2000) Ont i ryggen, ont i nacken. En evi- Papir-Kricheli D, Frey J, Laufer R, Gilon C, Chorev M, Selinger Z,
densbaserad kunskapssammanstllningom orsaker, diagnostik och Devor M (1987) Behavioural effects of receptor-specific substance
Behandling. SBU Statens Berednings fr medicinsk Utvrdering, P agonists. Pain 31(2):263276
Stockholm, Sweden Pattrick M, Hamilton E, Wilson R, Austin S, Doherty M (1993)
Nakamae T, Ochi M, Olmarker K (2011) Pharmacological inhibition of Association of radiographic changes of osteoarthritis, symptoms,
tumor necrosis factor may reduce pain behavior changes induced by and synovial fluid particles in 300 knees. Ann Rheum Dis
experimental disc puncture in the rat: an experimental study in rats. 52(2):97103
Spine 36(15):232236. doi:10.1097/BRS.0b013e3181d8bef3 Pedersen HE, Blunck CF, Gardner E (1956) The anatomy of lum-
Nakamura S, Takahashi K, Takahashi Y, Morinaga T, Shimada Y, bosacral posterior rami and meningeal branches of spinal nerve
Moriya H (1996) Origin of nerves supplying the posterior portion of (sinu-vertebral nerves); with an experimental study of their func-
lumbar intervertebral discs in rats. Spine (Phila Pa 1976) 21(8): tions. J Bone Joint Surg Am 38-A(2):377391
917924 Peng B, Wu W, Hou S, Li P, Zhang C, Yang Y (2005) The pathogenesis
Nakamura F, Kalb RG, Strittmatter SM (2000) Molecular basis of of discogenic low back pain. J Bone Joint Surg Br 87(1):6267
semaphorin-mediated axon guidance. J Neurobiol 44(2):219229 Peng B, Hao J, Hou S, Wu W, Jiang D, Fu X, Yang Y (2006a) Possible
Newman Dorland WA (2007) Dorlands illustrated medical dictionary, pathogenesis of painful intervertebral disc degeneration. Spine (Phila
31st edn. Saunders/Elsevier, Philadelphia Pa 1976) 31(5):560566. doi:10.1097/01.brs.0000201324.45537.46
Nishida K, Gilbertson LG, Robbins PD, Evans CH, Kang JD (2000) Peng B, Hou S, Wu W, Zhang C, Yang Y (2006b) The pathogenesis and
Potential applications of gene therapy to the treatment of interverte- clinical significance of a high-intensity zone (HIZ) of lumbar inter-
bral disc disorders. Clin Orthop Relat Res (379 Suppl):S234S241 vertebral disc on MR imaging in the patient with discogenic low
Ohtori S, Inoue G, Ito T, Koshi T, Ozawa T, Doya H, Saito T, Moriya H, back pain. Eur Spine J 15(5):583587
Takahashi K (2006) Tumor necrosis factor-immunoreactive cells Pope MH, Panjabi M (1985) Biomechanical definitions of spinal insta-
and PGP 9.5-immunoreactive nerve fibers in vertebral endplates of bility. Spine (Phila Pa 1976) 10(3):255256
patients with discogenic low back pain and Modic Type 1 or Type 2 Purmessur D, Freemont AJ, Hoyland JA (2008) Expression and regula-
changes on MRI. Spine (Phila Pa 1976) 31(9):10261031. tion of neurotrophins in the nondegenerate and degenerate human
doi:10.1097/01.brs.0000215027.87102.7c intervertebral disc. Arthritis Res Ther 10(4):R99. doi:10.1186/
Olmarker K (2005) Neovascularization and neoinnervation of subcuta- ar2487
neously placed nucleus pulposus and the inhibitory effects of cer- Rajasekaran S, Vidyadhara S, Subbiah M, Kamath V, Karunanithi R,
tain drugs. Spine 30(13):15011504 Shetty AP, Venkateswaran K, Babu M, Meenakshi J (2010) ISSLS
Olmarker K (2008) Puncture of a lumbar intervertebral disc induces prize winner: a study of effects of in vivo mechanical forces on
changes in spontaneous pain behavior: an experimental study in human lumbar discs with scoliotic disc as a biological model: results
rats. Spine 33(8):850855 from serial postcontrast diffusion studies, histopathology and bio-
Olmarker K (2011) Puncture of a disc and application of nucleus pulpo- chemical analysis of twenty-one human lumbar scoliotic discs.
sus induces disc herniation-like changes and osteophytes. An exper- Spine (Phila Pa 1976) 35(21):19301943. doi:10.1097/BRS.
imental study in rats. Open Orthop J 5:154159 0b013e3181e9a156
Olmarker K, Myers RR (1998) Pathogenesis of sciatic pain: role of her- Rang HP, Bevan S, Dray A (1991) Chemical activation of nociceptive
niated nucleus pulposus and deformation of spinal nerve root and peripheral neurones. Br Med Bull 47(3):534548
dorsal root ganglion. Pain 78(2):99105 Raoul S, Faure A, Robert R, Rogez JM, Hamel O, Cuillere P, Le Borgne
Olmarker K, Iwabuchi M, Larsson K, Rydevik B (1998) Walking analy- J (2003) Role of the sinu-vertebral nerve in low back pain and ana-
sis of rats subjected to experimental disc herniation. Eur Spine J tomical basis of therapeutic implications. Surg Radiol Anat
7(5):394399 24(6):366371. doi:10.1007/s00276-002-0084-8
Olmarker K, Storkson R, Berge OG (2002) Pathogenesis of sciatic pain: Rasminsky M (1987) Spontaneous activity and cross-talk in pathologi-
a study of spontaneous behavior in rats exposed to experimental cal nerve fibers. Res Publ Assoc Res Nerv Ment Dis 65:3949
disc herniation. Spine 27(12):13121317 Risbud MV, Shapiro IM, Vaccaro AR, Albert TJ (2004) Stem cell
Olmarker K, Nutu M, Storkson R (2003) Changes in spontaneous regeneration of the nucleus pulposus. Spine J 4(6 Suppl):348S
behavior in rats exposed to experimental disc herniation are blocked 353S. doi:10.1016/j.spinee.2004.07.031
by selective TNF-alpha inhibition. Spine 28(15):16351641; dis- Roberts S (2002) Disc morphology in health and disease. Biochem Soc
cussion 1642 Trans 30(Pt 6):864869. doi:10.1042/BST0300864
274 K. Olmarker
Roberts S, Eisenstein SM, Menage J, Evans EH, Ashton IK (1995) Takahashi Y, Ohtori S, Takahashi K (2009) Peripheral nerve pathways
Mechanoreceptors in intervertebral discs. Morphology, distribution, of afferent fibers innervating the lumbar spine in rats. J Pain
and neuropeptides. Spine (Phila Pa 1976) 20(24):26452651 10(4):416425. doi:10.1016/j.jpain.2008.10.012
Rodriquez-Enchandia EL, Broitman ST, Foscolo M, Gonzales A (1986) Takahashi Y, Ohtori S, Takahashi K (2010) Dorsoventral organization
Gamma-vinyl-GABA injection in the locus coeruleus region of the of sensory nerves in the lumbar spine as indicated by double label-
rat. Effect on spontaneous behavior and stress responses. Adv ing of dorsal root ganglion neurons. J Orthop Sci 15(4):578583.
Biochem Psychopharmacol 42:93102 doi:10.1007/s00776-010-1482-0
Rohm B, Ottemeyer A, Lohrum M, Puschel AW (2000) Plexin/neuropi- Takatalo J, Karppinen J, Niinimaki J, Taimela S, Nayha S, Jarvelin MR,
lin complexes mediate repulsion by the axonal guidance signal Kyllonen E, Tervonen O (2009) Prevalence of degenerative imaging
semaphorin 3A. Mech Dev 93(12):95104 findings in lumbar magnetic resonance imaging among young
Roofe PG (1940) Innervation of annulus fibrosusand posterior longitu- adults. Spine (Phila Pa 1976) 34(16):17161721. doi:10.1097/
dinal ligament. Arch Neurol Psychiatry 44:110113 BRS.0b013e3181ac5fec
Rose CP, Cockshott WP (1982) Anterior femoral erosion and patello- Takebayashi T, Cavanaugh JM, Kallakuri S, Chen C, Yamashita T
femoral osteoarthritis. J Can Assoc Radiol 33(1):3234 (2006) Sympathetic afferent units from lumbar intervertebral discs.
Ross JS, Modic MT, Masaryk TJ (1989) Tears of the anulus fibrosus: J Bone Joint Surg Br 88(4):554557. doi:10.1302/0301-
assessment with Gd-DTPA-enhanced MR imaging. AJNR Am J 620X.88B4.17194, 88-B/4/554 [pii]
Neuroradiol 10(6):12511254 Thompson KJ, Dagher AP, Eckel TS, Clark M, Reinig JW (2009)
Saifuddin A, Mitchell R, Taylor BA (1999) Extradural inflammation Modic changes on MR images as studied with provocative diskog-
associated with annular tears: demonstration with gadolinium- raphy: clinical relevance a retrospective study of 2457 disks.
enhanced lumbar spine MRI. Eur Spine J 8(1):3439 Radiology 250(3):849855. doi:10.1148/radiol.2503080474
Sakai D, Mochida J, Yamamoto Y, Nomura T, Okuma M, Nishimura K, Tolofari SK, Richardson SM, Freemont AJ, Hoyland JA (2010)
Nakai T, Ando K, Hotta T (2003) Transplantation of mesenchymal Expression of semaphorin 3A and its receptors in the human inter-
stem cells embedded in Atelocollagen gel to the intervertebral disc: vertebral disc: potential role in regulating neural ingrowth in the
a potential therapeutic model for disc degeneration. Biomaterials degenerate intervertebral disc. Arthritis Res Ther 12(1):R1.
24(20):35313541 doi:10.1186/ar2898
Samartzis D, Karppinen J, Mok F, Fong DY, Luk KD, Cheung KM Treptow K, Bondarenko NA, Waldmann AW, Oehme P (1986) The
(2011) A population-based study of juvenile disc degeneration and effect of substance P on the swimming behavior of rats under stress-
its association with overweight and obesity, low back pain, and ing and nonstressing conditions. Pharmazie 41(8):578580
diminished functional status. J Bone Joint Surg Am 93(7):662670. Twomey L, Taylor J (1985) Age changes in lumbar intervertebral discs.
doi:10.2106/jbjs.i.01568 Acta Orthop Scand 56(6):496499
Schellhas KP, Pollei SR, Gundry CR, Heithoff KB (1996) Lumbar disc van Roy P, Barbaix E, Clarijs JP, Mense S (2001) Anatomical back-
high-intensity zone. Correlation of magnetic resonance imaging and ground of low back pain: variability and degeneration of the lumbar
discography. Spine 21(1):7986 spinal canal and intervertebral disc. Schmerz 15(6):418424.
Schmorl G (1929) ber knorpelknoten an der Hinterflsche der doi:10.1007/s004820100026
Wirbelbandscheiben. Fortschr Geb Rontgenstr 40:629634 Videman T, Nurminen M (2004) The occurrence of anular tears and
Shao Z, Rompe G, Schiltenwolf M (2002) Radiographic changes in the their relation to lifetime back pain history: a cadaveric study using
lumbar intervertebral discs and lumbar vertebrae with age. Spine barium sulfate discography. Spine 29(23):26682676
(Phila Pa 1976) 27(3):263268 Vos BP, Strassman AM, Maciewicz RJ (1994) Behavioral evidence of
Skaper SD (2008) The biology of neurotrophins, signalling pathways, trigeminal neuropathic pain following chronic constriction injury to
and functional peptide mimetics of neurotrophins and their recep- the rats infraorbital nerve. J Neurosci 14(5 Pt 1):27082723
tors. CNS Neurol Disord Drug Targets 7(1):4662 Wang ZX, Hu YG, Chen XM (2008) Significance of the high-intensity
Smith L (1964) Enzyme dissolution of the nucleus pulposus in humans. zone located in the posterior annulus fibrosus for diagnosing disco-
JAMA 187:137140 genic low back pain. Zhonghua Yi Xue Za Zhi 88(35):24782481
Stadnik TW, Lee RR, Coen HL, Neirynck EC, Buisseret TS, Osteaux Weber R, Ramos-Cabrer P, Wiedermann D, van Camp N, Hoehn M
MJ (1998) Annular tears and disk herniation: prevalence and con- (2006) A fully noninvasive and robust experimental protocol for
trast enhancement on MR images in the absence of low back pain or longitudinal fMRI studies in the rat. Neuroimage 29(4):13031310.
sciatica. Radiology 206(1):4955 doi:10.1016/j.neuroimage.2005.08.028
Staiger TO, Paauw DS, Deyo RA, Jarvik JG (1999) Imaging studies for Wehling P, Schulitz KP, Robbins PD, Evans CH, Reinecke JA (1997)
acute low back pain. When and when not to order them. Postgrad Transfer of genes to chondrocytic cells of the lumbar spine. Proposal
Med 105(4):161162, 165166, 171162 for a treatment strategy of spinal disorders by local gene therapy.
Stout A (2010) Discography. Phys Med Rehabil Clin N Am 21(4):859 Spine (Phila Pa 1976) 22(10):10921097
867. doi:10.1016/j.pmr.2010.07.002 Weinstein J, Claverie W, Gibson S (1988) The pain of discography.
Sugiura A, Ohtori S, Yamashita M, Inoue G, Yamauchi K, Koshi T, Spine (Phila Pa 1976) 13(12):13441348
Suzuki M, Norimoto M, Orita S, Eguchi Y, Takahashi Y, Watanabe Weishaupt D, Zanetti M, Hodler J, Boos N (1998) MR imaging of the
TS, Ochiai N, Takaso M, Takahashi K (2008) Existence of nerve lumbar spine: prevalence of intervertebral disk extrusion and
growth factor receptors, tyrosine kinase a and p75 neurotrophin sequestration, nerve root compression, end plate abnormalities, and
receptors in intervertebral discs and on dorsal root ganglion neurons osteoarthritis of the facet joints in asymptomatic volunteers.
innervating intervertebral discs in rats. Spine (Phila Pa 1976) Radiology 209(3):661666
33(19):20472051. doi:10.1097/BRS.0b013e31817f8d58 Willburger RE, Wittenberg RH (1994) Prostaglandin release from lum-
Sward L, Hellstrom M, Jacobsson B, Peterson L (1990) Back pain and bar disc and facet joint tissue. Spine (Phila Pa 1976) 19(18):
radiologic changes in the thoraco-lumbar spine of athletes. Spine 20682070
(Phila Pa 1976) 15(2):124129 Wuttke W, Hoffmeister F (1968) Changes in the spontaneous behavior
Takahashi K, Miyazaki T, Ohnari H, Takino T, Tomita K (1995) of cats induced by chlordiazepoxide and chlorpromazine. Naunyn
Schmorls nodes and low-back pain. Analysis of magnetic resonance Schmiedebergs Arch Exp Pathol Pharmakol 260(2):221222
imaging findings in symptomatic and asymptomatic individuals. Yamada H, Honda T, Yaginuma H, Kikuchi S, Sugiura Y (2001)
Eur Spine J 4(1):5659 Comparison of sensory and sympathetic innervation of the dura
mater and posterior longitudinal ligament in the cervical spine after Zhang N, Li FC, Huang YJ, Teng C, Chen WS (2010) Possible key role
removal of the stellate ganglion. J Comp Neurol 434(1):86100 of immune system in Schmorls nodes. Med Hypotheses 74(3):552
Yoon ST, Park JS, Kim KS, Li J, Attallah-Wasif ES, Hutton WC, Boden 554. doi:10.1016/j.mehy.2009.09.044
SD (2004) ISSLS prize winner: LMP-1 upregulates intervertebral Zimmermann M (1984) Basic concepts of pain and pain therapy.
disc cell production of proteoglycans and BMPs in vitro and in vivo. Arzneimittelforschung 34(9A):10531059
Spine (Phila Pa 1976) 29(23):26032611
Yoshizawa H, OBrien JP, Smith WT, Trumper M (1980) The neuropa-
thology of intervertebral discs removed for low-back pain. J Pathol
132(2):95104
Clinical Features and Pathobiology
of Chordoma 17
John A. Abraham, Brian Neuman, and Francis J. Hornicek
Contents 17.1 Introduction and Historical Perspective

17.1 Introduction and Historical Perspective ..................... 277
Chordoma is the most common primary malignant bone
17.2 Epidemiology ................................................................. 278
tumor found in the spine and sacrum. While these tumors
17.3 Clinical Features............................................................ 278 are relatively slow growing, they do have the potential to
17.4 Tumor Imaging .............................................................. 280 recur and metastasize. Chordoma was described histologi-
17.5 Histopathology and Immunohistochemistry .............. 281
cally long before it was realized that they were probably
derived from notochordal precursor cells. Early reports
17.6 Differential Diagnosis.................................................... 282 dating back to Virchow in 1857 describe a vacuolated cell
17.7 Tumor Staging ............................................................... 282 type seen in these tumors. These cells were described as
17.8 Management .................................................................. 282 physaliferous, from Greek for having bubbles. It was
17.8.1 Surgery ............................................................................ 282 thought at that early time that they were a cartilaginous
17.8.2 Radiation ......................................................................... 283 tumor, which may have been a result of evaluation of a
17.8.3 Systemic Therapy and Future Directions ........................ 284
chondroid variant tumor. By 1923, Burrow and Stewart
17.9 Summary of Critical Concepts recognized that chordomas were a lowly malignant tumor
Discussed in This Chapter ............................................ 286 of slow growth, locally invasive and destructive, and only
References .................................................................................... 286 rarely giving rise to metastases. By then, the location at
either end of the spine correlated well with contemporary
descriptions of the location of vestigial notochordal rem-
nants by Muller. These observations led to the hypothesis
J.A. Abraham, MD (*)
that chordomas were not tumors of the intervertebral disc
Division of Orthopedic Oncology, Rothman Institute,
Philadelphia, PA, USA but rather malignant transformation of these notochordal
remnants. In 1858, Muller coined the current name by pro-
Department of Orthopedic Surgery and Radiation Oncology,
Thomas Jefferson University, posing the following hypothesis: A direct relation of these
925 Chestnut Street, Philadelphia, PA 19107, USA growths to the chorda dorsalis cannot be overlooked and I
e-mail: john.abraham@rothmaninstitute.com consider them to be excessively growing remnants of the
B. Neuman, MD chorda. Whosoever likes the name may designate these
Department of Orthopedic Surgery, masses as chordoid tumors, or chordomas. Since these
Thomas Jefferson University,
early descriptions, this concept has been supported by
925 Chestnut Street, Philadelphia, PA 19107, USA
e-mail: bneumy@gmail.com significant indirect evidence, although there is a paucity of
direct proof.
F.J. Hornicek, MD, PhD
Division of Orthopedic Oncology, The early treatment of chordoma focused on surgical
Center for Sarcoma and Connective Tissue Oncology, removal, but the difficulty in completely resecting these
Massachusetts General Hospital Cancer Center, tumors was very quickly recognized. Investigation of
Massachusetts General Hospital,
the addition of radiation to the treatment regimen with or
Boston, MA, USA
without surgery was described in the 1970s and continues to
Department of Orthopedic Surgery,
be investigated (Pearlman and Friedman 1970; Pearlman
Harvard Medical School,
55 Fruit Street, Boston, MA 02114, USA et al. 1972). Medical therapies failed to provide any
e-mail: fhornicek@partners.org significant benefit, and the search for an effective medical

278 J.A. Abraham et al.
agent continues today. This chapter provides a review of the with better survival. This observation was surprising as the
clinical features of chordoma and the current molecular Hispanic population was associated with low socioeconomic
understanding of its pathobiology. status, and this factor alone should have increased the risk of
death. Consistently, studies on gender distribution of chordo-
mas show a 2:1 male predominance (Ashwood et al. 1994;
17.2 Epidemiology Forsyth et al. 1993). The median age at the time of diagnosis
is 58.5 years, while diagnosis is rare in patients younger than
Chordoma is the most frequent primary bone tumor found in 30 years of age (McMaster et al. 2001; Weber and Sim 2002).
the spine. Nevertheless, these tumors are rare with the age- These tumors are very uncommon in pediatric populations
adjusted incidence rate of 0.08 per 100,000 (McMaster et al. comprising less than 5 % of all chordomas, with the majority
2001). Chordomas make up 14 % of all primary bone tumors of these being skull based (McMaster et al. 2001).
(Healey and Lane 1989; Unni 1996; Papagelopoulos et al.
2004). Approximately 50 % of these tumors are located
within the sacrum (Fig. 17.1). The remaining anatomic loca- 17.3 Clinical Features
tions are the skull-base, spheno-occipital region (35 %), and
mobile spine (15 %) (Bohlman et al. 1986; Bjornsson et al. Patients diagnosed with a chordoma most commonly present
1993; Bergh et al. 2000). The distribution of these tumors with pain, regardless of location (Bergh et al. 2000; Boriani
within the mobile spine was evaluated by Boriani et al. (2006) et al. 1996, 2006). The second most common presentation is
in a consecutive series over 50 years. This group demon- development of neurological symptoms and least commonly
strated the highest frequency of involvement in the lumbar a palpable mass (Bergh et al. 2000, Soo 2001). Symptom
spine (57.5 %), followed by the cervical region (29 %), and duration averages 2 years prior to diagnosis, highlighting the
least frequently in the thoracic spine (13.5 %). Chordomas slow-growing nature of these tumors (Bergh et al. 2000). If
comprise greater than 50 % of the primary bone tumors found untreated, pain can progress to the point of incapacitation,
in the sacrum (Boriani et al. 2006). In a recent analysis of 409 which in one study was found to be at approximately
patients identified utilizing the California Cancer Registry, 50 months from the onset of symptoms (Boriani et al. 2006).
the racial distribution was 65 % Caucasian, 23 % Hispanic, Up to 60 % of the time, chordomas can extend into the
10 % Asian or other, and 1.7 % African. In this study, in spinal canal and in some of these cases can cause significant
evaluating chordoma-specific survival, there was a neurologic symptoms, such as compressive myelopathy or
significantly decreased risk of death in Hispanics (Lee et al. cauda equina (Meyer et al. 1984) (Fig 17.2). Neurological
2012). In this group, chordomas were also found to be asso- symptoms are most commonly associated with chordomas of
ciated with younger age at diagnosis, cranial disease, and a
higher rate of surgical intervention, which were all associated
S149
Fig. 17.2 Large cervical spine chordoma at C2 with compression of

Fig. 17.1 Typical MRI appearance of large sacral chordoma spinal cord
17 Clinical Features and Pathobiology of Chordoma 279
the mobile spine. The spectrum of neurological symptoms throat irritation, dysphagia, esophageal compression, dys-
widely ranges depending on the location of the tumor. Severe phonia, or airway obstruction from compression of critical
compression of the spinal cord is a late complication leading local structures (Singh et al. 2007; Nicoucar et al. 2008).
to paralysis. Tumors which invade the neural foramina can Horner syndrome has been reported from lower cervical
cause a radicular pattern of symptoms, which includes weak- spine chordomas (Leone et al. 2002). Mass effect from of
ness and sensory deficits in the distribution of a particular sacral chordomas can cause compression and displacement
nerve root (Sundaresan et al. 1990; Mindell 1981). of the bladder or rectum leading to urinary stress inconti-
Non-neurologic symptoms from chordoma are most com- nence, constipation, or obstruction. Sacral chordoma of
monly due to local effects of the tumor mass. For example, sufficiently large size can be palpated on rectal exam
chordomas which develop in the cervical spine may cause (Fourney and Gokaslan 2003; Atalar et al. 2006).
Box 17.1 The Cauda Equina these critical nerve roots and is considered a surgical
Anatomically, the cauda equina, or horses tail, is the emergency.
collection of nerve roots that travels through the spinal Symptoms of cauda equina syndrome from any cause
canal beyond the termination of the spinal cord. The include weakness in the lower extremities, urinary reten-
spinal cord ends at approximately the L1 level in tion due to detrusor muscle weakness, loss of rectal tone
humans, at which point the lumbar (L15) and sacral due to anal sphincter weakness, and subsequent fecal
(S15) roots have already branched off the spinal cord. incontinence. Sexual dysfunction and saddle anesthesia
However, because these nerve roots exit the spinal canal and lower extremity pain may also be present. Lower
at successively more inferior levels in the lumbar and extremity reflexes are reduced or absent. The causes of
sacral bony spine, they travel together for a distance cauda equina syndrome are numerous and essentially
within the spinal canal as a bundle of nerve roots, which can be any inciting agent or problem which causes
is called the cauda equina. Injuries or other conditions pressure on the cauda equina. Commonly, acutely her-
which damage the cauda equina cause a specific niated discs can cause cauda equina syndrome. Other
constellation of symptoms related to dysfunction of degenerative spinal conditions such as spinal stenosis
Spinal cord
Conus medullaris
L1
L2
L3 Cauda equina
Lumbar spine
L4
L5
S1
Sacrum
S2
S3
The cauda equina is shown S4
schematically on the left, and the
S5
osseous anatomy of the lumbar
spine and sacrum is shown on
the right
or spondylolisthesis can contribute as well. Trauma is well as removal or correction of the inciting mechanical
another common cause, either by direct damage to nerve problem. In the case of degenerative and traumatic condi-
roots by fracture, dislocation, or penetrating trauma or tions, every effort is made to preserve the involved nerve
by hematoma secondary to the initial traumatic injury. roots. In the case of metastatic tumors, intralesional pro-
Tumors, such as chordoma or more commonly meta- cedures may be performed in conjunction with adjuvant
static disease, can also cause a cauda equina syndrome, therapies such as radiation. Sacral chordoma requires
although in this situation the presenting symptoms are wide resection, which usually involved resection of some
usually more chronic and develop over a longer period or all of the sacral nerve roots, so the expectation is that
of time as the tumor grows. the nerve deficits will not recover. It is important to dis-
Treatment of cauda equina syndrome is primarily cuss the specific expected deficits with any patient under-
surgical decompression of the involved nerve roots as going sacrectomy for chordoma.
Fig. 17.3 Coronal MRI image of sacral chordoma showing extensive Fig. 17.4 Typical MRI appearance of cervical spine chordoma, in this
calcifications, sometimes seen in chordoma case involving right-sided pedicle
17.4 Tumor Imaging sacrum. Intratumoral amorphous calcifications are seen on

plain radiographs within a chordoma in 5070 % of cases
Plain radiographs are often the initial image taken for a com- and in 90 % of cases on CT scan, but have no known prog-
plaint of back pain. However, detailed examination of the nostic significance (Fig 17.3). An associated large soft tissue
spine with plain imaging alone can be difficult. In particular, mass can also be seen on CT scan or MRI.
the obliquity of the sacrum and the overlying shadows from Evaluation of the central spinal canal and cord or nerve
bowel gas and contents can often limit its utility (Manaster root involvement is best done with MR imaging, and as such
and Graham 2003). Nonetheless, when evaluating the sacrum this modality is critical for evaluation of chordomas. MRI
on plain radiographs, there are particular features that should features of a chordoma on T1-weighted sequences consist of
be scrutinized: the paired sacral foramen should appear simi- an isointense or hypointense mass as compared to muscle
lar with distinct sacral boundaries outlining the foramen, the and on a T2-weighted sequences as a high-signal-intensity
anterior and posterior aspects of the sacroiliac joint should mass (Fig 17.4). If calcifications are present, they can appear
be distinct, and the posterior contour of the iliac wing should as areas of low signal intensity on T1- and T2-weighted
be seen underlying the sacral ala. Lack of any of these images. Chordomas enhance with gadolinium (Manaster and
findings on pelvic radiograph could suggest a lesion of the Graham 2003). Primary tumors and metastatic lesions both
show very high signal intensities on diffusion weighted been used to study chordoma treatment. This technology has
images, which may help distinguish metastatic nodules from shown some promise in pre- and posttreatment imaging of
nodules of other unrelated etiologies (Kishimoto et al. 2012). chordoma (Zhang et al. 2004).
Both MRI and CT scan can show bone destruction and exten-
sion of the tumor into the canal (Fig. 17.5). MR imaging can
be used to differentiate benign notochordal cell tumors 17.5 Histopathology and
(BNCTs) from chordomas based on the lack of gadolinium Immunohistochemistry
uptake, bone sclerosis, and completely intraosseous location
seen with BNCTs (Nishiguchi et al. 2011). Chordomas also Virchows original description of the physaliphorous cell,
demonstrate fluorodeoxyglucose avidity on F-18 PET scans a vacuolated cell clustered in sheets giving the appearance
(Lin et al. 2006; Miyazaway et al.2008; Park and Kim 2008). of soap bubbles and demonstrating a lobular pattern of
Carbon-11-methionine positron emission tomography (MET- growth, describes the classic histologic features of chordo-
PET), used to evaluate the effectiveness of carbon-ion radio- mas (Fig. 17.6). These cells have small round dark staining
therapy for assessment of rectal cancer and other tumors, has nuclei with few mitotic figures but demonstrate significant
atypia. The sheets of cells are separated by a fibrous sep-
tae with areas of calcification or hemorrhage representing
necrotic areas (Weber and Sim 2002). These tumors can
be separated into three classes: classical or conventional,
chondroid, and dedifferentiated (Chugh et al. 2007). The
classical form most commonly displays the typical physal-
iferous features. Chondroid type tumors may exhibit areas
of chondrosarcoma-like cartilage as well as features of clas-
sic chordoma. Dedifferentiated tumors may have a more
highly aggressive sarcomatous histological appearance.
Dedifferentiated tumors may display nuclear inclusions, bi- or
multinucleation, and sometimes mitotic figures (Crapanzano
et al. 2001). Immunohistochemical staining is useful in the
evaluation of chordomas. Significant immunoreactivity for
S-100, membrane antigen (MUC-1), and cytokeratin is seen
in these tumors.
Chondroid tumors are difficult to distinguish from chond-
rosarcomas, emphasizing the need for specific immunohis-
tochemical markers. Brachyury is a notochordal transcription
Fig. 17.5 Axial CT scan image showing extensive sacral bone destruc-
tion due to chordoma. Note that the tumor is a midline tumor, which can factor that is expressed in most sporadic chordomas, but not
differentiate it from other sacral tumors in chondrosarcomas (Vujovic et al. 2006). Interestingly, the
a b
Fig. 17.6 Photomicrograph of chordoma histology demonstrating vacuolated physaliferous cells. (a) Low power, 100; (b) high power, 400
T gene locus containing the brachyury gene has been found not approved as a first-line modality for this purpose. F-18
to be duplicated in rare familial instances of chordoma, sug- PET or MET-PET are potentially promising techniques as
gesting a critical role in the pathogenesis of this tumor (Yang described earlier.
et al. 2009). The addition of brachyury to the panel of bio-
markers used to identify this cell type has improved the sen-
sitivity and specificity for chordoma to 98 and 100 % (Oakley 17.8 Management
et al. 2008). Other important biomarkers such as ezrin,
MMP-9, and COX-2 have also been recently investigated 17.8.1 Surgery
and hold diagnostic promise (Froehlich et al. 2012). The next
step in the investigation of these markers is to study their The primary treatment of chordoma is wide surgical resec-
value as potential therapeutic targets. tion (Fig. 17.7) and the goal of surgery is wide excision with
negative margins. This is sometimes an unrealistic goal due
to the difficult locations of this tumor, especially clival or
17.6 Differential Diagnosis skull-base tumors, or high-level sacral tumors. Nonetheless,
multiple studies have demonstrated a correlation between
In the differential diagnosis of tumors of the spine, myeloma, recurrence rate and positive margins (Boriani et al. 2006;
plasmacytoma, benign notochordal cell tumor, lymphoma, Bilsky et al. 2004; Fuchs et al. 2005). Boriani et al. (2006)
osteomyelitis, giant cell tumor, and chondrosarcoma need to demonstrated that in the absence of negative margins, the
be considered (Sciubba et al. 2009). Key clinical, radio- recurrence rate was approximately 70 and 100 % following
graphic, and histologic findings help distinguish each type of radiation treatment alone, palliative care, or intralesional
tumor from a chordoma. Plasmacytoma may have a similar excision. When wide excision with adequate margins was
radiographic appearance as chordoma. A positive scintigra- obtained, there was a recurrence rate of 20 % diagnosed at
phy seen with a chordoma may differentiate these tumors 5694 months post surgery. In a group of patients all treated
(Greenspan et al. 2006). Radiographically, osteomyelitis and with en bloc excision for sacrococcygeal chordoma, Fuchs
lymphoma can be difficult to distinguish from a chordoma; et al. (2005) demonstrated an overall survival rate of 74 % at
however, their clinical course and laboratory data can usually 5 years, 52 % at 10 years, and 47 % at 15 years. Interestingly,
be used to distinguish these conditions (Sciubba et al. 2009). the survival rate in this group of patients was significantly
Benign notochordal cell tumors are usually asymptomatic, higher when negative margins were obtained, and the most
may demonstrate a more sclerotic appearance, and have no significant predictor of survival was a wide margin. The size
associated soft tissue mass. Giant cell tumors are benign but of the tumor, level of resection, and surgical approach did not
locally aggressive tumors that are often seen in the sacrum significantly impact the survival rate.
and can have an appearance similar to that of a chordoma. Sacral resection can be achieved using either a combined
Chordomas have more of a midline predilection, however, anterior-posterior approach or a posterior-only approach.
and giant cell tumors often demonstrate a thin rim of Generally, for tumors with significant anterior soft tissue
peripheral bone encompassing the soft tissue mass which
is not typically seen in a chordoma. Chondrosarcoma and
chondroid chordomas can have similar appearances, and the
biomarkers discussed above therefore play a critical role in
distinguishing these tumors.
17.7 Tumor Staging
Staging is the process of defining the local and distant

extension of a cancerous disease process. In the case of
chordoma, MRI is the primary imaging modality used to
evaluate the location and extent of the tumor. The presence
of metastatic disease is evaluated using CT scan of the
chest, abdomen, and pelvis with intravenous and oral con-
trast agents. Nuclear medicine bone scintigraphy can show
other skeletal lesions. PET metabolic imaging can be used
to demonstrate fluorodeoxyglucose avidity at sites of dis- Fig. 17.7 Intraoperative photograph showing sacral chordoma after
ease, but has some limitations based on lesion size, and is resection
mass, or high sacral resections, there is a benefit to first

performing an anterior dissection prior to resection from a
posterior approach. This was best shown by Fuchs et al.
(2005) in a series from the Mayo Clinic in which 81 % of
patients with combined anterior-posterior approaches had
negative margins. Based on these results, it was recom-
mended that a combined dual approach should be used for
any patients with tumors above S3.
After chordoma resection, functional deficit is dependent
on the extent of the surgery. To maximize survival rates and
to prepare patients for postoperative deficits, careful preop-
erative planning and discussion of possible neurological
problems is important. Whether partial or total sacrectomy is
performed, usually one or more sacral nerve roots will need
to be sacrificed, leading to a motor deficit, sensory impair-
ment, sphincter loss, and/or sexual dysfunction. Fourney
et al. (2005) developed a classification system describing the
type of sacrectomy based on the level of nerve root sacrificed, Fig. 17.8 An example of a postoperative reconstruction after resection
as opposed to the site of the osteotomy. The type of resection of entire sacrum
was defined as low, middle, high sacral amputation, total
sacrectomy, or hemicorporectomy. Sacral amputations are In the skull base, despite multiple surgical approaches,
considered low if at least one S4 nerve root is sacrificed, wide resection surgery is rarely possible (Singh et al. 2010;
middle if at least one S3 nerve root is sacrificed, and high Holzmann et al. 2010). Nonetheless, radiation therapy is
if at least one S2 nerve root is sacrificed. If neither S1 nerve effective in managing microscopic or limited gross amounts
root can be spared, then the required amputation is a total of residual tumor (Potluri et al. 2011). For this reason the
sacrectomy. A hemicorporectomy (translumbar amputation) recommended approach in these difficult anatomic areas is
is performed when the tumor extends to the lumbar spine. aggressive near-total intralesional excision that maximally
Functionally, if both S2 roots can be spared, half of the preserves neurologic function.
patients or more will have normal bowel and bladder func-
tion. If an S3 root is preserved as well, these odds improve.
However, perineal numbness and sexual dysfunction are still 17.8.2 Radiation
commonly observed, the latter more common in the elderly.
If one S2 root is sacrificed, typically some amount of volun- Radiation therapy has become increasingly important in the
tary control is compromised. Sacrifice of one S1 or S2 nerve management of chordomas. As a result of advances in radia-
root and all lower roots, as seen in high sacral amputations, tion techniques and modalities, it is now possible to deliver
commonly leads to urinary or fecal incontinence leading to higher doses to the tumor while sparing critical surrounding
the possibility for the need of indwelling urinary catheters, structures. The limiting factor is the tolerance of the spinal
intermittent straight catheterization, colostomy, or digital cord, in particular at its upper end, which is lower than the
stimulation to defecate (Hulen et al. 2006). Loss of ankle dose needed to effectively treat the tumor. In general, con-
plantar flexion is also seen after the loss of the S1 nerve ventional external beam radiation alone, at doses of 4060 Gy,
root. is suboptimal for treatment of chordoma, resulting in 5-year
If resection of the lumbar spine or pelvis is required at the local control of 1040 % (Catton et al. 1996; Cummings
time of sacral amputation, spinopelvic instability must be et al. 1983). Doses of up to 80 Gy have a high rate of associ-
assessed. If instability is observed, then instrumentation is ated radiation-induced myelopathy.
warranted (Fig. 17.8). In general, about 50 % of the sacro- Use of high-dose protons and charged particles such as
iliac joint can be removed before instability is seen, provided carbon, helium, or neon ions (collectively categorized as
the ligamentous structure of the remainder of the joint is hadrons) enables higher doses to be delivered to tumors with
preserved (Gunterberg et al. 1976; Stener and Gunteberg limited radiation damage to critical surrounding structures.
1978). Chordomas of the mobile spine are also best treated Indeed, as proton therapy has no measurable exit dose,
with resection with negative margins. To achieve this goal, a peripheral structures are spared. Proton beam treatment of
total spondylectomy is the surgery of choice and is generally chordomas, either alone or in combination with photons,
done through a combined anterior and posterior approach when used in conjunction with primary resection surgery,
(Boriani et al. 2006). has proven to be an excellent method for local control
(Hug et al. 1999; Nol et al. 2001; Fuji et al. 2011). This is sensitivity to aggressive chemotherapy (Fleming et al.
particularly relevant to skull-base tumors since the tolerance 1993), but overall there is no role for chemotherapy in treat-
is lower than in the peri-sacral regions. In the sacrum, pri- ment of localized disease. Currently, for metastatic disease,
mary surgery and radiation provide better results when com- both the timing of initiating treatment and the choice of
pared with treating chordoma reoccurrence, thereby chemotherapy regimen are generally made on an individu-
supporting a role for this approach as an effective first-line alized basis with significant consideration given to limiting
treatment option (Park et al. 2006). the side effect profile.
Carbon-ion radiotherapy has also been studied in chor- Based on the observation that chordomas have a high
doma management. Since carbon ions are heavier than pro- expression of platelet-derived growth factor receptors
tons, this approach is thought to provide a higher biological (PDGFRB and PDGFRA) and KIT receptors (Tamborini
effectiveness. Interestingly the effectiveness increases with et al. 2006), tyrosine kinase inhibitors have been used for
depth, reaching its peak at the end of the beams range. This the treatment of metastatic chordoma. Imatinib, a tyrosine
is an extremely attractive property for local control of cancer kinase inhibitor with specificity for PDGFRB and KIT
and as such has led to significant use of carbon-ion therapy receptors, was initially studied in a small group of chor-
in the management of chordoma. 5-year local control rates of doma patients with advanced disease (Casali et al. 2005);
7088 % and 10-year local control rates of 8082 % have the promising results of this study led to a larger phase II
been reported in skull-base tumors using carbon-ion therapy study with 50 patients (Stacchiotti et al. 2012). This recent
(Schulz-Ertner et al. 2007; Mizoe et al. 2009; Tsujii and study showed only one partial response obtained at 6 months;
Kamada 2012). however, there were 35 patients with stable disease and a
An alternate approach to using hadrons in treatment of 64 % clinical benefit rate, confirming the findings of smaller-
chordomas has been the use of highly conformal delivery tech- scale studies, and certainly warranting further investigation.
niques, such as intensity-modulated radiation therapy (IMRT) The EGFR pathway has also been implicated in the patho-
or stereotactic radiosurgery (SRS). In a recent study of the genesis of chordoma (Dewaele et al. 2011), leading to study
North American Gamma Knife Consortium, SRS was found of inhibitors of this pathway, including cetuximab, gefitinib,
to be an excellent option for small-sized chordomas, especially and erlotinib (Hof et al. 2006; Singhal et al. 2009). A multi-
for young patients, and when combined with surgery, provides center trial of sunitinib which chordomas, making up 19 %
an overall 80 % 5-year local control rate (Kano et al. 2011). of the study group, demonstrated a 44 % stable disease rate
for 16 weeks (George et al. 2009). Current ongoing sys-
temic trials include nilotinib, dasatinib, lapatinib, and
17.8.3 Systemic Therapy and Future Directions everolimus.
Preclinical studies are also accelerating, primarily due to
Chordomas are generally considered to be insensitive to the development of several cell lines which have been char-
conventional chemotherapy. Many conventional agents acterized, including CH8,GP60, and U-CH-1 (Yang et al.
have been tried, with varying levels of response, including 2010) and more recently CH22 (Liu et al. 2012). The devel-
anthracycline, cisplatin, alkylating agents, and camptothe- opment and characterization of these cell lines may help
cin analogues, but no single drug has emerged as a reliable identify as yet unknown targets and help clarify the role of
first-line agent. Some small-scale sporadic reports suggest suspected targets such as brachyury (Hsu et al. 2011), lead-
that dedifferentiated chordomas may have an increased ing to the development of further future clinical trials.
Box 17.2 RTK Inhibition in Chordoma the MAPK or PI3K/AKT pathways. As the schematic
There is significant interest in the role of receptor tyrosine indicates downstream of mTOR, the 40S ribosomal pro-
kinase (RTK) inhibition in the treatment of many sarco- tein S6 kinase (p70s6k) and the eukaryotic initiation factor
mas. Specifically, it has been demonstrated that chordomas 4E-binding protein-1 (4E-BP1) initiate protein synthe-
express activated platelet-derived growth factor recep- sis and promote cell growth and proliferation. eIF4E in
tors (PDGFRB) (Tamborini et al. 2006). Furthermore, an initiation factor that binds to the mRNA cap. When
a subset of chordomas are known to express EGFR and hyperphosphorylated, 4E-BP1 binds to eIF4E repressing
c-MET, both of which signal through a tyrosine kinase its translational initiation.
pathway (Weinberger et al. 2005). mTOR is downstream A recent study investigated the role of RTK inhibition
of the receptor tyrosine kinases and is activated via in chordomas (Tamborini et al. 2010). It was noted that
activated PDGFR, FLT3, CSF1-R, all components of the With respect to downstream effectors of mTOR,
PDGFR family, and EGFR family members EGFR, Her2/ Western blot analysis showed that 14 out of 22 chordomas
neu, and HER4 were present in chordoma tissue samples. cases demonstrated eIF4E release from translational
These findings are in strong support of the concept that the repression of 4E-BP1. In 13 of these cases, this was due
PDGFR and EGFR pathways are activated in chordoma. to hyperphosphorylation of 4E-BP1, and in one case there
Another observation was that EGFR and PDGFRB were was an absence of 4E-BP1. At the same time, phospho S6
co-immunoprecipitated, suggesting heterodimer forma- was only present at low or very low levels in 11 cases;
tion. This information could explain why some chordo- unphosphorylated S6 was found at low levels in 3 cases
mas are resistant to imatinib treatment. It is possible that a and not expressed at all in 8 cases. It is unclear if this sug-
bimodal approach using anti-PDGFR and anti-EGFR gests a tumor suppressor role for S6 or that the majority
agents may be required to fully silence the activation of of downstream effect of mTOR in chordoma is mediated
mTOR in chordomas. Interestingly, in two chordoma via 4E-BP1/eIF4E, but this discrepancy certainly war-
patients, there was a clinical response to cetuximab, an rants further study.
anti-EGFR monoclonal antibody (Hof et al. 2006).
Activated in chordoma
RTKs:
PDGFR family
EGFR family
PI3K
PIP2 PIP3 PTEN

mTOR inhibitors
mTORCO2 AKT AKT inhibitors

mTORCO1

S6K 4EBP1 Hyperphosphorylation or

absence seen in chordoma
Low phospho S6
Unphosphorylated S6 S6
Absence of S6 seen in chordoma
Schematic demonstrating
signaling through RTKs
via mTOR and critical
effectors PTEN, PI3K, and
Growth
AKT, and downstream
effectors S6K and 4EBP1
Box 17.3 Genetics of Chordoma been described and is consistent with loss of PTEN
Molecular studies of tumors from families with familial (Han et al. 2009, also see Box 17.2 for a review of RTK
chordoma syndromes have shed some light on the genet- signaling in chordoma). CDKN2A is a tumor suppressor
ics of chordoma. A recent study by Yang et al. (2009) gene that inhibits the function of cdk4- and cdk6-cyclin D
details the identification of T (brachyury) gene duplica- complexes. These cdk-cyclin complexes regulate the
tion and its role in conferring susceptibility to familial retinoblastoma protein, thereby controlling the G1-S
chordoma. These studies were performed using combined checkpoint of cell cycle progression. This gene was found
genetic linkage and high-resolution array CGH (compara- to be deficient in 80 % of sporadic tumors studied as
tive genomic hybridization) analyses to identify unique well.
duplications of a region on 6q27 in four families with Although the precise genetic mechanism for the
more than three cases or chordoma in each. This locus development of chordoma is unclear, these data taken
was found to contain the T (brachyury) gene. Brachyury together suggest that T/brachyury is a key figure in the
is a tissue-specific transcription factor expressed in the mechanism, at least in the case of familial chordoma. It is
nucleus of notochord cells. Chordomas express brachyury, possible that T/brachyury is important in sporadic chor-
but its expression has been studied, and it is not found to doma as well, but if this is the case, then the mechanism
be expressed in nonneoplastic tissues and in 42 other must be one other than copy number variation based on
types of neoplasm. Its exact role in the pathogenesis of these findings.
chordoma is unclear, but the finding represents a significant
advancement in the understanding of chordoma biology. References
The majority of chordomas, however, are sporadic. Han S, Polizzano C, Neilsen GP, Hornicek FJ, Rosenberg AE,
Sporadic chordomas do not display duplication or Ramesh V (2009) Aberrant hyperactivation of akt and
amplification of the brachyury gene. Karyotype analyses Mammalian target of rapamycin complex 1 signaling in
of sporadic chordomas demonstrate several abnormalities sporadic chordomas. Clin Cancer Res 15:19401946
that identify this disease as one of significant genetic Le LP, Nielsen GP, Rosenberg AE, Thomas D, Batten JM,
instability, as demonstrated in a study by Le et al. (2011). Deshpande V, Schwab J, Duan Z, Xavier RJ,
Copy number variations involving copy number losses Hornicek FJ, Iafrate AJ (2011) Recurrent chromo-
are seen more frequently than copy number gains. somal copy number alterations in sporadic chordomas.
The chromosomes with relevant losses include PLOS One 6:e18846
1p,3,4,9,10,13,14, and 18. PTEN, which is an important Yang XR, Ng D, Alcorta DA, Liebsch NJ, Sheridan E, Li
tumor suppressor gene located on 10q23.3, was found to S, Goldstein AM, Parry DM, Kelley MJ (2009) T
have hemizygous deletion in 80 % of the sporadic chor- (brachyury) gene duplication confers major suscepti-
domas studied. Interestingly, hyperactivation of Akt/ bility to familial chordoma. Nat Genet 41:
mTORC1 signaling in sporadic sacral chordomas has 11761178
17.9 Summary of Critical Concepts The role for chemotherapy is limited but advancements in
Discussed in This Chapter systemic therapy using agents targeting tyrosine kinase
pathways show significant promise.
Chordomas are rare tumors of notochordal origin.
Surgical treatment with wide resection is the mainstay of
therapy. References
Local recurrence and survival rates are dependent on
margins. Ashwood N, Hoskin PJ, Saunders MI (1994) Metastatic chordoma: pat-
tern of spread and response to chemotherapy. Clin Oncol (R Coll
Radiation is an important modality of treatment, gener-
Radiol) 6:341342
ally as an adjuvant to surgery. Atalar H, Selek H, Yildiz Y, Salik Y (2006) Management of sacrococ-
Alternate forms of radiation such as proton therapy, had- cygeal chordomas. Int Orthop 30(6):514518
ron therapy, or stereotactic radiation are all playing Bergh P, Kindblom LG, Gunterberg B, Remotti F, Ryd W, Meis-
Kindblom JM (2000) Prognostic factors in chordoma of the sacrum
increasingly important roles.
and mobile spine: a study of 39 patients. Cancer 88:21222134
Platelet-derived growth factor receptor signaling is impor- Bilsky MH, Yamada Y, Yenice KM, Lovelock M, Hunt M, Gutin
tant in the oncogenesis of chordoma. PH, Leibel SA (2004) Intensity-modulated stereotactic radiother-
apy of paraspinal tumors: a preliminary report. Neurosurgery 54: treatment of nongastrointestinal stromal tumor sarcomas. J Clin
823830 Oncol 27(19):31543160
Bjornsson J, World LE, Ebersold MJ, Laws ER (1993) Chordoma of the Greenspan A, Jundt G, Remagen W (2006) Chordoma, in differential
mobile spine. A clinicopathological analysis of 40 patients. Cancer diagnosis in orthopaedic oncology, 2nd edn. Lippincott Williams &
71:735740 Wilkins, Philadelphia, pp 445446
Bohlman HH, Sachs BL, Carter JR, Riley L, Robinson RA (1986) Gunterberg B, Stener B, Romanus B (1976) Pelvic strength after major
Primary neoplasms of the cervical spine. Diagnosis and treatment of amputation of the sacrum. An experimental study. Acta Orthop
twenty-three patients. J Bone Joint Surg Am 68:483494 Scand 47:635642
Boriani S, Chevalley F, Weinstein JN, Biagini R, Campanacci L, Healey JH, Lane JM (1989) Chordoma: a critical review of diagnosis
De Iure F, Piccill P (1996) Chordoma of the spine above the sacrum. and treatment. Orthop Clin North Am 20:417426
Treatment and outcome in 21 cases. Spine 21:15691577 Hof H, Welzel T, Debus J (2006) Effectiveness of cetuximab/gefitinib
Boriani S, Bandiera S, Biagini R, Bacchini P, Boriani L, Cappuccio M, in the therapy of a sacral chordoma. Onkologie 29(12):572574
Chevalley F, Gasbarrini A, Picci P, Weinstein JN (2006) Chordoma Holzmann D, Reisch R, Krayenbuhl H, Hug E, Bernays RL (2010)
of the mobile spine: fifty years of experience. Spine 31:493503 The transnasal transclival approach for clivus chordoma. Minim
Burrow JF, Stewart MJ (1923) Malignant spheno-occipital chordoma. Invasive Neurosurg 53:211217
J Neurol Psychopathol 4(15):205217 Hsu W, Mohyeldin A, Shah SR, Rhys CM, Johnson LF, Sedora-Roman
Casali PG, Messina A, Stacchiotti S, Tamborini E, Crippa F, Gronchi A, NI, Kosztowski TA, Awad OA, McCarthy EF, Loeb DM,
Orlandi R, Ripamonti C, Spreafico C, Bertieri R, Bertulli R, Wolinsky JP, Gokaslan ZL, Quiones-Hinojosa A (2011) Generation
Colecchia M, Fumagalli E, Greco A, Grosso F, Olmi P, Pierotti MA, of chordoma cell line JHC7 and the identification of Brachyury as a
Pilotti S (2005) Imatinib mesylate in 18 advanced chordoma novel molecular target. J Neurosurg 115(4):760769
patients. J Clin Oncol 23:9012 Hug EB, Loredo LN, Slater JD, DeVries A, Grove RI, Schaefer RA,
Catton C, OSullivan B, Bell R, Laperriere N, Cummings B, Fornasier V, Rosenberg AE, Slater JM (1999) Proton radiation therapy for chor-
Wunder J (1996) Chordoma: long-term follow up after radical pho- domas and chondrosarcomas of the skull base. J Neurosurg 91(3):
ton irradiation. Radiother Oncol 41:6772 432439
Chugh R, Tawbi H, Lucas DR, Biermann JS, Schuetze SM, Baker LH Hulen CA, Temple HT, Fox WP, Sama AA, Green BA, Eismont FJ
(2007) Chordoma: the nonsarcoma primary bone tumor. Oncologist (2006) Oncologic and functional outcome following sacrectomy for
12:13441350 sacral chordoma. J Bone Joint Surg Am 88:15321539
Crapanzano JP, Ali SZ, Ginsberg MS, Zakowski MF (2001) Chordoma: Kano H, Iqbal FO, Sheehan J, Mathieu D, Seymour ZA, Niranjan A,
a cytologic study with histologic and radiologic correlation. Cancer Flickinger JC, Kondziolka D, Pollock BE, Rosseau G, Sneed PK,
93:4051 McDermott MW, Lunsford LD (2011) Stereotactic radiosurgery for
Cummings BJ, Hodson DI, Bush RS (1983) Chordoma: the results of chordoma: a report from the North American Gamma Knife
megavoltage radiation therapy. Int J Radiat Oncol Biol Phys 9: Consortium. Neurosurgery 68(2):379389
633642 Kishimoto R, Omatsu T, Hasegawa A, Imai R, Kandatsu S, Kamada T
Dewaele B, Maggiani F, Floris G, Ampe M, Vanspauwen V, Wozniak A, (2012) Imaging characteristics of metastatic chordoma. Jpn J Radiol
Debiec-Rychter M, Sciot R (2011) Frequent activation of EGFR in 30(6):509516
advanced chordomas. Clin Sarcoma Res 1(1):4 Lee J, Bhatia NN, Hoang BH, Ziogas A, Zell JA (2012) Analysis of
Fleming GF, Heimann PS, Stephens JK, Simon MA, Ferguson MK, prognostic factors for patients with chordoma with use of the
Benjamin RS, Samuels BL (1993) Dedifferentiated chordoma. California Cancer Registry. J Bone Joint Surg Am 94(4):356363
Response to aggressive chemotherapy in two cases. Cancer Leone A, Cerase A, Tarquini E, Mul A (2002) Chordoma of the low
72(3):714718 cervical spine presenting with Horners syndrome. Eur Radiol
Forsyth PA, Cascino TL, Shaw EG, Scheithauer BW, OFallon JR, 12(Suppl 3):S43S47
Dozier JC, Piepgras DG (1993) Intracranial chordomas: a clinico- Lin CY, Kao CH, Liang JA, Hsieh TC, Yen KY, Sun SS (2006) Chordoma
pathological and prognostic study of 51 cases. J Neurosurg 78: detected on F-18 FDG PET. Clin Nucl Med 31(8):506507
741747 Liu X, Nielsen GP, Rosenberg AE, Waterman PR, Yang W, Choy E,
Fourney DR, Gokaslan ZL (2003) Current management of sacral Sassi S, Yang S, Harmon DC, Yang C, Schwab JH, Kobayashi E,
chordoma. Neurosurg Focus 15:E9 Mankin HJ, Xavier R, Weissleder R, Duan Z, Hornicek FJ (2012)
Fourney DR, Rhines LD, Hentschel SJ, Skibber JM, Wolinsky JP, Establishment and characterization of a novel chordoma cell line:
Weber KL, Suki D, Gallia GL, Garonzik I, Gokaslan ZL (2005) En CH22. J Orthop Res. doi:10.1002/jor.22113
bloc resection of primary sacral tumors: classification of surgical Manaster BJ, Graham T (2003) Imaging of sacral tumors. Neurosurg
approaches and outcome. J Neurosurg Spine 3:111122 Focus 15(2):E2
Froehlich EV, Scheipl S, Lazry A, Varga PP, Schmid C, Stammberger H, McMaster ML, Goldstein AM, Bromley CM, Ishibe N, Parry DM
Beham A, Bodo K, Schroettner H, Quehenberger F, Windhager R, (2001) Chordoma: incidence and survival patterns in the United
Liegl B, Leithner A (2012) Expression of ezrin, MMP-9, and COX-2 States, 19731995. Cancer Causes Control 12:111
in 50 chordoma specimens: a clinical and immunohistochemical Meyer JE, Lepke RA, Lindfors KK, Pagani JJ, Hirschy JC, Hayman LA,
analysis. Spine 37(13):E757E767 Momose KJ, McGinnis B (1984) Chordomas: their CT appearance in
Fuchs B, Dickey I, Yaszemski MJ, Inwards CY, Sim FH (2005) the cervical, thoracic and lumbar spine. Radiology 153(3):693696
Operative management of sacral chordoma. J Bone Joint Surg Am Mindell E (1981) Chordoma. J Bone Joint Surg Am 63:501505
87:22112216 Miyazawa N, Ishigame K, Kato S, Satoh Y, Shinohara T (2008) Thoracic
Fuji H, Nakasu Y, Ishida Y, Horiguchi S, Mitsuya K, Kashiwagi H, chordoma: review and role of FDG-PET. J Neurosurg Sci 52(4):
Murayama S (2011) Feasibility of proton beam therapy for chor- 117121
doma and chondrosarcoma of the skull base. Skull Base Mizoe JE, Hasegawa A, Takagi R, Bessho H, Onda T, Tsujii H (2009)
21(3):201206 Carbon ion radiotherapy for skull base chordoma. Skull Base
George S, Merriam P, Maki RG, Van den Abbeele AD, Yap JT, 19(3):219224
Akhurst T, Harmon DC, Bhuchar G, OMara MM, DAdamo DR, Muller H (1858) Uber das Vorkommen von Resten der Chorda dorsalis
Morgan J, Schwartz GK, Wagner AJ, Butrynski JE, Demetri GD, bei Menschen nach der Geburt und iiber ihr Verhiiltniss zu den
Keohan ML (2009) Multicenter phase II trial of sunitinib in the Gallertgeschwiilsten am Clivus. Zeitschr f rat Med ii:202229
Nicoucar K, Rausch T, Becker M, Dulguerov P (2008) Cervical chor- Singhal N, Kotasek D, Parnis FX (2009) Response to erlotinib in a
doma with retropharyngeal extension presenting with impaired patient with treatment refractory chordoma. Anticancer Drugs
voice. Tumori 94:873876 20(10):953955
Nishiguchi T, Mochizuki K, Ohsawa M, Inoue T, Kageyama K, Soo MY (2001) Chordoma: review of clinicoradiological features and
Suzuki A, Takami T, Miki Y (2011) Differentiating benign noto- factors affecting survival. Australas Radiol 45:427434
chordal cell tumors from chordomas: radiographic features Stacchiotti S, Longhi A, Ferraresi V, Grignani G, Comandone A,
on MRI, CT, and tomography. AJR Am J Roentgenol 196(3): Stupp R, Bertuzzi A, Tamborini E, Pilotti S, Messina A, Spreafico
644650 C, Gronchi A, Amore P, Vinaccia V, Casali PG (2012) Phase II
Nol G, Habrand JL, Mammar H, Pontvert D, Haie-Mder C, study of imatinib in advanced chordoma. J Clin Oncol 30(9):
Hasboun D, Moisson P, Ferrand R, Beaudr A, Boisserie G, 914920
Gaboriaud G, Mazal A, Krody K, Schlienger M, Mazeron JJ (2001) Stener B, Gunterberg B (1978) High amputation of the sacrum for extir-
Combination of photon and proton radiation therapy for chordomas pation of tumors. Principles and technique. Spine 3:351366
and chondrosarcomas of the skull base: the centre de Protonthrapie Sundaresan N, Schmidek HH, Schiller AL, Rosenthal D (1990) Tumors
DOrsay experience. Int J Radiat Oncol Biol Phys 51(2): of the spine: diagnosis and clinical management. W.B. Saunders,
392398 Philadelphia
Oakley GJ, Fuhrer K, Seethala RR (2008) Brachyury, SOX-9, and Tamborini E, Miselli F, Negri T et al (2006) Molecular and biochemical
podoplanin, new markers in the skull base chordoma vs chondrosar- analyses of platelet-derived growth factor receptor (PDGFR) B,
coma differential: a tissue microarray-based comparative analysis. PDGFRA, and KIT receptors in chordomas. Clin Cancer Res
Mod Pathol 21:14611469 12:69206928
Papagelopoulos PJ, Mavrogenis AF, Currier BL, Katonis P, Galanis EC, Tamborini E, Virdis E, Negri T, Orsenigo M, Brich s, Conca E, Gronchi
Sapkas GS, Korres DS (2004) Primary malignant tumor of the A, Stacchiotti S, Manenti G, Casali PG, Pierotti MA, Pilotti S (2010)
cervical spine. Orthopedics 27:10671075 Analysis of receptor tyrosine kinases (RTKs) and downstream path-
Park SA, Kim HS (2008) F-18 FDG PET/CT evaluation of sacrococ- ways in chordomas. Neuro Oncol 12(8):776789.
cygeal chordoma. Clin Nucl Med 33(12):906908 Tsujii H, Kamada T (2012) A review of update clinical results of carbon
Park L, Delaney TF, Liebsch NJ, Hornicek FJ, Goldberg S, Mankin H, ion radiotherapy. Jpn J Clin Oncol 42(8):670685
Rosenberg AE, Rosenthal DI, Suit HD (2006) Sacral chordomas: Unni KK (1996) Chordoma. In: Unni KK (ed) Dahlins bone tumors:
impact of high-dose proton/photon-beam radiation therapy com- general aspects and data on 11,087 cases, 5th edn. Lippincott-Raven,
bined with or without surgery for primary versus recurrent tumor. Philadelphia, pp 291305
Int J Radiat Oncol Biol Phys 65(5):15141521 Virchow RL (1857) Untersuchungen ueber die Entwicklung des
Pearlman AW, Friedman M (1970) Radical radiation therapy of Schaedelgrundes. G Rimer, Berlin
chordoma. Am J Roentgenol Radium Ther Nucl Med 108(2): Vujovic S, Henderson S, Presneau N, Odell E, Jacques TS, Tirabosco R,
332341 Boshoff C, Flanagan AM (2006) Brachyury, a crucial regulator
Pearlman AW, Singh RK, Hoppenstein R, Wilder J (1972) Chordoma: of notochordal development, is a novel biomarker for chordomas.
combined therapy with radiation and surgery: case report and new J Pathol 209:157165
operative approach. Bull Hosp Joint Dis 33(1):4757 Weber K, Sim FH (2002) Chordoma. In: Bulstrode C, Buckwalter J,
Potluri S, Jeffries SJ, Jena R, Harris F, Burton KE, Prevost AT, Carr A et al (eds) Oxford textbook of orthopaedics and trauma.
Burnet NG (2011) Residual postoperative tumor volume pre- Oxford University Press, New York, pp 294299
dicts outcome after high-dose radiotherapy for chordoma and Weinberger PM, Yu Z, Kowalski D, Joe J, Manger P, Psyrri A, Sasaki
chondrosarcoma of the skull base and spine. Clin Oncol 32: CT (2005) Differential expression of epidermal growth factor recep-
199208 tor, c-Met, and HER2/neu in chordoma compared with 17 other
Schulz-Ertner D, Karger CP, Feuerhake A, Nikoghosyan A, Combs SE, malignancies. Arch Otolaryngol Head Neck Surg 131(8):
Jkel O, Edler L, Scholz M, Debus J (2007) Effectiveness of carbon 707711.
ion radiotherapy in the treatment of skull-base chordomas. Int Yang XR, Ng D, Alcorta DA, Liebsch NJ, Sheridan E, Li S,
J Radiat Oncol Biol Phys 68(2):449457 Goldstein AM, Parry DM, Kelley MJ (2009) T (brachyury) gene
Sciubba D, Cheng JJ, Petteys RJ, Weber KL, Frassica DA, Gokaslan ZL duplication confers major susceptibility to familial chordoma. Nat
(2009) Chondroma of the sacrum and vertebral bodies. J Am Acad Genet 41(11):11761178
Orthop Surg 17(11):708717 Yang C, Hornicek FJ, Wood KB, Schwab JH, Choy E, Iafrate J,
Singh N, Soo M, De Cruz M, Gomes L, Maclean F, Dandie G (2007) Rosenberg A, Nielsen GP, Xavier RJ, Mankin H, Duan Z (2010)
Cervical chordoma presenting as retropharyngeal mass and dyspho- Characterization and analysis of human chordoma cell lines. Spine
nia: case report and literature review. Australas Radiol 51(suppl): 35(13):12571264
B183B188 Zhang H, Yoshikawa K, Tamura K, Sagou K, Tian M, Suhara T,
Singh H, Harrop J, Schiffmacher P, Rosen M, Evans J (2010) Ventral Kandatsu S, Suzuki K, Tanada S, Tsujii H (2004) Carbon-11-
Surgical approaches to craniovertebral junction chordomas. methionine positron emission tomography imaging of chordoma.
Neurosurgery 66(suppl 3):96103 Skeletal Radiol 33(9):524530
Part III
Models of Disc Disease
and Biological Regeneration
Large Animal Models of Disc
Degeneration 18
Shyam A. Patel, Christopher K. Kepler, Thomas P. Schaer,
and D. Greg Anderson
Contents 18.6 Guide to Selecting the Most Useful

Animal Model ................................................................ 300
18.1 Overview ........................................................................ 291
18.2 History ............................................................................ 292 in the Chapter ................................................................ 300
18.3 Types of Models Used in Intervertebral References .................................................................................... 301
Disc Research................................................................. 292
18.3.1 The Clinical Perspective ................................................. 292
18.3.2 Animal Models in General .............................................. 294
18.4 Large Animal Models ................................................... 295
18.4.1 Porcine ............................................................................ 295 18.1 Overview
18.4.2 Ovine ............................................................................... 297
18.4.3 Canine ............................................................................. 297 Intervertebral discs are largely of embryonic notochordal
18.4.4 Caprine ............................................................................ 298 origin and impart a unique biomechanical function to the
18.5 Advantages and Disadvantages of Large spine of vertebrate animals (Singh et al. 2005). The interver-
Animal Species Used in Disc Research ........................ 298 tebral disc is anatomically comprised of two constituents: a
18.5.1 Porcine/Miniature Pig ..................................................... 298
18.5.2 Ovine ............................................................................... 298
proteoglycan-rich nucleus pulposus contained within a col-
18.5.3 Canine ............................................................................. 299 lagen 1-rich annulus fibrosus (Singh et al. 2005). Together,
18.5.4 Caprine ............................................................................ 299 these structures effectively dissipate mechanical loads within
18.5.5 Nonhuman Primates ........................................................ 299 the spine while allowing controlled motion between adjacent
vertebrae. The biologic environment of the disc is uniquely
harsh due to the avascular nature of the tissue and the long
distance between metabolically active cells and their nutri-
tional source. Tissue degeneration progresses through a well-
S.A. Patel, BS
defined series of changes (Singh et al. 2005) including
Drexel University College of Medicine,
Philadelphia, PA 19107, USA breakdown of the long-chain proteoglycan constituents of
e-mail: sap.0521@gmail.com the extracellular matrix, loss of water-binding capacity,
C.K. Kepler, MD decreased cellularity, and annular disorganization/disruption
Department of Orthopaedic Surgery, (Lotz 2004).
Thomas Jefferson University, Rothman Institute, As discussed in other sections of this book, while degen-
Philadelphia, PA 19107, USA
erative changes within the intervertebral disc are essentially
e-mail: chris.kepler@gmail.com
universal, the association between disc degeneration and
T.P. Schaer, VMD (*)
symptomatic low back pain is unpredictable (Lotz 2004;
Department of Clinical Studies,
University of Pennsylvania, Alini et al. 2008). Indeed, back pain attributed to disc degen-
School of Veterinary Medicine, New Bolton Center, eration is a relatively common phenomenon; however, most
Kennett Square, PA 19348, USA individuals have few or no symptoms during the degenera-
e-mail: tpschaer@vet.upenn.edu
tive process. The lack of correlation between degeneration
D.G. Anderson, MD and clinical symptoms has made the study of disc degenera-
Departments of Orthopaedic Surgery and Neurological Surgery,
tion especially challenging (Lotz 2004). Despite the chal-
Thomas Jefferson University, Rothman Institute,
Philadelphia, PA 19107, USA lenge, the goal of researchers in the field is to identify an
e-mail: greg.anderson@rothmanisntitute.com effective biologic therapy for painful disc degeneration.

292 S.A. Patel et al.
Human discs are subjected to some unique physical and be useful in treating dislocations, fractures, and osteolytic
biomechanical factors that are not easily replicated in ani- processes of the human spine (Wagner et al. 1979).
mal models. First, humans are bipedal and subject their Current prosthetic cages used in spinal fusion can trace their
spines to unique physical forces associated with an upright roots to this experiment (Wagner et al. 1979).
posture. Second, human discs are significantly larger in size
than many common research animals such as rodents and
rabbits. This means that the diffusion of nutrients or phar- 18.3 Types of Models Used
macologic agents into the central region of the disc is sub- in Intervertebral Disc Research
stantially more challenging. In addition, because of the
larger size, certain types of physical interventions that Intervertebral disc research involving various regions of the
would be theoretically possible in humans are difficult to spine has been useful in assessing the physical, histological,
reproduce in small animal models. Third, although the pri- and biomechanical characteristics of the disc (Panjabi 1998).
mary symptom attributed to disc degeneration in humans is In vitro models have also played a substantial role in provid-
pain, it is challenging to study pain of spinal origin in ani- ing an understanding of the unique cellular biology of the
mals. Fourth, intervertebral discs from various species dif- disc. Unfortunately, in vitro models generally lack the com-
fer quite substantially in their cellular makeup. The large plexity of the intact system, which may limit the ability of
physaliferous cells that are felt to be helpful in maintaining researchers to draw definitive clinical conclusions from the
the health of the disc are largely absent in adult humans, but scientific findings. Additionally, in vitro models generally do
remain quite common in some other species. This may not provide an understanding of long-term experimental
affect the response of the disc to various experimental effects of an intervention.
interventions. Advances in computer technology have led to the devel-
Animal research is critical to understanding, developing, opment of some in silico models. These computer-based
and testing biologic interventions for human disc degen- simulations are suitable for forming initial hypotheses that
eration. Although most of the animal research done to date may serve as the rationale for further research. Computer
involves small animals, there are certain limitations in the simulations also allow the investigator a level of control over
use of these models mainly due to differences in the rela- experimental variables that is typically not possible with
tive dimensions of the disc or the cellular makeup of the in vitro and in vivo studies. With respect to intervertebral
disc. Thus, large animal disc research is both necessary and disc research, in silico models have seen limited utilization
important for the development of many disc therapies and is in the study of mechanical properties of the spine through
generally required prior to moving a promising therapy into finite element analysis modeling or the modeling of material
human usage. Large animal models are often specifically properties of the disc (Galbusera et al. 2011).
required during regulatory approval processes by agen- The testing of novel therapeutics for human applications
cies such as the Food and Drug Administration (FDA), the generally requires the use of in vivo studies. With in vivo
Central Office for Research Ethics Committees (COREC), research, investigators are able to observe the impact of treat-
the National Institute for Clinical Excellence (NICE), and ment on a living biologic system (Panjabi 1998). Given the
the Medicines and Healthcare Products Regulatory Agency complexity of intervertebral disc tissue and the multifactorial
(MHRA). This chapter will review the status of the large ani- nature of the degenerative process, in vivo research plays a
mal models that have been used to study disc degeneration crucial role in the development of human therapeutics (Lotz
or disc therapeutics. 2004; Singh et al. 2005). Additionally, before most therapeu-
tics are considered suitable for human use, it is required that
they are studied in an animal model with reasonable similar-
18.2 History ity to the human condition; this often requires the use of a
large animal model (Alini et al. 2008).
Historically, large animal veterinary care has played a role in
our current understanding of disc pathology and surgical
reconstruction. One example is the work of Wagner et al. 18.3.1 The Clinical Perspective
(1979) who, in 1979, performed cervical fusion using bone
from equine cadaver ilia to treat horses with cervical spine From a clinical perspective, the treatment of disc degenera-
disorders. Eleven of the 12 horses achieved a fusion of the tion itself is of limited value as most disc degeneration is not
operative levels and improved clinically; follow-up necrop- directly responsible for patient symptoms (this concept is
sies revealed restricted motion at the fusion levels (Wagner discussed in great length in Chap. 16). However, disc degen-
et al. 1979). The authors concluded that this technique could eration associated with significant pain symptoms is
18 Large Animal Models of Disc Degeneration 293
Box 18.1 Spinal Fusion Basket: Bagbys Basket Arabian colt who presented with grade III ataxia in all
Interestingly, the history of spine fusion cages begins with four limbs. Laboratory findings for EPM (equine proto-
horses. Four decades ago, Dr. George Bagby invented the zoal myeloencephalitis) were negative and as was the
first cage, also referred to as Bagbys basket. In collabo- remainder of cerebral spinal fluid analysis. A myelogram
ration with Dr. Barrie Grant, a veterinarian from under general anesthesia showed static compression at
Washington State University, the two surgeons explored C67 (B, solid arrow) as well as dynamic compression at
surgical techniques to treat cervical instability or wob- C34 and C45 (B, hashed arrow). Ventral stabilization
blers disease, a neurological condition caused by was performed at C67 (C&D) 1 month after diagnosis.
impingement of the spinal cord in the horses neck by After surgery, the colt was on stall rest for 1 month, then
malformed or degenerated vertebrae. While spinal fusion 1 month of hand-walking, followed by 1 month of round-
for human patients has rapidly evolved, cervical interbody pen turnout. Exercise level was slowly increased thereaf-
fusion using a stainless steel cage remains the only suc- ter, the ataxia steadily improved (grade 0.5 in LF/RF/LH
cessful surgical intervention to treat this debilitating dis- and grade 2 RH 9 months after surgery), and now he is
ease in horses. Digital radiographs1 (AD) of a yearling showing successfully with full recovery.
a b
c d
1
Courtesy of Drs. Reed and Woddie, Rood and Riddle Equine
Hospital, Lexington, KY
common, and it is the associated pain and decreased quality For many studies involving therapeutic interventions, an
of life that are the targets of therapy. important factor is identifying a model with similar
Unfortunately, a reliable means to quantify pain associ- characteristics in terms of disc size and geometry (Beckstein
ated with disc degeneration in any animal model has not yet et al. 2008). Wide variations in disc size across species are
been defined. Therefore, animal research is generally limited obvious, and this can affect the mechanical and biologic
to providing a means to study the effects of a potential ther- properties of the disc (Elliott and Sarver 2004; OConnell
apy on the physical, cellular, or chemical milieu of the disc. et al. 2007). However, Beckstein et al. (2008) demonstrated
The hypothesis of this line of research is that an intervention that mechanical tissue properties of many species are
which provides improvement in objective measures of disc conserved evolutionarily and that variations with regard to
health would be likely to provide improvement in the pain the biomechanical properties of the intervertebral disc are
associated with the degenerative process in affected human not as drastic as previously thought. Unfortunately, the
patients. Therefore, although animal testing of potential current literature fails to provide comprehensive compari-
human therapeutics is important, favorable results still sons between animal models across a wide range of species.
require validation in humans. Thus, animal model testing is Most comparative studies published to date compare the
just one step in the process of bringing a therapy from the properties of a specific animal species relative to the
bench top to the clinic (An et al. 2003). human.
18.3.2.3 Complications Associated

18.3.2 Animal Models in General with Validity/Fidelity
One factor which complicates animal model selection is
As stated, animal models play a vital role in translational related to intervertebral disc development. The nucleus pul-
research and provide a crucial step in ensuring that a new posus is embryologically derived from notochordal cells. In
therapy is valid and safe prior to considering human usage humans, notochordal cells which are common at birth
(Smith et al. 2011). become rare following the first decade of life (Urban et al.
2000). Notochordal cells are believed to produce greater
18.3.2.1 Standardization and Reliability amounts of proteoglycan than other disc cells and contribute
When an animal model is first introduced, it is essential to positively to the maintenance of disc extracellular matrix
evaluate its reliability. Using the new model, independent integrity (Palmer and Lotz 2004; Aguiar et al. 1999;
investigators performing similar experiments should find Kalichman and Hunter 2008). In addition to wide variation
comparable and reproducible results. Significant variation in in the persistence of these cells on a species-by-species basis,
results between investigators or unpredictable results from exposure to high biomechanical loads on the spine may has-
repetitive experiments may indicate that the model used is ten the disappearance of notochordal cells (Lotz et al. 1998;
unreliable. Iatridis et al. 1999; Palmer and Lotz 2004).
With animal models, standardization of the test animals Variability in the presence and number of notochordal
age plays a key role in achieving adequate reliability for disc cells across species is important given their potential in
research. Other factors such as breed, gender, size, living resisting disc degeneration (Lotz 2004). In fact, Berry (1961)
conditions, vertebral levels studied, and nutrition could demonstrated that a breed of pin-tailed mice with low noto-
potentially affect the outcome of a study intervention, and so chordal mitotic rates develops degeneration earlier than those
these features should be well controlled in the experimental with a more rapid rate of cell replication. This relationship
design. These factors should also be reported by the investi- between notochordal cells and disc degenerations may com-
gator performing the animal research so that others can rep- plicate the validity of many current animal models (Omlor
licate the study conditions. et al. 2009) as certain species, particularly small animals
(Higuchi et al. 1982) and dogs (Aguiar et al. 1999), have
18.3.2.2 Model Fidelity/Validity notochordal cells present throughout most of their lives.
Issues of validity and fidelity are important in the design of Differences in the number or qualitative characteristics of
animal trials. There is little benefit in using an animal model notochordal cells across species may manifest itself clini-
with such substantial differences from human disease that cally as variation in the onset of disc degeneration (Hunter
the information gained is not clinically relevant. Because all et al. 2004).
animal models are significantly different from the human Animal models may also differ in the mechanism leading
condition, this limitation is one of magnitude. The researcher to the onset of the degradative process in the disc. Generally
should seek to use a model that provides the closest possible speaking, the etiology of disc degeneration in an animal
replication of the human condition, understanding that a per- model can be divided into spontaneous and experimentally
fect match is not possible. induced degeneration (Lotz 2004) with the latter being
further subdivided into models which are induced by either process is well advanced. Specimens were stained using a
mechanical or structural alterations (Lotz 2004). Mechanical Sirius red/Alcian Blue stain developed in our laboratory as
perturbations involve changes in the distribution or magni- a method for morphologic identification of collagen and
tude of loads placed on a disc, whereas structural perturba- proteoglycans in sections of non-decalcified disc specimens
tions involve chemical or physical injury to the disc itself embedded in methyl methacrylate and sectioned on a rotary
(Lotz 2004). In contrast, human intervertebral disc degenera- microtome at 48 mm.
tion usually occurs spontaneously (Lotz 2004) or possibly in
response to a poorly defined mechanism of injury. Although
spontaneous disc degeneration is present in many animal 18.4.1 Porcine
species, well-characterized models of spontaneous degenera-
tion generally have involved smaller animals such as the Multiple studies of the intervertebral disc have relied on
Mediterranean sand rat (Silberberg 1988) and genetically porcine models. A number of lines of evidence suggest that
altered mice (Lotz 2004). The value of the sand rat in disc the porcine model is a reasonably good model of the human
research is discussed in considerable detail in Chap. 20. One disc. At the gene expression level, changes associated with
counterexample, however, is chondrodystrophic dogs which aging in swine are similar to those observed in humans (Cho
have an abnormality in chondrocyte proliferation and matu- et al. 2011). Substantial morphologic similarities have been
ration that also impacts the cells of the nucleus pulposus. observed in the porcine disc post-nucleotomy and the human
With chondrodystrophy, there is accelerated disc degenera- degenerative disc. Omlor et al. (2009) found significant
tion, making these dogs a potential preclinical model for the decreases in disc height, MRI signal intensity, and
human condition. notochordal cell number in the porcine disc after nucleot-
omy (Omlor et al. 2009). Kaapa et al. (1994b, c) studied the
effect of pig disc injury on collagen content and metabolism
18.4 Large Animal Models and found dramatic alterations in collagen levels and colla-
gen biosynthesis along with decreased intradiscal water
Sheep, goats, pigs, mini pigs, runt cows, nonhuman primates, content. Holm et al. (2007) used a structural porcine disc
dogs, chickens, kangaroos, and ostriches are all large animal injury model, in which intradiscal pressure was used to
species that have been used or discussed for disc research. assess degeneration. Using two separate injuries (one to the
However, the most commonly used species to date have been annulus and the other to the endplate), intradiscal pressure
the pig, sheep, goat, and dog (Wilke et al. 1997a, b). was measured while applying biomechanical loads (Holm
One concern regarding the use of the most commonly et al. 2007). The authors found that the ratio of injured to
used large animal models relates to their quadruped nature adjacent disc pressure was highest in the annulus injury
(Wilke et al. 1997a, b). Unlike quadrupeds, humans are model, signifying that the endplate injury was a more severe
bipedal, using unique motions and body positions during type of injury. Although this type of injury may lead to mor-
ambulation and rest (Zhang et al. 2011a, b). This biome- phologic and biochemical changes which replicate human
chanical limitation must be acknowledged when consid- disc degeneration, the traumatic effects of the disc injury
ering quadrupedal animals for research. Although some may also introduce unintended effects that limit the inter-
animals use bipedal ambulation (e.g., kangaroo, ostrich) pretation of this type of research.
and some could be considered partial bipeds (nonhuman Other studies have used pigs to assess the therapeutic
primates), none provide an accurate replication of human value of various interventions or to examine specific
gait including the upright torso position, sagittal balance changes associated with degeneration. Buser et al. (2011)
of the spine, and the almost horizontally positioned discs evaluated the use of fibrin sealant after nucleotomy to deter-
during ambulation. Tissue processing of large animal spine mine if intradiscal injection of the sealant had a therapeutic
specimen can be a challenge at times. To process isolated effect on the degenerating intervertebral disc. With injec-
motion segments containing various devices often requires tion of the sealant, the authors observed an inhibition of
resorting to un-decalcified histology using plastic embed- fibrosis and improvements in proteoglycan levels (Buser
ding. Careful tissue processing starting with adequate tis- et al. 2011). Similarly, Chiang et al. (2011) demonstrated
sue fixation yields good-quality histological sections at an that sealing of annular defects via the modified purse-string
average thickness of 8 mm. Figures 18.1, 18.2, and 18.3 suture reduced degenerative changes in the disc following
represent key features from ovine lumbar intervertebral the injury.
discs with various degrees of degeneration. While there is Kaapa et al. studied porcine disc innervation and the
little evidence of intervertebral disc degeneration days fol- effects of degeneration on nerve topography. The authors
lowing the surgical intervention (chemonucleolysis using characterized nerve density, finding that in common with
chondroitinase ABC), at 6 and 18 months, the degenerative humans, the outer annulus contained most of the nerve
Fig. 18.1 Sections of two ovine vertebral bodies (L4L5) and inter- bone and the distinctly heterogeneous staining of collagen (red) and
vertebral disc. Low-power (orig mag 1) view of a section stained with proteoglycan (blue) in the disc. Insert: high-power view of Picrosirius
Picrosirius Red/Alcian Blue demonstrates normal intervertebral disc Red/Alcian Blue stained section (40). Proteoglycan-positive chondro-
morphology consisting of vertebral endplates (VEP, annulus fibrosus cytes and surrounding matrix stain blue. Residual collagen stains red
AF, and nucleus pulposus NP). The cartilaginous disc stains an intense (SB subchondral bone, CEP cartilaginous endplate, OA outer annulus);
blue. Note the high content of collagen (red) in cortical and trabecular unstained areas (picric acid) appear yellow
Fig. 18.2 Plastic embedded

sagittal sections of ovine lumbar
motion segments (L45) stained
with Picrosirius Red/Alcian Blue
at day 0 (a), 6 months (b), and
18 months (c) postoperative.
b
Choosing a left lateral
percutaneous approach through a
19-g docking needle, sheep were
injected with 300 ml of
chondroitinase ABC via the
dorsolateral corner into the center
of L45 discs using a 22-g spinal
needle. Figure (ac) demonstrates
progressive loss of overall disc c
height associated with decreased
proteoglycan content in the
nucleus pulposus (NP) and loss
of vertebral endplate integrity
with established subchondral bone
thickening (b, c). VEP vertebral
endplate. Sections were
photographed at 4
a b
20 40
Fig. 18.3 Intervertebral disc degeneration (IVDD) per se is not a basis IVDD may provide clues to the pathophysiology of discogenic pain. A
for clinical intervention: identification of specific features underlying preclinical model correlating the clinical course of IVDD as it is seen in
discogenic pain is of the utmost importance to advance the current level the humans would greatly enhance the predictability of clinical out-
of care and identify novel therapeutic targets. One critical deficiency in come when evaluating novel therapeutic concepts. Figure (a) and (b)
large animal models (sheep, goat, and cattle) is the lack of perceptible demonstrates sparse positive immunoreactivity (arrows) to CGRP and
pain during surgically induced IVDD. Symptoms derived from degen- SP confined to the dorsal longitudinal ligament (DLL) and outermost
erating discs may be classified into two types: (1) radicular pain sec- annulus fibrosus (OAF) in the lumbar disc of dogs without clinical his-
ondary to stenosis and nerve-root or cord irritation and (2) discogenic tory of back problems. These findings support current knowledge of
pain due to internal disc disruption. These syndromes may have distinct neuroanatomic similarities between the healthy human and canine
etiologic bases. Certain dog breeds that exhibit an insidious onset of intervertebral discs
endings (Kaapa et al. 1994a). Interestingly, this study found lar lesion model can be extended to study neural and vascu-
that disc injury had no effect on nerve topography lar ingrowth during the degradative process, a phenomenon
(Kaapa et al. 1994c). hypothesized to play a role in the pain experienced by some
human patients.
18.4.2 Ovine
18.4.3 Canine
The ovine model has also been used widely as a model for
disc research. Melrose et al. (2012) used annular incisions The canine model has also been popular in disc research.
to create mechanical destabilization and degeneration in Various insults have been applied to canine discs in order to
sheep. Three months after surgery, MRI examination dem- induce disc degeneration. Hutton et al. (2004) hypothesized
onstrated a reduction in disc height. Biomechanical testing that disc degeneration could be produced in canine models
to assess range of motion and to measure the neutral zone by limiting nutritional flow through application of bone
was performed along with histopathological examination. cement to the endplates; however, 70 weeks postsurgery, no
Measurement of aggrecan breakdown along with the expres- changes were observed (Hutton et al. 2004). In other studies,
sion of proteoglycan and collagen indicated that there were canine lumbar discs were subjected to compressive forces
progressive degradative changes. In an earlier study, 8 months using springs (Hutton et al. 1998, 2000). Despite application
after annular injury of the lumbar discs of sheep, Melrose of springs over an entire year, the authors reported that there
et al. (1992) found that while proteoglycan and collagen levels were no gross discal changes. They did, however, find differ-
had decreased, the amount of non-collagenous protein had ences in proteoglycan and collagen content between the con-
increased. The changes were primarily found in the nucleus trol and experimental groups. A study by Hasegawa et al.
pulposus and not the annulus fibrosus (Melrose et al. 1992). (1995) demonstrated that fatigue loading was more detri-
A subsequent study by the same investigators revealed mental to vertebral bodies adjacent to nucleotomized discs
increased expression of decorin and biglycan after annular when compared to controls. Cyclic compression loading of
damage. These proteins exhibit adverse effects on cell pro- nucleotomized discs resulted in an increase in microcrack
liferation in vitro and may further suppress the limited ability density on the adjacent vertebral bodies suggesting a possible
of the disc to heal after injury (Melrose et al. 1997). A later microtrauma mechanism for progressive disc degeneration
study by Melrose et al. (2002) suggested that the ovine annu- (Hasegawa et al. 1995).
Several investigations have tried to slow or reverse content. It should be noted that the degree of degeneration in
intervertebral disc degeneration in canine models via direct these studies prior to injection was relatively mild.
transplantation of autologous cells (Ganey and Meisel 2002;
Ganey et al. 2003, 2009). One study assessed the value of
autologous adipose tissue-derived stem cells in treating dam- 18.5 Advantages and Disadvantages
aged discs (Ganey et al. 2009). In each dog, three sequential of Large Animal Species Used
discs were injured via partial nucleotomy, while two adjacent in Disc Research
discs were spared and served as controls (Ganey et al. 2009).
Of the three discs with partial nucleotomies, one was injected 18.5.1 Porcine/Miniature Pig
with stem cells in a hyaluronic acid carrier, one was injected
with hyaluronic acid alone, and the last underwent no inter- Pigs are relatively simple to obtain and care for and usually
vention. Based on assessments of the overall disc morphol- moderate in cost (Cho et al. 2011). Other advantages of the
ogy and matrix production subsequent to transplantation, the mini-porcine model include the size and lordotic curvature
authors reported that treatment with stem cells in a hyaluronic of the cervical spine which provides a reasonable model of
carrier provided superior results. In another study, Ganey the human cervical spine (Chiang et al. 2011). Moreover,
et al. (2003) harvested autologous disc cells, cultured them biomechanical studies have shown that porcine and human
ex vivo, and returned them to the original canine disc via a spines are similar with respect to spinal flexibility and disc
percutaneous method. Not only did the transplanted cells dimensions (Lundin et al. 2000; Park et al. 2005). Although
successfully reintegrate into the disc, they also produced pigs are quadrupeds, biomechanical evidence suggests that
extracellular matrix. With respect to disc water content, fol- the forces acting on their spines are primarily along the long
low-up MRI was consistent with disc regeneration. Similarly, axis, similar to humans (Alini et al. 2008).
Hiyama et al. (2008) were also able to show signs of regen- Pig spines are most similar to human spines between the
eration in a canine post-nucleotomy model through the injec- T6 and T10 vertebral levels making these levels ideal for
tion of autologous disc cells expanded in culture. modeling thoracic conditions (Sheng et al. 2010). Moreover,
In contrast to the canine disc injury models, chondrodystro- pig discs are relatively large, facilitating surgical interven-
phic breeds have the potential to serve as a model for sponta- tions (Omlor et al. 2009). From a biochemical perspective,
neous disc degeneration. The genetic defect which influences the extracellular matrix composition and chondroitin sulfate
disc cell proliferation and maturation leads to an increased to keratan sulfate ratios in humans and pig discs are similar
pace of disc degeneration. These dogs experience the early (Kaapa et al. 1994b). Pigs, like dogs, retain notochordal cells
disappearance of notochordal cells, and there is replace- through adult life which may make the pig disc more resis-
ment of the nucleus pulposus with fibrocartilage (Melrose tant to degeneration and afford them greater regenerative
et al. 1996). Thus, they provide a unique model to study a potential compared to humans (Omlor et al. 2009).
spontaneous degenerative process within the disc of a larger The primary disadvantage of using full-sized porcine
animal model. Additionally, investigations into disc innerva- models is the rapid growth rate of most breeds; because of
tions have thus far revealed the sparse presence of nocicep- the subsequent weight gain, they are costly for long-term
tive fibers (Fig. 18.3). Whether there is indeed an increase of studies and also difficult to use for surgical interventions
nociception during disc injury or degenerative processes in (Sheng et al. 2010). To overcome this limitation, some work-
this species is subject of ongoing investigations. ers have turned to the mini pig model. Yoon et al. (2008)
were able to induce slow progressive disc degeneration,
confirmed by histopathology and MRI imaging, in miniature
18.4.4 Caprine pigs after annular injury.
Another shortcoming of the pig (as with most large ani-
Attempts have also been made to induce disc disease in mal models) is that young animals generally do not experi-
caprine models. Hoogendoorn et al. (2008) successfully ence spontaneous degeneration. Therefore, the degenerative
injected chondroitinase ABC directly into goat discs. A sub- process must be induced by annular or endplate injury (Yoon
sequent study by Zhang et al. (2011a, b) demonstrated that a et al. 2008; Omlor et al. 2009).
4.5-mm drill bit inserted 15 mm into the disc resulted in
significantly greater histological changes consistent with
disc degeneration compared to disc injuries using a scalpel 18.5.2 Ovine
blade. This group also studied the effects of injecting caprine
bone marrow stromal cells into degenerating discs and Sheep are also relatively easy to obtain and moderate in cost.
reported this intervention increased the proteoglycan In contrast to dogs and pigs, sheep are similar to humans
with regard to the absence of notochordal cells in adulthood As previously mentioned, non-chondrodystrophic canine
(Osti et al. 1990; Melrose et al. 1992; Alini et al. 2008). They breeds have persistent notochordal cells which may be pro-
also exhibit similarities in water content, collagen content, tective against degenerative changes and promote regenera-
and fiber orientation angles (Reid et al. 2002). Despite the tion following mild injuries. Chondrodystrophic breeds
similarity in cellular composition, sheep have relatively experience spontaneous degeneration, although the genetic
smaller spines and substantial anatomic incongruences when basis for the degenerative process is quite different from that
compared to humans. Upon comparing the anatomic dimen- of normal human discs (Sakai et al. 2009).
sions of sheep and human discs, Wilke et al. (1997a, b) found
that the greatest similarities were in the thoracic and lumbar
regions; however, there were disparities in all regions in 18.5.4 Caprine
terms of disc height: disc height in humans is greatest in the
lumbar region, while sheep have greater disc heights in the Goats are robust animals that are relatively easy to obtain
cervical region. In addition, there are fundamental anatomic and moderate in cost. One advantage of the caprine model is
differences including variations of the atlas and axis verte- that the disc height, size, and shape of certain breeds are
brae, pedicles, and posterior vertebral body heights (Wilke similar to those of humans (Zhang et al. 2011a, b). The C34
et al. 1997a, b). The range of motion between vertebral intervertebral disc in goats is particularly useful for model-
regions (e.g., cervical, thoracic, lumbar) is similar to the ing the C56 disc in humans (Hu et al. 2006). Grossly, goat
human (Wilke et al. 1997a, b). Kandziora et al. (2001) evalu- spines have similar geometric, anatomic, and loading charac-
ated 20 human and Merino sheep spines and concluded that teristics, while the nucleus pulposus displays comparable
despite significant differences in anatomy, the two were mechanical properties (Zhang et al. 2011a, b).
similar enough to use for in vivo modeling of human dis- The weight range of goats is similar to humans, they
eases, specifically endorsing the C34 motion segment as a tolerate surgery and anesthesia well, and they do not gain
reliable model of human cervical motion. enough weight to interfere with surgical manipulation
(Zhang et al. 2011a, b). Due to the shape of the goat torso,
which is much thinner side to side than the anterior-pos-
18.5.3 Canine terior dimension, a lateral approach to the lumbar spine is
relatively easy and can be performed using a minimally
Although dogs are abundant and relatively inexpensive to invasive approach (Zhang et al. 2011a, b). A disadvan-
maintain, their use for research in many communities pro- tage of the goat model is that it has not been used as
vokes ethical scrutiny. Moreover, due to the unique relation- widely in disc research as some other large animal
ship between humans and dogs, attitudes regarding the use of models.
canine models in research vary (Zhang et al. 2011a, b). In
some cases, social norms complicate the use of dogs in bio-
medical research, and ethical scrutiny has led to an increased 18.5.5 Nonhuman Primates
regulatory burden when performing canine research (Zhang
et al. 2011). From an evolutionary perspective, nonhuman primates
Nonetheless, because dog intervertebral discs have been have the greatest genetic similarity to humans and theoreti-
observed to have certain favorable similarities to humans, the cally should have the most conserved disc biology. Not
canine model has been used extensively (Nguyen et al. 1989). surprisingly, the baboon and human cervical spines are
The major differences between the dog and human spine very similar in terms of geometry and anatomic shape
have been described by Nguyen et al. (1989) and include (Sheng et al. 2010).
their smaller size and dissimilar disc shape (Nguyen et al. However, the lumbar region of the spine does exhibit
1989; Zhang et al. 2011a, b). notable differences that relate to upright posture and ambula-
In terms of motion, lateral bending and axial rotation are tion in the human. Nonhuman primates do not require the
coupled differently in humans and dogs at the C45 vertebral lordotic posture of humans in the lumbar region. Additionally,
levels. The dog cervical spine appears to have more coupled human lumbar discs (especially in the lower two lumbar seg-
motion with regard to lateral bending and axial rotation comments) are wedge shaped with greater anterior disc height.
pared to human spinal motion segments (Hofstetter et al. This allows for a greater range of extension and reduced
2009). However, compressive and torsional stiffness along flexion (Farfan 1978).
with the relative contribution of the posterior elements, discs, In spite of the genetic similarities, the principal barriers to
and ligament is similar between dog and human (Zimmerman the use of nonhuman primates for disc research are the sub-
et al. 1992). stantially greater cost of acquiring and maintaining these
relatively rare and highly social animals and the ethical tion. Collapsing disc space is one radiographic hallmark of
scrutiny that must be met before utilizing these intelligent disc degeneration. Subjective radiographic assessments
animals for biomedical research. require perfect lateral projections and are imprecise at best.
A robust method to assess radiographic findings can be a
powerful tool to corroborate findings from preclinical ani-
18.6 Guide to Selecting the Most mal trials (Fig. 18.4). Careful pre-experimental planning is
Useful Animal Model required to ensure that the minimal numbers of animals used
for the study are adequate to address the research question
The selection of an animal model should be guided by the and avoid performing an underpowered study. Gene expres-
unique research objectives of the specific study. No single sion studies generally require DNA sequence information
animal model identically reproduces all of the clinically rel- which is more readily available in certain animal species.
evant variables associated with the human disc (Alini et al. Studies in which the research objectives can be completed
2008); however, different animal models share with humans using advanced imaging, without animal sacrifice, may allow
specific anatomic and biologic characteristics which may be animals to be used for additional research so long as it does
relevant to a particular investigation. If the topic of interest is not confound the results of the first set of experiments (Chan
focused on an area in which a particular large animal model et al. 2011).
shares favorable anatomic, biomechanical, or biochemical When selecting an animal model for a specific purpose,
features with the human, the choice is fairly straightforward such as medical device development, anatomic dimensions
(Alini et al. 2008). may be of particular importance. For example, in order to
Assessing disc torsional biomechanics and collagen con- test a novel posterior instrumentation system, the investiga-
tent across an assortment of species, Showalter et al. (2012) tor may want to use an animal model with more humanlike
provided loose guidelines as to which species are the most pedicle morphology and vertebral body size. For testing
favorable for modeling of human disc degeneration. With interbody cages or an intervertebral disc prosthesis, disc and
respect to torsional mechanics, they found that 9 out of 11 vertebral body dimensions must to be taken into account
species were similar to humans; however, sheep and pig (McLain et al. 2002).
discs exhibited a significantly higher torsional stiffness. Calf, For studies of spinal motion, it is important that the model
pig, baboon, goat, sheep, lumbar discs, and cow caudal discs is biomechanically similar to humans. As an example,
are the most preferable species for studies investigating tor- Schmidt et al. (2005) found that the porcine model was bio-
sional forces on the spine. With regard to collagen content, mechanically similar to humans in flexion/extension, but
although the authors noted differences between species with markedly different for lateral bending and axial rotation.
regard to specific disc regions, in general, they concluded Thus, the porcine model would be best suited for the study of
there was relatively little variation across species. flexion/extension and not for lateral bending and rotation.
Unfortunately, few other similar cross-species studies have Finally, when working with large animal models, safety is
been performed to date. always a concern. The most obvious cause for concern is the
Cost-effectiveness is an important factor in choosing an potential for injury (bites, kicks); however, there are other
animal species for investigational studies. Larger, more intel- less obvious hazards (Langley 1999). These include, but are
ligent animals are expensive to acquire and maintain. not limited to, needle puncture and scalpel cuts, allergies
Typically, such animals require specialized housing and may related to dander or hair, and fall-related injuries on floors
need trained handlers. Furthermore, some animals take lon- slippery from animal urine, food, or feces (Langley 1999).
ger to manifest degenerative changes after an experimental Zoonotic diseases are also a potential concern (Weigler et al.
intervention, which can contribute to both study cost and 2005). Appropriate protocols and training should be set in
length (Singh et al. 2005). place and communicated to all staff working with animals to
The methods used to evaluate experimental results may insure that precautions are taken to reduce the risk of injury
also play a role in model selection. Histological examina- or disease exposure (Weigler et al. 2005).
tion, mechanical testing, and gene expression analysis all
provide different levels of information regarding the extent
of disc degeneration and the success of potential therapeutic 18.7 Summary of Critical Concepts
interventions (Chan et al. 2011). In vivo postoperative moni- Discussed in the Chapter
toring is critical and helps reduce animal numbers.
Methodologies may include digital radiography, computed Many small animal models are useful for studies involving
tomography, or magnetic resonance imaging. Digital radiog- intervertebral disc disease; however, large animal models are
raphy can be a cost-effective tool to monitor, i.e., disc space required for certain types of research and for regulatory
as a function of efficacy testing of an experimental interven- agency approval of certain medical devices.
a b 5
4
1
6
2 3
c d
.34
.32
.3
Mean DHI
.28 .26
.24
0 90 180 270 360

Time (days)
Fig. 18.4 To evaluate disc height changes over time, a robust method- The DHI for normal discs was consistent and reliable for multiple
ology adapted from Frobin et al. (1997) was developed and validated in observers. The disc collapsed after C-ABC injection by 12 % after
a cohort of sheep enrolled in an IVDD study involving chemonucleoly- 12 weeks postoperative and maximally by 16 % at 6 months before sta-
sis with chondroitinase ABC at L45. Digital radiographs in the lateral bilizing (d). DHI at all time points following C-ABC injection were
projection of the lumbar spine were obtained at 0, 2, 12, 26, and lower than at time zero (p < 0.01). This imaging analysis allows for
52 weeks (a). Radiographs were imported into LabVIEW (c) where effective in vivo monitoring during preclinical intervertebral disc
disc height changes were measured using specific anatomic landmarks research trials
(b) and the disc height index (DHI) calculated by blinded observers.
Large animal models can be used for studying disc biol- For a specific study, the research objectives, ethical con-
ogy, biomechanics, regeneration strategies, and medical siderations, budget, and availability of the model should
device development. be considered when choosing the preferred animal model.
Validation of each animal model is necessary. Significant
differences exist between species with regard to the pres-
ence of notochordal cells, the biologic predilection
towards disc degeneration, and the anatomic and physio-
References
logic features of the intervertebral disc. Thus, selection of Aguiar DJ, Johnson SL et al (1999) Notochordal cells interact with
the best animal model for a particular investigation will nucleus pulposus cells: regulation of proteoglycan synthesis. Exp
depend on the goals of the study. Cell Res 246(1):129137
Large animal model potentially useful for the study of Alini M, Eisenstein SM et al (2008) Are animal models useful for study-
ing human disc disorders/degeneration? Eur Spine J 17(1):219
degenerative disc disease includes sheep, goat, pig/minia- An H, Boden SD et al (2003) Summary statement: emerging techniques
ture pigs, runt cow, nonhuman primates, dog, chicken, for treatment of degenerative lumbar disc disease. Spine (Phila Pa
kangaroo, and ostrich. 1976) 28(15 Suppl):S24S25
Beckstein JC, Sen S et al (2008) Comparison of animal discs used in Hutton WC, Toribatake Y et al (1998) The effect of compressive force
disc research to human lumbar disc: axial compression mechanics applied to the intervertebral disc in vivo. A study of proteoglycans
and glycosaminoglycan content. Spine (Phila Pa 1976) 33(6): and collagen. Spine (Phila Pa 1976) 23(23):25242537
E166E173 Hutton WC, Ganey TM et al (2000) Does long-term compressive load-
Berry RJ (1961) Genetically controlled degeneration of the nucleus ing on the intervertebral disc cause degeneration? Spine (Phila Pa
pulposus in the mouse. J Bone Joint Surg Br 43:387393 1976) 25(23):29933004
Buser Z, Kuelling F et al (2011) Biological and biomechanical effects Hutton WC, Murakami H et al (2004) The effect of blocking a nutri-
of fibrin injection into porcine intervertebral discs. Spine (Phila Pa tional pathway to the intervertebral disc in the dog model. J Spinal
1976) 36(18):E1201E1209 Disord Tech 17(1):5363
Chan DD, Khan SN et al (2011) Mechanical deformation and Iatridis JC, Mente PL et al (1999) Compression-induced changes in
glycosaminoglycan content changes in a rabbit annular puncture intervertebral disc properties in a rat tail model. Spine (Phila Pa
disc degeneration model. Spine (Phila Pa 1976) 36(18):14381445 1976) 24(10):9961002
Chiang CJ, Cheng CK et al (2011) The effect of a new anular repair Kaapa E, Gronblad M et al (1994a) Neural elements in the normal and
after discectomy in intervertebral disc degeneration: an experimen- experimentally injured porcine intervertebral disk. Eur Spine J 3(3):
tal study using a porcine spine model. Spine (Phila Pa 1976) 36(10): 137142
761769 Kaapa E, Holm S et al (1994b) Collagens in the injured porcine inter-
Cho H, Park SH et al (2011) Snapshot of degenerative aging of porcine vertebral disc. J Orthop Res 12(1):93102
intervertebral disc: a model to unravel the molecular mechanisms. Kaapa E, Holm S et al (1994c) Proteoglycan chemistry in experimentally
Exp Mol Med 43(6):334340 injured porcine intervertebral disk. J Spinal Disord 7(4):296306
Elliott DM, Sarver JJ (2004) Young investigator award winner: valida- Kalichman L, Hunter DJ (2008) Diagnosis and conservative manage-
tion of the mouse and rat disc as mechanical models of the human ment of degenerative lumbar spondylolisthesis. Eur Spine J
lumbar disc. Spine (Phila Pa 1976) 29(7):713722 17(3):327335
Farfan HF (1978) The biomechanical advantage of lordosis and hip Kandziora F, Pflugmacher R et al (2001) Comparison between sheep
extension for upright activity. Man as compared with other anthro- and human cervical spines: an anatomic, radiographic, bone mineral
poids. Spine (Phila Pa 1976) 3(4):336342 density, and biomechanical study. Spine (Phila Pa 1976) 26(9):
Frobin W et al (1997) Precision measurement of disc height, vertebral 10281037
height and sagittal plane displacement from lateral radiographic Langley R (1999) Physical hazards of animal handlers. Occup Med
views of the lumbar spine. Clin Biomech 12:S1S63 14(2):181194
Galbusera F, Schmidt H et al (2011) The mechanical response of the Lotz JC (2004) Animal models of intervertebral disc degeneration: les-
lumbar spine to different combinations of disc degenerative changes sons learned. Spine (Phila Pa 1976) 29(23):27422750
investigated using randomized poroelastic finite element models. Lotz JC, Colliou OK et al (1998) Compression-induced degeneration of
Eur Spine J 20(4):563571 the intervertebral disc: an in vivo mouse model and finite-element
Ganey TM, Meisel HJ (2002) A potential role for cell-based study. Spine (Phila Pa 1976) 23(23):24932506
therapeutics in the treatment of intervertebral disc herniation. Eur Lundin O, Ekstrom L et al (2000) Exposure of the porcine spine to
Spine J 11(Suppl 2):S206S214 mechanical compression: differences in injury pattern between
Ganey T, Libera J et al (2003) Disc chondrocyte transplantation in a adolescents and adults. Eur Spine J 9(6):466471
canine model: a treatment for degenerated or damaged interverte- McLain RF, Yerby SA et al (2002) Comparative morphometry of L4
bral disc. Spine (Phila Pa 1976) 28(23):26092620 vertebrae: comparison of large animal models for the human lumbar
Ganey T, Hutton WC et al (2009) Intervertebral disc repair using spine. Spine (Phila Pa 1976) 27(8):E200E206
adipose tissue-derived stem and regenerative cells: experiments in a Melrose J, Ghosh P et al (1992) A longitudinal study of the matrix
canine model. Spine (Phila Pa 1976) 34(21):22972304 changes induced in the intervertebral disc by surgical damage to the
Hasegawa K, Turner CH et al (1995) Effect of disc lesion on microdam- annulus fibrosus. J Orthop Res 10(5):665676
age accumulation in lumbar vertebrae under cyclic compression Melrose J, Taylor TK et al (1996) Variation in intervertebral disc serine
loading. Clin Orthop Relat Res 311:190198 proteinase inhibitory proteins with ageing in a chondrodystrophoid
Higuchi M, Kaneda K et al (1982) Age-related changes in the nucleus (beagle) and a non-chondrodystrophoid (greyhound) canine breed.
pulposus of intervertebral disc in mice. An electronmicroscopic Gerontology 42(6):322329
study. Nihon Seikeigeka Gakkai Zasshi 56(4):321329 Melrose J, Ghosh P et al (1997) Elevated synthesis of biglycan and
Hiyama A, Mochida J et al (2008) Transplantation of mesenchymal decorin in an ovine annular lesion model of experimental disc
stem cells in a canine disc degeneration model. J Orthop Res 26(5): degeneration. Eur Spine J 6(6):376384
589600 Melrose J, Roberts S et al (2002) Increased nerve and blood vessel
Hofstetter M, Gedet P et al (2009) Biomechanical analysis of the three- ingrowth associated with proteoglycan depletion in an ovine anular
dimensional motion pattern of the canine cervical spine segment lesion model of experimental disc degeneration. Spine (Phila Pa
C4-C5. Vet Surg 38(1):4958 1976) 27(12):12781285
Holm S, Ekstrom L et al (2007) Intradiscal pressure in the degenerated Melrose J, Shu C et al (2012) Mechanical destabilization induced by
porcine intervertebral disc. Vet Comp Orthop Traumatol 20(1): controlled annular incision of the intervertebral disc dysregulates
2933 metalloproteinase expression and induces disc degeneration. Spine
Hoogendoorn RJ, Helder MN et al (2008) Reproducible long-term disc (Phila Pa 1976) 37(1):1825
degeneration in a large animal model. Spine (Phila Pa 1976) 33(9): Nguyen CM, Haughton VM et al (1989) A model for studying interver-
949954 tebral disc degeneration with magnetic resonance and a nucleotome.
Hu N, Cunningham BW et al (2006) Porous coated motion cervical disc Invest Radiol 24(5):407409
replacement: a biomechanical, histomorphometric, and biologic OConnell GD, Vresilovic EJ et al (2007) Comparison of animals used
wear analysis in a caprine model. Spine (Phila Pa 1976) 31(15): in disc research to human lumbar disc geometry. Spine (Phila Pa
16661673 1976) 32(3):328333
Hunter CJ, Matyas JR et al (2004) Cytomorphology of notochordal and Omlor GW, Nerlich AG et al (2009) A new porcine in vivo animal
chondrocytic cells from the nucleus pulposus: a species compari- model of disc degeneration: response of anulus fibrosus cells,
son. J Anat 205(5):357362 chondrocyte-like nucleus pulposus cells, and notochordal nucleus
pulposus cells to partial nucleotomy. Spine (Phila Pa 1976) 34(25): Silberberg R (1988) Histologic and morphometric observations on
27302739 vertebral bone of aging sand rats. Spine (Phila Pa 1976) 13(2):
Osti OL, Vernon-Roberts B et al (1990) 1990 Volvo Award in experi- 202208
mental studies. Anulus tears and intervertebral disc degeneration. Singh K, Masuda K et al (2005) Animal models for human disc degen-
An experimental study using an animal model. Spine (Phila Pa eration. Spine J 5(6 Suppl):267S279S
1976) 15(8):762767 Smith LJ, Nerurkar NL et al (2011) Degeneration and regeneration of
Palmer EI, Lotz JC (2004) The compressive creep properties of normal the intervertebral disc: lessons from development. Dis Model Mech
and degenerated murine intervertebral discs. J Orthop Res 22(1): 4(1):3141
164169 Urban JP, Roberts S et al (2000) The nucleus of the intervertebral disc
Panjabi MM (1998) Cervical spine models for biomechanical research. from development to degeneration. Am Zool 40:5361
Spine (Phila Pa 1976) 23(24):26842700 Wagner PC, Bagby GW et al (1979) Surgical stabilization of the equine
Park C, Kim YJ et al (2005) An in vitro animal study of the biome- cervical spine. Vet Surg 8(1):712
chanical responses of anulus fibrosus with aging. Spine (Phila Pa Weigler BJ, Di Giacomo RF et al (2005) A national survey of labora-
1976) 30(10):E259E265 tory animal workers concerning occupational risks for zoonotic
Reid JE, Meakin JR et al (2002) Sheep lumbar intervertebral discs as diseases. Comp Med 55(2):183191
models for human discs. Clin Biomech (Bristol, Avon) 17(4): Wilke HJ, Kettler A et al (1997a) Are sheep spines a valid biomechani-
312314 cal model for human spines? Spine (Phila Pa 1976) 22(20):
Sakai D, Nakai T et al (2009) Differential phenotype of intervertebral 23652374
disc cells: microarray and immunohistochemical analysis of canine Wilke HJ, Kettler A et al (1997b) Anatomy of the sheep spine and its
nucleus pulposus and anulus fibrosus. Spine (Phila Pa 1976) 34(14): comparison to the human spine. Anat Rec 247(4):542555
14481456 Yoon SH, Miyazaki M et al (2008) A porcine model of intervertebral
Schmidt R, Richter M et al (2005) Limitations of the cervical disc degeneration induced by annular injury characterized with
porcine spine in evaluating spinal implants in comparison with magnetic resonance imaging and histopathological findings.
human cervical spinal segments: a biomechanical in vitro compari- Laboratory investigation. J Neurosurg Spine 8(5):450457
son of porcine and human cervical spine specimens with different Zhang Y, Drapeau S et al (2011a) Histological features of the degener-
instrumentation techniques. Spine (Phila Pa 1976) 30(11): ating intervertebral disc in a goat disc-injury model. Spine (Phila Pa
12751282 1976) 36(19):15191527
Sheng SR, Wang XY et al (2010) Anatomy of large animal spines Zhang Y, Drapeau S et al (2011b) Transplantation of goat bone marrow
and its comparison to the human spine: a systematic review. Eur stromal cells to the degenerating intervertebral disc in a goat disc
Spine J 19(1):4656 injury model. Spine (Phila Pa 1976) 36(5):372377
Showalter BL, Beckstein JC et al (2012) Comparison of animal discs Zimmerman MC, Vuono-Hawkins M et al (1992) The mechanical prop-
used in disc research to human lumbar disc: torsion mechanics and erties of the canine lumbar disc and motion segment. Spine (Phila
collagen content. Spine (Phila Pa 1976) 37(15):E900E907 Pa 1976) 17(2):213220
Intervertebral Disc Herniation:
Pathophysiology and Emerging 19
Therapies
Beth A. Winkelstein, Kyle D. Allen, and Lori A. Setton

19.1 Introduction ..................................................................... 305
Approximately 2.6 % of the US population visits a physician
19.2 Pathophysiology and Pain of Intervertebral
Disc Herniation ................................................................ 306
for treatment of spinal disorders annually (Fraser 2009) with
costs of $7.1 billion from lost work days alone (Ricci et al.
19.3 Animal Models of Intervertebral Disc Herniation ....... 308
2006). Herniation of the intervertebral disc is one of the
19.3.1 Mechanical Factors in Intervertebral Disc Herniation ...... 310
19.3.2 Role of Chemical and Inflammatory Mediators several spinal disorders that contribute to this very high inci-
in Disc Herniation ............................................................. 313 dence, with potential to cause significant pain, neurological
19.4 Emerging Therapies for Intervertebral deficit, and functional disability in affected individuals.
Disc-Associated Radiculopathy ...................................... 315 Herniation presents as a protrusion or extrusion of discal tis-
19.4.1 Cytokine Antagonism ....................................................... 317 sue into the epidural cavity, resulting in nerve root impinge-
19.4.2 Neuronal Receptor Blockers ............................................. 318 ment and disc tissue exposure (Fig. 19.1). Both mechanical
19.4.3 Cell Cycle Modifiers ......................................................... 319
19.4.4 Pathway Inhibitors ............................................................ 319 compression and tissue exposure contribute to pain and dis-
ability associated with intervertebral disc herniation (Goupille
19.5 Final Comments .............................................................. 319
et al. 1998; Mixter et al. 1934; Olmarker and Rydevik 1991).
19.6 Summary of Critical Concepts Discussed In areas innervated by the affected nerves, it is commonly
in the Chapter .................................................................. 319 seen as low back pain, radiating leg pain (i.e., radiculopathy
References ...................................................................................... 320 or sciatica), muscle weakness, gait abnormality, muscle atro-
phy, asymmetric reflexes, or loss of function (Atlas et al.
2005; Frymoyer 1988; Hart et al. 1995). The incidence of
sciatica related to intervertebral disc herniation peaks
between the fourth and fifth decades of life and is most fre-
quently associated with herniations between the L3 and S1
vertebral levels (Atlas et al. 2005; Awad and Moskovich
2006). The severity of herniation symptoms in the cervical or
lumbar regions has been shown to relate to the size or nature
of the herniated fragment, whether it is simply protruding
B.A. Winkelstein, PhD into the neural cavity, extruded, or completely sequestered
Departments of Bioengineering and Neurosurgery,
from the parent structure.
University of Pennsylvania,
Philadelphia, PA, USA Intervertebral disc herniation may also occur in associa-
e-mail: winkelst@seas.upenn.edu tion with disc degeneration, wherein degenerated nucleus
K.D. Allen, PhD pulposus fragments migrate into previously established
J. Crayton Pruitt Family Department of Biomedical Engineering, defects in the anulus fibrosus (Moore et al. 1996). A desic-
University of Florida, Gainesville, FL, USA cated and fibrous nucleus pulposus is associated with loss of
e-mail: kyle.allen@bme.ufl.edu
disc height and an increased axial disc bulge with compres-
L.A. Setton, PhD (*) sive loading (Adams and Roughley 2006); the altered tissue
Departments of Biomedical Engineering and Orthopaedic Surgery,
can generate untoward stresses upon the anulus fibrosus lead-
Duke University, 136 Hudson Hall, 90281,
Durham, NC 27708, USA ing to tissue fragment extrusion. Thus, while intervertebral
e-mail: setton@duke.edu disc degeneration is positively associated with disc herniation,

306 B.A. Winkelstein et al.
Spinous process
Meninges
Formen
Gray matter
(filled with adipose tissue)
White matter
Inferior articular process Dorsal root
Ventral root
Superior articular process
Spinal nerve
Posterior tubercle of
transverse process
Anterior tubercle of
transverse process Nucleus pulposus
Herniated disc impinging
Foramen transversium Vertebral body Disc anulus on spinal nerve
Fig. 19.1 Schema showing a cross section of the intervertebral disc and spinal nerves as they exit through the intervertebral foramen. (Right)
anatomy of the spinal cord and posterior vertebral processes. (Left) Normal Schema of anatomic changes typical of a posterolateral intervertebral disc
anatomy showing labels identifying the intervertebral disc anulus fibrosus herniation, suggesting nerve root impingement (Used under CCL3.0,
and nucleus pulposus, as well as the nerve roots which come together as http://en.wikipedia.org/wiki/File:Cervical_vertebra_english.png)
it can be difficult to identify the specific contributions of bio- anti-inflammatory and opioid analgesics. While more
mechanical, environmental, or genetic factors. Once the tis- invasive, epidural administration of anesthetics, such as bupi-
sue is protruded or herniated, there is evidence for increased vacaine, and/or corticosteroids, like methylprednisolone or
angiogenesis, macrophage infiltration, and proteinase pro- triamcinolone, shows some efficacy in providing symptom
duction in the tissue fragment that can contribute to its relief, although again without evidence of disease modification
resorption over a period of months to years (Komori et al. (Buenaventura et al. 2009; Staal et al. 2008).
1996). While nonsurgical care is the first treatment option (see
The factors listed above are also believed to play impor- Chap. 15), surgical intervertebral disc excision is the most
tant roles in the etiology and pathophysiology of interverte- frequently performed musculoskeletal procedure in the
bral disc herniation (Adams and Roughley 2006; Battie and USA, affecting 0.3 % of the population with hospitalization
Videman 2006). Mechanical loading of the lumbar spine in costs of $9.5 billion (Fraser 2009; Ricci et al. 2006). The
work-related loading conditions may be sufficient to cause surgical approach for painful disc herniation is simply to
disc herniation. Epidemiological studies further suggest a remove the inflammatory material and to unload and decom-
role for mechanical factors and nutrient transport in interver- press the adjacent neural structures (Loupasis et al. 1999).
tebral disc herniation with higher incidences of herniation While frequently successful in providing relief, compres-
and sciatica associated with obesity, smoking, and heavy sion-relieving discectomy may not prevent recurrence of
physical work (this topic is discussed in considerable detail pain, prompting the need for additional surgeries, or in
in Chap. 9) (Battie et al. 2009; Bostman 1993; Heliovaara 2060 % of patients, there may be repeat herniations
1987a, b; Heliovaara et al. 1987a, b). Genetic predisposition (Weinstein et al. 2006a, b).
to lumbar radiculopathy may also exist, with data suggesting
that disc herniation may be associated with genetic muta-
tions in the a2 and a3 chains of collagen IX or the regulatory 19.2 Pathophysiology and Pain of
cytokines interleukin-1b (IL-1b) and interleukin-6 (IL-6) Intervertebral Disc Herniation
(see Chap. 11).
For intervertebral disc herniation, the first treatment of Both chemical and mechanical factors are widely believed to
choice is conservative, unless motor weakness or loss of contribute to radicular pain subsequent to intervertebral disc
function is a significant concern. Lifestyle modifications, in herniation (Olmarker 2001). The herniated fragment may
particular exercise and physical therapy, can provide symp- impinge upon the exiting spinal nerve and contribute to nerve
tomatic relief and reduce the need for operative treatment, root compression with many associated deleterious effects.
although they are not generally considered to be disease- The root consists of both anterior and posterior rootlets exit-
modifying therapies (Weinstein et al. 2006a, b). In addition, ing from the spinal nerve that combine to form the dorsal and
conventional pharmacological interventions are widely ventral nerve roots containing sensory and efferent fibers,
prescribed for sciatica including orally administered respectively (Fig. 19.2). The roots come together in the region
19 Intervertebral Disc Herniation: Pathophysiology and Emerging Therapies 307
Sympathetic
ganglion
Cell of Dogiel
Spinal ganglion 2
6
7
Posterior
nerve root
Spinal nerve
Sympathetic
cord
2 Anterior
1
nerve root
6
3 1
3
Wh 4
ite 5
ram
Gr us
ay com
mun
ram ican
us s
com
mun Sympathetic
ican
s
ganglion
Fig. 19.2 Scheme showing structure of a typical spinal nerve and spinal ganglion. 1 Somatic efferent, 2 somatic afferent, 3, 4, 5 sympathetic
efferent, 6, 7 sympathetic afferent (Figure 799 from Grays Anatomy (Used under CCL3.0, http://en.wikipedia.org/wiki/File:Gray799.svg))
of the neural foramen and continue more distally into the decreased 41 % at as early as 1 min following positioning
periphery of the spinal nerve to innervate the structures out- and decreased by 63 % after 3 min (Takamori et al. 2011).
side of the spinal column. Unlike nerves, the nerve roots are The change in evoked action potential continued to develop
not enclosed by a thick epineural sheath, and so they lack the through the period of nerve root impingement, demonstrating
mechanical resilience of their peripheral counterparts; hence, that time-dependent electrophysiologic responses occur in
their structural anatomy places them at particular risk for the nerve root with compression. These changes in neuronal
injurious mechanical loading under intervertebral disc her- signaling probably contribute to radiculopathy symptoms.
niation. In addition, the cell bodies of peripheral nerves are Together, the magnitude, duration, and rate of compression
housed in the dorsal root ganglion (DRG) and can respond to of the nerve root modulate both the extent of the local tissue
even slight compression with sustained neuronal activity and damage and the degree and duration of the pain symptoms
pain (Hanai et al. 1996; Hu and Xing 1998; Van Zundert et al. (Kobayashi et al. 2005b; Olmarker et al. 1989; Rothman et al.
2006). Intraoperative studies in patients with disc herniation 2010; Rydevik et al. 1991; Winkelstein et al. 2002).
and associated root impingement demonstrate decreases in As indicated above, intervertebral disc herniation can con-
the amplitude of compound muscle action potentials follow- tribute to a type of neuropathic pain that has, as its most com-
ing electrical stimulation to the affected nerve root (Morishita mon characteristic symptom, radicular pain or radiculopathy.
et al. 2006; Takamori et al. 2011). This work provides direct Self-reported pain and disability scales, such as the visual ana-
support for the clinical observation that functional impair- logue scale (VAS) or Oswestry Disability Index, are often used
ment is associated with disc herniation. Reproduction of pain to provide measures of pain or functional loss associated with
generating straight-leg raises in patients undergoing surgery disc herniation. Clinically, the straight-leg raise test (or SLR)
showed that the amplitude of the evoked action potentials is considered the most sensitive approach for quantifying the
Table 19.1 Molecular mediators of radiculopathy identified in intervertebral disc tissues. Annotated findings in footnotes
Mediator Notes Citation
TNF-a 13,6 Weiler et al. (2005), Takahashi (1996), Demircan (2007), Ahn (2002), Le Maitre et al. (2007),
Nygaard (2007)
ICAM-1 1 Doita et al. (1996)
Interleukin-1a 1,2 Le Maitre (2005), Takahashi (1996), Ahn (2002)
Interleukin-1b 1,3,7 Demircan (2007), Takahashi (1996), Le Maitre et al. (2005, 2007)
Interleukin-17 3 Shamji et al. (2010)
Interleukin-4 3,5 Shamji et al. (2010), Park (2002)
Interleukin-6 1,3,6 Demircan (2007), Takahashi (1996), Kang (1997), Burke et al. (2002), Shamji et al. (2010),
Specchia (2002), Nygaard (2007)
Interleukin-8 1,2,4 Demircan (2007), Ahn (2002), Burke et al. (2002)
Interleukin-12 3,4 Shamji et al. (2010), Park (2002)
Interleukin-20 1 Huang (2008)
Interferon-g 1,3,4 Demircan (2007), Shamji et al. (2010), Park (2002)
Leukotriene-B4 Demircan (2007), Nygaard (2007), Willburger (1994)
Thromboxane-B2 1,4,7 Demircan (2007), Nygaard (2007)
Phospholipase A 1 Saal (1990)
Prostaglandin E2 1,3 ODonnell (1996), Kang (1997), Wilburger (1994)
1. Expression noted in degenerative and/or herniated IVD
2. Protein or mRNA expression revealed in herniated IVD alone
3. Herniated or degenerated IVD > non-degenerative or autopsy control
4. Uncontained > contained herniations or non-degenerative control
5. Expression in uncontained < contained herniations
6. No evidence of protein or mRNA expression in herniated discs
7. Little or no spontaneous expression detected in herniated IVD
degree of radicular pain originating in the lumbar region, as inflammatory mediators and proinflammatory cytokines,
pain is more pronounced upon elevation of the leg (van der such as IL-1a, IL-6, and TNF-a (Table 19.1). Elevated
Windt et al. 2010). Similarly, positions that reproduce pain expression of these molecules may activate the immune sys-
such as forward flexion, hyperextension, and slump are some- tem and upregulate the expression of proteinases important
times used to corroborate findings of radicular pain, although for fragment resorption; nerve root compression alone may
imaging is commonly needed to confirm the pain source is also upregulate the expression of many of these same
related to intervertebral disc herniation. In patients, physical inflammatory mediators (Kobayashi et al. 2005b). Many
tests of muscle weakness, impaired reflexes, and sensory mediators are known generators of pain, such that they have
deficits or hypersensitivity may also be used to detect impair- become therapeutic targets for new and emerging studies
ment associated with radiculopathy. These diagnostic and (reviewed at the end of this chapter).
quantitative assessments are relevant to animal models as no
diagnostic biomarkers that span human to animal models have
yet been developed for intervertebral disc herniation radicul- 19.3 Animal Models of Intervertebral
opathy (Brisby et al. 2002; Gajendran et al. 2011; Tokunaga Disc Herniation
et al. 2010). Tests of hypersensitivity to non-noxious (allo-
dynia) and noxious stimuli (hyperalgesia) serve as surrogate To advance therapeutic options that can change disc hernia-
measures of sensory changes with disc herniation and can tion outcomes, it is of critical importance to understand the
serve as indicators of neuropathic pain in both human subjects molecular mechanisms that regulate pain, muscle changes,
and animals. A heightened response to a light brush of the skin and dysfunction. Animal models of herniation have been
would be considered a sign of mechanical allodynia, while a extensively studied for this purpose. These models fall into
heightened response to pinch would be considered a sign of two categories: (1) models designed to mimic direct nerve
mechanical hyperalgesia. Patients presenting with herniation root compression in a well-controlled manner using plung-
may experience either or both mechanical or/and thermal allo- ers, constriction, or graded compression (Hou et al. 2003;
dynia, and these serve as important metrics of pain-related Kallakuri et al. 2005; Kawakami et al. 2003; Onda et al.
behaviors in animal models of intervertebral disc herniation. 2005; Sekiguchi et al. 2009); (2) models designed to mimic
Separate from nerve root compression, a herniated disc chemical injury that can elicit an inflammatory response
fragment may evoke an inflammatory and immune response through the use of chemical irritants (Colburn et al. 1999;
near an affected root (Olmarker and Larsson 1998). Indeed, Hashizume et al. 2000a; Hubbard and Winkelstein 2005;
the herniated tissue fragment is a known generator of many Kajander et al. 1996; Kawakami et al. 1994a, b; Maves et al.
1993; Olmarker et al. 1993; Winkelstein and DeLeo 2004); frequency of head turns toward a respective limb, leg lifts,
or (3) direct application of nucleus pulposus tissue to the duration of spontaneous grooming activity on a given limb,
nerve root (Allen et al. 2011; Brisby et al. 2000; Cuellar et al. and wet-dog shakes (see Chap. 16) (Nakamae et al. 2011;
2005; Kawakami et al. 1999; McCarron et al. 1987; Olmarker Nilsson et al. 2011; Olmarker 2008; Olmarker et al. 2002,
et al. 1997; Olmarker and Myers 1998; Otani et al. 1997; 2003). Alternatively, quantitative measures of gait have been
Sekiguchi et al. 2008; Shamji et al. 2009). All of these ani- used to identify compensations and disabilities resulting
mal models mimic key features of painful radiculopathy such from intervertebral disc herniation (Allen et al. 2011;
as limb allodynia or hyperalgesia. The most commonly Shamji et al. 2009), made easier by the recent introduction
reported measure of limb hypersensitivity in preclinical of digitized gait apparata including the CatWalkTM,
models of intervertebral disc herniation is mechanical allo- TreadscanTM, and DigigaitTM systems (Beare et al. 2009;
dynia, detected using von Frey microfilaments applied to the Berryman et al. 2009; Gensel et al. 2006; Piesla et al. 2009;
plantar region of an animals paw (Fig. 19.5) (Colburn et al. Vrinten and Hamers 2003). Tracking of spatial and temporal
1999; Kallakuri et al. 2005; Kobayashi et al. 2005b; Shamji gait parameters, such as the geometric position of the limb
et al. 2009; Rothman et al. 2010; van der Windt et al. 2010). and paw ground contact times, can be used to obtain mea-
By measuring the frequency or response of withdrawal from sures such as stride length, step width, toe-out angle, stance
a given filament, the sensitivity of a limb to non-noxious times, gait symmetry (a measure of limping), and running
mechanical stimuli can be assessed. Mechanical hyperalge- velocity (Fig. 19.3). Dynamic data describe forces and
sia is also a measure of limb sensitivity and can be deter- moments that occur during a gait cycle and include measures
mined by recording the time spent grooming following a such as ground reaction forces and moments (Crawley 2007;
pinprick or assessing the mechanical pinch force required to Whishaw and Kolb 2005). These parameters are standard
initiate an animal response (RandallSelitto device). These analytical tools in the study of musculoskeletal injury and
measures have some gross relationships to pain and serve as pathology and are more recently being used to objectively
surrogate measures of underlying neuropathology with inter- quantify pain-related behaviors and functional losses in ani-
vertebral disc herniation. mal models of disc herniation. The major models used to
Pain or dysesthesia can also elicit changes in grooming, study mechanisms that contribute to symptoms of interverte-
locomotion, or sensorimotor skills in animal models. Pain in bral disc herniation are reviewed here, followed by an
disc herniation models has been identified by recording key overview of developments in the area of emerging nonsurgi-
features of animal behavior over a period of time, such as the cal therapies.
Left foot Stance Swing Stance Swing
Left foot-strike Left foot-strike Balanced: (Left%Stance Right%Stance) 0

Normal gait
Right foot Swing Stance Swing Stance

(LeftFoot-strike2 RightFoot-strike1)
Right foot-strike Right foot-strike Symmetric: 0.5
(LeftFoot-strike2 LeftFoot-strike1)
Time
Left foot Stance Swing Stance Swing

Unilateral injury
Imbalanced: (Left%Stance Right%Stance) 0

Left foot-strike Left foot-strike
Right foot Swing Stance Swing Stance
(LeftFoot-strike2 RightFoot-strike1)
Assymmetric: 0.5
Right foot-strike Right foot-strike (LeftFoot-strike2 LeftFoot-strike1)
Time
6% * 0.58
Stance time balance
# Pre-operative trials (n = 12)

4% 0.56
Gait symmetry
Sham surgery (n = 6)
2% 0.54
NP on right L5 DRG (n = 6)
0.52 * Balance 0 %, p < 0.05
0%
0.50 # Symmetry 0.5, p < 0.05
-2 % Data presented as mean st. err.
0.48
Pre-op 1 week 4 week Pre-op 1 week 4 week
Fig. 19.3 Definitions of temporal parameters obtained from quantita- between two left foot-strike events. (Middle) For unilateral right limb
tive gait analysis. Rats freely ambulate across a clear gait chamber injury, increased time is spent on the left limb and decreased time is
while digital video is acquired for determination of geometric positions spent on the right limb, as seen relative shifts in stance times. Also, the
of both hindpaws during each gait cycle. As shown, parameters of foot-strike sequence becomes syncopated in time, with the right limb
stance time, gait symmetry, as well as self-selected velocity, stance foot strike occurring past the midway point of two left limb foot strikes.
width, stride length, and more can be measured. (Top) Schema of a (Bottom) Data for stance time imbalance and gait symmetry of rats in a
normal gait sequence wherein stance times are similar on the left and nucleus pulposus injury model of radiculopathy as compared to sham
right hind limbs and a right foot-strike event occurs at the midpoint controls (Plotted from Hwang et al. (2012), used under CCL3.0)
Box 19.1 DRG and Neuronal Cell Culture ing disc-mediated pain since they can provide a cellu-
Intervertebral disc herniation can initiate cellular-level lar-level characterization of neuronal responses to
changes in the neurons and afferents in the dorsal root relevant stimuli related to disc herniation. They enable
ganglion (DRG). Because many of these neuronal doseresponse studies and real-time functional assays,
changes can contribute to central sensitization that as well as provide an exciting platform for screening
induces sustained nociceptive responses in the spinal potential therapeutics.
cord as well as pain, DRG culture studies are a useful
proxy for investigating the pathophysiology of disc
herniation. In particular, cell culture systems have the 19.3.1 Mechanical Factors in Intervertebral
potential to recapitulate the cellular environment of Disc Herniation
primary afferents of the DRG via chemical and/or
mechanical stimuli. In some cases, multi-compartment Nerve root compression in association with disc herniation
in vitro systems enable more complicated systems can result from acute insult to the axonal, connective, and
approaches to modeling synaptic connections, and vascular tissues of the nerve root and initiate a cascade of
integration of multi-electrode arrays in these setups related and integrated neuronal, inflammatory, and degenera-
also enables neuron-level assessments of function. In tive changes (Kobayashi et al. 2004b; Rydevik et al. 1984,
addition, culture preparations have been used to evalu- 1994; Winkelstein et al. 2002). From this perspective,
ate neuronal responses to challenges such as cytokines mechanical injury and inflammatory injury are not unique,
and other inflammatory mediators known to be involved nor unlinked, events. Nerve root compression following
in disc-mediated pain. acute mechanical loading can induce long-term nerve root
A variety of specialized techniques have been devel- pathophysiology, such as edema, inflammation, and thicken-
oped to isolate, maintain, and manipulate isolated neu- ing of connective tissues (Beck et al. 2010; Jancalek and
rons and intact DRGs from rodents. Briefly, immediately Dubovy 2007; Kobayashi et al. 2004b; Mosconi and Kruger
following perfusion with KrebsRinger bicarbonate 1996), as well as the development of evoked pain via modified
buffer, laminectomies and facetectomies are performed communication with the spinal cord. Following mechanical
to expose the DRG which is removed and immediately trauma and compression, injured axons exhibit axonal swell-
placed in Krebs buffer on ice. For isolation of intact ing, loss of cytoskeleton proteins, separation and disorgani-
neurons, the DRG is digested in collagenase under zation of the myelin sheath, loss of axonal transport, Wallerian
sterile conditions in Hanks Balanced Salt Solution, degeneration, and a decrease in axon packing density
trypsinized, and dissociated by trituration. After trypsin (Guertin et al. 2005; Jancalek and Dubovy 2007; Kobayashi
inactivation, cells are resuspended in Dulbeccos et al. 2004a, b, 2005a, b, c, d; Mosconi and Kruger 1996;
Modified Eagle Medium (DMEM) supplemented with Myers et al. 1993). Like functional changes in neurons dur-
fetal bovine serum, growth factors, and antibiotics. ing and after compression, degenerative changes in axons
Intact rodent DRGs may also be incubated with organ develop at later times and are also dependent on the magni-
culture medium (DMEM and serum that is supple- tude of the compression (Hubbard et al. 2008b; Kobayashi
mented with glucose and nerve growth factor). Isolated et al. 2005b; Nicholson et al. 2011).
neurons can be cultured on a variety of different sub- Cells that respond to injury include microglia (resident
strates and exhibit varied responses depending on the macrophages in the central nervous system) and astrocytes.
substrate stiffness. Typically, cells are plated with cul- These cells have many roles in both the peripheral and central
ture media on glass-bottom dishes coated with poly-d- nervous systems, including the maintenance of homeostasis
lysine in borate buffer and laminin in borate buffer. It at neuronal synapses. Both types of glial cells respond to
is possible to phenotype DRG cultures to identify injury by changing their morphology, proliferating, upregu-
afferent populations using immunohistochemistry lating cell surface markers, and releasing several inflammatory
techniques to identify A fibers (by positive NF200 mediators (Cao and Zhang 2008; DeLeo et al. 2004; Saab
labeling), C fibers (negative for NF200 labeling), and et al. 2008; Suter et al. 2007). A peripheral stimulus by some
also peptidergic (substance P positive) and non-pepti- neurotransmitters/neuromodulators (e.g., excitatory amino
dergic fibers (IB4 positive). acids, substance P, ATP) (Cao and Zhang 2008; Marriott
Although these culture techniques have been wide- 2004) can activate early release of proinflammatory cytok-
spread in the neuroscience community for quite some ines as well as nitric oxide, prostaglandins, and nerve growth
time, they are becoming more common for understand- factor (DeLeo et al. 2004; Inoue 2006). These mediators, in
turn, induce an exaggerated release of neurotransmitters
Fig. 19.4 Both the spinal

extracellular potential (blue Sham
trace) and number of spikes 0.2
evoked by electrical stimulus
(green traces) generated
0.2
following a nerve root compres-
sion are modified by the duration 3-min compression (non-painful)
of the applied compression. The
longer duration (15 min) 0.2
compression that produces pain 50 m
symptoms also causes an 0.2
abolishment of evoked responses
and decrease in EC response. In 15-min compression (painful)
addition, that loading scenario
also produces axonal swelling in 0.2
the unmyelinated nerve fibers of
the root that is absent in the 0.2
3-min (non-painful) compression
0.04 s
Extracellular potential (V)

Forepaw stimulus (V)
Evoked spikes (V)
from presynaptic neurons, sensitize the postsynaptic mem- order of weeks (Kobayashi et al. 2008; Ramer et al. 2004).
brane, activate neighboring astrocytes, and enhance micro- The extent of degeneration is modulated by the mechanics of
glial activity (DeLeo et al. 2004; Inoue 2006). This positive the insult and is associated with persistent pain and hypersen-
feedback sustains the release of pain mediators, facilitating sitivity after a compression to the nerve root (Dyck et al. 1990;
the development of neuronal hypersensitivity and can lead to Hubbard et al. 2008a, b; Kobayashi et al. 2008; Nicholson
the persistent pain that is often associated with a herniated et al. 2011) (Fig. 19.5). Disruption to the axonal structure has
disc or inflamed nerve root. Following such neural insults, been found to be more pronounced for greater loads applied
spinal glial cells become activated and modulate other for longer durations (Dyck et al. 1990; Hubbard et al. 2008a;
immunologic changes via cytokine and growth factor pro- Kobayashi et al. 2005a, 2008; Nicholson et al. 2011). Further,
duction, leading to persistent pain (DeLeo and Yezierski the extent of damage and Wallerian degeneration of the axons
2001; Hashizume et al. 2000a; Obata et al. 2004; Winkelstein in the compressed nerve root is directly related to the develop-
et al. 2001a). In particular, greater nerve root compression ment of persistent hypersensitivity (Hubbard et al. 2008a, b;
leads to increased activation of spinal astrocytes that is appar- Hubbard and Winkelstein 2008).
ent as early as 1 day and is sustained in parallel with persis- The severity of nerve root injury and the intensity of pain
tent symptoms of mechanical allodynia (Rothman and after nerve root injury inflammation strongly relate to neuro-
Winkelstein 2007). These observations suggest that spinal peptide depletion in the DRG and spinal cord together with
astrocytes may directly respond to the changes in the dorsal axonal degeneration (Hubbard et al. 2008a, b; Rothman
horn that are induced by damaged primary afferents (Hogan et al. 2005). For example, persistent allodynia, due to higher
2007; Sapunar et al. 2005) (Fig. 19.4). magnitudes of dorsal nerve root compression, is associated
Severe axonal injury can also induce Wallerian degenera- with greater sensitivity at 1 week or later (Hubbard et al.
tion of the axonal process distal to the cell body (Stoll and 2008a, b). This is accompanied by a corresponding deple-
Jander 1999; Stoll and Muller 1999). For the central axons of tion of the nociceptive neuropeptide, substance P, in the
primary afferents, which make up the dorsal nerve root, DRG that similarly varies with the magnitude of the initiat-
Wallerian degeneration can occur proximal to the site of injury ing compressive load (Hubbard et al. 2008b; Kobayashi
(Hubbard and Winkelstein 2008; Kobayashi et al. 2008). et al. 2005b). Spinal expression of another potent neuropep-
Axonal degeneration, marked by neurofilament degradation tide for regulating pain, calcitonin gene-related peptide
and loss of axonal integrity, is evident as early as 15 min after (CGRP), also decreases with increased painful loading to
trauma, but is more commonly present at time points in the the nerve root (Hubbard et al. 2008a, b). Together with the
nerve root compression, elevated magnitudes of compres-

sion increased the levels of mechanical allodynia and reduced
axonal transport in the compressed root (Kobayashi et al.
2005a; Winkelstein et al. 2002). Although animal models of
nerve root compression have shown sustained mechanical
hypersensitivity in the affected limb (Colburn et al. 1999;
Hashizume et al. 2000a; Kobayashi et al. 2005a; Winkelstein
and DeLeo 2004; Winkelstein et al. 2002), mechanical
hypersensitivity can also be produced when the nerve root is
compressed for times as short as 2 s (Sekiguchi et al. 2003,
2009). Moreover, lumbar nerve root compression produces
an immediate change in evoked signal conduction along the
fibers of the compressed root (Fumihiko et al. 1996; Morishita
et al. 2006; Pedowitz et al. 1992; Rydevik et al. 1991;
Takahashi et al. 2003). Both compression rate and magnitude
contribute to edema production in the nerve root, such that
Example of 4 g von Frey filament the magnitude of pressure required to produce edema
decreases for higher loading rates (Hubbard et al. 2008b;
25 Pre-operative trials (n = 12) Hubbard and Winkelstein 2008; Nicholson et al. 2011, 2012;
Paw withdrawal threshold (g force)
Sham surgery (n = 6)
NP on right L5 DRG (n = 6)
Olmarker et al. 1989; Rothman et al. 2010; Rydevik et al.
20 1991). Moreover, specific loading parameters, such as
* Different from pre-op, p < 0.05
Data presented as mean st. dev. magnitude and duration, likely play a role in modulating
15
electrophysiologic responses. Animal models of nerve root
compression in the cauda equina demonstrate that evoked
10
neuronal signaling is altered during and after compression
*
* (Fumihiko et al. 1996; Garfin et al. 1990; Pedowitz et al.
15
1992; Rydevik et al. 1991) and decreases in the amplitude of
electrically evoked compound nerve action potentials due to
0
Pre-op 1 week 4 weeks cauda equina compression may persist after removal of the
Post-op Post-op compressive force (Pedowitz et al. 1992; Rydevik et al.
Fig. 19.5 Apparata and typical results for measuring mechanical allo- 1991). These changes are corroborated by measurements in
dynia in a rodent model of radiculopathy. (Left) von Frey filaments are human subjects with intervertebral disc herniation that simi-
logarithmically graded (e.g., 4g as shown) by buckling load upon com- larly show changes in evoked nerve action potentials upon
pressive application. (Right) When filaments of varying strengths are straight-leg raise.
applied to the hindpaw of a rat following placement of nucleus pulpo-
sus tissue upon the lumbar dorsal root ganglion (DRG), paw with- The results of studies of mechanically compressed nerve
drawal is recorded over multiple trials and filaments. Data for paw roots together with the widespread molecular changes can
withdrawal threshold of rats in an NP injury model of radiculopathy as be integrated into a generalized schema. As early as 1 h fol-
compared to sham controls. Data show that nucleus pulposus place- lowing even a transient nerve root compression that is
ment upon a nave DRG induces a paw withdrawal at lower filament
strength, indicative of a persistent sensitivity to non-noxious stimuli sufficient to produce persistent behavioral sensitivity,
(mechanical allodynia) (Plotted from Hwang et al. (2012), used under inflammatory cytokines, IL-6 and TNF-a, and mRNA
CCL3.0) expression levels are elevated in the ipsilateral DRG and also
in the spinal cord (Rothman et al. 2009b). Within 1 day of
effects of neural trauma described above, these mechani- that event, behavioral sensitivity develops along with hall-
cally deleterious events can dramatically contribute to marks of spinal inflammation, including activation and pro-
reduced transport of neuropeptides and neurotrophic factors liferation of microglia (Rothman et al. 2009a). By 7 days
from neurons, where these factors are synthesized, to their after injury, axons of the injured nerve root show signs of
release from the presynaptic terminals in the spinal cord. degeneration, and the spinal inflammation becomes even
In animal models, the magnitude, duration, and rate of the more pronounced, and both astrocytes and microglia become
nerve root compression have been shown to modulate the activated (Hubbard and Winkelstein 2005, 2008). These
extent of the local tissue damage to the root and the degree changes together with the decrease in neuropeptides in the
and duration of the pain-related behaviors (Kobayashi et al. spinal cord at this same time point (Hubbard et al. 2008b)
2005a; Olmarker et al. 1989; Rothman et al. 2010; Rydevik may lead to alterations in neuronal signaling in the spinal
et al. 1991; Winkelstein et al. 2002). In rodent models of cord following a painful injury.
19.3.2 Role of Chemical and Inammatory

Mediators in Disc Herniation Box 19.2 Historical Information on Sciatica
What we currently understand as sciatica was origi-
As mentioned above, contributing to a cascade of chemi- nally considered an evil display of demon magic by
cal injuries at the affected nerve are elevated levels of the earliest civilizations to provide written record (Sigerist
inflammatory mediators, infiltration of macrophages, and 1934). The Greeks of fifth century BC provide documen-
activation of glial cells. To further examine these effects, tation of early attempts at treating the sudden shots of
animal models have been developed that mimic features of pain that were common to the spinehipjoint complex.
both inflammation and immune system function in interver- In Hippocratic times, corrective traction was often under-
tebral disc herniation. Both behavioral hypersensitivity and taken, with more conservative treatment that included
widespread immune responses are produced when chromic massage, heat, dietary alterations, bed rest, and music to
gut suture pieces are placed in contact with the nerve root pipe away pain. A particular Roman physician of the
without any mechanical perturbation (Colburn et al. 1999; fourth century AD, Caelius Aurelianus, is noted for pro-
Hashizume et al. 2000a; Hou et al. 2003; Hubbard and viding some attempt at explaining the etiopathogenesis
Winkelstein 2005; Kajander et al. 1996; Kawakami et al. and anatomic origins of sciatica (Drabkin ed 1950).
1994a, b; Maves et al. 1993; Murata et al. 2004a, b; Olmarker Caelius Aurelianus believed that a sciatic attack could be
et al. 1993; Rothman and Winkelstein 2007; Rutkowski et al. caused by a sudden jerk or movement during exercise,
2002; Winkelstein and DeLeo 2004). Gut suture ligation lifting a heavy object, a sudden shock, or fall and that this
simultaneously compresses the nerve root by ligation, while presentation was most common for middle-aged persons.
the chromic salts and pyrogallol in the suture serve as irri- Surprisingly, these early Roman observations are all con-
tants (Colburn et al. 1999; Hashizume et al. 2000a; Kajander sistent with our present-day understanding. Further,
et al. 1996; Kawakami et al. 1994a, b; Maves et al. 1993; Caelius Aurelianus believed that sciatica could be caused
Robinson and Meert 2005; Winkelstein and DeLeo 2004; Xu by a deep-seated congelation, referring to a cutting off
et al. 1996). A consistent response to the chromic gut suture of squeezing of nervous tissue. In the day of Caelius
by the nerve root damage is the induction of thermal hyper- Aurelianus, patients afflicted with sciatica may have
algesia (decreased latency to withdraw from thermal stimuli) been treated with a mixture of sweet marjoram, rose-
that is transient and dose dependent; this form of hyperalge- mary leaf, wine, and olive oil, as well as bed rest, mas-
sia is not consistently present in intervertebral disc hernia- sage, heat, and passive range-of-motion exercises. If met
tion models that mimic the compressive stimuli alone (Maves with failure, a Roman may have undergone treatment
et al. 1993; Yamamoto and Nozaki-Taguchi 1995). The with leeches, hot coals, skin hooks, and bloodletting.
mechanisms governing the neuronal responses to chromic
gut salts appear to be related to early immune activation References
characterized by Schwann cell proliferation, macrophage Aurelianus, Caelius. Edited by Israel E. Drabkin, New
infiltration, and microglial activation. Along with these ed., based on the first printed editions, of Caelii
changes are molecular events including a depleted neuro- Aureliani Methodici Siccensis celerum vel acu-
peptide expression and an early expression of cell adhesion tarum passionum and Caelii Aureliani Methodici
molecules such as ICAM-1 and PECAM. These adhesion Siccensis tardarum passionum, Chicago : University
molecules would serve to promote recruitment of circulating of Chicago Press 1950.
monocytes to the injured nerve root (Chang and Winkelstein Sigerist HE. A history of medicine (German). Oxford
2011; Hashizume et al. 2000a; Rothman et al. 2010; U.P, 1951.
Rutkowski et al. 2002; Xu et al. 1996; Yamamoto and Source: Adapted from Karampelas et al. Neurosurgery
Nozaki-Taguchi 1995). These studies indicate that, despite Focus V16: pp. 14, 2004
similarities in sensitivity to chromic gut suture and mechani-
cal compression, differences in the pattern of hypersensitiv-
ity and molecular changes may reflect only limited injury
etiologies. Nonetheless, use of chromic gut sutures as a
chemical injury model for nerve root damage is popular for promote an autoimmune response. With degeneration and
evaluating treatments that can interfere in the perception of damage, the internal structures of the intervertebral disc
disc herniation-related pain. can come in contact with adjacent tissues, producing an
Soft tissues of the disc are believed to be immune privi- immune response that is the subject of extensive investiga-
leged, in that the internal structures of the intervertebral tion. Intervertebral disc tissues contain high levels of
disc do not come into contact with the systemic circulation CD68+ immunoreactive macrophages, as well as lesser
under normal conditions and express the Fas ligand that can quantities of T and B lymphocytes (Doita et al. 1996;
Box 19.3 Glossary of Terms Ground reaction force Ground reaction forces occur
Schober index Schober index measures a patients as a limb contacts the ground during motion and are
ability to flex the lumbar region of his or her back. In a typically described by three ground reaction force com-
normal standing posture, a mark is made 5 cm above ponents. Vertical ground reaction forces help to support
and 10 cm below the iliac spine. The patient then bends body weight during motion. Braking and propulsive
to full flexion, and the distance between the two marks ground reaction forces occur in the direction of travel
is measured. A distance above 20 cm in flexion is con- and help to propel the body forward during locomotion.
sidered normal, where distances below 20 cm during Mediolateral ground reaction forces are directed toward
flexion suggest a limited range of motion. an animals or persons midline and help to stabilize and
Oswestry Disability Index The Oswestry Disability balance the body during locomotion.
Index is a validated questionnaire commonly used to Wet-dog shake A characteristic shaking motion
evaluate the principal conditions associated with spinal resembling a wet dog shaking out its coat. The fre-
disorders. The Oswestry Disability Index evaluates quency of wet-dog shake motions increases in rat mod-
self-reports of pain intensity, emotional well-being, els of lumbar radiculopathy.
and ability to perform common tasks.
Visual analogue scale (VAS) Visual analogue scales
Freemont et al. 2002; Gronblad et al. 1994; Roberts et al.
are commonly used in questionnaires and surveys as
2006; Shamji et al. 2010). In addition, the disc tissues
ranking statistics, where subjective semiquantitative
secrete numerous proinflammatory cytokines and
ranks are used to assess a patients relative improve-
inflammatory mediators, such as interleukin-1a (IL-1a),
ment or decline.
interleukin-6 (IL-6), interleukin-17 (IL-17), and tumor
Radiculopathy A medical condition where one or necrosis factor-a (TNF-a); these cytokines may be pro-
more nerves do not function properly resulting in radic- duced by primary disc cells or by infiltrating monocytes
ular pain, numbness, and weakness and dysfunction. (Bachmeier et al. 2009; Burke et al. 2002; Le Maitre et al.
Sciatica A sharp pain sensation or series of painful 2007; Murata et al. 2004a; Saal 1995; Shamji et al. 2010;
sensations affecting the back, hip, or leg. Sciatica can Weiler et al. 2005, 2011) (Table 19.1). Numerous studies
result from compression or inflammation of the spinal have documented the molecular effects of the herniated dis-
nerve root in the lower back and is often described as cal tissues (Mulleman et al. 2006a, b). Models used for
shooting pain down the leg. these studies have included harvest of tail nucleus pulposus
N-Methyl-d-aspartic acid (NMDA) receptor NMDA tissue for placement at a lumbar nerve root or cauda equina
receptors are ionotropic glutamate receptors that help (Allen et al. 2011; Aoki et al. 2002; Brisby et al. 2000;
to control synaptic plasticity and regulate nociception. Cuellar et al. 2005; Kawakami et al. 1999; Shamji et al.
NMDA receptors have been identified as key regula- 2009; Skouen et al. 1999; Yabuki et al. 1998) or lumbar
tors of peripheral and central sensitization in lumbar disc puncture to promote nucleus pulposus herniation
radiculopathy. (Olmarker et al. 1998; Olmarker and Myers 1998; Otani
Allodynia Allodynia is the sensation of pain from et al. 1997). Physiological changes noted in these models
stimuli that would not typically cause pain. Mechanical include reduced nerve conduction velocities, lowered
allodynia, a heightened sensitivity to light touch, and endoneurial pressure for dorsal horn or sensory neurons,
cold allodynia, a heightened sensitivity to cold, have been and elevated nitric oxide synthase activity in spinal nerve
observed in several models of lumbar radiculopathy. roots at 14 weeks following exposure to discal tissues. In
addition, there is increased expression of neurotrophins,
Hyperalgesia Hyperalgesia is a prolonged or exag- IL-1b, TNF-a, phospholipase-1, and/or nitric oxide syn-
gerated pain response to stimuli that typically would thase in the applied nucleus pulposus tissues or in cell bod-
cause pain. Thermal hyperalgesia, a heightened sensi- ies and intercellular domains around the DRG (Kallakuri
tivity to heat, and mechanical hyperalgesia, a height- et al. 2005; Kawakami et al. 1999; Murata et al. 2004a;
ened sensitivity to a pinch or pinprick, have been Onda et al. 2002). Disc-associated cytokines (IL-1b,
observed in several models of lumbar radiculopathy. TNF-a, IFN-g) applied directly to nerve roots, as opposed
Dysesthesia Dysesthesia is a sensation described as to cytokines secreted by nucleus fragments, have also been
unpleasant or abnormal, but not considered painful. shown to induce electrophysiological changes, consistent
Dysesthesia is among the symptoms reported from with a heightened sensitivity (Ozaktay et al. 2002; 2006). It
neuropathies and is often associated with descriptions is likely that these cytokines bind to receptors for TNF-a
of limb weakness and numbness. and IL-1b on the sensory neurons (Binshtok et al. 2008;
Verri et al. 2006). While these findings clearly indicate a
role for inflammation and inflammatory mediators in regu- control or preoperative values at later times after surgery.
lating neuronal sensitivity in intervertebral disc herniation, Additional studies of CNS sensitization in human and ani-
it must be noted that many changes in the nucleus pulposus- mal models of disc herniation are needed to better under-
induced or chemical injury models of radiculopathy over- stand the role of these important pathway changes in the
lap with those reported for direct compression neuropathy. pathogenesis of disc pain.
As with direct nerve root compression or chromic gut Taken together, with knowledge gained from direct nerve
suture exposure, application of nucleus pulposus tissue to a root compression studies, it becomes clear that many of the
nave nerve root induces limb hypersensitivity that is largely sensitization and functional changes associated with disc
characterized by a mechanical allodynia (Hou et al. 2003; herniation may reflect contributions from both compressive
Mulleman et al. 2006a, b) (Fig. 19.5). Many studies of nucleus trauma as well as inflammatory agents in the disc. While
pulposus-induced radiculopathy following disc puncture to somewhat intuitive, the more severe nerve root injuries (with
induce nucleus herniation or placement of nucleus pulposus greater degrees of tissue impingement) produce more pain-
tissue upon the nerve root report evidence of allodynia as associated behavioral sensitivity than those injuries with less
early as 2 days that may persist out to 3 weeks (Kawakami tissue compression (Hubbard et al. 2008b; Hubbard and
et al. 1999; Obata et al. 2002). Early studies also report func- Winkelstein 2005; Winkelstein et al. 2001b, 2002). This
tional changes following nucleus pulposus-induced radicul- graded relationship has been shown to hold true regardless of
opathy, including altered footprints and visual evidence of the absence or presence of molecular challenges (Winkelstein
limping, paw lift, or rotation of the head (Olmarker et al. 1998, and DeLeo 2004). Indeed, evidence that the duration of the
2002). With the use of quantitative gait analysis, our laborato- mechanical trauma modulates several different pain path-
ries have further demonstrated gait asymmetries between ipsi- ways strongly supports the concept that a more permanent
and contralateral hind limbs out to 3 or 4 weeks postoperatively, mechanical insult would produce a similar or even more
indicative of limping (Shamji et al. 2009) (Fig 19.3). Static robust deleterious clinical effect. Based on these understand-
and dynamic gait analyses indicate that animals with radicul- ings, clinical interventions, and specifically nonsurgical ther-
opathy bear less weight on their affected hindpaw in both apies, are focused on alleviating the persistent sensitivity
stance and during locomotion (Allen et al. 2012). that is associated with a transient or more chronic compres-
The findings discussed above suggest that radiculopathy sive trauma to the nerve root and the associated inflammatory
induced by DRG exposure to nucleus pulposus tissue, as a events. This topic is briefly covered in the next section on
model of intervertebral disc herniation, can repeatedly mimic emerging pharmacological approaches to treat intervertebral
key characteristics of prolonged pain sensitivity in the human disc herniation-associated radiculopathy.
patient (Allen et al. 2011; Shamji et al. 2009). The finding of
persistence of sensitivity supports the notion that the molec-
ular responses of an inflamed disc may influence sensory 19.4 Emerging Therapies for Intervertebral
changes during intervertebral disc herniation. The persistent Disc-Associated Radiculopathy
sensitivity changes, extending beyond fragment removal or
resorption, imply that the widespread responses imitated in As mentioned in Sect. 19.1, in the absence of motor weak-
the CNS and even systemically are not ameliorated. Once ness or related loss of function, conservative care is the pre-
initiated, the neuroimmune cascade involves activation of ferred treatment for a patient presenting with pain and
many nonneuronal cells in the peripheral tissues (DeLeo and symptoms of radiculopathy secondary to an intervertebral
Yezierski 2001; Julius and Basbaum 2001; Moalem disc herniation. Conservative care most frequently involves
and Tracey 2006). These resident cells, including mast and lifestyle modifications as well as orally administered non-
Schwann cells, release mediators such as histamine, prosta- steroidals or opioid analgesics. Commonly prescribed selec-
glandins, cytokines, and chemokines that also lead to the tive or nonselective nonsteroidal drugs include ibuprofen,
recruitment of other infiltrating immune cells (e.g., neutro- indomethacin, diclofenac, piroxicam, diflunisal, and cele-
phils, macrophages, lymphocytes) (DeLeo and Yezierski coxib, few of which have demonstrated significant effects on
2001; Moalem and Tracey 2006; Verri et al. 2006). radicular pain associated with disc herniation (Chou and
Proinflammatory cytokines also trigger the release of many Huffman 2007) (also see Chap. 15). In animal models, indo-
other inflammatory mediators that can sensitize nociceptors, methacin, ibuprofen, diclofenac, and celecoxib have been
further maintaining neuronal excitability and sensitization shown to reverse allodynia following peripheral or nerve
and leading to dysfunction and pain (Moalem and Tracey root compression and may work by lowering prostaglandins
2006; Verri et al. 2006). Nonetheless, the persistence of pain and inhibiting the resulting decreases in nerve conduction
observed in some human subjects following intervertebral velocity and decreased intraneural blood flow (see
disc herniation that may continue for months and even years Table 19.2). Epidural administration of anesthetics such as
has not been replicated in these animal models, where bupivacaine and/or corticosteroids, including methylpredni-
mechanical allodynia or thermal hyperalgesia can recover to solone, is also widely used for treatment of symptoms with
Table 19.2 Summary of compounds under investigation for treatment of IVD herniation-associated radiculopathy
Agent Route Model description Reference Observations
Anti-inflammatories
Indomethacin Oral Canine lumbar nerve injury Arai et al. (2004) Reversed changes in intraneural blood
flow and nerve conduction
Ketoprofen Systemic Porcine nerve root constriction Cornefjord et al. Partly reversed decreased nerve
(i.m.) or NP tissue placement (2001) conduction velocity in constriction but
not NP exposure
Diclofenac Systemic Porcine nerve root constriction Cornefjord et al. Partly reversed decreased nerve
(i.m.) or NP tissue placement (2001) conduction velocity
Ibuprofen Oral Rat sciatic nerve compression Schafers et al. Reduced mechanical allodynia at
injury (2004) short times after injury, reduced PGE2
levels in nerve and DRG
COX-2 inhibitor celecoxib Oral Rat sciatic nerve compression Schafers et al. Reduced mechanical allodynia at
injury (2004) short times after injury, reduced PGE2
levels in nerve
NO synthase inhibitor Intrathecal Rat lumbar DRG compression Ding et al. (2010) Some reversal of thermal
(L-NAME) injury hyperalgesia, decreased nitrite in DRG
Prostaglandin E2 receptor Oral Rat exposure of lumbar DRG to Sekiguchi et al. Reduced mechanical allodynia,
antagonist (EP1-RA) NP (2011) attenuated increased activating
transcription factor-3 (ATF3)
immunoreactive positive cells induced
by NP
Thromboxane A2 synthetase Epidural Rat exposure of lumbar DRG to Kawakami et al. Reduced mechanical allodynia
inhibitor NP (2001)
Leukotriene B4 receptor Epidural Rat exposure of lumbar DRG to Kawakami et al. Reduced mechanical allodynia
antagonist (LTB4 receptor NP (2001)
antagonist)
COX-2 antibody Intrathecal Rat exposure of lumbar DRG to Ohtori et al. (2004) Reduced mechanical allodynia
NP
Neuronal receptor modifiers
NMDA receptor antagonist, Intraspinal Rat spinal nerve or sciatic nerve Chaplan et al. Reversed molecular changes in CNS,
MK-801 or systemic constriction injury (1997), Uceyler partly reversed motor changes
(i.p.) et al. (2008)
Gabapentin Systemic i.p. Rat sciatic nerve or lumbar Abe et al. (2002), Some reversal of mechanical
or local nerve root constriction injury Zanella et al. (2008) allodynia
(perineural)
delivery
Sarpogrelate hydrochloride Systemic Canine or rat DRG exposure to Sekiguchi et al. Decreased blood vessel diameter and
(5-HT2A receptor antagonist: NP (2008), Hashizume increased blood flow in nerve roots
5-HTRA) et al. (2007) inflamed by NP application, partly
reversed mechanical allodynia
Cell cycle modifiers
Minocycline Systemic i.v. Rat cervical nerve root Rothman et al. Some reversal of mechanical
and compression and chromic gut (2009b) allodynia, no changes in spinal
prophylactic exposure microglial proliferation
delivery
Methotrexate Intrathecal Rat lumbar DRG constriction Hashizume et al. Reduced allodynia, no changes in
or local with chromic gut (2000b) spinal glial activation
Pathway inhibitors
Ruthenium red (TRPV4 Intrathecal Rat lumbar DRG compression Ding et al. (2010) Some reversal of thermal
antagonist) or TRPV4 injury hyperalgesia, decreased nitrite in
antisense oligonucleotide DRG
(TRPV4 AS)
Soluble guanylate cyclase Intrathecal Rat lumbar DRG compression Song et al. (2006), Some reversal of thermal hyperalgesia
inhibitor (1H-[1,2,4]- or perineural injury, rat lumbar nerve root Ding et al. (2010)
oxadiazolo[4,3-a] quinoxalin- compression via steel rod
1-one, ODQ)
8-(4-chlorophenylthio)- Intrathecal Rat lumbar DRG compression Ding et al. (2010) Some reversal of thermal hyperalgesia
guanosine 3,5-cyclic injury
monophosphorothioate
(continued)
Table 19.2 (continued)

Agent Route Model description Reference Observations
PKA antagonist (SQ22536) Perineural Rat lumbar nerve root compres- Song et al. (2006) Some reversal of thermal hyperalgesia
sion via steel rod
NF-kappa B decoy-FITC Intrathecal Rat L5 DRG compression and Suzuki et al. (2009) Reversed molecular changes in DRG,
NP placement partly reversed mechanical allodynia,
and thermal hyperalgesia
PKG inhibitor Rp-isomer Intrathecal Rat lumbar nerve root compres- Song et al. (2006), Some reversal of thermal hyperalgesia
sodium salt (Rp-8-pCPT- sion via steel rod, rat lumbar Ding et al. (2010)
cGMPS) DRG compression injury
Protease or cytokine inhibitors
Hydroxamic acid-based Epineural Rat sciatic nerve constriction Sommer et al. Reduced thermal hyperalgesia and
metalloproteinase inhibitor, injury (1997) mechanical allodynia, reduced TNF
TAPI immunoreactivity in epineurium
sTNFRI Intrathecal Lumbar or cervical nerve root Winkelstein et al. Some reversal of mechanical
or systemic compression with, or without (2001a), Rothman allodynia, reduced spinal astrocytic
chromic gut exposure et al. (2009b, 2010) reactivity
sTNFRII Systemic i.p. Rat constriction nerve injury or Schafers et al. Some reversal of allodynia, restored
or local ligation, rat DRG exposure to (2003), Allen et al. normal gait
(perineural) NP (2011), Zanella et al.
(2008)
IL-1Ra Intrathecal Lumbar or cervical nerve root Winkelstein et al. Some reversal of allodynia, and
or systemic compression with or without (2001b), Rothman reduced spinal astrocytic reactivity
chromic gut exposure et al. (2009b, 2010)
Salmon fibrin and thrombin Perineural Rat cervical nerve root Weisshaar et al. Partly reversed mechanical allodynia
compression (2011) and reduced macrophage recruitment
radiculopathy, and they appear to have some efficacy as mea- and to identify those with the most promising therapeutic
sured with both objective (e.g., straight-leg raising test) and potential.
self-reported outcomes (e.g., VAS) (Buenaventura et al.
2009; Staal et al. 2008). Methylprednisolone and other corti-
costeroids may work by inhibiting the increase in endoneur- 19.4.1 Cytokine Antagonism
ial vascular permeability and the decrease in nerve conduction
velocity that follows injury to the nerve root upon exposure Given the documented increase of TNF-a and IL-1b expres-
to nucleus pulposus tissue or chemical injury (Byrod et al. sion in degenerated and herniated discal tissues, as well as
2000; Olmarker et al. 1994). Regardless of the mechanism, clear evidence of a role for the cytokine TNF-a in reproduc-
use of these NSAIDs and corticosteroids is generally consid- ing many symptoms of radiculopathy in animal models
ered safe and a first line of treatment for patients presenting (Allen et al. 2011; Olmarker 2001; Onda et al. 2002; Rothman
with pain from intervertebral disc herniation. and Winkelstein 2010; Winkelstein et al. 2001a), cytokine
The current standard of care does not address the underly- antagonism has received considerable attention. In our prior
ing pathology of intervertebral disc herniation, which clearly studies, we have demonstrated an ability for IL-1b antago-
involves inflammatory, proteolytic, and immune-mediated nists (IL-1Ra or KineretTM) to partially reverse mechanical
pathways. As new knowledge from animal models is gained, allodynia and spinal astrocytic reactivity following nerve
an understanding of the role of inflammatory mediators in root compression (Rothman and Winkelstein 2010;
mediating responses to nerve root injury has emerged; the Winkelstein et al. 2001a, Table 2). The finding that TNF
focus here is on cytokines that regulate immune system antagonists (blocking antibodies) influence peripheral nerve
involvement, such as TNF-a, or those that mediate pain sen- injury (DeLeo et al. 2000; Lindenlaub et al. 2000; Sommer
sitivity such as a2 adrenergic receptor and serotonin receptor et al. 2001) and attenuate allodynia in constriction injury
antagonists. Pharmacological approaches that target percep- models, and that overexpression of TNF could elevate allo-
tion of pain and restoration of functional losses with radicu- dynia in the same model, has evoked considerable interest in
lopathy, as well as those that influence disease pathways, this class of compounds. Application of exogenous TNF-a to
have great potential to reduce the duration of symptoms and lumbar nerve roots in the rat reproduces many of the neuro-
disability associated with disc herniation and may even play physiology changes described earlier for compression-
a role in reducing the need for surgical intervention. Here, induced nerve root injury, including decreased nerve
we will briefly review ongoing investigations to pharmaco- conduction velocity, glial activation, and inflammatory
logically treat radiculopathy associated with disc herniation changes in the ganglion (Aoki et al. 2002; Igarashi et al. 2000;
Onda et al. 2002; Ozaktay et al. 2002). Furthermore, sys- in nine or ten patients at 6 weeks after treatment, although no
temic treatment with the TNF-blocking antibody, Remicade control group was cited in this study. Over the years, multiple
(infliximab, intravenous or intraperitoneal), reduced pain- clinicians have individually described case reports of their
related movements in rats, as well as the expression of key experiences with delivery of infliximab or etanercept for the
neurotrophins in the DRG and spinal cord (Murata et al. treatment of disc herniation-related radiculopathy that was
2004b; Olmarker et al. 2003; Olmarker and Rydevik 2001; otherwise nonresponsive to treatment (Atcheson and Dymeck
Onda et al. 2004; Sasaki et al. 2007). In other animal studies, 2004; Tobinick and Britschgi-Davoodifar 2003). Decreases
including our own, local delivery of a soluble TNF receptor in pain scores and improved disability indices are generally
type II analogue (etanercept or Enbrel) has been shown to reported, leading to the emergence of TNF inhibitors as
restore normal gait patterns and reverse the heightened allo- available strategies for clinical treatment of pain associated
dynia response to nucleus pulposus placement upon the DRG with intervertebral disc herniation.
as a model of intervertebral disc herniation (Allen et al. 2011; The first randomized controlled trial of TNF antagonism
Cuellar et al. 2004). These results add strength to the notion for radiculopathy (termed FIRST II, Finnish Infliximab
that TNF inhibition can attenuate or reverse changes in ani- Related Study) (Korhonen et al. 2006; Korhonen et al. 2005)
mal locomotion, nerve electrophysiology, and pathology due was reported in 2006. Patients (n = 40 with symptomatic disc
to nucleus pulposus exposure in the short term (<7 days). For herniation on MRI, leg pain <12 weeks. 60 on the straight-
these reasons, the hypothesis has evolved that, in response to leg raising test) were treated with a single intravenous infu-
disc herniation, TNF-a serves as the critical cytokine medi- sion of infliximab (5 mg/kg) and compared against a placebo.
ating nerve sensitivity and inflammation. This plausible Straight-leg raise, motor and sensory defects, leg and back
mechanism is supported by the elevated tissue levels of pain (VAS), Oswestry disability, quality of life (RAND-36),
TNF-a in the degenerative disc as well the increased number and more parameters were compared between the treatment
of activated macrophages and lymphocytes in herniated tis- and placebo groups out to 1 year following treatment. Results
sue fragments that secrete the cytokine. Thus, it is plausible were not supportive of infliximab therapy, however, with 67
that TNF-a contributes, at least in part, to the perception of and 63 % of all patients reporting no pain in the infliximab
pain and promotes neuroinflammation following disc hernia- and placebo groups, respectively. Similar values were
tion. For this reason, to date, clinical studies that target symp- observed between treatment groups for other outcomes,
toms of disc herniation-associated radiculopathy have although a subgroup of patients in the infliximab group
focused largely on TNF antagonists. appeared to especially benefit from the infliximab treatment
In a first study of TNF inhibitors to antagonize pain asso- (an L4L5 or L3L4 herniation with a Modic change at the
ciated with intervertebral disc herniation, patients with a his- symptomatic level). The authors concluded that further study
tory of herniation-associated sciatica (average duration of of this subgroup of patients may yield insights about the
symptoms = 7 weeks) were given a single intravenous infu- potential for TNF antagonists to modify clinical outcomes
sion of infliximab (n = 10, 3 mg/kg, Karppinen et al. 2003) or for intervertebral disc herniation.
treated with periradicular saline as a historical control A recent study of epidural etanercept for radiculopathy
(n = 62). Outcomes of visual analogue scale (VAS), straight- (Cohen et al. 2009) provided a more promising result.
leg raising test, low back pain severity, and Oswestry Patients with unresolved symptoms (n = 12, >2 months in
Disability Index were measured at baseline and out to 3 duration) received 2 injections of etanercept (2, 4, or 6 mg).
months after treatment. All outcomes outperformed the A majority of patients noted complete resolution of pain by
saline control, with the exception of a decrease in leg pain 3 months after treatment with etanercept; this was much
reported at 1 h after the infusion, and sustained out to 3 higher than the 17 % for the epidural saline control group.
months (88 and 51 % change from baseline, infliximab vs. While a small study, it nonetheless illustrates the potential
saline). Similar changes supportive of infliximab treatment for local administration of TNF antagonists to modify radic-
were noted in low back pain severity, straight-leg raising test ular pain associated with disc herniation.
results, the Schber index, and the Oswestry index. No
patients required surgery and none experienced adverse
effects of infliximab, while motor and sensory loss resolved 19.4.2 Neuronal Receptor Blockers
in patients within 13 months. A later study evaluated the
efficacy of subcutaneous injections of etanercept (25 mg Another set of targets proposed for the treatment of interver-
every 3 days, 3 injections) in ten patients with acute severe tebral disc herniation-associated pain centers around block-
sciatica (mean symptom duration of 27 weeks (Genevay ing receptors involved in neuronal activation. Compounds
et al. 2004)). Visual analogue scales (VAS) for leg and back shown in Table 19.2 generally act by competitively binding
pain, as well as Oswestry Disability Index and modified to receptors that control neuronal excitation and downstream
RolandMorris Disability Questionnaire, were obtained after effectors. When delivered systemically, their activity is not
10 days and 6 weeks. A good clinical response was observed localized to the affected nerve and can involve the CNS
and other sites. In preclinical models, an NMDA receptor animal studies show that pathway inhibitors are able to partly
antagonist, administered systemically following sciatic nerve reverse thermal hyperalgesia induced by DRG compression
ligation, reversed mRNA changes induced by peripheral or nucleus pulposus placement. Importantly, inhibition of
nerve constriction (Uceyler et al. 2008). When administered NF-kB, a key mediator of TNF-a signaling, appears to also
locally, this NMDA receptor antagonist was also capable of reverse changes in the mRNA profile downstream of DRG
partly reversing motor deficits and allodynia following lum- compression. As for cell cycle modifiers, additional studies
bar spinal nerve ligation (Chaplan et al. 1997). In animal are required to promote the concept that any specific inhibi-
models, the GABA analogue, gabapentin, whether delivered tor can influence neuroinflammation linked to intervertebral
locally or systemically also caused a similar reversal of allo- disc herniation.
dynia (Abe et al. 2002; Zanella et al. 2008). Indeed, orally
prescribed gabapentin has been shown to relieve symptoms
associated with radiculopathy from lumbar disc herniation 19.5 Final Comments
when measured by the VAS or self-reported disability indi-
ces; on the other hand, systemic delivery is associated with This review focuses on pathological events associated with
frequent and adverse side effects (Kasimcan and Kaptan disc herniation, in particular the grade of nerve root com-
2010; Yildirim et al. 2009). Nevertheless, this class of recep- pression and the exposure and presence of nucleus pulposus
tor blockers appears to be of use for the treatment of radicu- tissue. The impact of herniation can thus be viewed as not
lopathy and superior to nonsteroidal anti-inflammatory one disorder, but perhaps two or three distinct conditions
drugs, especially when delivered locally. that can benefit from distinctly different therapeutic
The serotonin receptor, 5-HT(2A), blocker, sarpogrelate approaches. A contained or protruding herniation that is
hydrochloride, has been used for both animal model and linked to prolonged nerve root compression may be associ-
clinical studies for the treatment of sciatica. In a rat model, ated with electrophysiological changes that can benefit
5-HT(2A) receptor blockers were shown to decrease blood from early intervention including the use of agents to atten-
flow to a nerve inflamed by placement of nucleus pulposus uate glial activation and mediate pain sensitivity (e.g., neu-
tissue upon the nerve root and to reverse allodynia ronal receptor modifiers). An uncontained or extruded disc
(Hashizume et al. 2007; Sekiguchi et al. 2008). In patients, herniation that has components of both nerve root compres-
high doses (300 mg) of sarpogrelate hydrochloride given sion and tissue-mediated inflammation may be better
orally led to significant improvements in VAS scales of sci- addressed with a combined anti-inflammatory approach
atic pain in a majority of patients, with few patients (<20 %) including the use of cytokine antagonists and protein kinase
needing to go on to surgery (Kanayama et al. 2003). Patients or NF-kB inhibitors. In all cases, it appears that systemic
with an uncontained disc herniation, or one that is extruded delivery of these agents will be associated with lower
or sequestered from the parent disc, responded more favor- efficacy and higher risk of side effects than local drug deliv-
ably to the 5-HT(2A) blocker treatment with few side ery; this view is supported by the increased interest in epi-
effects, suggesting this approach may be useful for alleviat- dural administration of immunosuppressants like the TNF
ing symptoms associated with intervertebral disc antagonists and advancement of novel, depot-based deliv-
herniation. ery systems (Hubbard et al. 2009; Shamji et al. 2008;
Zanella et al. 2008). Of course, surgery for disc fragment
removal remains an option for patients that are nonrespon-
19.4.3 Cell Cycle Modiers sive to pharmacologic and alternate approaches. However,
it is becoming clear that residual and persistent hypersensi-
A number of therapeutic approaches have been proposed to tivity from long-term nerve root compression can be a del-
target the microglia that are believed to be activated early in eterious event and a constant source of intense pain. Thus,
the inflammation cascade that follows compression or chem- demand will continue for strategies that attenuate the neu-
ical injury to the nerve root. The neuroprotective antibiotic ropathy associated with intervertebral disc herniation,
minocycline, as well as the antimetabolite methotrexate, shorten the symptom period, and promote functional
attenuates allodynia induced by DRG constriction injury, yet recovery.
does not appear to act through attenuation of glial activation
following injury (Table 19.2).
in the Chapter
19.4.4 Pathway Inhibitors
Fragments of intervertebral disc that herniate into the
New developing therapies that target the NF-kB or protein extra-discal space and impinge upon, or contact, the spinal
kinase pathways have been proposed (Table 19.2). Results of nerve roots can cause significant pain, neurological deficits,
and functional disability in a large number of affected In summary, there is a growing demand for strategies that
individuals. attenuate neuropathy associated with intervertebral disc her-
Nerve root impingement due to a herniated disc can niation and shorten the duration of pain and functional
result in nerve root pathophysiology, including edema, recovery.
inflammation, disorganization of the myelin sheath,
decreased axonal packing, Wallerian degeneration of the Acknowledgments This work was supported by grants from the NIH
peripheral axons, and loss of axonal transport and R00AR057426 (KDA), R01AR047442 (LAS), P01AR050245 (LAS),
DOD (BAW), and the Catherine Sharpe Foundation (BAW).
decreased amplitudes of evoked action potentials.
Nerve root changes are associated with developing hyper-
sensitivity to non-noxious (allodynia) and noxious stimuli
(hyperalgesia) that serve as useful indicators of neuro-
pathic pain in both human subjects and animals. References
The magnitude, duration, and rate of compression of the
nerve root promote local nerve root damage and the Abe M, Kurihara T, Han W, Shinomiya K, Tanabe T (2002) Changes in
expression of voltage-dependent ion channel subunits in dorsal root
degree and duration of the pain symptoms. ganglia of rats with radicular injury and pain. Spine (Phila Pa 1976)
Herniated intervertebral disc exhibits increased angio- 27:15171524; discussion 1525
genesis, macrophage and lymphocyte infiltration, and Adams MA, Roughley PJ (2006) What is intervertebral disc degenera-
proinflammatory cytokine and proteinase activity that can tion, and what causes it? Spine (Phila Pa 1976) 31:21512161
Ahn SH, Cho YW, Ahn MW, Jang SH, Sohn YK, Kim HS (2002)
contribute to sustained painful symptoms over a period of mRNA expression of cytokines and chemokines in herniated lum-
months to years. bar intervertebral discs. Spine 27(9):911917
Herniated disc fragments may activate the immune Allen KD, Shamji MF, Mata BA, Gabr MA, Sinclair SM, Schmitt DO,
system when in contact with the systemic circulation, Richardson WJ, Setton LA (2011) Kinematic and dynamic gait
compensations in a rat model of lumbar radiculopathy and the
glial cells of the affected nerves, and the central ner- effects of tumor necrosis factor-alpha antagonism. Arthritis Res
vous system (microglia and astrocytes). The activated Ther 13:R137
microglia will release increased levels of neurotrans- Allen KD, Mata BA, Gabr MA, Huebner JL, Adams SB Jr, Kraus VB,
mitters from presynaptic neurons, sensitize the post- Schmitt DO, Setton LA (2012) Kinematic and dynamic gait com-
pensations resulting from knee instability in a rat model of osteoar-
synaptic membrane, activate neighboring astrocytes, thritis. Arthritis Res Ther 14:R78
and enhance microglial activity. Hence, the initiation Aoki K, Rydevik B, Kikuchi S, Olmarker K (2002) Local application
of nerve root damage by a herniated disc can drive a of disc-related cytokines on spinal nerve roots. Spine 27:
positive feedback loop that can promote a sustained 16141617
Arai I, Mao GP, Otani K, Konno S, Kikuchi S, Olmarker K (2004)
painful neuropathy. Indomethacin blocks the nucleus pulposus-induced effects on nerve
While surgical removal of the herniated disc fragment can root function. An experimental study in dogs with assessment of
alleviate many of the pathological symptoms, pharmaco- nerve conduction and blood flow following experimental disc her-
logical interventions and lifestyle modifications are the niation. Eur Spine J 13(8):691694
Atcheson SG, Dymeck T (2004) Rapid resolution of chronic sciatica
first line of treatment for pain from disc herniation. with intravenous infliximab after failed epidural steroid injections.
The use of NSAIDs and corticosteroids with herniation- Spine 29:E248E250
associated radiculopathy is safe, but provides no means to Atlas SJ, Keller RB, Wu YA, Deyo RA, Singer DE (2005) Long-term
modify pathology or outcomes. Emerging pharmacologi- outcomes of surgical and nonsurgical management of lumbar spinal
stenosis: 8 to 10 year results from the maine lumbar spine study.
cal therapies provide some promise here and are focused Spine (Phila Pa 1976) 30:936943
on antagonizing the many proinflammatory cytokines Awad JN, Moskovich R (2006) Lumbar disc herniations: surgical ver-
implicated in radiculopathy, most notably tumor necrosis sus nonsurgical treatment. Clin Orthop Relat Res 443:183197
factor (TNF-a). Bachmeier BE, Nerlich A, Mittermaier N, Weiler C, Lumenta C, Wuertz
K, Boos N (2009) Matrix metalloproteinase expression levels sug-
Pharmacologic interventions that show promise in the gest distinct enzyme roles during lumbar disc herniation and degen-
treatment of herniation-associated radiculopathy include eration. Eur Spine J 18:15731586
the class of competitive inhibitors of neuronal receptors Battie MC, Videman T (2006) Lumbar disc degeneration: epidemiol-
(e.g., serotonin receptor antagonists), inhibitors of glial ogy and genetics. J Bone Joint Surg Am 88(Suppl 2):39
Battie MC, Videman T, Kaprio J, Gibbons LE, Gill K, Manninen H,
proliferation (e.g., minocycline) or metabolism (e.g., Saarela J, Peltonen L (2009) The twin spine study: contributions to
methotrexate), and inhibitors of the cellular inflammatory a changing view of disc degeneration. Spine J 9:4759
pathways (e.g., NF-kB or PKC inhibitors). Beare JE, Morehouse JR, DeVries WH, Enzmann GU, Burke DA,
Additional work is needed to reveal the role of central Magnuson DS, Whittemore SR (2009) Gait analysis in normal and
spinal contused mice using the treadscan system. J Neurotrauma
nervous system sensitization and remodeling in response 26:20452056
to disc herniation and to better understand its role in the Beck KD, Nguyen HX, Galvan MD, Salazar DL, Woodruff TM,
persistent neuropathy. Anderson AJ (2010) Quantitative analysis of cellular inflammation
after traumatic spinal cord injury: evidence for a multiphasic DeLeo JA, Rutkowski MD, Stalder AK, Campbell IL (2000) Transgenic
inflammatory response in the acute to chronic environment. Brain expression of TNF by astrocytes increases mechanical allodynia in
133:433447 a mouse neuropathy model. Neuroreport 11:599602
Berryman ER, Harris RL, Moalli M, Bagi CM (2009) Digigait quantita- DeLeo JA, Tanga FY, Tawfik VL (2004) Neuroimmune activation and
tion of gait dynamics in rat rheumatoid arthritis model. neuroinflammation in chronic pain and opioid tolerance/hyperalge-
J Musculoskelet Neuronal Interact 9:8998 sia. Neuroscientist 10:4052
Binshtok AM, Wang H, Zimmermann K, Amaya F, Vardeh D, Shi L, Demircan MN, Asir A, Cetinkal A, Gedik N, Kutlay AM, Colak A et al.
Brenner GJ, Ji RR, Bean BP, Woolf CJ, Samad TA (2008) (2007) Is there any relationship between proinflammatory mediator
Nociceptors are interleukin-1beta sensors. J Neurosci 28: levels in disc material and myelopathy with cervical disc herniation
1406214073 and spondylosis? A non-randomized, prospective clinical study. Eur
Bostman OM (1993) Body mass index and height in patients requiring Spine J 16(7):983986
surgery for lumbar intervertebral disc herniation. Spine (Phila Pa Ding XL, Wang YH, Ning LP, Zhang Y, Ge HY, Jiang H, Wang R, Yue
1976) 18:851854 SW (2010) Involvement of TRPV4-NO-cGMP-PKG pathways in
Brisby H, Byrod G, Olmarker K, Miller VM, Aoki Y, Rydevik B (2000) the development of thermal hyperalgesia following chronic com-
Nitric oxide as a mediator of nucleus pulposus-induced effects on pression of the dorsal root ganglion in rats. Behav Brain Res
spinal nerve roots. J Orthop Res 18:815820 208(1):194201
Brisby H, Olmarker K, Larsson K, Nutu M, Rydevik B (2002) Doita M, Kanatani T, Harada T, Mizuno K (1996) Immunohistologic
Proinflammatory cytokines in cerebrospinal fluid and serum in study of the ruptured intervertebral disc of the lumbar spine. Spine
patients with disc herniation and sciatica. Eur Spine J 11:6266 21:235241
Buenaventura RM, Datta S, Abdi S, Smith HS (2009) Systematic review Dyck PJ, Lais AC, Giannini C, Engelstad JK (1990) Structural altera-
of therapeutic lumbar transforaminal epidural steroid injections. tions of nerve during cuff compression. Proc Natl Acad Sci U S A
Pain Physician 12:233251 87:98289832
Burke JG, Watson RW, McCormack D, Dowling FE, Walsh MG, Fraser I (2009) Statistics on hospital-based care in the United States.
Fitzpatrick JM (2002) Intervertebral discs which cause low back Agency for Healthcare Research and Quality, Rockville
pain secrete high levels of proinflammatory mediators. J Bone Joint Freemont AJ, Watkins A, Le Maitre C, Jeziorska M, Hoyland JA (2002)
Surg Br 84:196201 Current understanding of cellular and molecular events in interver-
Byrod G, Otani K, Brisby H, Rydevik B, Olmarker K (2000) tebral disc degeneration: implications for therapy. J Pathol
Methylprednisolone reduces the early vascular permeability increase 196:374379
in spinal nerve roots induced by epidural nucleus pulposus applica- Frymoyer JW (1988) Back pain and sciatica. N Engl J Med
tion. J Orthop Res 18:983987 318:291300
Cao H, Zhang YQ (2008) Spinal glial activation contributes to patho- Fumihiko H, Nobuo M, Nobuo H (1996) Changes in responses of wide
logical pain states. Neurosci Biobehav Rev 32:972983 dynamic range neurons in the spinal dorsal horn after dorsal root or
Chang YW, Winkelstein BA (2011) Schwann cell proliferation and dorsal root ganglion compression. Spine (Phila Pa 1976)
macrophage infiltration are evident at day 14 after painful cervical 21:14081414
nerve root compression in the rat. J Neurotrauma 28:24292438 Gajendran VK, Reuter MW, Golish SR, Hanna LS, Scuderi GJ (2011)
Chaplan SR, Malmberg AB, Yaksh TL (1997) Efficacy of spinal NMDA Is the fibronectin-aggrecan complex present in cervical disk dis-
receptor antagonism in formalin hyperalgesia and nerve injury ease? PM R 3:10301034
evoked allodynia in the rat. J Pharmacol Exp Ther 280:829838 Garfin SR, Cohen MS, Massie JB, Abitbol JJ, Swenson MR, Myers RR,
Chou R, Huffman LH (2007) Medications for acute and chronic low Rydevik BL (1990) Nerve-roots of the cauda equina. The effect of
back pain: a review of the evidence for an American Pain Society/ hypotension and acute graded compression on function. J Bone
American College of Physicians clinical practice guideline. Ann Joint Surg Am 72:11851192
Intern Med 147:505514 Genevay S, Stingelin S, Gabay C (2004) Efficacy of etanercept in the
Cohen SP, Bogduk N, Dragovich A, Buckenmaier CC 3rd, Griffith S, treatment of acute, severe sciatica: a pilot study. Ann Rheum Dis
Kurihara C, Raymond J, Richter PJ, Williams N, Yaksh TL (2009) 63:11201123
Randomized, double-blind, placebo-controlled, doseresponse, and Gensel JC, Tovar CA, Hamers FP, Deibert RJ, Beattie MS, Bresnahan
preclinical safety study of transforaminal epidural etanercept for the JC (2006) Behavioral and histological characterization of unilateral
treatment of sciatica. Anesthesiology 110:11161126 cervical spinal cord contusion injury in rats. J Neurotrauma
Colburn RW, Rickman AJ, DeLeo JA (1999) The effect of site and type 23:3654
of nerve injury on spinal glial activation and neuropathic pain Goupille P, Jayson MI, Valat JP, Freemont AJ (1998) The role of
behavior. Exp Neurol 157:289304 inflammation in disk herniation-associated radiculopathy. Semin
Cornefjord M, Olmarker K, Otani K, Rydevik B (2001) Effects of Arthritis Rheum 28:6071
diclofenac and ketoprofen on nerve conduction velocity in experi- Gronblad M, Virri J, Tolonen J, Seitsalo S, Kaapa E, Kankare J,
mental nerve root compression. Spine (Phila Pa 1976) Myllynen P, Karaharju EO (1994) A controlled immunohistochemi-
26(20):21932197 cal study of inflammatory cells in disc herniation tissue. Spine (Phila
Crawley JN (2007) Whats wrong with my mouse? Behavioral pheno- Pa 1976) 19:27442751
typing of transgenic and knockout mice. Wiley-Interscience, Guertin AD, Zhang DP, Mak KS, Alberta JA, Kim HA (2005)
Hoboken Microanatomy of axon/glial signaling during Wallerian degenera-
Cuellar JM, Montesano PX, Carstens E (2004) Role of TNF-alpha in tion. J Neurosci 25:34783487
sensitization of nociceptive dorsal horn neurons induced by applica- Hanai F, Matsui N, Hongo N (1996) Changes in responses of wide
tion of nucleus pulposus to L5 dorsal root ganglion in rats. Pain dynamic range neurons in the spinal dorsal horn after dorsal root or
110:578587 dorsal root ganglion compression. Spine 21:14081414; discussion
Cuellar JM, Montesano PX, Antognini JF, Carstens E (2005) Application 14141415
of nucleus pulposus to L5 dorsal root ganglion in rats enhances Hart LG, Deyo RA, Cherkin DC (1995) Physician office visits for
nociceptive dorsal horn neuronal windup. J Neurophysiol 94:3548 low back pain. Frequency, clinical evaluation, and treatment pat-
DeLeo JA, Yezierski RP (2001) The role of neuroinflammation and terns from a US national survey. Spine (Phila Pa 1976) 20:
neuroimmune activation in persistent pain. Pain 90:16 1119
Hashizume H, DeLeo JA, Colburn RW, Weinstein JN (2000a) Spinal Jancalek R, Dubovy P (2007) An experimental animal model of spinal
glial activation and cytokine expression after lumbar root injury in root compression syndrome: an analysis of morphological changes
the rat. Spine (Phila Pa 1976) 25:12061217 of myelinated axons during compression radiculopathy and after
Hashizume H, Rutkowski MD, Weinstein JN, DeLeo JA (2000b) decompression. Exp Brain Res 179:111119
Central administration of methotrexate reduces mechanical allo- Julius D, Basbaum AI (2001) Molecular mechanisms of nociception.
dynia in an animal model of radiculopathy/sciatica. Pain Nature 413:203210
87(2):159169 Kajander KC, Pollock CH, Berg H (1996) Evaluation of hindpaw posi-
Hashizume H, Kawakami M, Yoshida M, Okada M, Enyo Y, Inomata Y tion in rats during chronic constriction injury (CCI) produced with
(2007) Sarpogrelate hydrochloride, a 5-HT2a receptor antagonist, different suture materials. Somatosens Mot Res 13:95101
attenuates neurogenic pain induced by nucleus pulposus in rats. Kallakuri S, Takebayashi T, Ozaktay AC, Chen C, Yang S, Wooley PH,
Spine (Phila Pa 1976) 32:315320 Cavanaugh JM (2005) The effects of epidural application of
Heliovaara M (1987a) Body height, obesity, and risk of herniated lum- allografted nucleus pulposus in rats on cytokine expression, limb
bar intervertebral disc. Spine (Phila Pa 1976) 12:469472 withdrawal and nerve root discharge. Eur Spine J 14:956964
Heliovaara M (1987b) Occupation and risk of herniated lumbar inter- Kanayama M, Hashimoto T, Shigenobu K, Yamane S (2003) Efficacy
vertebral disc or sciatica leading to hospitalization. J Chronic Dis of serotonin receptor blocker for symptomatic lumbar disc hernia-
40:259264 tion. Clin Orthop Relat Res (411):159165
Heliovaara M, Impivaara O, Sievers K, Melkas T, Knekt P, Korpi J, Kang JD, Stefanovic-Racic M, McIntyre LA, Georgescu HI, Evans CH
Aromaa A (1987a) Lumbar disc syndrome in Finland. J Epidemiol (1997) Toward a biochemical understanding of human interverte-
Community Health 41:251258 bral disc degeneration and herniation. Contributions of nitric oxide,
Heliovaara M, Knekt P, Aromaa A (1987b) Incidence and risk factors of interleukins, prostaglandin E2, and matrix metalloproteinases.
herniated lumbar intervertebral disc or sciatica leading to hospital- Spine 22(10):10651073
ization. J Chronic Dis 40:251258 Karppinen J, Korhonen T, Malmivaara A, Paimela L, Kyllonen E,
Hogan QH (2007) Role of decreased sensory neuron membrane cal- Lindgren KA, Rantanen P, Tervonen O, Niinimaki J, Seitsalo S,
cium currents in the genesis of neuropathic pain. Croat Med J Hurri H (2003) Tumor necrosis factor-alpha monoclonal antibody,
48:921 infliximab, used to manage severe sciatica. Spine 28:750753; dis-
Hou SX, Tang JG, Chen HS, Chen J (2003) Chronic inflammation and cussion 753754
compression of the dorsal root contribute to sciatica induced by the Kasimcan O, Kaptan H (2010) Efficacy of gabapentin for radiculopathy
intervertebral disc herniation in rats. Pain 105:255264 caused by lumbar spinal stenosis and lumbar disk hernia. Neurol
Hu SJ, Xing JL (1998) An experimental model for chronic compression Med Chir (Tokyo) 50:10701073
of dorsal root ganglion produced by intervertebral foramen stenosis Kawakami M, Weinstein JN, Chatani K, Spratt KF, Meller ST, Gebhart
in the rat. Pain 77:1523 GF (1994a) Experimental lumbar radiculopathy. Behavioral and
Huang KY, Lin RM, Chen WY, Lee CL, Yan JJ, Chang MS (2008) histologic changes in a model of radicular pain after spinal nerve
IL-20 may contribute to the pathogenesis of human intervertebral root irritation with chromic gut ligatures in the rat. Spine (Phila Pa
disc herniation. Spine 33(19):20342040 1976) 19:17951802
Hubbard RD, Winkelstein BA (2005) Transient cervical nerve root Kawakami M, Weinstein JN, Spratt KF, Chatani K, Traub RJ, Meller
compression in the rat induces bilateral forepaw allodynia and spi- ST, Gebhart GF (1994b) Experimental lumbar radiculopathy.
nal glial activation: mechanical factors in painful neck injuries. Immunohistochemical and quantitative demonstrations of pain
Spine (Phila Pa 1976) 30:19241932 induced by lumbar nerve root irritation of the rat. Spine (Phila Pa
Hubbard RD, Winkelstein BA (2008) Dorsal root compression pro- 1976) 19:17801794
duces myelinated axonal degeneration near the biomechanical Kawakami M, Matsumoto T, Kuribayashi K, Tamaki T (1999) mRNA
thresholds for mechanical behavioral hypersensitivity. Exp Neurol expression of interleukins, phospholipase A2, and nitric oxide syn-
212:482489 thase in the nerve root and dorsal root ganglion induced by autolo-
Hubbard RD, Chen Z, Winkelstein BA (2008a) Transient cervical nerve gous nucleus pulposus in the rat. J Orthop Res 17:941946
root compression modulates pain: load thresholds for allodynia and Kawakami M, Matsumoto T, Tamaki T (2001) Roles of thromboxane
sustained changes in spinal neuropeptide expression. J Biomech A2 and leukotriene B4 in radicular pain induced by herniated
41:677685 nucleus pulposus. J Orthop Res 19(3):472477
Hubbard RD, Quinn KP, Martinez JJ, Winkelstein BA (2008b) The Kawakami M, Hashizume H, Nishi H, Matsumoto T, Tamaki T,
role of graded nerve root compression on axonal damage, neuro- Kuribayashi K (2003) Comparison of neuropathic pain induced by
peptide changes, and pain-related behaviors. Stapp Car Crash J the application of normal and mechanically compressed nucleus
52:3358 pulposus to lumbar nerve roots in the rat. J Orthop Res 21:535539
Hubbard RD, Martinez JJ, Burdick JA, Winkelstein BA (2009) Kobayashi S, Yoshizawa H, Yamada S (2004a) Pathology of lumbar
Controlled release of GDNF reduces nerve root-mediated behav- nerve root compression. Part 1: intraradicular inflammatory changes
ioral hypersensitivity. J Orthop Res 27:120127 induced by mechanical compression. J Orthop Res 22:170179
Hwang PY, Allen KD, Shamji MF, Jing L, Mata BA, Gabr MA, Kobayashi S, Yoshizawa H, Yamada S (2004b) Pathology of lumbar nerve
Huebner JL, Kraus VB, Richardson WJ, Setton LA (2012) Changes root compression. Part 2: morphological and immunohistochemical
in midbrain pain receptor expression, gait and behavioral sensitivity changes of dorsal root ganglion. J Orthop Res 22:180188
in a rat model of radiculopathy. Open Orthop J 6:383391. PubMed Kobayashi S, Baba H, Uchida K, Kokubo Y, Kubota C, Yamada S,
Central PMCID: PMC3434701 Suzuki Y, Yoshizawa H (2005a) Effect of mechanical compression
Igarashi T, Kikuchi S, Shubayev V, Myers RR (2000) 2000 Volvo on the lumbar nerve root: localization and changes of intraradicular
Award winner in Basic Science Studies: exogenous tumor necrosis inflammatory cytokines, nitric oxide, and cyclooxygenase. Spine
factor-alpha mimics nucleus pulposus-induced neuropathology. 30:16991705
Molecular, histologic, and behavioral comparisons in rats. Spine Kobayashi S, Kokubo Y, Uchida K, Yayama T, Takeno K, Negoro K,
25:29752980 Nakajima H, Baba H, Yoshizawa H (2005b) Effect of lumbar nerve
Inoue K (2006) The function of microglia through purinergic receptors: root compression on primary sensory neurons and their central
neuropathic pain and cytokine release. Pharmacol Ther 109: branches: changes in the nociceptive neuropeptides substance P and
210226 somatostatin. Spine (Phila Pa 1976) 30:276282
Kobayashi S, Meir A, Baba H, Uchida K, Hayakawa K (2005c) Imaging Mulleman D, Mammou S, Griffoul I, Watier H, Goupille P (2006a)
of intraneural edema by using gadolinium-enhanced MR imaging: Pathophysiology of disk-related low back pain and sciatica. II.
experimental compression injury. AJNR Am J Neuroradiol Evidence supporting treatment with TNF-alpha antagonists. Joint
26:973980 Bone Spine 73:270277
Kobayashi S, Sasaki S, Shimada S, Kaneyasu M, Mizukami Y, Kitade Mulleman D, Mammou S, Griffoul I, Watier H, Goupille P (2006b)
I, Ogawa M, Kawahara H, Baba H, Yoshizawa H (2005d) Changes Pathophysiology of disk-related sciatica. I Evidence supporting a
of calcitonin gene-related peptide in primary sensory neurons and chemical component. Joint Bone Spine 73:151158
their central branch after nerve root compression of the dog. Arch Murata Y, Onda A, Rydevik B, Takahashi K, Olmarker K (2004a)
Phys Med Rehabil 86:527533 Distribution and appearance of tumor necrosis factor-alpha in the
Kobayashi S, Uchida K, Kokubo Y, Takeno K, Yayama T, Miyazaki T, dorsal root ganglion exposed to experimental disc herniation in rats.
Nakajima H, Nomura E, Mwaka E, Baba H (2008) Synapse involve- Spine (Phila Pa 1976) 29:22352241
ment of the dorsal horn in experimental lumbar nerve root compres- Murata Y, Onda A, Rydevik B, Takahashi K, Olmarker K (2004b)
sion: a light and electron microscopic study. Spine (Phila Pa 1976) Selective inhibition of tumor necrosis factor-alpha prevents nucleus
33:716723 pulposus-induced histologic changes in the dorsal root ganglion.
Komori H, Shinomiya K, Nakai O, Yamaura I, Takeda S, Furuya K Spine 29:24772484
(1996) The natural history of herniated nucleus pulposus with radic- Myers RR, Yamamoto T, Yaksh TL, Powell HC (1993) The role of
ulopathy. Spine (Phila Pa 1976) 21:225229 focal nerve ischemia and wallerian degeneration in peripheral
Korhonen T, Karppinen J, Paimela L, Malmivaara A, Lindgren KA, nerve injury producing hyperesthesia. Anesthesiology 78:
Jarvinen S, Niinimaki J, Veeger N, Seitsalo S, Hurri H (2005) The 308316
treatment of disc herniation-induced sciatica with infliximab: results Nakamae T, Ochi M, Olmarker K (2011) Pharmacological inhibition of
of a randomized, controlled, 3-month follow-up study. Spine tumor necrosis factor may reduce pain behavior changes induced by
30:27242728 experimental disc puncture in the rat: an experimental study in rats.
Korhonen T, Karppinen J, Paimela L, Malmivaara A, Lindgren KA, Spine (Phila Pa 1976) 36:E232E236
Bowman C, Hammond A, Kirkham B, Jarvinen S, Niinimaki J, Nicholson KJ, Quindlen JC, Winkelstein BA (2011) Development of a
Veeger N, Haapea M, Torkki M, Tervonen O, Seitsalo S, Hurri H duration threshold for modulating evoked neuronal responses after
(2006) The treatment of disc-herniation-induced sciatica with nerve root compression injury. Stapp Car Crash J 55:124
infliximab: one-year follow-up results of first ii, a randomized con- Nicholson KJ, Guarino BB, Winkelstein BA (2012) Transient nerve
trolled trial. Spine 31:27592766 root compression load and duration differentially mediate behav-
Le Maitre CL, Freemont AJ, Hoyland JA (2005) The role of interleu- ioral sensitivity and associated spinal astrocyte activation and
kin-1 in the pathogenesis of human intervertebral disc degenera- mGLuR5 expression. Neuroscience 209:187195
tion. Arthritis Res Ther 7(4):R732R745 Nilsson E, Nakamae T, Olmarker K (2011) Pain behavior changes fol-
Le Maitre CL, Hoyland JA, Freemont AJ (2007) Catabolic cytokine lowing disc puncture relate to nucleus pulposus rather than to the
expression in degenerate and herniated human intervertebral discs: disc injury per se: an experimental study in rats. Open Orthop J
IL-1beta and TNFalpha expression profile. Arthritis Res Ther 5:7277
9:R77 Nygaard OP, Mellgren SI, Osterud B (1997) The inflammatory proper-
Lindenlaub T, Teuteberg P, Hartung T, Sommer C (2000) Effects of ties of contained and noncontained lumbar disc herniation. Spine
neutralizing antibodies to TNF-alpha on pain-related behavior and 22(21):24842488
nerve regeneration in mice with chronic constriction injury. Brain Obata K, Tsujino H, Yamanaka H, Yi D, Fukuoka T, Hashimoto N,
Res 866:1522 Yonenobu K, Yoshikawa H, Noguchi K (2002) Expression of neu-
Loupasis GA, Stamos K, Katonis PG, Sapkas G, Korres DS, rotrophic factors in the dorsal root ganglion in a rat model of lumbar
Hartofilakidis G (1999) Seven- to 20-year outcome of lumbar dis- disc herniation. Pain 99:121132
cectomy. Spine (Phila Pa 1976) 24:23132317 Obata K, Yamanaka H, Dai Y, Mizushima T, Fukuoka T, Tokunaga A,
Marriott I (2004) The role of tachykinins in central nervous system Noguchi K (2004) Activation of extracellular signal-regulated pro-
inflammatory responses. Front Biosci 9:21532165 tein kinase in the dorsal root ganglion following inflammation near
Maves TJ, Pechman PS, Gebhart GF, Meller ST (1993) Possible chemi- the nerve cell body. Neuroscience 126:10111021
cal contribution from chromic gut sutures produces disorders of ODonnell JL, ODonnell AL (1996) Prostaglandin E2 content in herni-
pain sensation like those seen in man. Pain 54:5769 ated lumbar disc disease. Spine 21(14):16531655; discussion
McCarron RF, Wimpee MW, Hudkins PG, Laros GS (1987) The 16551656
inflammatory effect of nucleus pulposus. A possible element in the Ohtori S, Takahashi K, Aoki Y, Doya H, Ozawa T, Saito T, Moriya H
pathogenesis of low-back pain. Spine (Phila Pa 1976) 12: (2004) Spinal neural cyclooxygenase-2 mediates pain caused in a
760764 rat model of lumbar disk herniation. J Pain 5(7):385391
Mixter CG, Blumgart HL, Berlin DD (1934) Total ablation of the thy- Olmarker K (2001) Radicular pain recent pathophysiologic concepts
roid for angina pectoris and congestive heart failure: results of eigh- and therapeutic implications. Schmerz 15:425429
teen months experience. Ann Surg 100:570577 Olmarker K (2008) Puncture of a lumbar intervertebral disc induces
Moalem G, Tracey DJ (2006) Immune and inflammatory mechanisms changes in spontaneous pain behavior: an experimental study in
in neuropathic pain. Brain Res Rev 51:240264 rats. Spine 33:850855
Moore RJ, Vernon-Roberts B, Fraser RD, Osti OL, Schembri M (1996) Olmarker K, Larsson K (1998) Tumor necrosis factor alpha and nucleus-
The origin and fate of herniated lumbar intervertebral disc tissue. pulposus-induced nerve root injury. Spine (Phila Pa 1976)
Spine 21:21492155 23:25382544
Morishita Y, Hida S, Naito M, Arimizu J, Matsushima U, Nakamura A Olmarker K, Myers R (1998) Pathogenesis of sciatic pain: role of herni-
(2006) Measurement of the local pressure of the intervertebral fora- ated nucleus pulposus and deformation of spinal nerve root and dor-
men and the electrophysiologic values of the spinal nerve roots in sal root ganglion. Pain 78:99105
the vertebral foramen. Spine (Phila Pa 1976) 31:30763080 Olmarker K, Rydevik B (1991) Pathophysiology of sciatica. Orthop
Mosconi T, Kruger L (1996) Fixed-diameter polyethylene cuffs applied Clin North Am 22:223234
to the rat sciatic nerve induce a painful neuropathy: ultrastructural Olmarker K, Rydevik B (2001) Selective inhibition of tumor
morphometric analysis of axonal alterations. Pain 64:3757 necrosis factor-alpha prevents nucleus pulposus-induced thrombus
formation, intraneural edema, and reduction of nerve conduction Roberts S, Evans H, Trivedi J, Menage J (2006) Histology and pathol-
velocity: possible implications for future pharmacologic treatment ogy of the human intervertebral disc. J Bone Joint Surg Am 88(Suppl
strategies of sciatica. Spine 26:863869 2):1014
Olmarker K, Rydevik B, Holm S (1989) Edema formation in spinal Robinson I, Meert TF (2005) Stability of neuropathic pain symptoms in
nerve roots induced by experimental, graded compression. An partial sciatic nerve ligation in rats is affected by suture material.
experimental study on the pig cauda equina with special reference to Neurosci Lett 373:125129
differences in effects between rapid and slow onset of compression. Rothman SM, Winkelstein BA (2007) Chemical and mechanical nerve
Spine (Phila Pa 1976) 14:569573 root insults induce differential behavioral sensitivity and glial acti-
Olmarker K, Rydevik B, Nordborg C (1993) Autologous nucleus pul- vation that are enhanced in combination. Brain Res 1181:3043
posus induces neurophysiologic and histologic changes in porcine Rothman SM, Winkelstein BA (2010) Cytokine antagonism reduces
cauda equina nerve roots. Spine (Phila Pa 1976) 18:14251432 pain and modulates spinal astrocytic reactivity after cervical nerve
Olmarker K, Byrod G, Cornefjord M, Nordborg C, Rydevik B (1994) root compression. Ann Biomed Eng 38(8):25632576
Effects of methylprednisolone on nucleus pulposus-induced nerve Rothman SM, Kreider RA, Winkelstein BA (2005) Spinal neuropeptide
root injury. Spine (Phila Pa 1976) 19:18031808 responses in persistent and transient pain following cervical nerve
Olmarker K, Brisby H, Yabuki S, Nordborg C, Rydevik B (1997) The root injury. Spine (Phila Pa 1976) 30:24912496
effects of normal, frozen, and hyaluronidase-digested nucleus pul- Rothman SM, Guarino BB, Winkelstein BA (2009a) Spinal microglial
posus on nerve root structure and function. Spine 22:471475; dis- proliferation is evident in a rat model of painful disc herniation both
cussion 476 in the presence of behavioral hypersensitivity and following mino-
Olmarker K, Iwabuchi M, Larsson K, Rydevik B (1998) Walking analy- cycline treatment sufficient to attenuate allodynia. J Neurosci Res
sis of rats subjected to experimental disc herniation. Eur Spine J 87:27092717
7:394399 Rothman SM, Huang Z, Lee KE, Weisshaar CL, Winkelstein BA
Olmarker K, Storkson R, Berge OG (2002) Pathogenesis of sciatic pain: (2009b) Cytokine mRNA expression in painful radiculopathy. J Pain
a study of spontaneous behavior in rats exposed to experimental 10(1):9099
disc herniation. Spine 27:13121317 Rothman SM, Nicholson KJ, Winkelstein BA (2010) Time-dependent
Olmarker K, Nutu M, Storkson R (2003) Changes in spontaneous behav- mechanics and measures of glial activation and behavioral sensitiv-
ior in rats exposed to experimental disc herniation are blocked by ity in a rodent model of radiculopathy. J Neurotrauma 27:
selective TNF-alpha inhibition. Spine 28:16351641; discussion 1642 803814
Onda A, Hamba M, Yabuki S, Kikuchi S (2002) Exogenous tumor Rutkowski MD, Winkelstein BA, Hickey WF, Pahl JL, DeLeo JA
necrosis factor-alpha induces abnormal discharges in rat dorsal horn (2002) Lumbar nerve root injury induces central nervous system
neurons. Spine (Phila Pa 1976) 27:16181624; discussion 1624 neuroimmune activation and neuroinflammation in the rat: relation-
Onda A, Murata Y, Rydevik B, Larsson K, Kikuchi S, Olmarker K ship to painful radiculopathy. Spine (Phila Pa 1976) 27:16041613
(2004) Infliximab attenuates immunoreactivity of brain-derived Rydevik B (1994) Sciatica and herniated disk. Current aspects of
neurotrophic factor in a rat model of herniated nucleus pulposus. pathophysiology and pain mechanisms. Nord Med 109:7476, 80
Spine 29:18571861 Rydevik B, Brown MD, Lundborg G (1984) Pathoanatomy and
Onda A, Murata Y, Rydevik B, Larsson K, Kikuchi S, Olmarker K pathophysiology of nerve root compression. Spine (Phila Pa 1976)
(2005) Nerve growth factor content in dorsal root ganglion as related 9:715
to changes in pain behavior in a rat model of experimental lumbar Rydevik BL, Pedowitz RA, Hargens AR, Swenson MR, Myers RR,
disc herniation. Spine (Phila Pa 1976) 30:188193 Garfin SR (1991) Effects of acute, graded compression on spinal
Otani K, Arai I, Mao GP, Konno S, Olmarker K, Kikuchi S (1997) nerve root function and structure. An experimental study of the pig
Experimental disc herniation: evaluation of the natural course. Spine cauda equina. Spine (Phila Pa 1976) 16:487493
(Phila Pa 1976) 22:28942899 Saab CY, Waxman SG, Hains BC (2008) Alarm or curse? The pain of
Ozaktay AC, Cavanaugh JM, Asik I, DeLeo JA, Weinstein JN (2002) neuroinflammation. Brain Res Rev 58:226235
Dorsal root sensitivity to interleukin-1 beta, interleukin-6 and tumor Saal JS (1995) The role of inflammation in lumbar pain. Spine (Phila Pa
necrosis factor in rats. Eur Spine J 11:467475 1976) 20:18211827
Ozaktay AC, Kallakuri S, Takebayashi T, Cavanaugh JM, Asik I, DeLeo Saal JS, Franson RC, Dobrow R, Saal JA (1990) High levels of
JA, Weinstein JN (2006) Effects of interleukin-1 beta, interleukin-6, inflammatory phospholipase A2 activity in lumbar disc herniations.
and tumor necrosis factor on sensitivity of dorsal root ganglion and Spine 15:674678
peripheral receptive fields in rats. Eur Spine J 15:15291537 Sapunar D, Ljubkovic M, Lirk P, McCallum JB, Hogan QH (2005)
Park JB, Chang H, Kim YS (2002) The pattern of interleukin-12 and Distinct membrane effects of spinal nerve ligation on injured and
T-helper types 1 and 2 cytokine expression in herniated lumbar disc adjacent dorsal root ganglion neurons in rats. Anesthesiology
tissue. Spine (Phila Pa 1976) 27(19):21252128 103:360376
Pedowitz RA, Garfin SR, Massie JB, Hargens AR, Swenson MR, Myers Sasaki N, Kikuchi S, Konno S, Sekiguchi M, Watanabe K (2007) Anti-
RR, Rydevik BL (1992) Effects of magnitude and duration of TNF-alpha antibody reduces pain-behavioral changes induced by
compression on spinal nerve root conduction. Spine (Phila Pa 1976) epidural application of nucleus pulposus in a rat model depending
17:194199 on the timing of administration. Spine (Phila Pa 1976) 32:
Piesla MJ, Leventhal L, Strassle BW, Harrison JE, Cummons TA, Lu P, 413416
Whiteside GT (2009) Abnormal gait, due to inflammation but not Schafers M, Svensson CI, Sommer C, Sorkin LS (2003) Tumor necro-
nerve injury, reflects enhanced nociception in preclinical pain mod- sis factor-alpha induces mechanical allodynia after spinal nerve
els. Brain Res 1295:8998 ligation by activation of p38 MAPK in primary sensory neurons.
Ramer LM, Richter MW, Roskams AJ, Tetzlaff W, Ramer MS (2004) J Neurosci 23(7):25172521
Peripherally-derived olfactory ensheathing cells do not promote pri- Schafers M, Marziniak M, Sorkin LS, Yaksh TL, Sommer C (2004)
mary afferent regeneration following dorsal root injury. Glia Cyclooxygenase inhibition in nerve-injury- and TNF-induced hype-
47:189206 ralgesia in the rat. Exp Neurol 185(1):160168
Ricci JA, Stewart WF, Chee E, Leotta C, Foley K, Hochberg MC (2006) Sekiguchi Y, Kikuchi S, Myers RR, Campana WM (2003) ISSLS prize
Back pain exacerbations and lost productive time costs in United winner: erythropoietin inhibits spinal neuronal apoptosis and pain
States workers. Spine (Phila Pa 1976) 31:30523060 following nerve root crush. Spine (Phila Pa 1976) 28:25772584
Sekiguchi M, Otoshi K, Kikuchi S, Konno S (2011) Analgesic effects Tobinick EL, Britschgi-Davoodifar S (2003) Perispinal TNF-alpha
of prostaglandin E2 receptor subtype EP1 receptor antagonist: inhibition for discogenic pain. Swiss Med Wkly 133:170177
experimental study of application of nucleus pulposus. Spine Tokunaga S, Yamanokuchi K, Yabuki A, Fujiki M, Misumi K (2010)
36(22):18291834 Cartilage oligomeric matrix protein in canine spinal cord appears in
Sekiguchi M, Konno S, Kikuchi S (2008) The effects of a 5-HT2A the cerebrospinal fluid associated with intervertebral disc hernia-
receptor antagonist on blood flow in lumbar disc herniation: application. Spine (Phila Pa 1976) 35:49
tion of nucleus pulposus in a canine model. Eur Spine J 17:307313 Uceyler N, Tscharke A, Sommer C (2008) Early cytokine gene expres-
Sekiguchi M, Sekiguchi Y, Konno S, Kobayashi H, Homma Y, Kikuchi sion in mouse CNS after peripheral nerve lesion. Neurosci Lett
S (2009) Comparison of neuropathic pain and neuronal apoptosis 436(2):259264
following nerve root or spinal nerve compression. Eur Spine J van der Windt DAWM, Simons E, Riphagen II, Ammendolia C,
18:19781985 Verhagen AP, Laslett M, Devill W, Deyo RA, Bouter LM, de Vet
Shamji MF, Whitlatch L, Friedman AH, Richardson WJ, Chilkoti A, HCW, Aertgeerts B (2010) Physical examination for lumbar radicu-
Setton LA (2008) An injectable and in situ-gelling biopolymer for lopathy due to disc herniation in patients with low-back pain.
sustained drug release following perineural administration. Spine Cochrane Database Syst Rev (2):CD007431
(Phila Pa 1976) 33:748754 Van Zundert J, Harney D, Joosten EA, Durieux ME, Patijn J, Prins MH,
Shamji MF, Allen KD, So S, Jing L, Adams SB Jr, Schuh R, Huebner J, Van Kleef M (2006) The role of the dorsal root ganglion in cervical
Kraus VB, Friedman AH, Setton LA, Richardson WJ (2009) Gait radicular pain: diagnosis, pathophysiology, and rationale for treat-
abnormalities and inflammatory cytokines in an autologous nucleus ment. Reg Anesth Pain Med 31:152167
pulposus model of radiculopathy. Spine (Phila Pa 1976) 34:648654 Verri WA Jr, Cunha TM, Parada CA, Poole S, Cunha FQ, Ferreira
Shamji MF, Setton LA, Jarvis W, So S, Chen J, Jing L, Bullock R, SH (2006) Hypernociceptive role of cytokines and chemokines: targets
Isaacs RE, Brown C, Richardson WJ (2010) Proinflammatory for analgesic drug development? Pharmacol Ther 112:116138
cytokine expression profile in degenerated and herniated human Vrinten DH, Hamers FF (2003) Catwalk automated quantitative gait
intervertebral disc tissues. Arthritis Rheum 62:19741982 analysis as a novel method to assess mechanical allodynia in the rat;
Skouen J, Brisby H, Otani K, Olmarker K, Rosengren L, Rydevik B a comparison with von Frey testing. Pain 102:203209
(1999) Protein markers in cerebrospinal fluid in experimental nerve Weiler C, Nerlich AG, Bachmeier BE, Boos N (2005) Expression and
root injury. A study of slow-onset chronic compression effects or distribution of tumor necrosis factor alpha in human lumbar inter-
the biochemical effects of nucleus pulposus on sacral nerve roots. vertebral discs: a study in surgical specimen and autopsy controls.
Spine 24:21952200 Spine (Phila Pa 1976) 30:4453; discussion 54
Sommer C, Schmidt C, George A, Toyka KV (1997) A metallopro- Weiler C, Lopez-Ramos M, Mayer HM, Korge A, Siepe CJ, Wuertz K,
tease-inhibitor reduces pain associated behavior in mice with exper- Weiler V, Boos N, Nerlich AG (2011) Histological analysis of surgi-
imental neuropathy. Neurosci Lett 237(1):4548 cal lumbar intervertebral disc tissue provides evidence for an asso-
Sommer C, Schafers M, Marziniak M, Toyka KV (2001) Etanercept ciation between disc degeneration and increased body mass index.
reduced hyperalgesia in experimental painful neuropathy. J Peripher BMC Res Notes 4:497
Nerv Syst 6:6772 Weinstein JN, Lurie JD, Tosteson TD, Skinner JS, Hanscom B, Tosteson
Song XJ, Wang ZB, Gan Q, Walters ET (2006) cAMP and cGMP contrib- AN, Herkowitz H, Fischgrund J, Cammisa FP, Albert T, Deyo RA
ute to sensory neuron hyperexcitability and hyperalgesia in rats with (2006a) Surgical vs nonoperative treatment for lumbar disk hernia-
dorsal root ganglia compression. J Neurophysiol 95(1):479492 tion: the spine patient outcomes research trial (sport) observational
Specchia N, Pagnotta A, Toesca A, Greco F (2002) Cytokines and cohort. JAMA 296:24512459
growth factors in the protruded intervertebral disc of the lumbar Weinstein JN, Tosteson TD, Lurie JD, Tosteson AN, Hanscom B,
spine. Eur Spine J 11:145151 Skinner JS, Abdu WA, Hilibrand AS, Boden SD, Deyo RA (2006b)
Staal JB, de Bie R, de Vet HC, Hildebrandt J, Nelemans P (2008) Surgical vs nonoperative treatment for lumbar disk herniation: the
Injection therapy for subacute and chronic low-back pain. Cochrane spine patient outcomes research trial (sport): a randomized trial.
Database Syst Rev (3):CD001824 JAMA 296:24412450
Stoll G, Jander S (1999) The role of microglia and macrophages in the Weisshaar CL, Winer JP, Guarino BB, Janmey PA, Winkelstein BA
pathophysiology of the CNS. Prog Neurobiol 58:233247 (2011) The potential for salmon fibrin and thrombin to mitigate
Stoll G, Muller HW (1999) Nerve injury, axonal degeneration and neu- pain subsequent to cervical nerve root injury. Biomaterials
ral regeneration: basic insights. Brain Pathol 9:313325 32(36):97389746
Suter MR, Wen YR, Decosterd I, Ji RR (2007) Do glial cells control Whishaw IQ, Kolb B (2005) The behavior of the laboratory rat: a hand-
pain? Neuron Glia Biol 3:255268 book with tests. Oxford University Press, Oxford/New York
Suzuki M, Inoue G, Gemba T, Watanabe T, Ito T, Koshi T, Yamauchi K, Willburger RE, Wittenberg RH (1994) Prostaglandin release from
Yamashita M, Orita S, Eguchi Y, Ochiai N, Kishida S, Takaso M, lumbar disc and facet joint tissue. Spine (Phila Pa 1976) 19(18):
Aoki Y, Takahashi K, Ohtori S (2009) Nuclear factor-kappa B decoy 20682070
suppresses nerve injury and improves mechanical allodynia and Winkelstein BA, DeLeo JA (2004) Mechanical thresholds for initia-
thermal hyperalgesia in a rat lumbar disc herniation model. Eur tion and persistence of pain following nerve root injury: mechani-
Spine J 18(7):10011007 cal and chemical contributions at injury. J Biomech Eng 126:
Takahashi H, Suguro T, Okazima Y, Motegi M, Okada Y, Kakiuchi T 258263
(1996) Inflammatory cytokines in the herniated disc of the lumbar Winkelstein BA, Rutkowski MD, Sweitzer SM, Pahl JL, DeLeo JA
spine. Spine (Phila Pa 1976) 21(2):218224 (2001a) Nerve injury proximal or distal to the DRG induces similar
Takahashi N, Yabuki S, Aoki Y, Kikuchi S (2003) Pathomechanisms of spinal glial activation and selective cytokine expression but differ-
nerve root injury caused by disc herniation: an experimental study ential behavioral responses to pharmacologic treatment. J Comp
of mechanical compression and chemical irritation. Spine (Phila Pa Neurol 439(2):127139
1976) 28:435441 Winkelstein BA, Rutkowski MD, Weinstein JN, DeLeo JA (2001b)
Takamori Y, Arimizu J, Izaki T, Naito M, Kobayashi T (2011) Combined Quantification of neural tissue injury in a rat radiculopathy
measurement of nerve root blood flow and electrophysiological model: comparison of local deformation, behavioral outcomes, and
values: intraoperative straight-leg-raising test for lumbar disc spinal cytokine mRNA for two surgeons. J Neurosci Methods 111:
herniation. Spine (Phila Pa 1976) 36:5762 4957
Winkelstein BA, Weinstein JN, DeLeo JA (2002) The role of mechani- constriction injury to the sciatic nerve in the rat. Neurosci Lett
cal deformation in lumbar radiculopathy: an in vivo model. Spine 202:8992
(Phila Pa 1976) 27:2733 Yildirim K, Deniz O, Gureser G, Karatay S, Ugur M, Erdal A, Senel K
Xu J, Pollock CH, Kajander KC (1996) Chromic gut suture reduces (2009) Gabapentin monotherapy in patients with chronic radicul-
calcitonin-gene-related peptide and substance P levels in the spinal opathy: the efficacy and impact on life quality. J Back Musculoskelet
cord following chronic constriction injury in the rat. Pain 64: Rehabil 22:1720
503509 Zanella JM, Burright EN, Hildebrand K, Hobot C, Cox M, Christoferson
Yabuki S, Kawaguchi Y, Nordborg C, Kikuchi S, Rydevik B, Olmarker L, McKay WF (2008) Effect of etanercept, a tumor necrosis factor-
K (1998) Effects of lidocaine on nucleus pulposus-induced nerve alpha inhibitor, on neuropathic pain in the rat chronic constriction
root injury: a neurophysiologic and histologic study of the pig cauda injury model. Spine (Phila Pa 1976) 33(3):227234
equina. Spine 23:23832389
Yamamoto T, Nozaki-Taguchi N (1995) Role of cholecystokinin-B recep-
tor in the maintenance of thermal hyperalgesia induced by unilateral
The Sand Rat (Psammomys obesus
obesus) Model of Spontaneous, 20
Age-Related Intervertebral Disc
Degeneration
Helen E. Gruber and Edward N. Hanley Jr.

20.1 Introduction ..................................................................... 327
20.1.1 Need for a Reliable, Economical Small Animal 20.1.1 Need for a Reliable, Economical Small
Model for Degenerative Disc Research ............................ 327 Animal Model for Degenerative Disc
20.1.2 Animal Models Using the Stab Technique to Induce
Degeneration Are Not Optimum ....................................... 328
Research
20.2 Review of the Early Literature With its associated low back pain, disc degeneration is a
on Disc Degeneration in the Sand Rat........................... 328
major health-care concern causing disability. It plays a major
20.3 Newer Findings on Sand Rat Disc Degeneration ......... 328 role in the US medical, social, and economic structure. Low
20.3.1 Morphologic Changes with Aging .................................... 328
20.3.2 Radiology and Micro-CT .................................................. 330
back pain is devastating and influences the quality of life for
20.3.3 Spin-Lock (T1r) Imaging ................................................... 332 millions. The lifetime prevalence of low back pain approxi-
20.3.4 Disc Cell Death and End Plate Bone mates 80 % with an estimated direct cost burden of $86 bil-
Mineral Density (BMD) .................................................... 332 lion per year. Patients with back pain incur higher health-care
20.3.5 Vascular Supply to the Disc .............................................. 333
costs and utilization and greater work loss than do patients
20.4 Spine Fusion Model......................................................... 333 without back pain. These health-care statistics point to the
20.5 Autologous Disc Cell Implantation need for appropriate animal models for translational research
and Annulus Fibrosus Cell Culture ............................... 333 efforts directed towards a greater understanding of disc
20.6 Utility and Challenges of the Sand Rat Model ............. 335 degeneration and models to test potential new therapies.
20.6.1 A Nontraditional Species .................................................. 335 In November 1995, an NIH workshop was held on New
20.6.2 Other Topics Related to Animal Husbandry ..................... 337 Horizons in Low Back Pain. The publication which grew
20.7 Importance of Animal Models Such from this workshop emphasized the importance of this costly
as the Sand Rat for Disc Research ................................. 337 health-care problem and recommended areas of future
20.8 Summary of Critical Concepts Discussed research, including the importance of animal models for the
in the Chapter .................................................................. 337 study of disc degeneration. In his review of animal models,
References ...................................................................................... 338 Krag commented that the usefulness of a particular model
derives not from the extent to which it mimics reality (because
that cannot be known in advance), but rather from the extent
to which it facilitates the formulation and subsequent testing
of hypotheses that lead to an improved understanding of that
reality (Krag 1996).
H.E. Gruber, PhD (*) In the search for a reliable and economical species, the
Department of Orthopaedic Surgery, authors reviewed the value of possible animal models that
Cannon Research Center, Room 304, Carolinas Medical Center, can be utilized for studies of disc degeneration. From those
32861, Charlotte, NC 28232, USA
discussions, the laboratory has chosen to utilize the sand rat
e-mail: helen.gruber@carolinashealthcare.org
(Psammomys obesus obesus). This desert mammal, indige-
E.N. Hanley Jr., MD
nous to land bounded by the eastern Mediterranean, is the
Department of Orthopaedic Surgery, Carolinas Medical Center,
1025 Morehead Medical Drive, Suite 300, Charlotte, NC, USA only small animal model with spontaneous, age-related disc
e-mail: edward.hanley@carolinashealthcare.org degeneration.

328 H.E. Gruber and E.N. Hanley Jr.
20.1.2 Animal Models Using the Stab Technique 20.3 Newer Findings on Sand Rat Disc
to Induce Degeneration Are Not Optimum Degeneration
The goal of this chapter is to review past literature and sum- 20.3.1 Morphologic Changes with Aging
marize recent findings using the sand rat model. There are a
number of reasons why this is an extremely important and Midsagittal histologic studies have documented the features of
valuable model for disc degeneration. First, as noted above, disc degeneration in older animals which include irregular
these animals undergo spontaneous, age-related disc degen- disc margins, disc space narrowing, wedging, and end plate
eration. Degeneration in the commonly employed rabbit calcification (Gruber et al. 2005). Morphologic changes
models must be induced by a scalpel blade stab. Sobajima beginning in younger animals have received less attention,
et al. (2005) have noted that the classic stab model causes however. As previously recognized with radiologic analyses,
immediate herniation of the nucleus pulposus, and thus it lower lumbar vertebrae show the earliest signs of degeneration
cannot mimic slower changes in the nucleus or annulus asso- (Gruber et al. 2002a). Figure 20.1 illustrates the striking mor-
ciated with onset and progression of degenerative disc dis- phologic changes in the nucleus pulposus with aging. At age 3
ease. It is also not appropriate as a model to test therapies weeks (Fig. 20.1a), the notochordal syncytium makes up the
which focus upon the early stages of degeneration. Second, nucleus pulposus. Notochordal cells can still be prominent up
as the sand rat is a small rodent, its use avoids the high costs to 23 months of age in upper lumbar discs. Notochordal cells
of larger animal models such as the canine or even the rabbit. are shown in high magnification in Fig. 20.1b. In lower discs
Finally, modern custom diets, such as that used in the at 3 months, notochordal cells begin to undergo fragmentation
authors laboratory, can be designed with low-calorie formu- and cell death (Fig. 20.1c); by 11.7 months the demarcation
lations (see below), thus avoiding the onset of nutrition- between annulus and nucleus becomes less distinct, and it
related type 2 diabetes associated with ad lib feeding of appears that annulus cells begin to occupy the proteoglycan-
high-caloric diets as evidenced in the diabetes-prone strain rich nucleus pulposus (Fig. 20.1d, e). By 17 months, the
of the sand rat. nucleus shows marked degenerative changes. Nuclear mate-
rial can be found projecting into regions of herniation in the
dorsal margin of the disc by 1836 months of age (Fig. 20.1g).
Often, these nucleus matrix sites become sequestered
20.2 Review of the Early Literature on Disc (Fig. 20.1h) and can be quite large (Fig. 20.1i) extending dra-
Degeneration in the Sand Rat matically towards the spinal cord (Fig. 20.1j).
Analyses of live and dead cells in the aging sand rat have
The sand rat (Psammomys obesus) is a very attractive model also been carried out using fluorescent markers (Gruber et al.
which avoids problems linked with chemonucleolysis or sur- 2008a). This work utilized animals aged 26, 1319, and
gical injury. In the older literature, spine studies on the sand 2638 months of age. The percentages of dead cells for the
rat were primarily derived from one research group, their entire annulus were 46.1, 48.1, and 76.8 %, respectively, for
subsequent collaborators (Adler et al. 1983; Moskowitz et al. the three age groups studied. The percentage of dead cells
1990; Silberberg et al. 1979, 1989; Silberberg 1988a, b; correlated significantly with end plate bone mineral density
Silberberg and Adler 1983; Ziran et al. 1994; Ziv et al. 1992), (p < 0.02). (Bone mineral density studies are discussed in
and notations in later reviews (Krag 1996; Matsuzaki and greater detail below.) This study also confirmed the progres-
Wakabayashi 1999). Published studies focused primarily on sive death of notochordal cells in the nucleus during aging as
histologic and radiologic changes in animals aged 2, 3, 6, previously noted in morphologic studies.
and 1830 months of age and documented the presence of Sand rat disc tissue has also been extremely valuable
granular debris in the nucleus, small annular tears, hernia- when used in parallel with human disc tissue for specialized
tions into and through cartilage end plates, and ligament studies and for immunohistochemical investigations. Like
calcification. human disc tissue (Gruber and Hanley 1998), sand rat discs
Fig. 20.1 Representative light microscopic images of the aging sand and 17-month-old (f) animals show progressive degeneration of matrix
rat lumbar disc: (a) Lumbar disc at 3 weeks of age: image shows nucleus and cells (scale on E = 50 mm and on F = 200 mm). (gj) Low
pulposus filled with notochordal cells, annulus, and end plates magnification images of lumbar discs showing herniations in the dorsal
(scale = 200 mm). (b) Higher-resolution view of the syncytial network portions of lumbar discs. Note the greatly narrowed nucleus regions. (g)
of the notochord (scale = 20 mm). (c) The nucleus pulposus in a 3-month- 18-month-old animal (scale = 200 mm). (h) 28-month-old animal with
old animal. Note that the notochordal cells have begun to fragment regions of sequestered nucleus pulposus matrix isolated in the hernia-
(scale = 50 mm). (d) Annulus-nucleus transition zone lacks clear tion region (scale = 200 mm). (i) 30-month-old animal. (scale = 500 mm).
definition in this specimen from an 11.7-month-old animal (j) 36-month-old animal (scale = 500 mm) (SP spinal cord, Masson-
(scale = 50 mm). (e, f) Images of the nucleus from an 11.7-month-old (e) trichrome staining)
20 The Sand Rat ( Psammomys obesus obesus) Model of Spontaneous, Age-Related Intervertebral Disc Degeneration 329
a b d
e
c
f g h
i j
Table 20.1 Analysis of the incidence of radiologic abnormalities (%) by age group
Group 1: 13.9 months Group 2: 4.011.9 months Group 3: 1223.9 months Group 4: 2446 months
(n = 36) (n = 18) (n = 19) (n = 26) P valuea
Narrowing
L12 4.0 5.8 38.8 34.6 0.01
L23 29.4 17.6 55.5 42.3 NS (0.08)
L34 29.4 35.2 55.5 88.4 <0.001
L45 35.2 47.0 61.1 88.4 0.0005
L56 38.2 41.1 66.6 96.1 <0.001
L67 50.0 94.1 88.8 100.0 <0.001
L7S 44.1 82.3 88.8 100.0 <0.001
Wedging
L12 11.7 5.8 33.3 26.9 NS (0.08)
L23 14.7 29.4 55.5 46.1 0.01
L34 29.4 41.1 50.0 65.3 0.04
L45 55.8 70.5 72.2 88.4 NS (0.056)
L56 70.5 70.5 94.4 100.0 0.005
L67 64.7 76.4 83.3 92.3 NS (0.075)
L7S 82.3 70.5 88.8 100.0 0.04
End plate calcification
L12 2.9 100.0 44.4 100.0 <0.001
L23 2.9 100.0 61.1 96.1 <0.001
L34 2.9 29.4 83.3 96.1 <0.001
L45 8.8 58.8 88.8 100.0 <0.001
L56 26.4 70.5 100.0 100.0 <0.001
L67 47.0 94.1 100.0 100.0 <0.001
L7S 47.0 82.3 100.0 100.0 <0.001
a
Analysis by chi-square
show the presence of programmed cell death (apoptosis) and performed (Gruber et al. 2002a, 2007a) (Table 20.1). Cross-
cell senescence (Gruber et al. 2007b). Immunolocalization sectional studies showed significant age-related changes
studies in the sand rat disc have shown patterns similar to comprising irregular disc margins, disc wedging, disc nar-
those present in human discs for the presence of thrombos- rowing, end plate and ligament calcification, and osteophyte
pondin (Gruber et al. 2006b), myocilin (Gruber et al. 2006a), formation. Figure 20.2 illustrates radiologic changes at 12,
and pregnancy-associated plasma protein-A (PAPP-A) which 29, and 30 months of age, including wedging (Fig. 20.2a,
unmasks insulin-like growth factor binding protein-4 in the arrow), osteophyte formation (Fig. 20.2b), and bone bridges
extracellular matrix (Gruber et al. 2008b). The sand rat disc between osteophytes (Fig, 20.2c). Males were more likely to
shows a pattern of localization of brain-derived neurotrophic develop osteophytes than were females. There was no gender
factor (BDNF) similar to that seen in the human disc (Gruber difference for other radiographic features.
et al. 2008c). Asporin, a member of the family of small leu- In the prospective study, 22 sand rats were followed with
cine-rich proteoglycan (SLRP) family which has been monthly X-rays from age 2 to 12 months; 11 were males and
reported to have a genetic association with osteoarthritis, is 11 were females (Gruber et al. 2002a). Statistical analysis
present in both the human and sand rat disc (Gruber et al. showed that wedging, narrowing, end plate calcification, and
2009b). Periostin, a matricellular protein in the fasciclin irregular disc margins were all significantly more common at
family expressed in tissues subjected to constant mechanical 12 months of age than at 2 months of age (p = 0.0001). Tests
stress, has also been shown to be present within both the sand for differences in radiologic disc features related to gender
rat and human disc (Gruber et al. 2011a). were performed at 2, 3, 6, and 12 months. Males showed a
statistically greater incidence (%) of wedging at 6 months
(6.4 1.6 vs. 4.8 1.1, p = 0.024) and at 12 months (7.8 0.6
20.3.2 Radiology and Micro-CT vs. 6.3 1.3, p = 0.004) than females. Narrowing was more
frequent in males at age 2 months (1.5 1.6 vs. 0.3 0.5,
20.3.2.1 Lumbar Spine p = 0.028) and age 6 months (5.8 1.8 vs. 4.1 1.6, p = 0.025)
The sand rat spine contains seven lumbar discs (Table 20.1). than in females. End plate calcification was more common in
Radiologic characterizations of degenerating discs in both 2-month-old males than females (1.8 1.5 vs. 1.2 1.7,
cross-sectional and prospective groups of animals have been p = 0.028). Females, however, showed a greater incidence of
Fig. 20.2 (ac) Representative

radiographic images of the
a
lumbar spine showing wedging
(arrow, a), osteophyte formation
(arrow, b), and bony bridging
across osteophytes (arrow, c).
(a) 12-month-old animal;
(b) 29-month-old animal;
(c) 30-month-old animal.
(d) Micro-CT models with cuts
revealing interior features of the b
end plate in 2-, 8-, and 23-month-
old animals. Note the progressive
loss of porosity in end plates.
(e) High-resolution spin-lock
(T1r) image of the lumbar spine
representative of degenerative
changes in older animals. Arrows
mark disc space narrowing and
loss of proteoglycan content
c
(Image courtesy of Dr. Ravinder
Regatte, Depts. of Orthopaedic
Surgery and Radiology, New
York University School of
Medicine)
2 months 8 months 23 months
end plate calcification at 6 months of age than did males In humans, and in the sand rat, advanced disc degenera-
(5.5 1.2 vs. 3.9 0.2, p = 0.009). No gender differences tive changes and end plate calcification (sclerosis) have been
were found for irregular disc margin incidence at 2, 3, 6, or documented. Most notably, Modic et al. have contributed the
12 months of age. major study of end plates in degenerating human discs
Lumbar spine radiographs showing degenerative features (Modic et al. 1988). As noted above, a positive correlation
in the sand rat have been utilized for development of an auto- was observed between the percentages of dead cells in the
mated computer-assisted procedure that analyzed digitized aging sand rat disc and end plate bone mineral density.
X-rays (Wilson et al. 2003). Techniques were developed to Micro-CT model formulation provided an additional
automate the quantitative processing of vertebral radiographic novel, nondestructive technique to assess the end plate-disc
images that may be applicable to modeling of human disc interface and the vascular canal network within the end plate
degeneration. Microcomputerized tomography (CT) has been of the sand rat spine (Gruber et al. 2005). This technique
applied to the development of three-dimensional (3D) mod- revealed a solid bony surface to the end plate that was not
els of vertebral bone and the disc space (Gruber et al. 2005). penetrated by vasculature. Models constructed from
Fig. 20.3 Representative

radiologic images of cervical
spine features in a 2-month-old
and a 22.7-month-old sand rat.
An osteophyte marked in the
older animal is shown in
histologic section in the insert
(Masson-trichrome stain; image
scale = 500 mm)
2.1 months 22.7 months

of age of age
specimens from older animals showed roughening and pit- 20.3.3 Spin-Lock (T1r) Imaging
ting of the end plate surface and the development of irregular
margins. Figure 20.2d illustrates end plate models examined Studies by Regatte et al. have utilized spin-lock (T1r) imag-
in sagittal section to make porosity of the end plate visible. ing to study the sand rat model of disc degeneration
Note the progressive decrease in canals within the end plate (Regatte et al. 2004; also see Chap. 12). High-spatial-
from age 2 months to 6 and 23 months of age. resolution images documented disc changes in both sagittal
and cross-sectional planes of the disc. Figure 20.2e illustrates
20.3.2.2 Cervical Spine a sagittal plane image that shows prominent disc space nar-
Data are now available on sand rat cervical disc degeneration rowing (arrows) in a 45-month-old animal. These investiga-
(Gruber et al. 2011b). Recent findings show that the same tors also quantified spatial variation in the T1r relaxation
major features seen in human disc degeneration and sand rat times in annulus and nucleus regions. These values were in
lumbar spine disc degeneration are also present in the aging the range of 100120 ms and 5060 ms, respectively, at
sand rat cervical discs. Scoring of digital X-rays of cervical 500 Hz. Proteoglycan imaging was excellent based on
and lumbar sites in the same animals showed that in younger exchange of protons on glycosaminoglycans with bulk
animals, cervical discs exhibited a significantly greater water.
proportion of irregular margins compared to lumbar sites
(96.5 % vs. 86.2 %, respectively, p = 0.001). In older ani-
mals, cervical discs also showed a significantly greater pro- 20.3.4 Disc Cell Death and End Plate Bone
portion of osteophytes (4.9 % vs. 0.7 %, respectively, Mineral Density (BMD)
p < 0.0001) compared to lumbar sites. Since animal models
for cervical disc degeneration are rare, the sand rat fills an As noted above, the degenerating disc exhibits increasing
important need and provides spontaneous, age-related degen- end plate calcification. The bone mineral density (BMD) of
eration in both lumbar and cervical spines. Figure 20.3 illus- cranial and caudal end plates of the sand rat lumbar verte-
trates cervical radiologic features in a 2.1-month-old animal brae was significantly greater in older animals (Table 20.2),
and compares that with degenerative changes present in a and caudal end plates usually had significantly greater BMD
22.7-month-old animal. In the older animal, the inset high- values than did cranial plates (Gruber et al. 2008a). These
lights the histologic features of the osteophyte marked in the same specimens were used to obtain viability assays for
digital radiograph. live/dead cells within the annulus. The percentage of dead
Table 20.2 Analysis of the averaged end plate BMD (g/cm2) per level in the age groups
Group 1: 13.9 months Group 2: 4.011.9 months Group 3: 1223.9 months Group 4: 2446 months
Level (n = 36) (n = 18) (n = 19) (n = 26) P valuea
L12 0.064 0.013 0.102 0.016 0.113 0.014 0.128 0.022 <0.001
L23 0.068 0.015 0.101 0.020 0.118 0.016 0.134 0.023 <0.001
L34 0.084 0.016 0.120 0.019 0.133 0.019 0.153 0.028 <0.001
L45 0.094 0.018 0.138 0.022 0.146 0.023 0.177 0.027 <0.001
L56 0.100 0.018 0.145 0.026 0.159 0.026 0.197 0.031 <0.001
L67 0.108 0.019 0.157 0.028 0.167 0.032 0.204 0.036 <0.001
L7S 0.121 0.024 0.180 0.029 0.199 0.033 0.214 0.030 <0.001
a
Data are expressed as mean SD. Analysis by ANOVA for each disc site for the four groups. Note: P values were <0.001 for each disc site when
analyses were also run on either cranial or caudal BMD data separately. ANOVA analysis for all sites comparing the four age groups was also
performed. With this analysis, Group 1 was differed significantly from Groups 2, 3, and 4, and Group 3 vs. 4 was also significant (p = 0.05)
cells in the annulus correlated in a significant, positive man- outer annulus of lumbar discs was surgically removed, the
ner with end plate BMD (p = 0.02, r = 0.347). Vertebral end animals allowed to recover, and then allowed to age 126
plate sclerosis and disc cell viability are probably associ- months postsurgery. Analysis of adjacent segment changes
ated because diffusion of nutrients through the end plate was not able to identify significant differences in cranial vs.
and disc margin provides the only source of nutrition for the caudal disc changes based on age at surgery or time of har-
adult avascular disc. It is believed that end plate calcification, vest postsurgery.
followed by its replacement by bone, impedes nutrient flow
into the disc (Bernick et al. 1991; Bernick and Caillet 1982).
In these experiments the sand rat model provided excellent 20.5 Autologous Disc Cell Implantation
information directly relevant to human disc degeneration. and Annulus Fibrosus Cell Culture
With other researchers, the Gruber/Hanley laboratory

20.3.5 Vascular Supply to the Disc shares a great enthusiasm for the potential of biologic ther-
apies for disc degeneration, especially cell-based therapies
In order to more carefully and accurately study the relation- (see Chaps. 6, 23, and 27 and An et al. 2003; Anderson
ship between vascular supply through the end plate and disc et al. 2005; Fassett et al. 2009; Ganey and Meisel 2002;
degeneration, studies have been performed using a fluorescent Gruber and Hanley 2003; Phillips et al. 2003). The sand rat
vascular tracer in vivo with subsequent visualization of the has been shown to be an excellent model for proof-of-con-
tracer with UV microscopy (Gruber et al. 2005). No penetra- cept studies for cell-based autologous disc cell therapy
tion of the vascular tracer was seen into the disc; this was (Gruber et al. 2002b). For this purpose, cells were har-
verified by immunocytochemical staining for blood vessels. vested from a healthy disc, expanded in culture, and
Only small vasculature features were identified on the dorsal implanted into a degenerating disc in the older donor ani-
and ventral margins of the annulus. mal. Immunohistochemical identification of the implanted
cells showed that they integrated well into the normal
matrix at time points up to 8 months post-engraftment.
20.4 Spine Fusion Model Although this work was technically challenging, the sand
rat proved to be a valuable model for these autologous disc
The potential for long-term adjacent segment degeneration cell studies.
at a fused clinical site remains a complex topic of great Sand rat annulus cells have been cultured in both mono-
interest. An economical small animal model for lumbar layer and in three-dimensional agarose culture (see
fusion would be a useful research tool in investigations of Box 20.1). Annulus cells also attached well and proliferated
lumbar fusion and degenerative changes in adjacent discs or within collagen three-dimensional sponges and after 10
facet joints caused by mechanical loading after fusion. A days of culture produced extracellular matrix containing
recent study has shown the relevance of the sand rat model collagen I and II, keratin sulfate, and chondroitin sulfate
for this need (Gruber et al. 2009a). A small segment of the (Gruber et al. 2002b).
Box 20.1 Culture of Cells from the Sand Rat Annulus or 5-bromo-2-deoxyuridine (BrdU)), rinsed, trypsinized,
The sand rat has been utilized to demonstrate the feasibil- and carefully placed within a 2-mm3-3D carrier (Gelfoam,
ity of cell-based autologous disc cell therapy (Gruber Pfizer) for implantation in the donor animal. On average,
et al. 2002b). Figure (a) presents a view of a surgical field 10,00021,000 cells can be loaded into Gelfoam for
during opening of a lumbar disc for gentle curetting and implantation. In monolayer, sand rat annulus cells are
sterile removal of annulus fragments for culture. Annulus spindle-shaped (illustration B) but assume a rounded
fragments are then minced and cultured in 35 % Dulbeccos morphology when cultured in 3D in agarose (illustration
Modified Eagles Medium, 35 % Hams/F-12 nutrient C). In agarose culture, cells proliferate more slowly than
mixture, and 30 % fetal bovine serum. Fragments are in monolayer; over 10 days of culture in agarose, ~40 %
anchored in 24-well tissue culture plates using sterile of seeded cells form multicellular colonies. It is important
SpectraMesh (104 mm mesh opening, Spectrum to note that in our experience cultures derived from
Laboratories Inc.) trimmed to fit the well. Cells are fed younger animals (aged 23.9 months) yielded disc tissue
every 48 h and cultured until confluent (~68 weeks with successful long-term annulus cultures 60 % of the
of culture). Next, cells can be labeled for immuno- time. Old animals, aged 810 months, yielded successful
histochemical identification postimplantation (with long-term cultures ~20 % of the time (Figs. b and c, origi-
carboxyfluorescein diacetate succinimidyl ester (CFSE) nal magnification 200)
b
a c
Box 20.2 Advantages and Disadvantages of the Sand Dehydration/loss of recognizable nucleus pulpo-
Model for Disc Research sus in lower lumbar sites with progressive
Advantages of the sand rat model: degeneration.
Only natural model with spontaneous, age-related MRI examination shows loss of hydration features
disc degeneration. present in the degenerating human lumbar and cer-
Economically advantageous small animal model vical spine.
that is available commercially via Harlan
Laboratories, Inc.
Note that Harlan requires investigators to sign an
agreement that purchased animals and their
descendants be used solely for research purposes 20.6 Utility and Challenges
at that institution and not be further distributed to of the Sand Rat Model
any third party or utilized for development of
breeding colonies outside that institution. In addition to studies of age-induced spontaneous disc degen-
Radiologic and histologic changes mimic those in eration in the sand rat, this animal model is valuable for other
the aging/degenerating human lumbar and cervical types of research. Many laboratories utilize the sand rat
spines (see below). specified as diabetes prone in nutritionally induced type 2
Radiologic and histologic changes are reliable and diabetes research [see (Collier et al. 2002) for a review] and
well characterized in cross-sectional and prospec- for research on cataract formation (Chenault et al. 2002;
tive analyses. Pollack et al. 1999).
Small size advantageous for viewing the entire lum-
bar or cervical spine in a single radiologic or histo-
logic field.
20.6.1 A Nontraditional Species
Shortcomings of the sand rat model:
A nontraditional species with few colonies now in
Also called the fat or obese sand rat, desert sand rat, or the
research use.
diurnal sand rat, the common nomenclature is misleading
More costly than other rodent models (such as rats
because these animals are of the genus Psammomys, a ratlike
and mice).
gerbil with a heavy body type and a fully haired and tufted
Colony requires specialized low-caloric diet.
tail, small ears, and short hindfoot (Harrison 1964).
Colony requires separate housing with controlled
Figure 20.4a illustrates an adult female and offspring. The
access.
sand rat is found in Algeria, Libya, Egypt, Sudan, Israel, and
Small litter sizes (commonly four pups/litter).
Saudi Arabia (Strasser 1968).
Genome not fully sequenced.
Sand rats generally have a pleasant disposition and the
Small animal size may make disc surgical proce-
animal caretakers enjoy working with these animals. Animals
dures challenging.
in the facility are individually housed with the exception of
Parallels with human lumbar and cervical disc
old mated pairs and same-sex littermates which have been
degeneration:
housed together since weaning. All adult animals should
Age-related disc degeneration.
have implantable micro identification chips to assure correct
Degeneration of both cervical and lumbar levels.
individual animal identification.
Exhibits radiologic features present in the degener-
ating human lumbar and cervical spine, including:
Disc space narrowing, wedging, end plate 20.6.1.1 Need for Separate Housing
calcification, and irregular disc margins. with Restricted Access
Exhibits histologic disc features present in the There is the need to maintain restricted personnel access to
degenerating human lumbar and cervical spine, nontraditional species such as the sand rat. A dedicated hous-
including: ing room should be maintained and special animal handling
Development of disc cell death, disc cell senes- procedures employed. Animal caretakers usually work in
cence, and disc cell apoptosis. this room as their final daily task and may wear protective
disposable lab coats and booties.
Fig. 20.4 (a) Dam (right) and

offspring (left). Note the small a
ears and furred tail, with slight
tufting at the tip.
(b) Representative normal body
weights for male and female
animals
b 330
Male
280
230 Male
Weight (g)
Female
180
130
80
30
4-23-98
5-7-98
5-21-98
6-4-98
6-18-98
7-2-98
7-16-98
7-30-98
8-13-98
8-27-98
9-10-98
9-24-98
10-8-98
10-22-98
11-5-98
11-19-98
12-3-98
12-17-98
12-30-98
1-14-99
1-28-99
2-11-99
2-24-99
3-11-99
3-25-99
Table 20.3 Age and body weight demographics

Group Mean age SD (n) in months Mean body weight SD (n) in grams Gender distributiona
Group 1: 13.9 months 2.13 0.50 (36) 137.86 23.41 (36) 15 F, 21 M
Group 2: 4.011.9 months 6.92 1.57 (18) 189.72 30.22 (18) 5 F, 13 M
Group 3: 1223.9 months 16.05 4.06 (19) 212.37 53.76 (19) 9 F, 10 M
Group 4: 2446 months 33.54 5.01 (26) 175.04 33.74 (26) 20 F, 6 M
a
F female, M male. Chi-square analysis showed a significant difference in gender distribution between groups, p = 0.0069
20.6.1.2 Colony Monitoring Challenges trimmed and weight monitored for a return to normal. The
Sentinel animals occasionally housed within the sand rat state of other dental issues in older sand rats have been
colony have never shown positive disease serology, but the reported in the literature (Ulmansky et al. 1984).
value of available non-sand-rat-specific serology tests con-
tinues to be an issue of discussion. Due to lack of specificity, 20.6.2.3 Blood Testing
such monitoring may not be accurate. Since the sand rat has a fully haired tail, blood sampling is
accomplished via orbital eye sinus puncture or via the
20.6.1.3 Breeding saphenous vein.
In captivity, the sand rat breeds best in the springtime, pos-
sibly related to airborne pollen which reaches the colony.
Animals are fed ad lib only when they are paired for breed-
ing or when the dam has pups in the cage. Animals can be 20.7 Importance of Animal Models
considered sexually mature by 12 weeks of age. When first Such as the Sand Rat for Disc Research
paired, males and females should be checked for compatibil-
ity. Several articles in the older literature have addressed the The discovery and maintenance of relevant animal models
topic of sand rat breeding (Adler et al. 1976; Frenkel et al. for human disease should remain a high priority among
1972; Strasser 1968). researchers. Although the creation of animal models through
genetic engineering continues to be highly important
(Schulhof 2000), support and maintenance of naturally
20.6.1.4 Litter Size occurring animal models needs to be encouraged. Animal
Litter sizes are small; rarely are there six pups/litter, and four models which closely resemble the progression of human
pups/litter is more common. It is advisable to always have diseases, such as the sand rat model of spontaneous, age-
pairs set up for mating within the colony. related disc degeneration, are an especially important tool in
orthopedic research. In addition, as seen in the use of autolo-
20.6.1.5 Low-Calorie Diet Essential gous disc cell therapy in the sand rat model, they can bridge
All animals should be fed the Purina special formulated the gap between in vitro studies and initial human clinical
sand rat diet (Purina custom sand rat diet #5L09 with trials (An and Friedman 1999; Gruber et al. 2002b).
2.42 kcal/g metabolizable energy). 100120 g (5060 g/ As utilization of the sand rat model increases, knowledge
sand rat) is placed in cages twice a week. This low-caloric will be gained of its molecular biology. This will be aided by
diet serves to minimize development of diabetes (El Aoufi ongoing research into genome sequencing that is being
et al. 2007), and adults, breeders, and young animals do well developed for diabetes research at various centers (IREN
on it due to their ability to ferment materials in their hind cDNA libraries of insulin-producing cells 2011).
gut for nutrition.

20.6.2 Other Topics Related to Animal in the Chapter
Husbandry
The sand rat (Psammomys obesus) exhibits age-related
20.6.2.1 Body Weight Is the Best Health Indicator spontaneous degeneration of the intervertebral disc. As
Experience has shown that body weight is the best index of such it is a very important, economical model which
sand rat health (Fig. 20.4b, Table 20.3). For adult breeders, avoids the use of chemonucleolysis or surgical injury to
if 20 % of initial body weight is lost, this may be seen as a induce disc degeneration.
criterion for euthanization. Body weight should be moni- Radiologic cross-sectional and prospective studies have
tored weekly and water consumption and body fur evalu- characterized disc changes with aging in the sand rat.
ated (animals which are not thriving look scruffy in These changes mimic human disc degeneration, including
appearance). disc wedging, disc space narrowing, end plate calcification,
and osteophyte formation.
20.6.2.2 Teeth Need to Be Clipped Periodically Micro-CT models and spin-lock (T1r) imaging studies
If animals are losing weight, they should be promptly exam- provided additional confirmation of the presence of wedg-
ined for possible overgrowth of incisors which might be ing, disc space narrowing, end plate calcification, and
interfering with chewing. If malocclusion is noted, teeth are osteophyte formation.
Micro-CT models also provide an excellent experimental Frenkel G, Shaham Y, Kraicer PF (1972) Establishment of conditions
approach for the analysis of end plate porosity and vascu- for colony-breeding of the sand-rat Psammomys obesus. Lab Anl
Sci 22:4047
lar canals. Ganey TM, Meisel HJ (2002) A potential role for cell-based therapeu-
Because of the importance of cell-based therapies for disc tics in the treatment of intervertebral disc herniation. Eur Spine J
degeneration, the sand rat has been used to demonstrate 11(suppl 2):S206S214
the practicality of autologous disc cell transplantation. Gruber HE, Hanley EN Jr (1998) Analysis of aging and degeneration of
the human intervertebral disc comparison of surgical specimens
Morphologic studies of the aging sand rat disc show with normal controls. Spine 23:751757
strong parallels with the aging human disc, including the Gruber HE, Hanley EN Jr (2003) Biologic strategies for the therapy of
presence of programmed cell death and cell senescence, intervertebral disc degeneration. Expert Opin Biol Ther 3:
the formation of concentric layers of extracellular matrix 12091214
Gruber HE, Johnson T, Norton HJ, Hanley EN Jr (2002a) The sand rat
around disc cells, and cell clustering. Asporin, pregnancy- model for disc degeneration: radiologic characterization of age-
associated plasma protein-A, and brain-derived neu- related changes. Cross-sectional and prospective analyses. Spine
rotrophic factor are present in both human and sand rat 27:230234
discs. Gruber HE, Johnson TL, Leslie K, Ingram JA, Martin D, Hoelscher G,
Banks D, Phieffer L, Coldham G, Hanley EN Jr (2002b) Autologous
Highlights on animal husbandry issues with the nontradi- intervertebral disc cell implantation: a model using Psammomys
tional species are provided. obesus, the sand rat. Spine 27:16261633
Gruber HE, Ashraf N, Kilburn J, Williams C, Norton HJ, Gordon BE,
Acknowledgments The authors wish to thank the Brooks Back Pain Hanley EN (2005) Vertebral endplate architecture and vasculariza-
Research Endowment for support. We thank Michelle V. Chenault, tion: application of micro-computerized tomography, a vascular
Ph.D., for her assistance in providing animals for the early establishment tracer, and immunocytochemistry in analyses of disc degeneration
of our colony, and Brian Gordon, D.V.M., Kim Mihalko, D.V.M., and in the aging sand rat. Spine 30:25932600
Dr. Chenault for assistance with animal husbandry. Thanks are also due Gruber HE, Ingram JA, Hanley EN Jr (2006a) Cellular immunohis-
to Cliff Williams, Karen D. Fay, and Juanita C. Jolly for colony mainte- tochemical localization of the matricellular protein myocilin in the
nance and to Jane A. Ingram and Natalia Zinchenko for assistance with intervertebral disc. Biotech Histochem 81:119124
morphology and radiology studies. We also thank our previous coau- Gruber HE, Ingram JA, Hanley EN Jr (2006b) Immunolocalization of
thors on past sand rat publications for their many contributions. thrombospondin in the human and sand rat intervertebral disc. Spine
31:25562561
Gruber HE, Gordon B, Williams C, Norton HJ, Hanley EN (2007a)
Vertebral endplate and disc changes in the aging sand rat lumbar
References spine: cross-sectional analyses of a large male and female popula-
tion. Spine 32:25292536
Gruber HE, Ingram JA, Hoelscher G, Norton HJ, Hanley EN (2007b)
Adler JH, Roderig C, Gutman Z (1976) Breeding sand rats (Psammomys
Senescence in cells of the aging and degenerating intervertebral
obesus) with a diabetic predisposition for laboratory investigation.
disc: immunolocalization of senescence associated--galactosidase
Refuah Vet 33:18
in human and sand rat discs. Spine 32:321327
Adler JH, Schoenbaum M, Silberberg R (1983) Early onset of disk
Gruber HE, Gordon B, Norton HJ, Kilburn J, Williams C, Zinchenko N,
degeneration and spondylosis in sand rats (Psammomys obesus).
Heath J, Ingram J, Hanley EN Jr (2008a) Analysis of cell death and
Vet Pathol 20:1322
vertebral endplate bone mineral density in the annulus of the aging
An YH, Friedman RJ (1999) Animal selections in orthopaedic research.
sand rat. Spine J 8:475481
In: An YH, Friedman RJ (eds) Animal models in orthopaedic
Gruber HE, Hoelscher GL, Ingram JA, Hanley EN (2008b)
research, 1st edn. CRC Press LLC, Boca Raton, pp 3957
Immunolocalization and biochemical evidence of pregnancy-asso-
An HS, Thonar EJ, Masuda K (2003) Biological repair of intervertebral
ciated plasma protein-A (PAPP-A) in the intervertebral disc. Spine
disc. Spine 28:S86S92
33:E447E454
Anderson DG, Yoon T, Shen F, Kang J, Phillips F, An H (2005)
Gruber HE, Ingram JA, Hoelscher G, Zinchenko N, Norton HJ, Hanley
Summary statement: biologic repair or regeneration of the interver-
EN Jr (2008c) Brain-derived neurotrophic factor and its receptor in
tebral disc. Spine J 5:304s
the human and the sand rat intervertebral disc. Biomed Cent Arthritis
Bernick S, Caillet R (1982) Vertebral end-plate changes with aging of
Res Ther 10:R82
human vertebrae. Spine 7:97102
Gruber HE, Gordon B, Williams C, Ingram JA, Norton HJ, Hanley EN
Bernick S, Walker JM, Paule WJ (1991) Age changes to the anulus
Jr (2009a) A new small animal model for the study of spine fusion
fibrosus in human intervertebral discs. Spine 16:520524
in the sand rat: pilot studies. Lab Anim 43:272277
Chenault VM, Ediger MN, Ansari RR (2002) In vivo assessment of
Gruber HE, Ingram JA, Hoelscher GL, Zinchenko N, Hanley EN Jr,
diabetic lenses using dynamic light scattering. Diabetes Technol
Sun Y (2009b) Asporin, a susceptibility gene in osteoarthritis, is
Ther 4:651659
expressed at higher levels in the more degenerated human interver-
Collier G, Walder K, De Silva A, Tenne-Brown J, Sanigorski A, Segal
tebral disc. Arthritis Res Ther 11:R47
D, Kantham L, Augert G (2002) New approaches to gene discovery
Gruber HE, Norris RA, Kern MJ, Hoelscher GL, Ingram JA, Zinchenko
with animal models of obesity and diabetes. Ann N Y Acad Sci
N, Hanley EN Jr (2011a) Periostin is expressed by cells of the
967:403413
human and sand rat intervertebral discs. Biotech Histochem
El Aoufi S, Gendre P, Sennoune SR, Rigoard P, Maxixent JM, Griene L
86:199206
(2007) A high calorie diet induces type 2 diabetes in the desert sand
Gruber HE, Phillips R, Zinchenko N, Ingram JA, Norton HJ, Hanley
rat (Psammomys obesus). Cell Mol Biol 53:943953
EN (2011b) Spontaneous, age-related cervical spine degeneration in
Fassett DR, Kurk MF, Vaccaro AR (2009) Biologic solutions for degen-
the sand rat: a new research application for this animal model.
erative disk disease. J Spinal Disord Tech 22:297308
Cervical Spine Research Society, Scottsdale
Harrison DL (1964) The mammals of Arabia, 1st edn. Stechert-Hafner Silberberg R (1988b) The vertebral column of diabetic sand rats
Service Agency, New York, pp 597602 (Psammomys obesus). Expt Cell Biol 56:217220
IREN cDNA libraries of insulin-producing cells (2011) Sand rat islet Silberberg R, Adler JH (1983) Comparison of truncal and caudal lesions
cDNA library. http://www99.mh-hannover.de/institute/clinbio- in the vertebral column of the sand rat (Psammomys obesus). Isr J
chemistry/iren/libraries.htm. Med Sci 19:10641071
Krag M (1996) In: Weinstein JN, Gordon SL (eds) Low back pain: a Silberberg R, Aufdermaur M, Adler JH (1979) Degeneration of the
scientific and clinical overview, 1st edn. American Academy of intervertebral disks and spondylosis in aging sand rats. Arch Pathol
Orthopaedic Surgeons, Rosemont, pp 479492 Lab Med 103:231235
Matsuzaki H, Wakabayashi K (1999) In: An YH, Friedman RJ (eds) Silberberg R, Meier-Ruge W, Odermatt B (1989) Age-related changes
Animal models in orthopaedic research, 1st edn. CRC Press, Boca in fibronectin in annulus fibrosus of the sand rat (Psammomys obe-
Raton, pp 539547 sus). Expl Cell Biol 57:233237
Modic MT, Steinberg PM, Ross JS, Masaryk TJ, Carter JR (1988) Sobajima S, Kompel JF, Kim JS, Wallach CJ, Robertson DD, Vogt MT,
Degenerative disk disease: assessment of changes in vertebral body Kang JD, Gilbertson LG (2005) A slowly progressive and reproduc-
marrow with MR imaging. Radiology 166:193199 ible animal model of intervertebral disc degeneration characterized
Moskowitz RW, Ziv I, Denko CW, Boja B, Jones PK, Adler JH (1990) by MRI, x-ray, and histology. Spine 30:1524
Spondylosis in sand rats: a model of intervertebral disc degeneration Strasser H (1968) A breeding program for spontaneously diabetic
and hyperostosis. J Orthop Res 8:401411 experimental animals: psammomys obesus (sand rat) and acomys
Phillips FM, An H, Kang JD, Boden SD, Weinstein J (2003) Biologic cahirinus (spiny mouse). Lab Anim Care 18:328338
treatment for intervertebral disc degeneration summary statement. Ulmansky M, Ungar H, Adler JH (1984) Dental abnormalities in aging
Spine 28:S99 sand rats (Psammomys obesus). J Oral Pathol 13:366372
Pollack A, Oren P, Stark AH, Eisner Z, Nyska A, Madar Z (1999) Wilson C, Brown D, Najarian K, Hanley EN Jr, Gruber HE (2003)
Cataract development in sand and galactosemic rats fed a natural Computer aided vertebral visualization and analysis: a methodology
tomato extract. J Agric Food Chem 47:51225126 using the sand rat, a small animal model of disc degeneration. BMC
Regatte RR, Akella SVS, Dodge GR, Reddy R (2004) Spin-lock (T1r) Musculoskelet Disord 4:4
MRI of the sand rat: A new approach to study this model of disc Ziran BH, Pineda S, Pokharna H, Esteki A, Mansour JM, Moskowitz
degeneration. 5th Combined Meeting of the Orthopaedic Research RW (1994) Biomechanical, radiologic, and histopathologic correla-
Societies of Canada, USA, Japan and Europe, podium presentation tions in the pathogenesis of experimental intervertebral disc disease.
No. 048 Spine 19:21592163
Schulhof J (2000) Restocking the ark: supporting the discovery, cre- Ziv I, Moskowitz RW, Kraise I, Adler JH, Maroudas A (1992)
ation, and maintenance of new animal models. Lab Anim 29:2529 Physicochemical properties of the aging and diabetic sand rat inter-
Silberberg R (1988a) Histologic and morphometric observations on vertebral disc. J Orthop Res 10:205210
vertebral bone of aging sand rats. Spine 13:202208
Use of Knockout and Transgenic Mouse
Models in Disc Research 21
Laura Mangiavini, Rita Gerard-ORiley,
and Ernestina Schipani

21.1 Introduction .................................................................... 341
21.1.1 Brief Anatomical Description of the Intervertebral 21.1.1 Brief Anatomical Description
Disc and Its Embryonic Origin ........................................ 341 of the Intervertebral Disc
21.1.2 Genetic Factors Involved in Disc Degeneration
in Humans ........................................................................ 342
and Its Embryonic Origin
21.2 Animal Models................................................................ 342 The intervertebral disc is a fibrocartilaginous structure that is
21.2.1 General Considerations .................................................... 342
21.2.2 Genetically Modified Mice .............................................. 342 located between two vertebral bodies; it transmits loads
21.2.3 Studies of Disc Development and Function through the spine and allows bending, flexion, and torsion of
Using Genetically Modified Mice .................................... 345 the column. The intervertebral disc consists of an outer ring
21.3 Summary of Critical Concepts Discussed of fibrous cartilage called the annulus fibrosus and an inner
in the Chapter ................................................................. 349 gelatinous structure, the nucleus pulposus. The nucleus pul-
References ..................................................................................... 349 posus is a gelatinous structure mainly formed by sparse
chondrocyte-like cells embedded in a highly hydrated aggre-
can-containing gel (Raj 2008). For more details of the struc-
ture of the disc, see Chaps. 4 and 5. The annulus fibrosus is
composed of an outer layer of collagen I and an inner region
containing collagen II. The collagen fibers are arranged in
parallel lamellae, and elastin fibers are interposed between
the lamellae.
The annulus fibrosus and the nucleus pulposus follow dif-
ferent developmental pathways. The annulus fibrosus is
derived from the condensation of sclerotome cells. Signals
from the notochord induce migration, condensation, and dif-
ferentiation of sclerotome cells into the annulus fibrosus and
vertebrae (Risbud et al. 2010). Still uncertain is the origin of
the nucleus pulposus, although recent experimental evidence
showed that the nucleus pulposus cells in mice derive from
the notochord (Choi et al. 2008); notochordal cells probably
undergo hypertrophy to form the nucleus pulposus (Aszdi
L. Mangiavini, MD R. Gerard-ORiley, BS et al. 1998; Hunter et al. 2004; Sakai et al. 2009). However,
E. Schipani, MD, PhD (*) the mechanism by which notochordal cells differentiate into
Department of Medicine, nuclei pulposi is not completely understood.
Cancer Center, Indiana University, 980 W. Walnut Street, R3 C102,
Indianapolis, IN 46202, USA
The notochord thus appears to have a double function:
directly forming the nucleus pulposus and indirectly control-
Department of Anatomy and Cell Biology,
Cancer Center, Indiana University,
ling the formation of vertebral bodies and the annulus
980 W. Walnut Street, R3 C102, Indianapolis, IN 46202, USA fibrosus. The development of the disc is described in detail in
e-mail: laumangi@iu.edu; rloriley@iu.edu; eschipan@iu.edu Chap. 3.

342 L. Mangiavini et al.
21.1.2 Genetic Factors Involved in Disc different developmental stages and/or to generate disc degen-
Degeneration in Humans eration. Moreover, husbandry and housing are often
cost-prohibitive (Singh et al. 2005). For further discussion of
Aging leads to numerous changes in the structure of the the pros and cons of large animal models, see Chap. 18.
intervertebral disc such as shape, vascular supply, and matrix On the other hand, smaller animals like rats and mice are
composition including proteoglycan accumulation and water less expensive and easy to handle. Moreover, intervertebral
content. With degeneration and age, the extracellular matrix disc development and subsequent degeneration occur in a
of the nucleus pulposus is progressively reduced. It is likely shorter time period, making the use of these models more
that the decreased aggrecan content in the aging nucleus pul- cost-effective. In addition, a variety of markers and probes
posus impairs its hydration status resulting in unequal distri- are available for murine tissues. Lastly, mouse models are
bution of compressive forces on the spine and, eventually, to ideal for studies of the intervertebral disc because they can
disc herniation. For detailed analysis of this topic, see Chaps. be genetically manipulated. Taken altogether, genetically
2, 7 and 19. engineered mice provide an important new tool to test
Degenerative disc disease (DDD) and disc herniation are hypotheses related to disc function, degeneration mecha-
the most common cause of low back pain in modern society nisms, and nonsurgical treatments for disc degeneration. Of
(Frymoyer 1988; Frymoyer and Cats-Baril 1991; Wisneski course there are limitations for their use: first, anatomical
et al. 1992; Anderson and Weinstein 1996; Komori et al. differences exist between humans and mice; second, because
1996). Curiously, however, their precise etiology is still of costs, young mice are often analyzed and they may not
largely unknown (Sambrook et al. 1999; Ala-Kokko 2002; faithfully reproduce adult human conditions (Alini et al.
Battie et al. 2004). Familial studies have established that 2008). In this chapter, mouse models that have promoted a
disc herniation is influenced, not surprisingly, by genetic critical new understanding of the human intervertebral disc
and familial factors (Varlotta et al. 1991; Battie et al. 1995; will be reviewed.
Matsui et al. 1997; Zhang et al. 2008). In the last decade,
degenerative disc disease in humans has been associated
with different mutations in genes encoding matrix proteins 21.2.2 Genetically Modied Mice
such as collagen I (Pluijm et al. 2004; Tilkeridis et al. 2005),
collagen IX (Annunen et al. 1999; Paassilta et al. 2001), Genetically manipulated mice have contributed enormously
and aggrecan (Kawaguchi et al. 1999; Roughley et al. 2006). to identification of genes controlling intervertebral disc
Moreover, the three master transcription factors of development and to the elucidation of their mechanisms of
chondrogenesis, Sox5, Sox6, and Sox9; cytokines such as action (Table 21.1).
interleukin-1 (Solovieva et al. 2004) and interleukin-6 A genetically modified mouse is a mouse whose genome
(Noponen-Hietala et al. 2005); and, lastly, the vitamin D has been altered through the use of genetic engineering tech-
receptor (Videman et al. 2001) have been also involved in niques. Different genetic techniques are available to examine
the pathogenesis of DDD. The role of each of these genes is the effects of mutant gene products and their roles in the
described in detail in Chap. 10. intervertebral disc. In the next section, we will briefly dis-
cuss some of them.
21.2 Animal Models 21.2.2.1 Transgenic and Global Knockout

Approaches
21.2.1 General Considerations There are two basic genetic technical approaches to produce
either transgenic mice or gene-targeted mutant mice, respec-
In recent years, different animal models have been character- tively. In the transgenic approach, new genetic information is
ized in order to reach a better understanding of the develop- inserted into the mouse genome via injection of a full-length
ment and function of the intervertebral disc in humans. coding sequence (cDNA) of the gene of interest cloned
Generally speaking, a series of variables need to be consid- downstream of a specific promoter sequence into the pronu-
ered when choosing an adequate animal model to study the clei of fertilized mouse oocytes, with subsequent random
nucleus pulposus, such as availability, size, and cost, but integration of this cDNA sequence into the host genome.
knowledge of the similarities and differences in biomechani- A major limitation of the transgenic approach is that overex-
cal and biochemical properties between the model and the pression models often produce nonphysiological levels of
human discs is of utmost importance. Larger animals are the protein of interest, and this can confound interpretation
good models as they recapitulate the human disc with respect of the physiological role of the gene. Also, the site of trans-
to biomechanics, geometry, and structure. However, large gene integration can have consequences on tissue specificity
animals require long time frames in order to analyze the and levels of expression of the transgene. Moreover, even
21 Use of Knockout and Transgenic Mouse Models in Disc Research 343
Table 21.1 Knockout and transgenic mouse models in the intervertebral disc
Gene of interest Knockout Annulus fibrosus Nucleus pulposus References
GDF-5 Universal Abnormal Abnormal Li et al. (2004)
Col2a1 Universal Abnormal Normal Sahlman et al. (2001)
Col1a1 Universal Abnormal Normal Sarver and Elliott (2004)
Col9a1 Universal Abnormal Normal Boyd et al. (2008)
Allen et al. (2009)
Biglycan Universal Abnormal Abnormal Furukawa et al. (2009)
Danforths short tail Universal Abnormal Absent Lane and Birkenmeiser
(Sd) (1993), Alfred et al. (1997)
Sickle tail (Skt) Universal Abnormal Abnormal Semba et al. (2006)
Sox5 and Sox6 Universal Deficiency of the inner annulus Absent Smits and Lefebvre (2003)
Pax-1 Universal Abnormal Abnormal Wallin et al. (1994)
Pax-9 Universal Abnormal Abnormal Peters et al. (1998)
Has2 Conditional Abnormal Abnormal Roughley et al. (2011)
Knockout in cartilage
c-Jun Conditional Normal Abnormal Behrens et al. (2003)
Knockout in axial skeleton,
sclerotome, notochord
Ext1 Conditional Abnormal Abnormal Mundy et al. (2011)
Knockout in the developing joints
Smoothened Conditional Normal Abnormal Choi and Harfe (2011)
Knockout in the notochord
Tgfbr2 Conditional Deficiency of the inner annulus Normal Jin et al. (2011)
Knockout in growth plate
chondrocytes and inner annulus
fibrosus cells in the postnatal stage
Wnt/b-catenin Conditional Abnormal Abnormality due Kondo et al. (2011)
Knockout in axial skeleton to degeneration
of the growth
plate
well-characterized promoters can be expressed at low levels triple knockouts may be necessary. Another consideration is
in nontarget tissues, so rigorous analysis should include that global knockouts usually contain a modified allele in
examination of expression of the transgene in a large range which the selectable cassette used to screen the ES colonies
of tissues. is often retained in the locus of interest. In this case, this
The second approach, homologous recombination, could have effects on neighboring genes.
involves modifying a specific gene in embryonic stem cells If the mutated gene carries loss-of-function or gain-of-
with a DNA construct containing DNA sequences homolo- function mutations, a global knock-in for that targeted gene
gous to the target gene. Embryonic stem cells carrying the is then generated.
recombined genomic DNA are selected and are then injected Instead of using a targeted strategy, global knockouts
into mouse blastocysts. As a result, mutant genes with either can be generated using a high-throughput mutagenesis strat-
loss-of-function or gain-of-function mutations can be gener- egy. One of the most widely used strategies involves the pro-
ated (Fig. 21.1). duction of random insertional mutations in ES cells using
If the mutated gene is a null allele, a global knockout for vectors that contain a promoterless reporter gene.
that targeted gene is then generated. Global knockouts pro-
vide direct insight into the physiological role of the ablated 21.2.2.2 Tissue-Specic Knockout Models
gene product. Moreover, novel actions of the targeted genes Conventional gene targeting generates a modified allele in all
can emerge because, unlike transgenic models, global knock- cells of the mouse from fertilization on; therefore, it is an
out models are not limited to a particular tissue or system. extremely useful tool for investigating gene function during
A disadvantage of the global knockout approach is that dele- development and adulthood. However, if the inactivation of
tion of genes that are essential for early development may the target gene results in early embryonic lethality, the func-
result in early lethality. On the other hand, because of functions of the gene in specific tissues cannot be studied. Further,
tional redundancy, many knockout strains do not exhibit an universal gene targeting can make it difficult to distinguish
obvious phenotype. In this case, creation of double or even direct effects of ablating a gene in a particular tissue from the
Endogenous gene
Targeting vector
3. Targeting vector introduced
into ES cells by electroporation
Positive-negative selection
1. Mouse blastocyst 2. ES cells 4. Rare cells carrying recombined 5. Pure population of ES cells
gene carrying recombined
gene
7. Blastocysts implanted into foster

mothers
6. Mutant ES cells injected into

8. Chimeric mice blastocysts
9. Heterozygous mutant 9. Wild type mouse

mouse
Fig. 21.1 Schematic representation of gene targeting in mice. carrying the recombined gene are selected (5). The targeted ES are
Embryonic stem cells (ES) are collected from mouse blastocysts (1). ES injected into blastocysts (6). The blastocysts are implanted into foster
are cultivated in vitro (2) and a targeting vector is introduced by elec- mothers (7) and they give birth to chimeric mice (8). The breeding
troporation. The targeting vector contains fragments of DNA that are between chimeric mice produces mice heterozygous for the targeted
homologous to the endogenous gene (3). Homologous recombination gene and wild-type mice (9)
occurs between the targeting vector and the endogenous gene (4). ES
more indirect effects of ablating the gene in all tissues. The conditional gene targeting can also be used to investigate
Cre recombinase-loxP (Cre-loxP) system was developed to gene function at late embryonic stages or in adulthood.
overcome these limitations and to inactivate genes in a con- To date, the Cre-loxP system is the best-characterized
ditional manner in the living mouse as well (Orban et al. system for conditional gene inactivation in mice (Wilson and
1992; Sauer 1998). Conditional gene targeting refers to a Kola 2001; Le and Sauer 2001). The Cre-loxP system is
gene modification in the mouse that is restricted either to comprised of two elements: the Cre recombinase enzyme
certain cell types (tissue specific), to a specific developmen- and a small stretch of DNA recognized by the recombinase
tal stage (temporally specific), or to both (Lewandoski (loxP site) (Stark et al. 1992; Van Duyne 2001). The Cre
2001). recombinase is produced by the bacteriophage P1 and is a
The regional and temporal specificity provided by condi- member of the integrase superfamily of site-specific recom-
tional gene targeting allows a better analysis of the gene binases that cleave DNA at a distinct target sequence and
function in different ways. The tissue specificity allows the ligate it to the cleaved DNA of a second identical site to gen-
study of a gene in a particular cell lineage without being erate a contiguous strand. The loxP site consists of 34 base
influenced by gene loss in adjacent tissues, as the rest of the pairs, a size that is unlikely to occur randomly in even the
embryo is genetically wild type. Moreover, the temporal largest vertebrate genome and yet small enough to be effec-
specificity does not permit the organism to adapt to the tively neutral toward gene expression when positioned in
genetic change, as the wild-type gene product was previ- chromosomal DNA for genetic manipulations. The orienta-
ously present. Therefore, compensatory responses, which tion of these target sites relative to each other on a segment
can alter interpretation of conventional germ line mutations, of DNA directs the type of modification catalyzed by the
are mitigated, providing a more precise link between geno- recombinase; more specifically, in order to achieve excision
type and phenotype. Lastly, if null mutations lead to a of the intervening DNA, the two loxP sites must be oriented
severe or lethal phenotype during embryonic development, in the same direction.
Fig. 21.2 Cre-loxP strategy. 1

Two separate mouse strains are ATG
Recombinase prom cre pA Recombinase
typically generated and transgene
intercrossed for a conditional
gene targeting experiment. One
Intron
mouse strain expresses the Cre
recombinase in selected tissues
(1); the mate carries a gene 2
Floxed gene loxP loxP
segment flanked (floxed) by the Floxed
loxP sites (2). In offspring (3), gene
cells expressing the recombinase
delete the target gene segment
3
Recombinase
Recombined
gene
Recombined gene
Two separate mouse strains are typically generated and intervertebral discs. This phenotype may be related to aging
intercrossed for a conditional gene targeting experiment. but is mainly attributed to an altered metabolism. In fact, dia-
One mouse strain expresses the Cre recombinase in selected betic sand rats show a decrease in disc hydration, which
tissues, depending on which promoter has been selected to leads to less advantageous biomechanical properties and
drive recombinase expression; the mate carries a gene seg- eventually to degeneration of the disc (Silberberg et al. 1979;
ment flanked (floxed) by the loxP sites (Fig. 21.2). The loca- Ziv et al. 1992).
tion of the loxP sites must be appropriately chosen so that the Another spontaneous disc degeneration mouse model is
function of the gene is not affected and that deletion of the the pintail mouse (Anas acuta). The nuclei pulposi in the
floxed gene segment will lead to inhibition of transcription subcervical region have a low mucopolysaccharide content
and/or translation of the gene of interest or to the synthesis of as in the degenerate human discs. This phenotype is stronger
a nonfunctional protein. In offspring, cells expressing the in homozygotes. This mouse model was the first evidence
recombinase delete the target gene segment, while the target that deterioration of the human disc could indeed have a
gene remains functional in cells of all other tissues where genetic correlate (Berry 1961).
Cre is not expressed (Schipani 2002). Transcription factors of the Pax family play a major role
The ability to achieve site-specific recombination has in intervertebral disc development. Pax-1 is a transcription
revolutionized genetic analysis of skeletal cell function; it is factor critically involved in various aspects of mammalian
important to bear in mind that it may be difficult to find a organogenesis (Chalepakis et al. 1992). It is expressed in the
promoter that drives Cre expression with sufficient activity sclerotome and is implicated in the formation of ventral ver-
to result in complete excision of the target gene. tebral structures. Pax-1 is a critical modulator of the cross-
talk between notochord and sclerotome in early development
(Wallin et al. 1994). For further information on this gene,
21.2.3 Studies of Disc Development and please see Chap. 3.
Function Using Genetically Modied Mice Mice carrying a naturally occurring mutation of the Pax-1
gene demonstrate impaired development of both the vertebral
21.2.3.1 Spontaneous Gene Mutations Causing bodies and intervertebral discs (Wallin et al. 1994). In partic-
Impaired Disc Function and/or Structure ular, both vertebral bodies and intervertebral discs may be
Numerous animal models with spontaneous disc degenera- either absent or severely deformed. The defects are especially
tion have been described. In these animals, the mutation is evident in the lumbar region and in the tail. The malforma-
not artificially induced as the animals develop the pathologi- tions appear at an early embryonic stage and they affect both
cal condition naturally. the sclerotome and the notochord. However, this phenotype
The sand rat and the pintail mouse are the first spontane- most likely involves additional genes, as a targeted null muta-
ous models that have been reported (Singh et al. 2005). As tion of Pax-1 causes a less severe phenotype (see below).
was discussed in Chap. 20, in the sand rat (Psammomys obe- Watanabe et al. (1997) and, later, Wai et al. (1998) reported
sus), cysts and tears in the annulus fibrosus and even hernia- that the cmd (cartilage matrix deficiency) mouse carries an
tion of the nucleus pulposus are evident. These conditions autosomal recessive mutation in the gene encoding aggre-
resemble the pathological aspects of the degenerate human can, which is expressed in both the nucleus pulposus and the
annulus fibrosus. Homozygous mice for this mutation dis- evidence that the disc in Mov13 mutant mice is mechanically
play dwarfism and cleft palate, and they die shortly after inferior to controls when subjected to compression and tor-
birth, whereas modest dwarfism, age-associated hyperlordo- sion tests. This finding suggests that collagen I in the inter-
sis, and disc herniation are typically found in heterozygous vertebral disc is particularly important in absorbing torsional
mice. Aggrecan is thus likely to be an extremely important loads.
molecule in the development and homeostasis of the inter- Another essential matrix protein of the annulus fibrosus is
vertebral disc. collagen IX. This collagen acts as a bridge between the
Li et al. (2004) analyzed instead a spontaneous loss-of- fibrillar collagens and the other components of the matrix;
function mutation of growth differentiation factor-5 (GDF-5) together with collagens II and XI, it forms a heterofibril,
in mice. This growth factor has been shown to play a role in which stabilizes the cartilaginous tissue. Nakata et al. (1993)
a variety of musculoskeletal processes, including joint for- reported that mice in which a gene construct has been inserted
mation, endochondral ossification, and tendon and ligament to generate a truncated a1 chain of collagen IX develop
maintenance and repair (Francis-West et al. 1999). Mutant chondrodysplasia, osteoarthritis, and corneal abnormalities.
mice carrying a spontaneous loss of function of GDF-5 are In particular, homozygous mice exhibit spine deformities
characterized by short limbs, abnormalities of the joints, and characterized by shortening of the vertebrae, matrix
a reduction in the number of phalanges in the second through disorganization within the annulus fibrosus, and end-plate
fifth digits (Merino et al. 1999). More importantly, these irregularities (Kimura et al. 1996).
mice demonstrate abnormalities of both the annulus fibrosus More recently, another study demonstrated that mice
and the nucleus pulposus, resembling changes seen in some homozygous for a loss-of-function mutation of the Col9a1
animal models of disc degeneration. In particular, young gene have early-onset osteoarthritis around 6 months of age,
adult mice displayed decreased water content of the nucleus secondary to degeneration both in the annulus fibrosus and in
pulposus as indicated by MRI analysis. Moreover, the mutant the end plate. Interestingly, in this mouse model, the changes
nucleus pulposus is smaller and the glycosaminoglycan con- in the intervertebral disc are already clearly detectable at 3
tent of the disc is diminished, although the amount of total months of age and thus precede the degeneration of the ver-
collagen is not altered. In addition, the annulus fibrosus in tebral end plate. Notably, disc degeneration occurs in the
these mutant mice is characterized by loss of the lamellar annulus fibrosus and it is mainly characterized by tears of
organization. All of these abnormalities can be corrected by this structure, whereas the nucleus pulposus is virtually unaf-
treatment with recombinant GDF-5. Consistent with these fected (Boyd et al. 2008; Allen et al. 2009).
data, conditional knockout (see subsequent paragraph) of Furukawa et al. (2009) analyzed the function of another
GDF-5 further highlighted the importance of this growth fac- component of the extracellular matrix in the intervertebral
tor in the biology of the nucleus pulposus. disc by knocking out biglycan, which is a member of the fam-
ily of small leucine repeat proteoglycans (SLRPs); SLRPs
21.2.3.2 Global Knockout of Genes Encoding bind to TGF-bs, collagens, and other matrix proteins. In
Extracellular Matrix Proteins humans, biglycan is mostly expressed in the outer layer of the
Sahlman et al. (2001) reported that mice heterozygous for a annulus fibrosus, while its concentration is very low in the
Col2a1 null allele develop abnormalities in the vertebral nucleus pulposus. Advancing age leads to an increase in big-
bodies and in the disc. In particular, their vertebral end plates lycan content in the early stages of degeneration and then,
undergo premature ossification, which is associated with a eventually, to a progressive decrease (Cs-Szabo et al. 2002).
mild grade of degeneration of the disc and with a significantly Previous studies showed that biglycan-deficient mice devel-
reduced ability to run, likely secondary to the discomfort oped premature osteoarthritis (Ameye et al. 2002). In this
caused by the anatomical abnormalities. Of note, collagen II study, the authors provide evidence that the absence of bigly-
is poorly expressed in the nucleus pulposus, though it is a can leads to an early degeneration of the intervertebral disc.
classical marker of the fibrocartilage that forms the annulus Using morphometrical and histological analyses, the authors
fibrosus. Mice homozygous for a null mutation of Col2a1 reported that the size of the nucleus pulposus decreases with
show abnormalities in both vertebral bodies and interverte- age in the mutant mice and that, notably, at 6 months of age,
bral discs: the vertebral bodies enlarge gradually and they mice display an abnormal proliferation of chondrocyte-like
never initiate endochondral ossification; moreover, the noto- cells within the nucleus pulposus. Later, both the nucleus pul-
chord persists, leading to a failure in the development of the posus and the annulus fibrosus in biglycan-deficient mice are
intervertebral disc (Aszdi et al. 1998). characterized by tears associated with mucous degeneration
Along these lines, Sarver and Elliott (2004) created a of the nucleus. Therefore, biglycan is likely to be an impor-
transgenic mouse with reduced collagen I expression (Mov13 tant factor in the maintenance of the intervertebral disc. There
strain). Collagen I is particularly abundant in the matrix of are different mechanisms to explain why loss of biglycan
the outer annulus fibrosus. Mechanical testing provided accelerates disc degeneration: its loss might cause instability
of the extracellular matrix, or it might increase the mechani-

cal stress on the intervertebral disc. The absence of biglycan Box 21.1 Procedures to Phenotype a Mutant Mouse Disc
may also inhibit TGF-b signaling, and this could suppress the 1. Genomic PCR analysis to confirm the genotype
damage repair response in the extracellular matrix. 2. Macroscopic evaluation of the spine
3. Whole mount Alizarin Red S/Alcian Blue stain to
21.2.3.3 Global Knockouts of Genes Affecting visualize cartilage and ossified skeletal elements
Disc Development 4. Harris hematoxylin and eosin stain to analyze the
In recent years, numerous mouse mutants with abnormalities morphology of the cells
of the intervertebral disc have been generated by deleting 5. Alcian Blue stain to identify the presence of
genes of interest by homologous recombination. This knock- glycosaminoglycans
out strategy has allowed researchers to define the role of dif- 6. PAS stain to identify the presence of glycogen
ferent genes in disc development. 7. BrdU immunohistochemistry to quantify cell
For instance, transcription factors important in chondro- proliferation
genesis have also been shown to play a critical role in disc 8. Tunel assay to examine cell death
development and homeostasis. In particular, Smits and 9. Immunohistochemistry to detect the presence of
Lefebvre (2003) have provided evidence that Sox5 and Sox6 specific proteins
are expressed not only in the chondrocyte precursors that 10. In situ hybridization to detect the presence of
form the vertebral bodies and the annulus fibrosus but also in specific mRNAs
the notochord and are critically important for disc develop- 11. Use of reporter mice
ment. Sox5 and Sox6 encode two identical transcription fac- 12. Biomechanical tests (tension, compression, torsion)
tors (L-Sox5 and Sox6); moreover, Sox5 also encodes a short
protein (Sox5) that lacks the N-terminal of L-Sox5. L-Sox5
and Sox6 are co-expressed in all cartilages and in a few other
tissues. In vitro data have shown that these two transcription to single Pax-1 null mice (Peters et al. 1999). This study indi-
factors cooperate with Sox9 in the activation of the collagen cates that Pax-1 can completely compensate for the absence
II gene (Lefebvre et al. 1998). Moreover, Sox5 and Sox6 of Pax-9 during early development of vertebral bodies and
have redundant roles in chondrogenesis, and knockout mice intervertebral discs, whereas Pax-9 has a redundant role in
for these two transcription factors are characterized by a the formation of these structures.
severe chondrodysplasia (Smits et al. 2001). In a more recent Gene-trap mutagenesis in mice has revealed a critical role
work, Smits and Lefebvre analyzed the spine phenotype of for genes located on chromosome 2 in development and
mice lacking Sox5, Sox6, or both through RNA in situ hybrid- homeostasis of the nucleus pulposus. In particular, the so-
ization, cell proliferation, and death assays. First, they called Danforths short tail (Sd) mouse shows a complete
showed that Sox5//Sox6/ mice lack the nucleus pulposus absence of the nucleus along with severe abnormalities of the
and that the annulus fibrosus has a deficient extracellular notochord and vertebral bodies as early as E9.5 that lead to
matrix. At earlier developmental ages, mutant embryos dis- the characteristic short tail (Lane and Birkenmeiser 1993;
play an abnormal development of the notochord secondary Alfred et al. 1997). In another mouse model (the enhancer
to massive cell death. Analysis of the possible different gen- trap line Etl4lacZ, located also on chromosome 2), a reporter
otypes led to the conclusion that Sox5 and Sox6 have redun- (lacZ) is inserted near the Sd locus, and the resulting pheno-
dant functions in notochord development, though Sox6 could type is characterized by kinks in the caudal region of the tail
be slightly more important. (Zachgo et al. 1998). Interestingly, loss-of-function muta-
Taken together, these findings demonstrate that Sox5 and tions in the Sickle tail (Skt) gene, which is located on chro-
Sox6 are not necessary in early notochord formation but in mosome 2 near the Sd locus, also lead to severe abnormalities
the survival of notochordal cells. Moreover, these data are of the notochord and of the nucleus pulposus, resulting in a
consistent with the notion that the nucleus pulposus is derived kinky-tail phenotype in adults (Semba et al. 2006). The
from the notochord (Choi et al. 2008). nucleus pulposus in homozygous mice is degenerated at its
Homozygous Pax-1/ mice display abnormalities in the periphery, and also the annulus fibrosus shows some abnor-
shape of the vertebrae, and the intervertebral discs are malities such as thin fibrous layers and lack of adhesion to
replaced by a ventral cartilaginous rodlike structure (Wilm the vertebral bodies: this phenotype is particularly pro-
et al. 1998). In contrast to Pax-1 mutants, Pax-9/ mutant nounced in the tail, although some degree of abnormality is
mice do not exhibit morphological abnormalities of the axial also present in other regions of the spine. Mutant mice carry-
skeleton (Peters et al. 1998). However, double mutants lacking both mutations, Sd and Skt, display a phenotype that
ing both Pax-1 and Pax-9 have a more severe phenotype in appears to be the cumulative result of phenotypic abnormali-
both vertebral bodies and intervertebral discs when compared ties due to the individual mutations. However, the detailed
functional and genetic relations between the two loci are still and loss-of-function mutations of b-catenin achieved using
largely unknown. Generally speaking, Sd controls an early transgenic mice in which Cre recombinase was driven by
stage of mesoderm development and thus affects both sclero- fragment of Col11a1 or Col2a1 promoters have provided
tome and notochord, while Skt is important at a later stage unequivocal evidence that Wnt/b-catenin signaling supports
and is essentially involved in the growth and hypertrophy of organization of the annulus fibrosus and that this signaling
cells of the nucleus pulposus. pathway is critical for maintenance of intervertebral disc
structures during development (Kondo et al. 2011).
21.2.3.4 Tissue-Specic Knockouts of Genes TGF-b signaling has also been involved in disc formation
Encoding Extracellular Matrix Proteins during embryonic development (Battie et al. 2004). Mice in
Conditional knockouts of matrix proteins, signaling mole- which Tgfb receptor 2 (Tgfbr2) had been conditionally
cules, or transcription factors have identified essential play- knocked out in the axial skeleton using a fragment of the
ers in the biology of the intervertebral disc. Roughley et al. Col2a1 promoter displayed abnormal intervertebral discs.
(2011) conditionally knocked out hyaluronan synthase-2 Along these lines, postnatal deletion of Tgfbr2, achieved
(Has2) in cartilage by using Col2-Cre transgenic mice. Has2 using Col2a1-CreERT2 transgenic mice, led to severe abnor-
is responsible for most of the hyaluronan production in the malities of both the vertebral end plate and the inner annulus
body and its universal knockout is embryonic lethal. In the fibrosus, whereas the nucleus pulposus was virtually intact
intervertebral disc, hyaluronan forms the core of the proteo- (Jin et al. 2011). These findings are consistent with the higher
glycan aggregates that are responsible for the osmotic prop- content of collagen II, and thus likely of Cre recombinase
erties that allow the disc to resist compression. The knockout activity, in the vertebral bodies and in the annulus fibrosus in
of Has2 specifically in cartilage confers a phenotype charac- comparison with the nucleus pulposus. Of course, they do
terized by a shortening in the spine, long bones, ribs, and not exclude the possibility that Tgfbr2 could also play an
snout. Notably, disappearance of the notochord appears to be important role in homeostasis of the nucleus upon expression
delayed. of Cre recombinase specifically in the nucleus pulposus
Mundy et al. (2011) have analyzed the role of heparan itself.
sulfate proteoglycans (HSPGs) in the development of inter- Behrens et al. conditionally deleted the transcription fac-
vertebral discs. HPSGs modulate numerous developmental tor c-Jun in the notochord, sclerotome, and cartilage using a
processes, and they regulate cell-matrix interactions. Mundy Cre recombinase driven by a fragment of the collagen II pro-
et al. conditionally knocked out Ext1, which is responsible moter (Col2a1; JunD/D). In this way, they provided clear evi-
for heparan sulfate synthesis, in developing fibrous/synovial dence that this transcription factor has an essential role in the
joints by using GDF5-Cre transgenic mice. Of note, by using development of the intervertebral disc. The Col2a1; JunD/D
ROSA26 reporter mice, the authors provided evidence that mice exhibited a scoliotic vertebral column and abnormal
expression of the Cre was restricted to the annulus fibrosus morphogenesis of the ribs. Interestingly, the overall length
and did not involve the nucleus pulposus. The mutant mice and morphology of bones in the limbs appear normal in the
die from respiratory failure at birth, and they demonstrate adult mice. Moreover, the absence of c-Jun does not severely
severe abnormalities of the intervertebral discs; some discs affect chondrogenesis, and the skeletal defects are mainly
are missing and the adjacent vertebrae are fused, while the limited to the developing spine. Consistent with the high lev-
remaining discs contain a hypocellular annulus fibrosus. els of expression of c-Jun in the notochord, the mutant mice
These abnormalities could be due, at least in part, to a failure displayed a dramatic reduction in the size of the nucleus pul-
of annulus fibrosus progenitor cells to migrate or to prolifer- posus, but not of the annulus fibrosus. Notably, the reduced
ate. In conclusion, these data indicate that Ext1 and, conse- size of the nucleus pulposus is likely the result of increased
quently, heparan sulfate proteoglycans are necessary for the cell death (Behrens et al. 2003).
formation of the annulus fibrosus. Sonic Hedgehog (Shh) is expressed in the notochord, in
the floorplate, and in the posterior margins of the limb buds
21.2.3.5 Tissue-Specic Knockouts of Genes (Chiang et al. 1996). Embryos (E9.5) homozygous for a Shh
Affecting Disc Development null allele do not form both the vertebral column and the
The conditional knockout of a variety of genes has revealed nucleus pulposus, whereas differentiation of the annulus
an important function for many of these genes in notochord fibrosus is unaffected. Notably, their notochord progressively
and/or nucleus pulposus development. degenerates as demonstrated by the loss of expression of
b-catenin-dependent Wnt signaling is one of the central Brachyury. Two elegant lineage studies achieved by knock-
regulators of endochondral bone development (Tamamura ing in the cDNA encoding Cre recombinase into the Shh
et al. 2005; Enomoto-Iwamoto et al. 2002); but little is known locus (Choi et al. 2008; Maier et al. 2011) have provided the
about the function of this pathway in intervertebral disc first genetic evidence that nucleus pulposus cells are derived
development. A series of conditional in vivo gain-of-function from the notochord. Moreover, deletion of Smoothened,
receptor for Shh, achieved using a conditional knockout in Alfred JB, Rance K, Taylor BA, Phillips SJ, Abbott CM et al (1997)
Mapping in the region of Danforths short tail and the localization of
which Cre recombinase expression is driven by Shh promoter tail length modifiers. Genome Res 7:108117
(Choi and Harfe 2011), has shown that Hedgehog signaling Alini M, Eisenstein SM, Ito K, Little C, Kettler AA, Masuda K, Melrose
is indeed required for notochord sheath formation and for J, Ralphs J, Stokes I, Wilke HJ (2008) Are animal models useful for
development of the nucleus pulposus. Interestingly, induc- studying human disc disorders/degeneration? Eur Spine J
17(1):219
tion of Cre recombinase at a later stage of development when Allen KD, Griffin TM, Rodriguiz RM, Wetsel WC, Kraus VB, Huebner
the notochord sheath had already formed using a tamoxifen- JL, Boyd LM, Setton LA (2009) Decreased physical function and
inducible Cre allele (ShhcreERT2) did not affect differentia- increased pain sensitivity in mice deficient for type IX collagen.
tion of the nucleus pulposus. Taken together, these data Arthritis Rheum 60(9):26842693
Ameye L, Aria D, Jepsen K et al (2002) Abnormal collagen fibrils in
strongly support the notion that the notochord gives origin to tendons of biglycan/fibromodulin- deficient mice lead to gait
the nucleus pulposus and indicate that hedgehog signaling, impairment, ectopic ossification, and osteoarthritis. FASEB J
though not required for differentiation of the notochord into 16:673680
nucleus pulposus cells, is essential for subsequent develop- Anderson GBJ, Weinstein JN (1996) Disc herniation. Spine 21:1S
Annunen S, Paassilta P, Lohiniva J, Perl M, Pihlajamaa T, Karppinen
ment of the notochord. In this regard, a transgenic mouse in J, Tervonen O, Krger H, Lhde S, Vanharanta H, Ryhnen L,
which Cre recombinase is expressed under the control of the Gring HH, Ott J, Prockop DJ, Ala-Kokko L (1999) An allele of
Foxa2 promoter (Uetzmann et al. 2008; Park et al. 2008) has COL9A2 associated with intervertebral disc disease. Science
been recently generated. Foxa2-Cre recombination activity 285:409412
Aszdi A, Chan D, Hunziker E, Bateman JF, Fssler R (1998) Collagen
can be detected in the endoderm but also in the node, II is essential for the removal of the notochord and the formation of
floorplate, and notochord. Therefore, in the future, this par- intervertebral discs. J Cell Biol 143(5):13991412
ticular mouse line could be an extremely useful tool for Battie MC, Videman T, Gibbon LE, Fisher LD, Manninen H, Gill K
notochord-specific knockouts. (1995) Determinants of lumbar disc degeneration: a study relating
lifetime exposures and magnetic resonance imaging findings in
identical twins. Spine 20:26012612
Battie MC, Videman T, Parent E (2004) Lumbar disc degeneration: epi-
21.3 Summary of Critical Concepts demiology and genetic influences. Spine 29:26792690
Discussed in the Chapter Behrens A, Haigh J, Mechta-Grigoriou F, Nagy A, Yaniv M, Wagner
EF (2003) Impaired intervertebral disc formation in the absence of
Jun. Development 130(1):103109
Spontaneous mutation models have enhanced the under- Berry R (1961) Genetically controlled degeneration of the nucleus pul-
standing of the role of genes in disc function. posus in the mouse. J Bone Joint Surg 43B:387393
Global gene targeting has significantly expanded our Boyd LM, Richardson WJ, Allen KD, Flahiff C, Jing L, Li Y, Chen J,
Setton LA (2008) Early-onset degeneration of the intervertebral
knowledge of the molecular mechanisms that regulate disc and vertebral end plate in mice deficient in type IX collagen.
disc development, homeostasis, and degeneration. Arthritis Rheum 58(1):164171
Conditional gene targeting has overcome the limitations Chalepakis G, Tremblay P, Gruss P (1992) Pax genes, mutants and
of universal gene targeting, allowing the analysis of the molecular function. J Cell Sci 16(Suppl):6167
Chiang C, Litingtung Y, Lee E, Young KE, Corden JL, Westphal H,
genes of interest at different embryological stages and in Beachy PA (1996) Cyclopia and defective axial patterning in mice
different tissues, without affecting mouse viability. This lacking sonic hedgehog gene function. Nature 383(6599):
technique has also led to the discovery of the embryologi- 407413
cal origin of the nucleus pulposus. Choi KS, Harfe BD (2011) Hedgehog signaling is required for formation
of the notochord sheath and patterning of nuclei pulposi within the
Future research related to the intervertebral disc will intervertebral discs. Proc Natl Acad Sci U S A 108(23):94849489
involve a considerable extension of the use of tissue- Choi KS, Cohn MJ, Harfe BD (2008) Identification of nucleus pulposus
specific and temporally inducible knockout models, fur- precursor cells and notochordal remnants in the mouse: implications
ther expanding our knowledge of the development of for disk degeneration and chordoma formation. Dev Dyn
237(12):39533958
these important tissues and providing direction in the Cs-Szabo G, Ragasa-San Juan D, Turumella V, Masuda K, Thonar EJ,
search for therapies. An HS (2002) Changes in mRNA and protein levels of proteogly-
cans of the annulus fibrosus and nucleus pulposus during interver-
Acknowledgements This work was supported by the NIH RO1 tebral disc degeneration. Spine (Phila Pa 1976) 27(20):
AR04819 grant (to E.S.). 22122219
Enomoto-Iwamoto M, Kitagaki J, Koyama E et al (2002) The Wnt
antagonist Frzb-1 regulates chondrocyte maturation and long bone
development during limb skeletogenesis. Dev Biol 251:142156
Francis-West PH, Abdelfattah A, Chen P et al (1999) Mechanisms of
References GDF-5 action during skeletal development. Development
126:13051315
Ala-Kokko L (2002) Genetic risk factors for lumbar disc disease. Ann Frymoyer JW (1988) Back pain and sciatica. N Engl J Med
Med 34:4247 318:291300
Frymoyer JW, Cats-Baril WL (1991) An overview of the incidence and Paassilta P, Lohiniva J, Goring HH, Perala M, Raina SS, Karppinen
costs of low back pain. Orthop Clin North Am 22:263271 J et al (2001) Identification of a novel common risk factor for lum-
Furukawa T, Ito K, Nuka S, Hashimoto J, Takei H, Takahara M, Ogino bar disc disease. JAMA 285:18431849
T, Young MF, Shinomura T (2009) Absence of biglycan accelerates Park EJ, Sun X, Nichol P, Saijoh Y, Martin JF, Moon AM (2008) System
the degenerative process in mouse intervertebral disc. Spine (Phila for tamoxifen-inducible expression of cre-recombinase from the
Pa 1976) 34(25):E911E917 Foxa2 locus in mice. Dev Dyn 237(2):447453
Hunter CJ, Matyas JR, Duncan NA (2004) Cytomorphology of noto- Peters H, Neubser A, Kratochwil K, Balling R (1998) Pax9- deficient
chordal and chondrocytic cells from the nucleus pulposus: a species mice lack pharyngeal pouch derivatives and teeth and exhibit cran-
comparison. J Anat 205:357362 iofacial and limb abnormalities. Genes Dev 12:27352747
Jin H, Shen J, Wang B, Wang M, Shu B, Chen D (2011) TGF- signal- Peters H, Wilm B, Sakai N, Imai K, Maas R, Balling R (1999) Pax1 and
ing plays an essential role in the growth and maintenance of inter- Pax9 synergistically regulate vertebral column development.
vertebral disc tissue. FEBS Lett 585(8):12091215 Development 126(23):53995408
Kawaguchi Y, Osada R, Kanamori M, Ishihara H, Ohmori K, Matsui H Pluijm SM, van Essen HW, Bravenboer N, Uitterlinden AG, Smit JH,
et al (1999) Association between an aggrecan gene polymorphism Pols HA, Lips P (2004) Collagen type I alpha1 Sp1 polymorphism,
and lumbar disc degeneration. Spine 24:24562460 osteoporosis, and intervertebral disc degeneration in older men and
Kimura T, Nakata K, Tsumaki N, Miyamoto S, Matsui Y, Ebara S, Ochi women. Ann Rheum Dis 63:7177
T (1996) Progressive degeneration of articular cartilage and inter- Raj PP (2008) Intervertebral disc: anatomy-physiology-pathophysiol-
vertebral discs. An experimental study in transgenic mice bearing a ogy-treatment. Pain Pract 8(1):1844. Review
type IX collagen mutation. Int Orthop 20(3):177181 Risbud MV, Schaer TP, Shapiro IM (2010) Toward an understanding of
Komori H et al (1996) The natural history of herniated nucleus pulpo- the role of notochordal cells in the adult intervertebral disc: from
sus with radiculopathy. Spine 21:225229 discord to accord. Dev Dyn 239(8):21412148. Review
Kondo N, Yuasa T, Shimono K, Tung W, Okabe T, Yasuhara R, Pacifici Roughley P, Martens D, Rantakokko J, Alini M, Mwale F, Antoniou
M, Zhang Y, Iwamoto M, Enomoto-Iwamoto M (2011) Intervertebral J (2006) The involvement of aggrecan polymorphism in degenera-
disc development is regulated by Wnt/-catenin signaling. Spine tion of human intervertebral disc and articular cartilage. Eur Cell
(Phila Pa 1976) 36(8):E513E518 Mater 11:17; discussion 7. Review
Lane PW, Birkenmeiser CS (1993) Urogenital syndrome (us): a devel- Roughley PJ, Lamplugh L, Lee ER, Matsumoto K, Yamaguchi Y (2011)
opmental mutation on chromosome 2 of the mouse. Mamm Genome The role of hyaluronan produced by Has2 gene expression in devel-
4:481484 opment of the spine. Spine (Phila Pa 1976) 36(14):E914E920
Le Y, Sauer B (2001) Conditional gene knockout using Cre recombi- Sahlman J, Inkinen R, Hirvonen T, Lammi MJ, Lammi PE, Nieminen
nase. Mol Biotechnol 17(3):269275 J et al (2001) Premature vertebral endplate ossification and mild
Lefebvre V, Li P, de Crombrugghe B (1998) A new long form of Sox5 disc degeneration in mice after inactivation of one allele belonging
(L-Sox5), Sox6 and Sox9 are co-expressed in chondrogenesis and to the Col2a1 gene for type II collagen. Spine 26:25582565
cooperatively activate the type II collagen gene. EMBO Sakai D, Nakai T, Mochida J, Alini M, Grad S (2009) Differential phe-
J 17:57185733 notype of intervertebral disc cells: microarray and immunohis-
Lewandoski M (2001) Conditional control of gene expression in the tochemical analysis of canine nucleus pulposus and annulus fibrosus.
mouse. Nat Rev Genet 2(10):743755. Review Spine (Phila Pa 1976) 34(14):14481456
Li X, Leo BM, Beck G et al (2004) Collagen and proteoglycan abnor- Sambrook PN, MacGregor AJ, Spector TD (1999) Genetic influences
malities in the GDF-5- deficient mice and molecular changes when on cervical and lumbar disc degeneration: a magnetic resonance
treating disk cells with recombinant growth factor. Spine (Phila Pa imaging study in twins. Arthritis Rheum 42:366372
1976) 29(20):22292234 Sarver JJ, Elliott DM (2004) Altered disc mechanics in mice genetically
Maier JA, Harfe BD (2011) Nuclei pulposi formation from the embry- engineered for reduced type I collagen. Spine 29:10941098
onic notochord occurs normally in GDF5-deficient mice. Spine Sauer B (1998) Inducible gene targeting in mice using the Cre/lox sys-
(Phila Pa 1976) 36(24):E1555E1561 tem. Methods 14(4):381392. Review
Matsui H, Maeda A, Tsuji H, Naruse Y (1997) Risk indicators of low Schipani E (2002) Conditional gene inactivation using cre recombinase.
back pain among workers in Japan. Association of familial and Calcif Tissue Int 71(2):100102. Epub 2002 Jul 24. Review
physical factors with low back pain. Spine 22:12421247 Semba K, Araki K, Li Z, Matsumoto K, Suzuki M, Nakagata N, Takagi
Merino R, Macias D, Ga an Y, Economides AN, Wang X, Wu Q, K, Takeya M, Yoshinobu K, Araki M, Imai K, Abe K, Yamamura K
Stahl N, Sampath KT, Varona P, Hurle JM (1999) Expression and (2006) A novel murine gene, Sickle tail, linked to the Danforths
function of Gdf-5 during digit skeletogenesis in the embryonic short tail locus, is required for normal development of the interver-
chick leg bud. Dev Biol 206(1):3345 tebral disc. Genetics 172(1):445456
Mundy C, Yasuda T, Kinumatsu T, Yamaguchi Y, Iwamoto M, Enomoto- Silberberg R, Aufdermaur M, Adler JH (1979) Degeneration of the
Iwamoto M, Koyama E, Pacifici M (2011) Synovial joint formation intervertebral disks and spondylosis in aging sand rats. Arch Pathol
requires local Ext1 expression and heparan sulfate production in Lab Med 103(5):231235
developing mouse embryo limbs and spine. Dev Biol 351(1): Singh K, Masuda K, An HS (2005) Animal models for human disc
7081 degeneration. Spine J 5(6 Suppl):267S279S. Review
Nakata K, Ono K, Miyazaki J, Olsen BR, Muragaki Y, Adachi E, Smits P, Lefebvre V (2003) Sox5 and Sox6 are required for notochord
Yamamura K, Kimura T (1993) Osteoarthritis associated with mild extracellular matrix sheath formation, notochord cell survival and
chondrodysplasia in transgenic mice expressing alpha 1(IX) colla- development of the nucleus pulposus of intervertebral discs.
gen chains with a central deletion. Proc Natl Acad Sci U S A Development 130(6):11351148
90(7):28702874 Smits P, Li P, Mandel J, Zhang Z, Deng JM, Behringer RR, de
Noponen-Hietala N, Virtanen I, Karttunen R, Schwenke S, Jakkula E, Crombrugghe B, Lefebvre V (2001) The transcription factors
Li H et al (2005) Genetic variations in IL6 associated with interver- L-Sox5 and Sox6 are essential for cartilage formation. Dev Cell
tebral disc disease characterized by sciatica. Pain 114:186194 1(2):277290
Orban PC, Chui D, Marth JD (1992) Tissue- and site-specific DNA Solovieva S, Leino-Arjas SK, Ala-Kokko L, Luomo K, Raininko R,
recombination in transgenic mice. Proc Natl Acad Sci U S A Saarela J et al (2004) Interleukin 1 polymorphisms and interverte-
89(15):68616865 bral disc degeneration. Epidemiology 15:626633
Stark WM, Boocock MR, Sherratt DJ (1992) Catalysis by site-specific mouse cartilage matrix deficiency (cmd) mutant. Dev Genet 22(4):
recombinases. Trends Genet 8(12):432439. Review 349358
Tamamura Y, Otani T, Kanatani N, Koyama E, Kitagaki J, Komori Wallin J, Wilting J, Koseki H, Fritsch R, Christ B, Balling R (1994) The
T, Yamada Y, Costantini F, Wakisaka S, Pacifici M, Iwamoto role of Pax-1 in axial skeleton development. Development
M, Enomoto-Iwamoto M (2005) Developmental regulation of 120(5):11091121
Wnt/-catenin signals is required for growth plate assembly, carti- Watanabe H, Nakata K, Kimata K, Nakanishi I, Yamada Y (1997)
lage integrity, and endochondral ossification. J Biol Chem. 280(19): Dwarfism and age-associated spinal degeneration of heterozygote cmd
1918519195 mice defective in aggrecan. Proc Natl Acad Sci U S A 94(13):
Tilkeridis C, Bei T, Garantziotis S, Stratakis C (2005) Association of a 69436947
COL1A1 polymorphism with lumbar disc disease in young military Wilm B, Dahl E, Peters H, Balling R, Imai K (1998) Targeted disrup-
recruits. J Med Genet 42:e44 tion of Pax1 defines its null phenotype and proves haploinsufficiency.
Uetzmann L, Burtscher I, Lickert H (2008) A mouse line expressing Proc Natl Acad Sci U S A 95(15):86928697
Foxa2-driven Cre recombinase in node, notochord, floorplate, and Wilson TJ, Kola I (2001) The LoxP/CRE system and genome
endoderm. Genesis 46(10):515522 modification. Methods Mol Biol 158:8394. Review
Van Duyne GD (2001) A structural view of cre-loxp site-specific Wisneski RJ, Garfin SR, Rothman RH (1992) Lumbar disc disease. In:
recombination. Annu Rev Biophys Biomol Struct 30:87104. Rothman RH, Simeone FA (eds) The spine, 3rd edn. W.B. Saunders
Review Co, Philadelphia, p 671
Varlotta GP, Brown MD, Kelsey JL, Golden AL (1991) Familial predis- Zachgo J, Korn R, Gossler A (1998) Genetic interactions suggest that
position for herniation of a lumbar disc in patients who are less than Danforths short tail (Sd) is a gain-of-function mutation. Dev Genet
twenty-one years old. J Bone Joint Surg Am 73:124128 23(1):8696
Videman T, Gibbons LE, Battie MC, Maravilla K, Vanninen E, Zhang Y, Sun Z, Liu J, Guo X (2008) Advances in susceptibility genet-
Leppavuori J et al (2001) The relative roles of intragenic polymor- ics of intervertebral degenerative disc disease. Int J Biol Sci
phisms of the vitamin D receptor gene in lumbar spine degeneration 4(5):283290
and bone density. Spine 26:A1A6 Ziv I, Moskowitz RW, Kraise I, Adler JH, Maroudas A (1992)
Wai AW, Ng LJ, Watanabe H, Yamada Y, Tam PP, Cheah KS (1998) Physicochemical properties of the aging and diabetic sand rat inter-
Disrupted expression of matrix genes in the growth plate of the vertebral disc. J Orthop Res 10:205210
Intervertebral Disc Culture Models
and Their Applications to Study 22
Pathogenesis and Repair
Svenja Illien-Jnger, Benjamin A. Walter, Jillian E. Mayer,

Andrew C. Hecht, and James C. Iatridis
Contents 22.8 Biomaterials for Intervertebral Disc Repair .............. 366

22.1 Introduction ................................................................... 353 22.9 Human Organ Culture Models .................................... 366
22.1.1 Animal, Cell, and Organ Culture Models ....................... 354
22.1.2 The Intervertebral Disc Niche ......................................... 354 22.10 Limitations of Organ Culture Models ......................... 367
22.10.1 Age Changes and Use of Human Tissues ....................... 367
22.2 Objectives ....................................................................... 354 22.10.2 Measurements of Predictors of Painful Conditions ........ 367
22.3 Organ Culture Model Development ............................ 355 22.11 Conclusions .................................................................... 368
22.3.1 Choice of Animal Models ............................................... 356
22.3.2 Intervertebral Disc Swelling ........................................... 357 22.12 Summary of Critical Concepts Discussed
22.3.3 Free-Swelling and Osmotic Loading Models ................. 358 in the Chapter ................................................................ 368
22.3.4 Static and Diurnal Compression Loading Models .......... 358 References .................................................................................... 369
22.3.5 Improved Dissection Procedures for
Disc Organ Culture ......................................................... 359
22.4 Dynamic Compression Loading Models ..................... 361
22.5 Insights into Mechanism of Degeneration
Using Disc Organ Culture ............................................ 361
22.5.1 Altered Media Conditions ............................................... 362
22.5.2 Mechanical and Nutritional Challenges .......................... 363 22.1 Introduction
22.5.3 Surgical and Chemical Challenges.................................. 364
22.6 Mechanisms of Repair and Organ Culture The intervertebral disc is the largest avascular and aneural
Models of Disc Regeneration ........................................ 365 structure in the human body (Holm et al. 1981) and com-
22.6.1 Screening of Potential Therapeutic Agents ..................... 365 posed of a number of interacting and interdependent tissues.
22.7 Stem Cells in the Intervertebral Disc .......................... 365 While there are numerous clinical studies of the human disc,
22.7.1 Fate of Injected Stem Cells ............................................. 366 there remains a great deal to learn concerning its basic biol-
22.7.2 Homing of MSCs ............................................................ 366 ogy, alterations during disease, and treatment strategies.
One important approach to studying the pathophysiology
and the repair of the degenerative disc is through the use of
animal models that approximate its unique anatomy and
physiology. Although many relevant animal systems are
available, because of the relatively small size of the disc,
differences in cell and tissue function, and confounding bio-
logic factors that include nutrient transport and metabolic
activity, it is difficult to generate models that are comparable
to the human. Simulating the disc niche is one of the great-
est challenges in intervertebral disc degeneration research
S. Illien-Jnger, PhD B.A. Walter, MS
A.C. Hecht, MD J.C. Iatridis, PhD (*) J.E. Mayer, BS and a major obstacle in preventing the direct translation of
Leni and Peter W. May Department of Orthopaedics, findings relevant to cell culture and small animal model
Icahn School of Medicine at Mount Sinai, studies to the human condition. This chapter describes his-
1 Gustave Levy Place, 1188, New York, NY 10029-6574, USA
torical advancements in techniques used for intervertebral
e-mail: svenja.illien-juenger@mssm.edu; benjamin.walter@mssm.edu;
andrew.hecht@mountsinai.org; james.iatridis@mssm.edu; disc organ culture models and their applications to study
jillian.mayer@mssm.edu disc physiology, pathogenesis of disease, and repair.

354 S. Illien-Jnger et al.
22.1.1 Animal, Cell, and Organ Culture Models boundary conditions of the disc compared with in vivo mod-
els. A key strength of organ culture techniques is that native
The current understanding of the biology of the interverte- cellular matrix connectivity and other niche characteristics
bral disc and its associated diseases has been established are retained, which is important because cellular behavior is
through studies using human cadaveric tissue, surgical greatly influenced by the surrounding environment (Risbud
samples, and animal models. While animal models are et al. 2003).
essential for hypothesis testing and therapeutic screening, as
will be discussed later, they can differ both structurally and
cellularly from the human disc. Small and large animal 22.1.2 The Intervertebral Disc Niche
in vivo models are ideal for investigating interactions
between the intervertebral disc and surrounding spinal tis- The intervertebral niche is formed by the intimate connec-
sues as well as developmental studies of systemic tion between cells and their surroundings, and in the mature
inflammation, healing, or pain. However, in vivo conditions human adult, this niche is strongly influenced by the nutri-
cannot be precisely controlled as in vitro models. Live ani- tional environment, the presence of cytokines, and the com-
mal testing is costly from both economic and ethical per- position of the extracellular matrix. The specific structure
spectives, and for large animal models, the capacity for and niche environment undergoes age-dependent adaptive
high-throughput or long-term in vivo studies is limited. changes that can lead to the accumulation of toxic metabo-
Despite these drawbacks, large animal models are most lites. Nutrients and metabolites are transported into the
appropriate for evaluating late-stage therapeutic interven- nucleus pulposus by fluid exchange mediated by the diffu-
tions and treatment strategies. sion of molecules through the cartilaginous endplates or
Studies of surgical and cadaveric tissue are of great rele- outer annulus fibrosus. As a result, oxygen tension and glu-
vance to the human condition. Human cadaveric tissue is cose concentrations within the central region of the disc are
required for spine biomechanical testing and for studies of generally low; the acidic by-products of anaerobic
surgical repair but is limited in terms of measurements of metabolism and the slow transport of waste products out of
biologic change in response to interventions. Biochemical the disc induce a decrease in environmental pH, typically
and histological analyses facilitate investigations of cellular- composed of lactic acid (Urban et al. 2004). It is assumed
ity, inflammation, matrix composition, and other biological, that the low concentrations of required nutrients and metab-
structural, and mechanical questions even on postmortem olites and the relatively high concentration of waste prod-
tissue. When cadaveric human tissue is obtained within ucts provide a physicochemical environment that is
approximately 24 h of death, intervertebral disc cells can supportive of a small cell number (4,0009,000 cells) and
also be harvested for long-term cell and organ culture experi- slow metabolic rate (Bibby et al. 2005; Maroudas
ments. Human surgical samples are the most clinically rele- et al. 1975).
vant, yet the quantity (i.e., for multiple dependent variable
measurements) is often limited. Surgical tissue quality can
also be highly variable preventing clear distinction between 22.2 Objectives
nucleus pulposus, annulus fibrosus, and herniation tissue.
Furthermore, surgical samples, almost by definition, exclude Based on its anatomy and physiology, there are many
healthy tissues. unique challenges for simulating and studying the mecha-
Intervertebral disc cell culture experiments are used for nism by which the intervertebral disc functions in health
therapeutic screening and also for hypothesis testing. Most and disease. While all model systems have limitations,
cell culture techniques extract and then harvest cells from the retaining mechanical, chemical, and biological characteris-
native matrix, thereby altering the important interactions tics that are similar to, or mimic, the human disc is a prior-
between matrix and cells. Both matrix constituents and novel ity; this chapter focuses on specific techniques and
biomaterials provide high-quality 3D cell culture environ- advantages of organ culture models. The purpose of this
ments, but these model systems all differ from the native tis- chapter, therefore, is to provide an overview of interverte-
sue niche. bral disc organ culture systems, the historical aspects of
Organ culture is a variant of more traditional cell culture method development, technical considerations required for
systems where instead of extracting and culturing the tissue successful implementation of organ culture, its importance
cellular components, the entire intervertebral disc is col- as an experimental model for the study of intervertebral
lected and maintained in culture. Ex vivo organ culture also disc pathology, and finally as a screening tool for imple-
allows more precise control over mechanical and chemical mentation of therapeutic repair.
22 Intervertebral Disc Culture Models and Their Applications to Study Pathogenesis and Repair 355
Box 22.1 Intervertebral Disc Organ Culture models an attractive option. Comparison with other ani-
System Development mal models can be found in Sect. 22.3.1.
A historical survey of published papers, found using Need to quickly test novel hypotheses and screen thera-
online search engines, demonstrates that there has been peutic strategies. The high degree of control over mechan-
a substantial increase in the number of papers published ical and chemical boundary conditions of organ culture
on intervertebral disc organ culture systems. Notably, models is an advantage over in vivo systems. The elabo-
approximately 10 years ago, there were less than two rate 3D cellular niche of intervertebral disc cells can be
papers on the topic published per year, and in 2011 maintained in organ culture and provide major advantages
there were seven papers published in the first 10 months when compared with more traditional cell culture systems
of the year (Table 22.1). Similar trends were found in especially with respect to testing novel scientific hypothe-
the Transactions of the Orthopaedic Research Society. ses and screening of therapeutics. Organ culture models
Why are so many research groups adopting these novel that describe pathophysiology of degeneration and inter-
techniques? There are at least four answers to this vertebral disc repair are described in Sects. 23.5 and 23.6.
question:
Because human intervertebral discs can be maintained in
Technical advance. Since the first disc organ culture paper culture. There are many practical and ethical constraints
was published in 1979 (Oegema et al. 1979), several dis- on human subject testing. The capacity to maintain viable
section procedures, bioreactor systems, and cell culture human intervertebral discs in organ culture provides a
technologies have been developed. It was necessary to new and important preclinical tool that offers promise for
overcome two major technical obstacles. The first involved screening therapeutic agents in a way that complements
tissue processing procedures that allow transport of nutri- live animal testing. The use of human discs in organ cul-
ents into and waste products out of the intervertebral disc. ture is described in Sect. 23.7.
The second involved the development of improved
dissection procedures and bioreactors to inhibit axial Table 22.1 Publication history of intervertebral disc organ culture
swelling and apply physiological loading. Comprehensive systems
techniques for organ culture are described in Sect. 22.3.5. Date range www.pubmed.gov www.ors.org
2011 7 5
Need for human-relevant models of intervertebral discs.
20062010 26 11
The discs of most animal models have nutrient transport
20012005 7 3
mechanisms, micro-architecture, and cell metabolic rates 19962000 8 0
that are different from the mature human intervertebral 19911995 6 na
disc. Rodents have small intervertebral discs with a more >1990 4 na
gelatinous nucleus pulposus structure, a predominantly Survey of the number of intervertebral disc organ culture papers
notochordal cell composition, and higher cell metabolism published over the last two decades based on electronic searches of
rates than the mature human intervertebral disc. In con- Medline (www.pubmed.gov; keywords intervertebral organ cul-
trast, the intervertebral discs of some large animals are ture; date 9/27/2011) and of the Transactions of the Orthopaedic
Research Society Meeting (www.ors.org, keywords intervertebral
closer in structure, size, and cellularity to the mature organ culture; date 9/27/2011); note that electronic search spans
human discs. The expenses and ethical costs of large ani- 19992011. The table demonstrates the recent growth in the number
mal in vivo testing make large animal organ culture of papers on the topic of organ culture
22.3 Organ Culture Model Development organ culture systems faced many challenges in establishing
effective and relevant culture conditions that were adequate
Organ culture models have been developed over the last few to maintain tissue integrity and intervertebral disc cell viabil-
decades using animal discs of various sizes in either simple ity and metabolism. The most modern systems are computer
or complex bioreactor systems to provide controlled in vitro controlled and capable of maintaining multiple intervertebral
conditions that simulate the in vivo situation. The earliest discs from large animals with loading in multiple degrees of
freedom. The following sections provide an overview of the instrumentation design. Even loading on the tail is domi-
historical progression and diversity of organ culture models nated by axial musculature, so that loading differences
that have been developed. between ovine, bovine, rodent, and other coccygeal interver-
tebral discs models are typically no more significant than the
biological differences between species. These biomechani-
22.3.1 Choice of Animal Models cal similarities are highlighted by the findings that the effec-
tive axial stress (axial force normalized by cross-sectional
When compared to the human, small animal models (includ- area) required to restrain swelling of tail intervertebral discs
ing mice, rats, and rabbits) have enhanced nutrient and (0.10.3 MPa) is similar to the in vivo intradiscal pressure
waste transport characteristics. Small animal models also measured in human lumbar discs in the prone position (0.1
have higher matrix water content, greater distinction 0.3 MPa; see also Sect. 22.3.4). The range of motion and
between annulus fibrosus and nucleus pulposus, and a larger peak stresses do vary across intervertebral discs levels and
number of notochordal cells that persist to older ages; the species (Alini et al. 2008; Demers et al. 2004; MacLean et al.
latter cells exhibit an elevated metabolic activity when com- 2005; Oshima et al. 1993; Wilke et al. 1999). These differ-
pared with chondrocytic nucleus pulposus cells (Aguiar ences are paramount when considering final geometry and
et al. 1999; Oegema et al. 2000). Indeed, the notochordal material choices for instrumentation and implants at final
population in human intervertebral discs is greatly reduced stages of product development.
if not entirely absent by puberty (Hunter et al. 2004). Sakai Most commonly, the intervertebral discs of large ani-
et al. (2009) found considerable interspecies differences mals are smaller in size than human lumbar discs. However,
when comparing gene expression levels of notochordal and human intervertebral discs have a greater disc height
non-notochordal animal discs with cells of human interver- (height = 11.3 0.3 mm) and they are elliptical in cross
tebral discs. The high metabolic rate, relatively rapid aging section (medial-lateral diameter = 55.9 9.4 mm and ante-
process, and low cost of small animal models are all very rior-posterior diameter = 37.2 4.7 mm). Bovine caudal
attractive research features. However, these advantages discs are smaller than human lumbar discs (bovine disc
ignore major differences between animal and human disc height = 6.90 0.35 mm) and circular in cross section (diam-
cells. From this perspective, therapeutic interventions that eter = 28.9 2.0 mm). The normalized transport distance or
may be successful in small animal models may not succeed aspect ratios (disc height normalized to lateral diameter) are
in large animal models or humans. Depending on the bio- similar for human (0.202) and bovine (0.239; OConnell
logic or biomechanical question, small animal models can et al. 2007). A comparison of intervertebral disc dimensions
also be of limited value as they lack sufficient tissue for and differences in tissue composition of commonly used ani-
multiple measurements. mal models is shown in Fig. 22.1. The most obvious differ-
The intervertebral discs of large animal models are of ence between the discs is the size: bovine coccygeal discs
comparable size to the human, and nutrient transport, disc have diameter of ~22 mm (30 % smaller than human discs),
structure, cellularity, and metabolic rates are similar. followed by lumbar porcine (~18 mm), coccygeal ovine
Nutritional and transport limitations in the large human (~15 mm), rat (~4 mm), and mice (~2 mm). The coccygeal
intervertebral discs have long been considered a key factor in discs of bovine and ovine are approximately round in the
slowing successful repair (Kandel et al. 2008) and in contrib- transverse plane compared to the human and porcine which
uting to the uniquely challenging disc cell niche. These large are kidney bean shaped and typically approximate to an
diffusion distances that human disc cells experience are ellipse. Similar to the healthy human nucleus pulposus, the
likely to become a critical issue as research becomes more nucleus of bovine and ovine coccygeal intervertebral discs
clinically translatable. Recently developed bioreactor sys- appears white and fibrous, whereas the gelatinous nucleus
tems are designed to handle the size and loading require- of the porcine and rodent lumbar discs clearly differs in
ments of bovine and ovine intervertebral coccygeal discs composition. Several authors have provided a more compre-
which are among the most commonly used tissue sources. hensive comparison of bovine coccygeal intervertebral discs
There are differences in the types of loading that interver- with the human (Demers et al. 2004; Oshima et al. 1993).
tebral discs of quadrupeds experience when compared to Furthermore, most smaller species retain notochordal cells
bipeds. This difference reflects the orientation of the spine throughout aging, while bovine and ovine species do not
with respect to gravity and biomechanical differences (Hunter et al. 2004).
between anatomic regions (tails vs. spines). Surprisingly, Taken together, coccygeal intervertebral discs from bovine
since the dominant loading forces on the spine of all animals and ovine are of large size with similar aspect ratios, trans-
are associated with muscle action along the axis of the spine, port distances, and composition as human discs and are
these biomechanical differences are often smaller than one becoming accepted tissues for large animal organ culture
might expect and are most relevant when considering systems.
a b c d e f
Fig. 22.1 Disc species comparison. Transverse sections from (a), human (lumbar); (b), bovine (coccygeal); (c), porcine (lumbar); (d), ovine (coc-
cygeal); (e), rat (lumbar); and (f), mouse (lumbar) intervertebral discs
BEP noEP
Without endplate
With endplate
Pre-swelling
50
40
Weight gain(%)
30
Post-swelling
20
10
0
0 10 20 30 40 50 60 70
Time (h)
Fig. 22.2 Free swelling of disc explants. Left: comparison of swelling endplates and cartilage endplate cultures after 66 h (n = 4, error bars
capacity and deformation of intervertebral discs with and without bone indicate standard deviation; modified from Gawri et al. 2011)
endplates. Right: change in wet weight of intervertebral discs without
22.3.2 Intervertebral Disc Swelling and 90 min was 10, 16, and 22 % for intervertebral discs with-
out endplates, while the change was 2, 4, and 5 % for discs
A significant challenge for intervertebral disc research, with bone-covered endplates (Fig. 22.2)). After 15 h, the
regardless of species, is to maintain the original tissue struc- weight of discs without endplates increased around 22 %
ture. When muscular loading is removed and intervertebral compared to 10 % when cultured with bone-covered end-
discs are isolated, the tissue exhibits a propensity to swell in plates. A similar finding was observed by Gawri et al. (2011)
standard culture media as a result of development of large who compared swelling of intervertebral discs with cartilage
osmotic gradients (Urban and Maroudas 1981). Notably, if endplates vs. discs without endplates; after 66 h in culture, the
the disc is isolated without the superior and inferior endplates, observed increase in wet weight was 21 % in the cartilage
its weight increases drastically compared to discs with bone- endplate group, compared with 44 % for the discs without
covered endplates. This rapid increase in wet weight of inter- endplates (Gawri et al. 2011). When disc tissue was isolated
vertebral discs without endplates is apparent within the first as fragments, the wet weight was elevated to an even greater
1.5 h (% increase of initial wet weight for discs after 10, 20, extent (Urban and Maroudas 1981). In response to swelling,
disc cells increase expression of catabolic and pro- the concept that reduced nutrient and metabolite transport
inflammatory genes and can undergo cell death (Haschtmann through the vertebrae and annulus in these models is a prob-
et al. 2008a). Conversely, as the disc dehydrates in air, expo- lem. It is also notable that endplate transport may be a limita-
sure to aqueous media is required to maintain tissue structure tion even in these small animal models which have diffusional
and biomechanical integrity (Pflaster et al. 1997). To counter transport distances much smaller than human and large ani-
or prevent this propensity for swelling and preserve the physi- mal models.
ological structure of the intervertebral disc, ex vivo, several Haschtmann et al. (2006a) cultured rabbit intervertebral
strategies have been developed. discs with retained endplates in standard media under free-
swelling conditions. It was hypothesized that the endplate
served to naturally constrain the nucleus pulposus. These
22.3.3 Free-Swelling and Osmotic workers observed that cell viability could be maintained for
Loading Models at least 4 weeks without losing structural integrity or matrix
composition. However, degenerative gene expression pat-
To evaluate cell metabolism and protein synthesis, interver- tern and lowered metabolic rate were reported, thus limiting
tebral discs have been cultured in media as intact organs the use of this model for the study of the pathophysiological
(Oegema et al. 1979). Despite the swelling which can disrupt and long-term changes in the disc (Haschtmann et al. 2006a).
structure and promote loss of glycosaminoglycans (GAGs) In a subsequent study, rabbit intervertebral discs were cul-
and structural proteins, cells in the disc remain metabolically tured with endplates with diurnal changes in hyperosmotic
active (Chiba et al. 1998). Chiba and coworkers cultured conditions. These conditions were better able to maintain
whole rabbit intervertebral disc for up to 1 month in alginate aggrecan gene expression and inhibit swelling and collagen
to inhibit swelling using techniques adapted from more I expression than conventional static osmolarity conditions
traditional cell culture protocols. In this system, tissue (Haschtmann et al. 2006b). This finding suggests that the
integrity and biosynthetic activities were maintained. amount of loading and the way the load was applied has a
However, other investigators indicated that this technique, large effect on disc biosynthetic activity.
and its general nonphysiological approach to inhibit swell- In a further study by Van Dijk et al. (2011), swelling of
ing, required further optimization. A new approach using nucleus pulposus tissue explants was inhibited using culture
hyperosmotic media (supplemented with TGFb) to inhibit media made hypertonic by supplementation with polyethyl-
swelling was described by Risbud et al. who cultured rat ene glycol. Tissue integrity and cell viability of these bovine
lumbar intervertebral discs for 1 and 3 weeks. After 1 week, explants was maintained for 21 days. However, the decrease
no differences in cell viability, morphology, or gene and pro- in matrix gene expression observed in this tissue explant
tein expression profiles were found between cultured and model (a best case scenario for nutrient transport since there
control discs, whereas after 3 weeks in culture, a loss of cell were no endplates or annulus fibrosus to impede nutrient
function was observed (Risbud et al. 2003). This was a novel transport) highlighted that biomechanical stimulation, in
approach for investigating the nucleus pulposus function addition to nutrient transport, is important to promote ana-
under physiological and pathophysiological conditions. The bolic metabolism of the intervertebral disc.
model also demonstrated that even if swelling of the inter-
vertebral disc could be prevented, cell viability could not be
maintained, suggesting that inhibition of swelling was not 22.3.4 Static and Diurnal Compression
sufficient to maintain physiological levels of nucleus pulpo- Loading Models
sus cell viability and function in long-term culture.
To prevent swelling of the intervertebral disc in small ani- Oshima et al. (1993) evaluated the axial load required to pre-
mals, a further refinement of this technique involved retention vent swelling. These workers estimated that the interverte-
of the surrounding tissues. Lim et al. (2006) cultured rat bral disc experiences an effective mechanical stress of
motion segments (the disc with the adjacent vertebrae) for ~0.3 MPa in vivo. In a short 12-h study, both tissue hydration
4 weeks under free-swelling conditions; they reported that and matrix synthesis were affected by varying the static axial
after 14 days, nucleus pulposus and annulus fibrosus cell via- loading. In vivo levels were maintained when the effective
bility (~95 %) was maintained; after 21 days, only 25 % of stress ranged between 0.13 and 0.26 MPa (Ohshima et al.
the cells were viable and viability decreased even further after 1995). This early work established what is now considered to
28 days (9 %). Thus, the authors demonstrated that rigid be appropriate baseline loads for large animal disc organ cul-
fixation of the disc through the vertebral bodies inhibited tis- ture and showed the feasibility of this model for studies of
sue swelling in media. However, similar to the study by the mechanobiology of the intervertebral disc (Fig. 22.3a).
Risbud and coworkers, the loss of cell viability was problem- Baseline axial compression levels of similar values
atic at longer time intervals. Together, these studies support ~0.15 MPa were also found to retain in vivo disc height in rat
Porous disc biomechanical integrity of the disc. However, culturing

a
Load discs with endplates drastically reduced cell viability within
Dialysis membrane
1 week, especially in the nucleus. The reduced viability
Perfusate in Plate Perfusate out was hypothesized to be associated with formation of blood
medium
clots in the vertebral endplates that inhibited nutrient trans-
Microdialysis tube Disc Free isotope monitor
port. A further validation of bovine organ culture explants
Saline without endplates was performed by Korecki et al. (2007;
Fig. 22.3b) who found that static as well as diurnal loading
conditions maintained GAG content and had similar cell
metabolism rates.
b
22.3.5 Improved Dissection Procedures
Weights
for Disc Organ Culture
Weight support
Peristaltic pump
As nucleus pulposus viability in organ culture requires the
retention of the endplate cartilage, it was necessary to
Bioreactor with improve current methods for isolating these regions of the
sealing membrane intervertebral disc. Gantenbein et al. (2006) obtained inter-
vertebral discs from sheep that had been treated with a
systemic anticoagulant before death. The endplates were
cleaned thoroughly with a debridement tool to ensure
vasculature channel competence and facilitate nutrient
transport. Intervertebral discs with endplates were cultured
either under static (0.2 MPa/24 h) or diurnal loading
(0.2 MPa/6 h and 0.8 MPa/16 h) for 4 days in media supple-
Media tubing mented with 50 mol/l of dextran conjugated with a
fluorescent label. They demonstrated that when discs are
Fig. 22.3 Static and diurnal organ culture models. (a) Schematic rep-
loaded diurnally, the fluorophores diffused to a greater
resentation of the first axial compression bioreactor for large animal
organ culture designed by Ohshima et al. (1995). (b) Load frame to extent through the annulus fibrosus and endplate towards
allow diurnal loading on the Ohshima-style static bioreactor the disc center than when loaded statically. This was the
first study to demonstrate a technique in which (1) large
animal intervertebral discs could be cultured with endplates
caudal discs (MacLean et al. 2005). From these and other under diurnal loading conditions and (2) cell viability and
studies, it was concluded that for a number of animal species GAG synthesis could be maintained for 7 days. However,
and vertebrae levels, to prevent disc swelling, an effective gene expression data indicated possible catabolic changes
compressive stress of ~0.10.2 MPa is a reasonable baseline in cell activity, which was likely associated with a lack of
loading condition. cyclic loading stimulus.
Lee et al. (2006) created a bioreactor system for longer- Free-swelling studies by Gawri et al. (2011) on human
term culture using the same static compression loading intervertebral discs and Jim et al. (2011) on bovine discs
design of Ohshima and coworkers. This system successfully evaluated if the cartilaginous endplates promoted tissue
maintained cell viability for up to 7 days; although the structure and cell viability and metabolism. In both studies,
GAG synthesis rates dropped by 7080 % after only 2 days the vertebral endplates were prepared by removing bone with
of culture, the cells remained sensitive to mechanical stim- a high-speed drill until only the cartilage endplates remained.
ulation. Lee and coworkers concluded that this drop in Disc explants with cartilage endplates in free-swelling con-
GAG synthesis was the result of static loading, since in vivo ditions were cultured for up to 4 months in medium contain-
experiments demonstrated decreased cell metabolism with ing high or low glucose levels with high or low concentrations
loading and enhanced biosynthesis under dynamic com- of fetal bovine serum (FBS) (high glucose = 4.5 g/l; low glu-
pression loading (Maclean et al. 2004; MacLean et al. cose = 1 g/l; high FBS = 5 %; low FBS = 1 %). Importantly,
2003). Lee and coworkers also evaluated the possibility of cell viability of these free-swelling explants with cartilage
culturing intervertebral discs with intact endplates, which endplates was very high (>95 %) for up to 4 months for all
by anchoring the annulus fibrosus fibers and retaining nutritional culture conditions, indicating that adequate waste
nucleus pulposus, pressurization would maintain the and nutrition exchange was taking place. However, after
4 weeks in culture, the increase in disc height was about disc cell death. Several investigators have developed improved
21 %, and matrix degradation was observed to have increased, methods of dissections that allow nutrient transport while
independent of the various culture conditions. Similar to retaining the anatomic structure more completely (Gantenbein
other studies, they reported high cell viability but GAG loss et al. 2006; Gawri et al. 2011). The overall processes are simi-
in free-swelling culture conditions, suggesting that physio- lar and involve coarse dissection, sterilization and isolation of
logical axial loading may be necessary to regulate cellular the disc by dissection, thorough cleaning of the vertebral end-
metabolism and retain GAG content. plates, and eventual placement in a bioreactor system.
22.3.5.1 Dissection Procedures That Retain 22.3.5.2 Procedures for Cleaning the Endplate
Structural Integrity of the Explants The procedures described by Gantenbein et al. (2006) used
The method of removing and preparing discs from the sur- intervertebral discs obtained from experimental sheep (used
rounding tissues is an intricate multistep process. The proce- primarily for an alternate experiment) that were heparinized
dure utilizes bovine caudal intervertebral discs; these discs prior to death, making cleaning of the endplates relatively
are freely available and similar to the human lumbar disc easy. After dissection, residual blood within the endplate
(Sect. 22.3.1). It can be adapted for use with other animals, was aspirated with a Pasteur pipette and using a syringe with
other spine levels, and human specimens. As demonstrated an 18-gauge needle and a jet of phosphate-buffered saline.
by Lee et al. (2006), without thorough cleaning of blood clots Since the availability of large experimental animals is lim-
within the endplates, there is inhibition of nutrient and waste ited and obtaining only the tail can be costly both economi-
transport and nucleus pulposus cell death. In contrast to the cally and ethically, it was necessary to develop methods for
earlier techniques that serve to completely dissect the disc cleaning vertebral endplates that were effective for animals
from the vertebrae using a straight edge razor or cutting tool that were not heparinized prior to death, thus creating the
to insure straight, parallel cuts (Korecki et al. 2007), the cur- possibility of using bovine tails which are easily obtained
rent procedure maintains the cartilaginous and/or vertebral from the abattoir. Different cleaning procedures are available
endplates as part of the disc organ system. This technique cre- to clean the vertebral capillaries and enable diffusion through
ates a more anatomically complete structure; it inhibits aber- the endplates. Our protocol for removal of clotted blood is
rant swelling, maintains collagen connectivity, and reduces described in Box 22.2.
Box 22.2 Rened Dissection Procedures for Isolation of

Disc Explants with Vertebral Endplates A high precision histological band saw (e.g., Exakt
The whole tail should be disinfected before dissection. Apparatebau, GmbH, Norderstedt, Germany)
To access the discs, soft tissue should be removed from should be used to make two parallel transverse cuts
the caudal spine. through the vertebral endplate just proximal and
For better accessibility, the spinous and transverse pro- distal to the cartilage (12 mm adjacent to the disc;
cesses of the vertebrae can be removed. In our labora- Fig. 22.4a).
tory, we use a simple histological saw. After removal After dissection, residual tissue (Fig. 22.4b), cutting
of surrounding tissue, muscles, nerves, and blood ves- debris, and blood clots need to be removed.
sels should be removed carefully to avoid cutting into A commercially available wound debridement irriga-
the disc. tion system (Inter Pulse handpiece set; Stryker,
To make exact cutting possible, the vertebrae should Kalamazoo, MI, USA), commonly used in many surgi-
be rough-cut approximately in half to facilitate the cal procedures, can be used (Fig. 22.4c).
successive precision cutting.
a b c
Fig. 22.4 Dissection procedure for cultures with vertebral endplates. (a) Isolation of the intervertebral discs from vertebrae; (b) growth
plate removal; (c) intervertebral discs bone-covered endplate cleaning
An alternate procedure used for organ culture loading sufficient in size for bovine and human intervertebral discs
involves removal of vertebral endplates and retention of (Fig. 22.5b, Haglund et al. 2011). Another multi-chamber
just the cartilaginous portion of the plate. Instead of clean- organ culture device is being developed by our group that is
ing the residual ossified vertebral tissue, the discs are fur- capable of dynamically loading bovine coccygeal or human
ther processed by surgically ablating the vertebral bone lumbar discs in organ culture in axial compression under
with a surgical round-end bit attached to a high-speed drill, relatively high loading conditions. The device is designed
until the cartilage surface is exposed (Gawri et al. 2011; to fit in an incubator; it is hydraulically actuated, controlled
Jim et al. 2011). Intervertebral discs with cartilage end- via proportional solenoids, and capable of applying high
plates and no calcified bone can be maintained in culture forces required for large human lumbar intervertebral discs
or loaded into a bioreactor (Haglund et al. 2011). It is nota- (up to ~5.5 kN; ~2 MPa; Fig. 22.5c). An additional device
ble that for the dissections that retain the cartilage end- under development allows loading both in axial compres-
plates, the loading surfaces are not straight and parallel, so sion and in torsion so that the explants can be loaded in
loading platens that accommodate the unique curvature multiple degrees of freedom simultaneously (Fig. 22.5d;
and size of the cartilaginous EPs are required (Haglund Walser et al. 2012).
et al. 2011) and as such are a challenge for precise Overall, organ culture has evolved from the first studies,
loading. where intervertebral discs were simply placed in culture
media and allowed to free swell, to the current elegant
computer-controlled systems capable of loading several
22.4 Dynamic Compression Loading Models disc organs simultaneously, in multiple degrees of free-
dom. The major advances involved controlling swelling,
The addition to dynamic compression loading to interverte- retaining endplates to allow nutrient transport, and apply-
bral disc cultures with retained bone-covered endplates pro- ing physiological loads. The next sections describe the
vided another significant technical advance in organ culture usage of these models for degenerative and regenerative
techniques (Jnger et al. 2009). This method comprised a studies.
pneumatically actuated bioreactor loading system capable of Organ culture permits the investigation of concurrent
loading and culturing whole intervertebral discs with bone changes in both structure and cellular responses in a specific
endplates and able to simultaneously apply compression and controlled environment. As these models are well con-
loads to 4 individually controlled discs. Physiological loading trolled, they can be used to determine how multiple factors
was simulated with a rest phase at force equivalent of 0.2 MPa individually and together contribute to changes in the inter-
for 8 h (representing sleep), an active phase of 0.6 MPa for the vertebral disc biology and degeneration. These organ culture
remaining 16 h (to simulate time awake), and 2 simulated models have been used to address the following scientific
bouts of exercise lasting 4 h at 0.2 Hz and 0.2 MPa (i.e., questions:
spanning from 0.4 to 1.0 MPa during the active phase). These 1. What factor(s) causes and/or contributes to pathogenesis
simulated physiological loading cycles proved beneficial to of disc degeneration?
ovine disc explant cultures as there were no changes in meta- 2. What potential therapeutic approaches can be used to pre-
bolic activity observed after 3 weeks of culture (Jnger et al. vent and/or slow the degeneration process?
2009). The design specifications of the chambers and pneu- 3. Can human intervertebral discs be maintained in culture?
matic actuator were optimized for culturing relatively small Research on human tissues addresses issues most directly
discs with a diameter of ~15 mm (Fig. 22.5a). relevant to health.
Haglund et al. (2011) developed an organ culture load-
ing device for culturing intervertebral discs with cartilage
endplates (based on the dissection technique described by 22.5 Insights into Mechanism of
Gawri et al. (2011)) for 4 weeks. During the first week, the Degeneration Using Disc Organ Culture
discs were cultured without external load, followed by
2-day static load (0.1 MPa) and 19 days with a dynamic Intervertebral disc degeneration has been defined as an
loading regimen consisting of 0.1 MPa static loading with aberrant, cell-mediated response to progressive structural
2 periods of sinusoidal dynamic loading between 0.1 and failure (Adams and Roughley 2006) with multiple fac-
0.3 MPa load at 0.1 Hz for 2 h separated by a 6-h 0.1 MPa tors thought to be involved in its initiation and progres-
static load period (Haglund et al. 2011). In this model, cell sion. The role of these factors is reviewed in other
viability was maintained for up to 4 weeks, and Western chapters of this book. The majority of current degenera-
blot analysis revealed that the dynamic loading regimen tion models are induced using mechanical, structural,
preserved intact aggrecan and prevented the increase in nutritional, or inflammatory challenges. This work is pro-
degraded aggrecan that was observed under free-swelling viding an improved understanding of individual and
culture conditions. The culture chambers were designed to interacting factors that cause or lead to degenerative disc
accommodate discs between 20 and 60 mm, making them disease.
a b
c d
Fig. 22.5 Computer-controlled organ culture loading devices. (a) bovine or human intervertebral discs (Ben Walter and James Iatridis).
Pneumatically actuated multi-chamber system for axially loading inter- (d) Pneumatically actuated device for axial and torsional loading on
vertebral discs of medium size (up to 15 mm in diameter) (system bovine discs (With kind permission of Benjamin Gantenbein-Ritter;
described in Jnger et al. 2009). (b) Pneumatically actuated multi- system described in Walser et al. 2012). All of these loading devices
chamber system for axially loading of bovine or human discs (Figure incorporate mechanical parts including a displacement sensor, a load
modified from Haglund et al. 2011). (c) Hydraulically actuated multi- sensor, and a pneumatic or hydraulic actuator
chamber system for applying high magnitudes of axial loading on
22.5.1 Altered Media Conditions models have failed to show that limited nutrition induces
degeneration (Hutton et al. 2004; Krebs et al. 2007), possibly
When compared with in vivo studies that are directed at due to the lack of precise control over chemical or nutritional
assessing nutrient transport and uptake by the tissues of the boundary conditions. A whole-organ culture study in which
disc, organ culture models provide a much simpler approach there was excellent control over nutrient levels was able to
to elucidating the importance of this confounding factor. The address this controversy (Jnger et al. 2009). Jnger et al.
relevance of nutrient transport as a predisposing factor was (2009) cultured intervertebral discs under limiting glucose
demonstrated using nucleus pulposus cell culture (Bibby conditions and found that low glucose concentrations (2 g/l
et al. 2005; Horner and Urban 2001). However, some in vivo vs. 4.5 g/l in normal media) reduced the viability of ovine
caudal discs by 4050 %. However, they also noted that the high-frequency dynamic axial compression) and 0.2 Hz
remaining viable cells exhibited no changes in cellular (considered a beneficial moderate dynamic compression) on
metabolism, suggesting that limited nutrition alone was not intervertebral disc explants. These explants were also cul-
sufficient to initiate a degenerative cascade. tured in media with sufficient (4.5 g/l) or limited (2 g/l)
Pro-inflammatory cytokines, specifically tumor necrosis glucose concentrations (to simulate reduced nutrition result-
factor-a (TNFa) and interleukin-1-b (IL-1b), are associated ing from endplate sclerosis and calcification). Both high-
with increased degeneration (Le Maitre et al. 2005). Since frequency loading and limited glucose nutrition resulted in
both cytokines are known to induce catabolic shifts in gene increased cell death, and the combination of high frequency
expression and influence the activation of proteases (Hoyland and limited nutrition caused an additive increase in cell death
et al. 2008; Seguin et al. 2005), these pro-inflammatory and increased MMP13 gene expression. These findings sug-
cytokines have served as an attractive target for studies of the gested that high-frequency loading may increase cell metab-
degenerative process. Bovine discs that were treated with olism without sufficient time to allow transport of nutrients
media containing 200 ng/ml of TNFa for 7 days experienced to the disc cells. As a result, there would be an accumulation
dramatic losses of aggrecan and altered gene expression that of acidic metabolites within the explants, and the reduced pH
continued following cytokine removal (Walter et al. 2012). of the microenvironment would accelerate cell death. This
Thus, from a mechanistic viewpoint, the effects of injury phenomenon was further enhanced when disc explants were
may be directly linked to upregulation of this and other loaded under nutrient-depleted conditions. The shift in
cytokines. mRNA expression suggested that if cultured for longer dura-
tions, more substantial structural changes would be evident
(Illien-Jnger et al. 2010).
22.5.2 Mechanical and Nutritional Challenges The response of intervertebral disc explants to short-term
repetitive cyclic torsion was evaluated using bovine caudal
Mechanical loading is known to alter metabolic activity and tissue in organ culture. It was found that small angles (2)
matrix integrity (Iatridis et al. 2006; Walsh et al. 2000), and of torsion increased nucleus pulposus cell viability, whereas
hence a threshold of healthy loading has been described in apoptosis was observed at larger torsion angles (5) (Chan
which immobilization (or underloading) and overloading et al. 2011). These results are consistent with the in vivo rat
can both lead to deleterious or degenerative changes (Stokes tail studies of Barbir et al. (2011) who found that high mag-
and Iatridis 2004). Immobilization reduces anabolic metabo- nitudes of cyclic torsion increased pro-inflammatory cytokine
lism, and overloading can induce micro-injuries (or even production. While these tail model studies applied torsional
larger injuries) which can accumulate and contribute to the magnitudes well above the physiological range for human
degenerative process. lumbar discs of 2 (Adams and Hutton 1981), it is also
Korecki et al. (2008) found that when bovine interverte- known that tail discs have a larger physiological range of
bral discs (without endplates) were loaded in compression motion than those in the lumbar spine (Elliott and Sarver
at 1 Hz, even up to 2.5 MPa, there were no detrimental 2004). In contrast, low and moderate dynamic axial com-
effects on cellular metabolism or disc structure and there pression loading modes such as torsion, that produce high
was a dose-dependent increase in sulfate incorporation and fiber strains with minimal pressurization, have the capacity
collagen I and II gene expression in the annulus fibrosus to increase cell death, promote the expression of pro-
(Korecki et al. 2008). This finding indicated that moderate inflammatory cytokines, and enhance catabolism, without
dynamic compression increases disc cell metabolism and provoking an anabolic response.
can thus be considered as healthy loading conditions. These Noteworthy, complex loading involves multiple loading
observations were consistent with mechanobiology studies modes, for example, compression with bending. When com-
using in vivo rat tail models that showed that only small plex loading is excessive, e.g., hyperflexion, it can cause her-
amounts of damage accumulated when loading was applied niation in the motion segment (Adams and Hutton 1982). In
at excessive magnitudes and duty cycles in dynamic com- the bovine intervertebral disc, complex loading of 0.2 MPa
pression (Maclean et al. 2004; Wuertz et al. 2009). Taken compression at an applied wedge angle of 15 was shown to
together, these investigations support the concept that end- induce a rapid increase in apoptosis, a raised level of cata-
plates are the point of greatest weakness under axial com- bolic and pro-inflammatory gene expression, and disruption
pression in the motion segment, as was demonstrated of the annulus fibrosus (Walter et al. 2011). Although verte-
previously with in vitro biomechanical testing (Adams et al. bral endplates were absent in this model making it a hyper-
2000). physiological simulation, the results indicated that annulus
To evaluate how biomechanical loading interacts with fibrosus structural disruption can rapidly induce apoptosis
reduced nutrition, Illien-Jnger et al. (2010) examined the and inflammation and promote a strong catabolic response
effects of high-frequency loading at 10 Hz (extreme (Fig. 22.6a).
a b
Control
Wedgel
Convex Concave
c d
Trypsin Papain Control

Outer AF Inner AF
14G needle 25G needle
Fig. 22.6 Intervertebral discs organ culture injury models. (a) Complex degenerative changes (Roberts et al. 2008); (d) needle injury simulation
loading of 0.2 MPa axial compression applied at a wedged angle of 15 in organ culture (Figure modified from Korecki et al. 2008); (e) surgical
(Figure modified from Walter et al. 2011); (b) vertebral endplates frac- micro-discectomy simulation (Illien-Jnger et al. 2012a)
ture (Haschtmann et al. 2008a, b); (c) enzyme digestion to induce
22.5.3 Surgical and Chemical Challenges (necrosis) and apoptosis. Continued expression of pro-
apoptotic proteins FasL and TNFa was observed in the
Endplate fractures after spinal trauma can induce rapid nucleus pulposus, cytokines that might be expected to cause
degenerative changes in the intervertebral disc. To investi- apoptosis. As cell death is an early response of disc degen-
gate posttraumatic degenerative changes, Haschtmann et al. eration, this system could represent a reproducible model for
(2008b) developed an endplate trauma model based on their investigating degenerative processes after trauma.
established rabbit organ culture system. Endplate burst frac- Since one sign of degeneration is loss of pressurization of
tures (Fig. 22.6b) were induced in the disc with a custom- the nucleus pulposus and overall decrease in structural
made fracture device and cultured for 9 days. Within the first integrity, some studies have directly altered the disc structure
3 days, the injuries resulted in immediate cell damage by treatment with proteases such as trypsin and papain.
Bovine intervertebral discs that were injected with trypsin assess the effects of an intervention. In contrast, for studies
were found to have a reduced swelling pressure in the nucleus of growth factor therapy, screening can be performed in a
and a corresponding loss of intact GAGs (Fig. 22.6c; Jim small animal model, but the potential clinical efficacy may
et al. 2011; Roberts et al. 2008). Trypsin had no significant require additional testing using a large animal. A critical
effect on cell viability, but significantly altered the disc struc- scientific question has focused on long-term evaluation of
ture (Jim et al. 2011); however, the effect on cellular pheno- biosynthesis rates, cell survival, and mechanobiology. For
type remains uninvestigated. These models have the advantage this purpose, rigorous dissection and loading techniques are
of quickly creating a structurally degenerate disc and suggest required that enhance transport as well as simulated physio-
that they could be used to evaluate or screen therapeutic logical loading.
agents. However, this technique does not reflect the typical
pathophysiological history of disc degeneration.
Needle puncture is a mild surgical intervention that is 22.6.1 Screening of Potential
important therapeutically and in the context of disc degen- Therapeutic Agents
eration warrants investigation (Fig. 22.6d). Needle injection
simulates discography, a common diagnostic procedure used The activities of cells of the intervertebral disc are regulated
to inform surgical treatments for low back pain. Needle by the interplay of cytokines, proteolytic enzymes, enzyme
injection through the annulus fibrosus is also known to inhibitors, and growth factors (Masuda and An 2004) (see
induce degenerative changes in several animal models also Chaps. 8 and 25). Recent in vitro and in vivo studies
(Masuda et al. 2005). It was shown that bovine caudal discs have generated methods to upregulate the synthesis of the
that were punctured with a needle prior to culturing experi- major structural proteins like aggrecan or to downregulate
enced a rapid change in mechanical function, assessed via catabolic proteins that are induced by pro-inflammatory
measurement of the dynamic modulus, and there was local- cytokines, such as interleukin-1 (IL-1) or tumor necrosis
ized cell death at the injection site (Korecki et al. 2008). This factor-a (TNFa) (Masuda and An 2004; Seguin et al. 2005).
finding suggested that the degenerative changes observed Some studies have investigated these effects in vitro using
in vivo can be associated with compromised mechanical organ culture. In 2006, Risbud et al. cultured rat lumbar
behavior and that following large or small injuries, it is criti- intervertebral discs with endplates for 1 week in media sup-
cal that function is restored to the intervertebral disc. plemented with TGFb-1 or TGFb-3. They found that by
Micro-discectomies can provide back and leg pain relief by elevating the levels of activated ERK1/2, which in turn pro-
removing herniated fragments of tissue that impact nerve roots moted the expression of receptors for TGFb-1 and TGFb-1I,
in the spinal canal of back pain patients. While micro-discec- TGFb-3 maintained the in vivo phenotype and biological
tomy surgery can provide immediate benefits, the treated disc properties of the disc tissues.
continues to degenerate. Post-discectomy disc degeneration Using mouse functional spine units, Wang et al. (2011)
often occurs rapidly and leads to significant back pain and studied the effect of an anesthetic commonly used for the
occasionally to re-herniation. So as to develop repair strate- management of back pain. Discs of functional spinal units
gies, we are currently investigating the direct response of the were treated with the anesthetic for up to 4 weeks in organ
cells remaining after discectomy using a bovine organ culture culture. Compared to the control group, there was dramatic
system (Fig. 22.6e; Illien-Jnger et al. 2012a). decrease in cell viability and matrix protein synthesis, dem-
Several organ culture models have been used to study the onstrating that the anesthetic, at clinical relevant doses, is
pathology of intervertebral disc degeneration. The systems toxic to tissues of the intervertebral disc.
permit analysis of multiple causes of degeneration, including
intrinsic factors like altered nutrition and inflammation and
extrinsic initiators represented by different loading patterns. 22.7 Stem Cells in the Intervertebral Disc
Based on these models, novel regeneration therapies are
being developed, as described in the next section. Adult mesenchymal stem cells (MSCs) are a promising cell
source for tissue repair and regeneration therapies. While these
cells are found in various tissues including skeletal muscle,
22.6 Mechanisms of Repair and Organ periosteum, and synovium, the most well-studied sources of
Culture Models of Disc Regeneration MSCs are from bone marrow and adipose. One practical
advantage of using MSCs is that they proliferate rapidly, while
Disc organ culture models have been used to assess the fea- retaining their multi-lineage potential to commit to a number
sibility of potential therapeutic treatments, with the aim of of phenotypes including bone, cartilage, muscle, ligament,
maintaining the extracellular matrix and/or promoting its tendon, adipose, and stromal tissues (Pittenger et al. 1999). In
nutritional status. To investigate the utility of drugs, mouse vitro and in vivo studies have indicated the potential of MSCs
disc models can be a useful and relatively simple means to to differentiate towards a phenotype similar to nucleus
pulposus cells (Risbud et al. 2004; Steck et al. 2005). In addi- 10 Hz high-frequency dynamic compression. Bovine inter-
tion, MSCs are known to have immunosuppressive properties vertebral disc was pre-cultured either under simulated physi-
and are regarded as nonimmunogenic, rendering them particu- ological or degenerating conditions. At 812 days,
larly attractive for tissue engineering and cell and gene thera- fluorescently labeled MSCs were added to the culture, and
pies (Di Nicola et al. 2002). This general topic is discussed discs were further cultured under static conditions. After
further in Chaps. 23 and 24. 14 days, migration of MSCs into the discs was visualized
using fluorescent microscopy (Illien-Jnger et al. 2012b).
The results of this study confirm that organ culture is a useful
22.7.1 Fate of Injected Stem Cells paradigm in which to investigate MSC homing and migra-
tion into the disc as well as offering a promising tool for
In 2009, Le Maitre et al. presented a model in which labeled potential regenerative treatments.
human MSCs were injected into healthy bovine nucleus pul-
posus explants and cultured for up to 4 weeks. MSCs were
identified in all tissue samples, and viability was maintained 22.8 Biomaterials for Intervertebral
throughout the whole culture period. Immunohistochemical Disc Repair
staining for Sox-9, aggrecan, and collagen II revealed a spon-
taneous expression of nucleus pulposus markers, whereas no Several types of biomaterials and gels have been developed
collagen I or X or alizarin red staining was detected at any to support the regeneration of the herniated or degenerated
time point. The authors concluded that cellular hypertrophy intervertebral disc (Grad et al. 2010). The ideal biomaterial
and calcification were not induced, i.e., the MSCs differenti- should resemble the composition of the native disc in that it
ated into nucleus pulposus cells rather than annulus fibrosus should provide a 3D matrix to retain cells or drugs and par-
cells or hypertrophic chondrocytes (Le Maitre et al. 2009). tially replace the loss of disc tissue, while also providing
Disc organ culture is therefore a promising tool for future mechanical support.
studies to evaluate the fate of injected cells. The technology The screening of biomaterials in organ culture provides a
would help define optimal conditions for cell survival and strategy that complements in vivo implantation studies in ani-
phenotypic maintenance following injection. For further mal models. Recent in vitro and in vivo studies have shown
information on the use of MSC and stem cells in repair of the that hydrogels consisting of collagen and hyaluronan, a com-
discal tissue, see Chaps. 23 and 24. position similar to the native components of the nucleus pul-
posus extracellular matrix, are suitable biomaterial substitutes
(Collin et al. 2011; Sakai et al. 2006). Peroglio et al. (2011)
22.7.2 Homing of MSCs designed an injectable thermoreversible hyaluronan-based
hydrogel (HA-pNIPAM) as a nucleus pulposus cell carrier. In
Besides their potential to differentiate into a nucleus pulpo- the first step they compared bovine nucleus pulposus cells
sus-like phenotype, MSCs also have the ability to engraft seeded in HA-pNIPAM and pNIPAM alone. They showed
into different tissues after site-directed and systematic that cell morphology and the GAG synthesis rate were similar
administration, in particular after injury or disease. The in both hydrogels. Gene expression analyses revealed that
direct migration into areas of tissue damage was demon- after expansion, the hydrogels induced a redifferentiation
strated in bone defects (Kitaori et al. 2009), myocardial towards the nucleus pulposus phenotype. Compared to alg-
infarction (Barbash et al. 2003; Kawada et al. 2004), isch- inate beads, HA-pNIPAM promoted hyaluronan synthase 1
emic cerebral injury (Ji et al. 2004), nephropathy (Hauger gene expression. In a bovine disc culture, they showed that
et al. 2006), pulmonary fibrosis, and wound healing when cells were suspended in the thermoreversible hydrogel,
(Mackenzie and Flake 2001). It has been suggested that they could be injected through a 22-G needle with cell viabil-
trophic factors and cytokines released from those tissues are ity above 80 % after 1 week in culture. The authors concluded
involved in this migration process. Several studies have that HA-pNIPAM is a suitable, injectable carrier that is capa-
reported the functional expression of various chemokine ble of maintaining the nucleus pulposus phenotype and pro-
receptors on MSCs (Honczarenko et al. 2006; Nasef et al. moting disc matrix generation (Peroglio et al. 2011).
2007), and inflammatory cytokines have been shown to trig-
ger MSC chemotactic migration through extracellular matrix
structures (Ries et al. 2007). Recently, it was demonstrated 22.9 Human Organ Culture Models
that migration of MSCs into injured discs in organ culture
was significantly higher than migration into healthy tissue Human intervertebral disc organ culture models remain at
(Illien-Jnger et al. 2012b). In this model, disc degeneration an early stage of development. As with all research involv-
was induced using needle puncture injury combined with ing human tissue, sample procurement and usage must
follow federal guidelines including removal of any identi- 22.10.1 Age Changes and Use of Human Tissues
fying information and receive the approval of relevant
ethical committees. Whole intervertebral discs with end- The structure and composition of the human intervertebral
plates are currently obtained from cadavers following disc changes with age, with older discs typically showing
autopsy or from organ donors usually 1230 h postmor- more signs of degeneration. When large animal discs are used
tem. Due to the avascular nature of the tissues, cadaveric to model the human, as they are obtained from abattoirs, they
discs remain viable for 30 h postmortem with cell viabil- are typically younger and healthier than the clinically relevant
ity >90 %. human intervertebral disc. Not surprisingly, the healthy
After obtaining specimens, it is important to assess the human disc is structurally similar to the bovine and ovine
degenerative grade of the tissue. This can be achieved using (Fig. 22.1), but compared to the animal spine, the moderately
X-ray and MRI analysis and graded using the modified degenerate human discs display a more fibrous nucleus pul-
Thompson and Pfirrmann classification (Pfirrmann et al. posus, the tissue appears brownish in color, and the demarca-
2001; Thompson et al. 1990). While it is not known whether tion between the nucleus and the annulus is less distinct
all grades can be used, the clinical relevance of culturing an (Fig. 22.7). With progressing degeneration, the nucleus pul-
end-stage degenerated disc is arguable. Nevertheless, posus becomes more fibrous and it is hardly distinguishable
depending on the question being assessed, the development from the annulus; in some severely degenerated discs, there is
of criteria for including human discs in culturing experi- also vascular ingrowth (Fig. 22.7c). In the light of these path-
ments is helpful and requires degenerative grade assess- ological differences, it is difficult to extrapolate information
ment. Thorough cleaning of the bone endplate is absolutely generated from discs of young healthy animals to answer
necessary to enhance transport of nutrients and waste questions concerning severely degenerate human discs.
products. Large animal organ culture models of disc degeneration
Organ donor intervertebral disc explants with cartilage differ substantially from the human. The intervertebral spec-
endplates have been cultured in free-swelling conditions, for imens are from the young, meaning that they did not develop,
up to 4 months with high cell viability (Gawri et al. 2011). grow, remodel, accumulate injury, undergo endplate
This method has been further improved for culturing human calcification, and respond to pro-inflammatory cytokines. In
discs with cartilage endplates under diurnal (0.1 MPa/0.3 MPa) other words, animal intervertebral discs are generally
and sinusoidal (0.2 Hz) loads (Fig. 22.5b; Haglund et al. healthier, and most suitable for studies aimed at under-
2011). A high-speed grinding drill is used to completely standing the biology of the young and healthy human inter-
remove vertebral endplates and isolate whole discs with car- vertebral tissues (up to ~40 years of age), but are probably
tilage endplates. In this case, bioreactors must be of sufficient not reliable models for aging studies (Demers et al. 2004).
size to accommodate human lumbar intervertebral discs. This topic is considered in detail in Chap. 18.
These discs have an average diameter of approx. 3435 mm Accordingly, because of their high reproducibility, uni-
in length and 4147 mm in width (L1L5; Panjabi et al. 1992). formity, and availability, animal disc models have consider-
In turn, the bioreactors must accommodate a large volume of able advantages when used for screening therapeutic agents
media and to simulate daily loading to be able to withstand or addressing mechanistic questions regarding degenerative
forces of 1,500 N (Adams and Hutton 1983). For dynamic changes and responsiveness to injurious and environmental
compression, computer-controlled loading devices have conditions. Human discs are the gold standard when investi-
been developed as shown in Fig. 22.5. The bioreactor cham- gating aging and degeneration changes. However, inconsis-
bers are composed of a bottom chamber and a top loading tencies between human disc specimens are high, and the
piston with sintered porous loading platens on both ends. limited number of available discs exacerbates this variabil-
Tubing underneath/above the lower/upper platen provides ity. Nevertheless, human rather than animal organ culture
equal distribution of media through and around the disc models are most valuable for evaluating disc treatment
(Fig. 22.3). options.
22.10 Limitations of Organ Culture Models 22.10.2 Measurements of Predictors

of Painful Conditions
Whole disc organ culture models have become an essential
tool that is available to address important intervertebral Low back pain is often associated with disc degeneration
disc research questions. Some intrinsic drawbacks of these (Biering-Sorensen et al. 1985) (see also Chaps. 16 and 19).
model systems impact the types of questions that can be It is difficult, if not impossible, to directly measure pain in
addressed. The largest limitations relate to questions of age a whole-organ culture model due to its subjective nature.
and pain. However, while direct clinical measures of pain cannot be
a b c
Fig. 22.7 Human healthy and degenerate intervertebral discs. (a) Healthy intervertebral discs (grade I); (b) moderate degeneration (grade III); (c)
severe degeneration (grade IV). Grading is based on the Thompson degeneration scale
assessed in vitro, it is possible to measure the secretion or inhibit the high swelling propensity of the matrix and enable
expression of selected neurotrophic agents and growth nutrient transport. The high repeatability, uniformity, simi-
factors (Jung et al. 2011; Purmessur et al. 2008). Assessing lar aspect ratios, and metabolic rates of bovine or ovine cau-
these factors may indicate potential induction or relief of dal discs make them valuable tissues for disc organ culture
pain; however, the actual pain cannot be measured. to be used for screening therapeutic agents that are in devel-
Interestingly, most patients who have degenerative discs opment and for answering mechanistic questions concern-
are asymptomatic. What makes the degenerative disc ing the pathogenesis of degenerative events and progress in
painful in one patient and asymptomatic in another response to physicochemical challenges. These models rep-
remains unknown. Differentiating the normal aging pro- resent the initial battleground for exploring therapies that
cess from pathologic disc degeneration also remains a will inhibit catabolic and promote anabolic processes in the
very challenging if not impossible endeavor at this time. disc. Experimental variation in human discs is high, and
Further details on neurogenic pain are discussed in specimen procurement is challenging so that human organ
Chap. 16. cultures are best suited for evaluating refined treatment
options.
While animal organ culture models are valuable tools for
22.11 Conclusions the evaluation of therapeutic interventions, almost all models
require degeneration to be induced before a repair strategy
Disc degeneration is a widely investigated disorder strongly can be implemented. The use of human intervertebral discs
associated with low back pain. The highly structured extra- provides a model with naturally occurring degeneration and
cellular matrix and harsh physicochemical microenviron- can therefore serve as a useful tool in validating therapeutic
ment of human intervertebral tissues create a unique niche interventions for repair of early- to moderate-stage disc
with low cellularity and metabolic activity. This niche is not degeneration. Due to slow disc cell metabolism, long-term
well simulated with current cell culture techniques or small cultures will be needed to measure a significant change in
animal models. The development of organ culture models disc matrix composition during repair. Thus, organ cultures
has the capacity to better simulate the important 3D con- provide an important link between conventional in vitro and
nectivity and other niche characteristics found in the human in vivo technologies and offer a means of significantly reduc-
disc with precise control over the mechanical and chemical ing the use of animal models.
boundary conditions. Intervertebral disc organ culture tech-
niques have advanced from simple free-swelling tissue
models to computer-controlled and elegantly designed mul- 22.12 Summary of Critical Concepts
tiaxis and multi-chamber systems with the capacity to main- Discussed in the Chapter
tain disc structure, control nutritional factors, and simulate
physiological or damaging loads. Successful organ culture The native cellular niche of disc cells can be maintained
experiments can now be designed to focus on critical in organ culture providing an important advantage over
scientific questions, having developed procedures that the more traditional 2D and 3D cell culture studies.
Intervertebral disc organ cultures allow precise control biomechanics, using a rat tail model. Spine (Phila Pa 1976)
over mechanical and chemical boundary conditions, facil- 36(8):607614
Bibby SR, Jones DA, Ripley RM, Urban JP (2005) Metabolism of the
itating more targeted mechanistic hypothesis testing com- intervertebral disc: effects of low levels of oxygen, glucose, and pH
pared with in vivo models. on rates of energy metabolism of bovine nucleus pulposus cells.
Intervertebral disc culture under free-swelling conditions Spine (Phila Pa 1976) 30(5):487496
results in GAG loss, a decrease in anabolic activity, and Biering-Sorensen F, Hansen FR, Schroll M, Runeborg O (1985) The
relation of spinal x-ray to low-back pain and physical activity
an increase in the expression of catabolic genes. Retention among 60-year-old men and women. Spine (Phila Pa 1976)
of the endplate in culture maintains collagen connectivity, 10(5):445451
reduces disc cell death near the cut edge of the disc, and Chan SC, Ferguson SJ, Wuertz K, Gantenbein-Ritter B (2011)
inhibits swelling. Biological response of the intervertebral disc to repetitive short term
cyclic torsion. Spine (Phila Pa 1976) 36:20212030
Axial loading is required to fully inhibit swelling, pre- Chiba K, Andersson GB, Masuda K, Momohara S, Williams JM,
serve the complex extracellular environment of the inter- Thonar EJ (1998) A new culture system to study the metabolism of
vertebral disc, and simulate diurnal loading and daily the intervertebral disc in vitro. Spine (Phila Pa 1976) 23(17):1821
activities that are necessary to maintain physiological bio- 1827; discussion 28
Collin EC, Grad S, Zeugolis DI, Vinatier CS, Clouet JR, Guicheux JJ,
synthesis rates of disc cells. Weiss P, Alini M, Pandit AS (2011) An injectable vehicle for nucleus
Development of animal models enhances the study of pulposus cell-based therapy. Biomaterials 32(11):28622870
agents that promote degeneration and promote tissue Demers CN, Antoniou J, Mwale F (2004) Value and limitations of using
regeneration models in a defined and reproducible man- the bovine tail as a model for the human lumbar spine. Spine (Phila
Pa 1976) 29(24):27932799
ner with minimal donor variability. Di Nicola M, Carlo-Stella C, Magni M, Milanesi M, Longoni PD,
Human discs can be maintained in organ culture and pro- Matteucci P, Grisanti S, Gianni AM (2002) Human bone marrow
vide an important preclinical technique for screening stromal cells suppress T-lymphocyte proliferation induced by
therapies in a way that complements and minimizes cellular or nonspecific mitogenic stimuli. Blood 99(10):
38383843
in vivo animal testing. Elliott DM, Sarver JJ (2004) Young investigator award winner: valida-
tion of the mouse and rat disc as mechanical models of the human
Acknowledgments The research reported in this chapter was sup- lumbar disc. Spine (Phila Pa 1976) 29(7):713722
ported by the National Institute of Arthritis and Musculoskeletal and Gantenbein B, Grunhagen T, Lee CR, van Donkelaar CC, Alini M, Ito
Skin Diseases of the National Institutes of Health under Award Numbers K (2006) An in vitro organ culturing system for intervertebral disc
R01AR057397 and R01AR051146. The authors would like to thank explants with vertebral endplates: a feasibility study with ovine cau-
Marianna Peroglio, Benjamin Gantenbein-Ritter, and Jochen Walser. dal discs. Spine (Phila Pa 1976) 31(23):26652673
Gawri R, Mwale F, Ouellet J, Roughley P, Steffen T, Antoniou J,
Haglund L (2011) Development of an organ culture system for long
term survival of the intact human intervertebral disc. Spine (Phila Pa
References 1976) 36:18351842
Grad S, Mauro A, Eglin D, Sakai D, Mochida J, Mahor S, Collin E,
Adams MA, Hutton WC (1981) The relevance of torsion to the mechan- Dash B, Pandit A (2010) Cells and biomaterials for intervertebral
ical derangement of the lumbar spine. Spine (Phila Pa 1976) disc regeneration. In: Athanasiou KA (ed) Cell therapy for nucleus
6(3):241248 pulposus regeneration. Morgan & Claypool, Davis, pp 139
Adams MA, Hutton WC (1982) Prolapsed intervertebral disc. Haglund L, Moir J, Beckman L, Mulligan KR, Jim B, Ouellet JA,
A hyperflexion injury 1981 Volvo Award in basic science. Spine Roughley P, Steffen T (2011) Development of a bioreactor for axi-
(Phila Pa 1976) 7(3):184191 ally loaded intervertebral disc organ culture. Tissue Eng Part C
Adams MA, Hutton WC (1983) The effect of fatigue on the lumbar Methods 17:10111019
intervertebral disc. J Bone Joint Surg Br 65(2):199203 Haschtmann D, Stoyanov JV, Ettinger L, Nolte LP, Ferguson SJ (2006a)
Adams MA, Roughley PJ (2006) What is intervertebral disc degenera- Establishment of a novel intervertebral disc/endplate culture model:
tion, and what causes it? Spine (Phila Pa 1976) 31(18):21512161 analysis of an ex vivo in vitro whole-organ rabbit culture system.
Adams MA, Freeman BJ, Morrison HP, Nelson IW, Dolan P (2000) Spine (Phila Pa 1976) 31(25):29182925
Mechanical initiation of intervertebral disc degeneration. Spine Haschtmann D, Stoyanov JV, Ferguson SJ (2006b) Influence of diurnal
(Phila Pa 1976) 25(13):16251636 hyperosmotic loading on the metabolism and matrix gene expres-
Aguiar DJ, Johnson SL, Oegema TR (1999) Notochordal cells interact sion of a whole-organ intervertebral disc model. J Orthop Res
with nucleus pulposus cells: regulation of proteoglycan synthesis. 24(10):19571966
Exp Cell Res 246(1):129137 Haschtmann D, Ferguson SJ, Stoyanov JV (2008a) Apoptosis and gene
Alini M, Eisenstein SM, Ito K, Little C, Kettler AA, Masuda K, Melrose expression of collagenases but not gelatinases in rabbit disc frag-
J, Ralphs J, Stokes I, Wilke HJ (2008) Are animal models useful for ment cultures. J Neurosurg Spine 8(6):552560
studying human disc disorders/degeneration? Eur Spine J 17(1):219 Haschtmann D, Stoyanov JV, Gedet P, Ferguson SJ (2008b) Vertebral
Barbash IM, Chouraqui P, Baron J, Feinberg MS, Etzion S, Tessone A, endplate trauma induces disc cell apoptosis and promotes organ
Miller L, Guetta E, Zipori D, Kedes LH, Kloner RA, Leor J (2003) degeneration in vitro. Eur Spine J 17(2):289299
Systemic delivery of bone marrow-derived mesenchymal stem cells Hauger O, Frost EE, van Heeswijk R, Deminiere C, Xue R, Delmas Y,
to the infarcted myocardium: feasibility, cell migration, and body Combe C, Moonen CT, Grenier N, Bulte JW (2006) MR evaluation
distribution. Circulation 108(7):863868 of the glomerular homing of magnetically labeled mesenchymal
Barbir A, Godburn KE, Michalek AJ, Lai A, Monsey RD, Iatridis JC stem cells in a rat model of nephropathy. Radiology 238(1):
(2011) Effects of torsion on intervertebral disc gene expression and 200210
Holm S, Maroudas A, Urban JP, Selstam G, Nachemson A (1981) Korecki CL, MacLean JJ, Iatridis JC (2007) Characterization of an
Nutrition of the intervertebral disc: solute transport and metabolism. in vitro intervertebral disc organ culture system. Eur Spine J
Connect Tissue Res 8(2):101119 16(7):10291037
Honczarenko M, Le Y, Swierkowski M, Ghiran I, Glodek AM, Korecki CL, MacLean JJ, Iatridis JC (2008) Dynamic compression
Silberstein LE (2006) Human bone marrow stromal cells express a effects on intervertebral disc mechanics and biology. Spine (Phila
distinct set of biologically functional chemokine receptors. Stem Pa 1976) 33(13):14031409
Cells 24(4):10301041 Krebs J, Aebli N, Goss BG, Wilson K, Williams R, Ferguson SJ (2007)
Horner HA, Urban JP (2001) 2001 Volvo Award Winner in Basic Cardiovascular changes after pulmonary cement embolism: an
Science Studies: effect of nutrient supply on the viability of cells experimental study in sheep. AJNR Am J Neuroradiol.
from the nucleus pulposus of the intervertebral disc. Spine (Phila Pa 28(6):10461050
1976) 26(23):25432549 Le Maitre CL, Freemont AJ, Hoyland JA (2005) The role of interleu-
Hoyland JA, Le Maitre C, Freemont AJ (2008) Investigation of the role kin-1 in the pathogenesis of human intervertebral disc degeneration.
of IL-1 and TNF in matrix degradation in the intervertebral disc. Arthritis Res Ther 7(4):R732R745
Rheumatology (Oxford) 47(6):809814 Le Maitre CL, Baird P, Freemont AJ, Hoyland JA (2009) An in vitro
Hunter CJ, Matyas JR, Duncan NA (2004) Cytomorphology of noto- study investigating the survival and phenotype of mesenchymal
chordal and chondrocytic cells from the nucleus pulposus: a species stem cells following injection into nucleus pulposus tissue. Arthritis
comparison. J Anat 205(5):357362 Res Ther 11(1):R20
Hutton WC, Murakami H, Li J, Elmer WA, Yoon ST, Minamide A, Lee CR, Iatridis JC, Poveda L, Alini M (2006) In vitro organ culture of
Akamaru T, Tomita K (2004) The effect of blocking a nutritional the bovine intervertebral disc: effects of vertebral endplate and
pathway to the intervertebral disc in the dog model. J Spinal Disord potential for mechanobiology studies. Spine (Phila Pa 1976)
Tech. 17(1):5363 31(5):515522
Iatridis JC, MacLean JJ, Roughley PJ, Alini M (2006) Effects of Lim TH, Ramakrishnan PS, Kurriger GL, Martin JA, Stevens JW, Kim
mechanical loading on intervertebral disc metabolism in vivo. J, Mendoza SA (2006) Rat spinal motion segment in organ culture:
J Bone Joint Surg Am 88(Suppl 2):4146 a cell viability study. Spine (Phila Pa 1976) 31(12):12911297
Illien-Jnger S, Gantenbein-Ritter B, Grad S, Lezuo P, Ferguson SJ, Mackenzie TC, Flake AW (2001) Human mesenchymal stem cells per-
Alini M, Ito K (2010) The combined effects of limited nutrition and sist, demonstrate site-specific multipotential differentiation, and are
high-frequency loading on intervertebral discs with endplates. Spine present in sites of wound healing and tissue regeneration after trans-
(Phila Pa 1976) 35(19):17441752 plantation into fetal sheep. Blood Cells Mol Dis 27(3):601604
Illien-Jnger SM, Mayer JE, Walter BA, Qureshi S, Hecht AC, MacLean JJ, Lee CR, Grad S, Ito K, Alini M, Iatridis JC (2003) Effects
Iatridis JC (2012a) Intervertebral disc whole organ culture model to of immobilization and dynamic compression on intervertebral disc
investigate the effect of TGFb3 after discectomy. Trans Orthop Res cell gene expression in vivo. Spine (Phila Pa 1976) 28(10):973981
Soc 12: Abstract 1192 Maclean JJ, Lee CR, Alini M, Iatridis JC (2004) Anabolic and catabolic
Illien-Jnger S, Pattappa G, Peroglio M, Benneker LM, Stoddart MJ, mRNA levels of the intervertebral disc vary with the magnitude and
Sakai D, Mochida J, Grad S, Alini M (2012b) Homing of mesen- frequency of in vivo dynamic compression. J Orthop Res
chymal stem cells in induced degenerative intervertebral discs in a 22(6):11931200
whole organ culture system. Spine (Phila Pa 1976) 15;37(22): MacLean JJ, Lee CR, Alini M, Iatridis JC (2005) The effects of short-
18651873 term load duration on anabolic and catabolic gene expression in the
Ji JF, He BP, Dheen ST, Tay SS (2004) Interactions of chemokines and rat tail intervertebral disc. J Orthop Res 23(5):11201127
chemokine receptors mediate the migration of mesenchymal stem Maroudas A, Stockwell RA, Nachemson A, Urban J (1975) Factors
cells to the impaired site in the brain after hypoglossal nerve injury. involved in the nutrition of the human lumbar intervertebral disc: cel-
Stem Cells 22(3):415427 lularity and diffusion of glucose in vitro. J Anat 120(Pt 1):113130
Jim B, Steffen T, Moir J, Roughley P, Haglund L (2011) Development Masuda K, An HS (2004) Growth factors and the intervertebral disc.
of an intact intervertebral disc organ culture system in which degen- Spine J 4(6 Suppl):330S340S
eration can be induced as a prelude to studying repair potential. Eur Masuda K, Aota Y, Muehleman C, Imai Y, Okuma M, Thonar EJ,
Spine J 20(8):12441254 Andersson GB, An HS (2005) A novel rabbit model of mild, repro-
Jung WW, Kim HS, Shon JR, Lee M, Lee SH, Sul D, Na HS, Kim JH, ducible disc degeneration by an annulus needle puncture: correla-
Kim BJ (2011) Intervertebral disc degeneration-induced expression tion between the degree of disc injury and radiological and
of pain-related molecules: glial cell-derived neurotropic factor as a histological appearances of disc degeneration. Spine (Phila Pa 1976)
key factor. J Neurosurg Anesthesiol 23:329334 30(1):514
Jnger S, Gantenbein-Ritter B, Lezuo P, Alini M, Ferguson SJ, Ito K Nasef A, Chapel A, Mazurier C, Bouchet S, Lopez M, Mathieu N,
(2009) Effect of limited nutrition on in situ intervertebral disc cells Sensebe L, Zhang Y, Gorin NC, Thierry D, Fouillard L (2007)
under simulated-physiological loading. Spine (Phila Pa 1976) Identification of IL-10 and TGF-beta transcripts involved in the
34(12):12641271 inhibition of T-lymphocyte proliferation during cell contact with
Kandel R, Roberts S, Urban JP (2008) Tissue engineering and the inter- human mesenchymal stem cells. Gene Expr 13(45):217226
vertebral disc: the challenges. Eur Spine J 17(Suppl 4):480491 OConnell GD, Vresilovic EJ, Elliott DM (2007) Comparison of ani-
Kawada H, Fujita J, Kinjo K, Matsuzaki Y, Tsuma M, Miyatake H, mals used in disc research to human lumbar disc geometry. Spine
Muguruma Y, Tsuboi K, Itabashi Y, Ikeda Y, Ogawa S, Okano (Phila Pa 1976) 32(3):328333
H, Hotta T, Ando K, Fukuda K (2004) Nonhematopoietic mes- Oegema TR Jr, Bradford DS, Cooper KM (1979) Aggregated proteo-
enchymal stem cells can be mobilized and differentiate into glycan synthesis in organ cultures of human nucleus pulposus.
cardiomyocytes after myocardial infarction. Blood 104(12): J Biol Chem 254(21):1057910581
35813587 Oegema TR Jr, Johnson SL, Aguiar DJ, Ogilvie JW (2000) Fibronectin
Kitaori T, Ito H, Schwarz EM, Tsutsumi R, Yoshitomi H, Oishi S, and its fragments increase with degeneration in the human interver-
Nakano M, Fujii N, Nagasawa T, Nakamura T (2009) Stromal cell- tebral disc. Spine (Phila Pa 1976) 25(21):27422747
derived factor 1/CXCR4 signaling is critical for the recruitment of Ohshima H, Urban JP, Bergel DH (1995) Effect of static load on matrix
mesenchymal stem cells to the fracture site during skeletal repair in synthesis rates in the intervertebral disc measured in vitro by a new
a mouse model. Arthritis Rheum 60(3):813823 perfusion technique. J Orthop Res 13(1):2229
Oshima H, Ishihara H, Urban JP, Tsuji H (1993) The use of coccygeal pulposus cells forming nucleus pulposus-like tissue in vitro:
discs to study intervertebral disc metabolism. J Orthop Res 11(3): influence on the proliferation and proteoglycan production of
332338 HNPSV-1 cells. Biomaterials 27(3):346353
Panjabi MM, Goel V, Oxland T, Takata K, Duranceau J, Krag M, Price Sakai D, Nakai T, Mochida J, Alini M, Grad S (2009) Differential phe-
M (1992) Human lumbar vertebrae. Quantitative three-dimensional notype of intervertebral disc cells: microarray and immunohis-
anatomy. Spine (Phila Pa 1976) 17(3):299306 tochemical analysis of canine nucleus pulposus and annulus fibrosus.
Peroglio M, Grad S, Mortisen D, Sprecher CM, Illien-Jnger S, Alini Spine (Phila Pa 1976) 34(14):14481456
M, Eglin D (2011) Injectable thermoreversible hyaluronan-based Seguin CA, Pilliar RM, Roughley PJ, Kandel RA (2005) Tumor
hydrogels for nucleus pulposus cell encapsulation. Eur Spine J necrosis factor-alpha modulates matrix production and catabolism
21:S839S849 in nucleus pulposus tissue. Spine (Phila Pa 1976) 30(17):
Pfirrmann CW, Metzdorf A, Zanetti M, Hodler J, Boos N (2001) 19401948
Magnetic resonance classification of lumbar intervertebral disc Steck E, Bertram H, Abel R, Chen B, Winter A, Richter W (2005)
degeneration. Spine (Phila Pa 1976) 26(17):18731878 Induction of intervertebral disc-like cells from adult mesenchymal
Pflaster DS, Krag MH, Johnson CC, Haugh LD, Pope MH (1997) Effect stem cells. Stem Cells 23(3):403411
of test environment on intervertebral disc hydration. Spine (Phila Pa Stokes IA, Iatridis JC (2004) Mechanical conditions that accelerate
1976) 22(2):133139 intervertebral disc degeneration: overload versus immobilization.
Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca Spine (Phila Pa 1976) 29(23):27242732
JD, Moorman MA, Simonetti DW, Craig S, Marshak DR (1999) Thompson JP, Pearce RH, Schechter MT, Adams ME, Tsang IK, Bishop
Multilineage potential of adult human mesenchymal stem cells. PB (1990) Preliminary evaluation of a scheme for grading the gross
Science 284(5411):143147 morphology of the human intervertebral disc. Spine (Phila Pa 1976)
Purmessur D, Freemont AJ, Hoyland JA (2008) Expression and regula- 15(5):411415
tion of neurotrophins in the nondegenerate and degenerate human Urban JP, Maroudas A (1981) Swelling of the intervertebral disc
intervertebral disc. Arthritis Res Ther 10(4):R99 in vitro. Connect Tissue Res 9(1):110
Purmessur D, Walter BA, Roughley PJ, Laudier DM, Hecht AC, Iatridis Urban JP, Smith S, Fairbank JC (2004) Nutrition of the intervertebral
J (2013) A role for TNF in intervertebral disc degeneration: a non- disc. Spine (Phila Pa 1976) 29(23):27002709
recoverable catabolic shift. Biochem Biophys Res Commun 433(1): van Dijk B, Potier E, Ito K (2011) Culturing bovine nucleus pulposus
151156 explants by balancing medium osmolarity. Tissue Eng Part C
Ries C, Egea V, Karow M, Kolb H, Jochum M, Neth P (2007) MMP-2, Methods 17:10891096
MT1-MMP, and TIMP-2 are essential for the invasive capacity of Walser J, Ferguson SJ, Gantenbein-Ritter B (2012) Design of a mechan-
human mesenchymal stem cells: differential regulation by ical loading device to culture intact bovine caudal motional seg-
inflammatory cytokines. Blood 109(9):40554063 ments of the Spine under twisting motion. In: Davies J (ed)
Risbud MV, Izzo MW, Adams CS, Arnold WW, Hillibrand AS, Replacing animal models: a practical guide to creating and using
Vresilovic EJ, Vaccaro AR, Albert TJ, Shapiro IM (2003) An organ biomimetic alternatives. John Wiley & Co., Chichester
culture system for the study of the nucleus pulposus: description of Walsh WR, Harrison J, Loefler A, Martin T, Van Sickle D, Brown MK,
the system and evaluation of the cells. Spine (Phila Pa 1976) Sonnabend DH (2000) Mechanical and histologic evaluation of
28(24):26522658; discussion 5859 Collagraft in an ovine lumbar fusion model. Clin Orthop Relat Res
Risbud MV, Albert TJ, Guttapalli A, Vresilovic EJ, Hillibrand AS, 375:258266
Vaccaro AR, Shapiro IM (2004) Differentiation of mesenchymal Walter BA, Korecki CL, Purmessur D, Roughley PJ, Michalek AJ,
stem cells towards a nucleus pulposus-like phenotype in vitro: Iatridis JC (2011) Complex loading affects intervertebral disc
implications for cell-based transplantation therapy. Spine (Phila Pa mechanics and biology. Osteoarthritis Cartilage 19(8):
1976) 29(23):26272632 10111018
Risbud MV, Di Martino A, Guttapalli A, Seghatoleslami R, Denaro V, Wang D, Vo NV, Sowa GA, Hartman RA, Ngo K, Choe SR, Witt WT,
Vaccaro AR, Albert TJ, Shapiro IM (2006) Toward an optimum sys- Dong Q, Lee JY, Niedernhofer LJ, Kang JD (2011) Bupivacaine
tem for intervertebral disc organ culture: TGF-beta 3 enhances decreases cell viability and matrix protein synthesis in an interverte-
nucleus pulposus and annulus fibrosus survival and function through bral disc organ model system. Spine J 11(2):139146
modulation of TGF-beta-R expression and ERK signaling. Spine Wilke HJ, Neef P, Caimi M, Hoogland T, Claes LE (1999) New in vivo
(Phila Pa 1976) 31(8):884890 measurements of pressures in the intervertebral disc in daily life.
Roberts S, Menage J, Sivan S, Urban JP (2008) Bovine explant model Spine (Phila Pa 1976) 24(8):755762
of degeneration of the intervertebral disc. BMC Musculoskelet Wuertz K, Godburn K, MacLean JJ, Barbir A, Donnelly JS, Roughley
Disord 9:24 PJ, Alini M, Iatridis JC (2009) In vivo remodeling of intervertebral
Sakai D, Mochida J, Iwashina T, Watanabe T, Suyama K, Ando K, discs in response to short- and long-term dynamic compression.
Hotta T (2006) Atelocollagen for culture of human nucleus J Orthop Res 27(9):12351242
Use of Stem Cells for Regeneration
of the Intervertebral Disc 23
Daisuke Sakai and Joji Mochida
Contents 23.1 Evolution of Stem Cell Biology

23.1 Evolution of Stem Cell Biology in Musculoskeletal Tissues
in Musculoskeletal Tissues ............................................. 373
23.2 Evidence of a Stem Cell System
The ability of a tissue to respond to stress or injury requires
in the Intervertebral Disc ............................................... 374 the involvement and functions of stem cells resident in tissue-
specific microenvironmental niches. Aging has been shown
23.3 Commitment of Stem Cells Toward
an Intervertebral Disclike Cell ....................................... 375 to result in a decrease in stem cell number as well as loss of
ability to maintain tissue homeostasis and regenerate lost tis-
23.4 Stem Cell Promotion of Disc Cell Differentiation ........ 376
sue function. Therefore, it is of critical importance to iden-
23.5 Stem Cell Transplantation for Intervertebral tify stem/progenitor cell populations in different tissues,
Disc Tissue Engineering and Regeneration .................. 377
determine how these cells function in tissue homeostasis,
23.6 Perspectives on the Role of Stem Cell Biology and ascertain their potential utility in tissue engineering.
in the Treatment of Intervertebral Disc Disease .......... 379 Classically, stem cells of the musculoskeletal region were
23.7 Summary of Critical Concepts Discussed defined as undifferentiated cells, found in small numbers in
in the Chapter .................................................................. 381 the periosteum or the bone marrow (Caplan 1991; Deans and
References ...................................................................................... 381 Moseley 2000). Subsequently, these small populations of
cells have been found in other stem cell pools, such as the
adipose tissues, synovial tissues, perivascular regions in the
surrounding tissue environment, and even in the matrix com-
ponent of tissues (Bianco et al. 2001; Crisan et al. 2008).
They were designated mesenchymal stem cells because
from a developmental perspective, they were thought to be
mesenchymal in origin. These cells can undergo sustained
proliferation in vitro and potentially give rise to multiple mes-
enchymal cell lineages, including osteocytes, chondrocytes,
and adipocytes. Since then, many researchers have reported
studies of mesenchymal stem cells using different methods of
isolation and expansion. With the increasing difficulties
encountered in comparing and contrasting study outcomes,
these cells are now preferentially called multipotent mesen-
chymal stromal cells (multipotent MSCs). The Tissue Stem
Cell Committee of the International Society for Cellular
Therapy has proposed minimal criteria to define human MSC
(Dominici et al. 2006). First, MSC must be adherent to plastic
D. Sakai, MD, PhD (*) J. Mochida, MD, PhD when maintained under standard culture conditions. Second,
Department of Orthopaedic Surgery, Surgical Science, MSC must express the surface molecules CD105, CD73, and
Research Center for Regenerative Medicine,
CD90, and not express CD45, CD34, CD14, CD11b, CD79a,
Tokai University School of Medicine,
Shimokasuya 143, Isehara, Kanagawa 259-1193, Japan CD19, or HLA-DR. Third, MSC must differentiate into osteo-
e-mail: daisakai@is.icc.u-tokai.ac.jp; jomo@is.icc.u-tokai.ac.jp blasts, adipocytes, or chondroblasts in vitro. Despite these

374 D. Sakai and J. Mochida
stem cell system in the intervertebral disc is still in its infancy,

Definition of a stem/progenitor cell
recent studies have shown evidence of a potential stem cell
Clonogenicity niche in the intervertebral disc region. Henriksson et al. (2009)
attempted to detect cell proliferation zones and label-retaining
High proliferation
cells with in vivo 5-bromo-2-deoxyuridine (BrdU) labeling
Self-renewal in rabbit discs. They also investigated the localization of pro-
Multipotency genitor cell markers (Notch1, Delta4, Jagged1, C-KIT, KI67,
and Stro-1) with immunohistochemistry in the degenerated
intervertebral discs of rabbits, rats, mini pigs, and humans.
Although slow, ongoing proliferation was detectable in both
the nucleus pulposus and annulus fibrosus regions. They also
found a high number of BrdU-positive cells at the annulus
fibrosus borders with the ligament zone and the perichon-
drium region at early time points, whereas only a few label-
Fig. 23.1 Definition of a stem/progenitor cell. Criteria of stem/pro-
retaining cells were observed at later time points, identifying
genitor cell include clonogenicity, high proliferative capacity, self- a stem cell niche-like pattern. This may support the findings
renewal capability (from a single cell to reconstitute a tissue in which of Melrose et al. (2007), who demonstrated spatial annular
the cell was derived), and multipotency remodeling in experimentally injured ovine discs.
The recruitment of cells from the surrounding environ-
minimal criteria to identify MSC, it is difficult to define MSC ment is another phenomenon often seen in the regenerative
other than by the operational definition of self-renewal and process. Using the BrdU in vivo labeling technique described
differentiation potential in vitro. Therefore, our knowledge of above, Henriksson et al. (2011) observed BrdU-labeled cells
MSC is based solely on the characterization of cultured cells. in the intervertebral disc niche, adjacent to the epiphyseal
We still lack any knowledge of their in vivo characteristics, plate at early time points, but at later time points, these cells
such as their development, exact tissue localization, and were mainly in the outer region of the annulus fibrosus, sug-
physiological roles. gesting possible migratory pathway. Tzaan and Chen (2011)
Despite these difficulties, the considerable therapeutic have extended this concept by enhancing bone marrow cell
potential of human MSC has generated marked and increas- migration to the disc by stimulation with granulocyte colony-
ing interest within a wide variety of biomedical disciplines, stimulating factor. However, their results only demonstrated
including studies of the intervertebral disc. Published studies increased bone marrow cells in the endplate. Cell migration
involving MSC have increased markedly in this field and processes are thus underdefined and require substantial
may in the long run provide new and useful information on investigation.
disc biology, the pathogenesis of intervertebral disc There is increasing evidence that stem/progenitor cells
degeneration, and possibly offer new therapeutic strategies are present in the intervertebral disc. Risbud et al. (2007)
(Sakai 2011). However, as with any area of study, resear- reported that cells isolated from degenerative human disc tis-
chers must share similar definitions, terminologies, and sues expressed CD105, CD166, CD63, CD49a, CD90,
methods to ensure that their findings can be compared. It is CD73, p75 low-affinity nerve growth factor receptor, and
important for scientists to know the essential features of a CD133/1, proteins that are characteristic of MSC and repre-
stem cell: its capacity for self-renewal and the ability of a sent the differentiation capacity of these cells toward osteo-
single cell to regenerate all the cell types and matrices of the genesis, adipogenesis, and chondrogenesis. Feng et al. (2010)
lineage from which the cell was derived (Fig. 23.1) (Blanpain have demonstrated that the human degenerative annulus con-
et al. 2004). Accordingly, the experimental design and the tains cells that express the MSC markers CD29, CD49e,
types of stem cells used in investigational studies and trans- CD51, CD73, CD90, CD105, CD166, CD184, and Stro-1
lational cell therapies involving the intervertebral disc must and two neuronal stem cell markers, nestin and neuron-
be selected with care. specific enolase. In an in vitro assay, they differentiated
annulus fibrosus-derived cells into adipocytes, osteoblasts,
chondrocytes, neurons, and endothelial cells. Blanco et al.
23.2 Evidence of a Stem Cell System (2010) investigated MSC markers in nucleus pulposus cells
in the Intervertebral Disc isolated from degenerated discs and compared their differen-
tiation capacities with those of bone marrow-derived MSC
As mentioned above, investigation of the tissue-specific stem from the same patient. They found that nucleus pulposus-
cell niche is a key factor in understanding the degenerative derived MSC fulfilled almost all the morphological, immu-
processes and tissue regeneration. Although research into the nophenotypical, and differentiation criteria for MSC
23 Use of Stem Cells for Regeneration of the Intervertebral Disc 375
Fig. 23.2 Utilization of Utilization of MSC for IVD regeneration

multipotent mesenchymal
stromal cells for disc regenera- Induction of MSC
towards IVD cell-like cells IVD cell activation by MSC Direct transplanation
tion. Mesenchymal stem cells
(MSCs) can be directly induced
to differentiate toward an
intervertebral disc cell (IVD) Culture insert
using growth factor and coculture
techniques (left). In addition, Activation IVD cell
activation can be achieved using MSC MSC
coculture systems in which there
is direct cellcell contact Growth factor
(middle). Alternatively, MSC can Coculture etc. MSC
be directly transplanted into the
disc and commitment enhanced
by the local conditions and the
presence of resident host cells
(right)
Activated IVD cell
IVD cell-like cell
described by the International Society of Cell Therapy, et al. 2011; Purmessur et al. 2011; Ruan et al. 2012; Stoyanov
except that nucleus-derived MSCs were unable to differenti- et al. 2011). Various methods to induce MSC differentiation
ate into adipocytes. Likewise, Liu et al. (2011) investigated have been evaluated such as stimulation with growth factors
the characteristics of MSC derived from degenerated human under specific culture conditions, coculture with terminally
disc cartilage endplates and reported that the morphology, differentiated intervertebral disc cells, or seeding cells onto a
proliferation rate, cell cycle, immunophenotype, and expres- microenvironment-mimicking scaffold. Steck et al. (2005)
sion of stem cell genes were similar to those isolated from compared the molecular phenotypes of human disc cells and
the bone marrow. These research findings suggest that the articular chondrocytes to determine whether MSC can dif-
stimulation of endogenous stem cell populations may be an ferentiate toward both cell types after transforming growth
effective strategy for treating intervertebral disc degenera- factor-b (TGFb)-mediated induction in vitro. Their results
tion or to provide cells for the allogeneic transplantation of demonstrated that the gene expression profile adopted by
somatic-tissue-specific stem cells. MSC resembled that of native disc tissue more closely than
that of native joint cartilage. As well as TGFb3, insulin-like
growth factor 1 (IGF1), fibroblast growth factor 2, and plate-
23.3 Commitment of Stem Cells Toward let-derived growth factor BB have been shown to induce
an Intervertebral Disclike Cell MSC differentiation toward nucleus pulposus-like cells
(Ehlicke et al. 2010). Richardson et al. (2006) assessed the
In contrast to the small steps yet taken in investigating the value of a coculture system with or without cellcell contact
endogenous stem cell system in the intervertebral disc, a to induce MSC differentiation toward a nucleus pulposus-
number of studies have reported the utilization of MSC in like phenotype. Real-time quantitative polymerase chain
new, targeted therapeutic strategies (Fig. 23.2). MSC utiliza- reaction (RT-PCR) demonstrated a significant increase in the
tion can be subdivided into three main types of studies. The expression of nucleus pulposus marker genes in stem cells
first approach utilizes the multipotent differentiation capac- when the cells were cocultured with contact for 7 days, and
ity of MSC to induce stem cells to commit toward an inter- this change was regulated by the cell ratio. However, no
vertebral disclike cell. A considerable number of studies in significant changes in marker gene expression was observed
the literature describe this strategy (Risbud et al. 2004; Li when the cells were cultured with either the nucleus pulpo-
et al. 2005; Steck et al. 2005; Richardson et al. 2006; sus cells or stem cells without contact, indicating a require-
Sobajima et al. 2008; Vadal et al. 2008; Wuertz et al. 2008; ment for cellcell contact in this induction process.
Chen et al. 2009; Kim et al. 2009; Le Maitre et al. 2009; Wei In a follow-up study, Strassburg et al. (2010) used human
et al. 2009a; Korecki et al. 2010; Strassburg et al. 2010; nucleus pulposus cells from degenerate and nondegenerate
Bertolo et al. 2012; Choi et al. 2011; Feng et al. 2011a; Luo intervertebral discs to show that MSC differentiated into a
nucleus pulposus-like phenotype following direct coculture presence of TGFb1. The nanofibrous scaffold supported the
with either of the tissues with significantly upregulated differentiation of rabbit MSC toward a nucleus pulposus-like
expression of the genes encoding SOX9, collagen VI, aggre- phenotype in vitro, with the upregulated expression of a few
can, and versican, together with the simultaneous upregula- important nucleus-associated genes (encoding aggrecan, col-
tion of growth differentiation factor 5 (GDF5), TGFb1, lagen II, and Sox-9), the abundant deposition of ECM (gly-
IGF1, and connective tissue growth factor expression. The cosaminoglycan and collagen II), and the continuous
direct coculture of normal nucleus pulposus cells with MSC expression of the nucleus pulposus-specific marker, hypoxia-
had no effect on the phenotype of the nucleus cells, whereas inducible factor 1-a (HIF1-a).
coculture with degenerated nucleus pulposus cells resulted The effect of physiological stimulation on MSC differen-
in enhanced matrix gene expression in the degenerate cells, tiation toward intervertebral disclike cells has also been stud-
accompanied by increases in both TGFB1 and GDF5 gene ied. Luo et al. (2011) cultured MSC under simulated
expressions. These results suggest that cellular interactions microgravity in a chemically defined medium supplemented
between MSC and degenerated nucleus pulposus cells both with TGFb1 (positive control group). The results showed
stimulate MSC differentiation to a nucleus pulposus-like that MSC independently and spontaneously differentiated
phenotype and promote the endogenous nucleus pulposus toward a nucleus pulposus-like phenotype under simulated
cell population to regain a nondegenerated phenotype, con- microgravity without the addition of any external bioactive
sequently enhancing matrix synthesis for self-repair. stimulant, such as factors from the TGFb family, which were
Similarly, when human nucleus pulposus cells were main- previously considered necessary.
tained in a pellet coculture with human MSC in different Recently, another unique approach to inducing MSC was
ratios, the 75:25 and 50:50 NP:MSC ratios yielded the great- reported by Korecki et al. (2010). Because notochord-derived
est increases in extracellular matrix (ECM) production cells play a major role in nucleus development, they tested
(Sobajima et al. 2008). Vadal et al. (2008) also reported a whether the culture medium from porcine notochordal cells
reduction in the expression of collagen I and an increase in could direct the differentiation of MSCs. Human MSC pel-
the expression of collagen II and aggrecan in MSCs after lets were cultured first in serum-free medium for 4 days and
coculture with nucleus pulposus cells in alginate hydrogels, then in notochordal cell-conditioned medium for 7 days. It
which allowed short-distance paracrine cell interactions. was found that there was glycosaminoglycan accumulation
Stoyanov et al. (2011) have shown the important role of and increased gene expression toward a nucleus pulposus-
hypoxia, together with the addition of GDF5, in enhancing like phenotype, together suggesting that there was differen-
the expression of nucleus pulposus markers in a bone mar- tiation toward that of the intervertebral disc phenotype.
row-derived MSC coculture. The results of these studies suggest that MSC from any
MSCs from sources other than the bone marrow have source can be forced to express some of the molecular mark-
also been examined. Lu et al. (2007) studied the interac- ers of intervertebral disc cells in vitro and many external
tions between human nucleus pulposus cells and adipose- stimuli can accelerate differentiation. However, at the end of
tissue-derived stem cells in Transwell coculture, using both the day, they do not develop beyond nucleus pulposus-like
monolayer and micromass configurations. A similar cocul- cells because the default differentiation pathway is deter-
ture study was performed by Chen et al. (2009) utilizing syn- mined by the localization of the MSC. The most important
ovium-derived MSC. Vadal et al. (2008) investigated mus- factor in studying the induction of MSC toward interverte-
cle-derived MSC, and Ruan et al. (2012) assessed whether bral disclike cells is to demonstrate their functional capacity
Whartons jelly cells could be induced to differentiate toward in vivo.
nucleus pulposus-like cells in similar coculture studies.
The effects of scaffold properties on the differentiation of
MSC toward intervertebral disc cells have also been investi- 23.4 Stem Cell Promotion of
gated. Bertolo et al. (2012) evaluated four types of matrices Disc Cell Differentiation
as scaffolds, approved as medical devices for other applica-
tions: two made of equine or porcine collagen, one of gela- Another approach to the utilization of stem cells is to exploit
tin, and one of chitosan. They showed that although the their capacity to nourish other cells. Stem cells can act as
collagen scaffolds induced better chondrogenic differentia- feeder cells to stimulate target cells directly through cellcell
tion than the other scaffolds, the phenotype of the MSC was contact or indirectly through the secretion of various factors.
not fully equivalent to that of nucleus pulposus cells. Feng In a rabbit disc cell culture, Yamamoto et al. (2004) showed
et al. (2011a) investigated the use of three-dimensional (3D) that direct cellcell contact between nucleus pulposus cells
nanofibrous poly(l-lactide) scaffolds seeded with rabbit and MSC occurred across a membrane with 0.45 mm pores,
MSCs which were induced to differentiate along the nucleus which allowed only the cell processes to adhere to each other,
pulposus pathway in a hypoxic chamber (2 % O2) in the without any more extensive contact between the cultured
cells. Compared with the culture systems with no cellcell immunohistochemistry, and gene expression profiling
contact, a coculture system that allowed intercellular adhe- revealed the time-dependent development of the chondro-
sion between nucleus pulposus cells and MSC yielded a cytic phenotype of the seeded cells. The cells also maintained
marked increase in target cell proliferation, DNA synthesis, the microarchitecture of the native disc. An electrospun
and proteoglycan synthesis. This is possibly attributable to nanofibrous scaffold has also been used to examine the native
the increased secretion of cytokines found in the culture biomechanics of the annulus fibrosus (Driscoll et al. 2011).
system. Previous studies have shown that the tensile and shear prop-
An interesting study by He and Pei (2012) tested the erties of the native tissue are dependent on the fiber angle
effects of the ECM deposited by synovium-derived MSC on and the sample aspect ratio, respectively, so the effects of
the rejuvenation of nucleus pulposus cells. The nucleus pul- changing the fiber angle and the sample aspect ratio on the
posus cells that were expanded on ECM grew much faster, shear properties of aligned electrospun poly(e-caprolactone)
were smaller, and had a more fibroblastic shape than those scaffolds were evaluated. How ECM deposition by the resi-
expanded in plastic flasks. ECM-treated nucleus pulposus dent MSC modulates the measured shear response was also
cells acquired enhanced CD90 expression and higher mRNA determined. This team showed that the fiber orientation and
levels of collagen I, collagen II, and collagen X and aggrecan sample aspect ratio significantly influenced the response of
as well as a robust redifferentiation capacity for up to six pas- scaffolds in shear; indeed, the shear properties of both cel-
sages. The authors concluded that the ECM provides a tissue- lular and cell-seeded formulations can match or even exceed
specific microenvironment for the rejuvenation of nucleus the native tissue.
pulposus cells with a higher proliferation rate and greater A different approach to tissue engineering the annulus
redifferentiation capacity. These characteristics may improve fibrosus was reported by See et al. (2011). They constructed
any future autologous disc cell-based minimally invasive cell sheets from bone marrow MSC and incorporated them
therapeutic approach to the physiological reconstruction of a onto silk scaffolds to simulate the native lamellae of the
biologically functional disc in the clinical setting. annulus fibrosus. They then wrapped the construct around
Another use of MSC is in the delivery of bioactive factors silicone nucleus pulposus, mimicking an intervertebral disc
to resident disc cells. Meyerrose et al. (2010) suggested the construct, and used a bioreactor to provide compressive
use of MSC for the sustained in vivo production of supra- mechanical stimulation to the silicone disc. Under static con-
physiological levels of cytokines to support co-transplanted ditions, the MSC sheets remained viable, with no significant
stem cells and resident cells in intervertebral disc therapies. change in cell numbers for 4 weeks. A histological analysis
showed that the MSC sheets adhered well to the silk scaf-
folds and glycosaminoglycans were detected within the
23.5 Stem Cell Transplantation for ECM. The ratio of collagen I to collagen II within the ECM
Intervertebral Disc Tissue Engineering of the MSC sheets also decreased significantly over the
and Regeneration period of culture. These results suggest that extensive remod-
eling of the ECM occurred within the simulated disclike
The final approach to MSC utilization in intervertebral disc assembly and that this assembly is suitable for the regenera-
tissue engineering and regeneration involves the construc- tion of the inner annulus fibrosus.
tion of a disclike tissue in vitro and its transplantation or the Finally, there have been several attempts to supplement
direct delivery of stem cells into the intervertebral disc. viable cells in the disc by direct stem cell transplantation.
Gaetani et al. (2008), Nesti et al. (2008), Driscoll et al. The thought behind this concept is the finding that degenera-
(2011), and See et al. (2011) have all attempted to construct tion is initiated by reductions in the numbers, viability, and
this type of tissue in vitro. To improve the quality of in vitro- functions of disc cells, especially those of the nucleus pulpo-
reconstructed tissue, Gaetani et al. (2008) constructed a sus. Basic in vitro studies have shown that disc cells have a
nucleus pulposus-like tissue using a 3D culture of nucleus low proliferative capacity and that most cells in the adult
cells and adipose-derived MSC. To develop a biphasic con- human disc are in a senescent state. These facts have led
struct, Nesti et al. (2008) seeded MSC onto a hyaluronic researchers to focus on the idea of transplanting stem cells,
acidnanofibrous scaffold that consisted of an electrospun, which may show transplanted site-dependent differentiation
biodegradable nanofibrous scaffold enveloping a hyaluronic and function to produce a functional ECM.
acid hydrogel center. The seeded MSCs were induced to Sakai et al. (2003) reported the first MSC transplantation
undergo chondrogenesis in vitro in the presence of TGFb for study in a rabbit model of disc degeneration. In follow-up
up to 28 days. The cartilaginous hyaluronic acidnanofibrous studies (Sakai et al. 2005, 2006), they showed that trans-
scaffold construct resembled a native disc architecturally, planted MSCs survive, proliferate, and differentiate into cells
with an outer annulus fibrosus-like region and an inner expressing chondroitin sulfate; keratan sulfate; collagen I,
nucleus-like region. Histological and biochemical analyses, collagen II, and collagen IV; HIF1-a and HIF1-b; HIF2-a
Table 23.1 Summary of in vivo animal experimental studies on stem cell transplantation therapy for intervertebral disc degeneration
Stem cell type Mode Animal model Author Year
Bone marrow MSCs (Autologous MSCs expanded) Rabbit (nucleus aspiration) Sakai et al. 2003, 2005, 2006
Bone marrow MSCs (MSCs expanded) Rat (no injury) Crevensten et al. 2004
Bone marrow MSCs (Allogeneic MSCs expanded) Rabbit no injury Zhang et al. 2005
Bone marrow MSCs (Allogeneic MSCs expanded) Rabbit (nucleus puncture) Leung et al. 2006
Bone marrow MSCs (Autologous MSCs expanded) Canine (nucleotomy) Hiyama et al. 2008
Adipose MSCs (Autologous MSCs expanded) Goat (ABC chondroitinase) Hoogendoorn et al. 2008
Adipose MSCs (Autologous MSCs expanded) Canine (nucleotomy) Ganey et al. 2009
Bone marrow human MSCs (Xenogeneic MSC expanded) Rat (no injury) Wei et al. 2009a and 2009b
Bone marrow human MSCs (Xenogeneic MSC expanded) Porcine(nucleus aspiration) Henriksson et al. 2009
Synovial MSCs (Allogeneic MSCs expanded) Rabbit (nucleus puncture) Miyamoto et al. 2010
Bone marrow MSCs (Autologous MSCs expanded) Rabbit (nucleus aspiration) Yang et al. 2010
Bone marrow human MSCs (Xenogeneic MSC expanded) Rat (nucleotomy) Allon et al. 2010
Bone marrow MSC + (Autologous MSCs expanded) Rabbit (nucleus aspiration) Feng et al. 2010
autologous NP cells
Bone marrow MSCs (Autologous MSCs expanded) Mini pig (annulus incision) Bendtsen et al. 2011
Adipose human MSCs (Xenogeneic MSCs expanded) Rabbit (nucleus puncture) Chun et al. 2012
and HIF2-b; glucose transporter type 1 (GLUT1) and nucleus pulposus aspiration in mature beagle dogs (a chon-
GLUT3; and matrix metalloproteinase 2 (MMP2), proteins drodystrophoid breed), and the animals were followed for
that characterize the native nucleus pulposus cells. They also another 8 weeks. Radiological, histological, biochemical,
used RT-PCR to quantify the expression of the genes encod- immunohistochemical, and gene expression analyses dem-
ing aggrecan, versican, collagen I and II, interleukin 1b onstrated the clear regenerative effects of the transplanted
(IL1b), IL6, tumor necrosis factor-a, MMP9, and MMP13, cells. Importantly, MSC found in the nucleus 8 weeks after
thus demonstrating the increased expression of nucleus pul- transplantation expressed Fas-ligand protein, which has not
posus cell markers and the downregulated expression of been detected in MSCs before transplantation. Fas-ligand,
inflammatory genes. Magnetic resonance imaging (MRI) which is present in tissues with isolated immune privilege,
and radiography confirmed the regenerative effects of the including the nucleus pulposus, plays an important role in
procedure, showing that MSC transplanted into degenerating nucleus maintenance. The expression of Fas-ligand indicates
discs in vivo can survive, proliferate, and differentiate into that MSC transplantation may also contribute to the preser-
cells that express the phenotype of nucleus pulposus cells, vation of the immune privilege of the disc environment. In a
with suppressed inflammatory gene expression. Since then, follow-up study using the same model, Serigano et al. (2010)
numerous similar studies have used different animal models investigated the effects of the numbers of MSC transplanted.
and cell carriers and MSC from various sources (Table 23.1) To determine the optimal number of MSCs for transplanta-
(Crevensten et al. 2004; Zhang et al. 2005; Hiyama et al. tion, 105, 106, and 107 MSCs were transplanted, and their sur-
2008; Hoogendoorn et al. 2008; Ganey et al. 2009; Wei et al. vival and apoptosis after transplantation were compared,
2009a). together with their regenerative effects, which were assessed
Crevensten et al. (2004) used a 15 % hyaluronan gel as with histology, radiography, and MRI images. The transplan-
the carrier and injected fluorescently labeled MSC into rat tation of 106 MSCs, rather than 105 or 107 cells, best main-
coccygeal discs. Although the number of MSCs retained tained the structure of the intervertebral disc and optimally
decreased significantly during the first 2 weeks after injec- inhibited degeneration. This study demonstrates that the
tion, the initial cell number was restored after 4 weeks, and number of cells transplanted affects the regenerative capacity
cell viability and disc height were maintained. These results of MSC transplants in animal models of disc degeneration
indicate that the injected cells had started to proliferate within and addresses the importance of cell number in the clinical
the rat disc. Zhang et al. (2005) implanted allogeneic MSC application of MSC transplantation.
containing the marker gene lacZ from young rabbits into rab- Leung et al. (2006) investigated the allogeneic transplan-
bit intervertebral discs to determine the potential of this cell- tation of MSC and reported many advantages of such trans-
based approach. As the transplanted allogeneic MSC survived plantations in treating disc disease. They reasoned that if the
and increased the proteoglycan content within the disc, it nucleus pulposus is an immunoprivileged environment, then
lent strong support to the notion that these cells could be by presenting fewer antigens, MSC should escape alloanti-
used as a potential treatment for intervertebral disc degenera- gen recognition. Wei et al. (2009b) have also reported the
tion. Along the same lines, Hiyama et al. (2008) injected one xenogeneic transplantation of bone marrow-derived MSC in
million autologous MSCs into lumbar discs 4 weeks after rats, and Henriksson et al. (2009) transplanted human bone
marrow MSC into a porcine model of disc degeneration, pro- function. They showed that MSC transplantation partly
ducing a regenerative effect. regenerated the degenerative disc and maintained the perfu-
Various stem cells have already been tested as cell sources. sion and permeability characteristics of the vertebral end-
Hoogendoorn et al. (2008) reported that adipose-derived plate and subchondral bone.
MSC may be beneficial in cell-based therapies for interverte- With increasing evidence of the efficacy of MSC trans-
bral disc disease because they are isolated more easily than plantation therapy in promoting intervertebral disc regenera-
are bone marrow MSC. Ganey et al. (2009) studied the tion, three reports of stem cell transplantation in humans have
efficacy of autologous adipose-tissue-derived stem cells in been published. Haufe and Mork (2006) injected hematopoi-
promoting disc regeneration in a canine model of disc injury etic stem cells from the bone marrow into ten patients with
and found enhanced disc matrix production and improved discogenic back pain, diagnosed with provocative discogram.
overall disc morphology. A recent study by Chun et al. (2012) There was no improvement in any patient after 1 year. This
in a rabbit trauma model has also shown that the transplanta- study cautions that the appropriate selection of patients, cells,
tion of adipose-derived MSC increased ECM secretion, with and methods must be made with care before the clinical
little ossification of the damaged cartilage in the nucleus pul- application of this technique. Yoshikawa et al. (2010) trans-
posus compared with degenerated control discs. MSC purified planted autologous bone marrow MSC into discs showing
from the synovial tissue of knee joints demonstrated superior the vacuum phenomenon and instability in two patients
proliferative ability and greater chondrogenic differentiation undergoing decompression surgery for spinal stenosis. The
capacity than those of bone marrow MSC. Miyamoto et al. MSCs were cultured in medium containing autologous
(2010) transplanted allogeneic synovial MSC into rabbits and serum. During surgery, the stenosed spinal canal was fenes-
found that synovial MSC injected into the nucleus pulposus trated, and pieces of collagen sponge containing autologous
space promoted the synthesis of collagen II in the remaining MSCs were then grafted percutaneously onto the degener-
nucleus pulposus cells and inhibited the expressions of degra- ated disc. Two years after surgery, radiography and computed
dative enzymes and inflammatory cytokines. This allowed tomography showed improvements in the vacuum phenome-
the structure of the intervertebral disc to be maintained. non in both patients. T2-weighted MRI showed high signal
Although evidence of the regenerative effects of MSC intensity in the intervertebral discs treated with cell grafts,
transplantation has accrued, whether MSCs perform better indicating a high moisture content, and dynamic radiography
than transplanted nucleus pulposus cells is as yet unknown. showed less lumbar disc instability. Orozco et al. (2011) also
Feng et al. (2011b) compared the regenerative effects of transplanted autologous bone marrow MSC into ten patients
nucleus pulposus cell and MSC transplantation in a rabbit with chronic back pain, diagnosed with lumbar disc degen-
model. No significant differences in gene expression were eration with an intact annulus fibrosus. The feasibility and
observed between the groups transplanted with nucleus cells safety of the treatment were confirmed and strong indications
or MSCs, supporting the role of MSC as a useful cell therapy of its clinical efficacy identified. The patients displayed rapid
substitute for nucleus pulposus cells for intervertebral disc improvement in their pain and disability (to 85 % of maxi-
degeneration. Furthermore, recent studies have explored vari- mum in 3 months), which approached 71 % of optimal
ous modifications to this method to enhance its effect. In vitro efficacy, comparing favorably with the results for other pro-
studies have shown that the coculture of nucleus pulposus cedures, such as spinal fusion and total disc replacement.
cells and MSC promotes ECM production. The use of bilami- MRI evaluations 1 year after transplantation showed that
nar coculture pellets (BCPs) of MSC and nucleus pulposus although the disc height was not restored, the water content
cells has also been explored by Allon et al. (2010). BCPs, in was significantly elevated. They concluded that MSC therapy
which MSCs are enclosed in a shell of nucleus pulposus cells, might be a valid alternative treatment for chronic back pain
were transplanted into denucleated rat-tail discs with a fibrin- caused by degenerative disc disease. Its advantages over the
sealant carrier. Cell retention was higher in the BCP-treated current gold standard treatments include that it is a simpler
group than in the control group. The disc height and disc grade and more conservative intervention, without surgery, it pre-
increased over time only in the BCP-treated group. Only after serves the normal biomechanics of the intervertebral disc,
transplantation with BCPs was proteoglycan staining evident and it offers the same or better pain relief.
in the nuclear space after 35 days. This approach has been
combined with growth factor therapy (Yang et al. 2010). MSC
transplanted with pure fibrinous gelatin-TGFb1 resulted in 23.6 Perspectives on the Role of Stem Cell
less degeneration and a reduction in apoptotic cells in rabbits. Biology in the Treatment of
The modalities used to evaluate disc degeneration have Intervertebral Disc Disease
advanced markedly with the evolution of MRI technology.
Bendtsen et al. (2011) investigated the regenerative effects of Recent advances in stem cell biology offer a number of entic-
MSC transplantation with or without hydrogel in mini pigs, ing possible avenues for intervertebral disc research, with
using dynamic contrast-enhanced MRI to focus on endplate therapeutic potential. Undoubtedly, studies of the applications
of stem cell biology will increase; however, as already dis- recruitment and function of disc stem/progenitor cells may
cussed, it is important that investigators share a common stem enhance the longevity of the tissue itself. This area of research
cell terminology. There is an increasing interest in this field is still underdefined and awaits rigorous investigation.
and evidence of endogenous stem/progenitor cells within the Supporting this insight, an epoch making finding that nucleus
intervertebral disc. Identification of these stem/progenitor pulposus progenitor cell exhaustion relates to aging and
cells in their niche and the fates of these cells should provide degeneration has been recently reported (Sakai et al. 2012). In
insight into the key elements in the pathogenesis of disc this study, the authors sorted human and mouse nucleus pul-
degeneration and the endogenous repair system of the inter- posus cells using various surface markers and then scored
vertebral disc. This insight is required for potential therapeu- their ability to form colonies. They identified nucleus pulpo-
tic interventions. For instance, molecules that enhance the sus progenitor cells that formed spheroid colonies in Tie2+
and disialoganglioside GD2+ population. These spheroid col-
onies demonstrated superior aggrecan and collagen II produc-
tion capability compared to other colonies (Fig. 23.3). Clonal
analyses of Tie2+GD2+ human nucleus pulposus cell demon-
strated that these cells are highly proliferative and clonally
capable of differentiation into multiple mesenchymal and
nucleus pulposus lineages. Moreover, this multipotent ability
is sustained through long-term engraftment and was demon-
strated to include the ability for self-renewal. Analyses in
clinically obtained tissue provided new concept in the
underdefined pathology of disc degeneration by showing a
gradual exhaustion of Tie2+ cells in the disc correlating with
aging and degeneration in humans. Moreover, they defined
the importance of Tie2/angiopoietin-1 niche (Fig. 23.4) in
maintenance of Tie2+ nucleus pulposus progenitor cells,
which may lead to new treatment targets. To induce MSC to
differentiate into intervertebral disc cells, an obstacle that
must be overcome is that induced MSCs never proceed beyond
intervertebral disclike cells. This problem is exacerbated by
the lack of a specific marker or assays that defines a cell as an
intervertebral disc cell. Fundamental studies that can define a
cell as a disc cell that can function in vitro will be most useful
Fig. 23.3 Spheroid colony derived from nucleus pulposus progenitor in this field. The use of MSC as an activation tool for drug
cells. These cells express abundant aggrecan (red), the major extracel-
delivery systems may still offer a novel strategy for tissue
lular matrix component of the nucleus pulposus (blue indicates nuclear
staining with diamidino-2-phenylindole) (Reproduced from Sakai et al. repair. In the developmental and maturation stages of the disc,
2012) the mesenchymal cells surrounding the nucleus pulposus are
Fig. 23.4 Tie2/angiopoietin-1

signaling maintains progenitor
cells in the nucleus pulposus of
the intervertebral disc. Both Tie2
(red) and Ang-1 (green) are
expressed in cluster of cells
found in the nucleus pulposus
(blue indicates nuclear staining
with diamidino-2-phenylindole)
(Reproduced from Sakai et al.
2012)
thought to be influenced by notochord-derived cells. the Ministry of Education, Culture, Sports, Science and Technology of
Knowledge of the interactions between the components that Japan and a grant from AOSpine International and AO Foundation to
D.S.
participate in disc development will certainly provide new
information on tissue regeneration and tissue engineering.
Finally, more clinical trials of actual MSC transplantation for
intervertebral disc repair will also undoubtedly be reported in References
the future, using various techniques and stem cells from vari-
Allon AA, Aurouer N, Yoo BB, Liebenberg EC, Buser Z, Lotz JC
ous sources. However, caution is strongly warranted. For (2010) Structured coculture of stem cells and disc cells prevent disc
example, using a rabbit model, Vadal et al. (2012) reported degeneration in a rat model. Spine J 10:10891097
that cell leakage after MSC injection into the intervertebral Bendtsen M, Bnger CE, Zou X, Foldager C, Jrgensen HS (2011)
Autologous stem cell therapy maintains vertebral blood flow and
disc can cause osteophyte formation. They suggest that cell
contrast diffusion through the endplate in experimental interverte-
carriers or annulus-sealing techniques should be assessed or bral disc degeneration. Spine (Phila Pa 1976) 36:E373E379
perhaps postsurgical rehabilitation protocols investigated, to Bertolo A, Mehr M, Aebli N, Baur M, Ferguson SJ, Stoyanov JV (2012)
minimize leakage. Unintended differentiation and tumorigen- Influence of different commercial scaffolds on the in vitro differen-
tiation of human mesenchymal stem cells to nucleus pulposus-like
esis are also potential risks usually faced in stem cell thera-
cells. Eur Spine J 21:S826S838
pies. Although many animal studies and preliminary human Bianco P, Riminucci M, Gronthos S, Robey PG (2001) Bone marrow
trials have supported the promising future of stem cell therapy stromal stem cells: nature, biology, and potential applications. Stem
for disc disease, careful application of cells and techniques, Cells 19:180192
Blanco JF, Graciani IF, Sanchez-Guijo FM, Muntion S, Hernandez-
with the selection of the appropriate types of patients in a
Campo P, Santamaria C, Carrancio S, Barbado MV, Cruz G,
strictly controlled manner, are key elements in ensuring the Gutierrez-Cosio S, Herrero C, San Miguel JF, Brinon JG, del Canizo
success of these therapies. MC (2010) Isolation and characterization of mesenchymal stromal
cells from human degenerated nucleus pulposus: comparison with
bone marrow mesenchymal stromal cells from the same subjects.
Spine (Phila Pa 1976) 35:22592265
23.7 Summary of Critical Concepts Discussed Blanpain C, Lowry WE, Geoghegan A, Polak L, Fuchs E (2004) Self-
in the Chapter renewal, multipotency, and the existence of two cell populations
within an epithelial stem cell niche. Cell 118:635648
Caplan AI (1991) Mesenchymal stem cells. J Orthop Res 9:641650
While definition of stem cells in the mesenchymal tissues
Chen WH, Liu HY, Lo WC, Wu SC, Chi CH, Chang HY, Hsiao SH, Wu
has changed with time, there is still a lack of knowledge CH, Chiu WT, Chen BJ, Deng WP (2009) Intervertebral disc regen-
of their in vivo characteristics, and therefore, selection of eration in an ex vivo culture system using mesenchymal stem cells
stem cells for regeneration of the intervertebral disc needs and platelet-rich plasma. Biomaterials 30:55235533
Choi EH, Park H, Park KS, Park KS, Kim BS, Han IB, Shin DA, Lee
to be carefully evaluated.
SH (2011) Effect of nucleus pulposus cells having different pheno-
It is important that investigators share a common stem types on chondrogenic differentiation of adipose-derived stromal
cell terminology in designing experiments to evaluate the cells in a coculture system using porous membranes. Tissue Eng
use of stem cells for intervertebral disc research. Part A 17:24452451
Chun HJ, Kim YS, Kim BK, Kim EH, Kim JH, Do BR, Hwang SJ,
Laboratory investigations have revealed that a stem cell
Hwang JY, Lee YK (2012) Transplantation of human adipose-
system is present in the intervertebral disc and that its derived stem cells in a rabbit model of traumatic degeneration of
exhaustion may be related to aging and degeneration. lumbar discs. World Neurosurg 78:364371
Stem cells can be induced to express some of the charac- Crevensten G, Walsh AJ, Ananthakrishnan D, Page P, Wahba GM, Lotz
JC, Berven S (2004) Intervertebral disc cell therapy for regenera-
teristics of intervertebral disc cells.
tion: mesenchymal stem cell implantation in rat intervertebral discs.
Stem cells are capable of stimulating intervertebral disc Ann Biomed Eng 32:430434
cell metabolism. Crisan M, Yap S, Casteilla L, Chen CW, Corselli M, Park TS, Andriolo
Stem cells can facilitate intervertebral disc tissue engi- G, Sun B, Zheng B, Zhang L, Norotte C, Teng PN, Traas J, Schugar
R, Deasy BM, Badylak S, Buhring HJ, Giacobino JP, Lazzari L,
neering and repair.
Huard J, Pault B (2008) A perivascular origin for mesenchymal
Investigation of intervertebral disc stem/progenitor cells stem cells in multiple human organs. Cell Stem Cell 3:301313
in their niche and the fates of these cells should provide Deans RJ, Moseley AB (2000) Mesenchymal stem cells: biology and
insight into key events in the pathogenesis of disc degen- potential clinical uses. Exp Hematol 2:875884
Dominici M, Le Blanc K, Mueller I, Slaper-Cortenbach I, Marini F,
eration and the endogenous repair system. This informa-
Krause D, Deans R, Keating A, Prockop DJ, Horwitz E (2006)
tion should be of great use in the design of new biological Minimal criteria for defining multipotent mesenchymal stromal
therapies. cells. The International Society for Cellular Therapy position state-
ment. Cytotherapy 8:315317
Driscoll TP, Nerurkar NL, Jacobs NT, Elliott DM, Mauck RL (2011)
Acknowledgments The authors are grateful to the staff of the Fiber angle and aspect ratio influence the shear mechanics of ori-
Education and Research Support Center, Tokai University. This work ented electrospun nanofibrous scaffolds. J Mech Behav Biomed
was supported in part by a Grant-in-Aid for Scientific Research from Mater 4:16271636
Ehlicke F, Freimark D, Heil B, Dorresteijn A, Czermak P (2010) Liu LT, Huang B, Li CQ, Zhuang Y, Wang J, Zhou Y (2011)
Intervertebral disc regeneration: influence of growth factors on dif- Characteristics of stem cells derived from the degenerated human
ferentiation of human mesenchymal stem cells (hMSC). Int J Artif intervertebral disc cartilage endplate. PLoS One 6:e26285
Organs 33:244252 Lu ZF, Zandieh Doulabi B, Wuisman PI, Bank RA, Helder MN (2007)
Feng G, Yang X, Shang H, Marks IW, Shen FH, Katz A, Arlet V, Laurencin Differentiation of adipose stem cells by nucleus pulposus cells:
CT, Li X (2010) Multipotential differentiation of human annulus configuration effect. Biochem Biophys Res Commun 359:991996
fibrosus cells: an in vitro study. J Bone Joint Surg Am 92:675685 Luo W, Xiong W, Qiu M, Lv Y, Li Y, Li F (2011) Differentiation of
Feng G, Jin X, Hu J, Ma H, Gupte MJ, Liu H, Ma PX (2011a) Effects mesenchymal stem cells toward a nucleus pulposus-like phenotype
of hypoxias and scaffold architecture on rabbit mesenchymal stem utilizing simulated microgravity in vitro. J Huazhong Univ Sci
cell differentiation toward a nucleus pulposus-like phenotype. Technolog Med Sci 31:199203
Biomaterials 32:81828189 Melrose J, Smith SM, Fuller ES, Young AA, Roughley PJ, Dart A,
Feng G, Zhao X, Liu H, Zhang H, Chen X, Shi R, Liu X, Zhao X, Little CB (2007) Biglycan and fibromodulin fragmentation corre-
Zhang W, Wang B (2011b) Transplantation of mesenchymal stem lates with temporal and spatial annular remodelling in experimen-
cells and nucleus pulposus cells in a degenerative disc model in rab- tally injured ovine intervertebral discs. Eur Spine J 16:21932205
bits: a comparison of 2 cell types as potential candidates for disc Meyerrose T, Olson S, Pontow S, Kalomoiris S, Jung Y, Annett G, Bauer
regeneration. J Neurosurg Spine 14:322329 G, Nolta JA (2010) Mesenchymal stem cells for the sustained in vivo
Gaetani P, Torre ML, Klinger M, Faustini M, Crovato F, Bucco M, delivery of bioactive factors. Adv Drug Deliv Rev 62:11671174
Marazzi M, Chlapanidas T, Levi D, Tancioni F, Vigo D, Rodriguez Miyamoto T, Muneta T, Tabuchi T, Matsumoto K, Saito H, Tsuji K,
y Baena R (2008) Adipose-derived stem cell therapy for interverte- Sekiya I (2010) Intradiscal transplantation of synovial mesenchy-
bral disc regeneration: an in vitro reconstructed tissue in alginate mal stem cells prevents intervertebral disc degeneration through
capsules. Tissue Eng Part A 14:14151423 suppression of matrix metalloproteinase-related genes in nucleus
Ganey T, Hutton WC, Moseley T, Hedrick M, Meisel HJ (2009) pulposus cells in rabbits. Arthritis Res Ther 12:R206
Intervertebral disc repair using adipose tissue-derived stem and Nesti LJ, Li WJ, Shanti RM, Jiang YJ, Jackson W, Freedman BA, Kuklo
regenerative cells: experiments in a canine model. Spine (Phila Pa TR, Giuliani JR, Tuan RS (2008) Intervertebral disc tissue engineer-
1976) 34:22972304 ing using a novel hyaluronic acid-nanofibrous scaffold (HANFS)
Haufe SM, Mork AR (2006) Intradiscal injection of hematopoietic stem amalgam. Tissue Eng Part A 14:15271537
cells in an attempt to rejuvenate the intervertebral discs. Stem Cells Orozco L, Soler R, Morera C, Alberca M, Snchez A, Garca-Sancho J
Dev 15:136137 (2011) Intervertebral disc repair by autologous mesenchymal bone
He F, Pei M (2012) Rejuvenation of nucleus pulposus cells using extra- marrow cells: a pilot study. Transplantation 92:822828
cellular matrix deposited by synovium-derived stem cells. Spine Purmessur D, Schek RM, Abbott RD, Ballif BA, Godburn KE, Iatridis
(Phila Pa 1976) 37:459469 JC (2011) Notochordal conditioned media from tissue increases
Henriksson H, Thornemo M, Karlsson C, Hagg O, Junevik K, Lindahl proteoglycan accumulation and promotes a healthy nucleus pulpo-
A, Brisby H (2009) Identification of cell proliferation zones, pro- sus phenotype in human mesenchymal stem cells. Arthritis Res
genitor cells and a potential stem cell niche in the intervertebral disc Ther 13:R81
region: a study in four species. Spine (Phila Pa 1976) 34:22782287 Richardson SM, Walker RV, Parker S, Rhodes NP, Hunt JA, Freemont
Henriksson HB, Svala E, Skioldebrand E, Lindahl A, Brisby H (2011) AJ, Hoyland JA (2006) Intervertebral disc cell-mediated mesenchy-
Support of concept that migrating progenitor cells from stem cell mal stem cell differentiation. Stem Cells 24:707716
niches contribute to normal regeneration of the adult mammal inter- Risbud MV, Albert TJ, Guttapalli A, Vresilovic EJ, Hillibrand AS,
vertebral disc: a descriptive study in the New Zealand white rabbit. Vaccaro AR, Shapiro IM (2004) Differentiation of mesenchymal
Spine (Phila Pa 1976) 37:722732 stem cells towards a nucleus pulposus-like phenotype in vitro:
Hiyama A, Mochida J, Iwashina T, Omi H, Watanabe T, Serigano K, implications for cell-based transplantation therapy. Spine (Phila Pa
Tamura F, Sakai D (2008) Transplantation of mesenchymal stem 1976) 29:26272632
cells in a canine disc degeneration model. J Orthop Res Risbud MV, Guttapalli A, Tsai TT, Lee JY, Danielson KG, Vaccaro AR,
26:589600 Albert TJ, Gazit Z, Gazit D, Shapiro IM (2007) Evidence for skel-
Hoogendoorn RJ, Lu ZF, Kroeze RJ, Bank RA, Wuisman PI, Helder etal progenitor cells in the degenerate human intervertebral disc.
MN (2008) Adipose stem cells for intervertebral disc regeneration: Spine (Phila Pa 1976) 32:25372544
current status and concepts for the future. J Cell Mol Med Ruan D, Zhang Y, Wang D, Zhang C, Wu J, Wang C, Shi Z, Xin H, Xu
12:22052216 C, Li H, He Q (2012) Differentiation of human Whartons jelly cells
Kim DH, Kim SH, Heo SJ, Shin JW, Lee SW, Park SA, Shin JW (2009) toward nucleus pulposus-like cells after coculture with nucleus pul-
Enhanced differentiation of mesenchymal stem cells into NP-like posus cells in vitro. Tissue Eng Part A 18:167175
cells via 3D co-culturing with mechanical stimulation. J Biosci Sakai D (2011) Stem cell regeneration of the intervertebral disk. Orthop
Bioeng 108:6367 Clin North Am 42:555562, viiiix. Review
Korecki CL, Taboas JM, Tuan RS, Iatridis JC (2010) Notochordal cell Sakai D, Mochida J, Yamamoto Y, Nomura T, Okuma M, Nishimura K,
conditioned medium stimulates mesenchymal stem cell differentia- Nakai T, Ando K, Hotta T (2003) Transplantation of mesenchymal
tion toward a young nucleus pulposus phenotype. Stem Cell Res stem cells embedded in Atelocollagen gel to the intervertebral disc:
Ther 1:18 a potential therapeutic model for disc degeneration. Biomaterials
Le Maitre CL, Baird P, Freemont AJ, Hoyland JA (2009) An in vitro 24:35313541
study investigating the survival and phenotype of mesenchymal Sakai D, Mochida J, Iwashina T, Watanabe T, Nakai T, Ando K, Hotta
stem cells following injection into nucleus pulposus tissue. Arthritis T (2005) Differentiation of mesenchymal stem cells transplanted to
Res Ther 11:R20 a rabbit degenerative disc model: potential and limitations for stem
Leung VY, Chan D, Cheung KM (2006) Regeneration of intervertebral cell therapy in disc regeneration. Spine (Phila Pa 1976)
disc by mesenchymal stem cells: potentials, limitations, and future 30:23792387
direction. Eur Spine J 15(Suppl 3):S406S413 Sakai D, Mochida J, Iwashina T, Hiyama A, Omi H, Imai M, Nakai T,
Li X, Lee JP, Balian G, Greg AD (2005) Modulation of chondrocytic Ando K, Hotta T (2006) Regenerative effects of transplanting mes-
properties of fat-derived mesenchymal cells in co-cultures with enchymal stem cells embedded in atelocollagen to the degenerated
nucleus pulposus. Connect Tissue Res 46:7582 intervertebral disc. Biomaterials 27:335345
Sakai D, Nakamura Y, Nakai T, Mishima T, Kato S, Grad S, Alini M, cells and nucleus pulposus cells modulate gene expression profile
Risbud M, Chan D, Cheah K, Yamamura K, Masuda K, Okano H, without cell fusion. Spine (Phila Pa 1976) 33:870876
Ando K, Mochida J (2012) Exhaustion of nucleus pulposus pro- Vadal G, Sowa G, Hubert M, Gilbertson LG, Denaro V, Kang JD
genitor cells with ageing and degeneration of the intervertebral disc. (2012) Mesenchymal stem cells injection in degenerated interverte-
Nat Commun 3:1264 bral disc: cell leakage may induce osteophyte formation. J Tissue
See EY, Toh SL, Goh JC (2011) Simulated intervertebral disc-like Eng Regen Med 6:348355. doi:10.1002/term.433
assembly using bone marrow-derived mesenchymal stem cell sheets Wei A, Chung SA, Tao H, Brisby H, Lin Z, Shen B, Ma DD, Diwan AD
and silk scaffolds for annulus fibrosus regeneration. J Tissue Eng (2009a) Differentiation of rodent bone marrow mesenchymal stem
Regen Med. doi:10.1002/term.457 cells into intervertebral disc-like cells following coculture with rat
Serigano K, Sakai D, Hiyama A, Tamura F, Tanaka M, Mochida J (2010) disc tissue. Tissue Eng Part A 15:25812595
Effect of cell number on mesenchymal stem cell transplantation in a Wei A, Tao H, Chung SA, Brisby H, Ma DD, Diwan AD (2009b) The
canine disc degeneration model. J Orthop Res 28:12671275 fate of transplanted xenogeneic bone marrow-derived stem cells in
Sobajima S, Vadal G, Shimer A, Kim JS, Gilbertson LG, Kang JD rat intervertebral discs. J Orthop Res 27:374379
(2008) Feasibility of a stem cell therapy for intervertebral disc Wuertz K, Godburn K, Neidlinger-Wilke C, Urban J, Iatridis JC (2008)
degeneration. Spine J 8:888896 Behavior of mesenchymal stem cells in the chemical microenviron-
Steck E, Bertram H, Abel R, Chen B, Winter A, Richter W (2005) ment of the intervertebral disc. Spine (Phila Pa 1976) 33:
Induction of intervertebral disc-like cells from adult mesenchymal 18431849
stem cells. Stem Cells 23:403411 Yamamoto Y, Mochida J, Sakai D, Nakai T, Nishimura K, Kawada H,
Stoyanov JV, Gantenbein-Ritter B, Bertolo A, Aebli N, Baur M, Alini Hotta T (2004) Upregulation of the viability of nucleus pulposus
M, Grad S (2011) Role of hypoxia and growth and differentiation cells by bone marrow-derived stromal cells: significance of direct
factor-5 on differentiation of human mesenchymal stem cells toward cell-to-cell contact in coculture system. Spine (Phila Pa 1976)
intervertebral nucleus pulposus-like cells. Eur Cell Mater 21: 29:15081514
533547 Yang H, Wu J, Liu J, Ebraheim M, Castillo S, Liu X, Tang T, Ebraheim
Strassburg S, Richardson SM, Freemont AJ, Hoyland JA (2010) NA (2010) Transplanted mesenchymal stem cells with pure fibrinous
Co-culture induces mesenchymal stem cell differentiation and mod- gelatintransforming growth factor-b1 decrease rabbit interverte-
ulation of the degenerate human nucleus pulposus cell phenotype. bral disc degeneration. Spine J 10:802810
Regen Med 5:701711 Yoshikawa T, Ueda Y, Miyazaki K, Koizumi M, Takakura Y (2010)
Tzaan WC, Chen HC (2011) Investigating the possibility of interverte- Disc regeneration therapy using marrow mesenchymal cell trans-
bral disc regeneration induced by granulocyte colony stimulating plantation: a report of two case studies. Spine (Phila Pa 1976)
factor-stimulated stem cells in rats. Adv Orthop 2011:602089. Epub 35:E475E480
2010 Nov 21 Zhang YG, Guo X, Xu P, Kang LL, Li J (2005) Bone mesenchymal
Vadal G, Studer RK, Sowa G, Spiezia F, Iucu C, Denaro V, Gilbertson stem cells transplanted into rabbit intervertebral discs can increase
LG, Kang JD (2008) Coculture of bone marrow mesenchymal stem proteoglycans. Clin Orthop Relat Res 430:219226
Gene Therapy Approaches
for Disc Regeneration 24
Zulma Gazit, Nadav Kimelman-Bleich, Olga Mizrahi,
and Dan Gazit
Contents 24.4.2 In Vivo Gene Delivery of Growth

and Transcription Factors: Animal Models....................... 394
24.1 Introduction: Use of Gene Therapy
for the Treatment of Disease .......................................... 385 24.5 Use of Gene Therapy for Intervertebral
Disc Repair: Theoretical
24.2 Importance of Reporter Genes and Practical Consideration........................................... 394
for Studies of Gene Delivery........................................... 386
24.2.1 In Vitro Approaches Using Animal Cells ......................... 386 24.6 Summary of Critical Concepts Discussed
24.2.2 Use of Reporter Genes for Assessment of in the Chapter .................................................................. 397
Gene Delivery: In Vitro Studies Using
Human Cells...................................................................... 387 References ...................................................................................... 398
24.2.3 Use of Reporter Genes for Assessment
of Gene Delivery: In Vivo Studies
Using Animal Models ....................................................... 387
24.3 Delivery and Activity of Anti-inflammatory/ 24.1 Introduction: Use of Gene Therapy
Anti-degeneration Genes ................................................ 389 for the Treatment of Disease
24.3.1 Anti-inflammatory/Anti-degeneration
Gene Delivery into Animal and Human Cells:
In Vitro and In Vivo Studies.............................................. 389 One of the first articles describing the use of gene therapy
24.3.2 Fas-Fas Ligand (L) and the Intervertebral was published in Science in 1972. In that paper, Friedmann
Disc as an Immune-Privileged Tissue ............................... 390 and Roblin described how they induced mammalian cells
24.4 Gene Targeting of Growth and to express bacterial and viral DNA, demonstrating the
Transcription Factors ..................................................... 391 possibility that this method could later be used to promote
24.4.1 In Vitro Gene Delivery of Growth therapies for treatment of human genetic diseases
and Transcription Factors .................................................. 392
(Friedmann and Roblin 1972). Today, scientists are explor-
ing the therapeutic use of gene therapy for the treatment
Z. Gazit (*) D. Gazit of numerous pathological conditions including Parkinson
Skeletal Biotechnology Laboratory, disease (Yasuda et al. 2011), glioma (Lee et al. 2011),
Hadassah Faculty of Dental Medicine, Hebrew University,
X-linked severe combined immunodeficiency (SCID-X)
12272, Ein Kerem, Jerusalem 91120, Israel
e-mail: zulma.gazit@csmc.edu; dan.gazit@csmc.edu (Huston et al. 2011), diabetes (Rowzee et al. 2011), bone
fractures (Sheyn et al. 2008; Kimelman-Bleich et al.
Department of Surgery,
Regenerative Medicine Institute, Cedars-Sinai Medical Center, 2011), and heart failure (Muona et al. 2012; Pleger et al.
8700 Beverly Blvd AHSP-8108, Los Angeles, CA 90048, USA 2011). During the last two decades, efforts have been
N. Kimelman-Bleich O. Mizrahi made to enhance the safety and efficiency of gene therapy.
Skeletal Biotechnology Laboratory, Despite numerous advances, the immune response, which
Hadassah Faculty of Dental Medicine, Hebrew University, led to the death of a patient in one clinical trial (Somia
12272, Ein Kerem, Jerusalem 91120, Israel
and Verma 2000), and insertional mutagenesis, which
e-mail: nadav.kimelman@gmail.com; golzand@gmail.com

386 Z. Gazit et al.
triggered the development of leukemia in several patients preliminary proof of the therapeutic potential of gene
(Hacein-Bey-Abina et al. 2003; Cavazzana-Calvo et al. therapy in the organ. However, until now no clinical trial
2004), continue to pose a major barrier to the clinical use involving gene therapy for disc disease or disorder has
of gene therapy. Fortunately, improvements are constantly been listed in the clinical trial database of the NIH
being made in protocols as well as in the safety features of (www.clinicaltrails.gov).
some vectors, such as exogenous control of gene expres- Currently, three groups of genes are being used in stud-
sion or the use of nonviral gene delivery methods (Herzog ies that explore intervertebral disc gene therapy. First,
et al. 2010). Hundreds of clinical trials involving gene reporter genes (Wehling et al. 1997), such as green
delivery are currently being conducted in the USA for the fluorescence protein (GFP) and luciferase, are used for
treatment of various diseases, ranging from cystic fibrosis proof-of-concept studies to demonstrate the feasibility of
to HIV (Flotte 2007). gene delivery and to further understand how gene insertion
Although most developments in gene therapy have as influences the cells phenotype. Second, anti-inflammatory
their aim treatment of systemic diseases and disorders, a or anti-degeneration genes, such as interleukin-1 (IL-1),
few focus on tissue regeneration (Edelstein et al. 2007). In tissue inhibitors of metalloproteinases-1 (TIMP-1), and Fas
orthopedics, gene therapy is studied for its usefulness in ligand (FasL) (Wallach et al. 2003b; Le Maitre et al. 2007;
bone-tissue engineering (Bueno and Glowacki 2009), Seki et al. 2009; Suzuki et al. 2009), are used to stop the
because even transient expression of the bone morphoge- degeneration process or to minimize it. Third, anabolic
netic protein (BMP)-2, BMP-6, or BMP-9 gene is sufficient genes, such as growth and differentiation factor-5 (GDF- 5),
to promote bone generation. Gene therapy is also being BMPs, transforming growth factor-beta (TGF-b), and Sox-
explored for its value in creating cartilage (Menendez 9, are used not only to mitigate the degenerative process but
et al. 2011) and tendon (Xia et al. 2010), although success also to promote tissue regeneration (Nishida et al. 1999;
has been limited when direct gene delivery approaches are Lee et al. 2001; Paul et al. 2003; Cui et al. 2008; Reddi and
used. Reddi 2009; Wang et al. 2011). In this chapter, insight is
Viral vectors, primarily adenovirus and retrovirus, are provided into the current status of intervertebral disc gene
the main vectors used in clinical trials involving gene ther- therapy. Also discussed are future directions and current
apy (Edelstein et al. 2007). However, to avoid the safety hurdles that need to be surmounted to promote gene therapy
issues associated with viral vectors (discussed earlier), as a therapeutic option for intervertebral disc
there is increasing reliance on nonviral techniques (Edelstein degeneration.
et al. 2007). Nonviral vectors are easy to prepare and scale
up for production (Schmidt-Wolf and Schmidt-Wolf 2003;
Aslan et al. 2006). They also exhibit a high potential for 24.2 Importance of Reporter Genes
orthopedic applications, since bone formation requires for Studies of Gene Delivery
short-term expression of osteogenic genes (Noel et al.
2004). Gene therapy can be performed either directly, by 24.2.1 In Vitro Approaches Using Animal Cells
injection of a gene-carrying vector into the target area, or
ex vivo, by inserting the gene into harvested cells and then Several studies have been conducted to deliver reporter
implanting the modified cells into the desired tissue genes into disc cells that have been isolated from animal
(Wehling 2001). Current studies are underway to explore tissues. The goal of these experiments is usually to evaluate
various gene delivery techniques. the feasibility and limitations of gene delivery into disc-
One of the first studies in which gene therapy was used derived cells, specifically annulus fibrosus, nucleus pulpo-
for the treatment of intervertebral disc degeneration was sus, and end plate chondrocytes. Early attempts at delivery
published in 1997 (Wehling et al. 1997). End plate chon- of reporter genes to intervertebral disc cells used a retrovi-
drocytes were isolated and induced to express an anti- ral vector with a bacterial LacZ gene which was transfected
inflammatory gene using a retrovirus. This was a into bovine end plate cells (Wehling et al. 1997) and deliv-
feasibility study demonstrating that gene therapy can be ery of the same gene into rabbit nucleus pulposus cells
applied to intervertebral disc cells. Since then, many using adenovector (Nishida et al. 1998). While only 1 % of
studies have been conducted to locate target genes, the end plate cells were successfully transduced, the use of
improve gene delivery mechanisms, and provide adenovector yielded much higher levels of LacZ expression
24 Gene Therapy Approaches for Disc Regeneration 387
(detected by X-gal staining) after in vivo infection (Wehling reporter into human nucleus pulposus cells, with high
et al. 1997; Nishida et al. 1998). In another study, the effect efficiency (Morrey et al. 2008).
of different doses of baculovirus encoding the GFP reporter
gene in rabbit nucleus pulposus cells was evaluated. A
direct correlation between GFP expression and the multi- 24.2.3 Use of Reporter Genes for Assessment
plicity of infection (MOI) of the vector was noted up to a of Gene Delivery: In Vivo Studies Using
maximum of 87 % expressing cells at an MOI of 200, with Animal Models
gene expression observable up to 21 days after infection
(Liu et al. 2006). RNA interference (RNAi), first described Nishida and colleagues reported the first successful in vivo
at 1998 (Fire et al. 1998), was also evaluated as a gene ther- gene delivery into an intervertebral disc in 1998 (Wallach
apy tool, aimed at silencing specific genes involved in the et al. 2003a, b). After in vitro delivery of the LacZ reporter
disc degeneration process. Kakutani and colleagues used gene into rabbit nucleus pulposus cells using an adenovec-
small interfering RNA (siRNA, 21 nt in length) to assess tor, the same virus was injected into the intervertebral discs;
the use of this strategy in nucleus pulposus cells that had gene delivery was analyzed up to 3 months postinjection.
been isolated from rats and humans (Kakutani et al. 2006). Gene expression was evident, and no pathological changes
These researchers showed that after a single transfection, were noted in the injected discs (Nishida et al. 1998).
siRNA could prevent luciferase and GFP expression for Moreover, in a follow-up study, protein expression was
2 weeks. Moreover, no differences were found between rat detected up to 1 year after gene delivery. The fact that no
and human nucleus pulposus cells (Kakutani et al. 2006). pathological changes were noted was encouraging (Wallach
Based on these studies, it is concluded that siRNA may repet al. 2003a, b). Another paper described the use of the
resent a therapeutic strategy which could stop or slow disc luciferase reporter gene in noninvasive monitoring of gene
degeneration. expression following both ex vivo (cell mediated) and
in vivo (adenovector injection) gene delivery into rat discs
(Leo et al. 2004). The authors compared direct injection of
24.2.2 Use of Reporter Genes for Assessment the viral vector with the use of three different cell types as
of Gene Delivery: In Vitro Studies Using gene vehicles: (1) disc cells, a mix of nucleus pulposus and
Human Cells annulus fibrosus cells; (2) adipose-derived rat mesenchymal
stem cells (ASCs); and (3) bone marrow-derived rat mesen-
Following studies performed on animal cells, researchers chymal stem cells (MSCs). Of the three groups of cells, the
turned their attention to the delivery of reporter genes to most potent gene expressers were the genetically induced
human intervertebral disc cells. Moon and colleagues disc cells (Leo et al. 2004). This was the first report of bio-
were among the first to analyze the value of adenovectors luminescence imaging performed in the intervertebral disc,
for gene delivery into human disc cells (nucleus pulposus and it demonstrated the feasibility of noninvasive imaging
and annulus fibrosus cells) isolated from healthy and of gene expression in this tissue. The use of GFP-encoding
degenerate discs. They reported that in vitro, an aden- baculovirus as a gene delivery vehicle was also tested
ovector harboring the LacZ or luciferase reporter genes in vivo. GFP expression in the intervertebral disc was noted
was able to transduce 100 % of cultured cells, regardless as long as 13 days after gene delivery (Liu et al. 2006).
of the discs degenerative state or the anatomical location Following these feasibility studies, the effect of dosing and
of the transfected intervertebral disc cells (Moon et al. vector choice on the safety aspects of gene delivery into the
2000). As it became clear that human disc cells could be disc was evaluated. When different doses of both adenovec-
genetically engineered, efforts were made to use more tor and adeno-associated viral (AAV) vector encoding either
complex expression strategies, such as exogenous regula- GFP or LacZ were injected into rabbit intervertebral discs,
tion of gene expression. Vadala and colleagues reported no clinical, biochemical, or histological pathological deficits
that a Tet-Off system, in which gene expression is shut were noted. However, this was not the case when BMP-2 or
down by tetracycline could be used to control GFP TGF-b was delivered (Levicoff et al. 2008). One take home
expression in human nucleus pulposus cells (Vadala et al. message from these studies is that a range of markers need
2007). Lipid-based nonviral gene delivery systems were to be evaluated to assess the safety of gene delivery
also used to deliver plasmid DNA encoding the GFP systems.
388 Z. Gazit et al.
Box 24.1 Bioluminescence: BLI
4.5E+04
4.0E+04
3.5E+04
3.0E+04 Day 2
R.L.U
2.5E+04
2.0E+04
1.5E+04
1.0E+04
5.0E+03
Day 14
0.0E+00
Day 3 Day 6 Day 14 Day 20
4.5E+04
4.0E+04
3.5E+04
3.0E+04 Day 2
R.L.U
2.5E+04
2.0E+04
1.5E+04
1.0E+04
5.0E+03
Day 14
0.0E+00
Day 3 Day 6 Day 14 Day 20
Survival of luciferase-expressing stem cells following implantation in culture or injected into the implantation area in the
into the intervertebral disc region. Stem cells expressing the animal. In vitro and in vivo imaging is performed using
luciferase gene were implanted in a caudal disc space in the rat. At
each time point listed, luciferin was injected into the implantation a cooled charge-coupled device (CCCD), which can
site. Images and quantified data were generated using the BLI sys- detect light (photons) emitted from tissue following
tem. Arrows indicate luciferase-expressing cells in the disc region. luciferin degradation by luciferase (Sheyn et al.
RLU relative luminescence unit 2011).
Bioluminescence imaging can be used to monitor cell
Reference
survival over time both in vitro and in vivo. To apply
Sheyn D, Kallai I et al (2011) Gene-modified adult stem
this method, luciferase-expressing cells are required.
cells regenerate vertebral bone defect in a rat model.
Following gene delivery and cell culture or implanta-
Mol Pharm 8:15921601
tion, the luciferase substrate luciferin is added to cells
Following the use of viral vectors, there was an assess- usually transient was probably due to low cellular meta-
ment of nonviral vectors for gene therapy. When GFP- and bolic activity and proliferation in the disc. It also demon-
luciferase-encoding plasmid DNAs were delivered into rat strated the advantage of using nonviral gene delivery
discs via sonoporation (an ultrasound technique that techniques in this organ.
enhances gene delivery), prolonged expression of the Exogenously controlled gene delivery was assessed not
reporter genes was noted. Expression of GFP was evident only in vitro (Vadala et al. 2007) but also in vivo (Sowa et al.
after 7 days, and luciferase activity was recorded up to 2011). A novel ligand-based system for gene expression con-
24 weeks following gene delivery (Nishida et al. 2006). This trol (RheoSwitch) was successfully used in a rabbit model.
finding remarkable because nonviral gene delivery is GFP expression was controlled both in vitro and in vivo,
demonstrating this systems potential and enhanced safety 24.3.1 Anti-inammatory/Anti-degeneration

features (Sowa et al. 2011). Gene Delivery into Animal and Human
Finally, ex vivo gene delivery of reporter genes was used Cells: In Vitro and In Vivo Studies
to assess gene delivery into discs in a preclinical large-ani-
mal model. Omlor and colleagues used a retroviral vector In 2003, Wallach et al. (2003b) isolated cells from degener-
to generate stable luciferase-expressing porcine MSCs. ate intervertebral disc obtained from patients undergoing
These cells were subsequently implanted into the interver- elective surgical procedures and infected the cells with an
tebral discs of Gottingen minipigs. The animals were adenoviral-tissue inhibitor of metalloproteinase-1 (Adeno-
sacrificed on the same day or 3 days later, and a luciferase TIMP-1) construct. Gene delivery of TIMP-1 increased pro-
assay was performed to assess gene activity. Activity on teoglycan synthesis in a doseresponse manner. Disc cells
Day 3 postimplantation was reduced to 7 % of initial val- treated with Adeno-TIMP-1 demonstrated an optimal
ues, indicating massive cell loss following implantation response at an MOI of 100; cell pellets treated with Adeno-
(Omlor et al. 2010). TIMP-1 at 100 MOI were consistently larger than controls
Recently, siRNA was also used in vivo in a rat model to and pellets treated with other virus concentrations. Successful
analyze the feasibility of this gene-silencing tool in the inter- delivery of the anticatabolic gene TIMP-1 resulted in
vertebral disc. The siRNA downregulated the activity of the increased levels of proteoglycan in three-dimensional cul-
luciferase reporter gene up to 24 weeks following transfec- tures of degenerate disc cells. It is likely that TIMP-1 had no
tion performed using the ultrasound technique. At that time, direct effect on proteoglycan synthesis, but rather inhibited
gene activity remained at 80 % of control values (Suzuki proteoglycan degradation, thus promoting its accumulation
et al. 2009). These findings indicate that siRNA can be used in the culture. In addition, these results may reflect a second-
in the disc with high efficiency. ary effect, that is, inhibition of aggrecanase-1 (ADAMTS-4),
which is also inhibited by the TIMP family, as reported else-
where (Hashimoto et al. 2001).
24.3 Delivery and Activity of Anti- In a more recent study, Liu et al. (2010) investigated the
inammatory/Anti-degeneration Genes effect of combined connective tissue growth factor (CTGF)
and TIMP-1 expression mediated by AAV. These researchers
As discussed in other chapters of the book, during disc isolated rhesus monkey lumbar disc nucleus pulposus cells;
degeneration there is a progressive loss of water, proteogly- cultured and transduced the cells with rAAV2-CTGF-IRES-
cans, and collagen II from the matrix of the nucleus pulpo- TIMP-1 (a vector that elicits both CTGF and TIMP-1 expres-
sus. Although the events that lead to these degenerative sion), rAAV2-CTGF, or rAAV2-TIMP-1 at an MOI of 106;
changes have not been clearly defined, there is overwhelm- and then measured the expression of collagen II and proteo-
ing evidence to indicate that inflammatory mediators are glycan using reverse transcription polymerase chain reaction
elevated during degeneration. Kang et al. (1997) showed that (RT-PCR) analysis and Western blotting. By comparing
cells obtained from normal, nondegenerate human discs and transduced with nontransduced cells, these researchers
cultured for 72 h with interleukin (IL)-1b were biologically showed that CTGF induced the synthesis of collagen II and
responsive and increased their production of matrix metal- proteoglycan, whereas TIMP-1 enhanced expression of pro-
loproteinases, nitric oxide, IL-6, and prostaglandin E2. The teoglycan, but had no effect on collagen. Expression of both
effect was significantly higher in normal, nondegenerate genes in the lumbar disc nucleus pulposus cells significantly
discs, in which spontaneous synthesis of these mediators is increased the synthesis of proteoglycan and collagen II. In
low. These agents are inhibitors of proteoglycan synthesis, this particular case, single gene transduction of CTGF or
but IL-1 is likely to have a direct effect on proteoglycan TIMP-1 increased proteoglycan synthesis (Lee et al. 2001).
breakdown through activation of members of the metallopro- Overexpression of CTGF caused an increase in collagen II
teinase (MMP) family of enzymes. protein synthesis. Combined transduction of both CTGF and
Aside from the interleukins, Seguin et al. (2005) studied TIMP-1 significantly promoted the expression of proteogly-
the in vitro effect of TNF-a on nucleus pulposus tissue from can and collagen II to levels greater than that achieved using
bovine caudal spines. These studies reported decreased pro- each of the individual genes. This finding is of considerable
teoglycan and collagen gene expression and protein synthe- importance as it supports the concept that one approach to
sis and activation of aggrecanase-mediated proteoglycan preventing degradation is by promoting inhibition of cata-
degradation, involving MMPs, and ADAMTS. Since MMP bolic processes. It thus serves as a promising avenue of
activity is normally inhibited within the matrix by tissue research for the treatment of degenerative disc disease via
inhibitors of MMPs (TIMPs) (Nagase and Woessner 1999), gene therapy.
it is reasonable that members of this family were used to IL-1 is an inflammatory mediator implicated in the
explore whether candidate genes would stop or slow the disc degeneration of intervertebral discs, especially the nucleus
degradation process. pulposus matrix (Elfervig et al. 2001; Shen et al. 2003;
390 Z. Gazit et al.
Jimbo et al. 2005; Le Maitre et al. 2005). In 2006, Le Maitre Histological studies supported the MRI results, showing par-
et al. targeted the IL-1 receptor using an IL-1 receptor antag- tial maintenance of nucleus pulposus tissues in the ADAMTS5
onist (IL-1Ra) gene that was transferred ex vivo to the inter- siRNA-injected group, compared with the control group.
vertebral disc. Normal and degenerate human disc cells,
growing in a monolayer or in alginate beads, were infected
with an adenoviral vector carrying the IL-1Ra gene (Adeno- 24.3.2 Fas-Fas Ligand (L) and the Intervertebral
IL-1Ra). Protein production was measured, and the ability of Disc as an Immune-Privileged Tissue
the infected cells to inhibit the effects of IL-1 was assessed.
In addition, normal and degenerate intervertebral disc cells Since the concept of immune privilege is of high relevance in
infected with Adeno-IL-1Ra were injected into degenerate the biology of the intervertebral disc, the concept is discussed
disc tissue explants, and IL-1Ra production in these discs further in the accompanying box. Early in 1995, Griffith et al.
was evaluated. Nucleus pulposus cells and annulus fibrosus (1995) published a paper showing that Fas-FasL interactions
cells infected with Adeno-IL-1Ra produced elevated levels appear to be an important mechanism for the maintenance of
of IL-1Ra for prolonged time periods, and the infected cells immune privilege. That publication had as its focus the com-
were unaffected by IL-1. IL-1Ra protein expression was partmentalization of tissues of the eye. In a related publication,
increased and maintained for 2 weeks when infected cells Takada and his group reported that FasL exists in the interver-
were injected into disc explants. In a further study by the tebral disc and confers immune privilege (Takada et al. 2002).
same group (Le Maitre et al. 2007), Adeno-IL-1Ra was In their study, Takada et al. (2002) used immunohistochemical
introduced into degenerate disc explants directly or via and RT-PCR analysis to ascertain if FasL was present in human
genetically engineered nucleus pulposus cells. This resulted and rat disc specimens. They reported that nucleus pulposus
in cessation of degenerate intervertebral disc matrix, degra- cells of human and rat origin exhibited intense positive reac-
dation; in situ zymographic and immunohistochemical inves- tions for FasL. Outer annulus fibrosus cells and notochordal
tigations showed that there was downregulation of MMP-1, cells exhibited weak immunostaining. The results of RT-PCR
MMP-3, MMP-7, and MMP-13 expression as well as confirmed the existence of FasL in the rat disc cells, but analy-
ADAMTS4 expression. Single injections of disc cells engi- sis was not performed on human tissues.
neered to overexpress IL-1Ra significantly inhibited degra- Results of investigations reported by Han et al. (2009)
dation enzyme expression for 2 weeks. These studies showed that FasL-induced apoptosis of rat nucleus pulposus
indicated that since IL-1 is a key cytokine that promotes cells was sharply reduced by Fas siRNA, suggesting that it is
matrix degradation, direct delivery of IL-1Ra or direct deliv- feasible to suppress nucleus cell death using an RNAi approach.
ery by gene therapy provides a powerful new way to prevent The authors recognized that this was successful in short-term
or inhibit disc matrix degradation. in vitro assays, but for research purposes and especially as a
In 2005 Glasson and colleagues reported on the preven- therapeutic strategy a much longer period of inhibition may be
tion of cartilage degradation in a murine model of required. Because nucleus pulposus cells reportedly express
osteoarthritis by deletion of the ADAMTS5 a disintegrin and FasL constitutively and since FasL is recognized as a marker
metalloproteinase with thrombospondin motifs gene gene, a study led by Suzuki et al. (2009) investigated the appli-
(Glasson et al. 2005). To pursue the goal of looking for addi- cability of RNAi technology to downregulate the expression
tional candidate genes that would stop or slow the disc of this endogenous gene in rat intervertebral discs in vivo. As
degeneration process, the effect of ADAMTS5 was investi- mentioned earlier in this chapter, the Suzuki group previously
gated by Seki using siRNA oligonucleotide in a rabbit annu- used RNAi in cultured nucleus pulposus cells in vitro, target-
lus needle-puncture model (Seki et al. 2009). The efficiency ing two exogenous reporter luciferase plasmids, firefly and
of the siRNA constructs was first evaluated by growing rab- Renilla (Kakutani et al. 2006). For the endogenous FasL gene
bit NP cells in culture and transfecting them with the siRNA study, siRNAs that target rat FasL was injected into coccygeal
oligonucleotide specific to ADAMTS5. Compared with con- intervertebral discs. After the injection, therapeutic ultrasound
trol, the ADAMTS5 siRNA-transfected cells displayed an was applied to the skin surface covering the injected discs.
approximately 75 % decrease in levels of ADAMTS5 mRNA. There was significant inhibition (53 %) of endogenous FasL
For the in vivo portion of the study, the nucleus pulposus was gene expression compared with the control group, which was
punctured, and 7 days later the ADAMTS5 siRNA oligonu- transfected with nonspecific siRNA, at both 4 and 20 weeks
cleotide was injected into the rabbit nucleus. The early tim- posttransfection. One important conclusion from this study is
ing of the injection was to ascertain the impact of ADAMTS5 that there is long-term downregulation, which is mediated by
siRNA during the acute phase of disc degeneration. Eight siRNA for the endogenous gene Fas in discs in vivo. Besides
weeks after the siRNA oligonucleotide injection, the rabbit the techniques used for Fas-FasL genetic manipulation, it is
spines were examined ex vivo by using MRI. The T2 signal important to take into consideration the actual debate on the
intensity was stronger in ADAMTS5 siRNA-injected nucleus paradoxical role of FasL. Reports have indicated that FasL is
pulposi than in the control siRNA-injected nucleus. involved in the formation of the intervertebral disc and its
immune privilege, and also that the expression of FasL is

decreased in degenerate discs, but does not vary with age Gores PF, Hayes DH et al (2003) Long-term survival
(Kaneyama et al. 2008). On the other hand, FasL has also been of intratesticular porcine islets in nonimmunosup-
found to have a close relationship with disc cell apoptosis and pressed beagles Transplantation 75:613618
is seen as another factor involved in disc degeneration (Han Hong, SL, Van Kaer L (1999) Immune privilege: keep-
et al. 2009). ing an eye on natural killer T cells. J Exp Med
190:11971200
Hori J, Vega JL et al (2010) Review of ocular immune
privilege in the year 2010: modifying the immune
Box 24.2 Immune Privilege privilege of the eye. Ocul Immunol Inflamm
Immune privilege (IP) is the ability of certain organs to 18:325333
suppress the immune response to foreign bodies or Novak N, Haberstok J et al (2008) The immune privilege
organisms within their specific boundaries. IP was first of the oral mucosa. Trends Mol Med 14:191198
described by Sir Peter Medawar in 1948, based on the Zheng J, Umikawa M et al (2011) Ex vivo expanded
prolonged survival of an allogenic tissue graft that had hematopoietic stem cells overcome the MHC bar-
been placed in the anterior chamber of the eye (Hori rier in allogeneic transplantation. Cell Stem Cell
et al. 2010). Additional sites of IP include the brain, 9:119130
pregnant uterus, and testis. Recent studies have led to
the identification of more tissues with IP-like character-
istics (i.e., tissues that adopt IP mechanisms and hence
avoid a destructive immune response) such as the gut, 24.4 Gene Targeting of Growth
skin, and lung (Arck et al. 2008). The oral mucosa dis- and Transcription Factors
plays a similar tolerance: it rarely exhibits acute
inflammatory or allergic reactions to high levels of bac- To maintain the health and integrity of the intervertebral disc,
terial colonization or frequent contacts with allergens both in the nucleus pulposus and annulus fibrosus, there should
(Novak et al. 2008). Although IP was first thought to be be a balance among systems that regulate the synthesis, break-
a passive process, it soon became evident that many down, and accumulation of macromolecules in the matrix.
active immunoregulatory processes take place in main- When reviewing work published to date on the gene therapy
tenance of IP (Hong and Van Kaer 1999; Arck et al. approach to treat disc degeneration, four main intracellular
2008; Novak et al. 2008; Hori et al. 2010). This phe- regulators were found: BMPs and TGF-b as morphogens and
nomenon is believed to be an evolutionary protective LMP-1 and Sox-9. Growth factors are polypeptides that bind
mechanism designed to prevent a destructive to cell membranes and regulate matrix production and regen-
inflammatory immune response in critical parts of the eration of various cell types in paracrine and autocrine path-
body (Hong and Van Kaer 1999; Arck et al. 2008). The ways (Masuda and An 2004) (see also Chap. 25). In normal
potential clinical use of IP is clear. From allogenic cor- conditions, intervertebral disc cells express and secrete ana-
nea replacement (Hori et al. 2010) to implantation of bolic growth factors to maintain the balance between normal
xenogenic pancreatic islets performed in both rodents matrix synthesis and degradation. Included in this group of
and large-animal models (Gores et al. 2003), the possi- molecules is insulin-like growth factor (IGF), transforming
bility of exploiting sites of IP to avoid the immune growth factor-beta (TGF-b), and members of the BMP family
response to implanted tissue is attractive. In addition, (Thompson et al. 1991; Osada et al. 1996; Cui et al. 2008).
the inherent IP properties of stem cells overcome the Another group of targets includes transcriptional factors that
host immune rejection response, as in the case of ex are master regulators of the chondrogenic and osteogenic
vivo expanded hematopoietic stem cells, which when phenotype. Sox-9 is one such factor that has received study for
used in allogenic transplantation in a mouse model sur- cartilage and intervertebral disc basic and preclinical research;
mounts the major histocompatibility complex barrier retroviral expression of Sox-9 can also efficiently induce ASCs
(Zheng et al. 2011). Translated to humans, this strategy differentiation into chondrocyte-like cells. Thus, Sox-9 is a
can significantly enhance the availability of implanted candidate gene for use as a possible therapeutic tool for the
tissues and organs as well as promote novel therapies. treatment of degenerative disc diseases (Yang et al. 2011).
LIM mineralization protein-1 (LMP-1) is another regulatory
References protein that positively regulates BMP secretion (Boden et al.
Arck PC, Gilhar A et al (2008) The alchemy of immune 1998). The activities of both anabolic growth factors and
privilege explored from a neuroimmunological per- selected transcription factors have been used to promote regen-
spective. Curr Opin Pharmacol 8:480489 eration through augmentation of matrix production thereby
impacting disc height and function (Zhang et al. 2006).
392 Z. Gazit et al.
24.4.1 In Vitro Gene Delivery of Growth 11, BMP-12, BMP-13, BMP-14, and BMP-15) on extracel-
and Transcription Factors lular matrix accumulation by bovine intervertebral disc cells.
Of the 12 vectors evaluated by these researchers, BMP-2 and
Most experiments conducted in this field are still performed BMP-7 (also known as OP-1) were best at stimulating pro-
in vitro. Research is focused on finding the set of genes and teoglycan accumulation in nucleus pulposus cells, whereas
cells that will elicit the most promising results (Zhang et al. BMP-4 and BMP-14 (also known as GDF-5) optimally stim-
2006; Bron et al. 2009; Zhang et al. 2009). Most experiments ulated collagen production (Zhang et al. 2006, 2009). Similar
are conducted in 3D cultures known to support chondrogen- results were seen using the AAV-BMP-7 vector in canine
esis and nucleus pulposus-like differentiation (Risbud et al. nucleus pulposus cells (Wang et al. 2011). In another study,
2004). Another established method for in vitro differentia- researchers injected knee chondrocytes transduced with
tion is a whole-disc culture model, which mimics the in vivo Adeno-BMP-7 or Adeno-BMP-10 into whole intervertebral
environment (Zhang et al. 2008). disc explants cultured in vitro. Discs treated with chondro-
Originally identified by their ability to induce the forma- cytes expressing BMP-7 demonstrated a 50 % increase in
tion of bone and cartilage, BMPs are regarded today as key proteoglycan within the nucleus pulposus, whereas discs
signaling proteins responsible for organization of tissue treated with chondrocytes expressing BMP-10 evidenced no
architecture throughout the entire body. Today, 20 proteins increase in proteoglycan accumulation (Zhang et al. 2008).
are known to be associated with this group. The Food and GDF-5 (BMP-14) is known for its participation in joint
Drug Administration has recently allowed the initiation of formation and endochondral ossification (Cui et al. 2008). It
Investigational New Drug clinical trials on osteogenic pro- is also known that a deficiency in GDF-5 leads to abnormali-
tein-1 and growth differentiation factor-5 in the United States ties in intervertebral disc structure (Li et al. 2004). Studies
(Reddi and Reddi 2009; Zhang et al. 2011). Zhang and col- have shown that overexpression of GDF-5 in disc cells using
leagues compared the effects of overexpression of recombi- viral (adenoviral vector) and nonviral (nucleofection) meth-
nant adenoviral vectors expressing various BMPs (BMP-2, ods augments expression of proteoglycan proteins and col-
BMP-3, BMP-4, BMP-5, BMP-7, BMP-8, BMP-10, BMP- lagen (Cui et al. 2008; Feng et al. 2009). LMP-1 is an
Box 24.3 Viral Vector Used to Monitor Bone Formation
a
2
Luc
EcoRV
Lenti-hOc-
hOc Luc2
Ampr Scal
Eam1105
Kpnl
Notl Nrul
b
Day 1 2 Weeks 4 Weeks 6 Weeks
3.5 1.6
3.5
3.0 1.4 0.8
3.0
2.5 1.2
105 2.5 0.6
2.0 1.0 105
107 105
1.5 2.0 0.8 0.4
1.0 1.5 0.6
0.2
p/s/cm2/sr p/s/cm2/sr p/s/cm2/sr p/s/cm2/sr
c Luc expression in pASc-OC-Luc2 cells

*
*
1.E+08 *
8.E+07
Total flux (p/s)
6.E+07
4.E+07
2.E+07
0.E+00
Day 1 2 weeks 4 weeks 6 weeks
Vector used to monitor bone formation when applied in vertebral luciferase gene were implanted in a bone defect created in a rat
bone defect regeneration. (a) Osteocalcin-driven luciferase con- caudal vertebra. At each time point displayed, luciferin was injected
struct. To monitor bone formation noninvasively in real time, the in situ. Images and quantified data were generated using the BLI
osteocalcin promoter was cloned into the commercial reporter Luc2 system. (Figures b and c are reprinted (adapted) with permission
vector (pGL.4, Promega). The luciferase gene was expressed as new from Sheyn et al. (2011). Copyright (2011) American Chemical
bone formed. (b, c) Stem cells expressing the osteocalcin-driven Society
Osteocalcin is one of the pivotal genes expressed dur- References

ing new bone formation in rodents. In transgenic mice, Iris B, Zilberman Y et al (2003) Molecular imaging of the
when the osteocalcin promoter is used to control skeleton: quantitative real-time bioluminescence mon-
expression of the reporter gene luciferase, bone forma- itoring gene expression in bone repair and develop-
tion (or osteogenic activity) can be monitored using the ment. J Bone Miner Res 18:570578
BLI system (Iris et al. 2003; Kimelman-Bleich et al. Kimelman-Bleich N, Pelled G et al (2009) The use of a
2009). The activity of implanted cells can also be moni- synthetic oxygen carrier-enriched hydrogel to enhance
tored using the osteocalcin-driven luciferase gene mesenchymal stem cell-based bone formation in vivo.
(Sheyn et al. 2011). Biomaterials 30:46394648
Sheyn D, Kallai I et al (2011) Gene-modified adult stem
cells regenerate vertebral bone defect in a rat model.
Mol Pharm 8:15921601
intracellular regulatory molecule known to induce secretion development and tissue homeostasis, including growth, pat-
of multiple BMPs from leukocytes and osteoblasts (Boden terning, and cellular differentiation. Three isoforms of TGF-b
et al. 1998). Boden and colleagues used this activity to (1, 2, and 3) act through the same heteromeric receptor com-
increase proteoglycan production by intervertebral disc cells plex (Baffi et al. 2004). TGF-b signaling plays a key role in
(Yoon et al. 2004; Kuh et al. 2008). They have shown that the development and maintenance of cartilage and, in par-
transduction of rat disc cells in a 3D culture or in a monolayer ticular, the intervertebral disc (Nishida et al. 1999). Lee and
with Adeno-LMP1 results in significant increases in proteo- colleagues examined the effect of TGF-b on human disc
glycan and aggrecan mRNA levels as well as elevations in cells in vitro. Disc cells were transfected with an adenovec-
BMP-2 and BMP-7 mRNA levels. tor expressing TGF-b, after which the cells were grown in
A nonviral gene delivery system is a promising alternative 3D pellets. In response to this treatment, the cells increased
that avoids the risks of insertional mutagenesis of retrovi- synthesis of proteoglycan and collagen II (Lee et al. 2001).
ruses, immunogenicity of adenoviruses, and acquisition of Although positive results have been achieved with gene
replication competence (Nishida et al. 2000). Using a gene therapy, mainly when using adenoviral vectors, transfer of a
gun, Matsumoto et al. demonstrated successful transfection single gene using high doses of viral vector can produce haz-
of bovine intervertebral disc cells with BMP-7, which led to ardous systemic effects, in particular cytotoxicity, and an
a higher proteoglycan content (Matsumoto et al. 2001). immune response (Lohr et al. 2001). It is likely that these
In common with the BMPs discussed above, TGF-b is a effects can be mitigated by administration of several vectors
secreted signaling protein that regulates many aspects of so as to decrease the level of a single vector and possibly
394 Z. Gazit et al.
augment the regenerative potential. Moon and colleagues Gene expression of the delivered vector lasted up to 6 weeks. In
tested this concept in human intervertebral disc cells. The addition, delivery of the vector succeeded in arresting the
cells were transduced with different combinations of Adeno- decrease in disc height and loss of proteoglycan due to disc
TGF-b1, Adeno-BMP-2, and Adeno-IGF-1, after which the degeneration (Liang et al. 2010). Injection of Adeno-LMP-1 into
cells were placed in a 3D culture in alginate beads. Cells the native lumbar discs of a rabbit resulted in elevations of
treated with double and triple gene combinations were found LMP-1, BMP-2, and BMP-7 mRNA levels (Yoon et al. 2004).
to have higher proteoglycan synthesis rates than cells treated Surprisingly, to date, there is only one paper reporting on
with a single vector, hence allowing the use of less viral degenerated rabbit discs injected with adenoviral vector
vector (Moon et al. 2008). expressing the Sox-9 gene. The Sox-9-treated discs retained
Sox-9 is another transcription factor that plays a key role their chondrogenic appearance, unlike control discs
in the differentiation process. This transcription factor posi- injected with Adeno-GFP vector or control degeneration-
tively controls collagen II synthesis and is a key gene for induced discs (Paul et al. 2003).
chondrogenesis, thus suggesting that its expression could be
used to promote nucleus pulposus cell survival (Bell et al.
1997; Paul et al. 2003; Yang et al. 2011). After infection with 24.5 Use of Gene Therapy for Intervertebral
Adeno-Sox-9 vector, cells from a chondroblast line, human Disc Repair: Theoretical and Practical
disc cells derived from degenerated discs, and bovine nucleus Consideration
pulposus-derived cells demonstrated elevated collagen II
mRNA expression and raised protein levels (Paul et al. 2003; Concomitant with increased human life expectancy, disc
Zhang et al. 2006). In another study, rat ASCs were infected degeneration and associated spinal disorders are a major
with a retro-Sox-9 vector based on the murine leukemia health concern. Due to the limited regenerative capacity of
virus. In a 3D culture system, supplemented with TGF-b, the disc tissues, it is almost impossible to reverse or even
these genetically modified cells demonstrated increased lev- stop the process of degeneration, a topic extensively dis-
els of Sox-9 and collagen II. These researchers also cocul- cussed in this book. Thus, there is increasing interest in
tured these genetically modified ASCs with nucleus pulposus developing new biological approaches to regenerating the
cells. The coculturing technique has been shown to enhance degenerating disc. In this chapter, we have reviewed how this
differentiation toward a nucleus pulposus-like fate since the can be achieved through genetic modification of intradiscal
ASC secrete growth factors into the culture medium. When gene expression via gene therapy. Other therapeutic
this was performed, a marked increase in proteoglycan and approaches to disc regeneration include injection of growth
collagen II production by the nucleus pulposus cells was factors, with or without a carrier, and use of mature cells,
noted (Yang et al. 2011). progenitor cells, and stem cells, with or without scaffolding,
topics addressed in other chapters of this book.
Accordingly, we have reviewed studies of intervertebral
24.4.2 In Vivo Gene Delivery of Growth and disc gene therapy performed in the last decade, summarized
Transcription Factors: Animal Models in Table 24.1. The vast majority of the investigations per-
formed in vitro use reporter genes as proof of concept.
There are few published studies dealing with gene therapy Several studies were conducted to deliver reporter genes into
for intervertebral disc regeneration in vivo. The experiments disc cells isolated from animal tissues. Those experiments
described in this section are all focused on early stages of addressed the feasibility and limitations of gene delivery into
development and were performed to evaluate the feasibility intervertebral disc-derived cells. LacZ and GFP genes were
and safety of using gene therapy as a tool for disc regenera- used in various animal and human cells and were delivered
tion. The model used for these studies is based on intradiscal via viral and nonviral methods. Small iRNA technology was
injection of a therapeutic vector into the degenerated inter- also used in vitro in rat and human nucleus pulposus cells to
vertebral disc and takes advantage of the avascularity and downregulate luciferase activity. Following studies per-
immune privilege of the disc (Nishida et al. 2008). formed on animal cells, the delivery of reporter genes to
Nishida and colleagues evaluated injections of Adeno- human disc cells was also pursued. Adenovector harboring
TGF-b into rabbit IVDs. The discs were harvested 1 week the LacZ or luciferase reporter genes transduced 100 % of
after virus injection and were subjected to histological analy- cultured human intervertebral disc cells, healthy or degener-
sis and biochemical assays. The injected discs showed exten- ate. Another in vitro study demonstrated that it is possible to
sive increases in TGF-b expression and production, as well as exert exogenous control over gene activity using human cells
in proteoglycan synthesis (Nishida et al. 1999). In a similar in vitro. To date, both viral and nonviral vectors have been
study, Adeno-GDF-5 or Adeno-Luc vector was injected used to deliver reporter genes into intervertebral discs in vivo.
directly into mouse lumbar degenerated intervertebral discs. When different doses of both adenovector and AAV encoding
Table 24.1 Summary table of genes, vehicles, experimental models, and main results obtained in the last decade
24
Gene Vector Experimental system Findings Author (year)

ADAMTS5 siRNA, specific for (a) In vitro, rabbit NP cells both in monolayer and (a) ADAMTS5 gene suppression was 70 % compared with the Seki et al. (2009)
ADAMTS5 alginate bead cultures control upon IL-1 stimulation
(a) Transfection in vitro
(b) Injection, in vivo (b) In vivo by using the rabbit annular needle-punc- (b) The injection of anti-ADAMTS5 oligonucleotide in vivo
ture model and intradiscal injection resulted in improved MRI scores
BMP-2 Adenovirus Efficiency of different BMPs in bovine NP cells in Most effective in stimulating proteoglycan accumulation Zhang et al. (2006)
culture
Cells from human degenerated disc cultured in Cells treated with Ad-BMP-2 demonstrated a progressive Wallach et al. (2003a)
monolayer increase in proteoglycan synthesis with increasing viral
concentrations
BMP-7 (OP-1) Adenovirus Efficiency of different BMPs in bovine NP cells in Best at stimulating proteoglycan synthesis Zhang et al. (2006),
culture Zhang et al. (2009)
Adeno-associated virus Canine NP cells growing in culture Promoted proteoglycan and collage-II production Wang et al. (2011)
(AAV)
Gene gun Bovine IVD cells cultured in monolayer High proteoglycan synthesis in IVD cells Matsumoto et al. (2001)
Gene Therapy Approaches for Disc Regeneration
BMP-14 Nucleofection Mouse disc cells cultured in monolayer Elevated collagen type II and aggrecan genes expression Cui et al. (2008)
(GDF-5) Adenovirus Efficiency of different BMPs in bovine NP cells in Best in stimulating collagen production Zhang et al. (2006)
culture
Rabbit knee chondrocytes injected in disc explants 50 % increase in proteoglycan within the NP Zhang et al. (2008)
Intradiscal injection to degenerated mouse IVD Successful arrest of the typical decrease in disc height and Liang et al. (2010)
proteoglycan content
CTGF AAV Rhesus monkey lumbar IVD NP cells in monolayer CTGF promoted the synthesis of collagen type II and Liu et al. (2010)
proteoglycan
FasL siRNA specific for FasL, siRNA delivered in vivo into coccygeal IVD of The siRNA transfection inhibited endogenous FasL expression Suzuki et al. (2009)
delivered by ultrasound SpragueDawley rats by 53 % compared with the control
IL-1Ra Adenovirus (a) Normal and degenerate human IVD cells (a) Cells infected with Ad-IL-1Ra produced elevated levels of LeMaitre et al. (2007)
growing in monolayer or alginate cultures IL-1Ra for prolonged time periods, and the infected cells were
resistant to IL-1
(b) Normal and degenerate IVD cells infected with (b) IL-1Ra protein expression was increased which was
Ad-IL-1Ra were injected into degenerate disc tissue maintained for 2 weeks
explants
LMP-1 Adenovirus Rat IVD cells cultured in vitro in 3D and monolayer Significant increase in proteoglycan and BMP-2 and BMP-7 Yoon et al. (2004), Kuh
mRNA levels et al. (2008)
Injection into native rabbit lumbar discs Elevation of LMP-1, BMP-2, and BMP-7 mRNA levels Yoon et al. (2004)
(continued)
395
396
Gene Vector Experimental system Findings Author (year)

Sox-9 Adenovirus Chondroblastic cell line and human degenerated Increased level of RNA and protein of collagen type II Paul et al. (2003),
disc cells and bovine NP cells Zhang et al. (2006)
Retro (murine leukemia) Rat ASCs cultured in 3D Increase levels of collagen type II and Sox-9 Yang et al. (2011)
virus Coculture of the transfected cells with NP cells Increased collagen type II and proteoglycan production
TGF- Adenovirus Human IVD cells cultured in 3D pellets Increased synthesis of proteoglycan and collagen II Lee et al. (2001)
Human IVD cells treated with different combina- Cells treated with triple and double combinations showed Moon et al. (2008)
tions of TGF- , BMP-2, and IGF and grown in 3D higher proteoglycan content
alginate culture
Chondroblastic cell line and human degenerated Increased level of RNA and protein of collagen type II Paul et al. (2003),
disc cells and bovine NP cells Zhang et al. (2006)
Intradiscal injection to rabbit IVD Increase in TGF-b expression and production and in proteogly- Nishida et al. (1999)
can synthesis 1 week after injection
Intradiscal injection into degenerated rabbit IVDs The injected discs retained their chondrogenic appearance Paul et al. (2003)
TIMP-1 Adenovirus Degenerated human IVD cells cultured in 3D pellet Increased measured proteoglycan in cultured degenerated disc Wallach et al. (2003b)
cells
AAV Rhesus monkey lumbar IVD NP cells in culture Enhancing effect on the expression of proteoglycan but no Liu et al. (2010)
effect on collagen type II
NP nucleus pulposus, AF annulus fibrosus, IVD intervertebral disc
Z. Gazit et al.
either GFP or LacZ were injected into rabbit discs, no clini- before focusing on clinical efficacy. Given what we have
cal, biochemical, or histological deficits were noted. learned to date, we postulate that in the future a good treat-
However, this was not the case when BMP-2 or TGF-b was ment strategy may improve vectors viral- or nonviral-
delivered (Levicoff et al. 2008). It is therefore probable that encoding multiple genes. Perhaps a combination of
the safety of gene delivery cannot be assessed using only anti-inflammatory and anabolic genes, which in different
reporter genes. Cell- or stem cell-mediated gene therapy is combinations and stoichiometry, will regenerate the extra-
another tactic of choice and is viewed as a more physiologi- cellular matrix or, if prevention should prove feasible, will
cal approach to regulating gene expression. Luciferase- mitigate the altered homeostasis brought about by the degen-
expressing porcine MSCs were implanted into a minipig eration process in the intervertebral disc.
disc, but massive cell loss followed implantation. siRNA
against the exogenous Luciferase gene was used in a rat
model in vivo. Because the luciferase gene was silenced for 24.6 Summary of Critical Concepts Discussed
24 weeks, the siRNA activity was active for a considerable in the Chapter
time period (Suzuki et al. 2009).
Despite these positive results, transitions into preclinical Almost all of the gene therapy studies in the last decade
and clinical studies require further safety and doseresponse were performed in vitro, using reporter genes, LacZ and
studies designed to mimic conditions anticipated in human GFP, in various animal and human cells, delivered via
disease. There is always concern that the proximity of the viral and nonviral methods, as a proof of concept.
disc to the spinal canal may result in exposure of cord cells In specific cases, gene delivery of TIMP-1 increased pro-
to the viral vector and/or therapeutic cDNA/protein (Wallach teoglycan synthesis at each concentration assessed.
et al. 2006). Thus, we find it relevant at this point to con- Adenoviral vector carrying the IL-1Ra gene can be trans-
sider the safety aspect of gene therapy as a therapeutic ferred to nucleus pulposus cells and annulus fibrosus cells;
approach to overcome disease. Overall, safety studies that of IL-1Ra levels were elevated for prolonged time periods
mimic possible hazardous situations, such as injection of a and the infected cells were unaffected by the IL-1
wrong dose of vector, have produced contradictory results. Fas siRNA mediated a long-term downregulation of the
The groups of Wallach and Levicoff mimicked injections of endogenous gene FasL, showing that RNAi can be used to
an incorrect dosage to assess safety concerns and possible suppress apoptosis in cultured rat nucleus pulposus cells
complications involving viral vectors carrying BMPs; the and in discs in vivo.
results of the two investigations were contradictory (Wallach ADAMTS5 siRNA restrained intervertebral disc degen-
et al. 2006; Levicoff et al. 2008). Levicoff and colleagues eration in a rabbit model.
reported that up to 80 % of rabbits injected with adeno- Plasmids containing BMP-, LMP-1-, TGF-b-, and Sox-9-
based vector developed significant morbidity, whereas rab- encoding genes can be used to transfect nucleus pulposus
bits injected with AAV, a less immunogenic vector, displayed cells. Adenoviral vectors bearing the BMP-2 or BMP-7
no clinical signs of morbidity. In another study, only rabbits (OP-1) genes were found to be most potent in stimulating
that received a high dose of Adeno-TGF-b exhibited signs proteoglycan synthesis, whereas BMP-4 and BMP-14
of paralysis; rabbits injected with Adeno-BMP-2 did not. (GDF-5) genes showed more affinity and potency in stim-
However, both studies concluded that with appropriate dos- ulating collagen production.
ing, the therapeutic benefits of gene therapy may compen- Degenerated rabbit intervertebral discs served as a rele-
sate for its risks. vant model of disease for studies utilizing adenoviral vec-
Intradiscal injections of therapeutic vectors carrying ana- tor expressing Sox-9. The architecture and histology of
bolic genes/transcription factors provide a possible approach the nucleus pulposus were preserved over a 5-week study
to treating or preventing disc degeneration. Moreover, the period.
authors opine that injections of cells genetically manipulated Doses of both adenovector and AAV encoding either GFP
to express the genes discussed earlier (as shown in models or LacZ injected into rabbit intervertebral discs elicited no
mentioned in the in vitro section) provide an attractive detrimental clinical, biochemical, or histological changes.
approach to enhancing the abilities of cells and therapeutic Nevertheless, safety studies that mimic possible hazard-
genes to repair the degenerate intervertebral disc. ous situations, such as injection of a wrong dose of vector,
Undoubtedly, there are many obstacles to be overcome have produced contradictory results.
and issues to be resolved concerning the timing and the con-
trol of therapeutic gene delivery before this type of remedia-
Acknowledgments The authors wish to thank the NIH and NIAMS for
tion can be clinically relevant. As with all innovative ongoing support through grant RO3AR057143 and the Israel Science
therapeutic approaches, preclinical research, translational Foundation (ISF) grants and the Levi Eshkol Scholarship for PhD Students,
research, and clinical trials will be needed to assess safety the Ministry of Science, Culture and Sport of the State of Israel.
398 Z. Gazit et al.
References Herzog RW, Cao O, Srivastava A (2010) Two decades of clinical gene
therapysuccess is finally mounting. Discov Med 9(45):105111
Huston MW, van Til NP, Visser TP, Arshad S, Brugman MH, Cattoglio
Aslan H, Zilberman Y, Arbeli V, Sheyn D, Matan Y, Liebergal M, Li J,
C, Nowrouzi A, Li Y, Schambach A, Schmidt M, Baum C, von
Helm GA, Gazit D, Gazit Z (2006) Nucleofection-based ex vivo
Kalle C, Mavilio F, Zhang F, Blundell MP, Thrasher AJ, Verstegen
nonviral gene delivery to human stem cells as a platform for tissue
MM, Wagemaker G (2011) Correction of murine SCID-X1 by len-
regeneration. Tissue Eng 12(4):877889
tiviral gene therapy using a codon-optimized IL2RG gene and mini-
Baffi MO, Slattery E, Sohn P, Moses HL, Chytil A, Serra R (2004)
mal pretransplant conditioning. Mol Ther 19:18671877
Conditional deletion of the TGF-beta type II receptor in Col2a
Jimbo K, Park JS, Yokosuka K, Sato K, Nagata K (2005) Positive feed-
expressing cells results in defects in the axial skeleton without alter-
back loop of interleukin-1beta upregulating production of
ations in chondrocyte differentiation or embryonic development of
inflammatory mediators in human intervertebral disc cells in vitro.
long bones. Dev Biol 276(1):124142
J Neurosurg Spine 2(5):589595
Bell DM, Leung KK, Wheatley SC, Ng LJ, Zhou S, Ling KW, Sham
Kakutani K, Nishida K, Uno K, Takada T, Shimomura T, Maeno K,
MH, Koopman P, Tam PP, Cheah KS (1997) SOX9 directly regu-
Kurosaka M, Doita M (2006) Prolonged down regulation of specific
lates the type-II collagen gene. Nat Genet 16(2):174178
gene expression in nucleus pulposus cell mediated by RNA interfer-
Boden SD, Liu Y, Hair GA, Helms JA, Hu D, Racine M, Nanes MS,
ence in vitro. J Orthop Res 24(6):12711278
Titus L (1998) LMP-1, a LIM-domain protein, mediates BMP-6
Kaneyama S, Nishida K, Takada T, Suzuki T, Shimomura T, Maeno K,
effects on bone formation. Endocrinology 139(12):51255134
Kurosaka M, Doita M (2008) Fas ligand expression on human
Bron JL, Helder MN, Meisel HJ, Van Royen BJ, Smit TH (2009) nucleus pulposus cells decreases with disc degeneration processes.
Repair, regenerative and supportive therapies of the annulus fibrosus: J Orthop Sci 13(2):130135
achievements and challenges. Eur Spine J 18(3):301313 Kang JD, Stefanovic-Racic M, McIntyre LA, Georgescu HI, Evans CH
Bueno EM, Glowacki J (2009) Cell-free and cell-based approaches for (1997) Toward a biochemical understanding of human interverte-
bone regeneration. Nat Rev Rheumatol 5(12):685697 bral disc degeneration and herniation. Contributions of nitric oxide,
Cavazzana-Calvo M, Thrasher A, Mavilio F (2004) The future of gene interleukins, prostaglandin E2, and matrix metalloproteinases.
therapy. Nature 427(6977):779781 Spine (Phila Pa 1976) 22(10):10651073
Cui M, Wan Y, Anderson DG, Shen FH, Leo BM, Laurencin CT, Balian Kimelman-Bleich N, Pelled G, Zilberman Y, Kallai I, Mizrahi O,
G, Li X (2008) Mouse growth and differentiation factor-5 protein Tawackoli W, Gazit Z, Gazit D (2011) Targeted gene-and-host pro-
and DNA therapy potentiates intervertebral disc cell aggregation genitor cell therapy for nonunion bone fracture repair. Mol Ther
and chondrogenic gene expression. Spine J 8(2):287295 19(1):5359
Edelstein ML, Abedi MR, Wixon J (2007) Gene therapy clinical trials Kuh SU, Zhu Y, Li J, Tsai KJ, Fei Q, Hutton WC, Yoon ST (2008) The
worldwide to 2007an update. J Gene Med 9(10):833842 AdLMP-1 transfection in two different cells; AF cells, chondrocytes
Elfervig MK, Minchew JT, Francke E, Tsuzaki M, Banes AJ (2001) as potential cell therapy candidates for disc degeneration. Acta
IL-1beta sensitizes intervertebral disc annulus cells to fluid-induced Neurochir (Wien) 150(8):803810
shear stress. J Cell Biochem 82(2):290298 Le Maitre CL, Freemont AJ, Hoyland JA (2005) The role of interleu-
Feng G, Wan Y, Shen FH, Li X (2009) Nucleus pulposus explant cul- kin-1 in the pathogenesis of human intervertebral disc degeneration.
ture model. J Orthop Res 27(6):814819 Arthritis Res Ther 7(4):R732R745
Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC Le Maitre CL, Hoyland JA, Freemont AJ (2007) Interleukin-1 receptor
(1998) Potent and specific genetic interference by double-stranded antagonist delivered directly and by gene therapy inhibits matrix
RNA in Caenorhabditis elegans. Nature 391(6669):806811 degradation in the intact degenerate human intervertebral disc: an in
Flotte TR (2007) Gene therapy: the first two decades and the current situ zymographic and gene therapy study. Arthritis Res Ther
state-of-the-art. J Cell Physiol 213(2):301305 9(4):R83
Friedmann T, Roblin R (1972) Gene therapy for human genetic dis- Lee JY, Hall R, Pelinkovic D, Cassinelli E, Usas A, Gilbertson L, Huard
ease? Science 175(25):949955 J, Kang J (2001) New use of a three-dimensional pellet culture sys-
Glasson SS, Askew R, Sheppard B, Carito B, Blanchet T, Ma HL, tem for human intervertebral disc cells: initial characterization and
Flannery CR, Peluso D, Kanki K, Yang Z, Majumdar MK, Morris EA potential use for tissue engineering. Spine (Phila Pa 1976)
(2005) Deletion of active ADAMTS5 prevents cartilage degradation 26(21):23162322
in a murine model of osteoarthritis. Nature 434(7033): 644648 Lee EX, Lam DH, Wu C, Yang J, Ng WH, Wang S (2011) Glioma gene
Griffith TS, Brunner T, Fletcher SM, Green DR, Ferguson TA (1995) therapy using induced pluripotent stem cell-derived neural stem
Fas ligand-induced apoptosis as a mechanism of immune privilege. cells. Mol Pharm 8:15151524
Science 270(5239):11891192 Leo BM, Li X, Balian G, Anderson DG (2004) In vivo bioluminescent
Hacein-Bey-Abina S, Von Kalle C, Schmidt M, McCormack MP, imaging of virus-mediated gene transfer and transduced cell trans-
Wulffraat N, Leboulch P, Lim A, Osborne CS, Pawliuk R, Morillon plantation in the intervertebral disc. Spine (Phila Pa 1976)
E, Sorensen R, Forster A, Fraser P, Cohen JI, de Saint BG, Alexander 29(8):838844
I, Wintergerst U, Frebourg T, Aurias A, Stoppa-Lyonnet D, Romana Levicoff EA, Kim JS, Sobajima S, Wallach CJ, Larson JW 3rd, Robbins
S, Radford-Weiss I, Gross F, Valensi F, Delabesse E, Macintyre E, PD, Xiao X, Juan L, Vadala G, Gilbertson LG, Kang JD (2008)
Sigaux F, Soulier J, Leiva LE, Wissler M, Prinz C, Rabbitts TH, Le Safety assessment of intradiscal gene therapy II: effect of dosing
Deist F, Fischer A, Cavazzana-Calvo M (2003) LMO2-associated and vector choice. Spine (Phila Pa 1976) 33(14):15091516; discus-
clonal T cell proliferation in two patients after gene therapy for sion 1517
SCID-X1. Science 302(5644):415419 Li X, Leo BM, Beck G, Balian G, Anderson GD (2004) Collagen and
Han D, Ding Y, Liu SL, Wang G, Si IC, Wang X, Cui L, Huang D proteoglycan abnormalities in the GDF-5-deficient mice and molec-
(2009) Double role of Fas ligand in the apoptosis of intervertebral ular changes when treating disk cells with recombinant growth fac-
disc cells in vitro. Acta Biochim Biophys Sin (Shanghai) tor. Spine (Phila Pa 1976) 29(20):22292234
41(11):938947 Liang H, Ma SY, Feng G, Shen FH, Joshua Li X (2010) Therapeutic
Hashimoto G, Aoki T, Nakamura H, Tanzawa K, Okada Y (2001) effects of adenovirus-mediated growth and differentiation factor-5
Inhibition of ADAMTS4 (aggrecanase-1) by tissue inhibitors of in a mice disc degeneration model induced by annulus needle punc-
metalloproteinases (TIMP-1, 2, 3 and 4). FEBS Lett 494(3): 192195 ture. Spine J 10(1):3241
Liu X, Li K, Song J, Liang C, Wang X, Chen X (2006) Efficient and Omlor GW, Bertram H, Kleinschmidt K, Fischer J, Brohm K, Guehring
stable gene expression in rabbit intervertebral disc cells transduced T, Anton M, Richter W (2010) Methods to monitor distribution and
with a recombinant baculovirus vector. Spine (Phila Pa 1976) metabolic activity of mesenchymal stem cells following in vivo
31(7):732735 injection into nucleotomized porcine intervertebral discs. Eur Spine
Liu Y, Kong J, Chen BH, Hu YG (2010) Combined expression of CTGF J 19(4):601612
and tissue inhibitor of metalloprotease-1 promotes synthesis of pro- Osada R, Ohshima H, Ishihara H, Yudoh K, Sakai K, Matsui H, Tsuji H
teoglycan and collagen type II in rhesus monkey lumbar interverte- (1996) Autocrine/paracrine mechanism of insulin-like growth fac-
bral disc cells in vitro. Chin Med J (Engl) 123(15):20822087 tor-1 secretion, and the effect of insulin-like growth factor-1 on pro-
Lohr F, Huang Q, Hu K, Dewhirst MW, Li CY (2001) Systemic vector teoglycan synthesis in bovine intervertebral discs. J Orthop Res
leakage and transgene expression by intratumorally injected recom- 14(5):690699
binant adenovirus vectors. Clin Cancer Res 7(11):36253628 Paul R, Haydon RC, Cheng H, Ishikawa A, Nenadovich N, Jiang W,
Masuda K, An HS (2004) Growth factors and the intervertebral disc. Zhou L, Breyer B, Feng T, Gupta P, He TC, Phillips FM (2003)
Spine J 4(6 Suppl):330S340S Potential use of Sox9 gene therapy for intervertebral degenerative
Matsumoto T, Masuda K, Chen S, An H, Andersson G, Aota Y, Horvath disc disease. Spine (Phila Pa 1976) 28(8):755763
E, Thonar E (2001) Transfer of osteogenic protein-1 gene by gene Pleger ST, Shan C, Ksienzyk J, Bekeredjian R, Boekstegers P, Hinkel
gun system promotes matrix synthesis in bovine intervertebral disc R, Schinkel S, Leuchs B, Ludwig J, Qiu G, Weber C, Raake P, Koch
and articular cartilage cells. Trans Orthop Res Soc 26:30 WJ, Katus HA, Muller OJ, Most P (2011) Cardiac AAV9-S100A1
Menendez MI, Clark DJ, Carlton M, Flanigan DC, Jia G, Sammet S, gene therapy rescues post-ischemic heart failure in a preclinical
Weisbrode SE, Knopp MV, Bertone AL (2011) Direct delayed large animal model. Sci Transl Med 3(92):92ra64
human adenoviral BMP-2 or BMP-6 gene therapy for bone and car- Reddi AH, Reddi A (2009) Bone morphogenetic proteins (BMPs):
tilage regeneration in a pony osteochondral model. Osteoarthritis from morphogens to metabologens. Cytokine Growth Factor Rev
Cartilage 19(8):10661075 20(56):341342
Moon SH, Gilbertson LG, Nishida K, Knaub M, Muzzonigro T, Robbins Risbud MV, Albert TJ, Guttapalli A, Vresilovic EJ, Hillibrand AS,
PD, Evans CH, Kang JD (2000) Human intervertebral disc cells are Vaccaro AR, Shapiro IM (2004) Differentiation of mesenchymal
genetically modifiable by adenovirus-mediated gene transfer: impli- stem cells towards a nucleus pulposus-like phenotype in vitro:
cations for the clinical management of intervertebral disc disorders. implications for cell-based transplantation therapy. Spine (Phila Pa
Spine (Phila Pa 1976) 25(20):25732579 1976) 29(23):26272632
Moon SH, Nishida K, Gilbertson LG, Lee HM, Kim H, Hall RA, Rowzee AM, Cawley NX, Chiorini JA, Di Pasquale G (2011) Glucagon-
Robbins PD, Kang JD (2008) Biologic response of human interver- like Peptide-1 gene therapy. Exp Diabetes Res 2011:601047
tebral disc cells to gene therapy cocktail. Spine (Phila Pa 1976) Schmidt-Wolf GD, Schmidt-Wolf IG (2003) Non-viral and hybrid vec-
33(17):18501855 tors in human gene therapy: an update. Trends Mol Med 9(2): 6772
Morrey ME, Anderson PA, Chambers G, Paul R (2008) Optimizing Seguin CA, Pilliar RM, Roughley PJ, Kandel RA (2005) Tumor
nonviral-mediated transfection of human intervertebral disc chon- necrosis factor-alpha modulates matrix production and catabolism
drocytes. Spine J 8(5):796803 in nucleus pulposus tissue. Spine (Phila Pa 1976) 30(17):
Muona K, Makinen K, Hedman M, Manninen H, Yla-Herttuala S 19401948
(2012) 10-year safety follow-up in patients with local VEGF gene Seki S, Asanuma-Abe Y, Masuda K, Kawaguchi Y, Asanuma K,
transfer to ischemic lower limb. Gene Ther 19:392395 Muehleman C, Iwai A, Kimura T (2009) Effect of small interfer-
Nagase H, Woessner JF Jr (1999) Matrix metalloproteinases. J Biol ence RNA (siRNA) for ADAMTS5 on intervertebral disc degenera-
Chem 274(31):2149121494 tion in the rabbit anular needle-puncture model. Arthritis Res Ther
Nishida K, Kang JD, Suh JK, Robbins PD, Evans CH, Gilbertson LG 11(6):R166
(1998) Adenovirus-mediated gene transfer to nucleus pulposus Shen B, Melrose J, Ghosh P, Taylor F (2003) Induction of matrix metal-
cells. Implications for the treatment of intervertebral disc degen- loproteinase-2 and 3 activity in ovine nucleus pulposus cells grown
eration. Spine (Phila Pa 1976) 23(22):24372442; discussion in three-dimensional agarose gel culture by interleukin-1beta: a
2443 potential pathway of disc degeneration. Eur Spine J 12(1):6675
Nishida K, Kang JD, Gilbertson LG, Moon SH, Suh JK, Vogt MT, Sheyn D, Kimelman-Bleich N, Pelled G, Zilberman Y, Gazit D, Gazit Z
Robbins PD, Evans CH (1999) Modulation of the biologic activity (2008) Ultrasound-based nonviral gene delivery induces bone for-
of the rabbit intervertebral disc by gene therapy: an in vivo study of mation in vivo. Gene Ther 15(4):257266
adenovirus-mediated transfer of the human transforming growth Somia N, Verma IM (2000) Gene therapy: trials and tribulations. Nat
factor beta 1 encoding gene. Spine (Phila Pa 1976) 24(23): Rev Genet 1(2):9199
24192425 Sowa G, Westrick E, Pacek C, Coelho P, Patel D, Vadala G, Georgescu
Nishida K, Gilbertson LG, Robbins PD, Evans CH Kang JD (2000) H, Vo N, Studer R, Kang J (2011) In vitro and in vivo testing of a
Potential applications of gene therapy to the treatment of interver- novel regulatory system for gene therapy for intervertebral disc
tebral disc disorders. Clin Orthop Relat Res (379 Suppl): degeneration. Spine (Phila Pa 1976) 36(10):E623E628
S234S241 Suzuki T, Nishida K, Kakutani K, Maeno K, Yurube T, Takada T,
Nishida K, Doita M, Takada T, Kakutani K, Miyamoto H, Shimomura Kurosaka M, Doita M (2009) Sustained long-term RNA interfer-
T, Maeno K, Kurosaka M (2006) Sustained transgene expression ence in nucleus pulposus cells in vivo mediated by unmodified small
in intervertebral disc cells in vivo mediated by microbubble- interfering RNA. Eur Spine J 18(2):263270
enhanced ultrasound gene therapy. Spine (Phila Pa 1976) 31(13): Takada T, Nishida K, Doita M, Kurosaka M (2002) Fas ligand exists on
14151419 intervertebral disc cells: a potential molecular mechanism for
Nishida K, Suzuki T, Kakutani K, Yurube T, Maeno K, Kurosaka M, immune privilege of the disc. Spine (Phila Pa 1976) 27(14):
Doita M (2008) Gene therapy approach for disc degeneration and 15261530
associated spinal disorders. Eur Spine J 17(Suppl 4):459466 Thompson JP, Oegema TR Jr, Bradford DS (1991) Stimulation of
Noel D, Gazit D, Bouquet C, Apparailly F, Bony C, Plence P, Millet V, mature canine intervertebral disc by growth factors. Spine (Phila Pa
Turgeman G, Perricaudet M, Sany J, Jorgensen C (2004) Short-term 1976) 16(3):253260
BMP-2 expression is sufficient for in vivo osteochondral differenti- Vadala G, Sowa GA, Smith L, Hubert MG, Levicoff EA, Denaro V,
ation of mesenchymal stem cells. Stem Cells 22(1):7485 Gilbertson LG, Kang JD (2007) Regulation of transgene expression
400 Z. Gazit et al.
using an inducible system for improved safety of intervertebral disc stem cells into a chondrocyte-like phenotype in vitro. J Orthop Res
gene therapy. Spine (Phila Pa 1976) 32(13):13811387 29(8):12911297
Wallach CJ, Gilbertson LG, Kang JD (2003a) Gene therapy applica- Yasuda T, Hayakawa H, Nihira T, Ren YR, Nakata Y, Nagai M, Hattori
tions for intervertebral disc degeneration. Spine (Phila Pa 1976) N, Miyake K, Takada M, Shimada T, Mizuno Y, Mochizuki H
28(15 Suppl):S93S98 (2011) Parkin-mediated protection of dopaminergic neurons in a
Wallach CJ, Sobajima S, Watanabe Y, Kim JS, Georgescu HI, Robbins chronic MPTP-minipump mouse model of Parkinson disease.
P, Gilbertson LG, Kang JD (2003b) Gene transfer of the catabolic J Neuropathol Exp Neurol 70:686697
inhibitor TIMP-1 increases measured proteoglycans in cells from Yoon ST, Park JS, Kim KS, Li J, Attallah-Wasif ES, Hutton WC, Boden
degenerated human intervertebral discs. Spine (Phila Pa 1976) SD (2004) ISSLS prize winner: LMP-1 upregulates intervertebral
28(20):23312337 disc cell production of proteoglycans and BMPs in vitro and in vivo.
Wallach CJ, Kim JS, Sobajima S, Lattermann C, Oxner WM, McFadden Spine (Phila Pa 1976) 29(23):26032611
K, Robbins PD, Gilbertson LG, Kang JD (2006) Safety assessment Zhang Y, An HS, Thonar EJ, Chubinskaya S, He TC, Phillips FM
of intradiscal gene transfer: a pilot study. Spine J 6(2):107112 (2006) Comparative effects of bone morphogenetic proteins and
Wang C, Ruan DK, Zhang C, Wang DL, Xin H, Zhang Y (2011) Effects sox9 overexpression on extracellular matrix metabolism of
of adeno-associated virus-2-mediated human BMP-7 gene transfec- bovine nucleus pulposus cells. Spine (Phila Pa 1976) 31(19):
tion on the phenotype of nucleus pulposus cells. J Orthop Res 21732179
29(6):838845 Zhang Y, Phillips FM, Thonar EJ, Oegema T, An HS, Roman-Blas JA,
Wehling P (2001) Transfer of genes to intervertebral disc cells: pro- He TC, Anderson DG (2008) Cell therapy using articular chondro-
posal for a treatment strategy of spinal disorders by local gene ther- cytes overexpressing BMP-7 or BMP-10 in a rabbit disc organ cul-
apy. Joint Bone Spine 68(6):554556 ture model. Spine (Phila Pa 1976) 33(8):831838
Wehling P, Schulitz KP, Robbins PD, Evans CH, Reinecke JA (1997) Zhang Y, Markova D, Im HJ, Hu W, Thonar EJ, He TC, An HS, Phillips
Transfer of genes to chondrocytic cells of the lumbar spine. Proposal FM, Anderson DG (2009) Primary bovine intervertebral disc cells
for a treatment strategy of spinal disorders by local gene therapy. transduced with adenovirus overexpressing 12 BMPs and Sox9
Spine (Phila Pa 1976) 22(10):10921097 maintain appropriate phenotype. Am J Phys Med Rehabil 88(6):
Xia C, Ding C, Yang X, Sun K, Tian S (2010) Effects of antisense trans- 455463
forming growth factor-beta1 gene transfer on the biological activi- Zhang Y, Chee A, Thonar EJ, An HS (2011) Intervertebral disk repair
ties of tendon sheath fibroblasts. Orthopedics 33(8) by protein, gene, or cell injection: a framework for rehabilitation-
Yang Z, Huang CY, Candiotti KA, Zeng X, Yuan T, Li J, Yu H, Abdi S focused biologics in the spine. PM R 3(6 Suppl):S88S94
(2011) Sox-9 facilitates differentiation of adipose tissue-derived
Enhancing Disc Repair by Growth
Factors and Other Modalities 25
Won C. Bae and Koichi Masuda
Contents 25.1 Disc Degeneration: Imbalance

25.1 Disc Degeneration: Imbalance of Anabolic of Anabolic and Catabolic Activities
and Catabolic Activities ................................................ 401
25.2 Biologic Treatments of Degenerative
Recent studies have found an association between disc
Disc Diseases .................................................................. 402 degeneration and genes related to the extracellular matrices
of the disc, such as those encoding collagen I (Tilkeridis
25.3 Model Systems and Outcome Measures
to Evaluate Treatment Efficacy.................................... 404 et al. 2005), collagen IX (Annunen et al. 1999; Solovieva
et al. 2006), aggrecan (Kawaguchi et al. 1999), matrix metal-
25.4 Effect of Growth Factors on Intervertebral
Disc Cells and Tissues ................................................... 405 loprotease-2 (MMP-2) (Dong et al. 2007), MMP-3 (Takahashi
et al. 2001), interleukin-1b (IL-1b) (Solovieva et al. 2004),
25.5 Effect of Other Modalities on Intervertebral
Disc Cells and Tissues ................................................... 406
IL-6 (Noponen-Hietala et al. 2005), vitamin D receptor
(Videman et al. 1998), asporin (Song et al. 2008), and carti-
25.6 Limitations of Injection Therapy ................................ 409 lage intermediate layer protein (CILP) (Seki et al. 2005;
25.7 Conclusion ..................................................................... 410 Virtanen et al. 2007). These observations suggest that pro-
25.8 Summary of Critical Concepts Discussed teins in the extracellular matrix (ECM) or cytokines that may
in the Chapter ................................................................ 411 degrade the matrix may play a role in disc degeneration.
References .................................................................................... 411 Homeostasis of the extracellular matrix of the nucleus
pulposus and annulus fibrosus is conserved by the activities
of the resident disc cells that maintain a balance between
anabolism and catabolism. In the young human nucleus pul-
posus, there exists a large population of notochordal cells
(Trout et al. 1982), which by maturity is replaced by or dif-
ferentiates into a population of chondrocyte-like cells. Both
cell types synthesize proteoglycans (Kim et al. 2009) which
constitute the bulk of the nucleus pulposus tissue (Aguiar
et al. 1999). In common with articular cartilage, proteogly-
cans interact extracellularly with hyaluronan to form aggre-
gates that become entangled in a fibrillar network composed
primarily of collagen II (Scott and Haigh 1986). Together,
the large hydrophilic proteoglycan aggregates and the colla-
gen network provide the intrinsic mechanical properties of
W.C. Bae, PhD
Department of Radiology, the nucleus pulposus. The annulus fibrosus contains a rela-
University of California, San Diego, tively homogeneous population of cells that synthesize a
408 Dickinson Street, San Diego, CA 92103-8226, USA matrix richer in collagen and poorer in proteoglycans than
e-mail: wbae@ucsd.edu
the nucleus pulposus; as a result. The annulus fibrosus is a
K. Masuda, MD (*) dense lamellar tissue that contains fibrillar layers rich in col-
lagen I and in the deeper regions collagen II (Schollmeier
University of California, San Diego,
9500 Gilman Dr. MC0863, La Jolla, CA 92093-0863, USA et al. 2000). For a detailed discussion of collagen and proteo-
e-mail: koichimasuda@ucsd.edu glycans of the disc, see Chaps. 4 and 5.

402 W.C. Bae and K. Masuda
Molecules that regulate the anabolic state of the with thrombospondin motifs-4 (ADAMTS-4) (Patel et al.
intervertebral disc include polypeptide growth factors, 2007). It is also interesting to note that in articular cartilage,
such as insulin-like growth factor (IGF-1), transforming low concentrations of IL-1, in addition to inducing tissue
growth factor-b (TGF-b), and the bone morphogenetic degradation, can also inhibit the synthesis of proteoglycans,
proteins (BMPs) (Thompson et al. 1991; Osada et al. mainly aggrecan (Arner and Pratta 1989; Benton and Tyler
1996). Catabolic regulators include cytokines, such as 1988; Dingle et al. 1991).
IL-1 (Ahn et al. 2002a; Takahashi et al. 1996a) and tumor
necrosis factor-a (TNF-a) (Ahn et al. 2002a; Miyamoto
et al. 2000; Takahashi et al. 1996a), which influence the 25.2 Biologic Treatments
synthesis of matrix-degrading enzymes. Alterations in of Degenerative Disc Diseases
both anabolic and catabolic processes are thought to play
key roles in the onset and progression of disc degenera- There are several strategies being pursued for biologic repair
tion (Masuda and An 2006). or regeneration of the disc. A widely used strategy, and also
While the upregulation of anabolic genes and regulators the focus of this chapter, is to directly inject therapeutic
has been studied (Gokorsch et al. 2005; Desmoulin et al. agents into the intervertebral disc. The agents used include
2011), a far greater number of investigations have focused on those aimed at stimulating the anabolism of the host disc,
biochemical changes of the intervertebral disc under patho- providing an environment conducive for biologic repair, and
logic conditions. These conditions include intervertebral disc those that inhibit or compete with catabolic regulators and
degeneration (Patel et al. 2007), injury (e.g., puncture or stab enzymes that may exist in degenerated disc tissues. Anabolic
wounds) (Anderson et al. 2002; Sobajima et al. 2005), and agents include growth factors [e.g., osteogenic protein-1
abnormal mechanical loading (Yurube et al. 2010; Iatridis (OP-1; otherwise known as BMP-7), other BMPs, and
et al. 2011). In these conditions, the cells of the disc express TGF-b], materials containing multiple growth factors (e.g.,
and synthesize cytokines that can tilt the balance towards platelet-rich plasma), and cells transfected with therapeutic
catabolism. Increases in IL-1 mRNA and protein levels and genes (e.g., TGF-b). Gel-based biomaterials can provide
its major regulator, TNF-a, have been observed in degener- several benefits conducive to biologic repair: Gels that have
ated or herniated discal tissues in culture (Ahn et al. 2002b; similar mechanical properties to the nucleus pulposus tissue
Burke et al. 2003; Kang et al. 1996; Weiler et al. 2005; can provide immediate load support and may promote hydra-
Igarashi et al. 2000; Olmarker and Larsson 1998; Le Maitre tion. Moreover, they can be mixed or cross-linked with
et al. 2005, 2007a). The catabolic processes induced by growth factors to provide long-lasting anabolic effects com-
cytokines, such as IL-1, may be mediated by a number of pared to the growth factors alone. Inhibitory agents include
enzymes, including collagenase (Sakuma et al. 2002), those that inhibit expression of catabolic genes [e.g., small
cycloxygenase-2 (COX-2) (Kang et al. 1997), prostaglandin interference RNA (siRNA) for ADAMTS-5], antagonists for
E2 (PGE2) (Takahashi et al. 1996a; Rannou et al. 2000), IL-1 receptors, and TNF-a (i.e., soluble TNF receptor
MMP-1 (Jimbo et al. 2005), MMP-3, (Jimbo et al. 2005), type II). Tables 25.1 and 25.2 summarize some of the thera-
MMP-13 (Miyamoto et al. 2005; Le Maitre et al. 2004), and peutic agents that have been studied in vitro and in vivo,
the aggrecanases, such as a disintegrin and metalloproteinase respectively.
Table 25.1 The in vitro effects of injectable therapeutic agents

Agent Target Effect Reference
TGF-b Mature canine IVD PG synthesis increased up to 5X; higher in NP than AF Thompson et al. (1991)
IGF-1 Mature canine IVD PG synthesis marginally increased in NP Thompson et al. (1991)
TGF-b Human annulus cells, 3D culture Cell proliferation and PG synthesis increased; reduced Gruber et al. (1997)
apoptosis with serum depletion
IGF-1 Young and old bovine NP cells Cell proliferation and matrix synthesis stimulated; more Osada et al. (1996)
IGF-1 receptors
OP-1 Young rabbit NP and AF cells; alginate Increased PG and collagen production and content Masuda et al. (2003)
beads
BMP-2 Rat IVD cells monolayer Increased cell number, GAG, expression for collagen, Yoon et al. (2003)
aggrecan at higher doses
OP-1 Human NP and AF cells, alginate beads Maintained cell density, increased PG synthesis and Imai et al. (2007a)
accumulation
OP-1 Rabbit IVD cells; alginate beads: IL-1a IL-1 decreased PG and collagen; reversed and exceeded Takegami et al. (2002)
preexposure with OP-1
25 Enhancing Disc Repair by Growth Factors and Other Modalities 403

Agent Target Effect Reference
OP-1 Rabbit IVD cells; alginate bead: OP-1 upregulated PG synthesis. Greater effect on Takegami et al. (2005)
C-ABC preexposure C-ABC preexposure than control
BMP-2 Human IVD cells Increased PG synthesis, expression of aggrecan, Kim et al. (2003)
collagen types I and II; no bone formation
rhBMP-2 and Human IVD cells in monolayer PG, collagen synthesis increased in NP cells; minimal Gilbertson et al. (2008)
BMP-12 effect on AF cells
GDF-5 Bovine IVD cells; alginate bead Increased DNA and PG content; at higher dose, PG and Chujo et al. (2006)
collagen synthesis increased
PRP Porcine IVD cells; alginate bead Mild increase in cell proliferation; marked increase in Akeda et al. (2006)
PG and collagen synthesis and PG accumulation
TGF-b1 and PRP Human NP cells NP cell proliferation and aggregation; increase in Chen et al. (2006)
mRNA of SOX-9, collagen type II, aggrecan
Ad-TIMP-1, IVD cells from human degenerated IVD 2,000 pg/ml production of TIMP w/100 MOI at day 4. Wallach et al. (2003)
Ad-BMP-2 PG synthesis increased with both Ad-TIMP-1 and
Ad-BMP-2
Dexamethasone Human disc herniation tissue explants Decreased MMP-1 and MMP-3 levels Genevay et al. (2009)
IL-1ra Human disc herniation tissue explants Decreased MMP-3 levels Genevay et al. (2009)
IL-1ra Human normal and degenerated disc IL-1ra reduced cytokine levels (MMP-3, MMP-7, Le Maitre et al.
tissues in situ with IL-1 treatment MMP-13) and matrix degradation in all tissue types (2007b)
IL-1ra/ELP Human IVD cells (grades 23); alginate Reduced ADAMTS-4, MMP-3 transcription Shamji et al. (2007)
beads: IL-1ra pre-Tx then IL-1b insult
p38 MAPK Rabbit NP cells pretreated with IL-1 Decreased message for collagen, aggrecan, IGF-1. Studer et al. (2008)
inhibitor (SB Increased message for iNOS, COX-2, MMP-3, IL-6
202190)
TNF inhibitor Human IVD herniation tissue explants Decreased MMP-3 levels Genevay et al. (2009)
mAb
PDGF, bFGF, Human NP and AF cells Increased DNA synthesis via ERK and Akt pathways Pratsinis et al. (2012)
IGF-I
TGFb3 + Dex, Degenerated human NP cells Stimulated NP cell proliferation and decreased Abbott et al. (2012)
notochordal ADAMTS-5, MMP-1 expression
conditioned
media
Lactoferricin Bovine NP cells Increased PG accumulation and expression of SOX-9, Kim et al. (2012)
aggrecan, TIMP-family genes. Decreased expression of
MMPs and ADAMTSs in dose-dependent manner
IGF-1, BMP-7, Bovine NP cells Synergistically increased anabolic gene expression, PG Kim et al. (2010)
IGF-1 + BMP-7 synthesis, and PG accumulation
TGF-b transforming growth factor-b, IVD intervertebral disc, PG proteoglycan, NP nucleus pulposus, AF annulus fibrosus, IGF-1 insulin-like
growth factor-1, 3D three dimensional, OP-1 osteogenic protein-1, BMP-2 bone morphogenetic protein-2, GAG glycosaminoglycan, IL-1 interleu-
kin-1, C-ABC chondroitinase ABC, rh recombinant human, GDF-5 growth and differentiation factor-5, SOX-9 sex determining region Y-box 9
gene, PRP platelet-rich plasma, Ad-TIMP-1 adenoviral vector delivering cDNA of tissue inhibitor of matrix metalloproteinases-1, Ad-BMP-2
adenoviral vector delivering cDNA of BMP-2, MOI multiplicity of infection, MMP matrix metalloproteinase, IL-1ra IL-1 receptor antagonist, ELP
elastin-like polypeptide, Tx treatment, ADAMTS a disintegrin and metalloproteinase with thrombospondin motifs, MAPK mitogen-activated
protein kinase, iNOS inducible nitric oxide synthase, COX-2 cyclooxygenase-2, TNF tumor necrosis factor, mAb monoclonal antibody, PDGF
platelet-derived growth factor, bFGF basic fibroblast growth factor, ERK extracellular signal-regulated kinases, Akt protein kinases, Dex
dexamethasone
Table 25.2 The in vivo effects of intradiscal injection treatments

Agent Species Site Model Effect Reference
IGF-1 Rat Tail Static compression Clustering of inner annulus cells after single Walsh et al. (2004)
injection
GDF-5 Rat Tail Static compression Clustering of cells, increase in disc height Walsh et al. (2004)
(single injection)
TFG-b Rat Tail Static compression Proliferation of cells (multiple injections) Walsh et al. (2004)
bFGF Rat Tail Static compression No response Walsh et al. (2004)
OP-1 Rabbit Lumbar None (normal) Increased disc height and PG content in NP An et al. (2005)
(continued)

Agent Species Site Model Effect Reference
OP-1 Rabbit Lumbar C-ABC: co-injection Increased disc height and PG content in NP Imai et al. (2003)
OP-1 Rabbit Lumbar Needle puncture: Increased disc height and PG content in NP and Masuda et al. (2006)
Tx 4 weeks later AF, improvement of MRI and histology grades
OP-1 Rabbit Lumbar Needle puncture Increased disc height and viscoelastic properties Miyamoto et al. (2006b)
OP-1 Rabbit Lumbar C-ABC: Tx 4 weeks Increased disc height, PG content in NP and AF Imai et al. (2007b)
later
GDF-5 Rabbit Lumbar Needle puncture: Increased disc height, improvement of MRI and Chujo et al. (2006)
Tx 4 weeks later histology grades
GDF-5 Rabbit Lumbar Thrombin degraded: Increased disc height, improved T1rho and T2 Bae et al. (2009)
Tx 4 weeks later values. Decreased ADAMTS-4 and ADAMTS-5
and COX-2 expression
BMP-2 Rabbit Lumbar Annular stab More degeneration, vascularity, and fibroblast Huang et al. (2007)
(5 7 mm)
PRP Rabbit Lumbar Nucleotomy, PRP + GHM group had less degeneration and Nagae et al. (2007)
immediate Tx increased PG; PRP + PBS group showed no
differences
PRP Rabbit Lumbar Nucleotomy, PRP + GHM had greater disc height, water content, Sawamura et al. (2009)
immediate Tx mRNA for PG core protein, and collagen type II;
fewer apoptotic cells in NP
BMP-17 Sheep Lumbar Annular stab BMP-17 maintained disc height, MRI and Wei et al. (2009)
(3 6 mm), histology scores, NP cell density; increased PG and
immediate Tx collagen synthesis
ADAMTS-5 Rabbit Lumbar Needle puncture: Improved MRI and histology scores Seki et al. (2009)
siRNA Tx 4 weeks later
8K-NBD Mouse Lumbar Progeroid Ercc1 p65 Restored total NP GAG and PG synthesis Nasto et al. (2012)
peptide (NF-kB KO
inhibitor)
allograft MSC Minipig Lumbar 1 cm incision and JC Tx: high GAG content at 12 months, rich in Acosta et al. (2011)
or JC in fibrin nucleotomy; evaluate collagen type II
at 3, 6, 12 months
Link N protein Rabbit Lumbar Needle puncture: Tx Increased aggrecan gene expression and decreased Mwale et al. (2011)
4 weeks later proteinase gene expression
IGF-1 insulin-like growth factor-1, GDF-5 growth differentiation factor-5, TGF-b transforming growth factor-b, bFGF basic fibroblast growth
factor, OP-1 osteogenic protein-1, PG proteoglycan, NP nucleus pulposus, C-ABC chondroitinase ABC, Tx treatment, AF annulus fibrosus, MRI
magnetic resonance imaging, ADAMTS a disintegrin and metalloproteinase with thrombospondin motifs, COX cyclooxygenase, BMP bone mor-
pho genetic protein, PRP platelet-rich plasma, GHM gelatin hydrogel microspheres, PBS phosphate-buffered saline, siRNA small interference
RNA (siRNA), 8K-NBD Nemo-binding domain, NF-kB nuclear factor kappa-light-chain-enhancer of activated B cells, Ercc1 excision repair
cross-complementing rodent repair deficiency, complementation group 1, KO knock out, MSC mesenchymal stem cell, JC juvenile chondrocytes,
GAG glycosaminoglycan, Link N amino terminal peptide of link protein (DHLSDNYTLDHDRAIH)
25.3 Model Systems and Outcome techniques. Anabolic genes of interest may include those
Measures to Evaluate molecules comprising the extracellular matrix, such as col-
Treatment Efcacy lagen and aggrecan, and anabolic regulators, such as TGF-b,
IGF-1, and growth and differentiation factor-5 (GDF-5). At
For evaluation of the efficacy of various biologic treatments, the protein level, the synthesis of collagen and proteoglycans
a number of in vitro and in vivo model systems are routinely can be assessed, using radiolabeled 3H and 35S, respectively,
used. Isolated nucleus pulposus and annulus fibrosus cells by determining their incorporation from the media into the
from intervertebral discs are seeded in monolayer and treated cells. Other factors can be assessed by enzyme-linked immu-
with specific cytokines (e.g., IL-1, TNF-a) to model degen- nosorbent assay (ELISA) or multiplex cytokine assay.
erative conditions. Cells from discs with varying degrees of Pellets of cells surrounded by three-dimensional extra-
degeneration are also used. In these systems, the efficacy of cellular matrices, recovered after alginate culture (Masuda
therapeutic agents can be determined at several levels. In et al. 2000), for example, have also been used. The presence
cells, a decrease in the gene expression of catabolic enzymes, of the extracellular matrix better mimics the microenviron-
such as aggrecanases and proteinases, as well as vascular ment of disc cells. In addition to treatments using a mono-
factors and pain-related factors may be assessed using PCR layer system, pellet cultures may be treated with agents to
alter or deplete specific matrix components (e.g., using (Fig. 25.1), upregulates proteoglycan metabolism in
chondroitinase ABC to deplete proteoglycans); this is not intervertebral disc cells. OP-1 strongly stimulated the pro-
feasible using monolayer culture. For outcome measure- duction and formation of extracellular matrix by rabbit disc
ments, the biochemical content of the matrix can be deter- cells (Masuda et al. 2003); similar effects have been noted
mined. For example, the dimethylmethylene blue (DMMB) using human intervertebral disc cells (Imai et al. 2007a).
assay can be used to assess the proteoglycan content, while OP-1 also replenished proteoglycans and collagens after
western blot and PCR can be used to assess protein and gene depletion of the matrix following exposure of intervertebral
expression levels. disc pellets to IL-1 (Takegami et al. 2002) or chondroitinase
ABC (C-ABC) (Takegami et al. 2005). The efficacy of OP-1
injection has also been evaluated in a number of in vivo ani-
25.4 Effect of Growth Factors on mal models. In adolescent rabbits, an injection of recombi-
Intervertebral Disc Cells and Tissues nant human OP-1 (rhOP-1), but not the lactose vehicle,
reversed the reduction in disc height and improved the
A variety of anabolic growth factors and cytokines alter magnetic resonance imaging (MRI) grade caused by a needle
intervertebral disc homeostasis and stimulate extracellular puncture of the annulus fibrosus (Masuda et al. 2006).
matrix synthesis (Masuda et al. 2004). OP-1 (Masuda et al. In another rabbit study (Miyamoto et al. 2006b), OP-1
2003), a member of the BMP family and TGF-b superfamily restored dynamic viscoelastic biomechanical properties, in
Ligand
antagonists Ligands
P P
II I II I
P FKBP12
Type II Type I Non-canonical
SMAD pathway receptor receptor pathways
Cytoplasm
JNK/ MAPK AKT RhoA

p38
DNA-binding
Nucleus
P proteins
P
Cofactors
14
Coactivators
or P Inhibit
P SMAD
corepressors
pathway
Cofactors Regulates
cytoskeleton,
cell adhesion
Fig. 25.1 TGF-b signaling pathways. Ligand binds with types I and II coactivators or corepressors (9) to activate or repress transcription of
receptors (1), which releases FKBP12 (2) from the GS domain of the genes (10). The activated ligand-receptor complex can begin other non-
type I receptor and forms a ligand-receptor complex (3). The type I canonical pathways (11), which can affect cofactors (12), regulate
receptor is phosphorylated by type II receptor (4). The activated type I cytoskeletal organization and cell adhesion (13), or inhibit the SMAD
receptors phosphorylate receptor-regulated SMADs (rrSMAD) (5), pathway (14). In addition, ligand antagonists can sequester ligands
which associate with SMAD4 (6) and move into the nucleus (7). The extracellularly (15)
SMAD complex associates with DNA-binding cofactors (8) along with
needle-punctured intervertebral discs (Miyamoto et al.

2006b). It was also effective in restoring discs that have been 2004). Among these growth factors, TGF-b1 exists in
chemically degraded with C-ABC, which has been consid- the highest concentration (Weibrich et al. 2002) and
ered as an alternative to chymopapain for chemonucleolysis may be the core ingredient and the indicator for apply-
(Eurell et al. 1990; Fry et al. 1991; Henderson et al. 1991; ing PRP. These growth factors appear to play an impor-
Kato et al. 1992; Ando et al. 1995; Sugimura et al. 1996; tant role in wound healing and are assumed to facilitate
Takahashi et al. 1996b; Yamada et al. 2001), in animal mod- hard and soft tissue regeneration.
els of disc degeneration such as the rat tail (Norcross et al. The efficacy of PRP injection for disc regeneration
2003; Hoogendoorn et al. 2007, 2008; Boxberger et al. 2008) is still being investigated; however, promising results
and goat (Hoogendoorn et al. 2007, 2008). When OP-1 or such as upregulation of proteoglycan anabolism and
vehicle was injected into rabbit discs degraded with C-ABC restoration of disc height have been seen in a number
for 4 weeks, the disc height was initially decreased (~34 %), of in vitro studies on disc cell (Akeda et al. 2006; Chen
then recovered, and gradually approached the level of the et al. 2006), on explants (Chen et al. 2009), as well as
control (Imai et al. 2007b). in animal models (Nagae et al. 2007; Chen et al. 2009;
A number of autologous agents have been shown to be Obata et al. 2012). PRP is often used in conjunction
clinically useful. For example, platelet-rich plasma (PRP) with biomaterials such as gelatin (Nagae et al. 2007)
contains high levels of multiple growth factors and as such it and mesenchymal stem cells (Chen et al. 2006, 2009).
has been used in disc repair in animal studies and in a clinical PRP has been used without activation (Nagae et al.
study ongoing in Japan. Moreover, PRP can be easily gener- 2007); however, most commercial systems (Castillo
ated in the operating room by centrifugal separation of autol- et al. 2011) use PRP releasate.
ogous blood using a point-of-care device. In vitro,
PRP-stimulated porcine disc cell proliferation and matrix
synthesis (Akeda et al. 2006) and using isolated human disc In addition to PRP and OP-1, many other anabolic growth
cells provoked the formation of a nucleus pulposus-like tis- factors and cytokines have been investigated. Early studies
sue (Chen et al. 2006). The efficacy of injecting allograft indicated that TGF-b promoted disc cell proliferation (Gruber
PRP with or without a gelatin hydrogel (which provides slow et al. 1997) and stimulated proteoglycan synthesis (Thompson
release and mechanical support) was explored using a rabbit et al. 1991; Gruber et al. 1997). IGF-1 (Osada et al. 1996;
model of nucleotomy (Nagae et al. 2007). The PRP hydrogel Pratsinis and Kletsas 2007) and platelet-derived growth fac-
was found to markedly suppress further degeneration, com- tor (PDGF) (Pratsinis and Kletsas 2007) also showed a simi-
pared to PRP alone and a saline control group. A follow-up lar cell-proliferative effect. In addition, PDGF exerted a
study suggested that the hydrogel microspheres without PRP protective, antiapoptotic effect on annulus fibrosus cells
did not have therapeutic value. In contrast, animals treated induced by serum depletion (Gruber et al. 2000).
with PRP with microspheres benefited from increased disc
height, elevated water content, increased expression of pro-
teoglycan core protein and collagen II, and fewer apoptotic 25.5 Effect of Other Modalities on
cells in the nucleus pulposus (Sawamura et al. 2009). In less Intervertebral Disc Cells and Tissues
severe models without nucleotomy, PRP (after activation
with autologous serum and calcium) injection alone has been Agents other than anabolic growth factors are being
effective for disc repair (Obata et al. 2012). considered for repair of the degenerate intervertebral disc.
In a recent study, ADAMTS-5 expression was silenced using
siRNA (Seki et al. 2009). In vitro studies, using rabbit
Box 25.1 Platelet-Rich Plasma nucleus pulposus cells, were performed to determine the
Platelet-rich plasma (PRP) injection is increasingly extent of reduction of ADAMTS-5 expression. Then, using
used as an off-label procedure before resorting to an annular needle puncture rabbit model, MRI and histologi-
orthopedic surgery. PRP contains a mixture of growth cal analysis were used to assess the efficacy of ADAMTS-5
factors (Weibrich et al. 2002; Okuda et al. 2003; siRNA to prevent tissue degradation. After 8 weeks, the
Dugrillon et al. 2002; Mazzocca et al. 2012) such as control group exhibited a complete loss of nucleus pulposus
transforming growth factor (TGF)-b1 and TGF-b2, tissue, while the ADAMTS-silenced animals maintained
vascular endothelial growth factor (VEGF), platelet- disc structure (Seki et al. 2009).
derived growth factor (PDGF), and insulin-like growth IL-1 receptor antagonist (IL-1ra) is another inhibitory
factor (IGF), which are naturally released from plate- agent that has been studied. When applied in vitro to
lets after being activated by calcium or thrombin degenerated (Le Maitre et al. 2007b) and herniated
(Tozum and Demiralp 2003; Arpornmaeklong et al. (Genevay et al. 2009) human disc tissues, IL-1ra reduced
the expression of MMP-3 (Le Maitre et al. 2007b;
AF NP
3 3
a P<0.01
b
P<0.01
(vs. IL -1)
(vs. IL -1)
2 2
(pg/ml media)
TNF-
1 1
0 0
50 50
c P<0.01 d
(vs. IL -1) P<0.01
40 40
(vs. IL -1)
(ng/ml media)
30 30
MMP3
20 20
10 10
0 0
40 40
e P<0.01
f P<0.01
(vs. IL -1) (vs. IL -1)
30 30
Caspase 3
(ng/ml)
20 20
10
0 10
0
0 0
IL-1 0 10 0 0 0 1 10 10 0 10 0 0 0 1 10 10
DN-TNF
(ng/ml) 0 0 0.1 1 10 0.1 1 10 0 0 0.1 1 10 0.1 1 10
Fig. 25.2 Dominant-negative tumor necrosis factor (DN-TNF) effec- Levels of TNF-a, matrix metalloproteinase-3 (MMP-3), and caspase 3
tively antagonizes interleukin-1b (IL-1b)-induced catabolic changes in released into the media were measured. DN-TNF treatment suppressed
human intervertebral disc cells. Human nucleus pulposus (NP) and release of (a, b) TNF-a, (c, d) MMP-3, and (e, f) caspase 3 by both AF
annulus fibrosis (AF) cells were isolated from cadaveric intervertebral and NP cells, suggesting DN-TNF may be effective in suppressing cata-
disc (IVD) tissues (average of ~60 years old) and cultured in alginate bolic cytokines, matrix-degrading enzymes, and cell apoptosis
beads for 7 days. Cells were serum starved for 1 day and treated for (Modified from Pichika et al. (2011))
2 days in media containing a combination of IL-1b and DN-TNF.
Genevay et al. 2009). In addition, following pretreatment of nitric oxide and IL-6 production (Sinclair et al. 2011).
degenerated human nucleus pulposus cells with IL-1ra and Although dominant-negative TNF (DN-TNF) does not acti-
subsequent treatment with IL-1, there was reduced expres- vate TNF receptors, it effectively competes with soluble
sion of ADAMTS-4 and MMP-3 (Shamji et al. 2007). TNF. When applied to human intervertebral disc cells chal-
Yet another target for inhibition is TNF-a. In human disc lenged with IL-1b (Pichika et al. 2011), DN-TNF effectively
cells, the use of soluble TNF receptors along with IL-1ra reduced the concentration of TNF-a (Fig. 25.2a, b), levels of
significantly upregulated proteoglycan synthesis (Kakutani MMP-3 in the media (Fig. 25.2c, d), the expression of intra-
et al. 2008), while co-treatment with TNF-a suppressed cellular caspase 3 (Fig. 25.2e, f), and the production of PGE2.
The use of a monoclonal antibody against TNF-a suppressed biomaterials in vivo and their long-term biologic effects, the
the expression and concentration of MMP-3 in explants of use of a suitable carrier material may synergize growth factor
herniated discs (Genevay et al. 2009). Other TNF-inhibiting activity.
agents are beginning to be used clinically for analgesic pur- While of limited use for development of injectable thera-
poses in patients with sciatica (Karppinen et al. 2003; peutic agents, a mechanism for stimulating disc cells anabo-
Genevay et al. 2004; Okoro et al. 2010) and discogenic pain lism would be of great value for countering the in vivo
(Tobinick and Britschgi-Davoodifar 2003). Anti-cytokine degradation of intervertebral disc tissues. One factor that
therapeutics include the p38 mitogen-activated protein kinase would be expected to impact anabolic events is local oxygen
(MAPK) inhibitor, which suppressed MMP-3 and IL-1 tension; in the disc this is low, between 2 and 5 % in vivo
expression (Studer et al. 2008) as well as a NF-kB decoy, (Urban 2002; Bartels et al. 1998). Low oxygen tension would
which reduced pain in a rat lumbar disc herniation model be expected to decrease mitochondrial function and oxida-
(Suzuki et al. 2009). tive activity (Bibby et al. 2005). It has been observed that it
Although the exact mechanism is unclear, the use of enhanced the anabolic effects of OP-1 (Miyamoto et al.
gelatinous biomaterials, alone or in combination with growth 2006a), BMP-7 (Tonomura et al. 2007), and TGF-b3 (Abe
factors, has shown some efficacy. These materials may work et al. 2008) by promoting extracellular matrix synthesis by
by providing immediate load support (Joshi et al. 2005) as bovine nucleus pulposus cells. In addition, in MSCs
well as a hydrated environment in which host or embedded (Stoyanov et al. 2011) and notochordal cells (Erwin et al.
cells can proliferate (Collin et al. 2011). Fibrin glue has been 2009), low oxygen tension has been shown to facilitate dif-
used to repair intervertebral discs in a porcine nucleotomy ferentiation and extracellular matrix production. This topic is
model: it suppressed IL-6 and TNF expression and restored discussed further in Chap. 6.
mechanical properties and the glycosaminoglycan (GAG) Low oxygen tension results in lactate accumulation and a
content (Buser et al. 2009). Hyaluronic acid-cross-linked concomitant decrease in media pH (Pichika et al. 2012).
hydrogels have been used in rabbits with less severe annular Dynamic mechanical stimulation in a physiologic range can
needle punctures and were found to improve disc height also stimulate anabolic activities. Thus, applying cycling
(Fig. 25.3a) and T2 MR properties (Bae et al. 2011) tensile strain to discs resulted in F-actin reorganization and
(Fig. 25.3ad), as well as safranin O staining (Nakashima an increase in collagen I expression in outer annulus fibrosus
et al. 2009). Mixing the hydrogel with TGF-b prior to injec- cells and an increase of collagen II expression in the nucleus
tion further increased disc height (Fig. 25.3a) (Bae et al. pulposus (Li et al. 2011). Rat caudal discs compressed in vivo
2011). While little is known about degradation of these at 1.0 MPa stress and 0.01 Hz frequency exhibited increased
a
100
Saline
Gel Only
high TGF+Gel
* *
Disc Height Index
90
*
80
70
0 2 4 6 8 10 12 14 16
Time (weeks after initial puncture)
Fig. 25.3 Treatment of intervertebral disc degeneration using samples than saline samples, consistent with (b) MRI showing larger
injectable hydrogel. Adult (~9 months old) rabbits received annular nucleus pulposus (NP) morphology for the hydrogel samples. (c) T2
puncture. After 4 weeks, animals were divided into three groups and property of the NP also showed a trend of being higher for the hydrogel
received an injection of either saline, hyaluronic acid-based hydrogel, group, while (d) that of the AF did not show much variation. The addi-
or transforming growth factor b3 (TGFb3) mixed with the hydrogel. tion of TGFb3 increased disc height even more (a) but had no effect on
Lateral plane radiographs were taken biweekly. At week 16, animals the T2 property. ROI region of interest (Reproduced from Bae et al.
were sacrificed and MRI was performed to determine T2 relaxation (2011), abstract)
properties of the discs. (a) Disc height index was higher for hydrogel
Fig. 25.3 (continued)

b
L2/ 3
L3/ 4
L4/ 5
Saline Gel Only Gel+TGF
c d
0.8 1.5
0.6
(normalized to L3/4)
1
T2 value
0.4
0.5
Gel+TGF
0.2
Saline
Gel
0 0
Level L2/3 L4/5 L2/3 L4/5
ROI Location Nucleus Anulus
expression of anabolic genes (Maclean et al. 2004). For This is consistent with a study suggesting that OP-1 binds to
tissue-engineering applications, these techniques may be collagen molecules (Reddi 2000); a physical interaction
combined with growth factors and scaffold or gel biomateri- between the proteins would slow degradation and thereby
als to help optimize the function of disc tissues (this topic is explain the long residence time. There are other confounding
discussed in detail in Chap. 26). factors, such as physical characteristics of the intervertebral
disc tissue, the nature of the carrier or vehicle, as well as the
injection location and technique. In addition, the duration
25.6 Limitations of Injection Therapy and time course of the anabolic effect resulting from a single
exposure to a growth factor remain to be established.
While the effects of growth factors on disc cells and tissues Many injectable therapies have inherent limitations that
have been widely studied, questions still remain regarding are dependent on the stage of disc degeneration. Growth fac-
their long-term effects. A major factor influencing efficiency tor injection requires the presence of viable and functional
is the residence time of an injected protein/compound in the cells in the intervertebral disc. In end-stage disc degenera-
disc, which for most agents has not been established. For tion, very little tissue and cells remain (Gruber and Hanley
OP-1, while some studies have suggested a short half-life 1998). Even if viable cells are present in the degenerate discs,
(~minutes) (Larson et al. 2006), others, using radiolabeling, they may exhibit a poor response to the growth factor
have noted times longer than 1 month (Pierce et al. 2006). (Le Maitre et al. 2008). For intervertebral disc with very low
Fig. 25.4 Ultrashort time-

to-echo (UTE) MRI of cartilage
a b
endplates. Cadaveric human spine
was imaged using (a) a
conventional spin-echo
T2-weighted MRI and (b) UTE
MRI. Square intervertebral disk,
arrows cartilage endplates, and
circle bone marrow. While the
region of cartilage endplate is (a)
invisible in conventional MRI, it
is shown with high signal
intensity and contrast in (b) UTE
MRI. Novel imaging techniques
such as UTE MRI may be useful
for evaluation of nutritional
environment of the disks, as well
as selection of suitable subjects
for receiving biological
therapeutics
cellularity, it may be possible to utilize a tissue-engineering trials. One successful attempt to address this issue was an
approach to increase cell number. For example, functional indirect experiment using the rat tail compression model. In
cells recovered from herniated tissues could be used or even this study, OP-1 was injected into the disc, which was then
MSCs derived from bone marrow or other tissue sites transplanted into the dorsal root ganglion in the same rat
(Nishimura and Mochida 1998; Okuma et al. 2000; Anderson (Kawakami et al. 2005). It was noted that the animal exhib-
et al. 2005; Gruber et al. 2002; Ganey et al. 2003). ited less allodynia after OP-1 injection. Other approaches
Another consideration is that the level of nutrients in the include behavioral (Olmarker 2008) or gait analysis (Shamji
disc may be low (Urban et al. 2004) due to vertebral endplate et al. 2009). Results of these studies provide preliminary evi-
sclerosis or cartilage endplate calcification (Bae et al. 2010). dence that an injection therapy is effective in reducing pain.
This region can be visualized using novel imaging techniques Moreover, they permit pain generation to be related to
such as ultrashort time-to-echo MRI (Fig. 25.4) (Bae et al. changes in the biochemical profile of disc tissues following
2010, 2012). The perfusion of solutes from the vertebral treatment.
body into intervertebral disc has been assessed indirectly
using contrast-enhanced MRI in human subjects (Rajasekaran
et al. 2004, 2008). It remains to be determined if these tech- 25.7 Conclusion
niques provide information on the time course of subsequent
disc degeneration. This chapter outlined approaches using injectable therapeu-
Lastly, assessment of pain reduction is difficult when tics to enhance biologic repair or regeneration of the inter-
using a preclinical animal model; this is of critical impor- vertebral disc and described methods available to evaluate
tance, as pain is the primary outcome measure for clinical the efficacy of the treatment. Using intervertebral disc cells
in vitro and discs in animal models in vivo, there is abundant cells in alginate beads. In: 54th annual meeting of Orthopedic
evidence supporting the use and efficacy of treatment using Research Society, San Francisco, p 1413
Acosta FL Jr, Metz L, Adkisson HD, Liu J, Carruthers-Liebenberg E,
growth factors and agents that exert anti-catabolic activity. In Milliman C, Maloney M, Lotz JC (2011) Porcine intervertebral disc
current preclinical animal studies, outcomes focus mainly on repair using allogeneic juvenile articular chondrocytes or mesen-
structural modification, and little is known about pain reduc- chymal stem cells. Tissue Eng Part A 17(2324):30453055.
tion. More behavioral studies, animal models of pain, and doi:10.1089/ten.tea.2011.0229
Aguiar DJ, Johnson SL, Oegema TR (1999) Notochordal cells interact
other novel methods such as functional MRI may be needed with nucleus pulposus cells: regulation of proteoglycan synthesis.
to better understand this important issue. In addition, further Exp Cell Res 246(1):129137
studies on large animals with intervertebral disc biology sim- Ahn S-H, Teng P-N, Niyibizi C, Gilbertson L, Kang J (2002a) The
ilar to humans are desired. Ultimately, using agents discussed effects of BMP-12 and BMP-2 on proteoglycan and collagen syn-
thesis in nucleus pulposus cells from human degenerated discs. In:
earlier, intradiscal injections offer great potential for treat- The International Society for the Study of the Lumbar Spine, 29th
ment of patients with chronic discogenic low back pain. Annual Meeting Proceeding, Cleveland, 1418 May 2002, p 49
Results of clinical trials currently underway will provide Ahn SH, Cho YW, Ahn MW, Jang SH, Sohn YK, Kim HS (2002b)
important initial safety and efficacy information. MRNA expression of cytokines and chemokines in herniated lum-
bar intervertebral discs. Spine 27(9):911917
Akeda K, An HS, Pichika R, Attawia M, Thonar EJ, Lenz ME, Uchida A,
Masuda K (2006) Platelet-rich plasma (PRP) stimulates the extra-
25.8 Summary of Critical Concepts cellular matrix metabolism of porcine nucleus pulposus and anulus
Discussed in the Chapter fibrosus cells cultured in alginate beads. Spine 31(9):959966
An HS, Takegami K, Kamada H, Nguyen CM, Thonar EJ, Singh K,
Andersson GB, Masuda K (2005) Intradiscal administration of
Intervertebral disc degeneration involves an imbalance osteogenic protein-1 increases intervertebral disc height and proteo-
between the anabolic and catabolic activities of disc cells glycan content in the nucleus pulposus in normal adolescent rabbits.
and a decrease in the number of functional cells. Spine 30(1):2531
Anderson DG, Izzo MW, Hall DJ, Vaccaro AR, Hilibrand A, Arnold W,
Therapeutic agents that can alter the activities of the inter- Tuan RS, Albert TJ (2002) Comparative gene expression profiling
vertebral disc cells by direct injections are being actively of normal and degenerative discs: analysis of a rabbit annular lac-
pursued. eration model. Spine 27(12):12911296
Therapeutic agents include anabolic growth factors, bio- Anderson DG, Albert TJ, Fraser JK, Risbud M, Wuisman P, Meisel HJ,
Tannoury C, Shapiro I, Vaccaro AR (2005) Cellular therapy for disc
stimulatory biomaterials, and agents antagonistic to degeneration. Spine 30(17 Suppl):S14S19
matrix-degrading enzymes and cytokines, some of which Ando T, Kato F, Mimatsu K, Iwata H (1995) Effects of chondroitinase
are currently under clinical trials. ABC on degenerative intervertebral discs. Clin Orthop 318:214221
The efficacy of candidate agents is initially evaluated Annunen S, Paassilta P, Lohiniva J, Perala M, Pihlajamaa T, Karppinen J,
Tervonen O, Kroger H, Lahde S, Vanharanta H, Ryhanen L,
in vitro using various cell and explant culture models and Goring HH, Ott J, Prockop DJ, Ala-Kokko L (1999) An allele of
in vivo using preclinical animal models. COL9A2 associated with intervertebral disc disease. Science
There is a need for continued development of model sys- 285(5426):409412
tems and evaluation techniques to assess the efficacy and Arner EC, Pratta MA (1989) Independent effects of interleukin-1 on
proteoglycan breakdown, proteoglycan synthesis, and prostaglandin
safety aspects of injected agents, as well as an understand- E2 release from cartilage in organ culture. Arthritis Rheum 32:
ing of the pathogenesis and progression of degenerative 288297
disc disease. Arpornmaeklong P, Kochel M, Depprich R, Kubler NR, Wurzler KK
(2004) Influence of platelet-rich plasma (PRP) on osteogenic dif-
ferentiation of rat bone marrow stromal cells. An in vitro study. Int
Acknowledgements This work was supported in part by NIH J Oral Maxillofac Surg 33(1):6070. doi:10.1054/ijom.2003.0492
P01AR48152 and K01AR059764. Bae WC, Yoshikawa T, Kakutani K, Hammed A, Znamirowski R,
Chung CB, Bydder GM, Masuda K (2009) Effect of rhGDF-5 on
Disclosures Koichi Masuda, M.D.: Research/Institutional Support: the thrombin model of rabbit intervertebral disc degeneration; T2
Medtronic, Nippon Zoki, Nuvasive, Biotime Quantification Using 3T MRI. Orthop Res Soc Trans:2129
Bae WC, Du J, Bydder GM, Chung CB (2010) Conventional and
ultrashort time-to-echo magnetic resonance imaging of articular
cartilage, meniscus, and intervertebral disk. Top Magn Reson
Imaging 21(5):275289. doi:10.1097/RMR.0b013e31823ccebc
References Bae WC, Yoshikawa T, Yamaguchi T, Hemmad AR, Rana S, Inoue N,
Bydder GM, Masuda K (2011) Repair of intervertebral disc degen-
Abbott RD, Purmessur D, Monsey RD, Iatridis JC (2012) Regenerative eration using injectable hydrogel: imaging assessments. Trans
potential of TGFbeta3 + Dex and notochordal cell conditioned Orthop Res Soc 36:362
media on degenerated human intervertebral disc cells. J Orthop Res Bae WC, Statum S, Zhang Z, Yamaguchi T, Wolfson T, Gamst AC,
30(3):482488. doi:10.1002/jor.21534 Du J, Bydder GM, Masuda K, Chung CB (2012) Morphology of
Abe Y, Asanuma K, Tonomura H, Kimura T, An H, Masuda K (2008) the cartilaginous endplates in human intervertebral disks with
Transforming growth factor-3 and low oxygen tension stimulate ultrashort echo time MR imaging. Radiology. doi:10.1148/
synthesis of proteoglycan and collagen by bovine intervertebral disc radiol.12121181
Bartels EM, Fairbank JC, Winlove CP, Urban JP (1998) Oxygen and Fry TR, Eurell JC, Johnson AL, Brown MD, Losonsky JM, Schaeffer DJ
lactate concentrations measured in vivo in the intervertebral discs of (1991) Radiographic and histologic effects of chondroitinase ABC
patients with scoliosis and back pain. Spine 23(1):17 on normal canine lumbar intervertebral disc. Spine 16(7):816819
Benton HP, Tyler JA (1988) Inhibition of cartilage proteoglycan Ganey T, Libera J, Moos V, Alasevic O, Fritsch KG, Meisel HJ,
synthesis by interleukin I. Biochem Biophys Res Commun 154(1): Hutton WC (2003) Disc chondrocyte transplantation in a canine
421428 model: a treatment for degenerated or damaged intervertebral disc.
Bibby SR, Jones DA, Ripley RM, Urban JP (2005) Metabolism of the Spine 28(23):26092620
intervertebral disc: effects of low levels of oxygen, glucose, and pH Genevay S, Stingelin S, Gabay C (2004) Efficacy of etanercept in the
on rates of energy metabolism of bovine nucleus pulposus cells. treatment of acute, severe sciatica: a pilot study. Ann Rheum Dis
Spine 30(5):487496 63(9):11201123
Boxberger JI, Auerbach JD, Sen S, Elliott DM (2008) An in vivo model Genevay S, Finckh A, Mezin F, Tessitore E, Guerne PA (2009) Influence
of reduced nucleus pulposus glycosaminoglycan content in the rat of cytokine inhibitors on concentration and activity of MMP-1 and
lumbar intervertebral disc. Spine 33(2):146154. doi:10.1097/ MMP-3 in disc herniation. Arthritis Res Ther 11(6):R169.
BRS.0b013e31816054f8, 00007632-200801150-00005 [pii] doi:10.1186/ar2858
Burke JG, Watson RW, Conhyea D, McCormack D, Dowling FE, Walsh Gilbertson L, Ahn SH, Teng PN, Studer RK, Niyibizi C, Kang JD (2008)
MG, Fitzpatrick JM (2003) Human nucleus pulposus can respond to The effects of recombinant human bone morphogenetic protein-2,
a pro-inflammatory stimulus. Spine 28(24):26852693 recombinant human bone morphogenetic protein-12, and adenoviral
Buser Z, Kuelling F, Jane L, Liebenberg E, Tang J, Thorne K, bone morphogenetic protein-12 on matrix synthesis in human annu-
Coughlin D, Lotz J (2009) Fibrin injection stimulates early disc lus fibrosis and nucleus pulposus cells. Spine J 8(3):449456.
healing in the porcine model. Spine J 9(10 Suppl):105S. doi:10.1016/j.spinee.2006.11.006, S1529-9430(06)01068-0 [pii]
doi:10.1016/j.spinee.2009.08.241 Gokorsch S, Weber C, Wedler T, Czermak P (2005) A stimulation unit
Castillo TN, Pouliot MA, Kim HJ, Dragoo JL (2011) Comparison of for the application of mechanical strain on tissue engineered anulus
growth factor and platelet concentration from commercial platelet- fibrosus cells: a new system to induce extracellular matrix synthesis
rich plasma separation systems. Am J Sports Med 39(2):266271. by anulus fibrosus cells dependent on cyclic mechanical strain. Int J
doi:10.1177/0363546510387517 Artif Organs 28(12):12421250
Chen WH, Lo WC, Lee JJ, Su CH, Lin CT, Liu HY, Lin TW, Lin WC, Gruber HE, Hanley EN Jr (1998) Analysis of aging and degeneration of
Huang TY, Deng WP (2006) Tissue-engineered intervertebral disc the human intervertebral disc. Comparison of surgical specimens
and chondrogenesis using human nucleus pulposus regulated with normal controls. Spine 23(7):751757
through TGF-beta1 in platelet-rich plasma. J Cell Physiol Gruber HE, Fisher EC Jr, Desai B, Stasky AA, Hoelscher G, Hanley EN
209(3):744754. doi:10.1002/jcp.20765 Jr (1997) Human intervertebral disc cells from the annulus: three-
Chen WH, Liu HY, Lo WC, Wu SC, Chi CH, Chang HY, Hsiao SH, dimensional culture in agarose or alginate and responsiveness to
Wu CH, Chiu WT, Chen BJ, Deng WP (2009) Intervertebral disc TGF-beta1. Exp Cell Res 235(1):1321
regeneration in an ex vivo culture system using mesenchymal stem Gruber HE, Norton HJ, Hanley EN Jr (2000) Anti-apoptotic effects of
cells and platelet-rich plasma. Biomaterials 30(29):55235533. IGF-1 and PDGF on human intervertebral disc cells in vitro. Spine
doi:10.1016/j.biomaterials.2009.07.019 25(17):21532157
Chujo T, An HS, Akeda K, Miyamoto K, Muehleman C, Attawia M, Gruber HE, Johnson TL, Leslie K, Ingram JA, Martin D, Hoelscher G,
Andersson G, Masuda K (2006) Effects of growth differentiation Banks D, Phieffer L, Coldham G, Hanley EN Jr (2002) Autologous
factor-5 on the intervertebral discin vitro bovine study and in vivo intervertebral disc cell implantation: a model using Psammomys
rabbit disc degeneration model study. Spine 31(25):29092917 obesus, the sand rat. Spine 27(15):16261633
Collin EC, Grad S, Zeugolis DI, Vinatier CS, Clouet JR, Guicheux JJ, Henderson N, Stanescu V, Cauchoix J (1991) Nucleolysis of the rabbit
Weiss P, Alini M, Pandit AS (2011) An injectable vehicle for nucleus intervertebral disc using chondroitinase ABC. Spine 16(2):203208
pulposus cell-based therapy. Biomaterials 32(11):28622870. Hoogendoorn RJ, Wuisman PI, Smit TH, Everts VE, Helder MN (2007)
doi:10.1016/j.biomaterials.2011.01.018 Experimental intervertebral disc degeneration induced by chondroi-
Desmoulin GT, Hewitt CR, Hunter CJ (2011) Disc strain and resulting tinase ABC in the goat. Spine 32(17):18161825. doi:10.1097/
positive mRNA expression from application of a noninvasive treat- BRS.0b013e31811ebac5, 00007632-200708010-00004 [pii]
ment. Spine 36(14):E921E928. doi:10.1097/BRS.0b013e3181fd78b3 Hoogendoorn RJ, Helder MN, Kroeze RJ, Bank RA, Smit TH,
Dingle JT, Horner A, Shield M (1991) The sensitivity of synthesis of Wuisman PI (2008) Reproducible long-term disc degeneration in
human cartilage matrix to inhibition by IL-1 suggests a mechanism a large animal model. Spine 33(9):949954. doi:10.1097/
for the development of osteoarthritis. Cell Biochem Funct BRS.0b013e31816c90f0, 00007632-200804200-00004 [pii]
9(2):99102 Huang KY, Yan JJ, Hsieh CC, Chang MS, Lin RM (2007) The in vivo
Dong DM, Yao M, Liu B, Sun CY, Jiang YQ, Wang YS (2007) biological effects of intradiscal recombinant human bone morpho-
Association between the -1306C/T polymorphism of matrix metal- genetic protein-2 on the injured intervertebral disc: an animal
loproteinase-2 gene and lumbar disc disease in Chinese young experiment. Spine 32(11):11741180
adults. Eur Spine J 16(11):19581961. doi:10.1007/s00586-007- Iatridis JC, Godburn K, Wuertz K, Alini M, Roughley PJ (2011)
0454-3 Region-dependent aggrecan degradation patterns in the rat interver-
Dugrillon A, Eichler H, Kern S, Kluter H (2002) Autologous concen- tebral disc are affected by mechanical loading in vivo. Spine
trated platelet-rich plasma (cPRP) for local application in bone 36(3):203209. doi:10.1097/BRS.0b013e3181cec247
regeneration. Int J Oral Maxillofac Surg 31(6):615619 Igarashi T, Kikuchi S, Shubayev V, Myers RR (2000) 2000 Volvo
Erwin WM, Las Heras F, Islam D, Fehlings MG, Inman RD (2009) The Award winner in basic science studies: exogenous tumor necrosis
regenerative capacity of the notochordal cell: tissue constructs gen- factor-alpha mimics nucleus pulposus-induced neuropathology.
erated in vitro under hypoxic conditions. J Neurosurg Spine Molecular, histologic, and behavioral comparisons in rats. Spine
10(6):513521. doi:10.3171/2009.2.SPINE08578 25(23):29752980
Eurell JA, Brown MD, Ramos M (1990) The effects of chondroitinase Imai Y, An H, Thonar E, Muehleman C, Okuma M, Matsumoto T,
ABC on the rabbit intervertebral disc. A roentgenographic and his- Andersson G, Masuda K (2003) Co-injected recombinant human
tologic study. Clin Orthop 256:238243 osteogenic protein-1 minimizes chondroitinase ABC-induced
intervertebral disc degeneration: an in vivo study using a rabbit anti-catabolic effects in the intervertebral disc. J Cell Physiol
model. Trans Orthop Res Soc:1143 227(4):15121520. doi:10.1002/jcp.22867
Imai Y, Miyamoto K, An HS, Thonar EJ, Andersson GB, Masuda K Larson JW 3rd, Levicoff EA, Gilbertson LG, Kang JD (2006) Biologic
(2007a) Recombinant human osteogenic protein-1 upregulates modification of animal models of intervertebral disc degeneration.
proteoglycan metabolism of human anulus fibrosus and nucleus J Bone Joint Surg Am 88(Suppl 2):8387
pulposus cells. Spine 32(12):13031309; discussion 1310 Le Maitre CL, Freemont AJ, Hoyland JA (2004) Localization of
Imai Y, Okuma M, An H, Nakagawa K, Yamada M, Thonar E, degradative enzymes and their inhibitors in the degenerate human
Muehleman C, Masuda K (2007b) Restoration of disc height loss by intervertebral disc. J Pathol 204(1):4754
recombinant human osteogenic protein-1 injection into interverte- Le Maitre CL, Freemont AJ, Hoyland JA (2005) The role of interleu-
bral discs undergoing degeneration induced by an intradiscal injec- kin-1 in the pathogenesis of human intervertebral disc degeneration.
tion of chondroitinase ABC. Spine 32(11):11971205 Arthritis Res Ther 7(4):R732R745
Jimbo K, Park JS, Yokosuka K, Sato K, Nagata K (2005) Positive Le Maitre CL, Hoyland JA, Freemont AJ (2007a) Catabolic cytokine
feedback loop of interleukin-1beta upregulating production of expression in degenerate and herniated human intervertebral discs:
inflammatory mediators in human intervertebral disc cells in vitro. IL-1beta and TNFalpha expression profile. Arthritis Res Ther
J Neurosurg Spine 2(5):589595 9(4):R77. doi:10.1186/ar2275, ar2275 [pii]
Joshi A, Mehta S, Vresilovic E, Karduna A, Marcolongo M (2005) Le Maitre CL, Hoyland JA, Freemont AJ (2007b) Interleukin-1 recep-
Nucleus implant parameters significantly change the compressive tor antagonist delivered directly and by gene therapy inhibits matrix
stiffness of the human lumbar intervertebral disc. J Biomech Eng degradation in the intact degenerate human intervertebral disc: an in
127(3):536540 situ zymographic and gene therapy study. Arthritis Res Ther
Kakutani K, Kanaji A, Asanuma K, Pichika R, An H, Yamada H, 9(4):R83. doi:10.1186/ar2282, ar2282 [pii]
Masuda K (2008) Effect of IL-1 receptor antagonist and soluble Le Maitre CL, Frain J, Fotheringham AP, Freemont AJ, Hoyland JA
TNF receptor on the anabolism of human intervertebral disc cells. (2008) Human cells derived from degenerate intervertebral discs
In: 54th Annual Meeting of Orthopedic Research Society, San respond differently to those derived from non-degenerate interverte-
Francisco, p 442 bral discs following application of dynamic hydrostatic pressure.
Kang JD, Georgescu HI, McIntyre-Larkin L, Stefanovic-Racic M, Biorheology 45(5):563575
Donaldson WF 3rd, Evans CH (1996) Herniated lumbar Li S, Jia X, Duance VC, Blain EJ (2011) The effects of cyclic tensile
intervertebral discs spontaneously produce matrix metalloprotei- strain on the organisation and expression of cytoskeletal elements in
nases, nitric oxide, interleukin-6, and prostaglandin E2. Spine bovine intervertebral disc cells: an in vitro study. Eur Cell Mater
21(3):271277 21:508522
Kang JD, Stefanovic-Racic M, McIntyre LA, Georgescu HI, Evans CH Maclean JJ, Lee CR, Alini M, Iatridis JC (2004) Anabolic and catabolic
(1997) Toward a biochemical understanding of human interverte- mRNA levels of the intervertebral disc vary with the magnitude and
bral disc degeneration and herniation. Contributions of nitric oxide, frequency of in vivo dynamic compression. J Orthop Res 22(6):
interleukins, prostaglandin E2, and matrix metalloproteinases. 11931200
Spine 22(10):10651073 Masuda K, An HS (2006) Prevention of disc degeneration with growth
Karppinen J, Korhonen T, Malmivaara A, Paimela L, Kyllonen E, factors. Eur Spine J 15(Suppl 15):422432
Lindgren KA, Rantanen P, Tervonen O, Niinimaki J, Seitsalo S, Masuda K, Hejna M, Thonar EJ-MA (2000) Alginate-recovered-
Hurri H (2003) Tumor necrosis factor-alpha monoclonal antibody, chondrocyte method (ARC method): a novel two-step method for
infliximab, used to manage severe sciatica. Spine 28(8):750753; the formation of cartilage tissue. Ortho Res Soc Trans 25:620
discussion 753754 Masuda K, Takegami K, An H, Kumano F, Chiba K, Andersson GB,
Kato F, Mimatsu K, Kawakami N, Iwata H, Miura T (1992) Serial Schmid T, Thonar E (2003) Recombinant osteogenic protein-1
changes observed by magnetic resonance imaging in the interverte- upregulates extracellular matrix metabolism by rabbit annulus
bral disc after chemonucleolysis. A consideration of the mechanism fibrosus and nucleus pulposus cells cultured in alginate beads.
of chemonucleolysis. Spine 17(8):934939 J Orthop Res 21(5):922930
Kawaguchi Y, Osada R, Kanamori M, Ishihara H, Ohmori K, Matsui H, Masuda K, Oegema TR Jr, An HS (2004) Growth factors and treatment
Kimura T (1999) Association between an aggrecan gene polymor- of intervertebral disc degeneration. Spine 29(23):27572769
phism and lumbar disc degeneration. Spine 24(23):24562460 Masuda K, Imai Y, Okuma M, Muehleman C, Nakagawa K, Akeda K,
Kawakami M, Matsumoto T, Hashizume H, Kuribayashi K, Chubinskaya Thonar E, Andersson G, An HS (2006) Osteogenic protein-1 injec-
S, Yoshida M (2005) Osteogenic protein-1 (osteogenic protein-1/ tion into a degenerated disc induces the restoration of disc height
bone morphogenetic protein-7) inhibits degeneration and pain- and structural changes in the rabbit anular puncture model. Spine
related behavior induced by chronically compressed nucleus pulpo- 31(7):742754
sus in the rat. Spine 30(17):19331939 Mazzocca AD, McCarthy MB, Chowaniec DM, Cote MP, Romeo AA,
Kim DJ, Moon SH, Kim H, Kwon UH, Park MS, Han KJ, Hahn SB, Bradley JP, Arciero RA, Beitzel K (2012) Platelet-rich
Lee HM (2003) Bone morphogenetic protein-2 facilitates expres- plasma differs according to preparation method and human
sion of chondrogenic, not osteogenic, phenotype of human interver- variability. J Bone Joint Surg Am 94(4):308316. doi:10.2106/
tebral disc cells. Spine 28(24):26792684 JBJS.K.00430
Kim JH, Deasy BM, Seo HY, Studer RK, Vo NV, Georgescu HI, Miyamoto H, Saura R, Harada T, Doita M, Mizuno K (2000) The role
Sowa GA, Kang JD (2009) Differentiation of intervertebral noto- of cyclooxygenase-2 and inflammatory cytokines in pain induction
chordal cells through live automated cell imaging system in vitro. of herniated lumbar intervertebral disc. Kobe J Med Sci
Spine 34(23):24862493. doi:10.1097/BRS.0b013e3181b26ed1 46(12):1328
Kim JS, Ellman MB, An HS, van Wijnen AJ, Borgia JA, Im HJ (2010) Miyamoto K, Pichika R, An H, Thonar E-M, Akeda K, Lenz M,
Insulin-like growth factor 1 synergizes with bone morphogenetic Masuda K (2005) Tumor Necrosis Factor-a exhibits potent effects
protein 7-mediated anabolism in bovine intervertebral disc cells. on the metabolism of human intervertebral disc cells. Trans Orthop
Arthritis Rheum 62(12):37063715. doi:10.1002/art.27733 Res Soc:188
Kim JS, Ellman MB, An HS, Yan D, van Wijnen AJ, Murphy G, Miyamoto K, An HS, Urban JP, Chujo T, Akeda K, Masuda K (2006a)
Hoskin DW, Im HJ (2012) Lactoferricin mediates anabolic and Low oxygen tension enhances anabolic effects of osteogenic
protein-1 on the extracellular matrix metabolism of bovine interver- the human intervertebral disc at early and advanced stages of
tebral disc cells. Trans Orthop Res Soc 31:1205 disc degeneration. Spine 32(23):25962603. doi:10.1097/
Miyamoto K, Masuda K, Kim JG, Inoue N, Akeda K, Andersson GB, BRS.0b013e318158cb85, 00007632-200711010-00014 [pii]
An HS (2006b) Intradiscal injections of osteogenic protein-1 restore Pichika R, Penta K, Yoshikawa T, Hemmad A, Lenz ME,
the viscoelastic properties of degenerated intervertebral discs. Spine Szymkowski DE, Binette F, Masuda K (2011) A new approach for
J 6(6):692703 anti-TNF therapy: dominant-negative TNF effectively antagonizes
Mwale F, Masuda K, Pichika R, Epure LM, Yoshikawa T, Hemmad A, IL-1b-induced catabolic changes in human intervertebral disc cells.
Roughley PJ, Antoniou J (2011) The efficacy of Link N as a media- Trans Orthop Res Soc 36:709, abstract
tor of repair in a rabbit model of intervertebral disc degeneration. Pichika R, Yamaguchi T, Taborek A, Bae WC, Sah RL, Masuda K
Arthritis Res Ther 13(4):R120. doi:10.1186/ar3423 (2012) Interactive effect of IL-1 and BMP-7 on cytokines, matrix
Nagae M, Ikeda T, Mikami Y, Hase H, Ozawa H, Matsuda K, components and gene expression of human intervertebral disc cells:
Sakamoto H, Tabata Y, Kawata M, Kubo T (2007) Intervertebral implications of pH and oxygen tension on biological therapy. Trans
disc regeneration using platelet-rich plasma and biodegradable gela- Orthop Res Soc 37:64
tin hydrogel microspheres. Tissue Eng 13(1):147158 Pierce A, Feng M, Masuda K (2006) Distribution, pharmacokinetics
Nakashima S, Matsuyama Y, Takahashi K, Satoh T, Koie H, Kanayama K, and excretion of 125-Iodine labeled BMP-7 [OP-1] following a
Tsuji T, Maruyama K, Imagama S, Sakai Y, Ishiguro N (2009) single-dose administration in lumbar IVD or knee joint of NZW
Regeneration of intervertebral disc by the intradiscal application of rabbits. In: Sixth International Conference on Bone Morphogenetic
cross-linked hyaluronate hydrogel and cross-linked chondroitin sul- Proteins, Cavtat, Croatia, 1115 Oct 2006
fate hydrogel in a rabbit model of intervertebral disc injury. Biomed Pratsinis H, Kletsas D (2007) PDGF, bFGF and IGF-I stimulate the
Mater Eng 19(6):421429. doi:10.3233/BME-2009-0608 proliferation of intervertebral disc cells in vitro via the activation of
Nasto LA, Seo HY, Robinson AR, Tilstra JS, Clauson CL, Sowa GA, the ERK and Akt signaling pathways. Eur Spine J 16(11):1858
Ngo K, Dong Q, Pola E, Lee JY, Niedernhofer LJ, Kang JD, Robbins 1866. doi:10.1007/s00586-007-0408-9
PD, Vo NV (2012) ISSLS prize winner: inhibition of Nf-kb activity Pratsinis H, Constantinou V, Pavlakis K, Sapkas G, Kletsas D (2012)
ameliorates age-associated disc degeneration in a mouse model of Exogenous and autocrine growth factors stimulate human interver-
accelerated aging. Spine (Phila Pa 1976). doi:10.1097/BRS. tebral disc cell proliferation via the ERK and Akt pathways. J Orthop
0b013e31824ee8f7 Res 30(6):958964. doi:10.1002/jor.22017
Nishimura K, Mochida J (1998) Percutaneous reinsertion of the nucleus Rajasekaran S, Babu JN, Arun R, Armstrong BR, Shetty AP,
pulposus. An experimental study. Spine 23(14):15311538 Murugan S (2004) ISSLS prize winner: a study of diffusion in
Noponen-Hietala N, Virtanen I, Karttunen R, Schwenke S, Jakkula E, human lumbar discs: a serial magnetic resonance imaging study
Li H, Merikivi R, Barral S, Ott J, Karppinen J, Ala-Kokko L (2005) documenting the influence of the endplate on diffusion in normal
Genetic variations in IL6 associate with intervertebral disc disease and degenerate discs. Spine 29(23):26542667
characterized by sciatica. Pain 114(12):186194 Rajasekaran S, Venkatadass K, Naresh Babu J, Ganesh K, Shetty AP
Norcross JP, Lester GE, Weinhold P, Dahners LE (2003) An in vivo (2008) Pharmacological enhancement of disc diffusion and differ-
model of degenerative disc disease. J Orthop Res 21(1):183188 entiation of healthy, ageing and degenerated discs: results from in-
Obata S, Akeda K, Imanishi T, Masuda K, Bae W, Morimoto R, vivo serial post-contrast MRI studies in 365 human lumbar discs.
Asanuma Y, Kasai Y, Uchida A, Sudo A (2012) Effect of autologous Eur Spine J 17(5):626643
platelet-rich plasma-releasate on intervertebral disc degeneration in Rannou F, Corvol MT, Hudry C, Anract P, Dumontier MF, Tsagris L,
the rabbit anular puncture model: a preclinical study. Arthritis Res Revel M, Poiraudeau S (2000) Sensitivity of anulus fibrosus cells to
Ther 14(6):R241. doi:10.1186/ar4084 interleukin 1 beta. Comparison with articular chondrocytes. Spine
Okoro T, Tafazal SI, Longworth S, Sell PJ (2010) Tumor necrosis 25(1):1723
alpha-blocking agent (etanercept): a triple blind randomized con- Reddi AH (2000) Morphogenetic messages are in the extracellular
trolled trial of its use in treatment of sciatica. J Spinal Disord Tech matrix: biotechnology from bench to bedside. Biochem Soc Trans
23(1):7477. doi:10.1097/BSD.0b013e31819afdc4 28(4):345349
Okuda K, Kawase T, Momose M, Murata M, Saito Y, Suzuki H, Sakuma M, Fujii N, Takahashi T, Hoshino J, Miyauchi S, Iwata H
Wolff LF, Yoshie H (2003) Platelet-rich plasma contains high levels (2002) Effect of chondroitinase ABC on matrix metalloproteinases
of platelet-derived growth factor and transforming growth factor- and inflammatory mediators produced by intervertebral disc of rab-
beta and modulates the proliferation of periodontally related cells bit in vitro. Spine 27(6):576580
in vitro. J Periodontol 74(6):849857 Sawamura K, Ikeda T, Nagae M, Okamoto S, Mikami Y, Hase H,
Okuma M, Mochida J, Nishimura K, Sakabe K, Seiki K (2000) Ikoma K, Yamada T, Sakamoto H, Matsuda K, Tabata Y, Kawata M,
Reinsertion of stimulated nucleus pulposus cells retards interverte- Kubo T (2009) Characterization of in vivo effects of platelet-rich
bral disc degeneration: an in vitro and in vivo experimental study. plasma and biodegradable gelatin hydrogel microspheres on
J Orthop Res 18(6):988997 degenerated intervertebral discs. Tissue Eng Part A 15(12):
Olmarker K (2008) Puncture of a lumbar intervertebral disc induces 37193727. doi:10.1089/ten.TEA.2008.0697
changes in spontaneous pain behavior: an experimental study in Schollmeier G, Lahr-Eigen R, Lewandrowski KU (2000) Observations
rats. Spine 33(8):850855. doi:10.1097/BRS.0b013e31816b46ca, on fiber-forming collagens in the anulus fibrosus. Spine 25(21):
00007632-200804150-00005 [pii] 27362741
Olmarker K, Larsson K (1998) Tumor necrosis factor alpha and nucleus- Scott JE, Haigh M (1986) Proteoglycan-collagen interactions in
pulposus-induced nerve root injury. Spine 23(23):25382544 intervertebral disc. A chondroitin sulphate proteoglycan associates
Osada R, Ohshima H, Ishihara H, Yudoh K, Sakai K, Matsui H, Tsuji H with collagen fibrils in rabbit annulus fibrosus at the d-e bands.
(1996) Autocrine/paracrine mechanism of insulin-like growth Biosci Rep 6(10):879888
factor-1 secretion, and the effect of insulin-like growth factor-1 on Seki S, Kawaguchi Y, Chiba K, Mikami Y, Kizawa H, Oya T, Mio F, Mori
proteoglycan synthesis in bovine intervertebral discs. J Orthop Res M, Miyamoto Y, Masuda I, Tsunoda T, Kamata M, Kubo T, Toyama
14(5):690699 Y, Kimura T, Nakamura Y, Ikegawa S (2005) A functional SNP in
Patel KP, Sandy JD, Akeda K, Miyamoto K, Chujo T, An HS, Masuda K CILP, encoding cartilage intermediate layer protein, is associated with
(2007) Aggrecanases and aggrecanase-generated fragments in susceptibility to lumbar disc disease. Nat Genet 37(6):607612
Seki S, Asanuma-Abe Y, Masuda K, Kawaguchi Y, Asanuma K, Takahashi M, Haro H, Wakabayashi Y, Kawa-uchi T, Komori H,
Muehleman C, Iwai A, Kimura T (2009) Effect of small interfer- Shinomiya K (2001) The association of degeneration of the inter-
ence RNA (siRNA) for ADAMTS5 on intervertebral disc degenera- vertebral disc with 5a/6a polymorphism in the promoter of the
tion in the rabbit anular needle-puncture model. Arthritis Res Ther human matrix metalloproteinase-3 gene. J Bone Joint Surg Br
11(6):R166. doi:10.1186/ar2851 83(4):491495
Shamji MF, Betre H, Kraus VB, Chen J, Chilkoti A, Pichika R, Takegami K, Thonar EJ, An HS, Kamada H, Masuda K (2002)
Masuda K, Setton LA (2007) Development and characterization of Osteogenic protein-1 enhances matrix replenishment by interverte-
a fusion protein between thermally responsive elastin-like polypep- bral disc cells previously exposed to interleukin-1. Spine 27(12):
tide and interleukin-1 receptor antagonist: sustained release of a 13181325
local antiinflammatory therapeutic. Arthritis Rheum 56(11):3650 Takegami K, An HS, Kumano F, Chiba K, Thonar EJ, Singh K,
3661. doi:10.1002/art.22952 Masuda K (2005) Osteogenic protein-1 is most effective in
Shamji MF, Allen KD, So S, Jing L, Adams SB Jr, Schuh R, Huebner J, stimulating nucleus pulposus and annulus fibrosus cells to repair
Kraus VB, Friedman AH, Setton LA, Richardson WJ (2009) Gait their matrix after chondroitinase ABC-induced in vitro chemonu-
abnormalities and inflammatory cytokines in an autologous nucleus cleolysis. Spine J 5(3):231238
pulposus model of radiculopathy. Spine 34(7):648654. doi:10.1097/ Thompson JP, Oegema TJ, Bradford DS (1991) Stimulation of mature
BRS.0b013e318197f013, 00007632-200904010-00004 [pii] canine intervertebral disc by growth factors. Spine 16(3):
Sinclair SM, Shamji MF, Chen J, Jing L, Richardson WJ, Brown CR, 253260
Fitch RD, Setton LA (2011) Attenuation of inflammatory Tilkeridis C, Bei T, Garantziotis S, Stratakis CA (2005) Association of
eventsin human intervertebral disc cells with a tumor necrosis a COL1A1 polymorphism with lumbar disc disease in young
factor antagonist. Spine 36(15):11901196. doi:10.1097/ military recruits. J Med Genet 42(7):e44
BRS.0b013e3181ebdb43 Tobinick EL, Britschgi-Davoodifar S (2003) Perispinal TNF-alpha
Sobajima S, Shimer AL, Chadderdon RC, Kompel JF, Kim JS, inhibition for discogenic pain. Swiss Med Wkly 133(1112):170
Gilbertson LG, Kang JD (2005) Quantitative analysis of gene 177, doi:2003/11/smw-10163
expression in a rabbit model of intervertebral disc degeneration by Tonomura H, Miyamoto K, Sah R, Urban J, Kubo T, Asanuma K,
real-time polymerase chain reaction. Spine J 5(1):1423 Lenz M, Masuda K (2007) Low oxygen tension and BMP-7 syner-
Solovieva S, Kouhia S, Leino-Arjas P, Ala-Kokko L, Luoma K, gistically stimulate synthesis and deposition of proteoglycan by
Raininko R, Saarela J, Riihimaki H (2004) Interleukin 1 polymor- bovine articular chondrocytes in alginate beads. Ortho Res Soc
phisms and intervertebral disc degeneration. Epidemiology Trans 32:257
15(5):626633 Tozum TF, Demiralp B (2003) Platelet-rich plasma: a promising
Solovieva S, Lohiniva J, Leino-Arjas P, Raininko R, Luoma K, innovation in dentistry. J Can Dent Assoc 69(10):664
Ala-Kokko L, Riihimaki H (2006) Intervertebral disc degeneration Trout J, Buckwalter J, Moore K, Landa S (1982) Ultrastructure of
in relation to the COL9A3 and the IL-1ss gene polymorphisms. Eur human intervertebral disc. I. Changes in notochordal cells with age.
Spine J 15(5):613619 Tissue Cells 14:359369
Song YQ, Cheung KM, Ho DW, Poon SC, Chiba K, Kawaguchi Y, Urban JP (2002) The role of the physicochemical environment in deter-
Hirose Y, Alini M, Grad S, Yee AF, Leong JC, Luk KD, Yip SP, mining disc cell behaviour. Biochem Soc Trans 30(Pt 6):858864
Karppinen J, Cheah KS, Sham P, Ikegawa S, Chan D (2008) Urban JP, Smith S, Fairbank JC (2004) Nutrition of the intervertebral
Association of the asporin D14 allele with lumbar-disc degeneration disc. Spine 29(23):27002709
in Asians. Am J Hum Genet 82(3):744747. doi:10.1016/j. Videman T, Leppavuori J, Kaprio J, Battie MC, Gibbons LE, Peltonen L,
ajhg.2007.12.017 Koskenvuo M (1998) Intragenic polymorphisms of the vitamin D
Stoyanov JV, Gantenbein-Ritter B, Bertolo A, Aebli N, Baur M, receptor gene associated with intervertebral disc degeneration.
Alini M, Grad S (2011) Role of hypoxia and growth and differentia- Spine 23(23):24772485
tion factor-5 on differentiation of human mesenchymal stem cells Virtanen IM, Song YQ, Cheung KM, Ala-Kokko L, Karppinen J,
towards intervertebral nucleus pulposus-like cells. Eur Cell Mater Ho DW, Luk KD, Yip SP, Leong JC, Cheah KS, Sham P, Chan D
21:533547 (2007) Phenotypic and population differences in the association
Studer RK, Gilbertson LG, Georgescu H, Sowa G, Vo N, Kang JD between CILP and lumbar disc disease. J Med Genet 44(4):
(2008) p38 MAPK inhibition modulates rabbit nucleus pulposus 285288. doi:10.1136/jmg.2006.047076, jmg.2006.047076 [pii]
cell response to IL-1. J Orthop Res. doi:10.1002/jor.20604 Wallach CJ, Sobajima S, Watanabe Y, Kim JS, Georgescu HI, Robbins P,
Sugimura T, Kato F, Mimatsu K, Takenaka O, Iwata H (1996) Gilbertson LG, Kang JD (2003) Gene transfer of the catabolic
Experimental chemonucleolysis with chondroitinase ABC in mon- inhibitor TIMP-1 increases measured proteoglycans in cells from
keys. Spine 21(2):161165 degenerated human intervertebral discs. Spine 28(20):
Suzuki M, Inoue G, Gemba T, Watanabe T, Ito T, Koshi T, Yamauchi K, 23312337
Yamashita M, Orita S, Eguchi Y, Ochiai N, Kishida S, Takaso M, Walsh AJ, Bradford DS, Lotz JC (2004) In vivo growth factor treatment
Aoki Y, Takahashi K, Ohtori S (2009) Nuclear factor-kappa B decoy of degenerated intervertebral discs. Spine 29(2):156163
suppresses nerve injury and improves mechanical allodynia and Wei A, Williams LA, Bhargav D, Shen B, Kishen T, Duffy N, Diwan AD
thermal hyperalgesia in a rat lumbar disc herniation model. Eur (2009) BMP13 prevents the effects of annular injury in an ovine
Spine J. doi:10.1007/s00586-009-0940-x model. Int J Biol Sci 5(5):388396
Takahashi H, Suguro T, Okazima Y, Motegi M, Okada Y, Kakiuchi T Weibrich G, Kleis WK, Hafner G, Hitzler WE (2002) Growth factor
(1996a) Inflammatory cytokines in the herniated disc of the lumbar levels in platelet-rich plasma and correlations with donor age, sex,
spine. Spine 21(2):218224 and platelet count. J Craniomaxillofac Surg 30(2):97102
Takahashi T, Kurihara H, Nakajima S, Kato T, Matsuzaka S, Sekiguchi T, Weiler C, Nerlich AG, Bachmeier BE, Boos N (2005) Expression and
Onaya M, Miyauchi S, Mizuno S, Horie K, Fujita Y, Hirose T distribution of tumor necrosis factor alpha in human lumbar
(1996b) Chemonucleolytic effects of chondroitinase ABC on intervertebral discs: a study in surgical specimen and autopsy
normal rabbit intervertebral discs. Course of action up to 10 days controls. Spine 30(1):4453; discussion 54
postinjection and minimum effective dose. Spine 21(21): Yamada K, Tanabe S, Ueno H, Oinuma A, Takahashi T, Miyauchi S,
24052411 Shigeno S, Hirose T, Miyahara K, Sato M (2001) Investigation of
the short-term effect of chemonucleolysis with chondroitinase ABC. Yurube T, Nishida K, Suzuki T, Kaneyama S, Zhang Z, Kakutani K,
J Vet Med Sci 63(5):521525 Maeno K, Takada T, Fujii M, Kurosaka M, Doita M (2010) Matrix
Yoon TS, Su Kim K, Li J, Soo Park J, Akamaru T, Elmer WA, metalloproteinase (MMP)-3 gene up-regulation in a rat tail com-
Hutton WC (2003) The effect of bone morphogenetic protein-2 on pression loading-induced disc degeneration model. J Orthop Res
rat intervertebral disc cells in vitro. Spine 28(16):17731780 28(8):10261032. doi:10.1002/jor.21116
Tissue Engineering
of the Intervertebral Disc 26
Rita Kandel, Paul Santerre, Eric Massicotte,
and Mark Hurtig
Contents 26.10 Summary of Critical Concepts Discussed

in the Chapter ................................................................ 428
26.1 Introduction ................................................................... 417
References .................................................................................... 428
26.2 What Is the Goal for Cell-Based Therapies
and Tissue Engineering?............................................... 418
26.3 Intervertebral Disc Composition 26.1 Introduction
and Architecture: What Will Bioengineered
Tissues Have to Recapitulate?...................................... 418
In an autopsy study, 97 % of individuals 50 years or older had
26.3.1 Annulus Fibrosus ............................................................ 418
26.3.2 Nucleus Pulposus ............................................................ 419 histological evidence of intervertebral disc degeneration, a
26.3.3 Cartilage Endplate ........................................................... 419 disease process that involves the annulus fibrosus, nucleus
26.4 The Role of Tissue Architecture in Disc Function ..... 419 pulposus, and cartilaginous endplate (Miller et al. 1988). The
back pain that can develop as a result of this disease has a
26.5 Tissue-Engineering Components ................................. 419
26.5.1 Cells ................................................................................ 420
lifetime prevalence of up to 80 % (Manchikanti et al. 2009;
26.5.2 Scaffolds.......................................................................... 421 Takatalo et al. 2011, 2012). Approximately 1 in 50 Canadians
26.5.3 Growth Factors ................................................................ 422 becomes disabled by back pain which is responsible for 40 %
26.6 Tissue-Engineering Approaches .................................. 422 of all workplace absences (Iron et al. 2004; Lee 1994; Rapoport
26.6.1 Nucleus Pulposus Tissue Engineering ............................ 422 et al. 2004). Although rarely life threatening, the annual total
26.6.2 Annulus Fibrosus Tissue Engineering ............................ 423 costs in 2002 in the USA as a result of back pain were over
26.7 Engineering the Cartilage Endplate ............................ 425 $100 billion (Asche et al. 2007; Dagenais et al. 2008; Iron
et al. 2004; Lee 1994) (for a detailed discussion, see Chap. 9).
26.8 Engineering the Intervertebral Disc ............................ 425
It has been estimated that in the USA alone, there are up to
26.9 Additional Challenges for Tissue Engineering four million adults with chronic back pain who have failed
a Functional Intervertebral Disc.................................. 427
conservative therapy (Masuda and Lotz 2010) and although
there are a number of surgical options for these individuals,
they all have limitations (Chou et al. 2009; Kishen and Diwan
2010; Raj 2008). Discectomy which relieves pain (325,000
operations performed in 2004 in the USA) (2008) does not
R. Kandel, MD (*) restore disc height or its original load-bearing capacity (Putzier
Department of Pathology and Laboratory Medicine, et al. 2005). Spinal fusion (375,000 surgeries in 2004 in the
Mount Sinai Hospital, University of Toronto,
USA) (2008) is another commonly performed treatment; as
Toronto, ON, Canada
has been discussed earlier, this is not always successful, often
P. Santerre, PhD
leading to pseudoarthrosis and limited flexibility (Kishen and
Institute of Biomaterials and Biomedical Engineering,
University of Toronto, Toronto, ON, Canada Diwan 2010; Mirza and Deyo 2007). Although controversial,
some studies suggest that this type of treatment may induce
E. Massicotte, MD
Department of Neurosurgery, Toronto Western Hospital, degenerative changes in adjacent vertebrae (Huang et al. 2006;
Toronto, ON, Canada Javedan and Dickman 1999; Kim and Branch 2006; Lee et al.
Department of Surgery, University of Toronto, Toronto, ON, Canada 2012a, b; Schulte et al. 2007). Disc replacement with a motion-
preserving prosthesis, such as partial or total disc replacement,
M. Hurtig, DVM
Department of Clinical Studies, Ontario Veterinary College, is another option (Fekete and Porchet 2010; Kishen and Diwan
University of Guelph, Guelph, ON, Canada 2010; Shim et al. 2007; So et al. 2007; Vernengo et al. 2008).

418 R. Kandel et al.
However, there are numerous clinical contraindications to this

treatment (Fekete and Porchet 2010; Shim et al. 2007), and Box 26.1 What Is Regenerative Medicine?
these replacements do not fully restore kinematics (Fekete and Tissue engineering (regenerative medicine) is an inter-
Porchet 2010). This topic is discussed in considerable detail in disciplinary field that applies the principles of engi-
Chap. 14. As well, disc prostheses can generate wear debris neering (materials and biomedical engineering) and
(van Ooij et al. 2007) and the reaction that this incites has the life sciences (genetics, cell and molecular biology) to
potential to be catastrophic given its location near major ves- the development of biological substitutes that can
sels and nerves. The long-term exposure to ions, such as cobalt restore, maintain, or enhance tissue function that has
and chromium, which can be detected in the serum after disc been damaged by congenital abnormalities, disease,
replacement is unknown (Zeh et al. 2009). Given all these age, or trauma. It involves any combination of cells,
issues, there has been great interest in developing new thera- materials, biomolecules, and/or mechanics (Langer
peutic options such as biological approaches for the treatment and Vacanti 1993). First recognized as a field in the
of chronic symptomatic disc disease (Bron et al. 2009; Kandel late 1980s, it has grown to incorporate the advances in
et al. 2008; Masuda and Lotz 2010; OHalloran and Pandit basic science to enhance biological repair. Tissue-
2007; Richardson et al. 2007). Cell-based therapies that regen- engineered products being used clinically now include
erate the disc avoid the functional impact on the disc and/or skin and bladder and cartilage repair products are in
adjacent tissues as well as the consequences of metal fatigue clinical trials. Bioengineered tissues have other uses.
and the reaction to wear debris (Guyer et al. 2011; Tumialan For example, they can be used to investigate the mech-
and Gluf 2011). Importantly the regenerated tissue can remodel anisms regulating tissue formation or for preclinical
and respond to load in a way synthetic prostheses or fused drug testing.
discs cannot.
success the goal should be repair and/or replacement of all the

26.2 What Is the Goal for Cell-Based damaged tissues of the disc.
Therapies and Tissue Engineering?
The intervertebral disc is a specialized structure consisting of 26.3 Intervertebral Disc Composition and
interdependent tissues: the annulus fibrosus which surrounds Architecture: What Will Bioengineered
the centrally placed nucleus pulposus and is integrated into the Tissues Have to Recapitulate?
cartilage endplate (Hukins 1994; Simon 1994). The outer annu-
lus fibrosus is responsible for withstanding circumferential ten- 26.3.1 Annulus Fibrosus
sile forces while the nucleus pulposus and inner annulus
fibrosus resist compressive forces (Hukins 1994; Simon 1994). While considerable information is provided elsewhere in the
Together these tissues can handle more load than each tissue book concerning the detailed structure of the annulus, it is
alone, stressing the importance of properly integrated struc- worthwhile reminding the reader that the annulus is a very
tures (Bogduk 1997). The intact cartilage endplate is also nec- well-engineered tissue and exquisitely adapted to its physio-
essary. This tissue contributes to the maintenance of nucleus logical role. It surrounds the nucleus pulposus and has a cross-
pulposus cell viability, absorbs the water that extrudes from the ply laminate structure consisting of between 10 and 25 lamellae,
nucleus pulposus during loading, and prevents protrusion of each composed of collagen fibers oriented parallel to each other
the nucleus into the adjacent vertebral body (Benneker et al. and about 65 from the vertical (Bron et al. 2009; Marchand
2005; Grunhagen et al. 2006; Horner and Urban 2001; Moore and Ahmed 1990). The direction of the inclination alternates
2006; Roberts et al. 1996; Shirazi-Adl et al. 2010). The impor- with each layer such that the fibers in one lamella are 65 to the
tance of a properly functioning cartilaginous endplate (Miller right, while in the next lamella they are 65 to the left, so every
et al. 1988) is demonstrated by the fact that animal models of second lamella has the same orientation. Lamella, which can
disc degeneration are created by damaging the endplate (Cinotti be discontinuous, varies in thickness depending on the location
et al. 2005; Holm et al. 2004) or by endplate calcification in the disc and ranges from 100 to 600 mm (Marchand and
(Gruber et al. 2007; Peng et al. 2001; Sahlman et al. 2001). A Ahmed 1990). They have a unique insertion into the vertebral
recent population study showed that the presence of Schmorls body and this differs somewhat between the outer and inner
nodes correlated with disc degeneration and low back pain annulus (Nosikova et al. 2012). The collagen concentration is
(Takatalo et al. 2011, 2012). The organization of the interverte- highest in the outer annulus and gradually decreases with pro-
bral disc allows the disc to rotate, flex, and resist tensile and gression towards the nucleus (Eyre 1979). The outer annulus
shear forces and is critical for proper disc function, especially, fibrosus is composed predominately of collagen I with increas-
as the disc under certain conditions has to withstand up to ten ing amounts of collagen II present within the inner annulus. It
times body weight. Thus, it would appear that for long-term also contains elastin both within the lamellae and between
26 Tissue Engineering of the Intervertebral Disc 419
lamellae (Yu et al. 2002, 2005, 2007), a small amount of pro- predominately of water, proteoglycans, and collagen, but
teoglycans of which the most abundant is aggrecan (Inerot and differs from that tissue in organization and amounts of these
Axelsson 1991) but others such as decorin, biglycan, versican, molecules (Gruber et al. 2007; Moore 2006). Vascular chan-
and fibromodulin are also present (Gotz et al. 1997). The pro- nels penetrate the cartilaginous endplate (Miller et al. 1988),
teoglycans have a differential distribution which is opposite to but after growth has ceased, these are obliterated and nutri-
that of collagen so that the inner annulus fibrosus lamellae are ents and oxygen must diffuse from blood vessels in the ver-
separated by more proteoglycan-rich matrix than those in the tebral body through the tissue (Benneker et al. 2005;
outer annulus (Bron et al. 2009). The inner annulus fibrosus Rajasekaran et al. 2004). In vitro studies suggest that the
merges almost imperceptibly with the nucleus pulposus. The chondrocytes produce a factor that inhibits TNFa produc-
difference in composition between the inner and outer aspects tion by nucleus pulposus cells, and thus it may play a role in
of the annulus fibrosus may reflect the different functions of preventing many of the changes that are characteristic of
these regions (Bron et al. 2009; Hsieh and Twomey 2010). aging (and degeneration) (Arana et al. 2010). The multiple
Translamellar bridges are seen in the outer annulus and recent functions of the cartilage layer in terms of nutrition and oxy-
work by Elliot et al. suggested that these represent areas of gen diffusion as well as protection of the nucleus stresses
blood vessel regression (Melrose et al. 2008; Schollum et al. the necessity for an intact cartilage endplate when engineer-
2010; Schollum et al. 2009; Smith and Elliott 2011). It is not ing a substitute tissue.
known whether they play a role in the biomechanical function-
ing of the disc. The interface of the annulus fibrosus with the
vertebral body is complex, as a portion of the collagen fibers 26.4 The Role of Tissue Architecture
pass through the cartilage endplate into the calcified zone and in Disc Function
the remainder sweep laterally to merge with the periosteum
(Nosikova et al. 2012). Aside from addressing the needs of the molecular architec-
ture briefly described above, the engineered disc will also
need to replace many if not all of the functional requirements
26.3.2 Nucleus Pulposus of the spine. As a major component of this functional unit,
the annulus fibrosus can be viewed as an anisotropic, non-
Probably the most important tissue to be engineered in the disc linear, viscoelastic tissue (Hsieh and Twomey 2010). As dis-
is the nucleus pulposus. It consists of proteoglycans within a cussed in detail in Chaps. 2 and 7, compressive loading
loose network of collagen that does not demonstrate the same causes disc narrowing and outward bulging, placing axial
level of organization as the annulus fibrosus (Bibby et al. 2001; compressive forces as well as tensile (biaxial) strains along
Chan et al. 2011). Proteoglycans comprise up to 65 % of the the circumference of the annulus fibrosus (Nerurkar et al.
dry weight of the nucleus. Aggrecan is the major proteoglycan 2010a, b). Part of the compressive force is transmitted by
and responsible for binding water thereby enabling the nucleus the nucleus from one vertebral body to the next, decreasing
to resist compressive loads (Chan et al. 2011). Other proteo- the load borne by the annulus. Sudden increases in compres-
glycans such as versican, decorin, and fibromodulin are also sive stress are propagated through the vertebral column;
present (Melrose et al. 2001; Singh et al. 2009; Smith et al. because of interpositions of the intervertebral disc, changes
2009). The nucleus contains predominately collagen II and in compressive stress are converted into tensile stain towards
lesser amounts of collagens III, VI, IX, and XI (Chan et al. the outer annulus (Markolf and Morris 1974). This region
2011). The nucleus contains notochordal cells but the latter experiences tensile strain with some compression in the
cell type either disappears or its morphology changes with age inner aspect; the inner annulus fibrosus is compressed as it
(for a detailed discussion of this topic, see Chaps. 3 and 21) resists the circumferential expansion of the nucleus pulpo-
(Choi et al. 2008; McCann et al. 2012; Risbud and Shapiro sus. With compression, collagen crimp angles change from
2011; Weiler et al. 2010). The notochordal cells likely contrib- outer to inner annulus fibrosus, resulting in both depth-
ute to disc function and/or maintenance, but how this is dependent and linear region stiffness (Holzapfel et al. 2005).
achieved is not entirely known at this time (Abbott et al. 2012; It has also been suggested that inter-collagen fiber shear is
Cappello et al. 2006; Risbud et al. 2010). important in strain redistribution (Desrochers and Duncan
2010) which may affect nutrient diffusion (Ambard and
Cherblanc 2009). These biomechanical events are relevant
26.3.3 Cartilage Endplate to the healthy disc; however, with degeneration of the
nucleus pulposus, reorientation of the annulus fibrosus is
The cartilage endplate is a thin (about 1 mm) layer of hya- compromised and tensile forces are transferred to the annu-
line cartilage, rich in collagen II. It is integrated with the lus fibrosus (Adams et al. 1996, 2009; Guerin and Elliott
vertebral body, nucleus pulposus, and annulus fibrosus 2006). Finite element modeling suggests that strains in the
(Moore 2006). Like articular cartilage, the plate consists inner annulus increase with degeneration and that with
progressive disease this strain is transferred to the vertebral

endplate (Schmidt et al. 2009). Thus, intact and properly Box 26.2 Requirements of Cells for Regenerative
oriented and interfaced annulus fibrosus lamellae are neces- Medicine
sary for proper responses to loading (Adams et al. 2009; One of the factors preventing clinical application of
Guerin and Elliott 2006). Furthermore, it appears that annu- regenerative medicine is a suitable source of cells to
lus fibrosus fiber tension not only limits axial rotation generate the tissue. Clearly primary cells are not an
(Krismer et al. 1996) but also limits susceptibility to degen- option. Stem cells will likely be critical to overcome
eration (Lotz et al. 2008). These biomechanical principles this limitation. However, the mechanisms regulating
are critical architectural features to be borne in mind in differentiation to the appropriate cell types have yet to
engineering a new intervertebral disc. be fully elucidated. Delineation of the conditions to
accomplish this in vitro is necessary, and factors such
as the culture conditions, choice of culture media,
26.5 Tissue-Engineering Components presence of biomolecules, appropriate biomaterial,
proper oxygenation status, and mechanical stimulation
Diseases that afflict articulating joints in the appendicular are all currently under investigation.
skeleton involve cartilage and in some settings bone. In the
disc, three tissues, the nucleus pulposus, the annulus fibrosus,
and the cartilaginous endplate, are all impacted by the dis- have to be expanded in culture to get sufficient numbers
ease process (Miller et al. 1988). From this perspective, in (Gruber et al. 1997; Maroudas et al. 1975). In the healthy
order to achieve proper function, all of these tissues may nucleus, the cell density has been reported to be 4,000/mm2;
require repair/replacement. Having to recapitulate three tis- however, it may even been even lower since, with age, a pro-
sues, each with very different molecular architecture and portion of the cells become senescent (Roberts et al. 2006).
function adds a level of complexity to successful biological Not only do these cells have a slower proliferative rate, they
disc repair. also have an altered phenotype (Le Maitre et al. 2007). A
The strategies for tissue engineering of disc tissues are sufficient number of cells may be obtained from herniated
similar to that of other tissues, namely, to use a combination disc tissue; this approach may not be straightforward since
of cells, scaffolds, growth factors, and/or mechanical signals. these cells may be dedifferentiated with limited ability to
The choice is dictated by the tissue(s) under repair. Given regenerate tissue in vitro compared to cells obtained from the
that the disc is always loaded, the repair/replacement tissue nucleus pulposus compartment (Hegewald et al. 2011a).
will likely experience mechanical forces and signals imme- For the reasons stated above, it may be necessary to uti-
diately upon implantation. lize allogeneic or xenogeneic sources; these cells carry the
risk for immune reaction and disease transmission. If cells
from the nucleus pulposus are to be used for tissue-engineer-
26.5.1 Cells ing purposes, then genetic problems must be accommodated.
As discussed in Chap. 10, polymorphism (Trp2 allele) in the
One of the major issues limiting the clinical application of COL9A2 gene encoding the alpha2 chain of collagen IX can
biological disc repair/replacement therapies is the identification predispose an individual to disc degeneration (Richardson
of a suitable source of cells. Cells could be obtained from a and Hoyland 2008; Jim et al. 2005). Non-degenerated discs
number of tissues, such as the disc itself, cartilage, or bone from these individuals were found to be mechanically infe-
marrow. It is not yet clear whether the cells used for nucleus rior to non-degenerated normal discs (Aladin et al. 2007).
pulposus repair must include notochordal cells. In vitro stud- This finding suggests that the use of genetically altered pri-
ies have shown that notochord cells produce factors, such as mary cells isolated from an otherwise healthy intervertebral
CTGF, which enhances matrix production by nucleus pulpo- disc may not be appropriate for regeneration of normal tis-
sus cells (Erwin 2008; Purmessur et al. 2011). Some investi- sue. Use of nucleus pulposus cells also may be limited as
gations indicate that notochordal cells are critical for the they have been shown to lose their regenerative capacity with
maintenance of nucleus tissue (Risbud et al. 2010; Risbud and age (Kandel et al. 2007). A recent study demonstrated that
Shapiro 2011; Shapiro and Risbud 2010). If these cells are matrix retention by nucleus pulposus cells obtained from
required, obtaining sufficient numbers from any of the sources adolescent cows is impaired and that compared to those
listed above may be problematic, although recent studies in obtained from younger animals (calves), matrix gene expres-
our laboratory suggest that under the appropriate culture con- sion is decreased (Kandel et al. 2007). If similar alterations
ditions, notochord cell numbers in nucleus pulposus tissue occur in human cells, it may not be appropriate to use nucleus
can be increased in vitro (Kandel 2011, unpublished data). pulposus cells from older patients unless conditions are
Although nucleus pulposus cells would be ideal to use for identified that prime the cells and encourage reversion to a
tissue engineering, they are limited in number and would more juvenile geno/phenotype.
Many of the issues limiting the use of differentiated cells developing new approaches to use hESC as a source of
may be overcome using autologous stem cells. chondrocytes is an area of intense investigation, little work
Developmentally, the annulus cells are derived from meso- has been done in this regard with disc cells. To date, deriv-
derm, whereas nucleus pulposus cells are derived from the ing chondrocytes from embryonic stem cells that can form
notochord (Chan et al. 2011; Choi et al. 2008; McCann et al. articular cartilage tissue has not been fully accomplished,
2012). Thus, these cells could be obtained from a number of suggesting that identifying conditions that induce differen-
different tissue types, the most common being bone marrow tiation of hESC to disc cells in cell culture is a long way off.
(marrow stromal cells, MSCs) or adipose tissue (adipose- Interestingly, in one study mouse ESC were differentiated
derived stem cells) (Ahmed et al. 2007; Bieback et al. 2008; in vitro to chondrocyte-type cells; these cells transdifferen-
Zuk et al. 2002). MSCs have the capability to commit to sev- tiated to notochord cells following implantation into rabbit
eral different lineages including cartilage, bone, or adipose discs (Sheikh et al. 2009). These results suggest that ESC
tissue in vitro, and it is speculated that they can also give rise may be a source of notochordal cells, but this topic requires
to disc cells although the evidence for this is limited. further study. Another issue associated with the use of ESC
Generating nucleus pulposus or annulus fibrosus cells from cells is that they can give rise to teratomas an unacceptable
MSCs has been stymied by our lack of knowledge of tissue complication for a treatment of a benign disease (Schriebl
phenotype such as the characteristic molecular profile of et al. 2012). To circumvent this problem, it will be necessary
these cells, a topic considered in detail in Chap. 11. to ensure that there are no residual hESCs after induction of
Despite an ever-increasing number of studies of stem cells differentiation (Schriebl et al. 2012). Consideration will
and cells of the nucleus pulposus, difficulties still exist in also have to be given to the ethical concerns related to the
confirming that a stem or stromal cell has differentiated along collection of these cells which could limit their application
the disc lineage. For example, Steck et al. suggested that clinically.
MSCs can adopt a gene expression profile resembling native Inducible progenitor cells (iPSC) hold much promise for
disc cells but the molecules they examined are also present in disc repair and obviate the ethical issues related to the use of
the cartilage (Steck et al. 2005). It has been suggested that embryonic stem cells. iPSC can be generated by the transfec-
co-culturing MSC with disc or notochordal cells may be tion of the four transcription factors Oct3/4, Sox2, c-Myc,
another way to effect conversion to a chondrogenic/disco- and Klf4 (Yamanaka factors) into somatic cells such as
genic phenotype (Korecki et al. 2010; Le Visage et al. 2006; human fibroblasts (Maherali et al. 2008; Takahashi et al.
Purmessur et al. 2011; Richardson et al. 2006; Strassburg 2007; Takahashi and Yamanaka 2006). The transduced cells
et al. 2010; Vadala et al. 2008), possibly via paracrine effects return to an undifferentiated, pluripotent state and resemble
or cell-cell interactions (Vadala et al. 2008). Given that MSC ES cells (Takahashi et al. 2007). A new technology using
can differentiate to chondrocytes, ensuring that the MSC virus-independent, transposon-mediated reprogramming of
commit entirely to the nucleus pulposus phenotype has not somatic cells into iPS cells has recently been developed
yet been possible. An alternative approach is to use MSC to (Woltjen et al. 2009), eliminating one of the potential con-
secrete factors that activate and stimulate nucleus pulposus traindications to the clinical use of these cells. Successful
cell matrix production and/or accumulation and thus potenti- differentiation into cell types such as chondrocytes and mel-
ate tissue repair (Doorn et al. 2012; Miyamoto et al. 2010; anocytes suggests that iPSC have the potential to be an
Strassburg et al. 2010). However, there are further issues appropriate cell source for tissue engineering of the disc
associated with the use of MSCs: with age they can lose their (Wei et al. 2012; Yang et al. 2011). However, recent studies
regenerative capability (Choumerianou et al. 2010; Erickson have identified genetic and methylation changes in iPSCs
et al. 2011). Extensive culture and population doublings may and issues related to immunogenicity, all of which raise
promote senescence, and cell karyotype and gene expression questions as to their suitability for tissue engineering
can become altered (Wang et al. 2005; Wilson et al. 2010). (Barrilleaux and Knoepfler 2011; Lister et al. 2011; Zhao
For example, a subpopulation of cells in MSC culture was et al. 2011). Clearly, there is much work to be done before
noted to appear morphologically distinct from typical human these cells can be used in clinical trials (Nakagawa et al.
MSCs (Wang et al. 2005). These cells showed a high level of 2008; Yamanaka 2007).
telomerase activity compared with typical MSCs and formed Chondrocytes are another source of cells to use for nucleus
tumors when transplanted into NOD/SCID mice. pulposus repair. A study by Acosta et al. (2011) showed that
Embryonic stem cells (hESC), another potential source juvenile chondrocytes formed cartilage tissue when implanted
of cells, are derived from the inner cell mass of the embryo into porcine discs whereas MSC did not. However, the
(Munoz et al. 2008). ESC can be readily maintained as efficacy of these cells in the long term is not known. As there
undifferentiated cells under defined conditions, providing is increasing evidence that nucleus pulposus cells are differ-
an unlimited supply of pluripotent stem cells. They are ent than chondrocytes, even though they produce many of
capable of differentiating into all three germ layers of the the same molecules, perhaps they should not be used for
embryo and hence all cell types of the human body. While nucleus repair (Mwale et al. 2004).
There has been little investigation into which cells are suit- which negatively affects cell matrix synthesis (Razaq et al.
able to use for annulus fibrosus repair. MSC would seem to be 2003). One way to overcome some of the problems linked to
an appropriate source of cells as the annulus fibrosus develop- cell source is to functionalize acellular scaffolds with biomol-
mentally arises from the mesoderm. Annulus cells themselves ecules or genes. For example, in a study of intervertebral disc
could be used as when grown in monolayer culture, annulus degeneration in rabbits, injection of platelet-rich plasma
cells maintain their phenotype for up to two passages (Chou (containing growth factors) encapsulated in gelatin micro-
et al. 2006). These cells would either have to be obtained spheres into the nucleus pulposus slowed down the disease
from allogeneic or xenogeneic tissues similar to nucleus pul- process compared to untreated animals (Nagae et al. 2007).
posus cells unless the entire disc is being replaced.
Another potential problem with growing the cells in vitro
prior to implantation is their contact with animal products 26.5.3 Growth Factors
such as fetal calf serum (Tekkatte et al. 2011). Bovine pro-
teins incorporated into the cell membrane could induce an The role of growth factors in maintaining the disc phenotype
immune response. There is also the risk of infection follow- during cell expansion in vitro and for tissue formation has not
ing implantation (Harrison et al. 2000). Interest in develop- been studied extensively. Several studies have shown that
ing alternatives to animal serum to avoid these problems, growth factors, for example, OP-1 (BMP-7) (Masuda and
such as growing cells in autologous serum (Harrison et al. Lotz 2010), enhance matrix production by nucleus pulposus
2000; Lange et al. 2007; Shahdadfar et al. 2005; Tekkatte and annulus fibrosus cells. Treatment of MSCs with growth
et al. 2011) or human platelet plasma, is being investigated factors may influence differentiation: one example is TGFb3
(Bieback et al. 2009; Schallmoser et al. 2007). treatment of MSCs maintained on a photo-cross-linked car-
boxymethyl cellulose hydrogel scaffold (Gupta et al. 2011).
Whether the treated cells commit to a nucleus pulposus cell
26.5.2 Scaffolds lineage or chondrocyte phenotype has yet to be determined. It
has been shown that treatment with a growth factor, such as
For disc tissue engineering, roles for scaffolds include reten- GDF5, or transduction with growth factor genes can enhance
tion of cells in a desired location, provision of mechanical disc repair in vivo. These results lend strong support to the
properties (sufficient for weight bearing while tissue is grow- notion that growth factors have a major role to play in biologi-
ing and maturing), and/or delivery of biochemical cues to the cal repair (Masuda 2008; Zhang et al. 2008). Further discus-
tissue as it is developing or to guide tissue ingrowth (Ikada sion of this topic is presented in Chaps. 23, 24, and 25.
2006). Scaffold requirements are such that the material must
be biocompatible and biodegradable at a rate that mirrors tis-
sue regeneration. For the nucleus pulposus, a liquid-based 26.6 Tissue-Engineering Approaches
scaffold would likely be preferable, whereas for the annulus
fibrosus, a fiber-type scaffold would be better. The choice of 26.6.1 Nucleus Pulposus Tissue Engineering
scaffold is critical as it can affect the tissue that develops.
Scaffold fiber diameter and stiffness have been shown to Repair of the nucleus pulposus by tissue engineering has
influence cell phenotype, function, proliferation, and orien- received more attention than the annulus fibrosus for two
tation (Hadjipanayi et al. 2009; Hsia et al. 2011; reasons: the nucleus is a relatively less complex tissue and
Saino et al. 2011). These responses could be utilized in the early disc degeneration can occur in the nucleus before being
design of disc scaffolds that enhance cell differentiation. evident in the annulus. A number of different approaches
Adhesion molecules used to coat the scaffolds can also have been taken to engineer the nucleus pulposus such as
affect cell function (Attia et al. 2010). For example, as shown injections of cells (nucleus pulposus cells, chondrocytes, or
by Attia et al. (2010), polyurethane scaffolds coated with mesenchymal stromal cells or stem cells) alone or in a scaf-
fibronectin result in an annulus fibrosus-like structure, cells fold or implantation of nucleus pulposus tissue formed
that are aligned and elongated similar to native tissue, and in vitro prior to implantation.
importantly the newly synthesized collagen is also properly
oriented parallel to the cell. In contrast if the scaffold is 26.6.1.1 Cell Therapy
coated with collagen I, the cells assume a polygonal shape A strategy for early disease repair is to inject cells into the disc
and collagen production is delayed and poorly oriented. In at a time when only the nucleus pulposus is damaged and the
addition, the scaffold can influence the recipient tissue reac- annulus fibrosus and cartilaginous endplate (Miller et al. 1988)
tion after implantation. For example, breakdown products are intact. The implanted cells are postulated to work by one
could induce fibrosis or lower the local pH; this is seen when of two ways: either by synthesizing matrix or stimulating the
polylactic/glycolic acid scaffolds degrade and release acid endogenous remaining nucleus pulposus cells to synthesize
matrix. There is increasing evidence that cell therapy for (see reviews) (OHalloran and Pandit 2007; Yang and Li
nucleus pulposus repair may be an appropriate approach. 2009). In summary, most have been shown to support cell
Injection of autologous nucleus pulposus cells or stem cells growth and, in some cases, tissue formation. Agarose and
into the disc has been shown to delay degeneration in both small alginate have been utilized, but the inability of agarose to
and large (dog and monkey) animal models (Allon et al. 2010; degrade and the impurities in alginate may limit clinical
Crevensten et al. 2004; Ganey et al. 2003; Gruber et al. 2002; application (Bron et al. 2011; Chou and Nicoll 2009). Other
Hiyama et al. 2008; Meisel et al. 2007; Okuma et al. 2000; examples of scaffolds that have been developed for this pur-
Sakai et al. 2005; Sheikh et al. 2009; Sobajima et al. 2008). In pose include collagen I, atelocollagen type II, collagen II
a rat study, the injection of a pellet of both nucleus pulposus enriched with hyaluronan and cross-linked with polyethyl-
cells and MSC was shown to preserve disc height, suggesting ene glycol (4S-StarPEG), polylactic acid, combined thiol-
that the use of both cell types maybe advantageous (Allon modified hyaluronan and elastin-like polypeptide,
et al. 2010). The duration of these individual studies was vari- chitosan-glycerophosphate, and collagen II/hyaluronan/chon-
able; the longest was a dog study in which the implant was droitin-6-sulfate copolymer sponge which have been seeded
assessed at 1 year (Ganey et al. 2003). Although the cells used with either nucleus pulposus or MSC cells (Bowles et al. 2010;
in these trials were for the most part autologous, there were at Collin et al. 2011; Halloran et al. 2008; Hegewald et al.
least two studies that used allogeneic or xenogeneic cells. One 2011b; Huang et al. 2011; Lee et al. 2012a; Nettles et al.
was a rat study in which allogeneic MSC were used to repair 2010; Richardson et al. 2008; Sakai et al. 2005). Some of
the disc. There was no evidence of an immune reaction at 1 these constructs have been evaluated in animal models. For
month, raising the possibility that the disc does provide an example, nucleus pulposus cells seeded into collagen type II/
immunoprivileged environment and that it may be possible hyaluronan/chondroitin-6-sulfate copolymer sponge were
to use allogeneic or xenogeneic cells in this setting (Crevensten grown in culture for 1 week and then implanted into rabbit
et al. 2004). In keeping with this premise, mouse ESC which lumbar discs immediately after nucleotomy. At 6 months, the
were partially differentiated in vitro to chondrocytes were disc height was greater in rabbits that received the cell-seeded
injected into rabbit degenerated discs: at 8 weeks no immune scaffold compared to scaffold alone; comparison to the nor-
reaction was observed (Sheikh et al. 2009). Cell labeling stud- mal unoperated disc was not provided. The implanted cells
ies confirmed that the injected cells remained in the disc. Since were present and there was evidence of tissue formation
both of these studies were short term, longer-term studies are (Huang et al. 2011). However, as rabbit nucleus pulposus
required. cells are predominately notochordal, the relevance of this
Human cell therapy studies are ongoing. In one clinical study to humans is not clear. In another study, MSC were
trial cervical discs were injected with nucleus pulposus cells seeded into type I atelocollagen and implanted into rabbit
harvested from herniated disc tissue. While the 2-year results discs 2 weeks after nucleus pulposus tissue aspiration. There
were promising (Hohaus et al. 2008), the longer-term out- was evidence of disc repair as the proteoglycan content
come has yet to be reported. More recently, a pilot study was returned to 83 % of normal. Cells were shown to be neces-
performed in which ten patients received intra-discal injec- sary as the scaffold alone resulted in only 13 % of the proteo-
tions of MSCs; there was significant pain relief by 3 months glycan content of the native disc (Sakai et al. 2005). Studies
and at 12 months there was increased water content in the in larger animals have not been performed. There has been
disc even though disc height was not fully restored (Orozco increasing interest in hydrogels as they are injectable, able to
et al. 2011). swell and retain water, and thus withstand loading (Pereira
Another approach to cellular repair is to inject MSC, not et al. 2013). The plethora of scaffolds that have been gener-
into the disc, but intravascularly (Alini M, personal commu- ated would suggest that the ideal scaffold has yet to be
nication 2012). It is expected that the cells would then home identified.
to the disc obviating the need for a delivery system or annu-
lus fibrosus needle puncture, which has been shown in ani- 26.6.1.3 Tissue Formation In Vitro
mals to alter disc mechanics and in humans to possibly Nucleus pulposus tissue can be formed in the presence or
promote disc degeneration (Carragee et al. 2009; Iatridis absence of a scaffold in vitro prior to implantation. Supporting
et al. 2009; Michalek et al. 2010; Zhang et al. 2009). this approach, a study in rabbits showed that implantation of
Validation of this approach is lacking at this time, but an nucleus pulposus tissue was more effective in delaying disc
in vitro study showing that MSC can migrate into the disc is degeneration than isolated cells (Nomura et al. 2001). There
promising (Illien-Junger et al. 2012). are several advantages to developing tissue prior to implanta-
tion. As the disc is always loaded in vivo, a formed tissue
26.6.1.2 Cell-Seeded Scaffolds that is able to withstand some degree of loading after implan-
A large number of scaffolds have been developed for nucleus tation is an obvious advantage. In addition, the tissue does
pulposus repair and an overview of these have been published not have to form in a harsh environment, e.g., presence of
proinflammatory cytokines. Furthermore, this approach is not surprising that by midlife, about 50 % of individuals
facilitates formation of more than one tissue type. For exam- have annular tears (Videman and Nurminen 2004). However,
ple, the authors have developed an in vitro system to form a the healing potential is limited and it is usually replaced by a
portion of the intervertebral disc consisting of nucleus pulpo- fibrous tissue which is biomechanically inferior and hernia-
sus tissue adherent to a subjacent layer of cartilaginous tissue tion of the implant is not uncommon. Thus, once damaged, in
(representing the endplate), thus recreating the inner disc tis- order to restore its complex architecture, the annulus fibrosus
sue (Hamilton et al. 2006). requires repair and more likely replacement (Bron et al.
2009). Moreover, given the complex loading patterns of the
annulus fibrosus, it is critical that the replacement tissue is
26.6.2 Annulus Fibrosus Tissue Engineering engineered to closely match the original. Relevant to this
issue, the inability of the annulus fibrosus to regenerate also
The annulus fibrosus experiences complex loading patterns, influences the type of nucleus replacements that can be used.
for example, under compression on the side of applied bend-
ing, there is both tensile radial and compressive axial strain, 26.6.2.1 Cell Therapy
whereas the opposite side undergoes tensile axial strains Given the complexities of the engineered architecture of the
(OConnell et al. 2011). As the annulus fibrosus experiences annulus fibrosus and the continual strains it experiences, cell
residual strain even when unloaded (Michalek et al. 2012), it therapies are not suitable to use to repair this tissue.
a b
Scanning electron microscopy
Bovine AF
S S
20 m
100 m
Scaffold
Fig. 26.1 Annulus fibrosus tissue engineering. (a) Appearance of electron microscopy shown as control. (b, c) Histological appearance of
electrospun nanofibrous scaffold grossly and by scanning electron AF cell-seeded multilayer scaffold with AF tissue rich in collagen
microscopy. Bovine annulus fibrosus (AF) visualized by scanning (C trichrome stain, S scaffold)
26.6.2.2 Cell-Seeded Scaffolds

A number of biological and synthetic scaffolds have been Box 26.3 Tissue Engineering of Articular Cartilage
shown to support annulus fibrosus cell attachment and Articular cartilage is very different than the interverte-
subsequent synthesis of extracellular matrix (Chang et al. 2007; bral disc, as it is a simpler organ in that it is composed
Chou et al. 2008; Gruber et al. 2004; Helen and Gough 2008; of only one tissue containing a small number of chon-
Saad and Spector 2004; Shao and Hunter 2007; Yang et al. drocytes surrounded by a large amount of extracellular
2009). Contraction of a rat tail collagen solution around a poly- matrix rich in proteoglycans, collagen, and water.
ethylene disc resulted in aligned collagen that supported growth Although similar in some ways to the nucleus pulpo-
of oriented annulus cells (Bowles et al. 2010). However, given sus, the cells of these two tissues are very different and
the need to form multi-ply tissues with alternating angle of ori- they can respond differently to stimuli. Tissue engi-
entation (angle-ply multi-lamellated tissues), electrospinning neering of articular cartilage is further advanced than
has been gaining interest as a way to generate scaffolds suitable the disc, and the principles can be translated to engi-
for the formation of annulus fibrosus tissue. Electrospun neering a functional endplate cartilage. However, both
nanofibrous scaffolds can be generated on continually rotating cartilage and disc tissue engineering have the same
circular mandrels, so they are aligned and resemble the scale of issue which is the recreation of the soft tissue to hard
native fibrillar collagen, the major component of the annulus tissue (bone) interface. It is not yet clear whether this
fibrosus matrix (see Fig. 26.1). Several studies have shown that integration should be developed prior to implantation
annulus fibrosus cells will align on these scaffolds and produce or whether it can be induced to develop in vivo, but it
a properly oriented matrix (Koepsell et al. 2011; Nerurkar et al. is a necessary goal if tissue-engineering approaches
2009; Vadala et al. 2012). A variety of biomaterials can be elec- are to be successful.
trospun including polyurethane, polycaprolactone, poly(l-
lactide), and poly-e-caprolactone. As shown by Nerurkar et al.,
these cell-seeded scaffolds can form multi-lamellated struc- approach. A study in rabbits demonstrated that insertion of
tures and will be discussed further below (Nerurkar et al. 2009, nucleus pulposus tissue was more effective than nucleus pul-
2010a). How to develop annulus fibrosus tissue with inner and posus cells in delaying disease (Nomura et al. 2001).
outer zones has yet to be determined. However, implantation of cells alone, or even nucleus pulpo-
sus tissue, cannot reverse changes such as cartilaginous end-
plate calcification, which may be the cause of the disease
26.7 Engineering the Cartilage Endplate process, nor will it be able to restore structural functionality
when all three tissues are degenerated. In these settings the
The cartilage endplate is critical to disc function and yet it has optimal approach would be to replace the entire disc.
not been intensively studied. A recent study by Broom et al. Intervertebral disc and vertebral body graft transplants
demonstrated some of the features of collagen organization in have been used successfully in animal models (Frick et al.
this tissue (Wade et al. 2012). The authors have developed an 1994; Luk et al. 1997; Olson et al. 1991) and more recently in
in vitro system to form a layer of cartilaginous tissue (repre- humans (Ruan et al. 2007) and serve as proof of concept for
senting the cartilaginous endplate) which is integrated into the biological disc replacement. Autografts are not an option in
top surface of a biodegradable porous bone substitute material humans, and the use of allografts is limited by tissue avail-
(calcium polyphosphate) (Waldman et al. 2002). The intent of ability they also generate their own problems, such as their
the bone substitute is to allow for fixation of the construct into ability to transmit disease. Hence, research groups have been
the vertebral body, an approach that has been used success- developing methods to create a functional spinal unit using
fully in cartilage implant studies (Kandel et al. 2006). tissue-engineering principles. Mizuno et al. generated the first
Generating nucleus pulposus tissue on this layer of cartilage, model of an intervertebral disc using nude mice as the biore-
a composite structure is generated that exhibits enhanced actor. They formed a disclike structure by generating a mesh
interfacial shear strength to the bone substitute compared to of polyglycolic/polylactic acid polymers seeded with annulus
nucleus pulposus tissue only (Hamilton et al. 2006). fibrosus cells which was placed around an alginate hydrogel
seeded with nucleus pulposus cells (Mizuno et al. 2006). This
composite formed tissue only when grown subcutaneously in
26.8 Engineering the Intervertebral Disc athymic mice. After 16 weeks, the biochemical properties
approximated the mouse disc but the mechanical properties
Although cell therapy studies have provided promising evi- (Youngs modulae) were inferior (Mizuno et al. 2006).
dence in support of this approach in restoring disc function, Since the inability to form tissues in vitro limited fur-
the cellular and architectural aspects of the repair tissue ther study, Bowles et al. developed another model which
formed have not been clarified. Moreover, it is likely that uses collagen gels, rather than a synthetic polymer, to
transplantation of tissue rather than cells may be a better mimic the circumferential alignment of the annulus fibrosus
(Bowles et al. 2011). In this construct, annulus-seeded col- silk scaffolds surrounding a chondrocyte-seeded porous
lagen surrounded a nucleus pulposus-seeded alginate plug. fibrin-hyaluronan hydrogel was engineered by Park et al.
The cells used in this study were xenogeneic as they were (2012). Likewise, Lazebnik et al. (2011) generated a disclike
obtained from sheep. Although this construct did not mimic composite by electrospinning circumferentially orientated
the cross-ply alignment of the native annulus fibrosus, when polycaprolactone fibers (an annulus fibrosus analog), seed-
implanted into rat caudal spines, there was integration with ing it with porcine chondrocytes which had been passaged
adjacent vertebral bodies, and disc height was restored. There three times and then gelling a chondrocyte (first passage
was evidence of tissue formation while the modulus indi- cells)-agarose solution in the center (nucleus pulposus ana-
cated that load-bearing properties were maintained. log). Cells were present and remained viable in this con-
Interestingly, there was less energy dissipation which may be struct, but no further characterization was performed. An
important for prevention of adjacent disc disease (Bowles electrospun poly(l-lactic acid) (PLLA) nanofibrous scaffold
et al. 2011). Although it could be argued that the implant was seeded with MSC was injected into the center of a MSC-
small and would not be subjected to the loads or nutritional containing slurry of hyaluronan to generate a composite
demands of human discs, the study demonstrates that a bio- structure. Nesti et al. (2008) showed that at 28 days, cells
logical disc replacement can integrate. To evaluate biological were embedded in a matrix rich in proteoglycans. The
functionality and potential for use in a clinical setting, assess- authors group has taken a somewhat different approach to
ment in a large animal will be necessary. generating a disc. An annulus fibrosus cell-seeded aligned
A number of other groups have also generated biological multi-lamellated polyurethane nanofibrous scaffold was
discs in vitro, although these are yet to be evaluated in vivo. wrapped around scaffold-free nucleus pulposus tissue which
Mauck and Elliot have generated a composite disclike struc- is integrated to the top surface of a porous bone substitute
ture of cross-ply poly-e-caprolactone aligned scaffolds cul- (see Fig. 26.2). The bone substitute will allow for fixation of
tured with mesenchymal stromal cells wrapped around an the construct into the vertebral body by bone ingrowth into
agarose plug containing MSCs (Nerurkar et al. 2010a). There the pores. In aggregate, although these studies suggest that it
was tissue formation after 6 weeks of culture which showed will be possible to engineer biological disc replacements,
some resemblance to the native intervertebral disc. A con- further investigations are still required to generate an
struct measuring 8 mm, inner diameter of 3.5 mm, and a intervertebral disc that mimics the native tissue and is shown
height of 3 mm using annulus fibrosus cell-seeded fibrous to be functional in a large animal model.
NP AF
Nucleus pulposus
Bone substitute
AF
bilayer
Fig. 26.2 Tissue-engineered
intervertebral disc tissue. (a)
Appearance of nucleus pulposus
tissue on bone substitute. (b) Bone
Appearance of intervertebral disc substitute
model composed of nucleus
pulposus surrounded by annulus
fibrosus
26.9 Additional Challenges for Tissue using finite element modeling, that the nucleus pulposus
Engineering a Functional should have a Youngs modulus of between 5 and 10 MPa
Intervertebral Disc and be able to withstand compressive stress in excess of
1.67 MPa. It has been suggested that the annulus fibrosus
A number of questions related to tissue engineering an inter- should be able to withstand strains of up to 15 % (Bass et al.
vertebral disc still need to be answered. Firstly, since the 2004). Cortes and Elliott (2012) demonstrated that the inter-
annulus fibrosus is composed of an inner and outer zone, lamellar tissue has an aggregate modulus of 10.2 3.3 kPa;
will it be necessary to recapitulate both zones or will they although nonlinearity was detected for compression less than
form spontaneously in vivo? Attempts to address this are 0.8. The contribution of the interlamellar tissue (defined as
ongoing. For example, Wan et al. (2008) has developed a extrafibrillary matrix) to the total aggregate modulus of the
biphasic construct. The outer phase of the construct was a annulus fibrosus decreased from 70 to 30 % for an applied
ring-shaped demineralized bone matrix gelatin scaffold compression of 50 % of the initial thickness. These data pro-
extracted from cortical bone (rich in collagen I to mimic the vide baseline information about what is desired from mature
outer annulus fibrosus) and an inner phase composed of disc tissue.
poly(polycaprolactone triol malate) orientated in concentric Thirdly, it is not known at what rate the regenerating tis-
sheets and seeded with chondrocytes to recapitulate the sues will acquire the desired native mechanical properties. In
inner aspect of the annulus fibrosus (contains collagen II a previous study, we demonstrated that in vitro-formed carti-
and proteoglycans). This in vitro study demonstrated that lage implanted into a focal articular joint defect in sheep
the cells on these scaffolds remain viable; no further charac- with a compressive equilibrium modulus 1 % of native tissue
terization was performed. survive showed a 30-fold improvement in mechanical prop-
Secondly, it is not known if the tissue-engineered disc tis- erties by 9 months (Kandel et al. 2006). This raises the pos-
sue will exhibit the minimal mechanical properties necessary sibility that in vitro-formed disc tissue will remodel similarly
for survival following implantation. We have some rudimen- in response to the mechanical environment into which it is
tary appreciation of the requirements for mature tissue, but placed as is seen in the maturation and growth of native disc
little about the implanted tissue. Yao et al. (2006) predicted, tissue that occurs with aging (Fig. 26.3).
Polarized
Light microscopy light microscopy
Fetal
(second trimester)
200 m
Adult
200 m
Fig. 26.3 Histological appearance of annulus fibrosus tissue in fetus and adult showing tissue growth with age (H&E stain)
Fourthly, the required strengths of the interlamellar inter- Adams MA, McNally DS, Dolan P (1996) Stress distributions inside
faces and the nucleus pulposus-annulus fibrosus interface intervertebral discs. The effects of age and degeneration. J Bone
Joint Surg Br 78:965972
need to be determined; information that is critical for Adams MA, Dolan P, McNally DS (2009) The internal mechanical
defining design parameters for tissues that will function functioning of intervertebral discs and articular cartilage, and its rel-
in vivo. evance to matrix biology. Matrix Biol 28:384389
Fifthly, the effect of the environment into which the disc Ahmed N, Stanford WL, Kandel RA (2007) Mesenchymal stem and
progenitor cells for cartilage repair. Skeletal Radiol 36:909912
will be placed has yet to be considered (Smith et al. 2011). Aladin DM, Cheung KM, Chan D, Yee AF, Jim JJ, Luk KD, Lu WW
It is likely that this will be unfavorable particularly if the (2007) Expression of the Trp2 allele of COL9A2 is associated with
disease process is long standing and the sclerotic osseous alterations in the mechanical properties of human intervertebral
endplate is not replaced. Time to vascularization of the discs. Spine (Phila Pa 1976) 32:28202826
Allon AA, Aurouer N, Yoo BB, Liebenberg EC, Buser Z, Lotz JC
implant and nutrient diffusion are also unknown factors. (2010) Structured coculture of stem cells and disc cells prevent disc
Finally, clinical issues such as which individuals should degeneration in a rat model. Spine J 10:10891097
receive an implant and how should these discs be implanted Ambard D, Cherblanc F (2009) Mechanical behavior of annulus
and the subsequent rehabilitation protocol need to be fibrosus: a microstructural model of fibers reorientation. Ann
Biomed Eng 37:22562265
determined. Arana CJ, Diamandis EP, Kandel RA (2010) Cartilage tissue enhances
proteoglycan retention by nucleus pulposus cells in vitro. Arthritis
Rheum 62:33953403
26.10 Summary of Critical Concepts Discussed Asche CV, Kirkness CS, McAdam-Marx C, Fritz JM (2007) The soci-
etal costs of low back pain: data published between 2001 and 2007.
in the Chapter J Pain Palliat Care Pharmacother 21:2533
Attia M, Santerre JP, Kandel RA (2010) The response of annulus
No consensus exists as to which approach is optimal to fibrosus cell to fibronectin-coated nanofibrous polyurethane-anionic
treat symptomatic disc degeneration: implantation of dihydroxyoligomer scaffolds. Biomaterials 32:450460
Barrilleaux B, Knoepfler PS (2011) Inducing iPSCs to escape the dish.
cells, tissues, or an entire disc. Cell Stem Cell 9:103111
The biological repair treatment paradigm will likely be per- Bass EC, Ashford FA, Segal MR, Lotz JC (2004) Biaxial testing of
sonalized to the individual and influenced by extent of dis- human annulus fibrosus and its implications for a constitutive for-
ease. For example, early symptomatic degeneration, mulation. Ann Biomed Eng 32:12311242
Benneker LM, Heini PF, Alini M, Anderson SE, Ito K (2005) 2004
characterized by pathogenic changes in the nucleus pulposus, Young Investigator Award Winner: vertebral endplate marrow con-
may be better treated with cells alone, especially if a delivery tact channel occlusions and intervertebral disc degeneration. Spine
method other than injection (needle puncture) is developed. (Phila Pa 1976) 30:167173
However, when the disease affects the nucleus pulposus and Bibby SR, Jones DA, Lee RB, Yu J, Urban JPG (2001) The pathophysi-
ology of the intervertebral disc. Joint Bone Spine 68:537542
annulus fibrosus, the entire disc may have to be replaced. Bieback K, Kern S, Kocaomer A, Ferlik K, Bugert P (2008) Comparing
Concomitant with the development of tissue-engineering mesenchymal stromal cells from different human tissues: bone mar-
approaches to biological repair, it will also be necessary row, adipose tissue and umbilical cord blood. Biomed Mater Eng
to delineate the mechanism(s) leading to back pain so that 18:S71S76
Bieback K, Hecker A, Kocaomer A, Lannert H, Schallmoser K, Strunk
the right patients are selected for treatment to ensure that D, Kluter H (2009) Human alternatives to fetal bovine serum for the
biological treatment will be effective in ameliorating expansion of mesenchymal stromal cells from bone marrow. Stem
symptomatology. Cells 27:23312341
It will be necessary to determine how to best facilitate Bogduk N (1997) The inter-body joints and the intervertebral discs. In:
Bogduk N (ed) Clinical anatomy of the lumbar spine and sacrum.
biological repair/replacement and what is the optimal Churchill Livingstone, New York, pp 1331
posttreatment rehabilitation protocol. Bowles RD, Williams RM, Zipfel WR, Bonassar LJ (2010) Self-
assembly of aligned tissue-engineered annulus fibrosus and inter-
Acknowledgements Canadian Institutes of Health Research (CIHR) vertebral disc composite via collagen gel contraction. Tissue Eng
Grant MOP114991 Part A 16:13391348
Bowles RD, Gebhard HH, Hartl R, Bonassar LJ (2011) Tissue-
engineered intervertebral discs produce new matrix, maintain disc
height, and restore biomechanical function to the rodent spine. Proc
Natl Acad Sci U S A 108:1310613111
References Bron JL, Helder MN, Meisel HJ, Van Royen BJ, Smit TH (2009)
Repair, regenerative and supportive therapies of the annulus fibrosus:
Abbott RD, Purmessur D, Monsey RD, Iatridis JC (2012) Regenerative achievements and challenges. Eur Spine J 18:301313
potential of TGFbeta3 + Dex and notochordal cell conditioned media on Bron JL, Vonk LA, Smit TH, Koenderink GH (2011) Engineering alg-
degenerated human intervertebral disc cells. J Orthop Res 30:482488 inate for intervertebral disc repair. J Mech Behav Biomed Mater 4:
Acosta FL Jr, Metz L, Adkisson HD, Liu J, Carruthers-Liebenberg E, 11961205
Milliman C, Maloney M, Lotz JC (2011) Porcine intervertebral disc Cappello R, Bird JL, Pfeiffer D, Bayliss MT, Dudhia J (2006)
repair using allogeneic juvenile articular chondrocytes or mesen- Notochordal cell produce and assemble extracellular matrix in a dis-
chymal stem cells. Tissue Eng Part A 17:30453055 tinct manner, which may be responsible for the maintenance of
healthy nucleus pulposus. Spine (Phila Pa 1976) 31:873882; Frick SL, Hanley EN Jr, Meyer RA Jr, Ramp WK, Chapman TM (1994)
discussion 883 Lumbar intervertebral disc transfer. A canine study. Spine (Phila Pa
Carragee EJ, Don AS, Hurwitz EL, Cuellar JM, Carrino JA, Herzog 1976) 19:18261834; discussion 18341835
R (2009) 2009 ISSLS Prize Winner: does discography cause Ganey T, Libera J, Moos V, Alasevic O, Fritsch KG, Meisel HJ, Hutton
accelerated progression of degeneration changes in the lumbar WC (2003) Disc chondrocyte transplantation in a canine model: a
disc: a ten-year matched cohort study. Spine (Phila Pa 1976) 34: treatment for degenerated or damaged intervertebral disc. Spine
23382345 (Phila Pa 1976) 28:26092620
Chan WC, Sze KL, Samartzis D, Leung VY, Chan D (2011) Structure Gotz W, Barnert S, Bertagnoli R, Miosge N, Kresse H, Herken R (1997)
and biology of the intervertebral disk in health and disease. Orthop Immunohistochemical localization of the small proteoglycans deco-
Clin North Am 42:447464, vii rin and biglycan in human intervertebral discs. Cell Tissue Res
Chang G, Kim HJ, Kaplan D, Vunjak-Novakovic G, Kandel RA (2007) 289:185190
Porous silk scaffolds can be used for tissue engineering annulus Gruber HE, Stasky AA, Hanley EN Jr (1997) Characterization and phe-
fibrosus. Eur Spine J 16:18481857 notypic stability of human disc cells in vitro. Matrix Biol 16:
Choi KS, Cohn MJ, Harfe BD (2008) Identification of nucleus pulposus 285288
precursor cells and notochordal remnants in the mouse: implications Gruber HE, Johnson TL, Leslie K, Ingram JA, Martin D, Hoelscher G,
for disk degeneration and chordoma formation. Dev Dyn Banks D, Phieffer L, Coldham G, Hanley EN Jr (2002) Autologous
237:39533958 intervertebral disc cell implantation: a model using Psammomys
Chou AI, Nicoll SB (2009) Characterization of photocrosslinked alg- obesus, the sand rat. Spine (Phila Pa 1976) 27:16261633
inate hydrogels for nucleus pulposus cell encapsulation. J Biomed Gruber HE, Leslie K, Ingram J, Norton HJ, Hanley EN (2004) Cell-
Mater Res A 91:187194 based tissue engineering for the intervertebral disc: in vitro studies
Chou AI, Bansal A, Miller GJ, Nicoll SB (2006) The effect of serial of human disc cell gene expression and matrix production within
monolayer passaging on the collagen expression profile of outer and selected cell carriers. Spine J 4:4455
inner anulus fibrosus cells. Spine (Phila Pa 1976) 31:18751881 Gruber HE, Gordon B, Williams C, Norton HJ, Hanley EN Jr (2007)
Chou AI, Reza AT, Nicoll SB (2008) Distinct intervertebral disc cell Vertebral endplate and disc changes in the aging sand rat lumbar
populations adopt similar phenotypes in three-dimensional culture. spine: cross-sectional analyses of a large male and female popula-
Tissue Eng Part A 14:20792087 tion. Spine (Phila Pa 1976) 32:25292536
Chou R, Baisden J, Carragee EJ, Resnick DK, Shaffer WO, Loeser JD Grunhagen T, Wilde G, Soukane DM, Shirazi-Adl SA, Urban JP (2006)
(2009) Surgery for low back pain: a review of the evidence for an Nutrient supply and intervertebral disc metabolism. J Bone Joint
American Pain Society Clinical Practice Guideline. Spine (Phila Pa Surg Am 88(Suppl 2):3035
1976) 34:10941109 Guerin HA, Elliott DM (2006) Degeneration affects the fiber reorienta-
Choumerianou DM, Martimianaki G, Stiakaki E, Kalmanti L, Kalmanti tion of human annulus fibrosus under tensile load. J Biomech
M, Dimitriou H (2010) Comparative study of stemness characteris- 39:14101418
tics of mesenchymal cells from bone marrow of children and adults. Gupta MS, Cooper ES, Nicoll SB (2011) Transforming growth factor-
Cytotherapy 12:881887 beta 3 stimulates cartilage matrix elaboration by human marrow-
Cinotti G, Della Rocca C, Romeo S, Vittur F, Toffanin R, Trasimeni G derived stromal cells encapsulated in photocrosslinked
(2005) Degenerative changes of porcine intervertebral disc induced carboxymethylcellulose hydrogels: potential for nucleus pulposus
by vertebral endplate injuries. Spine (Phila Pa 1976) 30:174180 replacement. Tissue Eng Part A 17:29032910
Collin EC, Grad S, Zeugolis DI, Vinatier CS, Clouet JR, Guicheux JJ, Guyer RD, Shellock J, Maclennan B, Hanscom D, Knight RQ,
Weiss P, Alini M, Pandit AS (2011) An injectable vehicle for nucleus McCombe P, Jacobs JJ, Urban RM, Bradford D, Ohnmeiss DD
pulposus cell-based therapy. Biomaterials 32:28622870 (2011) Early failure of metal-on-metal artificial disc prostheses
Cortes DH, Elliott DM (2012) Extra-fibrillar matrix mechanics of annu- associated with lymphocytic reaction: diagnosis and treatment
lus fibrosus in tension and compression. Biomech Model experience in 4 cases. Spine (Phila Pa 1976) 36:E492E497
Mechanobiol 11:781790 Hadjipanayi E, Mudera V, Brown RA (2009) Close dependence of
Crevensten G, Walsh AJ, Ananthakrishnan D, Page P, Wahba GM, Lotz fibroblast proliferation on collagen scaffold matrix stiffness. J Tissue
JC, Berven S (2004) Intervertebral disc cell therapy for regenera- Eng Regen Med 3:7784
tion: mesenchymal stem cell implantation in rat intervertebral discs. Halloran DO, Grad S, Stoddart M, Dockery P, Alini M, Pandit AS
Ann Biomed Eng 32:430434 (2008) An injectable cross-linked scaffold for nucleus pulposus
Dagenais S, Caro J, Haldeman S (2008) A systematic review of low regeneration. Biomaterials 29:438447
back pain cost of illness studies in the United States and internation- Hamilton DJ, Seguin CA, Wang J, Pilliar RM, Kandel RA (2006)
ally. Spine J 8:820 Formation of a nucleus pulposus-cartilage endplate construct
Desrochers J, Duncan NA (2010) Strain transfer in the annulus fibrosus in vitro. Biomaterials 27:397405
under applied flexion. J Biomech 43:21412148 Harrison PE, Ashton IK, Johnson WE, Turner SL, Richardson JB,
Doorn J, Moll G, Le Blanc K, van Blitterswijk C, de Boer J (2012) Ashton BA (2000) The in vitro growth of human chondrocytes. Cell
Therapeutic applications of mesenchymal stromal cells: paracrine effects Tissue Bank 1:255260
and potential improvements. Tissue Eng Part B Rev 18:101115 Hegewald AA, Endres M, Abbushi A, Cabraja M, Woiciechowsky C,
Erickson IE, van Veen SC, Sengupta S, Kestle SR, Mauck RL (2011) Schmieder K, Kaps C, Thome C (2011a) Adequacy of herniated
Cartilage matrix formation by bovine mesenchymal stem cells in disc tissue as a cell source for nucleus pulposus regeneration.
three-dimensional culture is age-dependent. Clin Orthop Relat Res J Neurosurg Spine 14:273280
469:27442753 Hegewald AA, Enz A, Endres M, Sittinger M, Woiciechowsky C,
Erwin WM (2008) The Notochord, Notochordal cell and CTGF/CCN- Thome C, Kaps C (2011b) Engineering of polymer-based grafts
2: ongoing activity from development through maturation. J Cell with cells derived from human nucleus pulposus tissue of the lum-
Commun Signal 2:5965 bar spine. J Tissue Eng Regen Med 5:275282
Eyre DR (1979) Biochemistry of the intervertebral disc. Int Rev Connect Helen W, Gough JE (2008) Cell viability, proliferation and extracellular
Tissue Res 8:227291 matrix production of human annulus fibrosus cells cultured within
Fekete TF, Porchet F (2010) Overview of disc arthroplasty-past, present PDLLA/Bioglass composite foam scaffolds in vitro. Acta Biomater
and future. Acta Neurochir (Wien) 152:393404 4:230243
Hiyama A, Mochida J, Iwashina T, Omi H, Watanabe T, Serigano K, Kishen TJ, Diwan AD (2010) Fusion versus disk replacement for
Tamura F, Sakai D (2008) Transplantation of mesenchymal stem degenerative conditions of the lumbar and cervical spine: quid est
cells in a canine disc degeneration model. J Orthop Res 26: testimonium? Orthop Clin North Am 41:167181
589600 Koepsell L, Remund T, Bao J, Neufeld D, Fong H, Deng Y (2011)
Hohaus C, Ganey TM, Minkus Y, Meisel HJ (2008) Cell transplantation Tissue engineering of annulus fibrosus using electrospun fibrous
in lumbar spine disc degeneration disease. Eur Spine J 17(Suppl scaffolds with aligned polycaprolactone fibers. J Biomed Mater Res
4):492503 A 99:564575
Holm S, Holm AK, Ekstrom L, Karladani A, Hansson T (2004) Korecki CL, Taboas JM, Tuan RS, Iatridis JC (2010) Notochordal cell
Experimental disc degeneration due to endplate injury. J Spinal conditioned medium stimulates mesenchymal stem cell differentia-
Disord Tech 17:6471 tion toward a young nucleus pulposus phenotype. Stem Cell Res
Holzapfel GA, Schulze-Bauer CA, Feigl G, Regitnig P (2005) Single Ther 1:18
lamellar mechanics of the human lumbar anulus fibrosus. Biomech Krismer M, Haid C, Rabl W (1996) The contribution of anulus fibers to
Model Mechanobiol 3:125140 torque resistance. Spine (Phila Pa 1976) 21:25512557
Horner HA, Urban JP (2001) 2001 Volvo Award Winner in Basic Lange C, Cakiroglu F, Spiess AN, Cappallo-Obermann H, Dierlamm J,
Science Studies: effect of nutrient supply on the viability of cells Zander AR (2007) Accelerated and safe expansion of human mes-
from the nucleus pulposus of the intervertebral disc. Spine enchymal stromal cells in animal serum-free medium for transplan-
26:25432549 tation and regenerative medicine. J Cell Physiol 213:1826
Hsia HC, Nair MR, Mintz RC, Corbett SA (2011) The fiber diameter of Langer R, Vacanti JP (1993) Tissue engineering. Science 260:920926
synthetic bioresorbable extracellular matrix influences human Lazebnik M, Singh M, Glatt P, Friis LA, Berkland CJ, Detamore MS
fibroblast morphology and fibronectin matrix assembly. Plast (2011) Biomimetic method for combining the nucleus pulposus and
Reconstr Surg 127:23122320 annulus fibrosus for intervertebral disc tissue engineering. J Tissue
Hsieh AH, Twomey JD (2010) Cellular mechanobiology of the interver- Eng Regen Med 5:e179e187
tebral disc: new directions and approaches. J Biomech 43:137145 Le Maitre CL, Freemont AJ, Hoyland JA (2007) Accelerated cellular
Huang RC, Tropiano P, Marnay T, Girardi FP, Lim MR, Cammisa FP Jr senescence in degenerate intervertebral discs: a possible role in the
(2006) Range of motion and adjacent level degeneration after lum- pathogenesis of intervertebral disc degeneration. Arthritis Res Ther
bar total disc replacement. Spine J 6:242247 9:R45
Huang B, Zhuang Y, Li CQ, Liu LT, Zhou Y (2011) Regeneration of the Le Visage C, Kim SW, Tateno K, Sieber AN, Kostuik JP, Leong KW
intervertebral disc with nucleus pulposus cell-seeded collagen II/ (2006) Interaction of human mesenchymal stem cells with disc
hyaluronan/chondroitin-6-sulfate tri-copolymer constructs in a rab- cells: changes in extracellular matrix biosynthesis. Spine (Phila Pa
bit disc degeneration model. Spine (Phila Pa 1976) 36:22522259 1976) 31:20362042
Hukins DW (1994) In: Ghosh P (ed) Biology of intervertebral disc. Lee P (1994) The economic impact of musculoskeletal disorders. Qual
Disc Structure and Function. CRC Press, Boca Raton, p 237 Life Res 3(Suppl 1):S85S91
Iatridis JC, Michalek AJ, Purmessur D, Korecki CL (2009) Localized Lee KI, Moon SH, Kim H, Kwon UH, Kim HJ, Park SN, Suh H, Lee
intervertebral disc injury leads to organ level changes in structure, HM, Kim HS, Chun HJ, Kwon IK, Jang JW (2012a) Tissue engi-
cellularity, and biosynthesis. Cell Mol Bioeng 2:437447 neering of the intervertebral disc with cultured nucleus pulposus
Ikada Y (2006) Challenges in tissue engineering. J R Soc Interface cells using atelocollagen scaffold and growth factors. Spine (Phila
3:589601 Pa 1976) 37:452458
Illien-Junger S, Pattappa G, Peroglio M, Benneker LM, Stoddart MJ, Lee MJ, Dettori JR, Standaert CJ, Brodt ED, Chapman JR (2012b) The
Sakai D, Mochida J, Grad S, Alini M (2012) Homing of mesenchy- natural history of degeneration of the lumbar and cervical spines: a
mal stem cells in induced degenerative intervertebral discs in a systematic review. Spine (Phila Pa 1976) 37:S18S30
whole organ culture system. Spine (Phila Pa 1976) 37:18651873 Lister R, Pelizzola M, Kida YS, Hawkins RD, Nery JR, Hon G,
Inerot S, Axelsson I (1991) Structure and composition of proteoglycans Antosiewicz-Bourget J, OMalley R, Castanon R, Klugman S,
from human annulus fibrosus. Connect Tissue Res 26:4763 Downes M, Yu R, Stewart R, Ren B, Thomson JA, Evans RM, Ecker
Iron KS, Manuel DG, Williams J (2004) Using a linked data set to JR (2011) Hotspots of aberrant epigenomic reprogramming in
determine the factors associated with utilization and costs of family human induced pluripotent stem cells. Nature 471:6873
physician services in Ontario: effects of self-reported chronic condi- Lotz JC, Hadi T, Bratton C, Reiser KM, Hsieh AH (2008) Anulus
tions. Chronic Dis Can 24:124132 fibrosus tension inhibits degenerative structural changes in lamellar
Javedan SP, Dickman CA (1999) Cause of adjacent-segment disease collagen. Eur Spine J 17:11491159
after spinal fusion. Lancet 354:530531 Luk KD, Ruan DK, Chow DH, Leong JC (1997) Intervertebral disc
Jim JJ, Noponen-Hietala N, Cheung KM, Ott J, Karppinen J, autografting in a bipedal animal model. Clin Orthop Relat Res
Sahraravand A, Luk KD, Yip SP, Sham PC, Song YQ, Leong JC, 337:1326
Cheah KS, Ala-Kokko L, Chan D (2005) The TRP2 allele of Maherali N, Ahfeldt T, Rigamonti A, Utikal J, Cowan C, Hochedlinger
COL9A2 is an age-dependent risk factor for the development and K (2008) A high-efficiency system for the generation and study of
severity of intervertebral disc degeneration. Spine (Phila Pa 1976) human induced pluripotent stem cells. Cell Stem Cell 3:340345
30:27352742 Manchikanti L, Singh V, Datta S, Cohen SP, Hirsch JA (2009)
Kandel RA, Grynpas M, Pilliar R, Lee J, Wang J, Waldman S, Zalzal P, Comprehensive review of epidemiology, scope, and impact of spi-
Hurtig M (2006) Repair of osteochondral defects with biphasic nal pain. Pain Physician 12:E35E70
cartilage-calcium polyphosphate constructs in a sheep model. Marchand F, Ahmed AM (1990) Investigation of the laminate struc-
Biomaterials 27:41204131 ture of lumbar disc anulus fibrosus. Spine (Phila Pa 1976) 15:
Kandel RA, Hamilton D, Seguin C, Li SQ, Arana C, Pilliar R (2007) An 402410
in vitro tissue model to study the effect of age on nucleus pulposus Markolf KL, Morris JM (1974) The structural components of the inter-
cells. Eur Spine J 16:21662173 vertebral disc. A study of their contributions to the ability of the disc
Kandel R, Roberts S, Urban JP (2008) Tissue engineering and the inter- to withstand compressive forces. J Bone Joint Surg Am 56:675687
vertebral disc: the challenges. Eur Spine J 17(Suppl 4):480491 Maroudas A, Stockwell RA, Nachemson A, Urban J (1975) Factors
Kim PK, Branch CL Jr (2006) The lumbar degenerative disc: confu- involved in the nutrition of the human lumbar intervertebral disc:
sion, mechanics, management. Clin Neurosurg 53:1825 cellularity and diffusion of glucose in vitro. J Anat 120:113130
Masuda K (2008) Biological repair of the degenerated intervertebral Nesti LJ, Li WJ, Shanti RM, Jiang YJ, Jackson W, Freedman BA, Kuklo
disc by the injection of growth factors. Eur Spine J 17(Suppl TR, Giuliani JR, Tuan RS (2008) Intervertebral disc tissue engineer-
4):441451 ing using a novel hyaluronic acid-nanofibrous scaffold (HANFS)
Masuda K, Lotz JC (2010) New challenges for intervertebral disc treat- amalgam. Tissue Eng Part A 14:15271537
ment using regenerative medicine. Tissue Eng Part B Rev Nettles DL, Chilkoti A, Setton LA (2010) Applications of elastin-like
16:147158 polypeptides in tissue engineering. Adv Drug Deliv Rev
McCann MR, Tamplin OJ, Rossant J, Seguin CA (2012) Tracing noto- 62:14791485
chord-derived cells using a Noto-cre mouse: implications for inter- Nomura T, Mochida J, Okuma M, Nishimura K, Sakabe K (2001)
vertebral disc development. Dis Model Mech 5:7382 Nucleus pulposus allograft retards intervertebral disc degeneration.
Meisel HJ, Siodla V, Ganey T, Minkus Y, Hutton WC, Alasevic OJ Clin Orthop Relat Res (389):94101
(2007) Clinical experience in cell-based therapeutics: disc chondro- Nosikova YS, Santerre JP, Grynpas M, Gibson G, Kandel RA (2012)
cyte transplantation A treatment for degenerated or damaged inter- Characterization of the annulus fibrosus-vertebral body interface:
vertebral disc. Biomol Eng 24:521 identification of new structural features. J Anat 221:577589
Melrose J, Ghosh P, Taylor TK (2001) A comparative analysis of the OConnell GD, Vresilovic EJ, Elliott DM (2011) Human intervertebral
differential spatial and temporal distributions of the large (aggrecan, disc internal strain in compression: the effect of disc region, loading
versican) and small (decorin, biglycan, fibromodulin) proteoglycans position, and degeneration. J Orthop Res 29:547555
of the intervertebral disc. J Anat 198:315 OHalloran DM, Pandit AS (2007) Tissue-engineering approach to
Melrose J, Smith SM, Appleyard RC, Little CB (2008) Aggrecan, ver- regenerating the intervertebral disc. Tissue Eng 13:19271954
sican and type VI collagen are components of annular translamellar Okuma M, Mochida J, Nishimura K, Sakabe K, Seiki K (2000)
crossbridges in the intervertebral disc. Eur Spine J 17:314324 Reinsertion of stimulated nucleus pulposus cells retards interverte-
Michalek AJ, Funabashi KL, Iatridis JC (2010) Needle puncture injury bral disc degeneration: an in vitro and in vivo experimental study.
of the rat intervertebral disc affects torsional and compressive bio- J Orthop Res 18:988997
mechanics differently. Eur Spine J 19:21102116 Olson EJ, Hanley EN Jr, Rudert MJ, Baratz ME (1991) Vertebral column
Michalek AJ, Gardner-Morse MG, Iatridis JC (2012) Large residual allografts for the treatment of segmental spine defects. An experi-
strains are present in the intervertebral disc annulus fibrosus in the mental investigation in dogs. Spine (Phila Pa 1976) 16:10811088
unloaded state. J Biomech 45:12271231 Orozco L, Soler R, Morera C, Alberca M, Sanchez A, Garcia-Sancho J
Miller JA, Schmatz C, Schultz AB (1988) Lumbar disc degeneration: (2011) Intervertebral disc repair by autologous mesenchymal bone
correlation with age, sex, and spine level in 600 autopsy specimens. marrow cells: a pilot study. Transplantation 92:822828
Spine 13:173178 Park SH, Gil ES, Cho H, Mandal BB, Tien LW, Min BH, Kaplan DL
Mirza SK, Deyo RA (2007) Systematic review of randomized trials (2012) Intervertebral disk tissue engineering using biphasic silk
comparing lumbar fusion surgery to nonoperative care for treatment composite scaffolds. Tissue Eng Part A 18:447458
of chronic back pain. Spine (Phila Pa 1976) 32:816823 Peng B, Hou S, Shi Q, Jia L (2001) The relationship between cartilage
Miyamoto T, Muneta T, Tabuchi T, Matsumoto K, Saito H, Tsuji K, end-plate calcification and disc degeneration: an experimental study.
Sekiya I (2010) Intradiscal transplantation of synovial mesenchy- Chin Med J (Engl) 114:308312
mal stem cells prevents intervertebral disc degeneration through Pereira D, Silva-Correia J, Oliveira J, Reis R (2013) Hydrogels in acel-
suppression of matrix metalloproteinase-related genes in nucleus lular and cellular strategies for intervertebral disc regeneration.
pulposus cells in rabbits. Arthritis Res Ther 12:R206 J Tissue Eng Regen Med 7:8598
Mizuno H, Roy AK, Zaporojan V, Vacanti CA, Ueda M, Bonassar LJ Purmessur D, Schek RM, Abbott RD, Ballif BA, Godburn KE, Iatridis JC
(2006) Biomechanical and biochemical characterization of compos- (2011) Notochordal conditioned media from tissue increases proteo-
ite tissue-engineered intervertebral discs. Biomaterials 27:362370 glycan accumulation and promotes a healthy nucleus pulposus pheno-
Moore RJ (2006) The vertebral endplate: disc degeneration, disc regen- type in human mesenchymal stem cells. Arthritis Res Ther 13:R81
eration. Eur Spine J 15(Suppl 3):S333S337 Putzier M, Schneider SV, Funk JF, Tohtz SW, Perka C (2005) The surgi-
Munoz M, Rodriguez A, De Frutos C, Caamano JN, Diez C, Facal N, cal treatment of the lumbar disc prolapse: nucleotomy with addi-
Gomez E (2008) Conventional pluripotency markers are unspecific tional transpedicular dynamic stabilization versus nucleotomy
for bovine embryonic-derived cell-lines. Theriogenology 69: alone. Spine (Phila Pa 1976) 30:E109E114
11591164 Raj PP (2008) Intervertebral disc: anatomy-physiology-pathophysiol-
Mwale F, Roughley P, Antoniou J (2004) Distinction between the extra- ogy-treatment. Pain Pract 8:1844
cellular matrix of the nucleus pulposus and hyaline cartilage: a req- Rajasekaran S, Babu JN, Arun R, Armstrong BR, Shetty AP, Murugan
uisite for tissue engineering of intervertebral disc. Eur Cell Mater S (2004) ISSLS prize winner: a study of diffusion in human lumbar
8:5863, discussion 6364 discs: a serial magnetic resonance imaging study documenting the
Nagae M, Ikeda T, Mikami Y, Hase H, Ozawa H, Matsuda K, Sakamoto influence of the endplate on diffusion in normal and degenerate
H, Tabata Y, Kawata M, Kubo T (2007) Intervertebral disc regenera- discs. Spine (Phila Pa 1976) 29:26542667
tion using platelet-rich plasma and biodegradable gelatin hydrogel Rapoport J, Jacobs P, Bell NR, Klarenbach S (2004) Refining the mea-
microspheres. Tissue Eng 13:147158 surement of the economic burden of chronic diseases in Canada.
Nakagawa M, Koyanagi M, Tanabe K, Takahashi K, Ichisaka T, Aoi T, Chronic Dis Can 25:1321
Okita K, Mochiduki Y, Takizawa N, Yamanaka S (2008) Generation Razaq S, Wilkins RJ, Urban JP (2003) The effect of extracellular pH on
of induced pluripotent stem cells without Myc from mouse and matrix turnover by cells of the bovine nucleus pulposus. Eur Spine
human fibroblasts. Nat Biotechnol 26:101106 J 12:341349
Nerurkar NL, Baker BM, Sen S, Wible EE, Elliott DM, Mauck RL Richardson SM, Hoyland JA (2008) Stem cell regeneration of degener-
(2009) Nanofibrous biologic laminates replicate the form and func- ated intervertebral discs: current status. Curr Pain Headache Rep
tion of the annulus fibrosus. Nat Mater 8:986992 12:8388
Nerurkar NL, Sen S, Huang AH, Elliott DM, Mauck RL (2010a) Richardson SM, Walker RV, Parker S, Rhodes NP, Hunt JA, Freemont
Engineered disc-like angle-ply structures for intervertebral disc AJ, Hoyland JA (2006) Intervertebral disc cell-mediated mesenchy-
replacement. Spine (Phila Pa 1976) 35:867873 mal stem cell differentiation. Stem Cells 24:707716
Nerurkar NL, Elliott DM, Mauck RL (2010b) Mechanical design criteria Richardson SM, Mobasheri A, Freemont AJ, Hoyland JA (2007)
for intervertebral disc tissue engineering. J Biomech 43:10171030 Intervertebral disc biology, degeneration and novel tissue
engineering and regenerative medicine therapies. Histol Histopathol Shao X, Hunter CJ (2007) Developing an alginate/chitosan hybrid fiber
22:10331041 scaffold for annulus fibrosus cells. J Biomed Mater Res A
Richardson SM, Hughes N, Hunt JA, Freemont AJ, Hoyland JA (2008) 82:701710
Human mesenchymal stem cell differentiation to NP-like cells in Shapiro IM, Risbud MV (2010) Transcriptional profiling of the nucleus
chitosan-glycerophosphate hydrogels. Biomaterials 29:8593 pulposus: say yes to notochord. Arthritis Res Ther 12:117
Risbud MV, Shapiro IM (2011) Notochordal cells in the adult interver- Sheikh H, Zakharian K, De La Torre RP, Facek C, Vasquez A, Chaudhry
tebral disc: new perspective on an old question. Crit Rev Eukaryot GR, Svinarich D, Perez-Cruet MJ (2009) In vivo intervertebral disc
Gene Expr 21:2941 regeneration using stem cell-derived chondroprogenitors.
Risbud MV, Schaer TP, Shapiro IM (2010) Toward an understanding of J Neurosurg Spine 10:265272
the role of notochordal cells in the adult intervertebral disc: from Shim CS, Lee SH, Shin HD, Kang HS, Choi WC, Jung B, Choi G, Ahn
discord to accord. Dev Dyn 239:21412148 Y, Lee S, Lee HY (2007) CHARITE versus ProDisc: a comparative
Roberts S, Urban JP, Evans H, Eisenstein SM (1996) Transport proper- study of a minimum 3-year follow-up. Spine (Phila Pa 1976)
ties of the human cartilage endplate in relation to its composition 32:10121018
and calcification. Spine (Phila Pa 1976) 21:415420 Shirazi-Adl A, Taheri M, Urban JP (2010) Analysis of cell viability in
Roberts S, Evans EH, Kletsas D, Jaffray DC, Eisenstein SM (2006) intervertebral disc: effect of endplate permeability on cell popula-
Senescence in human intervertebral discs. Eur Spine J 15(Suppl tion. J Biomech 43:13301336
3):S312S316 Simon SR (1994) In: Simon SR (ed) Orthopedic basic science.
Ruan D, He Q, Ding Y, Hou L, Li J, Luk KD (2007) Intervertebral disc Kinesiology. Am Academy of Orthopedic Surgeons, Rosemont,
transplantation in the treatment of degenerative spine disease: a pre- pp 558568
liminary study. Lancet 369:993999 Singh K, Masuda K, Thonar EJ, An HS, Cs-Szabo G (2009) Age-related
Saad L, Spector M (2004) Effects of collagen type on the behavior of changes in the extracellular matrix of nucleus pulposus and anulus
adult canine annulus fibrosus cells in collagen-glycosaminoglycan fibrosus of human intervertebral disc. Spine (Phila Pa 1976) 34:1016
scaffolds. J Biomed Mater Res A 71:233241 Smith LJ, Elliott DM (2011) Formation of lamellar cross bridges in the
Sahlman J, Inkinen R, Hirvonen T, Lammi MJ, Lammi PE, Nieminen J, annulus fibrosus of the intervertebral disc is a consequence of vas-
Lapvetelainen T, Prockop DJ, Arita M, Li SW, Hyttinen MM, cular regression. Matrix Biol 30:267274
Helminen HJ, Puustjarvi K (2001) Premature vertebral endplate Smith SM, Whitelock JM, Iozzo RV, Little CB, Melrose J (2009)
ossification and mild disc degeneration in mice after inactivation of Topographical variation in the distributions of versican, aggrecan
one allele belonging to the Col2a1 gene for type II collagen. Spine and perlecan in the foetal human spine reflects their diverse func-
(Phila Pa 1976) 26:25582565 tional roles in spinal development. Histochem Cell Biol 132:
Saino E, Focarete ML, Gualandi C, Emanuele E, Cornaglia AI, Imbriani 491503
M, Visai L (2011) Effect of electrospun fiber diameter and align- Smith LJ, Chiaro JA, Nerurkar NL, Cortes DH, Horava SD, Hebela
ment on macrophage activation and secretion of proinflammatory NM, Mauck RL, Dodge GR, Elliott DM (2011) Nucleus pulposus
cytokines and chemokines. Biomacromolecules 12:19001911 cells synthesize a functional extracellular matrix and respond to
Sakai D, Mochida J, Iwashina T, Watanabe T, Nakai T, Ando K, Hotta inflammatory cytokine challenge following long-term agarose cul-
T (2005) Differentiation of mesenchymal stem cells transplanted ture. Eur Cell Mater 22:291301
to a rabbit degenerative disc model: potential and limitations So K, Takemoto M, Fujibayashi S, Neo M, Kyomoto M, Hayami T,
for stem cell therapy in disc regeneration. Spine (Phila Pa 1976) Hyon SH, Nakamura T (2007) Antidegenerative effects of partial
30:23792387 disc replacement in an animal surgery model. Spine (Phila Pa 1976)
Schallmoser K, Bartmann C, Rohde E, Reinisch A, Kashofer K, 32:15861591
Stadelmeyer E, Drexler C, Lanzer G, Linkesch W, Strunk D (2007) Sobajima S, Vadala G, Shimer A, Kim JS, Gilbertson LG, Kang JD
Human platelet lysate can replace fetal bovine serum for clinical- (2008) Feasibility of a stem cell therapy for intervertebral disc
scale expansion of functional mesenchymal stromal cells. degeneration. Spine J 8:888896
Transfusion 47:14361446 Steck E, Bertram H, Abel R, Chen B, Winter A, Richter W (2005)
Schmidt H, Heuer F, Wilke HJ (2009) Dependency of disc degeneration Induction of intervertebral disc-like cells from adult mesenchymal
on shear and tensile strains between annular fiber layers for complex stem cells. Stem Cells 23:403411
loads. Med Eng Phys 31:642649 Strassburg S, Richardson SM, Freemont AJ, Hoyland JA (2010)
Schollum ML, Robertson PA, Broom ND (2009) A microstructural Co-culture induces mesenchymal stem cell differentiation and mod-
investigation of intervertebral disc lamellar connectivity: detailed ulation of the degenerate human nucleus pulposus cell phenotype.
analysis of the translamellar bridges. J Anat 214:805816 Regen Med 5:701711
Schollum ML, Appleyard RC, Little CB, Melrose J (2010) A detailed Takahashi K, Yamanaka S (2006) Induction of pluripotent stem cells
microscopic examination of alterations in normal anular structure from mouse embryonic and adult fibroblast cultures by defined fac-
induced by mechanical destabilization in an ovine model of disc tors. Cell 126:663676
degeneration. Spine (Phila Pa 1976) 35:19651973 Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K,
Schriebl K, Satianegara G, Hwang A, Tan HL, Fong WJ, Yang HH, Yamanaka S (2007) Induction of pluripotent stem cells from adult
Jungbauer A, Choo A (2012) Selective removal of undifferentiated human fibroblasts by defined factors. Cell 131:861872
human embryonic stem cells using magnetic activated cell sorting Takatalo J, Karppinen J, Niinimaki J, Taimela S, Nayha S, Mutanen P,
followed by a cytotoxic antibody. Tissue Eng Part A 18:899909 Sequeiros RB, Kyllonen E, Tervonen O (2011) Does lumbar disc
Schulte TL, Leistra F, Bullmann V, Osada N, Vieth V, Marquardt B, degeneration on magnetic resonance imaging associate with low
Lerner T, Liljenqvist U, Hackenberg L (2007) Disc height reduction back symptom severity in young Finnish adults? Spine (Phila Pa
in adjacent segments and clinical outcome 10 years after lumbar 360 1976) 36:21802189
degrees fusion. Eur Spine J 16:21522158 Takatalo J, Karppinen J, Niinimaki J, Taimela S, Mutanen P, Sequeiros
Shahdadfar A, Fronsdal K, Haug T, Reinholt FP, Brinchmann JE RB, Nayha S, Jarvelin MR, Kyllonen E, Tervonen O (2012)
(2005) In vitro expansion of human mesenchymal stem cells: Association of modic changes, Schmorls nodes, spondylolytic
choice of serum is a determinant of cell proliferation, differentia- defects, high intensity zone lesions, disc herniations, and radial tears
tion, gene expression, and transcriptome stability. Stem Cells with low back symptom severity among young Finnish adults. Spine
23:13571366 (Phila Pa 1976) 37:12311239
Tekkatte C, Gunasingh GP, Cherian KM, Sankaranarayanan K (2011) Nagy A (2009) PiggyBac transposition reprograms fibroblasts to
Humanized stem cell culture techniques: the animal serum con- induced pluripotent stem cells. Nature 458:766770
troversy. Stem Cells Int 2011:504723 Yamanaka S (2007) Strategies and new developments in the generation
Tumialan LM, Gluf WM (2011) Progressive vertebral body osteolysis of patient-specific pluripotent stem cells. Cell Stem Cell 1:3949
after cervical arthroplasty. Spine (Phila Pa 1976) 36:E973E978 Yang X, Li X (2009) Nucleus pulposus tissue engineering: a brief
Vadala G, Studer RK, Sowa G, Spiezia F, Iucu C, Denaro V, Gilbertson review. Eur Spine J 18:15641572
LG, Kang JD (2008) Coculture of bone marrow mesenchymal stem Yang L, Kandel RA, Chang G, Santerre JP (2009) Polar surface chem-
cells and nucleus pulposus cells modulate gene expression profile istry of nanofibrous polyurethane scaffold affects annulus fibrosus
without cell fusion. Spine (Phila Pa 1976) 33:870876 cell attachment and early matrix accumulation. J Biomed Mater Res
Vadala G, Mozetic P, Rainer A, Centola M, Loppini M, Trombetta M, A 91:10891099
Denaro V (2012) Bioactive electrospun scaffold for annulus fibrosus Yang R, Jiang M, Kumar SM, Xu T, Wang F, Xiang L, Xu X (2011)
repair and regeneration. Eur Spine J 21(Suppl 1):S20S26 Generation of melanocytes from induced pluripotent stem cells.
van Ooij A, Kurtz SM, Stessels F, Noten H, van Rhijn L (2007) J Invest Dermatol 131:24582466
Polyethylene wear debris and long-term clinical failure of the Yao J, Turteltaub SR, Ducheyne P (2006) A three-dimensional nonlin-
Charite disc prosthesis: a study of 4 patients. Spine (Phila Pa 1976) ear finite element analysis of the mechanical behavior of tissue engi-
32:223229 neered intervertebral discs under complex loads. Biomaterials
Vernengo J, Fussell GW, Smith NG, Lowman AM (2008) Evaluation of 27:377387
novel injectable hydrogels for nucleus pulposus replacement. Yu J, Winlove PC, Roberts S, Urban JP (2002) Elastic fibre organiza-
J Biomed Mater Res B Appl Biomater 84:6469 tion in the intervertebral discs of the bovine tail. J Anat
Videman T, Nurminen M (2004) The occurrence of anular tears and 201:465475
their relation to lifetime back pain history: a cadaveric study using Yu J, Fairbank JC, Roberts S, Urban JP (2005) The elastic fiber network
barium sulfate discography. Spine (Phila Pa 1976) 29:26682676 of the anulus fibrosus of the normal and scoliotic human interverte-
Wade KR, Robertson PA, Broom ND (2012) On how nucleus-endplate bral disc. Spine (Phila Pa 1976) 30:18151820
integration is achieved at the fibrillar level in the ovine lumbar disc. Yu J, Tirlapur U, Fairbank J, Handford P, Roberts S, Winlove CP, Cui
J Anat 221:3946 Z, Urban J (2007) Microfibrils, elastin fibres and collagen fibres in
Waldman SD, Grynpas MD, Pilliar RM, Kandel RA (2002) the human intervertebral disc and bovine tail disc. J Anat 210:
Characterization of cartilagenous tissue formed on calcium poly- 460471
phosphate substrates in vitro. J Biomed Mater Res 62:323330 Zeh A, Becker C, Planert M, Lattke P, Wohlrab D (2009) Time-
Wan Y, Feng G, Shen FH, Laurencin CT, Li X (2008) Biphasic scaffold dependent release of cobalt and chromium ions into the serum fol-
for annulus fibrosus tissue regeneration. Biomaterials 29:643652 lowing implantation of the metal-on-metal Maverick type artificial
Wang Y, Huso DL, Harrington J, Kellner J, Jeong DK, Turney J, lumbar disc (Medtronic Sofamor Danek). Arch Orthop Trauma Surg
McNiece IK (2005) Outgrowth of a transformed cell population 129:741746
derived from normal human BM mesenchymal stem cell culture. Zhang Y, Phillips FM, Thonar EJ, Oegema T, An HS, Roman-Blas JA,
Cytotherapy 7:509519 He TC, Anderson DG (2008) Cell therapy using articular chondro-
Wei Y, Zeng W, Wan R, Wang J, Zhou Q, Qiu S, Singh SR (2012) cytes overexpressing BMP-7 or BMP-10 in a rabbit disc organ cul-
Chondrogenic differentiation of induced pluripotent stem cells from ture model. Spine (Phila Pa 1976) 33:831838
osteoarthritic chondrocytes in alginate matrix. Eur Cell Mater 23:112 Zhang H, La Marca F, Hollister SJ, Goldstein SA, Lin CY (2009)
Weiler C, Nerlich AG, Schaaf R, Bachmeier BE, Wuertz K, Boos N Developing consistently reproducible intervertebral disc degenera-
(2010) Immunohistochemical identification of notochordal markers tion at rat caudal spine by using needle puncture. J Neurosurg Spine
in cells in the aging human lumbar intervertebral disc. Eur Spine J 10:522530
19:17611770 Zhao T, Zhang ZN, Rong Z, Xu Y (2011) Immunogenicity of induced
Wilson A, Shehadeh LA, Yu H, Webster KA (2010) Age-related molec- pluripotent stem cells. Nature 474:212215
ular genetic changes of murine bone marrow mesenchymal stem Zuk PA, Zhu M, Ashjian P, De Ugarte DA, Huang JI, Mizuno H,
cells. BMC Genomics 11:229 Alfonso ZC, Fraser JK, Benhaim P, Hedrick MH (2002) Human
Woltjen K, Michael IP, Mohseni P, Desai R, Mileikovsky M, Hamalainen adipose tissue is a source of multipotent stem cells. Mol Biol Cell
R, Cowling R, Wang W, Liu P, Gertsenstein M, Kaji K, Sung HK, 13:42794295
About the Editors
Irving M. Shapiro is the Director of the Division of Makarand V. Risbud is the leader of a team of investigators
Orthopaedic Research in the Department of Orthopaedic studying the pathogenesis of intervertebral disc disease. He
Surgery and the Anthony and Gertrude DePalma Professor is an international authority on the molecular biology of cells
of Orthopaedic Surgery, Jefferson Medical College, of the nucleus pulposus. Dr. Risbud did his post doctoral
Philadelphia. Dr. Shapiro is a biochemist with a long term training at Harvard Medical School and is currently Professor
interest in skeletal tissues, especially the biology of bone, the of Orthopaedic Surgery at Jefferson Medical College in
growth plate and the intervertebral disc. His current work is Philadelphia where he is also an active member of the pro-
focused on new approaches to restoring function to the gram in Cell and Developmental Biology of the Graduate
degenerative disc. School of Biomedical Sciences.

DOI 10.1007/978-3-7091-1535-0, Springer-Verlag Wien 2014
Index
A caprine model, 298, 299

Acetaminophen, 215 clinical perspective, 292, 294
Acroflex, 27 considerations, 342
A disintegrin and metalloproteinase with thrombospondin motifs genetically modified mouse, 342349
(ADAMTS), 181 history, 292
A disintegrin-like and metalloproteinase with thrombospondin type 1 nonhuman primates, 299300
motif (ADAMTS) ovine model, 297299
domain structures and differential cell localization, 127 porcine models, 295, 297, 298
intervertebral disc biology and disease, 132133 reliability, 294
roles of, 128, 130 selection of, 300
structure and function in silico models, 292
ADAMTS10, 129130 standardization, 294
ADAMTS4 and ADAMTS5, 129 validity and fidelity, 294295
ADAMTS7 and ADAMTS12, 129 in vitro models, 292
ADAMTS-like proteins, 130 in vivo research, 292
dermatosparactic collagen, 128 disc herniation
domain structure, 127 categories, 308309
mammalian ADAMTS proteases, 127, 128 chemical and inflammatory mediators, 313315
molecular functions, 128, 129 dysesthesia, 309, 314
procollagen-processing enzymes, 128 limb hypersensitivity, 309
proteoglycanase cluster, 130 mechanical allodynia, 309, 312
TIMPs, 127128 mechanical factors, 310312
Adjacent segment degeneration, 226, 227 quantitative gait analysis, 309
Adjacent segment disease gene delivery, of growth/transcription factors, 394
description, 226 Anisotropy, 18
incidence, 227 Annulus fibrosus, 9, 18, 2123
in lumbar spine, 227 biaxial loading, 22
prevalence, 227 collagens in, 80, 81 (see also Collagen)
TDA, 226 elasticity, 21
Age-related spontaneous intervertebral disc degeneration. mechanical behavior, 22
See Sand rat model proteoglycans and collagen, 21
Aggrecan shear tests, 23
degradation, 59 tensile properties of, 22
gene expression, 58 transport properties, 22
immunolocalization of, 55, 56 Annulus fibrosus cell culture, 333334
mutations in, 5859 Annulus fibrosus tissue engineering
structure and function cell therapy, 424
disulfide-bonded loops, 5556 complex loading patterns, 424
domains, 56 healing potential, 424
organization, 55, 58 scaffolds, 425
swelling properties, 55 Anterior cervical discectomy and fusion
variation in, 55, 57 (ACDF), 226
Aging Antidepressants, 253254
cartilage, molecular characteristics, 268 Anti-inflammatory/anti-degeneration genes, 386
genetic studies, 161 ADAMTS5 siRNA, 390
IVD degeneration, 265 IL-1Ra, 390
Allele, 158 TIMP-1, 389
Allodynia, 314 Apoptosis, 187188
Anabolic genes, 386 Aquaporins (AQP), 102
Animal models, 295296. See also Sand rat model Asporin, 69
disc degeneration Autologous disc cell therapy, 333
canine model, 297299 Autophagy, 188189

DOI 10.1007/978-3-7091-1535-0, Springer-Verlag Wien 2014
438 Index
B hemicorporectomy, 283
Back pain histologic features, 281
epidemiology, 248 history, 277278
etiology of, 247248 immunohistochemical markers, 281, 282
natural history, 248 MRI and CT scan, 280281
nonmechanical, 248 neurological symptoms, 278279
nonoperative treatment (see Nonoperative treatment, of back pain) non-neurologic symptoms, 279
nonorganic etiology, 248 plain radiographs, 280
Bagbys basket, 293 platelet-derived growth factor receptors, 284
Behavioral therapy, 249 radiation therapy, 283284
Biglycans, 69, 346347 RTK inhibition, 284285
Bioluminescence imaging, 388 sacral amputations, 283
Biomaterials, for intervertebral disc repair, 366 sacral resection, 282283
Botulinum toxin A injection, 254255 tumor staging, 282
Brachyury, 3536, 281, 282, 286 wide surgical resection, 282
Bryan cervical disc arthroplasty device, 235236, 238, 239 Chymopapain, 269
Collagen
amyloid deposits, 8485
C biosynthesis of
Camptodactyly-arthropathy-coxa vara-pericarditis syndrome chaperone-assisted folding, 85
(CACP), 66 domain structure, 8687
Canine model, 297299 folding, 87
Caprine model, 298, 299 intracellular processing, 85
Carbon-ion radiotherapy, in chordoma management, 284 posttranslational modifications, 85
Cartilage endplates registration peptides, 8586
collagens in, 80 cartilage oligomeric matrix protein, 84
tissue engineering, 425 classification of
UTE-MRI, 410 collagen I, 82, 83
Cartilage oligomeric matrix protein (COMP), 83, 84 collagen II, 82
b-Catenin-dependent Wnt signaling, 348 collagen IX, 84
Cauda equina, 279280 collagen VI, 8283
Cell ageing and death collagen X, 83
apoptosis, 187188 discoidin domain receptors binding, 88
autophagy, 188189 elastin, 85
cell clustering and proliferation, 189 extracellular processing, 87
cellular phenotype, 190 fibronectin, 84
matrix degradation, 187 hereditary disorders, 8889
replicative senescence, 190 homeostasis, 88
Cell-based autologous disc cell therapy, 333 integrin- and DDR-dependent signaling, 88
Cell culture integrin binding, 88
DRG and neuronal, 310 self-assembly of, 8788
ex vivo organ culture, 354 structure
hypothesis testing, 354 amino acids, 8081
organ culture (see Organ culture models) lamellae, 81
sand rat annulus cells, 333334 organization, 80, 81
therapeutic screening, 354 triple helix, 81
Cervical TDA devices synthesis, 82
Bryan cervical disc, 235236, 238, 239 tenascin, 85
Prestige ST, 234235, 237, 238 Collagen I and IX, 346
ProDisc-C, 233235 Compression loading models
Charite, 27 dynamic, 361, 362
Chemical exchange saturation transfer (CEST), 206207 static/diurnal, 358359
Chemonucleolysis, 269 Computed tomography (CT) scans
Chondroadherin, 70 chordoma, 280281
Chondroitin sulfate (CS), 59 degenerative disc disease, 202203
Chordoma intervertebral disc herniation, 217
brachyury, 281, 282, 286 Computer-controlled organ culture loading devices, 361, 362
cauda equina, 279280 Continuous cyclic stretch (CCS), 119
chemotherapy, 284 Cre recombinase-loxP (Cre-loxP) system, 344345
chondroid tumors, 281
clinical features, 278279
dedifferentiated tumors, 281 D
description, 277 Danforths short tail (Sd), 36, 347348
differential diagnosis, 282 Decorin, 6869
epidemiology, 278 Degenerative disc disease (DDD), 342
F-18 PET scans, 281 compositional changes, 202
genetics, 286 early and late stage, 202
Index 439
growth factors and inflammatory stimulation, 202 low oxygen tension results, 408
imaging modalities, 201202 matrix changes, 180181
computed tomography, 202203 matrix degradation, 181
MRI (see Magnetic resonance imaging (MRI)) mechanical load, 190191
radiographic imaging, 202 metabolites accumulation, 264
mechanical changes, 202 morphological features of, 179
Disc arthroplasty devices, history and evolution of, 227228. novel therapeutic agents, 191192
See also Total disc arthroplasty (TDA) nutrition and oxygen tension
Disc degeneration cartilaginous end-plate calcification, 186
alterations in cell biology cigarette smoke, 186
anabolic growth factors, 185186 decreased pH, 187
cellular function regulators, 183 energy-generating pathway, 186
interleukin-1, 183185 glucose limitation, 187
TNF-a, 185 oxygen levels, 186187
anabolic and catabolic activities, imbalance of, solute size and charge, 186
401402 sand rat model (see Sand rat model)
animal models, 295296, 353 secondary changes of
canine model, 297299 annular tears/fissures formation, 267
caprine model, 298, 299 height reduction, 266
clinical perspective, 292, 294 spontaneous behavior analysis, 267268
considerations, 342 spine studies, 328
genetically modified mouse, 342349 stem cell transplantation therapy for, 378
history, 292 strategies, 268269
nonhuman primates, 299300 using stab technique, 328
ovine model, 297299 vascular and nerve ingrowth
porcine models, 295, 297, 298 neovascularisation and innervation, 182
reliability, 294 nerve fibres, 182
selection of, 300 NGF and BDNF expression, 182183
in silico models, 292 pleiotrophin, 182
standardization, 294 semaphorins, 182
validity and fidelity, 294295 Disc herniation
in vitro models, 292 animal models
in vivo research, 292 categories, 308309
biologic treatments, efficacy of, 404405 chemical and inflammatory mediators,
cell ageing and death 313315
apoptosis, 187188 dysesthesia, 309, 314
autophagy, 188189 limb hypersensitivity, 309
cell clustering and proliferation, 189 mechanical allodynia, 309, 312
cellular phenotype, 190 mechanical factors, 310312
matrix degradation, 187 quantitative gait analysis, 309
replicative senescence, 190 clinical presentation of, 217
challenges, 291 dorsal root ganglion and neuronal cell
chondrolytic enzyme, injection of, 269 culture, 310
controversies, 262 epidemiology, 216
definition, 361 evoked action potential, 307
description, 261262, 269270 imaging studies, 217218
dimethylmethylene blue assay, 405 lifestyle modifications, 306
evaluation, MRI technology, 379 mechanical loading, of lumbar spine, 306
fibrils disarrangement, 269 molecular mediators, of radiculopathy, 308
genetic factors, in humans, 342 nerve root impingement, 305, 306
genetic influences, 179180 nonsurgical care, 306
genetic studies (see Genetic studies) ODI, 307
growth factors pathoetiology of, 217
dominant-negative tumor necrosis factor, 407 pharmacological interventions, 306
IL-1ra, 406407 radicular pain, 306308
osteogenic protein-1, 402, 405, 406, 409, 410 radiculopathy (see Radiculopathy)
platelet-derived growth factor, 406 sciatica
platelet-rich plasma, 406 definition, 314
TGF-b signaling pathways, 405 historical information, 313
human discs, 292 incidence, 305, 306
injectable hydrogel treatment, 408 severity of symptoms, 305
injectable therapeutic agents spinal nerve/ganglion structure, 306, 307
in vitro effects, 402403 straight-leg raise test, 307308
in vivo effects, 402404 thermal/mechanical allodynia, 308
injection therapy, limitations of, 409410 treatment for, 218219
LBP and, 265266, 269 uncontained/extruded herniation, 319
lifetime prevalence, 417 VAS, 307
440 Index
Discogenic low back pain immune privilege, 390391

clinical presentation, 214215 reporter genes for, 386389
epidemiology, 214 Gene therapy
imaging efficiency and safety, 385
AP and lateral radiographs, 215 human genetic disease treatment, 385
discography, 215, 216 for intervertebral disc repair
MRI, 215 safety and doseresponse studies, 397
pathogenesis of, 214 summary table, 394396
treatment of in orthopedics, 386
medication, 215216 viral vectors, 386
physical therapy, 216 Genetically modified mouse
surgical options, 216 cDNA sequence, 342
Discoidin domain receptors (DDRs), 88 gene targeting, 343, 344
Disc-triggered pain, 264265 global gene knockout
Dissection procedures, in organ culture models extracellular matrix proteins, 346347
clotted blood removal, 360 factors affecting disc development, 347348
endplate cleaning, 360 global knockout approach, 343
for explants structural integrity, 360 homologous recombination, 343
fluorophore diffusion, 359 limitations, 342
free-swelling culture conditions, 359, 360 spontaneous gene mutations, impaired disc function/structure,
nucleus pulposus viability, 359 345346
vertebral endplates, 360, 361 tissue-specific gene knockouts, 348349
Diurnal compression loading models, 358359 tissue-specific knockout models, 343345
Dominance, 158 transgenic approach, 342343
Dominant-negative form of TonEBP (DN-TonEBP), 101 Genetic studies
Dynamic compression loading models, 361, 362 aging process, 161
Dysesthesia, 309, 314 candidate genes, 162
case-control association studies, 162
cell function and survival, 167
E DNA methylation, 170171
Elastin, 85 epigenetics, 170171
Endplate cartilage, 9 extracellular matrix proteins
Epidural corticosteroid injections, 254 aggrecan, 163164
Epigenome-wide association studies (EWAS), asporin, 165
158, 171 candidate gene selection, 163
Exercise therapy, 249250 cartilage intermediate layer protein, 165
collagen I and IX, 164
extracellular matrix, 163
F matrilins, 165166
Facet joint injections, 255 thrombospondin-2, 166
F-actin, 116 Trp2 and Trp3 alleles, 164165
Fas ligand (FasL), 390391 family linkage analysis, 162
Fibril-associated collagen with interrupted triple helices (FACIT) future investigations, 171173
collagen, 84 heritability test, 160
Fibromodulin human genetic studies
cleavage, MMP-13, 70 complex diseases, 159
immunolocalization, 70 Mendelian genetics, 158159
interactions, 69 large-scale GWAS, 169170
structure, 69 matrix metalloproteinases, 166
Fibronectin, 84 microRNAs, 171
Fluid-induced shear stresses, 120 phenotypic parameters
Foxa2-Cre recombination activity, 349 disc degeneration, 160
MRI images, 161
radiographs, 160161
G variability of phenotypic parameters, 161
Gene delivery studies proinflammatory cytokines, 165166
anti-inflammatory/anti-degeneration genes risk factors, 167168
ADAMTS5 siRNA, 390 technologies in
IL-1Ra, 390 genome-wide association study, 169
TIMP-1, 389 International HapMap Project, 168169
Fas-FasL interactions, 390391 next-generation sequencing, 169
growth and transcription factors SNP genotyping, 169
animal models, 394 Gene-trap mutagenesis, 347
bone morphogenetic proteins, 392394 Genome-wide association study (GWAS), 158
GDF-5, 392 Global gene knockout
Sox-9, 393 extracellular matrix proteins
transforming growth factor-b signaling, 393 biglycans, 346347
Index 441
collagen I and IX, 346 mechanics

factors affecting disc development compression, 2425
gene-trap mutagenesis, 347 degeneration, 2627
Sd and Skt, 347348 flexion/extension, 25
Sox5 and Sox6, 347 lateral bending, 25
Global knockout approach, for genetically modified mouse, 343 stress and strain, 2324
Glucosamine, 59 torsion, 2526
Glycosaminoglycan nerve endings/axions, 264
chondroitin sulfate/dermatan sulfate, 55 normal and degenerated, innervation of, 262265
heparan sulfate/heparin, 55 spinal curvature
historical perspective of, 5455 axial curvature, 67
keratan sulfate, 55 hunchback spine, 7
Green fluorescence protein (GFP), 386388, 394, 397 lordosis, 7
Ground reaction force, 314 natural curvature, 6
Growth and differentiation factor-5 (GDF-5), 167, 346, 392 scoliosis, 78
somitogenesis, 8
tissues, 9
H annulus fibrosus, 18, 2123, 418419
HA-pNIPAM hydrogel, 366 cartilaginous endplates, 18, 23, 419
Heparan sulfate proteoglycans (HSPGs), 348 nucleus pulposus, 1821, 419
Herniation, 2728. See also Disc herniation osmotic effects, 19
5-HT(2A) blocker treatment, for disc herniation, 319 vertebrae, 17
Human cadaveric tissue, 354 zygapophysial joint, 17
Human organ culture models, 366367 Intervertebral disc development
Hyaluronan synthase-2 (Has2), 348 developmental engineering, 3435
Hyperalgesia, 314 endochondral bone formation, 34
Hypoxia-inducible factor (HIF) notochord and nucleus pulposus
cell survival and function, 99100 brachyury protein, 3536
energy conservation, 9899 Danforths short-tail (Sd) mutation, 36
hypoxia, 9596 lineage, 3738
Sox family, 95 mice and humans, 35
stability control, 9698 sheath, 3637
target genes, 95, 96 Sickle tail gene (Skt) mutation, 36
transactivation, 94 postnatal disc, 46
sclerotome, 33
vs. annulus fibrosus, 4445
I derivatives, 4243
IL-1 receptor antagonist (IL-1Ra), gene delivery studies, 390 formation, 42
Imaging modalities resegmentation, 42
computed tomography, 202203 syndetome, 4546
MRI (see Magnetic resonance imaging (MRI)) vs. vertebral cartilage, 45
radiographic imaging, 202 somitogenesis
Immune privilege (IP), 390391 clock regulation, 3940
Immunohistochemical staining, of chordoma, 281 epithelialization, 41
Injection therapy, back pain Klippel-Feil syndrome, 3839
epidural corticosteroid injections, 254 presomitic mesoderm, 38
facet joint injections, 255 segment identity, 4142
sacroiliac joint injections, 255 wavefront, 4041
soft tissue injections, 254255 Intervertebral disc niche, 354
Insulin-like growth factor 1 receptor (IGF1R), 167 cell biology, 94
International HapMap Project, 168169 in disc cell renewal, 104
International Society for the Study of the Lumbar Spine elevated osmolarity, 94
(ISSLS), 141 HIF proteins
Intervertebral disc cell survival and function, 99100
anatomical description, 341 energy conservation, 9899
in animals, 1213 hypoxia, 9596
biochemical and structural changes, 18 Sox family, 95
degeneration of, 265 stability control, 9698
development, 5 target genes, 95, 96
evolutionary considerations, 34 transactivation, 94
form and function of, 56 hypoxic nature, 94
functional role, 8 matrix synthesis control, nucleus pulposus
gross morphology and dimensions, 8 CCN2 expression, 103104
herniation (see Disc herniation) GAG synthesis, 103
mechanically induced injury proteoglycan synthesis, 103
endplate fracture, 28 microangiographical and immunohistochemical studies, 94
herniation, 2728 nucleus pulposus cell survival, 9394
442 Index
Intervertebral disc niche (cont.) Lumbar TDA devices

tonicity enhancer-binding protein instantaneous axis of rotation, 233
in annulus fibrosus, 101, 102 ProDisc-L, 233, 234
AQP2 expression, 102 SB Charit III, 232233
DN-TonEBP, 101 Lumican, 68, 70
in nucleus pulposus, 101103
osmoregulation, 101
M
Magnetic resonance imaging (MRI)
K biochemical-based MRI
Klippel-Feil syndrome (KFS), 3839 chemical exchange saturation transfer, 206207
magnetization transfer MRI, 205206
path to clinical utility, 209
L sodium MRI, 207209
LBP. See Lower back pain (LBP) T1r MRI, 204205
Linkage disequilibrium (LD), 158 UTE MRI, 207, 208
Lower back pain (LBP) chordoma, 280281
description, 261, 269270 discogenic low back pain, 215
and disc degeneration, 265266, 269 (see also Disc degeneration) drawback of, 204
Modic changes and, 265266 lumbar degenerative disease
strategies, 268269 intervertebral disc herniation, 217, 218
Lubricin lumbar stenosis, 220
gene organization and mutation, 66 spondylolisthesis, 222
in nucleus pulposus, 6465 phenotypic parameters, in genetic studies, 161
protein structure, 6566 sacral chordoma, 278, 280
superficial zone protein, 64 scanners, 203204
Luciferase reporter gene, 386, 387 spinal morphology
adenovector harboring, 394 disc degeneration, 203
bioluminescence imaging, 388 T2-weighted MR images, 203
osteocalcin promoter, 393 Magnet therapy, 256
siRNA, 389 Mechanical forces
Lumbar degenerative disease, 213 caveolae, 120
discogenic low back pain, 214216 cellular and molecular levels
intervertebral disc herniation, 216219 biaxial stretch, 118120
lumbar stenosis, 219220 cell response, 116117
spondylolisthesis, 220222 mechanical perturbation, 117
Lumbar disc degeneration mechanocoupling, 113115
discography, 143 mechanotransduction, 115
etiological factors pressure, 117118
age, 146147 signal transmission, 115116
behavioral and environmental factors, 145 fluid-induced shear stresses, 120
cigarette smoking, 149 glycocalyx, 120
constitutional risk factors, 145 nucleus pulposus and annulus fibrosus
familial aggregation, 150151 poroviscoelasticity, 111113
heritability, 151153 tissue homeostasis, 110
physical demands and spinal loading, 147148 spinal loading, 110
trauma, 149 Mechanical perturbation, 117
vertebral endplate, 149150 Mechanocoupling, 113115
whole-body vibration (WBV), 148149 Mendelian genetics, 158159
historical perspectives Mesenchymal stem cells (MSCs), 365366
degenerative disc disease, 140142 criteria, 373374
disc degeneration, 139140 differentiation of
macroscopic and histological observations, 142 physiological stimulation effects, 376
MR imaging, 143 scaffold properties effects, 376
plain radiography, 142143 drug delivery systems, use of, 380381
prevalence of human, 374
imaging approaches, 143144 markers, 374
mean scores, 144145 synovium-derived, 377
visual rating systems, 143 transplantation study, 377379
Lumbar posterior dynamic devices, 243 transplantation therapy, 379
Lumbar stenosis utilization of, 375, 377
clinical presentation of, 219 Mesenchymal to epithelial transition (MET), 41
epidemiology, 219 Microcomputerized tomography (Micro-CT), 13, 331
etiology of, 219 Micro-discectomies, 219, 365
imaging of, 219220 Modic changes and LBP, 265266
treatment, 220 Modulus, 18
Index 443
MSCs. See Mesenchymal stem cells (MSCs) sheath, 3637

Multipotent mesenchymal stromal cells, 373, 375 Sickle tail gene (Skt) mutation, 36
Muscle relaxants, 253 elastic behavior, 20
Mutation, 158 long fibers, 20
lubricin, 6465
mechanical properties, 20
N poroviscoelasticity (see Poroviscoelasticity)
Narcotic analgesics, 253 viscoelastic properties, 20
Nerve of Luschka, 62 Nucleus pulposus progenitor cells, spheroid colony, 380
Neurotrophins, 264 Nucleus pulposus tissue engineering
Next-generation sequencing (NGS), 169 cell therapy, 422423
NF-kB inhibitors, in disc herniation, 319 scaffolds, 423
N-Methyl-D-aspartic acid (NMDA) receptor, 314, 319 in vitro formation, 423424
Nonmechanical back pain, 248
Non-narcotic analgesics, 250
O
Nonoperative treatment, of back pain
ODI. See Oswestry Disability Index (ODI)
activity modification, 249
Opiate analgesics, 215
analgesic medications
Organ culture models
narcotic analgesics, 253
animal model selection, 356, 357
non-narcotic analgesics, 250
compression loading models
antidepressants, 253254
dynamic, 361, 362
behavioral therapy, 249
static/diurnal, 358359
exercise therapy, 249250
degeneration mechanism
injection therapy
altered media conditions, 362363
epidural corticosteroid injections, 254
endplate fractures, 364
facet joint injections, 255
mechanical and nutritional challenges, 363
sacroiliac joint injections, 255
needle injection, 365
soft tissue injections, 254255
surgical and chemical challenges, 364365
management options, 248, 249
disc regeneration and repair mechanisms, 365
muscle relaxants, 253
disc swelling, 357358
NSAIDs, 253
dissection procedures
oral corticosteroids, 253
clotted blood removal, 360
patient education, 248249
endplate cleaning, 360
physical treatment modality
for explants structural integrity, 360
acupuncture, 255256
fluorophore diffusion, 359
cold packs, 256
free-swelling culture conditions, 359, 360
magnet therapy, 256
nucleus pulposus viability, 359
orthoses, 256
vertebral endplates, 360, 361
spinal manipulation, 255
free-swelling, of disc explants, 357, 358
superficial heat, 256
human, 366367
traction, 255
limitations, 367368
transcutaneous electrical nerve stimulation, 256
osmotic loading models, 358
topical treatments, 254
publication history, 355
Nonsteroidal anti-inflammatory drugs (NSAIDs), 253
therapeutic agents, screening of, 365
Nontraditional species, sand rat model
Osmotic effects, 19
age and body weight demographics, 336, 337
Osteocalcin, 393
blood testing, 337
Oswestry Disability Index (ODI), 226, 307, 314
breeding, 337
Ovine model, 297299
colony monitoring challenges, 337
litter sizes, 337
low-calorie diet essentials, 337 P
periodic teeth trimming, 337 Pain. See also Back pain
restricted personnel access, 335 activation, 269
separate housing room maintenance, 335 basis of, 262265
NSAIDs. See Nonsteroidal anti-inflammatory drugs disc-triggered, 264265
(NSAIDs) Pax-1 transcription factor, 345
Nucleus pulposus, 9 Perlecan
aggrecan, 19 degradation, 64
collagen II, 20 extracellular matrix production, 62
collagens in, 80, 81 (see also Collagen) fibrillin and elastin assembly, 62
degeneration, 268269 gene organization and mutation, 64
development growth factors and morphogens, 61
brachyury protein, 3536 immunolocalization, 62
Danforths short-tail (Sd) mutation, 36 LTBP-2, 62
lineage, 3738 neonatal and adult disc, 61
mice and humans, 35 protein structure, 63, 65
444 Index
Phenotype, 158 fetal intervertebral disc, 60

Physical therapy, lumbar degenerative disease gene organization and mutation, 60
discogenic low back pain, 216 mature intervertebral disc, 60
lumbar stenosis, 220 protein structure and function, 60
Physical treatment modality, for nonoperative back pain Psammomys obesus. See Sand rat model
acupuncture, 255256
cold packs, 256
magnet therapy, 256 R
orthoses, 256 Radiculopathy
spinal manipulation, 255 animal model, 309, 312
superficial heat, 256 cell cycle modifiers, 319
traction, 255 compounds for treatment, 315317
transcutaneous electrical nerve stimulation, 256 cytokine antagonism, 317318
Physiological stimulation effects, MSC differentiation, 376 definition, 314
Pintail mouse model, 345 5-HT(2A) blockers, 319
Platelet-derived growth factor (PDGF), 406 molecular mediators, 308
Platelet-rich plasma (PRP), 406 neuronal receptor blockers, 318319
Pleiotrophin, 182 NF-kB inhibitors, 319
Poissons ratio, 18 nucleus pulposus-induced radiculopathy, 315
Porcine models, 295, 297, 298 tumor necrosis factor antagonism, 317318
Poroviscoelasticity Radiologic characterizations, sand rat model
nucleus pulposus and annulus fibrosus cervical spine, 332
collagen microstructure, 112113 lumbar spine, 330332
GAG degradation, 112 Real-time quantitative polymerase chain reaction (RT-PCR), 375, 390
internal pressure and shear, 113 Receptor tyrosine kinase (RTK) inhibition, in chordoma, 284285
interstitial fluid, 111 Recessive, 158
swelling properties, 111112 Regenerative medicine
time load history and age-related changes, 113 cell requirements, 420
Prestige ST cervical disc arthroplasty device, 234235, 237, 238 description, 418
Procollagens, extracellular processing of, 87 Reporter genes, 386
ProDisc-C cervical disc arthroplasty device, 233235 in vitro studies
ProDisc-L lumbar disc arthroplasty device, 233, 234 animal cells, 386387
Prosthetic nuclear replacement device, 243 human cells, 387
Proteinases in vivo studies, 387389
ADAMTS (see A disintegrin-like and metalloproteinase with Restriction fragment length polymorphism (RFLP), 158
thrombospondin type 1 motif (ADAMTS)) RT-PCR. See Real-time quantitative polymerase chain reaction
biological significance, 125126 (RT-PCR)
features of, 125126
matrix metalloproteinases, 126
Proteoglycans S
aggrecan (see Aggrecan) Sacral chordoma
glycosaminoglycan MRI, 278, 280
chondroitin sulfate/dermatan sulfate, 55 wide surgical resection, 282
heparan sulfate/heparin, 55 Sacroiliac joint injections, 255
historical perspective of, 5455 Sand rat model
keratan sulfate, 55 advantages, 335
historical perspective of, 5455 annulus fibrosus cell culture, 333334
lubricin autologous disc cell therapy, 333
gene organization and mutation, 66 disadvantages, 335
in nucleus pulposus, 6465 disc cell death, 332333
protein structure, 6566 discovery and maintenance, 337
superficial zone protein, 64 end plate bone mineral density, 332333
perlecan genome sequencing research, 337
degradation, 64 morphologic changes, with aging, 328330
extracellular matrix production, 62 nontraditional species
fibrillin and elastin assembly, 62 age and body weight demographics, 336, 337
gene organization and mutation, 64 blood testing, 337
growth factors and morphogens, 61 breeding, 337
immunolocalization, 62 colony monitoring challenges, 337
LTBP-2, 62 litter sizes, 337
neonatal and adult disc, 61 low-calorie diet essentials, 337
protein structure, 63, 65 periodic teeth trimming, 337
small leucine-rich repeat proteoglycan (see Small leucine-rich restricted personnel access, 335
repeat proteoglycans (SLRPS)) separate housing room maintenance, 335
versican nutritionally induced type 2 diabetes research, 335
vs. aggrecan, 5960 radiologic abnormalities, incidence of, 330
degradation, 61 radiologic characterizations
Index 445
cervical spine, 332 intervertebral disclike cell, 375376

lumbar spine, 330332 of musculoskeletal region, 374375
reliable species, 327 Stiffness, 18
spine fusion model, 333 Straight-leg raise (SLR) test, 307308
spin-lock (T1r) imaging, 332 Strain, 18
vascular supply, 333 Stress, 18
SB Charit III lumbar disc arthroplasty device, 232233 Stress-induced premature senescence (SIPS), 190
Scaffolds Syndetome, 4546
annulus fibrosus tissue engineering, 425
MSC differentiation, 376
nucleus pulposus tissue engineering, 423 T
Schober index, 314 TDA. See Total disc arthroplasty (TDA)
Sciatica, 314 Tenascin, 85
Semaphorins, 182, 264 Tgfb receptor 2, 348
Serotonin reuptake inhibitors (SSRIs), 253 Tie2/angiopoietin-1 signaling, 380
Sickle tail (Skt) , 36, 167, 347348 Tissue engineering
Single nucleotide polymorphism (SNP), 158 annulus cells, 422
Small leucine-rich repeat proteoglycans (SLRPS) biomechanical principles, 420
asporin, 69 challenges, 427428
biglycan, 69 chondrocytes, 421
chondroadherin, 70 collagen gels, 425426
decorin, 6869 disc functions, 419420
fibromodulin embryonic stem cells, 421
cleavage, MMP-13, 70 goals, 418
immunolocalization, 70 growth factors, 422
interactions, 69 inducible progenitor cells, 421
structure, 69 intervertebral disc tissue, schematic representation of, 426
knockout mice and gene mutation, 7071 MSCs, 421
lumican, 68, 70 notochordal cells, 420
PRELP, 68, 70 nucleus pulposus cells, 420
roles of, 68 polyglycolic/polylactic acid polymers, 425
structural organization, 67 scaffolds, 422
Sodium MRI, 207209 stem cell transplantation, 377379
Soft tissue injections, 254255 and vertebral body graft transplants, 425
Somitogenesis Tissue inhibitors of metalloproteinases (TIMP), 88, 181
clock regulation, 3940 Tissue inhibitors of metalloproteinases-1 (TIMP-1), 389
epithelialization, 41 Tissue-specific gene knockouts
Klippel-Feil syndrome, 3839 extracellular matrix proteins, 348
presomitic mesoderm, 38 factors affecting disc development
segment identity, 4142 b-catenin-dependent Wnt signaling, 348
wavefront, 4041 Foxa2-Cre recombination activity, 349
Sonic hedgehog, 348349 sonic hedgehog, 348349
Sox5 and Sox6, 347 Tgfb receptor 2, 348
Spheroid colony, 380 TGF-b signaling, 348
Spinal arthrodesis, 225 Tissue-specific knockout models
Spinal device registry, 236237 Cre-loxP system, 344345
Spinal fusion, 225227 gene targeting, 343344
Spinal manipulation therapy, 255 Tonicity enhancer-binding protein (TonEBP)
Spinal motion restoration devices. See Total disc arthroplasty in annulus fibrosus, 101, 102
(TDA) AQP2 expression, 102
Spine Patient Outcomes Research Trial (SPORT), 218 DN-TonEBP, 101
Spin-lock (T1r) imaging, 332 in nucleus pulposus, 101103
Spondylolisthesis osmoregulation, 101
clinical presentation of, 221 Total disc arthroplasty (TDA)
epidemiology, 220221 bearing surface requirements, 230
etiology of, 221 biomechanics
imaging of, 221 functional spinal unit, 228229
treatment, 221222 objectives, 229
SSRIs. See Serotonin reuptake inhibitors (SSRIs) cervical devices (see Cervical TDA devices)
Static compression loading models, 358359 clinical success, 226
Stem cells, 365366 cobalt chrome and titanium alloys, 230
biology evolution, 237
IVD disease treatment, role of, 379381 facet replacements, 243
in musculoskeletal tissues, evolution of, 373374 failures, 241242
criteria, 374 Food and Drug Administration, 232
definition, 374 implant fixation, 231
disc cell differentiation, 376377 indications
446 Index
Total disc arthroplasty (TDA) (cont.) VAS. See Visual analogue scale (VAS)
cervical TDA, 240241 Versican
lumbar TDA, 240, 241 vs. aggrecan, 5960
investigational device exemption, 232 degradation, 61
lumbar devices (see Lumbar TDA devices) fetal intervertebral disc, 60
lumbar posterior dynamic devices, 243 gene organization and mutation, 60
metal-on-metal articulation, 230231 mature intervertebral disc, 60
510(k) notifications, 232 protein structure and function, 60
operative technique Vertebrae
anterior retroperitoneal dissection, in lumbar spine, 238 in animals, 1314
implant positioning/sizing, 240 development, 45
posterior longitudinal ligament release, 240 evolutionary considerations, 34
preoperative planning, 238 structure
Smith-Robinson anterior approach, in cervical spine, 238 cervical vertebrae, 1011
polyethylene and polyurethane, 230 lumbar spine, 1112
premarket approval, 232 sacrum and coccygeal bones, 12
prostheses, material properties of, 229 thoracic vertebrae, 11
prosthetic nuclear replacement device, 243 uncovertebral joints, 10
revision strategies, for failed TDA vertebral foramen, 10
indications, 242 zygapophyseal (facet) joints, 10
long-term follow-up, 243 Viscoelasticity, 18
surgical options, 242 Visual analogue scale (VAS), 307, 314
symptomatic/asymptomatic patients, 242 Vitamin D receptor (VDR), 167
semiconstrained vs. unconstrained devices, 231 von Frey microfilaments, 309, 312
spinal motion-preserving devices, 243
X-stop, 243
Total disc replacement (TDR), 27 W
Transcutaneous electrical nerve stimulation therapy, 256 Water saturation shift referencing (WASSR), 206, 207
Transgenic approach, for genetically modified mouse, 342343 WDS. See Wet-dog shakes (WDS)
Tricyclic antidepressants (TCAs), 253254 Weill-Marchesani syndrome (WMS), 129130
Tumor necrosis factor (TNF) antagonism, 317318 Wet-dog shake motions, 309, 314
Wet-dog shakes (WDS), 268
Whole-body vibration (WBV), 148149
U
Ultrashort time-to-echo (UTE) MRI, 207, 208, 410
X
X-stop, 243
V
Vacuolated physaliferous cells, 281
Variable number of tandem repeat (VNTR), 158

The Intervertebral Disc

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Intervertebral Disc

Uploaded by

Copyright:

Available Formats

Irving M.

Molecular and Structural

The Intervertebral Disc

ISBN 978-3-7091-1534-3 ISBN 978-3-7091-1535-0 (eBook)

Library of Congress Control Number: 2013943012

Springer-Verlag Wien 2014

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Current understandings of the biology of the intervertebral disc is predicated on comprehen-

Richard H. Rothman, MD, PhD

March 2013 Irving M. Shapiro

Part I Biomechanical and Molecular Studies of the Intervertebral Disc

1 Introduction to the Structure, Function, and Comparative Anatomy

Part II Intervertebral Disc Disease: Pathogenesis and Current

9 Epidemiology of Lumbar Disc Degeneration . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

14 Spinal Motion Restoration Devices for the Degenerative Disc . . . . . . . . . . . . . 225

Part III Models of Disc Disease and Biological Regeneration

18 Large Animal Models of Disc Degeneration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291

About the Editors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435

Irving M. Shapiro and Makarand V. Risbud

Contents 1.1 Introduction

I.M. Shapiro, M.V. Risbud (eds.), The Intervertebral Disc, 3

are to allow movement of the individual vertebrae, transmit

diarthrodial joints that contain cartilage and fibrocartilage a c

1.3 Vertebral Structure

1.3.3 Lumbar Spine (Fig. 1.2)

vertebrae are the most massive of all: in most cases being

While considerable space is devoted to the sand rat (see

Table 1.1 Vertebral formula for animals and man

processes of C3 to T9/10, and the rhomboids which join the

Contents 2.1 Introduction

I.M. Shapiro, M.V. Risbud (eds.), The Intervertebral Disc, 17

endplates. Each of these tissues has a characteristic

2.2 StructureFunction and proteoglycans form a charged, porous, deformable solid

where R is the universal gas constant, T is the absolute tem-

with their special mechanical properties. For example, it is

provides the annulus with a series of important mechanical

Table 2.3 Shear modulus (kPa) Orientation

2.2.4 Cartilaginous Endplate

an applied torque of 7.5 Nm (Heuer et al. 2008). From in vitro

2.4 Effect of Degeneration

to a reduction of the disc height and an increase in insta-

Herniation is characterized by the prolapse of the nucleus

Contents 3.1 Overview

I.M. Shapiro, M.V. Risbud (eds.), The Intervertebral Disc, 33

The notochordal plate buds off of the endoderm to finally

expressed in the cartilage of the presumptive vertebrae. As a

occur via a clock and wavefront mechanism (Fig. 3.2).

Fig. 3.2 Somitogenesis RA

S-I S-I S-I

Somite and somitocoel Differentiation of vertebrae, Final structure

Somite Vertebral body

Notochord Dermomyotome Vertebral body

Pre-sclerotome Intervertebral disk Muscle

Pre-dermomyotome Syndetome Intervertebral disk

Contents 4.1 Introduction

I.M. Shapiro, M.V. Risbud (eds.), The Intervertebral Disc, 53

properties of aggrecan are related to its abundance, degree of

Versican b IAF e IAF h

a one fragment possessing a G1 region that remains bound to

4.3.2 Aggrecan Gene Organization,

Fig. 4.3 Schematic depiction G1 G2 G3

CS side chain KS side chain Core protein Disulphide stabilised loop