Biochemical, Genetic, and Molecular Interactions in Development - part 1 docx

The Skeleton
Biochemical, Genetic, and Molecular
Interactions in Development
and Homeostasis
Edited by
Edward J. Massaro
John M. Rogers
Edited by
Edward J. Massaro
John M. Rogers
T
HE
S
KELETON
T
HE
S
KELETON
B
IOCHEMICAL
, G
ENETIC
,
AND
M
OLECULAR
I
NTERACTIONS
IN
D

EVELOPMENT AND
H
OMEOSTASIS
HUMANA PRESS
TOTOWA, NEW JERSEY
Edited by
EDWARD J. MASSARO
and
JOHN M. ROGERS
Developmental Biology Branch,
Reproductive Toxicology Division,
The National Health and Environmental Effects
Research Laboratory, Office of Research Development,
United States Environmental Protection Agency,
Research Triangle Park, NC
© 2004 Humana Press Inc.
999 Riverview Drive, Suite 208
Totowa, New Jersey 07512
www.humanapress.com
All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or
by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise without written permission
from the Publisher.
The content and opinions expressed in this book are the sole work of the authors and editors, who have warranted due
diligence in the creation and issuance of their work. The publisher, editors, and authors are not responsible for errors
or omissions or for any consequences arising from the information or opinions presented in this book and make no
warranty, express or implied, with respect to its contents.
Production Editor: Jessica Jannicelli.
Cover design by Patricia F. Cleary.
Cover Illustration: Artwork supplied by József Zakany.
Due diligence has been taken by the publishers, editors, and authors of this book to ensure the accuracy of the information

published and to describe generally accepted practices. The contributors herein have carefully checked to ensure that
the drug selections and dosages set forth in this text are accurate in accord with the standards accepted at the time of
publication. Notwithstanding, as new research, changes in government regulations, and knowledge from clinical expe-
rience relating to drug therapy and drug reactions constantly occurs, the reader is advised to check the product information
provided by the manufacturer of each drug for any change in dosages or for additional warnings and contraindications.
This is of utmost importance when the recommended drug herein is a new or infrequently used drug. It is the respon-
sibility of the health care provider to ascertain the Food and Drug Administration status of each drug or device used in
their clinical practice. The publisher, editors, and authors are not responsible for errors or omissions or for any conse-
quences from the application of the information presented in this book and make no warranty, express or implied, with
respect to the contents in this publication.
For additional copies, pricing for bulk purchases, and/or information about other Humana titles, contact Humana at the above
address or at any of the following numbers: Tel.: 973-256-1699; Fax: 973-256-8341; E-mail: or
visit our website:
This publication is printed on acid-free paper. '
ANSI Z39.48-1984 (American National Standards Institute) Permanence of Paper for Printed Library Materials.
Photocopy Authorization Policy:
Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted
by Humana Press Inc., provided that the base fee of US $25.00 per copy is paid directly to the Copyright Clearance Center
at 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license from
the CCC, a separate system of payment has been arranged and is acceptable to Humana Press Inc. The fee code for users
of the Transactional Reporting Service is: [1-58829-215-0/04 $25.00].
Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1
Library of Congress Cataloging-in-Publication Data
E-ISBN: 1-59259-736-X
The skeleton : biochemical, genetic, and molecular interactions in
development and homeostasis / edited by Edward J. Massaro and John M.
Rogers.
p. ; cm.
Includes bibliographical references and index.
ISBN 1-58829-215-0 (alk. paper)

1. Bone. 2. Bones Physiology. 3. Bones Growth.
[DNLM: 1. Bone and Bones embryology. 2. Bone Development. 3.
Molecular Biology. WE 200 S6276 2004] I. Massaro, Edward J. II.
Rogers, John M.
QP88.2.S54 2004
612.7'5 dc22
2003027149
PREFACE
v
The skeleton is a complex multifunctional system. In addition to its mechanical/
structural support function, it contains the marrow in which blood cells are made and,
therefore, is a critical part of the circulatory and immune systems. Also, in that it is the
major reservoir for the essential element calcium, a critical component of intracellular
signaling pathways, the skeleton is an integral component of the endocrine system.
Furthermore, the skeleton is a dynamic system that is subject to modification (remodel-
ing) throughout life under the influence of both intrinsic (chemical) signals and extrinsic
(mechanical) signals. Therefore, it is axiomatic that properly regulated crosstalk between
the biochemistry and physiology of the skeleton and the chemical biology of the organism
is of critical importance both in the complex processes of development and the mainte-
nance of physiologic homeostasis. Thus, gaining insight in the nature and regulation of
these interactions is of considerable interest to researchers and clinicians in a broad
spectrum of biomedical disciplines.
Bone is formed during embryonic life and grows (formation exceeds resorption) rap-
idly through childhood. In humans, growth peaks around 20 yr of age. Thereafter, the
skeleton enters a prolonged period (lasting approx 40 yr) when bone mass remains rela-
tively stable. During this period, resorption and reformation (remodeling) of both cortical
and trabecular bone occur continuously and contemporaneously, resulting in an annual
turnover of approx 10% of the adult skeleton with essentially no net effect on bone mass.
The maintenance of skeletal mass is regulated through a balance between the activity of
cells that resorb bone (osteoclasts) and those that form bone (osteoblasts). Unfortunately,

the balance between resorption and formation degenerates with age and, if uncompen-
sated, can have debilitating consequences. For women, the balance terminates at meno-
pause. Bone loss also occurs in men, but usually later in life. Clinical disorders in which
bone resorption exceeds formation are common and include osteoporosis, Paget’s dis-
ease of bone, and bone wasting secondary to such cancers as myeloma and metastatic
breast cancer. Osteoporosis is the most common bone resorption disorder. It affects one
in three women after the fifth decade of life. The pathophysiology of this condition
includes genetic predisposition and alteration of systemic and local hormone levels
coupled with environmental influences. Treatment is based on drugs that inhibit bone
resorption either directly or indirectly: bisphosphonates, calcitonin, estrogens, and syn-
thetic estrogen-related compounds (SERMs—selective estrogen receptor modulators).
The search for more effective anti-osteoporosis drugs with fewer side effects continues.
In this regard, it is of both great interest and potentially enormous import to note that
recent evidence indicates that low bone mineral density (BMD) appears to protect women
over the age of 65 from primary breast cancer. It was reported that women in the highest
BMD quartile have approximately three times the risk of developing bone cancer than
those in the lowest quartile. Also, those with the highest BMD, obtained from measure-
ments of the wrist, forearm and heel, have almost six times the risk of advanced disease.
Less prevalent than disorders of bone loss are clinical disorders of reduced bone
resorption, such as osteopetrosis, and pycnodysostosis (owing to cathepsin K deficiency),
that are the consequence of genetic defects. Unfortunately, progress in the search for
effective treatments for these orphan diseases often is stymied by lack of support.
vi Preface
Significant insight into many aspects of vertebrate skeletal development has been
obtained through molecular and genetic studies of animal models and humans with
inherited disorders of skeletal morphogenesis, organogenesis, and growth. Morphogen-
esis, the developmental process of pattern formation and the establishment of the body
plan that is the template for the architecture of the adult form, is an exquisitely compli-
cated program. Our understanding of it contains many gaps. The information for the
pattern and form of the vertebrate skeleton emanates from mesenchymal cells during

embryonic development. Morphogenesis requires three key ingredients: inductive sig-
nals, responding stem cells and a supportive extracellular matrix. Within the vertebrate
morphogenetic program, skeletal development is controlled by sequence-dependent
activation/inactivation of specific genes that results in the distribution of cells from
cranial neural crest, sclerotomes, and lateral plate mesoderm into a pattern of mesenchy-
mal condensations at sites in which skeletal elements will develop. Condensation is the
earliest stage of organ formation at which tissue-specific genes are upregulated. It is
generated through interactions between molecules in the extracellular matrix such as the
cell adhesion molecules fibronectin, N-CAM and N-cadherin. Cell adhesion also is
mediated, albeit indirectly, via activation of particular CAM genes by the products of the
Hox genes, Hoxa-2 and Hoxd-13. Cells proliferate and differentiate, under the control of
transcription factors, into chondrocytes or osteoblasts forming, respectively, cartilage or
bone. Proliferation within the condensations is mediated through the activation of cell
surface receptors such as syndecan-3, a receptor for fibroblast growth factor 2 (FGF-2),
the antiadhesive matrix component, tenascin-C, a ligand for the epidermal growth factor
(EGF) receptor (EGFR), the Hox genes, Hoxd-11-13 and transcription factors such as
CFKH-1, MFH-1, and osf-2. Growth of condensations is regulated by BMPs, which
activate a number of genes including Pax-2, Hoxa-2, and Hoxd-11. Conversely, growth
is blocked via inhibition of BMP signaling by the BMP antagonist, Noggin. Defects in
the formation of specific bones and joints can occur through mutation of genes involved
in the control of bone and joint development. Information derived from ongoing and
future research focused on the identification of the genes/gene targets involved in skeletal
development and maintenance should open new avenues for the development of thera-
peutic measures for treating defects resulting either from mutation or trauma.
For most of the skeleton, bones develop from cartilage models comprised of assem-
blies of chondrocytes in an extracellular collagen-containing matrix that they secrete.
The replacement of cartilage by bone is the result of a genetic master program that
controls and coordinates chondrocyte differentiation, matrix alteration and mineraliza-
tion. During the conversion of the cartilage model into bone, the composition of the
matrix, including collagen types, is modified, ultimately becoming mineralized through

a process termed endochondral ossification and populated by osteocytes. Disruption of
the rate, timing, or duration of chondrocyte proliferation and differentiation results in
shortened, misshapen skeletal elements. In the majority of such disruptions, vasculariza-
tion also is perturbed. It has been proposed that vascularization plays a key role in the
synchronization of the processes involved in endochondral ossification. Bone formation
also occurs via intramembranous ossification, in which bone cells arise directly from
mesenchyme without an intermediate cartilage anlage. Data indicate that this process is
the result both of a positive selection for osteogenic differentiation and a negative selec-
tion against the progressive growth of chondrogenic cells in the absence of a permissive
Preface vii
or inductive environment. In any case, through the processes of bone growth and remod-
eling, an adult skeleton is shaped and molded and continually remolded in response to
environmental alterations. In effect, the adult skeleton is not a static entity. Bone is
metabolically active throughout life and, under the influence of mechanical stress, nutri-
tion, and hormones, bone remodeling occurs continually. However, bone remodeling is
compromised as a function of both post menopausal hormonal changes and aging, result-
ing in health problems of increasing magnitude as the proportion of the aged in the global
population increases.
Mutations in genes encoding structural proteins of the extracellular matrix can perturb
the coordination of events necessary for normal skeletal development. The magnitude of
the disruption of the process of ordered skeletal development is dependent on both the
role of the mutated gene product in the developmental process and the degree of its
functional perturbation. The range of mutational consequences is broad, including dis-
ruption of ossification/mineralization and linear growth and the structural integrity and
stability of articular cartilage. Evidence indicates that osteochondrodysplasias resulting
from defects in structural proteins are inherited in an autosomal dominant manner and
that a spectrum of related clinical phenotypes can be produced by different mutations in
the same gene. In addition, as might be expected, haploinsufficiency of a gene product
usually produces a milder clinical phenotype than do mutations resulting in the synthesis
of highly structurally abnormal proteins. The synthesis of structurally abnormal protein

can produce a dominant-negative effect that is the primary determinant of phenotype.
Thus, inherited defects that interfere with post-translational modification of matrix pro-
teins such as hydroxylation, sulfation and/or proteolytic cleavage, can result in distinct
osteochondrodysplasias. In the future, it may be possible to identify genes and pathways
that can maintain, repair, or stimulate the regeneration of bone and joint structures at post
patterning stages of development.
In this regard, it is to be noted that metabolites of vitamin A, including retinoic acid
(RA), comprise a class of molecules that are of critical importance in development and
homeostasis. Retinoic acid functions through a class of nuclear hormone receptors, the
RA receptors (RARs), to regulate gene transcription. Retinoic acid receptor-mediated
signaling plays a fundamental role in skeletogenesis. In the developing mammalian limb,
RA induces the differentiation of a number of cell lineages including chondrocytes.
However, excess RA is a potent teratogen that induces characteristic skeletal defects in
a stage- and dose-dependent manner. Genetic analyses have shown that RAR deficiency
results both in severe deficiency of cartilage formation in certain anatomical sites and the
promotion of ectopic cartilage formation in other sites. In the developing limbs of
transgenic mice expressing either dominant-negative or weakly constitutively active
RARs, chondrogenesis is perturbed, resulting in a spectrum of skeletal malformations.
Recently, RA was reported to bind two circadian clock proteins, Clock and Mop4, and
may play a role in regulating circadian rhythms. Thus, it may be possible to utilize these
interactions to manipulate the body’s response to therapeutic drugs, which is entrained
in the circadian flow.
A number of growth factors interact with osteoblasts or their precursors during bone
development, remodeling or repair. Traditionally, morphogenetic signals have been stud-
ied in embryos. However, it was observed that implantation of demineralized adult bone
matrix into subcutaneous sites in a variety of species resulted in local bone induction. Not
viii Preface
only did this model system mimic the process of limb morphogenesis, it also permitted
the isolation of bone morphogenetic proteins (BMPs). The BMPs constitute a large
family of morphogenetic proteins within the transforming growth factor-` (TGF-`)

superfamily. It is to be emphasized that these morphogens and related cartilage-derived
morphogenetic proteins (CDMPs) that initiate, promote, and maintain chondrogenesis,
have actions on systems other than bone. Indeed, bone morphogenetic proteins are mul-
tifunctional growth factors involved in many aspects of tissue development and morpho-
genesis, including, for example, regulation of FSH action in the ovary. The mechanism
underlying the phenomenon of bone matrix-induced bone induction is under intense
investigation by biomedical engineers and orthopedic researchers.
Growth/differentiation factor-5 (GDF-5), a BMP family member, has been shown to
be essential for normal appendicular skeletal and joint development in humans and mice.
It has been reported that GDF-5 promotes the initial stages of chondrogenesis by promot-
ing cell adhesion and increased cell proliferation. In the mouse GDF-5 gene mutant
brachypod, the defect is manifested early in chondrogenesis (embryonic day [E]12.5) as
a reduction in the size of the cartilage blastema. The defect is associated with a decrease
in the expression of cell surface molecules resulting in a decrease in cell adhesiveness
and, consequently, perturbation of cartilage model competence. Another member of the
family, BMP-6, has been shown to be overexpressed in prostate cancer and appears to be
associated with bone-forming skeletal metastases. In the United States, prostate cancer
became the number one cancer among white males in the mid-1980s and has increased
dramatically since then. A study of benign and malignant prostate lesions by in situ
hybridization showed that BMP-6 expression was high at both primary and secondary
sites in cases of advanced cancer with metastases. Does upregulation of BMP-6 promote
metastasis or is it involved in the body’s defense armamentarium? Is it a target for
therapeutics? Such questions are under active investigation by cancer researchers.
Two families of growth factors, the TGF-` superfamily and the insulin-like growth
factors (IGF) superfamily, appear to be the principal proximal regulators of osteogenesis.
However, these growth factors are not specific for cells of the osteoblast lineage. The
mechanism by which skeletal tissue is specifically induced and maintained involves both
complex interactions among circulating hormones, growth factors, and regulators of the
activity of specific genes. For example, nuclear transcription factors such as core binding
factor a1 (Cbfa1), a transcription factor essential for osteoblast differentiation and bone

formation, and CCAAT/enhancer binding protein b (C/EBPb), that function as regulators
of the expression/activity of specific bone growth factors and receptors, are activated in
response to glucocorticoids, sex steroids, parathyroid hormone (PTH), and prostaglandin
E2 (PGE2). Many environmentally available chemicals, both natural and man-made,
have either sex steroid or anti-sex steroid activity. Evidence suggests that such chemicals
have negatively impacted fish populations and other animals by interfering with the
mechanism of action of reproductive hormones. However, their impact on other mecha-
nisms such as growth have not been thoroughly investigated.
Members of the tumor necrosis factor (TNF) family of ligands and receptors have been
identified as critical regulators of osteoclastogenesis. Osteoprotegerin (OPG), a member
of the TNF receptor family, plays a key role in the physiological regulation of osteoclastic
bone resorption. OPG, a secreted decoy receptor produced by osteoblasts and marrow
stromal cells, acts by binding to its natural ligand, OPGL (also known as RANKL [recep-
tor activator of NF-gB ligand]), thereby preventing OPGL from activating its cognate
receptor RANK, the osteoclast receptor vital for osteoclast differentiation, activation and
survival. In vitro studies have suggested that estrogen stimulates OPG expression whereas
parathyroid hormone (PTH) inhibits its expression and stimulates the expression of
RANKL. This construct provides a molecular mechanism for the regulation of the osteo-
clastic bone resorption and osteoblastic bone formation couple and basis for the bone loss
of postmenopausal osteoporosis, aging and pathologic skeletal changes (e.g., osteopetro-
sis, glucocorticoid-induced osteoporosis, periodontal disease, bone metastases, Paget’s
disease, hyperparathyroidism, and rheumatoid arthritis). Environmental toxicants and
endocrine disruptors also may perturb the normal balance between osteoclastic and os-
teoblastic activity by interfering with homeostasis and/or accelerating aging processes.
With regard to endocrine disruption, OPG has been linked to vascular disease, particu-
larly arterial calcification in estrogen-deficient individuals, the aged, and those afflicted
with immunological deficits.
During skeletogenesis, cartilage matures either into permanent cartilage that persists
as such throughout the organism’s life or transient cartilage that ultimately is replaced by
bone. How cartilage phenotype is specified is not clear. In vitro studies have shown that

Cbfa1 is involved in induction of chondrocyte maturation. In this regard, it is of interest
to note that transgenic mice overexpressing either Cbfa1 or a dominant-negative (DN)-
Cbfa1 in chondrocytes exhibit dwarfism and skeletal malformations. These phenotypes
are mediated through opposing mechanisms. In the former case, Cbfa1 overexpression
accelerates endochondral ossification resulting from precocious chondrocyte maturation
whereas in the latter, DN-Cbfa1 overexpression suppresses maturation and delays endo-
chondral ossification. In addition, mice overexpressing Cbfa1 fail to form most of their
joints and what would be permanent cartilage in normal mice enters the endochondral
pathway of ossification. In contrast, in DN-Cbfa1 transgenic mice, most chondrocytes
exhibit a marker for permanent cartilage. It may be concluded from these observations that
proper temporal and spatial expression of chondrocyte Cbfa1 is required for normal
skeletogenesis, including formation of joints, permanent cartilage, and endochondral bone.
Both gain-of-function and loss-of-function mutations in fibroblast growth factor re-
ceptor 3 (FGFR3) have revealed unique roles for this receptor during skeletal develop-
ment. Loss-of-function alleles of FGFR3 lead to an increase in the size of the hypertrophic
zone, delayed closure of the growth plate and the subsequent overgrowth of long bones.
Gain-of-function mutations in FGFR3 have been linked genetically to autosomal domi-
nant dwarfing chondrodysplasia syndromes in which both the size and architecture of the
epiphyseal growth plate are altered. Analysis of these phenotypes and the biochemical
consequences of the mutations in FGFR3 demonstrate that FGFR3-mediated signaling
is an essential negative regulator of endochondral ossification.
Thorough understanding of bone physiology and how it is modified throughout all
stages of life, from in utero development to advanced age, is of great current interest for
its potential application to the establishment of criteria for the achievement and mainte-
nance of bone health and the reestablishment of bone health following trauma and dis-
ease. Other clinical applications include:
• Establishment of criteria for the achievement of optimal bone strength throughout
life, its maintenance in such long-term microgravity situations as space travel, and the
Preface ix
facilitation of readjustment to normogravity upon return to earth. This will require estab-

lishment of rapid and precise methods for distinguishing mechanically competent bone
from incompetent bone.
• Establishment of optimal conditions for the healing of fractures, osteotomies, and
arthrodeses.
• Understanding the mechanics of induction by falling of metaphyseal and diaphyseal
fractures of the radius in children, but primarily metaphyseal fractures in the aged.
• Improvement of the endurance of load-bearing implants.
• Understanding the mechanism(s) of osteopenia and osteoporosis and how and why,
during menopause, healthy women lose only bone adjacent to marrow.
Furthermore, because of the multifunctionality and interactions of the skeletal system,
biomedical researchers and practitioners of almost every clinical discipline have great
interest in bone biology. Even a cursory review of the bone biology literature will reveal
the depth of interest in the field. Publications emanate from a broad spectrum of biomedi-
cal areas that include: adolescent medicine, anatomy, anthropology, biochemistry, bio-
mechanics, biomedical engineering, biophysics, cardiology, cell and molecular biology,
clinical nutrition research, dentistry, developmental biology, endocrinology, enzymol-
ogy, epidemiology, food science, genetics, genetic counseling, gerontology, hematol-
ogy, histology, human nutrition, internal medicine, medicinal chemistry, metabolism,
microbiology, neurology, oncology, orthopedics, pediatric medicine, pharmacology and
therapeutics, physical and rehabilitation medicine, physiology, plastic surgery, public
health, radiology and imaging research, space and sports medicine, trace/essential ele-
ment research, vascular biology, vitaminology and cofactor research, women’s health,
teratology, and toxicology.
Bone biology is a diverse field, and our goal in developing The Skeleton: Biochemical,
Genetic, and Molecular Interactions in Development and Homeostasis was to provide
researchers and students with an overview of selected topics of current interest in bone
biology and to stimulate their interest in this fascinating and diverse field.
Edward J. Massaro
John M. Rogers
x Preface

Preface v
Contributors xv
I. CHONDROGENESIS, CHONDROCYTES, AND CARTILAGE
1 Molecular Basis of Cell–Cell Interaction and Signaling
in Mesenchymal Chondrogenesis 3
Rocky S. Tuan
2 Chondrocyte Cell Fate Determination in Response to Bone
Morphogenetic Protein Signaling 17
Lillian Shum, Yuji Hatakeyama, Julius Leyton,
and Kazuaki Nonaka
3 Regulation of Chondrocyte Differentiation 43
Andreia M. Ionescu, M. Hicham Drissi, and Regis J. O’Keefe
4 Continuous Expression of Cbfa1 in Nonhypertrophic
Chondrocytes Uncovers Its Ability to Induce Hypertrophic
Chondrocyte Differentiation and Partially Rescues
Cbfa1-Deficient Mice 55
Shu Takeda, Jean-Pierre Bonnamy, Michael J. Owen,
Patricia Ducy, and Gerard Karsenty
5 Molecular Biology and Biosynthesis of Collagens 77
Johanna Myllyharju
6 Mechanotransduction Pathways in Cartilage 89
Qian Chen
II. CONTROL OF SKELETAL DEVELOPMENT
7 Molecular Genetic Analysis of the Role of the HoxD Complex
in Skeletal Development: Impact of the loxP/Cre System
in Targeted Mutagenesis of the Mouse HoxD Complex 101
Marie Kmita, Denis Duboule, and József Zákány
8 Control of Development and Homeostasis Via Regulation
of BMP, Wnt, and Hedgehog Signaling 113
Renee Hackenmiller, Catherine Degnin, and Jan Christian

9
FGF4 and Skeletal Morphogenesis 131
Valerie Ngo-Muller, Shaoguang Li, Scott A. Schaller,
Manjong Han, Jennifer Farrington, Minoru Omi,
Rosalie Anderson, and Ken Muneoka
10 Retinoid Signaling and Skeletal Development 147
Andrea D. Weston and T. Michael Underhill
CONTENTS
xi
xii Contents
11 Retinoids and Indian Hedgehog Orchestrate Long Bone
Development 159
Maurizio Pacifici, Chiara Gentili, Eleanor Golden,
and Eiki Koyama
III. OSTEOBLASTIC CELL DIFFERENTIATION
12 Synergy Between Osteogenic Protein-1 and Osteotropic
Factors in the Stimulation of Rat Osteoblastic Cell
Differentiation 173
John C. Lee and Lee-Chuan C. Yeh
13 Bone Morphogenic Proteins, Osteoblast Differentiation,
and Cell Survival During Osteogenesis 185
Cun-Yu Wang
14 Osteoclast Differentiation 195
Sakamuri V. Reddy and G. David Roodman
IV. BONE INDUCTION, GROWTH, AND REMODELING
15 Soluble Signals and Insoluble Substrata: Novel Molecular
Cues Instructing the Induction of Bone 217
Ugo Ripamonti, Nathaniel L. Ramoshebi, Janet Patton,
Thato Matsaba, June Teare, and Louise Renton
16 Perichondrial and Periosteal Regulation of Endochondral

Growth 229
Dana L. Di Nino and Thomas F. Linsenmayer
17 Computer Simulations of Cancellous Bone Remodeling 249
Jacqueline C. van der Linden, Harrie Weinans,
and Jan A. N. Verhaar
18 Effects of Microgravity on Skeletal Remodeling
and Bone Cells 263
Pierre J. Marie
V. BONE MINERALIZATION
19 Quantitative Analyses of the Development of Different Hard
Tissues 279
Siegfried Arnold, Hans J. Höhling, and Ulrich Plate
20 Fetal Mineral Homeostasis and Skeletal Mineralization 293
Christopher S. Kovacs
21 Control of Osteoblast Function and Bone Extracellular Matrix
Mineralization by Vitamin D 307
Johannes P. T. M. van Leeuwen, Marjolein van Driel,
and Hulbert A. P. Pols
Contents xiii
VI. SKELETAL DYSMORPHOLOGY
22 Role of Pax3 and PDGF-_ Receptor in Skeletal
Morphogenesis and Facial Clefting 335
Simon J. Conway
23 Genetics of Achondroplasia and Hypochondroplasia 349
Giedre Grigelioniene
24 Effects of Boric Acid on Hox Gene Expression and the Axial
Skeleton in the Developing Rat 361
Michael G. Narotsky, Nathalie Wéry, Bonnie T. Hamby,
Deborah S. Best, Nathalie Pacico, Jacques J. Picard,
Françoise Gofflot, and Robert J. Kavlock

25 Toxicant-Induced Lumbar and Cervical Ribs in Rodents 373
John M. Rogers, R. Woodrow Setzer, and Neil Chernoff
26 Experimental Skeletal Dysmorphology: Risk Assessment
Issues 385
Rochelle W. Tyl, Melissa C. Marr, and Christina B. Myers
Index 415

CONTRIBUTORS
ROSALIE ANDERSON • Department of Biological Sciences, Loyola University, New
Orleans, LA
SIEGFRIED ARNOLD • Large Area Electronics, PerkinElmer Optoelectronics, Wiesbaden,
Germany
D
EBORAH S. BEST • Reproductive Toxicology Division, National Health and
Environmental Effects Research Laboratory, Office of Research and Development,
US Environmental Protection Agency, Research Triangle Park, NC
J
EAN-PIERRE BONNAMY • Department of Molecular and Human Genetics, Baylor
College of Medicine, Houston, TX
Q
IAN CHEN • Department of Orthopaedics, Brown Medical School, Rhode Island
Hospital, Providence, RI
N
EIL CHERNOFF • Developmental Biology Branch, Reproductive Toxicology Division,
National Health and Environmental Effects Research Laboratory, Office of Research
Development, US Environmental Protection Agency, Research Triangle Park, NC
S
IMON J. CONWAY • Department of Cell Biology and Anatomy, Institute of Molecular
Medicine and Genetics, Medical College of Georgia, Augusta, GA
JAN CHRISTIAN • Department of Cell and Developmental Biology, Oregon Health and

Science University, Portland, OR
C
ATHERINE DEGNIN • Department of Cell and Developmental Biology, Oregon Health
and Science University, Portland, OR
D
ANA L. DI NINO • Department of Anatomy and Cellular Biology, Tufts University
Medical School, Boston, MA
M. H
ICHAM DRISSI • Center for Musculoskeletal Research, University of Rochester,
Rochester, NY
DENIS DUBOULE • Department of Zoology and Animal Biology and NCCR ‘Frontiers
in Genetics,’ University of Geneva, Geneva, Switzerland
P
ATRICIA DUCY • Department of Molecular and Human Genetics, Baylor College of
Medicine, Houston, TX
J
ENNIFER FARRINGTON • Division of Developmental Biology, Department of Cell and
Molecular Biology, Tulane University, New Orleans, LA
C
HIARA GENTILI • Centro di Medicina Rigenerativa, Istituto Nazionale Ricerca sul
Cancro, Genova, Italy
F
RANÇOISE GOFFLOT • Unit of Developmental Genetics, Université Catholique de Louvain,
Bruxelles, Belgium
E
LEANOR GOLDEN • Department of Anatomy and Cell Biology, School of Dental
Medicine, University of Pennsylvania, Philadelphia, PA
GIEDRE GRIGELIONIENE • Paediatric Endocrinology Unit, Karolinska Hospital,
Stockholm, Sweden
R

ENEE HACKENMILLER • Department of Cell and Developmental Biology, Oregon
Health and Science University, Portland, OR
BONNIE T. HAMBY • RTI International, Center for Life Sciences and Technology,
Research Triangle Park, NC
xv
xvi Contributors
`
`
MANJONG HAN • Division of Developmental Biology, Department of Cell and Molecular
Biology, Tulane University, New Orleans, LA
Y
UJI HATAKEYAMA • Cartilage Biology and Orthopaedics Branch, National Institute
of Arthritis, and Musculoskeletal and Skin Disease, National Institutes of Health,
Bethesda, MD
H
ANS J. HÖHLING • Institut für Medizinische Physik und Biophysik, Westfälische
Wilhelms-Universität Münster, Germany
A
NDREIA M. IONESCU • Center for Musculoskeletal Research, University of Rochester,
Rochester, NY
G
ERARD KARSENTY • Department of Molecular and Human Genetics, Baylor College
of Medicine, Houston, TX
R
OBERT J. KAVLOCK • Reproductive Toxicology Division, National Health and
Environmental Effects Research Laboratory, Office of Research and Development,
US Environmental Protection Agency, Research Triangle Park, NC
M
ARIE KMITA • Department of Zoology and Animal Biology and NCCR “Frontiers in
Genetics,” University of Geneva, Geneva, Switzerland

C
HRISTOPHER S. KOVACS • Faculty of Medicine—Endocrinology, Memorial University
of Newfoundland Health Sciences Centre, St. John’s, Newfoundland, Canada
E
IKI KOYAMA • Department of Orthopaedic Surgery, Thomas Jefferson University
Medical School, Philadelphia, PA
J
OHN C. LEE • Department of Biochemistry, University of Texas Health Science Center,
San Antonio, TX
J
ULIUS LEYTON • Cartilage Biology and Orthopaedics Branch, National Institute
of Arthritis, and Musculoskeletal and Skin Disease, National Institutes of Health,
Bethesda, MD
S
HAOGUANG LI • The Jackson Laboratory, Bar Harbor, Maine
T
HOMAS F. LINSENMAYER • Department of Anatomy and Cellular Biology, Tufts
University Medical School, Boston, MA
PIERRE J. MARIE • INSERM U349 Lariboisière Hospital, Paris, France
M
ELISSA C. MARR • RTI International, Center for Life Sciences and Toxicology,
Research Triangle Park, NC
E
DWARD J. MASSARO • Developmental Biology Branch, Reproductive Toxicology Division,
National Health and Environmental Effects Research Laboratory, Office of Research
Development, US Environmental Protection Agency, Research Triangle Park, NC
T
HATO MATSABA • Bone Research Unit, Medical Research Council/University of the
Witwatersrand, Johannesburg, South Africa
C

HRISTINA B. MYERS • RTI International, Center for Life Sciences and Toxicology,
Research Triangle Park, NC
K
EN MUNEOKA • Division of Developmental Biology, Department of Cell and Molecular
Biology, Tulane University, New Orleans, LA
J
OHANNA MYLLYHARJU • Collagen Research Unit, Biocenter Oulu and Department of
Medical Biochemistry and Molecular Biology, University of Oulu, Oulu, Finland
MICHAEL G. NAROTSKY • Reproductive Toxicology Division, National Health and
Environmental Effects Research Laboratory, Office of Research and Development,
US Environmental Protection Agency, Research Triangle Park, NC
V
ALERIE NGO-MULLER • ICGM Cochin Port-Royal, Paris France
Contributors xvii
KAZUAKI NONAKA • Section of Pediatric Dentistry, Division of Oral Health, Growth,
and Development, Faculty of Dental Science, Kyushu University, Fukuoka, Japan
R
EGIS J. O’KEEFE • Center for Musculoskeletal Research, University of Rochester,
Rochester, NY
M
INORU OMI • Division of Developmental Biology, Department of Cell and Molecular
Biology, Tulane University, New Orleans, LA
M
ICHAEL J. OWEN • Imperial Cancer Research Fund, London, UK
N
ATHALIE PACICO • Unit of Developmental Genetics, Université Catholique de Louvain,
Bruxelles, Belgium
M
AURIZIO PACIFICI • Department of Orthopaedic Surgery, Thomas Jefferson University
Medical School, Philadelphia, PA

J
ANET PATTON • Bone Research Unit, Medical Research Council/University of the
Witwatersrand, Johannesburg, South Africa
JACQUES J. PICARD • Unit of Developmental Genetics, Université Catholique de Louvain,
Bruxelles, Belgium
U
LRICH PLATE • Klinik und Poliklinik für Mund- und Kiefer-Gesichtschirurgie,
Westfälische Wilhelms-Universität Münster, Germany
H
ULBERT A. P. POLS • Department of Internal Medicine, Erasmus Medical Center
Rotterdam, Rotterdam, The Netherlands
N
ATHANIEL L. RAMOSHEBI • Bone Research Unit, Medical Research Council/University
of the Witwatersrand, Johannesburg, South Africa
S
AKAMURI V. REDDY • Center for Bone Biology, Division of Hematology-Oncology,
Department of Medicine, University of Pittsburgh, Pittsburgh, PA
LOUISE RENTON • Bone Research Unit, Medical Research Council/University of the
Witwatersrand, Johannesburg, South Africa
U
GO RIPAMONTI • Bone Research Unit, Medical Research Council/University of the
Witwatersrand, Johannesburg, South Africa
J
OHN M. ROGERS • Developmental Biology Branch, Reproductive Toxicology
Division, National Health and Environmental Effects Research Laboratory,
Office of Research Development, US Environmental Protection Agency,
Research Triangle Park, NC
G. D
AVID ROODMAN • Center for Bone Biology, Division of Hematology-Oncology,
Department of Medicine, University of Pittsburgh, Pittsburgh, PA

SCOTT A. SCHALLER • Department of Cell and Molecular Biology, Division of
Developmental Biology, Tulane University, New Orleans, LA
R. W
OODROW SETZER • Pharmacokinetics Branch, Experimental Toxicology Division,
National Health and Environmental Effects Research Laboratory, Office of
Research and Development, US Environmental Protection Agency, Research
Triangle Park, NC
L
ILLIAN SHUM • Cartilage Biology and Orthopaedics Branch, National Institute
of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health,
Bethesda, MD
S
HU TAKEDA • Department of Molecular and Human Genetics, Baylor College of
Medicine, Houston, TX
J
UNE TEARE • Bone Research Unit, Medical Research Council/University of the
Witwatersrand, Johannesburg, South Africa
xviii Contributors
ROCKY S. TUAN • Cartilage Biology and Orthopaedics Branch, National Institute
of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health,
Bethesda, MD
R
OCHELLE W. TYL • RTI International, Center for Life Sciences and Toxicology,
Research Triangle Park, NC
T. M
ICHAEL UNDERHILL • Department of Physiology, and Division of Oral Biology,
Faculty of Medicine and Dentistry, University of Western Ontario, London,
Ontario, Canada
J
ACQUELINE C. VAN DER LINDEN • Orthopedic Research Laboratory, Erasmus University

Rotterdam, Rotterdam, The Netherlands
MARJOLEIN VAN DRIEL • Department of Internal Medicine, Erasmus Medical Center
Rotterdam, Rotterdam, The Netherlands
J
OHANNES P. T. M. VAN LEEUWEN • Department of Internal Medicine, Erasmus Medical
Center Rotterdam, Rotterdam, The Netherlands
J
AN A. N. VERHAAR • Orthopedic Research Laboratory, Erasmus University Rotterdam,
Rotterdam, The Netherlands
CUN-YU WANG • Department of Biologic and Materials Sciences, University of
Michigan School of Dentistry, Ann Arbor, MI
H
ARRIE WEINANS • Orthopedic Research Laboratory, Erasmus University Rotterdam,
Rotterdam, The Netherlands
N
ATHALIE WÉRY • Unit of Developmental Genetics, Université Catholique de Louvain,
Bruxelles, Belgium
A
NDREA D. WESTON • Department of Physiology, Faculty of Medicine and Dentistry,
University of Western Ontario, London, Ontario, Canada
L
EE-CHUAN C. YEH • Department of Biochemistry, University of Texas Health Science
Center, San Antonio, TX
JÓZSEF ZÁKÁNY • Department of Zoology and Animal Biology and NCCR ‘Frontiers in
Genetics,’ University of Geneva, Geneva, Switzerland
Regulation of Mesenchymal Chondrogenesis 1
I
Chondrogenesis,
Chondrocytes, and Cartilage
2 Tuan

4 Tuan
importance of cellular condensation in chondrogenesis has come from both in vivo and in vitro obser-
vations. Many classical studies have demonstrated a high cell density requirement for chondrogene-
sis to occur (10), correlated the extent of cell condensation with the level of chondrogenesis (11,12),
demonstrated the initiation of gap-junction-mediated cell–cell communication in condensing mesen-
chyme (13,14), and described characteristic limb skeletal abnormalities in genetic mutants defective in
mesenchymal cell condensation (reviewed in refs. 2 and 15).
The process of mesenchymal cell condensation is directed by cell–cell and cell–matrix interactions
as well as secreted factors interacting with their cognate receptors. Before condensation, mesenchy-
mal cells present in the limb secrete an extracellular matrix (ECM) rich in hyaluronan and collagen
type I that prevents intimate cell–cell interaction. As condensation begins, an increase in hyaluronid-
ase activity is observed with a decrease in hyaluronan in the ECM. Hyaluronan is thought to facilitate
cell movement, and the increase in hyaluronidase and subsequent decrease in hyaluronan allows for
close cell–cell interactions (16–18). The establishment of cell–cell interactions is presumably involved
in triggering one or more signal transduction pathways that initiates chondrogenic differentiation. Two
cell adhesion molecules implicated in this process are N-cadherin and neural cell adhesion molecule
(N-CAM). Both of these molecules are expressed in condensing mesenchyme and then disappear in
differentiating cartilage (19,20) and later are detectable only in the perichondrium. Perturbing the func-
tion(s) of N-cadherin (21) or N-CAM (22) causes reduction or alterations in chondrogenesis both in
vitro and in vivo, further supporting a role for these cell adhesion molecules in mediating the mesen-
chymal condensation step.
In addition to cell–cell interactions, cell–matrix interactions also appear to play an important role
in mesenchymal cell condensation. One ECM component implicated in this process is fibronectin.
Fibronectin expression is increased in areas of cellular condensation (23,24) and decreases as cytodif-
ferentiation proceeds. Fibronectin may facilitate a matrix-driven translocation of mesenchymal cells
into cellular condensations, and this process may be mediated by the amino terminal heparin binding
domain (25,26). Recent studies in our laboratory have demonstrated that fibronectin mRNA under-
goes alternative splicing during chondrogenesis (27–29). The isoform containing exon EIIIA is present
during condensation but disappears once differentiation begins, suggesting that this isoform switch-
ing is important for cytodifferentiation to occur. Antibodies specific for the region encoded by exon

EIIIA of the fibronectin gene inhibited chondrogenesis of limb micromass cultures in vitro, and when
injected into chick limb buds in vivo, caused moderate to severe skeletal malformations (27,28).
CELL ADHESION IN MESENCHYMAL CELL CONDENSATION
Cell adhesion is mediated by two major groups of cell–cell adhesion molecules, the Ca
2+
-indepen-
dent and the Ca
2+
-dependent adhesion molecules (30–38). The Ca
2+
-independent group is composed
of the large immunoglobulin supergene family of membrane glycoproteins known as CAMs, and the
Ca
2+
-dependent group consists largely of a class of transmembrane glycoproteins called the cadherins.
Two adhesion molecules, N-cadherin and N-CAM, have been shown to have an important role dur-
ing the precartilaginous condensation phase during endochondral ossification (21,22).
N-Cadherin
The cadherin superfamily has many members and can be divided into six gene subfamilies based
on structural homology: classical cadherins type I (e.g., E-, N-, P-, R-cadherin), classical cadherins
type II (cadherin-6 to -12), cadherins found in desmosomes (desmocollins, desmogleins), cadherins
with a very short cytoplasmic domain or none (LI-, T-cadherin), protocadherins, and the more dis-
tantly related gene products, including the Drosophila fat tumor-suppressor gene, the dachsous gene,
and the ret-proto-oncogene (39).
The classical cadherins are a group of Ca
2+
-dependent, single transmembrane glycoproteins that
mediate cell–cell adhesion by homotypic protein–protein interactions through their extracellular domain.
Regulation of Mesenchymal Chondrogenesis 5
Classic cadherins are synthesized as precursor polypeptides and are then processed into their mature

form. The extracellular domain of the mature protein consists of five tandem repeat domains termed
cadherin repeats, each of which consists of approx 110 amino acids. The cadherin repeats form four
Ca
2+
-binding domains, and the N-terminal repeat confers the cadherin-specific adhesive property of
the molecule. Classic cadherins have a single transmembrane domain followed by a highly conserved
cytoplasmic domain responsible for binding to the actin cytoskeleton via the catenin molecules (Fig. 1;
ref. 40).
The cadherin family of molecules exhibit spatiotemporally unique patterns of gene expression (37)
and demonstrate homotypic binding through their extracellular domain, suggesting that cadherins
may function as morphoregulatory molecules during development. N-Cadherin, named for its initial
identification in neural tissues, was one of the first identified cadherins, and its functional involve-
ment in cell–cell adhesion and development has been extensively studied (41–48). N-Cadherin plays
a major role in neural development but has also been shown to be expressed in other mesodermal tis-
sues, including developing limb mesenchyme. N-Cadherin is expressed in the developing embryonic
limb bud in a manner suggestive of a role in cellular condensation (Fig. 2; ref. 21). Immunohistochem-
ical localization of N-cadherin in the embryonic chick limb reveals a sparsely scattered expression pat-
tern in the central core mesenchyme during the precartilage stage (Hamburger–Hamilton stages 17/18
through 22/23; ref. 49). Expression dramatically increases in the condensing central core at stage 24/25,
and by stage 25/26, the condensed central core region begins to lose N-cadherin expression, whereas
cells along the periphery of the limb bud begin to express N-cadherin. By stage 29/30, the mature
cartilage is completely devoid of N-cadherin whereas the condensing, perichondral cells surrounding
the forming cartilage still exhibit high levels of N-cadherin. As the limb bud continues to develop,
the cartilaginous core region continues to grow appositionally, and it is likely that the N-cadherin–
positive cells along the periphery contribute to this growth (21).
Fig. 1. Schematic of N-cadherin–catenin complex. N-cadherin is a Ca
2+
-dependent, single-pass transmem-
brane protein that mediates cell–cell adhesion by homotypic protein–protein interactions through its extracellular
domain. The extracellular domain is composed of five tandem repeats, termed cadherin repeats, that form four

Ca
2+
binding sites. The fifth cadherin repeat confers its homotypic specificity by the HAV (histidine–alanine–
valine) amino acid sequence. The cytoplasmic domain binds the actin cytoskeleton via interactions with the cate-
nin family of proteins. The cytoplasmic domain binds `/a-catenin (`/_-cat) directly, which in turn binds _-catenin
(_-cat). Subsequently, _-cat binds the actin cytoskeleton directly or in conjunction with _-actinin. Other proteins
also bind the cytoplasmic domain of N-cadherin, such as p120
ctn
and the nonreceptor protein tyrosine phospha-
tase 1B (PTP1B). Both of these proteins bind the cytoplasmic domain and regulate cell adhesion.
6 Tuan
In high-density micromass cultures in vitro, dissociated limb mesenchymal cells aggregate to form
cellular condensations that ultimately differentiate into cartilaginous nodules, separated by fibroblasts
and myocytes (10). N-Cadherin protein is synthesized by the aggregating (condensing) mesenchyme
by 12 h after initiation of the culture, whereas the cells outside of the condensation centers display no
evident N-cadherin expression. Expression of N-cadherin becomes more intense as a function of time
with maximal expression at 18 h. As the cells in the center of the condensations differentiate, they
lose their N-cadherin protein, and the cells along the immediate periphery of the forming nodules main-
tain N-cadherin expression. Thus, the expression pattern of N-cadherin in vitro recapitulates that in
the developing limb in situ (21).
N-Cadherin expression is localized to the prechondroblastic cells of the limb bud, and maximal
expression is seen during mesenchymal cell condensation, after which it is downregulated, suggest-
ing that cellular condensation is dependent on N-cadherin–mediated cell–cell interactions (19,21).
Evidence to support this theory comes from studies designed to perturb N-cadherin function. We (19)
were able to demonstrate a significant inhibition of cellular condensation and chondrogenesis in vitro
and in vivo using a function-blocking monoclonal antibody, NCD-2, directed against N-cadherin (42).
These findings correlate well with previous findings that exogenous Ca
2+
significantly stimulates
chondrogenesis in vitro when added before condensation but has little effect when added after con-

Fig. 2. Spatiotemporal specificity of N-cadherin expression in the developing chick embryonic limb bud. A and
B, Hamburger–Hamilton Stage 24/25. NCD-2 immunofluorescent staining reveals N-cadherin expression to be
localized exclusively with the condensing mesenchyme (M; arrows). E, ectoderm. A, epifluorescent optics; B,
Nomarski optics. C and D, Stage 29/30. Mature cartilage (C) is formed and is negative for N-cadherin, whereas
the surrounding mesenchyme (M) remains positive. C, epifluorescent optics; D, Nomarski optics. Magnification:
bar = 100 µm. (Taken from ref. 21.)
Regulation of Mesenchymal Chondrogenesis 7
densation (11,12). In similar studies, the addition of transforming growth factor-` family member,
BMP-2, to chick limb bud or the C3H10T1/2 murine multipotential cell line plated at high-density
micromass cultures stimulated chondrogenesis on the basis of Alcian blue staining, collagen type II,
and link protein expression and led to an increase in [
35
S]sulfate incorporation (50–53). Further inves-
tigation revealed that BMP-2 treatment of C3H10T1/2 cells stimulated N-cadherin mRNA levels four-
fold within 24 h and protein levels eightfold by day 5 in culture, whereas an N-cadherin peptidomimic
containing the His-Ala-Val sequence was able to inhibit chondrogenesis in a dose-dependent manner
(54). To specifically examine the influence of altered N-cadherin expression or activity on chondro-
genesis, C3H10T1/2 cells were stably transfected with N-cadherin wild-type or dominant-negative
N-terminal deletion constructs. Cells expressing the wild-type N-cadherin at a moderate level (two-
fold) increased chondrogenesis, whereas cells expressing a fourfold increase in N-cadherin or the dom-
inant-negative construct had an initial, inhibitory effect on BMP-2 stimulation of chondrogenesis
(54). In recent studies (55,56) we have further examined the functional role of N-cadherin by analyz-
ing the effect of transfecting chick embryonic limb mesenchymal cells with expression constructs
that encode for wild-type or amino-deleted/carboxy-deleted mutant forms of N-cadherin. Plasmid and
retroviral (RCAS) vectors were used for transient and stable misexpression, respectively. Our results
showed that N-cadherin is crucial in mediating the initial cell–cell interaction in mesenchymal conden-
sation and requires both the extracellular homotypic binding site and the intracellular site involved in
adhesion complex formation. However, proper chondrogenic progression requires a subsequent down-
regulation of N-cadherin and cell adhesion, such that prolonged overexpression of wild-type N-cad-
herin in the stable transfectants actually results in a significantly reduced level of chondrogenesis.

Taken together, these data strongly support a functional and activity-dependent role for N-cadherin
in cellular condensation and chondrogenesis.
N-CAM
The glycoprotein N-CAM is a member of the immunoglobulin superfamily (57). N-CAM is com-
posed of five immunoglobulin-like domains, each consisting of aprrox 100 amino acids folded into `
sheets usually linked by a disulfide bond (58). There is only one N-CAM gene; however, different
forms of N-CAM can be generated through alternative splicing of its mRNA as well as varying degrees
of glycosylation (sialic acid; refs. 59–61). The major mRNA splicing differences occur near the car-
boxy-terminal with some forms displaying altered cytoplasmic domains or missing the transmembrane
domain. Homotypic binding of N-CAM occurs near the amino terminal (62) and does not appear to
be affected by alternative splicing.
N-CAM expression in the developing chick limb follows that of the previously described N-cadherin;
however, the expression of N-cadherin mRNA occurs earlier than that of N-CAM mRNA (20). N-CAM
expression in vivo is observed in all limb bud cells by stage 22 (63). N-CAM expression increases
and is enriched in the condensing mesenchyme at stage 27. By stage 30, the cells in the center of the
condensations differentiate and N-CAM expression is lost in mature cartilage, but strong N-CAM
expression is maintained in the surrounding perichondrium (22,63,64). The in vitro N-CAM expression
pattern parallels that of the in vivo expression. In micromass cultures in vitro, N-CAM is expressed
after 1.5 d in the aggregating, precartilage condensations, with a zone of moderately N-CAM–express-
ing cells surrounding the condensations. By 4 d in culture, the condensations have differentiated into
cartilaginous nodules and lose the expression of N-CAM in their center but retain N-CAM expres-
sion at their periphery (22).
The functional role of N-CAM in cellular condensation and chondrogenesis was determined by
perturbation studies in vitro by using aggregation assays and micromass cultures. Aggregation of dis-
sociated stage 23 chick limb bud cells was reduced when incubated with anti-N-CAM antibodies in
suspension culture compared with cells incubated with nonimmune Fab fragments (22). In the presence
of anti-N-CAM antibodies, both the number and size of aggregates is reduced 50–60%. Micromass
cultures of chick limb bud mesenchyme demonstrated a reduction in both the area occupied by con-