Dynamics of Bone and Cartilage Metabolism
Dynamics of Bone and Cartilage Metabolism
This Page Intentionally Left Blank
Dynamics of Bone and Cartilage Metabolism Edited by MARKUS J. SEIBEL Department of Medicine Division of Endocrinology and Metabolism University of Heidelberg Medical School Heidelberg, Germany
SIMON P. ROBINS Skeletal Research Unit Rowett Research Institute Aberdeen, United Kingdom
JOHN P. BILEZIKIAN Departments of Medicine and Pharmacology Division of Endocrinology College of Physicians and Surgeons Columbia University New York, New York
San Diego
New York
Boston
London
Sydney
Tokyo
Toronto
Academic Press is an imprint of Elsevier 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, California 92101-4495, USA 84 Theobald’s Road, London WC1X 8RR, UK This book is printed on acid-free paper. Copyright © 2006, Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone: (+44) 1865 843830, fax: (+44) 1865 853333, E-mail:
[email protected]. You may also complete your request on-line via the Elsevier homepage (http://elsevier.com), by selecting “Support & Contact” then “Copyright and Permission” and then “Obtaining Permissions.” Library of Congress Cataloging-in-Publication Data Dynamics of bone and cartilage metabolism / Markus J. Seibel, Simon P. Robins, and John P. Bilezikian, editors.— 2nd ed. p. ; cm. ISBN-13: 978-0-12-088562-6 (hardcover : alk. paper) ISBN-10: 0-12-088562-X (hardcover : alk. paper) 1. Bones—Metabolism. 2. Cartilage—Metabolism. 3. Extracellular matrix. 4. Bones—Metabolism—Disorders. 5. Biochemical markers. [DNLM: 1. Bone and Bones—metabolism. 2. Biological Markers. 3. Cartilage—metabolism. 4. Extracellular Matrix—metabolism. WE 200 D997 2006] I. Seibel, M. J. II. Robins, Simon P. III. Bilezikian, John P. QP88.2.D96 2006 612.7’5—dc22 2006004804 British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN 13: 978-0-12-088562-6 ISBN 10: 0-12-088562-X For information on all Academic Press publications visit our Web site at www.books.elsevier.com Printed in the United States of America 06 07 08 09 10 9 8 7 6 5
4
3
2
1
Contents Contributors xiii Preface xix
IV. Concluding Remarks References 50
CHAPTER 3 Vitamin K Dependent Proteins of Bone and Cartilage
PART I Components of the Organic Extracellular Matrix of Bone and Cartilage
CAREN M. GUNDBERG AND SATORU K. NISHIMOTO I. Abstract 55 II. Introduction 55 III. Osteocalcin 56 IV. Matrix Gla Protein 59 V. Gas6 63 VI. Vitamin K/Warfarin 64 References 65
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone I. II. III. IV. V. VI. VII. VIII.
KLAUS VON DER MARK Introduction 3 The Collagen Families 5 Bone Collagens 7 Cartilage Collagens 9 Collagen Biosynthesis 14 Collagen Genes and Transcriptional Regulation 17 Factors Regulating Collagen Biosynthesis Conclusions 25 References 26
50
CHAPTER 4 Noncollagenous Proteins; Glycoproteins and Related Proteins DICK HEINEGÅRD, PILAR LORENZO, AND TORE SAXNE I. Introduction 71 II. Cartilage Extracellular Matrix 72 III. Bone, Extracellular Matrix 77 IV. Concluding Remarks 79 References 80
21
CHAPTER 5 Proteoglycans and Glycosaminoglycans
CHAPTER 2 Fibrillogenesis and Maturation of Collagens
I. II. III. IV.
SIMON P. ROBINS I. Introduction 41 II. Fibrillogenesis 42 III. Cross-linking 44 v
TIM HARDINGHAM Introduction 85 Glycosaminoglycans 86 Proteoglycans in Cartilage 87 Aggrecan 88
vi
Contents
V. Leucine-rich Proteoglycans in Cartilage and Bone 93 VI. Perlecan in Cartilage 96 References 96
CHAPTER 6 I. II. III. IV. V. VI. VII.
Growth Factors
PHILIPPA HULLEY, GRAHAM RUSSELL, AND PETER CROUCHER Introduction 99 Insulin-like Growth Factors 99 The Transforming Growth Factor-Beta/Bone Morphogenetic Protein Superfamily 100 Fibroblast Growth Factors 103 Wnts 104 Additional Growth Factors 105 Summary 106 References 107
CHAPTER 10
CHAPTER 11 I. II. III. IV. V.
CHAPTER 7 Prostaglandins and Proinflammatory Cytokines I. II. III. IV.
I. II. III. IV. V. VI. VII.
Acid Phosphatases
HELENA KAIJA, LILA O.T. PATRIKAINEN, SARI L. ALATALO, H. KALERVO VÄÄNÄNEN, AND PIRKKO T. VIHKO I. Acid Phosphatases 165 II. Tartrate-Resistant Acid Phosphatase (TRACP) 166 III. Prostatic Acid Phosphatase (PAP) 173 References 176
Matrix Proteinases
IAN M. CLARK AND GILLIAN MURPHY Introduction 181 Aspartic Proteinases 181 Cysteine Proteinases 182 Serine Proteinases 184 Metalloproteinases 186 References 192
LAWRENCE G. RAISZ AND JOSEPH A. LORENZO Introduction 115 Prostaglandins 115 The Role that Cytokines have in Osteoclast Formation and Function 117 The Role that Proinflammatory Cytokines have in Bone and Cartilage Metabolism 118 References 121
PART II Structure and Metabolism of the Extracellular Matrix of Bone and Cartilage
CHAPTER 8 Integrins and Other Adhesion Molecules
I. II. III. IV.
M. H. HELFRICH AND M. A. HORTON Abstract 129 Introduction 129 Molecular Structure of Adhesion Molecules 130 Adhesion Molecules in Cells of the Osteoblast Lineage 134 Adhesion Molecules in Osteoclasts 139 Adhesion Molecules in Chondrocytes 142 Conclusion 144 References 144
CHAPTER 9
Alkaline Phosphatases
JOSÉ LUIS MILLÁN I. Introduction 153 II. Structure And Regulation of the TNAP Gene 153 III. Protein Structure 155 IV. Function of TNAP 157 V. Clinical Use 160 References 161
CHAPTER 12 Mineralization, Structure and Function of Bone ADELE L. BOSKEY Abstract 201 The Structure and Function of Bone 201 Bone Mineralization 204 Bone Modeling and Remodeling 209 References 209
CHAPTER 13 Bone Structure and Strength EGO SEEMAN I. Introduction 213 II. Gravity and the Need for Stiffness, Flexibility, Lightness and Speed 213 III. The Material Composition and Structural Design of Bone 214 IV. Bone Modeling and Remodeling – The Mechanism of Bone’s Construction during Growth and Decay with Advancing Age 216 V. Strength Maintenance 217 VI. Conclusion 219 References 219
vii
Contents
CHAPTER 14 I. II. III. IV. V.
VI.
JANE B. LIAN AND GARY S. STEIN Abstract 221 Introduction 221 Developmental Signals for Cartilage and Bone Tissue Formation 222 Osteogenic Lineage Cells 227 The Osteoclast: A Functionally Unique Cell for Physiologically Regulated Resorption of Bone Mineral 236 Perspectives 242 References 242
CHAPTER 15 I. II. III. IV. V. VI.
VII. VIII. IX.
X.
XI.
The Cells of Bone
Signaling in Bone
T. JOHN MARTIN AND NATALIE A. SIMS Abstract 259 Introduction 259 The Control of Osteoclasts 260 Signaling in the Control of Osteoclast Activity 261 Signals from the Osteoblast Lineage that Control Osteoclast Formation 261 Hormone and Cytokine Influences on the Contact-dependent Regulation of Osteoclasts 261 Discovery of the Physiological Signaling Mechanisms in Osteoclast Control 262 Rank Signaling 263 Coupling of Bone Formation to Resorption – Release of Growth Factors from bone Matrix 264 Coupling of Bone Formation to Resorption – Autocrine/Paracrine Regulation by Differentiating Osteoblasts 266 Coupling of Bone Formation to Resorption – Are Osteoclasts a Source of Coupling Activity? 266 References 267
IX. Activation of the Cyclic Adenosine Monophosphate Second-Messenger System by Parathyroid Hormone 281 X. Identification of a Second PTH Receptor 282 XI. Physiological Actions of PTH 282 XII. Cell-to-Cell Communication: Osteoblasts and Osteoclasts 284 XIII. Preferential Actions of PTH at Selected Skeletal Sites 285 References 287
CHAPTER 17 Interaction of Parathyroid Hormone-related Peptide with the Skeleton I. II. III. IV. V.
CHAPTER 18 The Vitamin D Hormone and its Nuclear Receptor: Mechanisms Involved in Bone Biology GEERT CARMELIET, ANNEMIEKE VERSTUYF, CHRISTA MAES, GUY EELEN, AND ROGER BOUILLON I. Introduction 307 II. Metabolism of Vitamin D 308 III. Nuclear Vitamin D Receptor 309 IV. Vitamin D and Bone Cells 312 V. Pathology and Therapy Related to Vitamin D Availability, Metabolism, and Function 314 VI. Conclusions 319 References 319
CHAPTER 19 Sex Steroid Effects on Bone Metabolism
CHAPTER 16 Parathyroid Hormone: Structure, Function and Dynamic Actions LORRAINE A. FITZPATRICK AND JOHN P. BILEZIKIAN I. Introduction 273 II. Structure of the PTH Gene 273 III. Chromosome Location 274 IV. Control of Gene Expression 275 V. Biosynthesis of Parathyroid Hormone 278 VI. Metabolism of Parathyroid Hormone 278 VII. Receptor Interactions of Parathyroid Hormone and Parathyroid Hormone-Related Protein 279 VIII. Structure of the PTH/PTHrP (PTH1R) Receptor 279
DAVID GOLTZMAN Abstract 293 Introduction 293 Molecular Biology and Mechanism of Action 294 The Skeletal Actions of PTHrP 298 Summary 301 References 302
I. II. III.
IV. V.
DAVID G. MONROE, THOMAS C. SPELSBERG, AND S. KHOSLA Abstract 327 Introduction 327 Molecular Structures, Synthesis, Mechanism of Action of Major Sex Steroids, and Transcriptional Coregulator Function 328 Effects of Sex Steroids on Bone Cells and Bone Turnover 331 Effects of Estrogens and Androgens on Bone Metabolism in Men Versus Women 335
viii
Contents
VI. Effects of Sex Steroids on Extraskeletal Calcium Homeostasis 336 VII. Summary 337 References 338
CHAPTER 20 Physiology of Calcium and Phosphate Homeostases I. II. III. IV. V.
VI. VII. VIII. IX. X. XI.
RENÉ RIZZOLI AND JEAN-PHILIPPE BONJOUR Abstract 345 Introduction 345 Body Distribution of Calcium 346 Determinants of Extracellular Calcium Concentration 346 Relative Importance of the Various Calcium Fluxes in Controlling Extracellular Calcium Homeostasis 349 Homeostatic Responses to Hypocalcemia 349 Calcium and Bone Growth 352 Body Distribution of Phosphorus 353 Determinants of Extracellular Phosphate Concentration 353 Homeostatic Responses to Changes in Phosphate Supply or Demand 356 Conclusions 357 References 357
CHAPTER 21 The Central Control of Bone Remodeling PAUL A. BALDOCK, SUSAN J. ALLISON, HERBERT HERZOG, AND EDITH M. GARDINER I. Introduction 361 II. Actions of Leptin 362 III. Sympathetic Nervous System 365 IV. Neuropeptide Y and the Y Receptors 369 V. Interaction between Leptin and Y2-Regulated Bone Antiosteogenic Pathways 371 VI. Concluding Remarks 373 References 373
CHAPTER 22 New Concepts in Bone Remodeling I. II. III. IV. V.
DAVID W. DEMPSTER AND HUA ZHOU Introduction 377 An Overview of the Remodeling Cycle 377 Functions of Bone Remodeling 380 The Role of Apoptosis in Regulating Bone Balance 381 Possible Mechanisms whereby a Reduction in Activation Frequency may Protect against Fracture 383 References 386
CHAPTER 23
Products of Bone Collagen Metabolism
JUHA RISTELI AND LEILA RISTELI I. Introduction 391 II. Products of Bone Collagen Synthesis, the Procollagen Propeptides 392 III. Degradation Products of Type I Collagen IV. Closing Remarks 402 References 403
397
CHAPTER 24 Supramolecular Structure of Cartilage Matrix I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.
PETER BRUCKNER Summary 407 Introduction 407 Light and Electron Micrography 408 Biochemistry of Cartilage 409 Studies of Fibril Structures by X-ray Diffraction 411 Structure of Fibril Fragments Obtained by Mechanical Disruption of Tissue 411 Studies of Collagen Cross-linking in Cartilage Fibrils 412 Reconstitution of Aggregates from Soluble Collagens and other Macromolecules 412 Studies of Transgenic Mice and of Human Genetic Matrix Diseases 414 Correlating Structure With the Biomechanical Role of Articular Cartilage 415 Models of Cartilage Fibril Structure 416 Future Perspectives 417 References 418
CHAPTER 25
Products of Cartilage Metabolism
DANIEL-HENRI MANICOURT, JEAN-PIERRE DEVOGELAER, AND EUGENE J.-M. A. THONAR I. Introduction 421 II. The Chondrocyte and its Extracellular Matrix 422 III. Products of Collagen Metabolism 424 IV. Products of Aggrecan Metabolism 429 V. Products of the Metabolism of other Proteoglycans 434 VI. Products of the Metabolism of Link Protein and Hyaluronan 435 VII. Other Products of Chondrocyte Metabolism 437 VIII. Concluding Statement 438 References 439
ix
Contents
CHAPTER 26 Fluid Dynamics of the Joint Space and Trafficking of Matrix Products PETER A. SIMKIN I. Introduction 451 II. Interpretation of Marker Data and Strategies for Dealing with Them 451 III. Conclusion 456 References 456
CHAPTER 27 Transgenic Models of Bone Disease I. II. III. IV.
BARBARA E. KREAM AND JOHN R. HARRISON Introduction 457 Generation of Mouse Models 457 Transgenic Models in Bone Biology 460 Perspectives and Future Directions 465 References 465
PART III Markers of Bone and Cartilage Metabolism
CHAPTER 30 Measurement of Parathyroid Hormone HARALD JÜPPNER AND GHADA EL-HAJJ FULEIHAN I. Abstract 507 II. Background 507 III. Different Immunometric Assays for the Detection of PTH 509 IV. Primary Hyperparathyroidism 510 V. Hyperparathyroidism in Renal Osteodystrophy 510 VI. Pseudohypoparathyroidism (PHP) 510 VII. Conclusion 511 References 511
CHAPTER 31 New Horizons for Assessment of Vitamin D Status in Man GARY L. LENSMEYER, NEIL BINKLEY, AND MARC K. DREZNER I. Measurement of Vitamin D3 (Cholecalciferol) and Vitamin D2 (Ergocalciferol) 513 II. Measurement of 25-Hydroxyvitamin D 513 III. Measurement of 1,25-Dihydroxyvitamin D 524 References 525
CHAPTER 28 The Role of Genetic Variation in Osteoporosis ANDRÉ G. UITTERLINDEN, JOYCE B.J. VAN MEURS, FERNANDO RIVADENEIRA, JOHANNES P.T.M.VAN LEEUWEN, AND HUIBERT A.P. POLS I. Abstract 471 II. Osteoporosis has Genetic Influences 471 III. Genome-wide Approaches to Find the Genes 473 IV. Association Analysis of Candidate Gene Polymorphisms 476 V. Haplotypes 478 VI. Meta-analyses 480 VII. Osteoporosis Candidate Genes: Collagen Type Iα1 and the Vitamin D Receptor 482 VIII. Summary 483 References 484
CHAPTER 29 Measurement of Calcium, Phosphate and Magnesium HEINRICH SCHMIDT-GAYK I. Measurement of Calcium 487 II. Measurement of Phosphate 495 III. Measurement of Magnesium 499 References 502
CHAPTER 32 Measurement of Biochemical Markers of Bone Formation I. II. III. IV. V. VI. VII.
KIM E. NAYLOR AND RICHARD EASTELL Abstract 529 Introduction 529 Propeptides of Type I Procollagen 530 Total Alkaline Phosphatase 533 Bone Alkaline Phosphatase 533 Osteocalcin 535 Discussion 537 References 537
CHAPTER 33 Measurement of Biochemical Markers of Bone Resorption MARIUS E. KRAENZLIN AND MARKUS J. SEIBEL I. Introduction 541 II. Collagen Related Markers 543 III. Non-Collagenous Proteins of the Bone Matrix 553 IV. Osteclast Enzymes 555 References 557
x
Contents
CHAPTER 34 Variability in the Measurement of Biochemical Markers of Bone Turnover
CHAPTER 38 Monitoring Anabolic Treatment
TUAN V. NGUYEN, CHRISTIAN MEIER, AND MARKUS J. SEIBEL Introduction 565 Sources of Pre-Analytical Variability in the Measurement of Biochemical Markers of Bone Turnover 566 Statistical Consideration of Variability 571 Summary 577 References 578
JOHN P. BILEZIKIAN AND MISHAELA R. RUBIN I. Introduction 629 II. Cellular and Regulatory Mechanisms of the Anabolic Actions of Parathyroid Hormone 630 III. Pharmacokinetics of Teriparatide in Human Subjects 631 IV. Actions of Parathyroid Hormone to Improve Bone Quality 632 V. Conclusions 642 References 642
CHAPTER 35 Validation of Biochemical Markers of Bone Turnover
CHAPTER 39 Monitoring of Antiresorptive Therapy
I. II.
III. IV.
I. II. III. IV.
KIM BRIXEN AND ERIK FINK ERIKSEN Introduction 583 Validation of Biochemical Markers by Calcium Kinetics 584 Validation of Biochemical Markers by Bone Histomorphometry 588 Conclusions 592 References 592
I. II. III.
IV.
CHAPTER 36 Genetic Markers of Joint Disease MICHEL NEIDHART, RENATE E. GAY, AND STEFFEN GAY I. Introduction 595 II. Ankylosing Spondylitis 595 III. Reactive Arthritis 597 IV. Rheumatoid Arthritis 598 V. Juvenile Rheumatoid Arthritis 605 VI. Conclusion 606 References 606
CHAPTER 37 Laboratory Assessment of Postmenopausal Osteoporosis PATRICK GARNERO AND PIERRE D. DELMAS I. Introduction 611 II. Postmenopausal Bone Loss 612 III. Management of Postmenopausal Osteoporosis 616 IV. Conclusion 624 References 624
CHRISTIAN MEIER, TUAN V. NGUYEN, AND MARKUS J. SEIBEL Introduction 649 Effects of Pretreatment Bone Turnover and Mineral Density on Therapeutic Outcomes The Role of Markers of Bone Turnover in Monitoring Antiresorptive Osteoporosis Therapy 651 Interpretation of Changes in Bone Turnover Markers 661 References 665
650
CHAPTER 40 Age-related Osteoporosis and Skeletal Markers of Bone Turnover I. II.
III. IV.
CLIFFORD J. ROSEN Introduction 671 Epidemiology and Pathogenesis of Age-Related Osteoporosis-Relationship to Bone Turnover 672 Markers of Bone Turnover and Age-Related Osteoporosis – Clinical Implications 679 Summary 683 References 683
CHAPTER 41 Steroid-Induced Osteoporosis I. II. III. IV. V.
IAN R. REID Summary 689 Introduction 689 Epidemiology 689 Pathogenesis 690 Effects of Glucocorticoids on Histological Indices of Bone Turnover 691
xi
Contents
VI. Effect of Glucocorticoids on Markers of Bone Turnover 692 VII. Evaluation of Steroid-Treated Patients 694 VIII. Treatment and Follow-up 694 References 695
IV. Summary 762 References 762
CHAPTER 46 Primary Hyperparathyroidism
CHAPTER 42 Transplantation Osteoporosis: Biochemical Correlates of Pathogenesis and Treatment CAROLINA A. MOREIRA KULAK AND ELIZABETH SHANE I. Introduction 701 II. Kidney Transplantation 702 III. Cardiac Transplantation 705 IV. Liver Transplantation 706 V. Lung Transplantation 706 VI. Bone Marrow Transplantation 707 VII. Mechanisms of Bone Loss After Transplantation 707 VIII. Prevention and Management of Transplantation Osteoporosis 709 IX. Conclusions 711 References 711
CHAPTER 43
Secondary Osteoporosis
JEAN E. MULDER, CAROLINA A. MOREIRA KULAK, AND ELIZABETH SHANE I. Introduction 717 II. Hyperthyroidism and Osteoporosis 717 III. Osteoporosis Secondary to Hypogonadism 724 IV. Anticonvulsant Drugs and Osteoporosis 730 References 732
SHONNI J. SILVERBERG AND JOHN P. BILEZIKIAN Introduction 767 Etiology 767 Clinical Presentation 768 Bone Markers in Primary Hyperparathyroidism 770 V. Cytokines in Primary Hyperparathyroidism 772 VI. Treatment of Primary Hyperparathyroidism 772 VII. Summary 775 References 775 I. II. III. IV.
CHAPTER 47
ANDREAS GRAUER, ETHEL SIRIS, AND STUART RALSTON I. Introduction 779 II. Etiology and Pathogenesis 779 III. Treatment 784 References 788
CHAPTER 48 I. II. III. IV. V.
CHAPTER 44 I. II. III. IV. V. VI.
Osteomalacia and Rickets
MARC K. DREZNER Definition 739 Etiology 741 Incidence and Epidemiology 742 Calciopenic Rickets and Osteomalacia 743 Phosphopenic Rickets and Osteomalacia 747 Normal Mineral Rickets and Osteomalacia 750 References 751
CHAPTER 45 Assessment of Bone and Joint Diseases: Renal Osteodystrophy ESTHER A. GONZÁLEZ, ZIYAD AL ALY, AND KEVIN J. MARTIN I. Introduction 755 II. Biochemical Assessment of Renal Osteodystrophy 757 III. Skeletal Imaging in Renal Osteodystrophy
Metastatic Bone Disease
JEAN-JACQUES BODY Abstract 793 Introduction 794 Use of Markers of Bone Turnover for the Diagnosis of Bone Metastases 796 Use of Markers of Bone Turnover for the Monitoring of Tumor Bone Disease 802 Prediction of the Development of Bone Metastases 806 References 806
CHAPTER 49 I. II. III. IV. V.
I. II. III. IV. V. VI.
Rare Bone Diseases
MICHAEL P. WHYTE Introduction 811 Osteopenia 812 Osteosclerosis and Hyperostosis Ectopic Calcification 823 Other Disorders 825 References 826
CHAPTER 50
761
Paget’s Disease of Bone
817
Osteogenesis Imperfecta
FRANCIS H. GLORIEUX AND FRANK RAUCH Introduction 831 Classification 832 Diagnosis 833 Differential Diagnosis 834 Pathogenesis 834 Bisphosphonate Therapy in OI 835
xii VII. Medical Therapies Other than Bisphosphonates 838 VIII. Potential Future Therapies 839 IX. Conclusions 839 References 839
CHAPTER 51 Rheumatoid Arthritis and other Inflammatory Joint Pathologies STEVEN R. GOLDRING AND MARY B. GOLDRING I. Abstract 843 II. Introduction 843 III. Effects of Joint Inflammation on Skeletal Remodeling 844 IV. Effects of Joint Inflammation on Cartilage Remodeling 852 V. Conclusion 858 References 858
Contents
CHAPTER 52 Osteoarthritis and Degenerative Spine Pathologies I. II. III. IV. V. VI. VII. VIII. IX. X. XI.
KRISTINA ÅKESSON Introduction 871 Characteristics of OA 872 Etiology 872 Treatment Options 873 Current Diagnostic Procedures 873 Biochemical Aspects of Osteoarthritis 873 Markers of Bone Turnover 874 Markers of Cartilage Metabolism 878 Spine Degeneration and Markers 882 Summary 883 Conclusions 884 References 884
Contributors
Kristina Åkesson Associate Professor, Department of Orthopedics, Malmö University Hospital, 205 02 Malmö, Sweden
John P. Bilezikian Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, NY, USA Departments of Medicine and Pharmacology, College of Physicians and Surgeons, Columbia University, New York, USA
Sari L. Alatalo Finnish Red Cross Blood Service, Helsinki, Finland Susan J. Allison Postgraduate Scholar, Bone Research Program, Garvan Institute of Medical Research, 384 Victoria Street, Sydney NSW 2010, Australia
Neil Binkley University of Wisconsin, Madison, Wisconsin, USA Jean-Jacques Body Dept of Internal Medicine and Endocrinology/ Bone Diseases Clinic, Institut J. Bordet, Univ. Libre de Bruxelles, Brussels, Belgium
Ziyad Al Aly Division of Nephrology, Saint Louis University School of Medicine, St. Louis, Missouri, USA
Jean-Philippe Bonjour Division of Bone Diseases, WHO Collaborating Center for Osteoporosis Prevention, Department of Rehabilitation and Geriatrics, University Hospitals, CH - 1211 Geneva 14 (Switzerland)
Paul A. Baldock Senior Research Officer, Bone Research Program, Garvan Institute of Medical Research,384 Victoria Street, Sydney NSW 2010, Australia John P. Bilezikian Departments of Medicine and Pharmacology, College of Physicians and Surgeons, Columbia University, New York, NY
Adele L. Boskey Starr Chair in Mineralized Tissue Research, Hospital for Special Surgery, New York, NY 10021 and Weill Medical College and Graduate School of Medical Sciences of Cornell University, New York, NY 10021
xiii
xiv Roger Bouillon Laboratorium for Experimental Medicine and Endocrinology, K. U. Leuven, Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium Kim Brixen Department of Endocrinology, Odense University Hospital, DK-5000 Odense C, Denmark Peter Bruckner Department of Physiological Chemistry and Pathobiochemistry, University of Münster, Münster, Germany Geert Carmeliet Laboratorium for Experimental Medicine and Endocrinology, K. U. Leuven, Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium Ian M Clark School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, UK Peter Croucher Academic Unit of Bone Biology, University of Sheffield Medical School, Sheffield S10 2RX, United Kingdom Pierre D. Delmas Professor of Medicine, Université Claude Bernard, Lyon, France and Director INSERM Research Unit 403, Lyon, France David W. Dempster Regional Bone Center, Helen Hayes Hospital, West Haverstraw, New York, USA Jean-Pierre Devogelaer Department of Rheumatology, Saint Luc University Hospital, Université Catholique de Louvain, 1200 Brussels, Belgium Marc K. Drezner Professor of Medicine, University of Wisconsin, Madison, Wisconsin, USA Richard Eastell From the Academic Unit of Bone Metabolism, University of Sheffield, Sheffield, UK Guy Eelen Laboratorium for Experimental Medicine and Endocrinology, K. U. Leuven, Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium
Contributors
Erik Fink Eriksen Novartis Pharma, Basel, Switzerland Lorraine A. Fitzpatrick Global Development, Amgen, Thousand Oaks, CA Ghada El-Hajj Fuleihan Calcium Metabolism and Osteoporosis Program, American University of Beirut-Medical Center, Beirut, Lebanon Edith M. Gardiner Associate Professor, School of Medicine, The University of Queensland, Head, Skeletal Biology Unit, Centre for Diabetes & Endocrine Research, Ground Floor, C Wing, Bldg 1, Princess Alexandra Hospital, Ipswich Road, Brisbane QLD 4102, Australia Patrick Garnero Research Scientist, INSERM research unit 403 and Vice-President Synarc Molecular Marker Division, Lyon, France Renate E. Gay Research Scientist, INSERM research unit 403 and Vice-President Synarc Molecular Marker Division, Lyon, France Steffen Gay Research Scientist, INSERM research unit 403 and Vice-President Synarc Molecular Marker Division, Lyon, France Francis H. Glorieux Genetics Unit, Shriners Hospital for Children, 1529 Cedar Avenue, Montréal, Québec, Canada H3G 1A6 Mary B. Goldring Medical Center, Harvard Medical School, Boston, MA; New England Baptist Bone and Joint Institute, Boston, MA 02215 Steven R. Goldring Department of Medicine, Rheumatology Division, Beth Israel Deaconess David Goltzman Calcium Research Laboratory, Department of Medicine, Royal Victoria Hospital, McGill University, Montreal, H3A 1A1, Canada
xv
Contributors
Esther A. González Division of Nephrology, Saint Louis University School of Medicine, St. Louis, Missouri, USA
Helena Kaija Research Center for Molecular Endocrinology, University of Oulu, Finland
Andreas Grauer Procter & Gamble Pharmaceuticals Mason, OH, USA
S. Khosla Division of Bone Diseases, WHO Collaborating Center for Osteoporosis Prevention, Department of Rehabilitation and Geriatrics, University Hospitals, CH - 1211 Geneva 14 (Switzerland)
Caren M. Gundberg Department of Orthopaedics and Rehabilitation, Yale University School of Medicine, New Haven, Connecticut 06510 Tim Hardingham Wellcome Trust Centre for Cell-Matrix Research, Faculty of Life Sciences, University of Manchester, Manchester M13 9PT, United Kingdom
Marius E. Kraenzlin Division of Endocrinology, Diabetology and Clinical Nutrition, University Hospital Basel, Petersgraben 4, CH-4031 Basel, Switzerland Barbara E. Kream Departments of Medicine and Genetics and Developmental Biology, University of Connecticut Health
John R. Harrison Division of Orthodontics, University of Connecticut Health Center, Farmington, CT 06030
Center, Farmington, Ct 06030
Dick Heinegård Departments of Experimental Medical Science and Clinical Science, BMC plan C12, SE-22184, Lund, Sweden
Carolina A. Moreira Kulak Department of Endocrinology, Federal University of Parana, Hospital de Clinicas, Curitiba, Brazil Division of Endocrinology and Metabology of Hospital de Clinicas, Federal university of Parana (SEMPR), Curitiba-/Brazil
M. H. Helfrich Department of Medicine and Therapeutics, University of Aberdeen, Institute of Medical Sciences, Foresterhill, Aberdeen AB25 2ZD, United Kingdom
Johannes P.T.M. van Leeuwen Department of Epidemiology & Biostatistics, Erasmus Medical Centre, Rotterdam, The Netherlands
Herbert Herzog Adjunct Professor, Faculty of Medicine, The University of New South Wales, Principal Research Fellow, Head Obesity and Energy Homeostasis Research Group, Director Neurobiology Program, Garvan Institute of Medical Research, 384 Victoria Street, Sydney NSW 2010, Australia
Gary L. Lensmeyer University of Wisconsin, Madison, Wisconsin, USA
M. A. Horton Bone and Mineral Centre, Department of Medicine, The Rayne Institute, London WC1E 6JJ, United Kingdom
Joseph A. Lorenzo Director, Bone Biology Research, Professor of Medicine, University of Connecticut Health Center, 263 Farmington Avenue, MC 1317, Farmington, CT 06030–1317.
Philippa Hulley Botnar Research Centre, Institute of Musculoskeletal Sciences, University of Oxford, Oxford OX3 7LD, United Kingdom
Pilar Lorenzo Departments of Experimental Medical Science and Clinical Science, BMC plan C12, SE-22184, Lund, Sweden
Harald Jüppner Endocrine and Pediatric Nephrology Units, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA
Christa Maes Laboratorium for Experimental Medicine and Endocrinology, K. U. Leuven, Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium
Jane B. Lian University of Massachusetts Medical School, Department of Cell Biology, 55 Lake Avenue North, Worcester, MA 01655, USA
xvi Daniel-Henri Manicourt Laboratoire de Chimie Physiologique (Metabolic Research Group, Connective Tissue Section), Christian de Duve Institute of Cellular Pathology and Department of Rheumatology, Saint Luc University Hospital, Université Catholique de Louvain, 1200 Brussels, Belgium Klaus von der Mark Dept. of Experimental Medicine and Connective Tissue Research, Friedrich-Alexander, University of Erlangen Nuremberg, Germany Kevin J. Martin Division of Nephrology, Saint Louis University School of Medicine, St. Louis, Missouri, USA T. John Martin St. Vincent’s Institute of Medical Research, 9 Princes Street, Fitzroy 3065, Australia
Contributors
Tuan V. Nguyen Bone and Mineral Research Program, Garvan Institute of Medical Research, St. Vincent’s Hospital and University of NSW Research Program University of New South Wales, Sydney, Australia Satoru K. Nishimoto Department of Molecular Sciences, The University of Tennessee College of Medicine, Memphis, Tennesse 38163 Lila O.T. Patrikainen Research Center for Molecular Endocrinology, University of Oulu, Finland Huibert A.P. Pols Department of Endocrinology, Laboratory Group, Im Breitspiel 15, 69126 Heidelberg, Germany
Christian Meier Endokrinologische Praxis & Labor, University Hospital Basel, Switzerland
Lawrence G. Raisz Interim Director, Musculoskeletal Institute, Board of Trustees Distinguished Professor of Medicine, University of Connecticut Health Center, 263 Farmington Avenue, MC-3805, Farmington, CT 06030
Joyce B.J. van Meurs Department of Epidemiology & Biostatistics, Erasmus Medical Centre, Rotterdam, The Netherlands
Stuart Ralston Rheumatic Diseases Unit, University of Edinburgh, Edinburgh, UK
José Luis Millán The Burnham Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037-1005, USA
Frank Rauch Genetics Unit, Shriners Hospital for Children, 1529 Cedar Avenue, Montréal, Québec, Canada H3G 1A6
David G. Monroe Biochemistry and Molecular Biology, Mayo Clinic College of Medicine, Rochester, MN
Ian R. Reid Department of Medicine, University of Auckland, Private Bag 92019, Auckland, New Zealand
Jean E. Mulder Department of Medicine, Brigham and Women’s Hospital, Harvard University, Boston, MA 02115
Juha Risteli Department of Clinical Chemistry, FI-90014 University of Oulu, Oulu, Finland
Gillian Murphy Dept of Oncology, Cambridge University, Cambridge CB2 2XY, UK
Leila Risteli Department of Clinical Chemistry, FI-90014 University of Oulu, Oulu, Finland
Kim E. Naylor From the Academic Unit of Bone Metabolism, University of Sheffield, Sheffield, UK
Fernando Rivadeneira Department of Epidemiology & Biostatistics, Erasmus Medical Centre, Rotterdam, The Netherlands
Michel Neidhart Research Scientist, INSERM research unit 403 and Vice-President Synarc Molecular Marker Division, Lyon, France
xvii
Contributors
René Rizzoli Division of Bone Diseases, WHO Collaborating Center for Osteoporosis Prevention, Department of Rehabilitation and Geriatrics, University Hospitals, CH - 1211 Geneva 14 (Switzerland) Simon P. Robins Matrix Biochemistry Group, Rowett Research Institute, Bucksburn, Aberdeen AB21 9SB Clifford J. Rosen Maine Center for Osteoporosis Research and Education, St. Joseph Hospital, 900 Broadway, Bldg #2, Bangor, Maine 04401 USA Mishaela R. Rubin Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, NY, USA Graham Russell Botnar Research Centre, Institute of Musculoskeletal Sciences, University of Oxford, Oxford OX3 7LD, United Kingdom Tore Saxne Departments of Experimental Medical Science and Clinical Science, BMC plan C12, SE-22184, Lund, Sweden Heinrich Schmidt-Gayk Department of Endocrinology, Laboratory Group, Im Breitspiel 15, 69126 Heidelberg, Germany
Natalie A. Sims University of Melbourne, Department of Melbourne, St. Vincent’s Health, 41 Victoria Pde, Fitzroy 3065, Australia Ethel Siris Department of Medicine, Columbia University College of P&S, New York, NY, USA Thomas C. Spelsberg Professor of Biochemistry, Mayo Clinic College of Medicine, Rochester, MN Gary S. Stein University of Massachusetts Medical School, Department of Cell Biology, 55 Lake Avenue North, Worcester, MA 01655, USA Eugene J.-M. A. Thonar Departments of Biochemistry, Internal Medicine and Orthopedic Surgery, Rush Medical College, RushPresbyterian-St. Luke’s Medical Center, Chicago, Illinois, 60612 USA André G. Uitterlinden Department of Internal Medicine, H. Kalervo Väänänen Institute of Biomedicine, Department of Anatomy, University of Turku, Turku, Finland
Ego Seeman Dept of Endocrinology and Medicine, Austin Hospital, University of Melbourne, Melbourne, Australia.
Annemieke Verstuyf Laboratorium for Experimental Medicine and Endocrinology, K. U. Leuven, Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium
Markus J. Seibel Bone Research Program, ANZAC Research Institute, University of Sydney and Concord Hospital, Sydney, Australia
Pirkko T. Vihko Department of Biological and Environmental Sciences, Division of Biochemistry, University of Helsinki, Finland
Elizabeth Shane Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, New York 10032
Michael P. Whyte Center for Metabolic Bone Disease and Molecular Research, Shriners Hospitals for Children; and Division of Bone and Mineral Diseases, Washington University School of Medicine At Barnes-Jewish Hospital; St. Louis, Missouri
Shonni J. Silverberg Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, USA Peter A. Simkin Professor of Medicine, Adjunct Professor of Orthopaedics, University of Washington
Hua Zhou Regional Bone Center, Helen Hayes Hospital, West Haverstraw, New York, USA
This Page Intentionally Left Blank
Preface to the Second Edition
The first edition of “Dynamics of Bone & Cartilage Metabolism”, published at the very end of the last century, in 1999, was successfully received by biomedical scientists in the bone field and by clinicians throughout the world. Since the first edition was published, research in bone and cartilage metabolism has progressed at a rapid pace leading to new insights into basic science as well as new ways in which markers of bone and cartilage metabolism can be used clinically. This second edition of “Dynamics of Bone & Cartilage Metabolism” incorporates these advances while maintaining the general structure of the first edition. It is a thorough update with all chapters either extensively revised or completely rewritten. To reflect the changing climate of knowledge, twelve new chapters have been added that, we believe, greatly enhance the substance and completeness of the book. The topics of these new chapters include: “Acid Phosphatases”, “Bone Structure, Architecture and Strength”, “Signalling Mechanisms in Bone”, “The Central Control of Bone Remodelling”, “Transgenic Models of Bone Disease”, “Models of Cartilage Metabolism and Disease”, “Measurement of
Parathyroid Hormone”, “Measurement of Vitamin D”, “Variability of Bone Markers”, “Monitoring of Anabolic Treatment”, “Monitoring of Anti-resorptive Treatment” and “Osteogenesis Imperfecta”. These new chapters add greatly to updated chapters and together provide a complete repository of information on this subject. We are grateful to all authors for their efforts to deliver their new or revised chapters within the time constraints, always a challenge. We are also grateful to the excellent staff at Elsevier-Academic Press, Tari Broderick, Karen Dempsey and Renske van Dijk, who helped us so enthusiastically throughout the preparation of this new edition of Dynamics of Bone and Cartilage Metabolism. MARKUS SEIBEL, Sydney SIMON ROBINS, Aberdeen JOHN BILEZIKIAN, New York June 2006
xix
This Page Intentionally Left Blank
Part I
Components of the Organic Extracellular Matrix of Bone and Cartilage
This Page Intentionally Left Blank
Chapter 1
Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone Klaus von der Mark
Department of Experimental Medicine and Connective Tissue Research, Friedrich-Alexander, University of Erlangen Nuremberg, Germany
VI. Collagen Genes and Transcriptional Regulation VII. Factors Regulating Collagen Biosynthesis VIII. Conclusions References
I. Introduction II. The Collagen Families III. Bone Collagens IV. Cartilage Collagens V. Collagen Biosynthesis
I. INTRODUCTION
collagen which may provide additional elasticity [8–11]. In addition, cartilage contains minor amounts of other collagen types depending on the cartilage type and location (see below). The basic function of collagens in cartilage and bone is to provide the structural scaffold to tissues into which minerals, proteoglycans, and glycoproteins can be firmly incorporated, thus being responsible for the unique physiological and mechanical properties of these tissues. But on top of biomechanical functions, collagens play an important role in all tissues, including cartilage and bone as biological substrates for cell adherence. Essential cell biological functions such as proliferation, cytoskeletal organization, migration, differentiation, and apoptosis are regulated by collagens, mediated by transmembrane receptors of the integrin and syndecan families [12]. Since collagens provide the major organic component with 90% of the dry mass in bone, or 60% in cartilage, respectively, it is obvious that defects in structure,
Collagens provide the structural framework of bones and cartilages and hold responsibility for shape and most of the biomechanical properties such as resistance to pressure, torsion, and tension [1]. In vertebrates, 27 genetically distinct collagen types with rather diverse structural and biochemical features have been identified, but only about half of them are represented in cartilage and bone [2–4] (Table I). Their specific functions in the tissues are only partially known. In cartilage and bone, fibril-forming collagens are dominant: the bone matrix consists basically of two collagen types, about 95% type I and 5% type V collagen which are assembled into heterofibrils [5]. Similarly, the backbone of all cartilages is made of types II/XI collagen heterofibrils which are decorated with so-called FACIT collagen types IX, XII, or XIV (see below) [3, 4, 6, 7]. These fibrils are interwoven with a microfibrillar mesh made of type VI Dynamics of Bone and Cartilage Metabolism
3
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
4
KLAUS VON DER MARK Table I.
Collagens in Cartilage and Bone
Type
Subunits
I
α1(I) α2(I)
α1(I)2 α2
I II
α1(I) α1(II)
[α1(I)]3 [α1(II)]3
III
α1(III)
[α1(III)] 3
V
[α1(V)]3 [α1(V)]2 α3(V)
XXIV
α1(V) α2(V) α3(V) α1(XI) α2(XI) α3(XI) α1(XXIV)
[α1(XI) α2(XI) α3(XI)] [α1(XI)]2 α2(V) n.d.
Bone, eye
XXVII
α1(XXVII)
n.d.
Cartilage; eye and ear
Forms fibers of high tensile strength; most abundant collagen Rare form Major cartilage collagen; forms heterofibrils with Col IX + XI Often in mixed fibrils with type I collagen; present in reticular fibers and elastic tissues; cystine bridges in triple helix Propeptide partially retained in the fibrils; forms hetero-fibrils with type I collagen controls fibril diameter Homologous to Col V; nucleates and controls cartilage coll. fibril formation; α3(XI) same gene as for α1(II) Similar to type V collagen, contains TSP motif in N-propeptide Col27a1 gene 156 kb, 61 exons
VI
α1(VI) α2(VI) α3(VI)
[α1(VI) α2(VI) α3(VI)]
Microfibrillar collagen Widespread, in cartilage (pericellular), intervert, disk dermis, placenta, lung vessel wall
Contains vWF and Kunitz type protein inhibitor domains; forms beaded filaments; highly disulfide crosslinked
X
α1(X)
[α1(X)]3
IX
[α1(IX) α2(IX) α3(IX)]
XII
α1(IX) α2(IX) α3(IX) α1(XII)
[α1(XII)]3
XIV
α1(XIV)
[α1(XIV)]3
XVI
α1(XVI)
[α1(XVI)]3
XX XXI XXII
α1(XX) α1(XXI) α1(XXII)
XI
Molecular forms
Tissue distribution Fibril-forming collagens Bone, dermis, tendon, ligaments cornea, most other tissues Dermis, dentin Cartilage, notochord, vitreous body embryonic epithelia, retina Reticular fibers of most tissues (lung, liver, dermis, spleen, vessel wall, etc.) In vitro: hamster lung cell cultures lung, cornea, bone, fetal membranes; together with Col I Cartilage, vitreous body Bone
Network forming, short chain collagens Hypertrophic cartilage
FACIT collagens Cartilage, vitreous humor Splice variant without NC-4 domain in cornea Perichondrium, ligaments, tendon Cartilage, dermis, tendon, vessel wall, placenta, lung, liver Cartilage (territorial matrix) papillary dermis Cornea; sternal cartilage Blood vessel wall Myotendinous junction, articular cartilage surface
biosynthesis, assembly, or turnover of collagens generally will lead to severe diseases such as osteoporosis, osteoarthritis, chondrodysplasias, or osteogenesis imperfecta. In order to understand the specific functions of the various collagens in cartilage and bone and the reasons for their failure in connective tissue diseases, it is necessary to have a close look at the specific structural and biochemical features of the collagens. An overview on the structural and functional features of cartilage and bone collagens, and their interactions with other matrix proteins and cell receptors, will therefore be given in the first part of this chapter. In order to understand the dynamics of collagen metabolism in bone and cartilage, it is helpful to gain
Characteristic features
Strong inter- and intramolecular interactions between NC1-domains Mutations in Col X-NC-1 →SMCD
Covalently linked to type II collagen fibrils; NC4 domain projects into cartilage matrix; contains chondroitin sulfate Large cruciform shaped NC3 domain; associated with type I collagen fibrils Associated with type I collagen Integrates into discrete Col II/XI fibrils Less FN III repeats than Col XIV TSP and vWFA domain, but no FNIII rep. Present only in tissue junctions
insight into the various levels of transcriptional and translational regulation of collagen biosynthesis and turnover. Therefore, in the second part of this chapter, the various steps of collagen biosynthesis and post-translational modifications will be summarized, and an overview is given on the structure of the collagen genes and their cis-acting regulatory element. How these are regulated by growth factors, cytokines, hormones, and transcription factors is one of the current challenges in understanding dynamics of connective tissue turnover and homeostasis. The scope of this chapter does not permit a detailed and complete presentation of our current knowledge on structure, biosynthesis, and regulation of collagens in bone
5
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
and cartilage. Fortunately, a number of excellent review articles and book chapters are available, such as the recent, most comprehensive book chapter on collagens by Kielty and Grant [2] and many other articles dealing with structure and function of collagen types [6, 13–16], with collagen gene [17–19], collagen biosynthesis and regulation [20–23], collagen degradation [24, 25], and collagenrelated diseases [26–29]. No further information can be given here on collagens which do not play a major role in cartilage and bone such as types IV, VII, VIII, and XIII–XIX. Aspects of extracellular assembly to supramolecular structures, in particular fibril formation and cross-link formation, will be dealt with in Chapters 2, 11, and 20. Collagen degradation by proteases will be covered in Chapter 10 on matrix proteases.
II. THE COLLAGEN FAMILIES Up to 2004, at least 27 genetically distinct collagen types had been identified in mammals, encyphered by 42 genes coding for their subunits [2, 6, 15] (Table I). Based on molecular structure and sequence homology, collagens have been grouped into seven or eight different families, including the fibril-forming collagens, FACIT collagens, microfibrillar collagens, network-forming collagens, transmembrane collagens, multiplexins, and others (see Table I). Since the first edition of this book seven new collagen types have been discovered, partially by screening sequence data banks for homology with collagen and procollagen peptide sequences [30–37]. Four of these new collagens (types XX, XXI, XXII, and XXIV) were found also in cartilage, but their functions are not yet known. All collagens consist of one or several collagenous, triple-helical domains, flanked or interrupted by noncollagenous domains which are largely removed by proteolytic processing in fibrillar collagen type I, II, and III procollagen. They are retained, however, in the mature molecule in most non-fibril-forming collagens.
A. The Collagen Triple Helix The key feature of all collagens is the triple helix, a coiled-coil structure in the form of a right-handed helix of 1.5 nm diameter, composed of three polypeptide chains (α-chains) [38–40] (Fig. 1). A structural requirement for the assembly of polypeptide chains into a collagen triple helix is the occupation of every third position by a glycine residue, resulting in the (Gly-X-Y)n repeat structure characterizing all collagens. The α-chains form a stretched, left-handed helix with a pitch of 18 amino acid residues per turn [41] and assemble around a central axis in a manner allowing all glycine residues to be positioned in the center of the triple helix. The more bulky side chains of the other amino acids in the X and Y position occupy the outer positions, where they are available for lateral interactions with adjacent collagen molecules to form fibrils. Another typical feature of the collagen triple helix is the high content of proline and hydroxyproline (ca. 20%). The hydroxyl groups of 4-hydroxy-proline are essential for the formation of intramolecular hydrogen bonds and thus critically determine the thermal stability of each collagen triple helix [40, 42]. The melting temperature of the type I collagen triple helix is 39⬚C at neutral pH, but can be considerably higher, e.g. 46⬚C in chicken type X collagen [43], or 65⬚C in the aminoterminal 7S domain of type IV collagen [44]. Triple helical domains vary considerably in their length: in fibril-forming collagens they span 300 nm, corresponding to 1000 amino acid residues per processed α-chain, while in other collagens, such as the multiplexins, they may include only nine triplets. Interruptions of the Gly-X-Y-Gly-X-Y- structure by one amino acid residue causes flexibility in the triple helix and renders the helix susceptible to proteolytic attack, e.g. in collagen type IV [45]. The native or denatured state of a triple helix may be measured by optical rotation, circular dichroism or resistance to proteases like pepsin, trypsin, or chymotrypsin [46]. For example, native type I collagen molecules have an optical rotation of ε = –1000⬚ at 405 nm, which drops
p t
p
Figure 1 Type I procollagen as a prototype of fibril-forming collagens. In types I, II, and III procollagens N- and C-terminal propeptides are removed after secretion by specific proteases, in types XI and V the N-propeptides are larger (see Fig. 3) and only partially cleaved. The C-propeptides contain AsN-linked mannose-rich oligosaccharides, while the triple helical part contains only hydroxylysine-linked glucosyl-galactosyl disaccharides or monosaccharides.
6
KLAUS VON DER MARK
to –336⬚ after denaturation. The resistance of the native triple helix to most proteases has been the basis for almost all biochemical isolation procedures of collagens in the past, using pepsin to destroy non-collagenous proteins while leaving the collagen triple helices intact. Triple helical collagenous domains are also found in proteins such as C1q, acetylcholine esterase and MARCO, a macrophage scavenger receptor [47, 48].
NC-domains of IX, XI, XII, and XVI collagens. Many noncollagenous domains, e.g. in collagen VI and XI, contain thrombospondin- and von Willebrand-factor-like domains or fibronectin-type III repeats [59]. In the non-fibril-forming collagens, the noncollagenous domains often show more specific and important structural and functional features than the triple helical domains. Thus, the C-propeptide of type XVIII collagen is retained after cleavage as a protein with antiangiogenic properties (endostatin) [60].
B. Noncollagenous Domains In the various collagens the triple helical domains are flanked or interrupted by noncollagenous domains. While the triple helix is a highly conserved structural protein element like the α-helix or the β-pleated sheet, there is a wide structural and functional diversity among the noncollagenous domains of the different collagen families, often bearing essential functions specific for each collagen. Fibril-forming collagens are synthesized as procollagens with a triple helix of 300 nm length, which is flanked by noncollagenous propeptides (NC-domains) at both ends [2, 49]. The globular C-propeptides, consisting of about 250 amino acid residues per α-chain, are all homologous and serve as a nucleation site for chain assembly and triple helix formation, while the N-propeptides regulate the fibril size when retained in the molecule [50, 51]. These procollagen peptides are largely removed by specific proteases after secretion, a prerequisite for fibril formation [52]. The C-propeptide of type II collagen remains in cartilage after cleavage from the procollagen molecule as a stable, hydroxyapatite-binding molecule (“chondrocalcin”) and may participate in cartilage calcification [53]. The C-terminal NC-1 domain of type X collagen, a highly compact and stable, bell-shaped trimer, also binds calcium and is highly homologous to TNFα [54], but does not bind to the TNFα receptor; its role is rather structural, serving as a nucleation site for triple helix assembly as well as for network assembly [55, 56]. In “FACIT” collagens (fibril-associated collagens with interrupted triple helices) [57], which include collagen types IX, XII, XIV, XVI, XIX, XX, XXI, and XXII, the collagenous domains are interrupted by two or three noncollagenous domains, while the aminoterminal domains form large, cross-shaped entities anchored in the extracellular matrix. In the more recently discovered “multiplexins” (collagens with multiple triple helix interruptions, [58] see Table I) the collagenous domains are short and separated by nine or ten noncollagenous domains. Despite the differences between fibril-forming and FACIT collagens, there is also sequence homology among the N-terminal
C. Structural and Functional Diversity of Collagens There is considerable complexity and diversity in the structure of the different collagen types, their splice variants, their triple helical and nontriple helical domains, and their assembly into extracellular matrix structures: fibrils, flexible meshworks, hexagonal sheets, beaded filaments, anchoring fibrils and perhaps other, yet unknown, structures [15]. The fibril-forming collagens represent with seven members (including the recently discovered collagen XXIV and XVII) and about 90% of the total collagens, the most abundant and widespread family of collagens in vertebrates. Type IV collagens with a more flexible triple helix assemble into supercoiled chicken-wire like meshes that are restricted to basement membranes. Types VIII and X collagens are short-chain collagens which form hexagonal sheets. Type VI collagen is highly disulfide cross-linked and assembles into a meshwork of beaded filaments interwoven with collagen fibrils. Collagens IX, XII, and XIV associate as single molecules with collagen fibrils (FACITcollagens) [3, 7], while others (Col XIII and XVII ) span cell membranes [61–63]. Little is known on the extracellular architecture of the more recently discovered collagens, except for collagens XXIV and XXVII, which have all the features of fibril-forming collagens [37]. Owing to their tensile strength and torsional stability, the major function of fibril-forming collagens is to support tissue architecture and stability. Despite structural similarity, however, the fibrillar collagen types I, II, and III have different, specific functions in different tissues, different immunological properties and show specific interactions with different cell types. For example, replacement of type II collagen by the similar type I collagen in cartilage, e.g. in joint repair by fibrous cartilage, causes severe loss of cartilage-specific features of the tissue. Initial concepts of specific collagen functions were mostly derived from their distribution in tissues and from in vitro experiments, i.e. from cell and organ culture studies. Recently, there is ample evidence for specific cell biological functions of
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
7
fibrillar collagens, such as types I and II collagen, in serving as substrates for cell adhesion, proliferation, migration, and differentiation, mediated by β1 integrins [12, 64–67]. More recently, natural and artificially introduced mutations in human and animal collagen genes, and the inactivation of collagen genes by homologous recombination in mice (“knockout mice”) [68–70], have provided valuable information on the role of some collagens in cartilage and bone. Surprising results of gene knockout experiments in mice or from human mutations have often caused a revision of common opinions on the function of various collagens; thus, after inactivation of the type II collagen gene Col2a1 in mice, rather normal development of long bones and of the eye was observed, although early expression of Col2a1 in embryonic corneal and retinal epithelia [71–73] and several in vitro studies had strongly suggested a critical role of collagen in early epithelial–mesenchymal interactions [71, 74]. The lack of type II collagen in the notochord of the Col2a1−/− mice did not prevent somite differentiation as expected [75], but the resorption of the notochord during development of the spine [69]. More and valid information on the function of individual collagens and their domains have been obtained from targeted mutations in distinct domains and tissue-specific inactivation (conditional knockout) of collagen genes.
a quarter-staggered heterofibrils with diameters between 25 and 400 nm (Fig. 2). Fibrils thicker than 50 nm show a characteristic banding pattern in the electron microscope with a periodicity of 65–67 nm (D-period) [16, 77, 78]. Adachi and Hayashi [79] have shown that inclusion of type V collagen controls the fibril diameter of type I collagen fibrils; in embryonic chick cornea, for example, a content of 20% type V collagen limits the fibril diameter to 25 nm [80]. In contrast, collagen fibrils in bone with a content of ca. 5% type V collagen reach diameters of 400 nm or more [76, 81]. Embedded in hydroxyapatite crystals and various bone-specific phosphoproteins and glycoproteins and SLRPs (small leucine-rich proteoglycans) such as osteoadherin (see Chapter 3), type I/V heterofibrils reveal unmatched biomechanical properties concerning load bearing, tensile strength, and torsional resistance. They serve also as a nucleation site for hydroxyapatite crystals [82] (see Chapter 12). In the osteons of compact bone, collagen fibrils seem to run parallel in two nearly orthogonal directions, forming twisted, nearly rectangular plywood-like layers [83]. As the assembly of collagen molecules into fibrils will be dealt with in detail in Chapter 2, here the structure of the bone collagen molecules, their biosynthesis and regulation, are focused on.
III. BONE COLLAGENS
A. Type I Collagen
The organic mass of the bone matrix comprises about 90% of type I and 5% of type V collagen [76], the remainder being bone-specific phospho- and glycoproteins such as osteopontin, bone sialoprotein, osteocalcin, osteonectin, and others. In bone, type I and V collagen assemble into
1. MOLECULAR STRUCTURE AND TISSUE DISTRIBUTION
Figure 2
Type I collagen is the most abundant, longest-known and best-studied collagen in vertebrates. It forms 90% of the organic mass of bone and tendon and is the major collagen of skin, ligaments, cornea, and many interstitial
Packing of types II, XI, and IX collagen in the cartilage collagen heterofibril, showing the location of the covalent cross-links between type II and IX collagen. The globular NC4 domains of type IX collagen reach out of the fibril. (From: D. Eyre (2001) Collagen of articular cartilage. Arthritis Res. 4, 30–35, with kind permission by BioMed Central, Ltd.)
8 connective tissues. Much of our information on biochemical and biophysical properties, cross-linking, and biosynthesis of collagens is based on research on this collagen, but may be applied to other collagens. The human type I procollagen is made of 2 proα1 chains of 1464 amino acid residues [84] and a somewhat shorter pro α2 subunit (1366 amino acid residues) (Fig. 1) [85]. It is synthesized in large quantities by fibroblasts, osteoblasts [86], and odontoblasts [87], and to a lesser extent by nearly all other tissue cells [12]. Although purified type I collagen can be reconstituted to crossbanded fibrils in vitro, in vivo type I collagen is always incorporated into heterofibrils containing either type III collagen, e.g. in skin and reticular fibers [88], or type V collagen in bone, tendon cornea and other tissues [76, 89] or both. Type I/III collagen heterofibrils are a constituent of reticular fibers of most parenchymal tissues such as lung, kidney, liver, muscle, or spleen, with the exception of hyaline cartilage, brain, and vitreous humor [90]. The key role of type I collagen in bone is most evident from mutations in the human type I collagen genes COL1A1 and COL1A2 as the cause of osteogenesis imperfecta (OI), a group of hereditary disorders characterized by a decrease in bone mass, enhanced bone fragility and multiple fractures (see also Chapter 54; for reviews see [28, 91]. The severity and progress of the disease is rather variable and ranges from mild forms to more severe and lethal forms. The mode of inheritance of the OI is in most cases autosomal dominant. More than 160 different mutations have been reported in the COL1A1 and COL1A2 genes, located on chromosomes 17q21.3 and 7q21.3–q22, respectively [28]. Most mutations affect glycine residues, leading to impaired triple helix formations, even in heterozygotes, but also exon skipping, frameshift mutations, RNA splicing mutations and basepair deletions or insertions have been identified. Generally it is difficult to predict the severity of the phenotype from the type of mutation. The general rule is that apparently mild mutations affecting the stability of the triple helix, owing to a glycine substitution, result in a more severe phenotype than entire exon deletions, allowing the formation of intact, although shortened, triple helices. The reason is that one affected α chain with a triple helical interruption may exert a dominant negative effect and impair seven type I collagen molecules. Furthermore, impaired triple helix formation in the rough endoplasmic reticulum leads to over-hydroxylation of proline and lysine residues [92, 93]. Premature chain terminations or deletions of exons in one COL1A1 allele which do not affect triple helix assembly may cause haplo-insufficiency, but will still allow the formation of intact collagen I molecules from the unaffected allele [28, 91].
KLAUS VON DER MARK 2. PHYSIOLOGICAL FUNCTIONS
Besides its biomechanical properties, type I collagen is important as adhesive substrate for many cells and plays a major role in tissue and organ development, in cell migration, proliferation and differentiation, in wound healing, tissue remodeling, and hemostasis [12, 74]. For example, in vitro studies have shown that many epithelial and endothelial cells acquire a polar cell shape and develop a luminal structure when cultured in a three-dimensional hydrated collagen lattice [94]. Cells recognize native type I collagen via α1β1, α2β1, and α11β1 integrins [65, 95] which are transmembrane receptors and confer signals from the extracellular matrix to intracellular signal cascades and to the actin cytoskeleton [96–98]. Furthermore, α11β1 integrin is able to organize the assembly of extracellular type I collagen molecules into fibrils [99]. Thus, reconstituted hydrated lattices consisting of native type I collagen fibrils are widely used for cell and tissue culture purposes, allowing cells to migrate, proliferate, and differentiate in a native, three-dimensional environment. Similarly, freezedried collagen sponges find wide applications in surgery and tissue engineering as scaffolds for wound and tissue repair by supporting adhesion and invasion of connective tissue cells [100–102]. Valuable and important information on the role of type I collagen has been gained from the MOV13 mouse strain, in which the expression of α1(I) chains is blocked owing to an insertion of the MULV Moloney virus into the first intron [103, 104]. Homozygous embryos die at day 13.5 owing to vessel rupture, but early development of organs is normal in the absence of type I collagen. Organ culture of salivary glands or lung buds showed that branching morphogenesis is also normal in the absence of type I collagen [105], possibly owing to a supplementing effect by type III collagen. Interestingly, in organ culture of tooth and bone anlagen α1(I) collagen is expressed despite the insertion of the Moloney virus in the Col1a1 gene [106, 107]. This observation led to the discovery of a bonespecific control of Col1a1 transcription starting at a site which differs from the transcriptional control in fibroblasts (see Section VI).
B. Type V Collagen Type V and XI collagen are closely related in structural and evolutionary terms and have therefore been grouped into a clade of fibril-forming collagens with similar biochemical properties and similar functions in the organism. Type V collagen usually co-distributes with type I collagen in bone, corneal stroma, interstitial matrix of smooth muscle, skeletal muscle, liver, lung, and placenta [80, 108], while type XI
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
collagen is attributed to cartilage collagen. Both collagens precipitate in their pepsin-treated form between 0.8 and 1.2 M NaCl at acidic pH, a feature which was decisive for their first discovery and separation from the dominant type I or type II collagens, respectively [114, 109]. Five of the pro-α-chains of collagen V and XI are further characterized by large amino terminal noncollagenous domains, which are partially retained in the fibrils [50]. The 400 amino acid residue globular domains of α1, α2, α3 (V), and α1 and α2(XI), located between signal peptide and the short triple helix of the N-propeptide, are about twice as large as the corresponding cysteine-rich region in α1(I), α1(II), and α1(III) (Fig. 2). They contain a proline/arginine rich domain (PARP domain) which is similar in α1(V), α2(V), α1(XI), and α2(XI), as well as in the N-terminal domain of collagen XII [110]. The domains are processed only partially after secretion, leaving stubs of 70–100 kDa in the fibril [50, 111] where they are critical for controlling fibril assembly and growth (see below). Unusual is the high content of tyrosine sulfate in the N-propeptide of α1(V) and α2(V) collagen chains [112]. With 40% of the tyrosine residues being O-sulfated, a strong regulating interaction with the more basic triple-helical part is likely to stabilize the fibrillar complex. In contrast to types I, II, and III collagen, the triple helical parts of types V and XI collagen are resistant to digestion with vertebrate collagenase (MMP1), but not to stromelysin [113]. Depending on the tissue, there is some heterogeneity in the composition of type V and XI molecules. Most tissues contain type V collagen molecules consisting of 2α1(V) and one α2(V) chain, but [α1(V)]3 homotrimers have been isolated from tumor cells [114], and some tissues contain α1(V), α2(V), α3(V) heterotrimers (see Table I). Also type V/XI hybrid molecules containing α1(V) and α2(XI) collagen chains were described in articular cartilage [4, 115], in bone [116], and in vitreous humor [117]. Immunohistochemical identification of type V collagen in tissue sections with antibodies requires demasking of the epitopes with acid or enzymes [118, 119], indicating a dense packing of type V collagen within the type I/V collagen heterofibrils [89].
IV. CARTILAGE COLLAGENS The backbone of all cartilaginous tissues of the vertebrate body is a heterofibril containing type II collagen as the predominant collagen type, into which type XI collagen is incorporated. The cartilage collagen fibril is decorated with FACIT collagens, mostly type IX collagen, which is covalently linked to type II collagen [7, 120]. The large N-terminal noncollagenous domains of FACIT
9
collagens, which are globular in the case of type IX collagen and cross-shaped in the case of types XII or XIV collagen, reach out of the fibril into the adjacent matrix space and may serve to anchor the collagen fibril in the proteoglycan matrix (see Fig. 2). Young growth cartilage contains about 85–90% type II collagen with 5–10% type XI and 5–10% collagen IX, while adult articular cartilage may have as little as 1% collagen IX and 3% collagen XI [4]. Substantial differences exist in the matrix composition of different cartilaginous tissues. Elastic cartilage contains collagens II, IX, and XI like hyaline cartilage, but elastin in addition. Fibrous cartilage contains a substantial amount of type I collagen besides type II collagen; type I collagen is also found in prechondrogenic tissue, e.g. in the perichondrium, and in tendon insertions [121]. In mammalian articular cartilage, type I collagen is restricted to the articular surface, where also the newly discovered collagen XXII was located [32], while chicken articular cartilage contains up to 30% type I collagen in the upper zone [122]. In osteoarthritic joints type I collagen was found in osteochondrophytes and fibrous repair tissue. There is still a debate on the question whether articular chondrocytes turn on type collagen I synthesis in osteoarthritis. While some studies report on the presence of type I collagen in osteoarthritic cartilage [123, 124], other immunohistological and in situ hybridization studies confirm the expression of α2(I) mRNA, but not of type I collagen protein in OA cartilage [125]. In contrast, type III collagen is an integral part of mammalian articular cartilage, where it has been located predominantly in the pericellular environment of chondrocytes [126]. Electron microscopic studies have shown that it is also incorporated in the type II collagen fibril [127]. Recently two new fibril-forming collagens (types XXIV and XXVII) [31, 37], and two new FACIT-like collagens (types XX and XXII) [30, 32] were found to be expressed in bone or cartilage, but their function is not yet known. The cartilage collagen fibrils are interwoven with a microfibrillar mesh consisting of type VI collagen, which is prominent in the chondrons of articular cartilage [9, 11]. Type X collagen, a network-forming collagen, is expressed predominantly in fetal and juvenile hypertrophic growth cartilage. It was also located to the upper zone of articular cartilage in certain joints in human, dog, and mouse cartilage [128–130], and in chondrocyte clusters of osteoarthritic cartilage and osteophytes [131, 132]. It supports endochondral ossification [55, 133, 134] and hematopoietic cell differentiation [135]. As a specific marker for hypertrophic chondrocytes it is widely used to analyze chondrocyte differentiation in skeletal development, but also to describe endochondral ossification in fracture callus and osteophyte formation [132, 136].
10
KLAUS VON DER MARK
A. Type II Collagen 1. STRUCTURE AND LOCALIZATION
Type II collagen is found predominantly, but not exclusively, in hyaline cartilage [137, 138], where it accounts for approximately 90% of the total collagen. It is a homotrimeric molecule with the composition [α1(II)]3 with similar size and biochemical features as type I collagen [3, 120], but it contains substantially more hydroxylysinelinked galactosyl-glucosyl disaccharides than type I collagen (ten disaccharides per α1(II) vs two per α1(I) chain). In early embryonic cartilage, type II collagen heterofibrils generally appear as 25–50-nm thin, unbanded fibrils, while in calcified cartilage or after reconstitution in vitro type II collagen may form up to 400-nm thick crossbanded fibrils with a 68-nm banding pattern repeat similar to type I collagen. Type II collagen exists in two splice variants; in the IIA splice form which is dominant in mature cartilage, exon 2 coding for a 69 amino acid residue, cysteine-rich domain in the N-terminal propeptide is spliced out. It is retained in the IIA splice variant, a transient embryonic form which was found in prechondrogenic mesenchyme, in perichondrium and vertebrae [139, 140]. Type II collagen is not only the major collagen of hyaline elastic [141] and fibrous cartilage [142, 143], but also represents the major collagen of vitreous humor [137, 144] and the nucleus pulposus of intervertebral disks. Furthermore, type II collagen is synthesized transiently by many embryonic epithelia such as notochord [73], cornea epithelium [72, 73], retina pigment epithelium, cranio facial mesenchyme, and endocardial and mesocardial tissues [71, 145–147]. 2. THE ROLE OF TYPE II COLLAGEN IN CARTILAGE FORMATION AND STABILITY
Much has been learned on the role of type II collagen in cartilage development and function from mutations in the human COL2A1 gene, causing a rather diverse spectrum of skeletal dysplasias such as achondrogenesis, hypochondroplasia, Stickler syndrome, spondyloepiphyseal dysplasia congenita, and Kniest syndrome [28, 148–150]. As in OI, the more severe forms of chondrodysplasias result from dominant negative mutations affecting triple helix stability such as glycine substitutions, overmodification of the proα1(II) and partial or complete intracellular degradation of type II procollagen molecules containing only one mutated α1(II) chain. For example, in the cartilages of achondrogenesis fetuses with a Gly769Ser mutation [151] or a Gly913Cys mutation causing hypochondrogenesis [152], no type II collagen was found, but instead a matrix containing types I and III collagen. Both mutations are lethal, demonstrating that type I and III collagen cannot replace the function and properties of type II collagen.
Inactivation of the type II collagen gene in mice by homologous recombination had severe consequences on skeletal development, in particular on notochord turnover [69] and vertebral development [153]. Interestingly, it did not affect early embryonic development of the eye, somites or craniofacial tissues, as had been predicted from the expression of type collagen in numerous early embryonic epithelia [71, 146]. Type II collagen-deficient mice die as a result of breathing and weaning inability due to thorax malformation and cleft palate formation, respectively. Long bones are shortened due to impaired endochondral ossification, and vertebral bodies and ribs are abnormal. Similar to achondrogenesis or hypochondrogenesis patients, the cartilaginous tissues contain chondrocyte-like cells, but a matrix consisting of types I and III collagen instead of type II collagen. Although the cartilaginous matrix contains aggrecan, the density of the collagen fibrils was lower and their structure abnormal [153]. Compared to other fibrillar collagens, type II collagen has unique antigenic properties: antibodies raised against chicken type II collagen cross-react with type II collagens from all other species including human, mouse, rat, calf, dog, sheep, and shark [90, 154], indicating highly conserved antigenic epitopes in the type II collagen molecule. Like other matrix components of articular cartilage, which has an immune privilege as a nonvascularized tissue, type II collagen is a major target for autoimmune responses in rheumatoid arthritis [155–157]. In animal models, purified native type II collagen has been shown to induce arthritis in certain strains of mice and rats [158, 159]. The major B-cell epitopes of type II collagen have been identified in the mouse, rat, and human system. They are confomationdependent and located near the integrin-binding site [160, 161]. Interestingly, the major T-cell epitope in rat and human type II collagen which are MHC restricted includes a galactosyl-glucosyl carbohydrate residue [162].
B. Type XI Collagen Type XI collagen is a heterotrimer consisting of α1, α2, and α3 (XI) subunits found predominantly in hyaline cartilage and vitreous humor associated with type II collagen. The α3(XI) subunit is identical in its amino acid sequence with α1(II) as it is translated from the Col2a1 gene, but it differs from α1(II) by a higher degree of hydroxylation and glycosylation [109, 163]. In mature cartilage, half of the α1(XI) molecules are replaced by α1(V) chains [164]. As in type V procollagen, the N-terminal domains of α1(XI) and α2(XI) chains are processed only partially after secretion, leaving stubs of 70–100 kDa in the fibril [163, 165]. Complex alternative splicing occurs within
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
the aminoterminal noncollagenous domains of α1(XI) and α2(XI) [166–168]. A proline- and arginine-rich subdomain (PARP-domain) of the aminopropetides of α1(XI) and α2(XI) seems to be rather stable and persists in the cartilage matrix [110, 169]. As an integral part of the cartilage collagen fibril, type XI collagen serves as a nucleation site of type II/XI collagen fibril formation and regulates the lateral growth of the fibrils [170, 171]. Within the fibril, type XI collagens are covalently cross-linked to each other through their N-telopeptide to helix interaction sites [7] and to the end of type II collagen molecules. The N-propeptide may stick out of the gap domains in the heterofibril (see Figs 2, 3) [172]. The pivotal role of type XI collagen in control of cartilage collagen fibril assembly became apparent also from the analysis of the gene defect of the cho-mouse mutant, which is affected with an autosomal recessive chondrodysplasia: a point mutation in the α1(XI) gene leading to chain determination caused absence of type XI collagen, resulting in a cartilage with irregular thick collagen fibrils, disorganized cartilage growth plate and disturbed chondrocyte differentiation [173]. Similarly, an in-frame deletion in the α2(XI) gene caused an autosomal recessive bone dysplasia or autosomal dominant Stickler syndrome [174].
C. Fibril-Associated Collagens with Interrupted Triple Helices (FACIT Collagens) 1. TYPE IX COLLAGEN
First evidence for the presence of additional collagenous proteins in cartilage was obtained in the form of various pepsin-resistant small collagenous fragments [175, 176]. Combined protein chemical and molecular biological efforts led to the elucidation of the complex structure of type IX collagen [177, 178]. It is a heterotrimeric molecule consisting of α1(IX), α2(IX) and α3(IX) chain, with three triple helical segments (COL1–COL3) that are interrupted and flanked by four globular domains NC1-NC4 [179, 180] (Figs 3, 4). Electron microscopical analysis of carefully dissected cartilage collagen fibrils revealed that the highly cationic NC4 domain, the largest domain with 243 amino acid residues and a pI of 9.7, reaches out from the fibril where it presumably interacts with proteoglycans [181]. A “hinge region” in NC3, caused by supernumerary amino acids in the NC3 domain of the α2(IX) chain, allows flexibility in the molecule. It is covalently linked to the surface of cartilage collagen fibrils with the collagenous domains COL1 and COL2 [182, 183]. The α2(IX) chain contains a chondroitin sulfate side chain [184, 185], linked to
-
Figure 3
11
Structural domains of fibril-forming collagens and the microfibrillar collagen type VI.
12
KLAUS VON DER MARK
Figure 4
Collagenous and noncollagenous domains of the FACIT collagens type IX, XII, and XIV. A = von Willebrand factor A domain.
a serine residue in the NC3 hinge region [186]. This CS chain is considerably longer in vitreous humor than in cartilage [187]. Owing to a second transcription start site between exons 6 and 7 in the α1(IX) collagen gene (see below), a shorter form of the α1(IX) collagen lacking the entire NC4 globule is expressed in the cornea and vitreous humor [188, 189]. The importance of type IX collagen for the integrity of cartilage matrix was underlined by the result of inactivation of the α2(IX) gene in mice. Animals homozygous for the deficiency in α2(XI) showed severe defects in cartilage development and revealed degenerative changes in adult articular cartilage similar to osteoarthritis [190]. 2. TYPES XII, XIV, AND XVI COLLAGEN
Type XII collagen is located predominantly in the perichondrium and articular surface, while type XIV collagen is more uniformly distributed throughout articular and tracheal cartilage [191]. Type XII collagen is a homotrimeric molecule with sequence homologies to type IX collagen in the C-terminal NC1 and COL1 domain, but only two collagenous domains which associate with type I or II collagen fibrils [192–194] (Fig. 4). The large, cross-shaped NC3 domain at the amino end reaches out into the perifibrillar
space [195]. This domain contains vWFA-domains, TSP and FN type III domains. Type XII collagen exists in two splice variants [196, 197]: the smaller IIB form with an α1(XII) chain of MW 220 kDa is found in skin, periosteum, and perichondrium. The larger XII form with an α-chain of MW 310 kDa was found in an epidermal cell line and contains a chondroitin sulfate chain. Collagen type XIV has a similar structure, although the cross-shaped NC3 domain is smaller than in Col XII [193, 198]. Type XIV collagen colocalizes with type I collagen in skin, tendon, lung, liver, placenta, and vessel walls by immunofluorescence [199, 200], and to some extent with type XII collagen in skin [201]. However, it does not bind directly to type I collagen, but to the dermatan sulfate side chain of decorin which associates with type I collagen [202, 203]. Collagen XVI is predominatly synthesized by fibroblasts and myoblasts, but also found in the territorial matrix of chondrocytes [204–206]. By immunohistochemistry and immunogold electron microscopy it was shown that collagen XVI is integrated in a discrete population of thin, D-banded collagen fibrils containing type II and XI that are distinct from type IX collagen-containing fibrils [204]. In the papillary dermis, however, collagen XVI is a component of fibrillin-1-containing microfibrils [204].
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone 3. TYPES XIX, XX, XXI, AND XXII COLLAGENS
These recently discovered collagens are phylogenetically and structurally related to FACIT collagens, but whether they are associated with fibrils in vivo remains to be shown. Type XIX collagen is restricted to muscle in the embryo, but not in the adult [207, 208]. Collagen XX is most abundant in the corneal epithelium, but by RTPCR it has also been detected in sternal cartilage and tendon [30]. The collagen XXI gene codes for a short FACIT collagen containing a vWFA and a Tsp (thrombospondin)-domain like the other FACIT collagens, but little is known on the protein [209]. Collagen XXII exhibits a restricted localization at tissue junctions such as myotendinous junctions, the hair follicle basement membrane or the articular cartilage–synovial fluid interface [32].
D. Microfibrillar Collagens: Type VI Collagen Type VI collagen is the major collagenous component of microfibrils in elastic fibers and in a larger variety of tissues including cartilage, skin, blood vessels (intima), cornea, placenta, uterus, ciliary body, iris, and others [2, 210]. The type VI collagen molecule is a highly glycosylated, cysteine-rich heterotrimer consisting of two α-chains of ca. 1000 amino acid residues (α1(VI) and α2(VI)), and the long α3(VI) chain with about 3000 amino acid residues; the short triple helical core accounts only for about 20% of the molecule. It was discovered first in the form of pepsinresistant “short-chain” collagen with three α-subunits in smooth muscle and placenta [211]. The three α-chains share three noncollagenous domains which are homologous to the von Willebrand factor-A domain [212, 213]; the α3(VI) which exists in multiple splice variants [214–216] contains an additional eight VWF-A-domains at the N-terminus [216]. The C-terminal domain of α3(VI) also contains a Kunitz-type inhibitor motif and is essential for collagen VI assembly and secretion [217]. Interestingly, the α3(VI) subunit is down-regulated by γ-interferon, but the other subunits are not [218]. Type VI collagen interacts with other matrix proteins and proteoglycans, e.g. hyaluronan, heparan sulfate, decorin or NG2-proteoglycan [219], but also with type IV collagen in basement membranes [220]. Examination of type VI collagen in tissues or cell cultures by electron microscopy often reveals beaded filaments with 25-nm beads aligned in 100-nm intervals [221, 222]. Such structures can be assembled in vitro from type VI collagen tetramers which connect and overlap at the globular ends when visualized by rotary shadowing [210, 223]. Complexes of matrilin-1 and biglycan or decorin decorate type VI collagen and link it to type II collagen [224].
13
In the presence of lumican, however, type VI collagen can also assemble into hexagonal networks similar to type X collagen [225]. In tissues like skin or cartilage the type VI collagen forms a highly disulfide cross-linked, branched network, interwoven with fibrillar collagens [9]. Type VI collagen expression is up-regulated already in early stages of chondrocyte differentiation [226]. In mature cartilage it is preferentially located in the pericellular space [8, 9, 11]. It is enhanced in osteoarthritic cartilage [227], but has not been identified yet in calcified bone tissues. In some tissues such as nucleus pulposus and in some tumors type VI collagen filaments may assemble in a parallel fashion to give rise to sheets with the characteristic 100-nm periodicity [222, 228]. The pericellular location of type VI collagen is consistent with its highly adhesive properties for many cell types. In contrast to other fibrillar collagens, several RGD-sequences in the α2-and α3(VI)-chain were found to be recognized by integrin receptors [229].
E. Network Forming Collagens: Type X Collagen Hypertrophic cartilage in the growth plate of fetal and juvenile long bones, ribs, and vertebrae contains a shortchain collagen, type X collagen [55, 230–232] which is unique to this tissue in the normal organism and only found elsewhere under pathological conditions, e.g. in osteoarthritic articular cartilage and in chondrosarcomas [131, 132, 233–235]. There is, however, recent evidence that type X collagen is also present in small amounts in the menisci and in the surface layer of articular cartilage of certain human, mouse, and dog joints [128–130]. Type X collagen is a homotrimeric collagen with a 130-nm triple helical core (460 amino acids per chain), a large C-terminal globular NC1 domain and a short amino terminal NC2 globule [55, 232]. It is homologous in both sequence and tertiary structure to type VIII collagen, which is produced by endothelial cells [236]. Type VIII collagen assembles into sheets with a hexagonal arrangement in the Descemet’s membrane [237], and in vitro reconstitution experiments with chicken type X collagen indicate that this collagen is able to form similar structures [238]. The three C-terminal NC1 domains of type X collagen molecule associate with unusually high affinity to a dense bell-shaped trimer [54, 56, 239] which is homologous to the complement factor C1q and adipoQ and TNFα, not only by amino sequence [239] but also by their crystal structure [54]. Mutations in the NC1 domains lead to cartilage growth abnormalities and waddling gait in patients affected with Schmid-type metaphyseal chondrodysplasia (SMCD) [240–245]. In vitro studies on the chain assembly
14
KLAUS VON DER MARK
of type X collagen demonstrated that most Schmid mutations located at the inner face of NC1 monomer severely affect triple helix assembly [240, 246]. The function of type X collagen is not entirely clear. The severe cartilage and joint defects in SMCD patients clearly indicate a critical role for type X collagen in endochondral ossification; similarly, transgenic expression of a truncated chicken type X collagen gene in mice causes growth plate abnormalities and a hunch back [134, 247]. There is also indirect evidence that type X collagen might be involved in matrix-vesicle-initiated calcification of hypertrophic cartilage [133]. Surprisingly, however, the phenotypic changes in the growth plate of two strains of type X collagen knockout mice were only moderate [133, 248]. A subset of the Col10a1 knockout mice suffered from severe impairment of hematopoiesis, causing prenatal or early postnatal death [135, 249].
F. Multiplexins and Other Collagens Collagen types XV, XVI, XVII, XVIII and XIX are composed of 8–12 collagenous domains interrupted by short noncollagenous domains, allowing a flexibility in the molecular shape which is not possible in fibril-forming collagens [14, 250–254]. The cellular location and tissue distribution of most of these collagens has been described; yet, little is known of the possible function of these collagens. Collagen type XIII is a type II transmembrane collagen with a short, noncollagenous carboxyterminal domain in the cytoplasm and a large extracellular part which contains three collagenous domains [61]. Collagen type XIII may be involved in cell–matrix interactions by binding to fibronectin [255] and α1β1 integrin [256], but it has also antiadhesive properties. Bone is a major physiological site for collagen XIII expression [257], but its role in this tissue is still unclear. Collagen type XVIII, a heparan-sulfate-containing collagen which anchors cells to the basement membrane has also been localized in hyaline and fibrous cartilage [258–260]. The C-terminal NC1 domain called endostatin of Col XVIII has received much attention recently as it persists as a separate entity in tissues and blood and was shown to inhibit tumor angiogenesis by interfering with endothelial cell sprouting [60, 261].
V. COLLAGEN BIOSYNTHESIS Most of our knowledge on the mechanisms of collagen biosynthesis is based on studies on fibril-forming collagens, in particular type I collagen. For this reason, in this
chapter I will focus on biosynthesis events and mechanisms elucidated for the fibrillar collagens, but it is likely that most of it will apply also for other collagens.
A. Steps of Collagen Biosynthesis 1. TRANSCRIPTION AND ALTERNATIVE SPLICING
Most collagen genes are transcribed into different mRNA species owing to alternative splicing, to a different extent of polyadenylation at the 3′ end as in the case of type I, II, and III collagen, or to several transcription start sites in some collagen genes, as in the case of type IX collagen α2(I) (for review see [21, 262]). In the cornea and in the vitreous, a shorter form of the α1(IX) collagen gene (Col9a1) is generated owing to an alternative transcription start site between exons 6 and 7 [189]. Alternative splicing of collagen mRNAs was first described for type II collagen which exists in an embryonic, longer form (IIB) and a shorter, adult form (IIA) from which exon 2 is spliced out [140, 263]. Alternative spliced forms have been reported for many collagens such as collagen VI (see above), XI [168] and many others, for example type XIII with more than 12 splice variants [264]. Further heterogeneity among collagen mRNA species is introduced by heterogeneity in polyadenylation at the 3′ end. Thus, two α1(I) mRNA species of 4.8 and 5.8 kb length can be observed in fibroblast cultures [265, 266], and also for α2(I) and α1(II), and α1(III) two different mRNA species are detected in cell cultures depending on the cell type and culture conditions [267, 268]. 2. Translation
Collagens are synthesized as procollagen molecules in the lumen of the rough endoplasmic reticulum (RER), transported to the Golgi and secreted (with the exception of membrane-spanning collagens) into the extracellular space via secretory vesicles. The nascent pre-proα-chains protrude after translation from the ribosome-bound collagen mRNA into the lumen of the RER with the help of signal-recognition particles and receptors. Immediately after intrusion, the signal peptide is cleaved off, and the nascent procollagen chains are modified by hydroxylation and glycosylation before they align into triple helical procollagen molecules, which are secreted and processed before assembling into complex extracellular aggregates (for reviews see [21, 49]). 3. Hydroxylation and Glycosylation
In the triple helical region of all procollagens about half of the proline residues in the Y-position of the Gly-XY-triplets are converted to 4-hydroxyproline by the enzyme
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
prolyl-4-hydroxylase, an event which occurs only on nascent collagen α-chains before the triple helix is formed [269–271] (Fig. 5). A few proline residues (all in X position) are converted to 3-hydroxyproline by the enzyme prolyl-3-hydroxylase [272, 273]. Two isoforms of 4-prolylhydroxylase have been described: the type I enzyme with the composition [α(I)]2β2 has been cloned from human, rat, and chicken sources [274] and accounts for 60–90% of the total prolylhydroxylase activity in most cell types, while the type II form [α(II)]2β2 is more prevalent in chondrocytes [275]. The four subunits of the prolylhydroxylase encircle and slide along nascent pro α-chains, but are unable to hydroxylate prolyl-residues after the triple helix has formed. Therefore, the time when the α-chains are kept in a single-chain conformation is critical for the extent of hydroxylation of prolyl residues; the same is true for hydroxylation of lysine residues. Prolyl-3- and -4-hydroxylase, as well as lysylhydroxylase, require Fe2+, oxygen and 2-oxoglutarate as cofactors [269]. Ascorbic acid is required nonstochiometrically for uncoupled decarboxylation of 2-oxoglutarate. Thus, lack of oxygen or ascorbic acid, as well as chelation of Fe2+ by chelators such as α,α′-dipyridyl or o-phenanthroline, effectively prevent hydroxylation of
15
proline and lysine during collagen synthesis [276]. Since non- or under-hydroxylated collagens are degraded to a large extent intracellularly [277], the inhibition of prolylhydroxylase activity is considered a potent tool to control excess of collagen synthesis in fibrotic diseases [278]. A number of competitive inhibitors of prolylhydroxylase have been developed and are being tested: pyridine 2,4dicarboxylate, 3,4-dihydroxylbenzoate or L-mimosine [278, 279]. The hydroxylation of lysine to δ-hydroxylysine by the enzyme lysylhydroxylase [2, 280–282] is not only important for the glycosylation of collagen, but also for the formation of the covalent pyrrolic and pyridinium crosslinks formed after oxidation of lysine and hydroxylysine to aldehydes [283] (see Chapter 2). The extent of lysyl hydroxylation varies with the collagen type, species, the cell and tissue type, and may change during development and aging or under pathological conditions. For example, two lysine residues in the α1(I) chain, both in the telopeptide region, but about ten lysine residues in the α1(II), are hydroxylated to 5-hydroxylysine. Subsequently, glucosyl and galactosyl residues are transferred to the hydroxyl groups of hydroxylysine by glycosyl transferases [270, 284]
l
Figure 5 Biosynthesis of procollagen in the lumen of the rough endoplasmic reticulum (RER). After hydroxylation of the nascent α-chains, triple helix assembly and glycosylation the procollagen molecules are transported to the Golgi apparatus, secreted in secretory vesicles to the cell surface and processed. PDI, protein disulfide isomerase; HSP47 and BiP, molecular chaperones; PPI, prolyl-peptidyl-cis-trans-isomerase.
16
KLAUS VON DER MARK
to form glucosyl-galactosyl disaccharides, while mannoserich oligosaccharides are coupled to a N-glycosylation receptor sequence in the C-propeptides of fibril-forming collagens [285]. 4. ALIGNMENT OF THE TRIPLE HELIX
Assembly of the procollagen chains to triple helical molecules occurs after hydroxylation of the nascent chains and completion of the C-propeptides. The correct conformation of the C-propeptides and of some N-propeptides is stabilized by the formation of intrachain disulfide bonds [286]. This is catalyzed by the enzyme protein disulfide isomerase (PDI) [287, 288], a reducing enzyme with a catalytic site similar to thioredoxin and present in many plant and animal cells. Interestingly, PDI is identical to the β-subunit of prolyl-4-hydroxylase [269, 271], and thus has a dual role: as a β2 dimer it catalyzes and controls the formation of intrachain and interchain disulfide bonds in the C- and N-propeptides; in combination with two α-subunits of the prolyl hydroxylase it catalyzes the hydroxylation of proline. Procollagen chain assembly involves a two-stage recognition event in the fibril-forming collagens [22]. (1) The first association and chain recognition in type I procollagen is driven by the C-propeptides [286]; essential for assembly are hydrophobic sequence patches with four hydrophobic amino acids that are conserved in their positions in all fibril-forming collagens. In type III procollagen, assembly of the procollagen chains and triple helix formation does not require the formation of interchain disulfide bonds which form later after the assembly of the trimer [289]. (2) Rate-limiting for the folding of the triple helix is a cis–trans-isomerization of prolyl peptide bonds in the α-chains, which is catalyzed by the enzyme peptidyl-prolyl cis–trans-isomerase (PPI) [288, 290]. Critical support for a key role of PPI in collagen chain-folding was the observation that cyclosporin, an inhibitor of PPI and other cyclins, blocks folding of procollagen I and III in cell cultures [290]. After alignment of the C-propeptides, in type I procollagen triple helix formation starts from the C-terminus and proceeds towards the N-terminus; it is driven by a patch of 3–4 Gly-Pro-Hypro triplets located at the C-terminal of the triple helix by providing a core stabilized by hydrogen bonds [22]. Although the role of C-propeptide in the chain assembly of fibril-forming collagens is firmly established, their role in chain selectivity in cells that produce several collagen types is not clear yet. Interestingly, only procollagen chains containing eight cysteines in the C-propeptide are able to form homotrimers (e.g. pro α1(I), α1(II), α1(III)), while procollagen chains lacking two or three of the eight cysteine residues such as pro α2(I) can only form heterotrimers [22].
5. Chaperones Involved in Folding and Processing of Procollagen
Several chaperones such as BiP and HSP47 are involved in the post-translational modification, folding, and processing of the procollagen molecule [291]. HSP47 is a heat shock protein and is expressed at ten-fold enhanced levels in chicken fibroblasts at 42–45⬚C [292]. It binds cotranslationally to the nascent pro-α-chain in the RER and prevents unspecific chain folding and aggregation [292, 293]. Interestingly, it also binds to native, triple helical collagen molecules in the RER and controls bundle formation, transport to the Golgi apparatus, and procollagen processing [291]. Although it has a basic pI of 9.0, the glycoprotein binds to basic collagen chains and to native collagen with a kDa of 10−6 to 10−7 M, but with high on and off rates [294]. In inflammatory situations, such as after treatment of cells with TNFα, HSP47 appears on the cell surface where it may invoke autoimmune reactions, causing rheumatoid arthritis [295]. 6. Procollagen Processing
Following secretion of procollagens in secretory vesicles, the C-propeptides of the fibril-forming collagen, and the N-propeptides of type I and II, partially also of type III collagen are cleaved by specific proteases. Both procollagen-Nproteinase and procollagen-C proteinases belong to a family of Zn2+-dependent metalloproteinases M12 [23, 296]. The N-proteinase exists in two splice variants in a longer form of 140 kDa (pNP1) and a shorter form of 70 kDa (pNP-2) [297, 298]. Both cleave the N-propeptides of types I and II procollagen [299], but only in the native state, leaving the N-telopeptide in the hairpin-loop configuration. The N-proteinase is a structural homolog to atrolysin and adamalysin II and contains an RGD-sequence and properdin-like repeats in the C-terminal part [23]. The procollagen-C-proteinase cleaves the C-propeptides of types I, II, and III collagen, as well as laminin 5 and lysyl oxidase precursor [23]. Curiously, the enzyme is partially identical to Bone Morphogenetic Protein 1 (BMP-1) or the mouse tolloid protease, and related to Drosophila tolloid, Astacin, and BP10 [300, 301]. All these enzymes share the Zn2+-binding proteinase domain, but the procollagen C-proteinases cCP-2 (mTld) (MW 110 kDa) or pCP-1 (BMP1) (70 kDa) differ in their C-terminal parts. The C-terminal propeptides that are cleaved off are retained for some while in the tissue. This has been shown in particular for type II collagen: the C-propeptide has been isolated from cartilage as an intact, disulfide-bonded entity which has been given the name chondrocalcin [53] as it is found in particular in calcified cartilage. Besides the described C-proteinases of the BMP-1 type, other proteinases seem to be able to cleave procollagen
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
propeptides, e.g. cathepsin D [302]. This possibility must be taken into consideration in view of the recent finding that inactivation of both alleles of BMP-1 in mice did not dramatically alter procollagen processing, but caused failure of body wall closure [301]. 7. Cross-linking
A key enzyme required for cross-linking of collagen α-chains and molecules is the enzyme lysyloxidase [282, 303], which catalyzes the oxidation of the ε-amino group of certain lysine and hydroxlysine residues to aldehyde groups. Lysine- or hydroxylysine-derived aldehyde groups initially form covalent Schiff-base-type bonds with ε-amino groups of other lysine and hydroxylysine residues in adjacent α-chains within the same or neighboring molecules. More detailed information about collagen cross-link formation will be provided in Chapter 2.
B. Translational and Post-translational Regulation of Collagen Biosynthesis Regulation of collagen biosynthesis occurs at several levels, involving a complex pattern of transcriptional, translational, and post-translational control mechanisms [304]. The following section will review studies showing the role of translational and post-translational mechanisms controlling the biosynthesis of fibrillar collagens. Little is known on translational or post-translational regulation of the other collagens. 1. Evidence for Translational Regulation
Evidence for translational control of collagen mRNA was provided by the observation that chondrocytes contain untranslated α1(I) mRNA [305]. Also splice variant of α2(I) mRNA was found in chondrocytes; curiously, it is translated into a noncollagenous protein [306, 307] using a different reading frame, but the function of this protein is as yet unclear. Another mechanism regulating collagen mRNA translation may be associated with the 3′-untranslated sequences. Polymorphic mRNA species exist for α1(I), α2(I), α1(II), α1(III), and α1(IV); for example Northern blotting of α1(I) reveals two bands of 4.7 and 5.7 kb in fibroblast monolayer [19, 265], which are expressed in different ratios depending on the cell type. This ratio increases significantly under the influence of TGFβ [308] which also increases the half life of the mRNA that is normally 9–10 h. In fibroblasts cultured in three-dimensional collagen gels, however, only the smaller form of α1(I) mRNA is expressed [267, 268]. As in other mRNA species, the 3′ untranslated segment of collagen mRNA is responsible for the stability and the translation rate of the collagen mRNA [309].
17
2. Regulation by MRNA Folding
There is experimental evidence that translation initiation of α1(I), α2(I), α1(II), and α1(III) mRNA may be controlled by highly conserved nucleotide sequences close to the translation initiation codon. RNA patches of about 50 nucleotides contain inverted repeat sequences which form stem loops or dimers and thus prevent translation initiation [310]. Deletion of these sequences in the α2(I) mRNA in vitro did not, however, always affect mRNA translation [311]. 3. Feedback Inhibition by N- and C-propeptides
Studies on fibroblasts with Ehlers–Danlos syndrome VII as well as cell culture studies indicated that the N-propeptides of type I and III collagen induce feedback inhibition of type I collagen synthesis [312]. The N-propeptide of α1(I) specifically inhibited cell-free translation of α1(I) mRNA [313] but, curiously, with fragmentation of the N-propeptide into smaller fragments, the inhibitory effect became unspecific [314]. Strong support for a specific role of the N-propeptide was the observation that it binds specifically to fibroblast membranes and is taken up by endocytosis [315]. Furthermore, expression of a metallothionine-inducible N-propeptide minigene in fibroblasts inhibited type I collagen synthesis [316]. Also, the type I procollagen C-propeptide has been implicated in the regulation of procollagen synthesis after internalization [317, 318]. It is secreted in significant amounts by osteoblasts and inhibits collagen synthesis to 80% of the control level [319]. In contrast to the N-propeptide, however, it is transported to the nucleus and seems to control collagen gene expression at the transcriptional level, while smaller fragments of the C-propeptide may regulate collagen synthesis both negatively or positively at the posttranscriptional level [318]. Similarly, the C-propeptide of type II collagen appears to inhibit translation of α1(II) mRNA [320].
VI. COLLAGEN GENES AND TRANSCRIPTIONAL REGULATION One of the current challenges in collagen research which needs to be solved for better understanding of the pathogenesis of collagen-related bone and cartilage diseases is the elucidation of mechanisms regulating tissue-specific expression of distinct collagen types. Tissue-specific expression of the various collagen types is regulated mostly, but not exclusively, at the transcriptional level and involves multiple control mechanisms at different levels. At the cell surface hormones, growth factors, cytokines, and their antagonists compete for their receptors; ligand binding
18
KLAUS VON DER MARK
triggers intracellular signaling pathways regulating expression and activity of transcription factors, and in the cytoplasm and nucleus complex interactions of transcription factors with promoter, enhancer, or silencer elements regulate expression of collagen genes [21, 304, 321, 322]. Since the deciphering of the first collagen cDNA sequences, coding for proα2(I) and proα1(I) mRNA [323–325], the gene structure of 42 vertebrate collagen genes has been resolved. Cis-acting regulatory domains and the corresponding trans-acting factors are, however, known only for few collagens genes, predominantly in the fibrillar collagens, collagens IV, VI, and X, which will be briefly summarized in the following section.
A. Genes Coding for Fibrillar Collagens The genes coding for fibril-forming collagens are rather conserved and characterized by numerous short exons interrupted by fairly long introns. The triple helices of the genes of α1(I) (COLlA1 for the human gene, Col1a1 for the murine), of α2(I) (Col1a2), α1(II) (Col2a1), α1(III) (Col3a1), and all three chains of type V and XI collagen are encoded by 44 short exons, flanked at the 5′-end by two longer exons coding for the N-propeptide, and at the 3′-end by four exons coding for the conserved C-propeptide and 3′-untranslated regions [19]. In contrast, most of the nonfibril-forming collagens and invertebrate collagens are composed of fewer and larger exons. The exons coding for triple helical sequences are all multiples of 9 bp corresponding to a Gly-Xaa-Yaa-triplet; most common are 45, 54, and 109 bp [18, 19, 326]. This feature is unique for fibril-forming collagens and suggests that the fibrillar collagen genes have evolved from an ancestral multi-exon gene coding for collagen triplets. In contrast to the highly conserved size and number of exons in fibrillar collagen genes, there is considerable variation in the intron lengths between species and collagen types. For example, the gene COL1A1, coding for human pro α1(I), has a total length of 18 kb [327, 328] while the gene size COL1A2 for α2(I) is 40 kb [328] . All genes coding for human fibrillar collagens, even those belonging to the same type, are located on different chromosomes (for overview see [19, 329]). 1. Regulatory Domains and Transcriptional Regulation
As in most regulated genes, transcription of collagen genes is regulated by multiple cis-acting elements located in the proximal promoter and by the tissue-specific action of enhancer and silencer elements located further upstream of the promoter or in the introns. Identification of these
elements in collagen genes has led in the past to the identification of DNA binding factors responsible for tissuespecific regulation of collagen gene expression, including the identification of transcription factors such as cbfa1/ runx2 or SOX9. At the same time, increasing knowledge has accumulated on growth factors, hormones, and cytokines and their receptors in the regulation of collagen gene expression. One of the current challenges is to bring together the hormone- and growth factor-regulated signaling pathways and the transcriptional mechanisms regulating tissuespecific expression of collagen genes. 2. Type I Collagen Genes
a. Col1a1 Genes Studies by Sokolov et al. [330, 331], and Rossert et al. [332, 333] indicate that for human COL1A1 expression in skin a minimal promoter is required which is located within −476 bp and −900 bp upstream of the transcription start site of the COL1A1 gene. Although in transgenic mice the −476 bp promoter seems sufficient for tissuespecific expression even without intron sequences [331], further positive and negative regulatory sequences were identified in the promoter and in the first intron of the COLA1 genes [322, 334]. Slack et al. [335] reported that a reporter gene construct containing a 440-bp segment of the human COL1A1 promoter lacked tissue specificity without the first intron. Several cis-acting intronic sequences with a highly conserved 29-bp domain involved in transcriptional regulation were located between +292 and +670 bp [336]. Between +820 and +1093 bp a silencer element is active when combined with a 332-bp promoter fragment. The effect of the intronic sequences varies with their size, location, and orientation within the reporter gene constructs, and the cell type used for transfection. First evidence for a key role of the first intron in fibroblast-specific expression of Col1a1 was obtained from the MoV13 mouse mutant which was generated by infection with the Moloney virus [103, 104]. Accidental insertion of the virus into the first intron of the Col1A1 gene caused a complete block of Col1a1 transcription in fibroblasts. Mice homozygous for the insertion of the virus did not contain any type I collagen and died at day 13.5 due to heart rupture [103, 104]. Interestingly, however, the α1(I) was expressed in tooth and bone anlagen when cultured in chorioallantoic membranes in vitro, thus bypassing the termination of bone differentiation in vivo at day 13.5. This indicated a different transcriptional regulation in osteoblasts and odontoblasts as compared to fibroblasts [87, 106]. In fact, Rossert et al. [332, 333] located a bone-specific enhancer element between −1656 and −1540 bp upstream of the transcription
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
start site which promotes expression in osteoblasts. The transcription factor which binds to this element remains to be elucidated; it is apparently not cbfa1 which binds to the OSe2 consensus sequences located in the proximal promoter of Col1a1 [337] (see below). Another bonespecific site may be the Vit.D3-responsive element within the 3.6 kb promoter sequences of rat Col1A1, which is down-regulated by Vit.D3 in ROS osteoblasts, but not in fibroblasts [338]. Similarly, glucocorticoids which have been shown to down-regulate collagen synthesis also decrease expression of Col1a1 promoter-reporter gene constructs [339, 340]. b. Col1a2 Genes Similar to the Col1a1 gene, regulatory sequences have been described in the first intron of Col1a2 between +418 and +524 bp upstream of the transcription start site [341]. Curiously, the intron of human COL1A2 suppressed reporter gene transcription in chicken fibroblast, while the murine intron enhances transcription in mouse NIH 3T3 fibroblasts [341]. A −2000 bp Col1a2 promoter-reporter gene is active in transiently transfected osteoblasts at −300 bp. TGFβ is one of the major factors stimulating type I collagen synthesis in fibroblasts (see below) [342]. A TGFβ-activating element (TAE) with a nuclear factor I-like sequence was identified about 1600 bp upstream of the Col1a1 transcription start site [343]. A NF-1 binding site mediates the inhibitory effect of TGFβ on the α2(I) promoter [344]. There are also data showing that an AP-1 site in the proximal promoter of COL1A2 is responsible for the regulation of α2(I) expression by TGFβ [345], while other data indicate that the transcription factor Sp1 but not Ap1 is required for the early response of COL1A2 expression to TGFβ [346]. A proximal element within the human α2(I) collagen promoter located between −161 to −125 bp relative to the transcription start site is responsible for down-regulation of Col1a2 promoter activity by interferon γ (IFNγ) [347]. This IFNγ-response element (TaRE) is located between −265 and −241 bp. A potent far upstream enhancer element between 15 and 17 kb 5′ of the Col1a2 gene strongly increased the expression of a reporter gene in transgenic mice in dermis, fascia, and other fibrous tissue [348]. The Col1a2 gene contains an alternative, cartilagespecific transcription start site with a 179-bp promoter in the third intron, which is transcribed in cartilage into a mRNA with an open reading frame, but does not translate into a collagenous protein [306, 307]. The role of this alternative transcript which is rapidly turned over in cartilage is still unclear. Of particular interest would be the mechanism controlling the shift from the use of the fibroblast-specific promoter of pro α2(I) in chondroprogenitor cells to
19
the use of the cartilage-specific promoter in differentiated chondrocytes. 2. Type II Collagen Genes
Unlike type I or III collagen, the expression of type II collagen is restricted to only a few tissues, predominantly to cartilage, vitreous humor, embryonic cardiac jelly, and some embryonic epithelia [71, 90]; therefore, the tissuespecific regulation of Col2a1 expression has received particular interest in the past. Horton et al. [349] observed that CAT reporter gene constructs containing only Col2a1 promoter fragments upstream of the transcription start site showed low expression levels, which was significantly increased when an 800-bp segment of the first intron was included. Cartilage-specific enhancer elements were located within a 182-bp sequence in the first intron of the murine Col2a1 gene [350] or a 119-bp fragment of the rat Col2a1 gene [351]. The enhancer element consists of three HighMobility Group (HMG) binding sequences located within a 48-bp element between +1878 to +2345 bp in the first intron [350, 352]. Four tandem repeats of this 48-bp element or 12 tandem copies of an 18-bp sub-element of this enhancer, in combination with a 309-bp Col2a1 promoter, allowed cell-specific expression of a β-galactosidase reporter gene in mouse chondrocytes [353]. The critical role of this enhancer is underlined by the finding that cartilagespecific expression of Col2a1 enhancer-reporter genes in transgenic mice even occurs when the 309-bp Col2a1 promoter is replaced by a minimal β-globin promoter [352]. The localization of this enhancer element allowed the identification of SOX-9 as a major transcription factor regulating type II collagen expression in chondrocytes [354] (see also below). Furthermore, a zinc-finger protein CII2FP was described, which binds to the enhancer within the first intron of Col2a1 and to Sp1 [355], but the role of this protein in the control of Col2a1 expression remains to be elucidated. The complexity of the transcriptional regulation of the type II collagen gene is underlined by the finding of two silencer elements in the rat Col2a1 promoter which seem to inhibit Col2a1 expression in nonchondrocytic cells [355]. 3. TYPE III COLLAGEN GENES
In the mouse Col3a1 gene two positive and one negative regulatory domains were located in the proximal promoter region. A 12-bp stretch binding at −122 to −106 bp to an AP-1-like protein enhances transcription in DNAtransfection assays, and a 22-bp segment B (−61 to −83 bp upstream of the transcription start site) binds a nuclear factor (BBF) extracted from Hela cells [356]. A negative regulatory element located between −350 and −300 bp repressed the activity of the Col 3a1 promoter [357].
20
KLAUS VON DER MARK 4. Type V and XI Collagen Genes
In contrast to the genes of type I, II, and III collagen, the promoter of the human of α1(V) collagen gene COL5A1 [358] lacks obvious TATA and CAAT boxes and has a number of features characteristic of the promoters of housekeeping genes [359]. It contains multiple transcription start sites, a high GC content and a number of consensus sites for the potential binding of the transcription factor SP1. The promoter seems to consist of an array of cis-acting elements, with a minimal promoter 212 bp upstream of the major transcription start site. The COL5A2 gene seems more closely related to the COL3A1 gene, but little is known of its regulating sequences [360]. Expression of the human COL5A3 gene is regulated by the transcription factor CBF/NF-Y binding to two repressors in the TATAless core promoter [361]. Similar to the COL5A1 gene, the genes of the human type XI collagens COL11A1 and COL11A2 lack TATA boxes, indicating that they are regulated similarly to housekeeping genes [362, 363]. The human Col11A1 gene has several minor transcriptional start sites clustered around one major start site at 318 bp upstream of the ATG codon [362, 363]. A transcriptional complex containing potential binding sites for ubiquitous nuclear proteins such as AP2 and Sp 1 is closely related to a similar complex in the proximal promoter of the human COL11A2 gene. The COL11A2 gene spans 30.5 kb with a minimum of 62 exons. Similarly to COL5A2, its promoter is GC-rich with two potential Sp1binding sites [167, 364]. A chondrocyte-specific enhancer element in the murine Col11a2 gene which binds Sox9 and resembles the Col2a1 enhancer confers cartilage-specific expression [365].
B. Type VI Collagen Genes The genes of the three type VI collagen subunits are composed of more than 60 small exons that are multiples of nine bp similar to those of the fibril-forming collagens. Of these, 19 exons in the range between 27–90 bp code for the triple helical part of chicken and human α1(VI) and α2(VI) [216, 366, 367]. Multiple splice variants have been described for α2(VI) and α3(VI) in chicken and human: in the chicken α2(VI) the major splice variant contains a vWFA domain at the carboxyterminal globule which is replaced by a FN-type III repeat in the minor splice variant expressed only in aortic tissue [368]. The human α2(VI) exists in three splice variants owing to alternative splicing at the C-terminus [216]. The chicken α3(VI) shows extensive size heterogeneity, with five bands in SDS-PAGE due to alternative splicing of vWFA
modules in the globular domain [369]. The α3(VI) splice variants show different tissue distribution depending on age and developmental stage [215]. In the chicken Col6a1 gene two major and several minor transcription sites were found [370]. The proximal promoter which lacks TATA or CAAT boxes contains three regulatory C-GGAGGG (GA box) boxes that bind several transcription factors including Sp1, AP1 [371, 372], and AP2 [373]. These elements are responsible for up-regulation of α1(VI) expression during C2C12 myoblast differentiation [371]. Expression of the murine Col6a1 promoterreporter gene constructs in transgenic mice indicated that the −215 bp minimal promoter is not sufficient for expression in tissue [370, 374]. Additional cis-acting elements between −0.4 to 1.4 kb are required for tissue expression. The cis-acting elements at –4.6 to −5.4 kb are responsible for α1(VI) expression in joints, while the region at −6.2 to −7.5 kb upstream of the transcription start site is required for expression in muscle, meninges, and other tissues except lung and uterus [370, 374]. Two promoters control the transcription of the human COL6A1 gene [373]. The chicken Col6a1 collagen promoter encompasses four regulatory sites; S1, S2, and S3 are recognized by transcription factors of the Sp1-family, while site X binds a 43 kDa transcription factor [375]. This type of cis-acting regulatory element found in the type VI collagen genes differs completely from those found in the genes of fibril-forming collagens. This is consistent with the distinct regulation of type VI collagen expression in tissues and cell culture in comparison to collagens I, II, or III. All three type VI collagen genes are up-regulated in fibroblasts when grown at increasing densities in culture, while Col I and III levels do not change [376]. Conversely, the phorbolester and cocarcinogen PMA reduces Col I and III levels, but not Col VI expression. Similar to Col I and III, however, Col VI expression is increased in systemic sclerosis and in keloids [377]. A strong up-regulation of collagen VI expression has been observed during chondrocyte differentiation [226]. Only in one study expression of α1(VI) in osteoblast-like cells been documented; it is up-regulated by interleukin 4 [378]. Interestingly, the human COL6A1 gene is not always coregulated with COL6A2 or COL6A3. Only α3(VI) expression, but not α1(VI) nor α2(VI), are up-regulated by TGFβ [379] or down-regulated by IFNγ [218] .
C. Type X Collagen Genes Unlike genes coding for fibril-forming collagens or type VI collagen chains, the genes for type X collagen contain only three exons that are fairly conserved in all
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
four species investigated: chicken [380], mouse [381–383], bovine [384], and human [385, 386]. Exon 1 of Col10a1 contains an 80–114 bp 5′ UTR sequences, exon 159–169 bp, while exon 3 with 2136–2900 bp codes for the majority of the triple helical part, the entire C-terminal NC1-1 domain and up to 939 (human) 3′ UTR sequences [387]. Nuclear run on experiments indicate that collagen X expression is regulated strictly at the transcriptional level [388]. As expected from the restricted temporal and spatial expression of type X collagen in the fetal cartilage growth plate [389, 390], strong positive and negative regulatory cisacting elements have been described in the promoter of collagen X genes, with major differences, however, between mammalian and chicken genes [387]. In the chicken, multiple negative elements in the proximal promoter (−580 bp) restrict expression of the Col10a1 gene to hypertrophic chondrocytes [391–393]. Also, a positive regulatory BMP-responsive 529-bp element was located 2.4–2.9 kb upstream of the Col10a1 transcription start site, which was found necessary and sufficient for BMP-stimulated Col10a1 expression in prehypertrophic chicken chondrocytes [394]. In the human COL10A1 promoter a strong enhancer element located between −1821 and −2400 bp upstream of the transcription site [395–397] stimulated transcription of reporter gene constructs in hypertrophic chondrocytes. In the mouse Col10a1 promoter an additional 2000-bp segment containing LINE elements, B1 and B2 and LTR elements, are inserted [398] moving the enhancer further upstream. This enhancer is highly homologous to the human and bovine enhancer and promotes specific expression of a COL10A1 reporter gene in hypertrophic cartilage both in in vitro reporter gene assays and in transgenic mice in vivo [398]. The enhancer contains an AP-1 site that is 100% conserved in mouse, human, and bovine Col10a1 enhancers. In vitro it is recognized by the AP-1 factors Fos B and Fra 1 [398]. Expression of type X collagen is down-regulated by PTH and PTHrP [397, 399] via the adenylate cyclase and PKA pathway which activates c-fos expression [397]. The proximal promoter regions of the Col10a1 genes contain a TATA box, one ETS binding site and an AP-2 binding site that are conserved between mouse, bovine and human species [386, 387]. Data by Harada et al. [400] indicate that the murine Col10a1 promoter is up-regulated by osteogenic protein 1, which is mediated by a MEF2-like sequence. The murine Col10a contains a second TATA box, and correspondingly, a second transcription start site, which is utilized only in the newborn mouse [382], while CCAAT boxes at −120 bp were found so far only in the mammalian Col10a1 genes. Furthermore, the proximal promoter of the murine Col10a1 gene contains OSE2 binding sites [401].
21
VII. FACTORS REGULATING COLLAGEN BIOSYNTHESIS A. Bone Collagen Regulation With collagens providing 90% of the organic mass of bone, many mechanisms and factors regulating growth, turnover, or repair of bone are likely to be involved in the regulation of type I and V collagen synthesis by osteoblasts or their turnover by osteoclasts and proteases. As shown above, collagen biosynthesis in bone is regulated at transcriptional, translational, and post-translational level. Here I will review briefly current knowledge on the role of growth factors, hormones, cytokines, oncogenes, and transcription factors in regulating collagen synthesis, but additional chapters in this volume are devoted to collagen turnover by proteases and the role of growth factors, hormones, and cartilage on other aspects of bone metabolism. In view of the ubiquitous distribution of type I and V collagen in the vertebrate organism, most factors regulating these collagens are not tissue specific. Recently, however, two bone specific transcription factors Cbfa1/ runx2 [402–404] and osterix [405] have been discovered which bind to distinct cis-acting elements in the promoters of type I collagen and other bone-specific genes. 1. Growth Factors
TGFβ factors are one of the longest-known and bestdescribed factors regulating expression of collagen types, although their specific effects on collagen expression seem to vary considerably in different cell culture systems. Generally, TGFβ1 stimulates type I and III collagen synthesis in fibroblasts, osteoblasts, hepatocytes, and many other cell types [342]. Incubation of cells with TGFβ1 in vitro causes enhanced α1(I) mRNA levels by stimulating gene transcription and by supporting the stability and translational activity of collagen mRNA [308, 406]. TGFβ3 stimulates collagen synthesis both in TGFβ1-dependent and -independent pathways [407] (see Table II). In hepatocytes, TGFβ-stimulated transcription of COL1A2 involves binding of the transcription factors SP1 and Smad 2, one of the TGFβ receptor regulated Smad factors, to the proximal promoter of the COL1A2 gene [408, 409]. In human skin fibroblasts, Smad-3 transmits the TGFβ signals from the receptor to the COL1A2 promoter, while Smad 7, an inhibitory Smad factor, may be involved in autocrine negative feedback in the regulation of COL1A2 promoter activity by TGFβ [410]. Bone morphogenic proteins (BMPs), in particular BMP-2, -3, -4, and -7, which belong to the TGFβ superfamily, induce bone formation in embryonic and adult stem cells, mostly via enchondral ossification, but also directly
22
KLAUS VON DER MARK Table II.
Cytokines, growth factors, oncogenes and transcription factors modulating collagen biosynthesis
Factor
Collagens affected
TGFβ1
I, III
TGFβ3
II II VI α1(I)
BMP -2, -3, -4, -7
I (?)
PDGF EGF IGF-I
II; X V I II
Il-1
I, III
IFNγ TNFα
II I II I
Vit.D3 Glucocorticoid
I α1(I)
PTH Retinoic acid
I X I, III
Ascorbate
X All collagens
c-fos
I
Cbfa1/Runx2
I X
SOX9
II, IX, XI
Cellular targets
Type of response
Growth factors Fibroblasts, osteoblasts Increases α1(I) + α1(III) mRNA level & stability, hepatoblasts, dediff. enhances translational activity of α2(I) mRNA chondrocytes Chondrocytes Suppression of Col2a1 expression Chondrocytes Up-regulation of Col2a1 expression Fibroblasts Up-regulates α3(VI), down-regulates α1 + α2(VI) Fibroblasts Stimulates α1(I) mRNA levels; down-regulates in presence of TGFβ1 Osteoblasts Induces osteoblast differentiation, stimulation of collagen I Mediated through smads and cbfa1β Chondrocytes Up-regulation of col2a1 expression in chondrocytic lines Gingival fibroblasts Stimulation of collagen protein synthesis Osteoblasts Inhibits collagen synthesis Chondrocytes Stimulates Col2a1 synthesis
Dediff. chondrocytes, fibroblasts Chondrocytes Fibroblasts Chondrocytes Fibroblasts
Cytokines Enhanced mRNA levels Suppresses CO2A1 transcription Suppression of collagen synthesis + mRNA levels Suppression of α1(II) mRNA levels Suppression of α1(I) mRNA
Hormones and other factors Osteoblasts Decreases α1(I) mRNA and protein levels Calvaria fibroblasts Suppresses collagen production in long-term cultures, reduced transcription rate of α1(I), multifactorial complex effects Fetal rat calvaria Suppression of α1(I) mRNA levels Hypertr. chondrocytes Reduced α1(X) mRNA levels Chondrocytes Induces switch from α1(II) to α1(I), α1(II), and α1(III) mRNA in chondrocytes Chondrocytes Transient stimulation of α1(X) expression Fibroblasts, Enhances prolyl and lysyl hydroxylation chondrocytes, stimulation of α2(I) mRNA transcription, enhanced osteoblasts, etc. mRNA stability Transcription factors and oncogenes c-fos overexpression inhibits α1(I) mRNA levels in osteoblasts Osteoblasts Induces transcription of Col1a1 in osteoblasts by binding to OSE2 elements in the promoter Hypertr. chondrocytes Stimulates Col10a1 expression by binding to OSE2 elements in the Col10a1 promoter Chondrocytes Induces Col2a1 and Col11a1 expression by binding to enhancer Osteoblasts
by membranous bone formation [411–413] (see also Chapter 6). BMP-induced osteoblast differentiation involves the Smad signaling pathway and receptors of the BMPRI group [414]. In BMP-treated cell and organ cultures the induction of osteogenesis is associated with a steep increase on type I collagen synthesis, as a result of enhanced osteoblast differentiation and proliferation. Recently Takagi et al.
Reference 308, 342, 406 408, 409, 410 461, 462 459, 460 379 407 411–415
466–468 416 417 482
439 490, 491 439 492 440–443
338, 418, 419 421, 422, 426
420 397, 488, 489 474–476 474, 477 494
419, 437 402, 403 430–432 401, 433 450–452
[415] have shown that BMP-2 induces the expression of osteogenic transcription factors such as Cbfa1/runx2 (see below), osterix, AJ18 or Msx2. There is no evidence, however, for a stimulation and direct regulation of type I collagen synthesis by BMPs as is the case with TGFβ. Another growth factor which has been shown to enhance collagen type I synthesis in gingival fibroblasts is PDGF
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
[416], while EGF inhibits collagen type I synthesis in osteoblasts [417]. 2. Steroid Hormones
Despite its essential role on bone mineralization, vitamin D3 (1,25-dihydroxycholecalciferol or 1,25(OH)2D3) inhibits type I collagen synthesis in vitro by suppressing the transcription of Col1a1 and Col1a2 genes [338, 418, 419]. Also, parathyroid hormone (PTH), another factor involved in calcium metabolism of bone, has been shown to inhibit type I collagen expression [420]. Glucocorticoids generally suppress synthesis of collagens in fibroblasts [421, 422] and in cartilage [423]. In calvarial bone dexamethasone suppresses collagen production in long-term cultures, but after an initial phase of stimulation [424, 425]. The effect of glucocorticoids can be complex in the presence of other factors, e.g. it potentiates the stimulatory effect of IGF-I on collagen synthesis [425, 426]. Estrogen, however, has been shown to stimulate type I collagen in osteoblasts predominantly in intermediate stages of cell differentiation [427]. 3. Transcription Factors and Oncogenes
A number of transcription factors are involved in controlling differentiation of preosteoblasts to osteoblasts. Dlx5, a homeodomain protein, is one of several Dlx proteins that are expressed in different craniofacial regions and in the limb during development [428]. Overexpression of Dlx5 in osteoblasts from chicken calvaria enhanced expression of type I collagen [429]. Cbfa1/Runx2 has been identified as a Runt-related, osteoblast-specific transcription factor which is essential for osteoblast differentiation and regulation of osteoblastspecific genes such as osteocalcin, osteopontin, and Col1a1 and Col1a2 genes [402, 403, 430–432]. Cbfa1 knockout mice initially develop a nearly complete skeleton. Endochondral and membraneous ossification are, however, completely impaired, indicating that, in the absence of Cbfa1, osteogenic differentiation is blocked at the nonossified preosteoblastic stage. The role of Cbfa1 in the regulation of bone-specific gene expression was shown by its specific binding to consensus sequences in the promoters of the mouse osteocalcin and Col1a1 genes [402]. A consensus Cbfa1 binding site called OSE2 is located at approximately –1350 bp in the promoter of human, mouse, and rat Col1a1 genes [337], but additional OSE2 elements are located also in the proximal promoter. Of several OSE2 elements in the Col1a2 promoter, only one is conserved in other species; this elements binds Cbfa1 and confers osteoblast-specific activity to the minimum Col1a1 promoter in reporter gene assays [337]. In view of its early expression in skeletal development, Cbfa1 seems
23
also involved in the regulation of genes that are expressed before overt osteogenic differentiation [337]. Furthermore, cbfa1/runx2 is one of the major factors regulating expression of type X collagen in hypertrophic chondrocytes [401, 433]. Recently, another osteogenic transcription factor called osterix, which belongs to the zinc finger family, has been identified. It is essential for bone formation [405], but acts downstream of Cbfa1. Similar to Cbfa1 null mice, osterix-deficient mice develop a cartilaginous skeleton with impaired endochondral ossification and complete absence of membraneous ossification. Osterix stimulates type I collagen gene expression in C2C12 cells [405], but it is nuclear whether it binds directly to regulatory elements in the Col1a1 gene. Also, the nuclear matrix seems involved in the regulation of type I collagen gene expression in osteoblasts. In reporter gene assays using UMR-106 osteoblasts it was shown that NP/NMP4 nuclear matrix proteins bind to a 3.5 kb promoter fragment of the human COL1A1 gene [434]. There is ample evidence for a key role of the AP-1 factors c-fos and Fos B in osteoblast differentiation [435, 436], but only few data are available on the regulation of type I collagen in osteoblasts by Ap-1 factors. Kuroki et al. [419, 437] have shown that overexpression of c-fos in the osteoblast cell line MC3T3-E1 inhibits α1(I) mRNA expression. Interestingly, vitamin D, which also inhibits type I collagen synthesis (see above), reverts the inhibitory effect of c-fos to a level of control cells, indicating inhibition of vitamin D binding to the vitamin D receptor [419]. 4. Cytokines
Most inflammatory cytokines such as interleukin-1β or interferon-α inhibit type I collagen synthesis [438]. Suppression of transcription of COL1A2 in hepatocytes by IFNα is exerted through the interaction of p300 with Stat1 [439]. Also, IFNγ plays an important physiological role in inflammation by down-regulating collagen gene expression. Interferon γ (IFNγ) and TNFα both suppress type I collagen mRNA levels in fibroblasts [440–443]. In analyzing the mechanism of IFNγ-induced collagen expression in lung fibroblasts, a trimeric DNA-binding complex named RFX5 was identified which is induced by IFNγ and binds to the Col1a1 transcription start site, thus repressing collagen expression [444, 445]. 5. Regulation of Type V Collagen
Little is known on factors which specifically regulate type V collagen synthesis. There is, however, experimental evidence from DNaseI footprinting and electrophoretic mobility shift, as well as from chromatin immunoprecipitation assays, that the CCAAT-binding
24
KLAUS VON DER MARK
transcription factor CBF/NF-Y regulates the core promoter of the human COL5A1 gene [361].
B. Cartilage Collagen Regulation 1. SOX9
Wright et al. [446] reported on the expression of SOX9, a transcription factor of the SRY (sex-determining region, Y-chromosome) family, in cartilaginous anlagen of the mouse embryo. In situ hybridization studies showed that Sox9 expression parallels expression of the Col2a1 gene in all chondroprogenitor cells and in early, but not hypertrophic, chondrocytes [447, 448], suggesting that it may be involved in chondrogenic development. Furthermore, mutations in the human Sox9 gene cause severe skeletal malformations in patients with campomelic dysplasia [449]. SOX9 has a DNA-binding domain that is homologous to the HMG (High Mobility Group) DNA-binding domain of the SRY gene [450]. Lefebvre et al. [451] and Bell et al. [452] finally demonstrated that SOX9 is a potent activator of Col2a1 gene expression, as it binds to a HMG box consensus sequence CATTCAT present in the 48-bp enhancer of the Col2a1 gene and strongly enhances transcription of Col2a1 reporter genes [354]. In mouse chondrocytes, high levels of SOX9 protein correlate with type II collagenexpressing cells, while dedifferentiated as well as hypertrophic chondrocytes lack SOX9 [354]. The strong chondrogenic potency of SOX9 is further supported by the observation that ectopic expression of SOX9 in transgenic mice activated Col2a1 expression in noncartilaginous tissues [452]. Transactivation of the Col2a1 enhancer by SOX9 is enhanced after phosphorylation of SOX9 by cAMP-dependent protein kinase A [453]. The key role of SOX9 in chondrogenesis is underlined by the fact that haploinsufficiency of SOX9 in Sox9 −/+ heterozygous mice results in defective cartilage primordia [454]. Sox9 gene deletion is lethal in the homozygotes, while in Sox9 −/− chimera all Sox9 −/− cells are excluded from cartilage [455]. SOX9 not only activates transcription of Col2a1, but also of other cartilage matrix genes such as Col11a1 [365], Col9a1, and aggrecan. The transcriptional complex includes also SOX5, L-SOX5, a larger splice variant, and SOX6, which are also essential for chondrogenic development, but cannot activate collagen type II gene expression in the absence of SOX9 [456]. 2. BMP and TGFb Family Proteins
TGFβ, which is the main stimulator of type I collagen expression, has also been shown to induce chondrogenic differentiation in rat periosteum in vivo and in vitro [457].
Conflicting reports exist, however, concerning the effect of TGFβ on type II expression in mature chondrocytes. Redini et al. [458] and Galera et al. [459, 460], have shown up-regulation of type II collagen expression in rabbit articular chondrocytes by TGFβ, while others report that TGFβ1 inhibits type II collagen [461]. Inhibition occurs in a protein kinase C-dependent manner and involves regulatory sequences in the promoter and first intron [462]. Bone morphogenic proteins have been shown to induce chondrogenic differentiation in limb bud mesenchymal cells [463, 464], in periosteal stem cells [465] and in various undifferentiated stem cell lines such as C3H10T1/2, ATDC5, and others [411, 466–468]. As a result of chondrogenic differentiation, type II collagen is up-regulated, and there are also data indicating that type II collagen expression in adult articular chondrocytes in agar cultures is stimulated by BMP-2. In reporter gene studies using MC615 chondrocytes and C3H10T1/2 cells, it was shown that Smad1 and 5 are involved in transactivation of Col2a1 reporter genes after stimulation with BMP-4 [469]. The target genes of Smad1 or 5 that are responsible for upregulation of type II collagen expression have not been identified yet, but several lines of evidence indicate that SOX9 is one of them. BMP responsive cis-acting elements have not been described in the type II collagen gene, but in the type X collagen gene. Several BMP factors, including BMP-4, -6, and -7, promote differentiation of proliferating to hypertrophic chondrocytes [399, 470, 471] as indicated by a steep increase in type X collagen expression. Several functional BMP-responsive elements [472] are, however, not active in the human or bovine COL10A1 gene. 3. Retinoic Acid
Retinoic acid (RA) has rigorous effects on both chondrogenic differentiation and on the stability of the chondrocyte phenotype, dependent on the RA dosis [473, 474]. At high concentrations (10–7 M) RA causes chondrocyte dedifferentiation, associated with a switch from type II to type I and III collagen expression [474, 475]. At low doses (10−9–10–8 M) RA stimulates chondrogenic differentiation of limb mesenchymal cells [476] and induces hypertrophic differentiation, including transient up-regulation of type X collagen expression in chondrocytes [474, 477]. RA inhibits limb chondrogenesis and causes malformation at high doses [478, 479]. There is, however, no information on direct interactions of RA receptors with collagen promoter sequences. The RA effect on type X collagen expression may be regulated through the induction of runx2/cbfa1, which not only induces osteogenic differentiation and the expression of bone-specific genes, but also
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
up-regulates expression of type X collagen in chicken and mouse chondrocytes by binding to OSE2 elements [401]. 4. Cytokines and Growth Factors Regulating Cartilage Collagen Synthesis
Insulin-like growth factor I (IGF-I) is one of the main growth factors regulating cartilage growth during skeletal development. It stimulates chondrocyte proliferation, but also proteoglycan synthesis [480, 481] and type II collagen synthesis [482]. FGF2 (= bFGF) inhibits chondrocyte differentiation [483], but no information is available on direct inteference with collagen gene expression. A key hormone regulating chondrocyte differentiation to hypertrophic cells is PTHrP (parathyroid hormonerelated peptide) which is synthesized in the perichondrium and prehypertrophic chondrocytes [484, 485]. Overexpression of PTHrP in transgenic mice delays endochondral ossification and bone development by stimulating proliferation and inhibiting differentiation of chondrocytes in the proliferative zone of the growth plate [486], while premature chondrocyte hypertrophy, including early onset of type X collagen, is observed in PTHrP-deficient mice [487]. In vitro, PTHrP inhibits expression of type X collagen in rabbit, chicken, and bovine hypertrophic chondrocytes [397, 488, 489]. This inhibition is mediated by c-fos, which is induced by PTHrP and involves a PTHrP-sensitive site located in the COL10A1 enhancer [397]. This enhancer contains an AP1 site, which is however not recognized by c-Fos, but by Fos B and Fra1 factors which stimulate the expression of Col10A1 reporter genes [398]. Interleukin 1 (IL-1) increases type I and III collagen mRNA levels in cultures of human chondrocytes [490], probably owing to the presence of dedifferentiated chondrocytes, while it suppresses the α1(II) mRNA levels [490, 491]. IFNγ also reduces type II collagen mRNA levels in chondrocytes [492]. Ascorbic acid is an essential factor in collagen biosynthesis as it is involved in hydroxylation of proline and hydroxyproline (for review see [493]). But beyond this role, ascorbic acid has been shown to enhance collagen α2(I) mRNA levels six-fold and to enhance mRNA stability in chicken tendon fibroblasts [494].
VIII. CONCLUSIONS With growing knowledge on the variety of collagen types, their structure, physiological features, and the manifold ways of their metabolic regulation, we begin to understand not only the unique biomechanical properties of bone and cartilage, but also the mechanisms regulating
25
their growth and turnover in development and diseases. In bone, densely packed and intensely cross-linked heterofibrils of type I and V collagen provide the architectural scaffold and the substrate for mineralization, and thus hold responsibility for the enormous resistance of bone to mechanical load, torsion, and tension. In hyaline cartilage, heterofibrils of collagen types II and XI, decorated with FACIT collagens, assemble into cross-linked meshworks which allow the incorporation of highly hydrated hyaluronan–proteoglycan complexes, thus providing the tissue with elasticity combined with high resistance to pressure. With better information on tissue-specific posttranslational modifications of collagens, including prolyl and lysyl-hydroxylation, glycosylation, sulfation, oxidation by lysyloxidases, and cross-linking, we begin to understand the unique biochemical and physiological properties of different collagen types in different tissues and may some day understand why, for example, types I or III collagen cannot replace the similar type II collagen in cartilage. A major extension in our understanding of the role of collagens in bone and cartilage was introduced by cell biological and transgenic mouse experiments, showing that collagens transmit – on top of their biomechanical function – specific information to cells by providing substrates for cell adhesion, proliferation, migration, arrangement in the tissue, differentiation, gene expression (e.g. of collagenases), or cell survival. Owing to recent discoveries of specific interactions of type I, II, III, and VI collagens with α1β1, α2β1, α11β1, and α10β1 integrins, the identification of the integrin-binding triple helical epitope GFQGER in type I collagen and the integrin-mediated signaling pathways, we are now in a position to elucidate entire molecular pathways of collagenregulated cell responses in bone and cartilage and other tissues. Impressed by the apparent eternal stability of mammoth bone discovered after 10000 years in Siberian ice, one should keep in mind, however, that bone and its collagens are subject to continuous renewal and turnover by osteoblasts and osteoclasts in the living organism. Numerous growth factors such as TGFβ, BMPs, FGFs, and IGF, steroid hormones such as vitamin D3, corticoids, and cytokines, are involved in the regulation of collagen biosynthesis and turnover in bone. The regulation of cartilage collagen metabolism by BMPs, IGFI, FGF, PTHrP, retinoic acid, and cytokines is not less complicated and varies in different cartilages and developmental stages. Boneand cartilage-specific transcription factors, such as cbfa1/ runx2 and osterix, or SOX9 for cartilage, respectively, were discovered which regulate not only the expression of
26
KLAUS VON DER MARK
collagen genes, but also of other bone- and cartilagespecific genes. Elucidation of the complete pathways of bone- and cartilage-inducing growth factors, such as BMPs and their receptor-mediated regulation of tissuespecific expression of collagens, seems one of the current challenges in bone and cartilage research. This task is matched by the elucidation of the equally complex regulation of collagen degradation by collagenases and the regulation of collagenases by cytokines and cell–matrix interactions, by activation and extracellular inhibitors, as described in detail in Chapter 10. Modern genetic tools have unraveled numerous new collagens also in cartilage and bone. The majority of these collagens, including types XII, XIV, and XVI, XX and XXII, classified as FACIT collagens and a new fibril-forming collagen (type XXIV), are also expressed in cartilage. The functions of these collagens and also of the long-known collagens such as collagen VI in cartilage are, however, yet unclear. Collagen knockout experiments often failed to reveal a clear phenotype, which leaves the field with numerous challenging problems to be solved.
References 1. Boskey, A. L., Wright, T. M., and Blank, R. D. (1999). Collagen and bone strength. J. Bone Miner. Res. 14, 330–335. 2. Kielty, C. M., and Grant, M. E. (2002). The collagen family: structure, assembly, and organisation in the extracellular matrix. In Connective Tisssue and its Heritable Disorders (P. M. Royce, and B. Steinmann, ed.) pp. 159–221, Wiley-Liss, New York. 3. Bruckner, P., and van der Rest, M. (1994). Structure and function of cartilage collagens. Microsc. Res. Tech. 28, 378–384. 4. Eyre, D. (2002). Collagen of articular cartilage. Arthritis Res. 4, 30–35. 5. Niyibizi, C., and Eyre, D. R. (1994). Structural characteristics of cross-linking sites in type V collagen of bone. Chain specificities and heterotypic links to type I collagen. Eur. J. Biochem. 224, 943–950. 6. Mayne, R., and Brewton, R. G. (1993). New members of the collagen superfamily. Curr. Opin. Cell Biol. 5, 883–890. 7. Eyre, D. R., Wu, J. J., Fernandes, R. J., Pietka, T. A., and Weis, M. A. (2002). Recent developments in cartilage research: matrix biology of the collagen II/IX/XI heterofibril network. Biochem. Soc. Trans. 30, 893–899. 8. Ayad, S., Evans, H., Weiss, J. B., and Holt, L. (1984). Type VI collagen but not type V collagen is present in cartilage. Collagen Rel. Res. 4, 165–168. 9. Keene, D. R., and Engvall, E. G. W. (1988). Ultrastructure of type VI collagen in human skin and cartilage suggests an anchoring function for this filamentous network. J. Cell Biol. 107, 1995–2006. 10. Horikawa, O., Nakajima, H., Kikuchi, T., Ichimura, S., Yamada, H., Fujikawa, K., and Toyama, Y. (2004). Distribution of type VI collagen in chondrocyte microenvironment: study of chondrons isolated from human normal and degenerative articular cartilage and cultured chondrocytes. J. Orthop. Sci. 9, 29–36. 11. Poole, C. A., Ayad, S., and Gilbert, R. T. (1992). Chondrons from articular cartilage. V. Immunohistochemical evaluation of type VI collagen organisation in isolated chondrons by light, confocal and electron microscopy. J. Cell. Sci. 103, 1101–1110.
12. von der Mark, K., and Sorokin, L. M. (2002). Adhesive Glycoproteins. In Connective Tissue and its Heritable Disorders (P. M. Royce, and B. Steinmann, eds.), pp. 293–328, Wiley-Liss, New York. 13. Mayne, R., and Burgeson, R. E. (1987). Structure and Function of Collagen Types. In (R. P. Mecham, ed), Academic Press, Inc. Orlando, FL. 14. Bateman, J. F., Lamandé, S. R., and Ramskans, J. A. M. (1996). Collagen superfamily. In “Extrazellular Matrix.” (W. D. Comper, ed.), Vol. 2, pp. 22–67, Harwood Acad. Pulls. 15. van der Rest, M., and Garrone, R. (1991). Collagen family of proteins. Faseb. J. 5, 2814–2823. 16. Piez, K. A. (1998). Molecular and aggregate structure of the collagens. In “Extracellular Matrix Biochemistry” (K. A. Piez, and H. Reddi, eds), Elsevier. 17. Olsen, B. R. (1995). New insights into the function of collagens from genetic analysis. Curr. Opin. Cell. Biol. 7, 720–727. 18. Sandell, L., and Boyd, C. D. (1990). Extracellular Matrix Genes. Academic Press, San Diego, CA. 19. Chu, M. L., and Prockop, D. J. (2002). Collagen: gene structure. In “Connective Tisssue and its Heritable Disorders” (B. Steinmann, and P. M. Royce, eds.), pp. 223–248, Wiley-Liss, New York. 20. Kivirikko, K. I. (1998). Collagen biosynthesis: a mini-review cluster. Matrix Biol. 16, 355–356. 21. Bornstein, P., and Sage, H. (1989). Regulation of collagen gene expression. Prog. Nucleic. Acid. Res. Mol. Biol. 37, 67–106. 22. McLaughlin, S. H., and Bulleid, N. J. (1998). Molecular recognition in procollagen chain assembly. Matrix Biol. 16, 369–377. 23. Prockop, D. J., Sieron, A. L., and Li, S. W. (1998). Procollagen N-proteinase and procollagen C-proteinase. Two unusual metalloproteinases that are essential for procollagen processing probably have important roles in development and cell signaling. Matrix Biol. 16, 399–408. 24. Johnson, L. L., Dyer, R., and Hupe, D. J. (1998). Matrix metalloproteinases. Curr. Opin. Chem. Biol. 2, 466–471. 25. Murphy, G., and Reynolds, J. J. (2002). Extracelllular matrix degradation. In “Connective Tissue and its Heritable Disorders” (P. M. Royce, and B. Steinmann, eds.), pp. 343–384, Wiley-Liss, New York. 26. Myllyharju, J., and Kivirikko, K. I. (2001). Collagens and collagenrelated diseases. Ann. Med. 33, 7–21. 27. Prockop, D. J., and Kivirikko, K. I. (1995). Collagens: molecular biology, diseases, and potentials for therapy. Annu. Rev. Biochem. 64, 403–434. 28. Kuivaniemi, H., and Tromp, G. (1997). Mutations in fibrillar collagens (types I, II, III, and XI), fibril-associated collagen (type IX), and network-forming collagen (type X) cause a spectrum of diseases of bone, cartilage, and blood vessels. Hum. Mut. 9, 300–315. 29. Royce, P. M., and Steinmann, B. (2002). “Connective tissue and its Heritable Disorders,” Wiley Liss, New York. 30. Koch, M., Foley, J. E., Hahn, R., Zhou, P., Burgeson, R. E., Gerecke, D. R., and Gordon, M. K. (2001). Alpha 1(XX) collagen, a new member of the collagen subfamily, fibril-associated collagens with interrupted triple helices. J. Biol. Chem. 276, 23120–23126. 31. Koch, M., Laub, F., Zhou, P., Hahn, R. A., Tanaka, S., Burgeson, R. E., Gerecke, D. R., Ramirez, F., and Gordon, M. K. (2003). Collagen XXIV, a vertebrate fibrillar collagen with structural features of invertebrate collagens: selective expression in developing cornea and bone. J. Biol. Chem. 278, 43236–43244. 32. Koch, M., Schulze, J., Hansen, U., Ashwodt, T., Keene, D. R., Brunken, W. J., Burgeson, R. E., Bruckner, P., and Bruckner-Tuderman, L. (2004). A novel marker of tissue junctions, collagen XXII. J. Biol. Chem. 279, 22514–22521. 33. Chou, M. Y., and Li, H. C. (2002). Genomic organization and characterization of the human type XXI collagen (COL21A1) gene. Genomics 79, 395–401. 34. Banyard, J., Bao, L., and Zetter, B. R. (2003). Type XXIII collagen, a new transmembrane collagen identified in metastatic tumor cells. J. Biol. Chem. 278, 20989–20994.
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone 35. Hashimoto, T., Wakabayashi, T., Watanabe, A., Kowa, H., Hosoda, R., Nakamura, A., Kanazawa, I., Arai, T., Takio, K., Mann, D. M., and Iwatsubo, T. (2002). CLAC: a novel Alzheimer amyloid plaque component derived from a transmembrane precursor, CLAC-P/collagen type XXV. Embo. J. 21, 1524–1534. 36. Sato, K., Yomogida, K., Wada, T., Yorihuzi, T., Nishimune, Y., Hosokawa, N., and Nagata, K. (2002). Type XXVI collagen, a new member of the collagen family, is specifically expressed in the testis and ovary. J. Biol. Chem. 277, 37678–37684. 37. Boot-Handford, R. P., Tuckwell, D. S., Plumb, D. A., Rock, C. F., and Poulsom, R. (2003). A novel and highly conserved collagen (pro(alpha)1(XXVII)) with a unique expression pattern and unusual molecular characteristics establishes a new clade within the vertebrate fibrillar collagen family. J. Biol. Chem. 278, 31067–31077. 38. Brodsky, B., and Ramshaw, J. A. (1997). The collagen triple-helix structure. Matrix Biol. 15, 545–554. 39. Berisio, R., Vitagliano, L., Mazzarella, L., and Zagari, A. (2002). Recent progress on collagen triple helix structure, stability and assembly. Protein Pept. Lett. 9, 107–116. 40. Mizuno, K., Hayashi, T., Peyton, D. H., and Bachinger, H. P. (2004). Hydroxylation-induced stabilization of the collagen triple helix. Acetyl-(glycyl-4(R)-hydroxyprolyl-4(R)-hydroxyprolyl)(10)-NH(2). forms a highly stable triple helix. J. Biol. Chem. 279, 38072–38078. 41. Hofmann, H., Fietzek, P. P., and Kuhn, K. (1978). The role of polar and hydrophobic interactions for the molecular packing of type I collagen: a three-dimensional evaluation of the amino acid sequence. J. Mol. Biol. 125, 137–165. 42. Berg, R. A., and Prockop, D. J. (1973). The thermal transition of a nonhydroxylated form of collagen. Evidence for a role for hydroxyproline in stabilizing the triple-helix of collagen. Biochem. Biophys. Res. Commun. 52, 115–120. 43. Schmid, T. M., and Linsenmayer, T. F. (1984). Denaturation-renaturation properties of two molecular forms of short-chain cartilage collagen. Biochem. 23, 553–558. 44. Risteli, J., Baechinger, H. P., Engel, J., Furthmayr, H., and Timpl, R. (1980). 7S-collagen: characterization of an unusual basement membrane structure. Eur. J. Biochem. 108, 239–250. 45. Kuehn, K., Wiedemann, H., Timpl, R., Risteli, J., Dieringer, H., Voss, T., and Glanville, R. (1981). Macromolecular structure of basement membrane collagen. FEBS Lett. 125, 123–128. 46. Bruckner, P., and Prockop, D. J. (1981). Proteolytic enzymes as probes for the triple-helical conformation of procollagen. Analyt. Biochem. 110, 360–368. 47. Kodama, T., Doi, T., Suzuki, H., Takahashi, K., Wada, Y., and Gordon, S. (1996). Collagenous macrophage scavenger receptors. Curr. Opin. Lipidol. 7, 287–291. 48. Elomaa, O., Sankala, M., Pikkarainen, T., Bergmann, U., Tuuttila, A., Raatikainen-Ahokas, A., Sariola, H., and Tryggvason, K. (1998). Structure of the human macrophage MARCO receptor and characterization of its bacteria-binding region. J. Biol. Chem. 273, 4530–4538. 49. Fessler, J. H., Doege, K. J., Duncan, K. G., and Fessler, L. I. (1985). Biosynthesis of collagen. J. Cell Biochem. 28, 31–37. 50. Fessler, J. H., Shigaki, N., and Fessler, L. I. (1985). Biosynthesis and properties of procollagens V. Ann. Acad. Sci. N.Y. 460, 181–186. 51. Fleischmajer, J. S., Perlish, J. S., and Olsen, B. R. (1987). Amino and carboxyl propeptides in bone collagen fibrils during embryogenesis. Cell Tissue Res. 247, 105–109. 52. Kadler, K. E., Hojima, Y., and Prockop, D. J. (1987). Assembly of collagen fibrils de novo by cleavage of the type I PC-collagen with procollagen C-proteinase – assay of critical concentration demonstrates that collagen self-assembly is a classical example of an entropy-driven process. J. Biol. Chem. 262, 15696–15701. 53. Hinek, A., Reiner, A., and Poole, R. (1987). The calcification of cartilage matrix in chondrocyte culture: Studies of the C-propetide of type II collagen (chondrocalcin). J. Cell Biol. 104, 1435–1441.
27
54. Bogin, O., Kvansakul, M., Rom, E., Singer, J., Yayon, A., and Hohenester, E. (2002). Insight into Schmid metaphyseal chondrodysplasia from the crystal structure of the collagen X NC1 domain trimer. Structure 10, 165–173. 55. Schmid, T. M., Cole, A. A., Chen, Q., Bonen, D. K., Luchene, L., and Linsenmayer, T. (1994). Assembly of type X collagen by hypertrophic chondrocytes. In “Extracellular Matrix Assembly and Structure” (P. Yurchenco, D. Birk, and R. Mecham, eds.), pp. 171–206, Academic Press. 56. Frischholz, S., Beier, F., Girkontaite, I., Wagner, K., Pöschl, E., Turnay, J., Mayer, U., and von der Mark, K. (1998). Characterization of human type X procollagen and its NC-1 domain expressed as recombinant proteins in HEK293 cells. J. Biol. Chem. 273, 4547–4555. 57. Shaw, L. M., and Olsen, B. R. (1991). FACIT collagens: diverse molecular bridges in extracellular matrices. Trends. Biochem. Sci. 16, 191–194. 58. Oh, S. P., Warman, M. L., Seldin, M. F., Cheng, S. D., Knoll, J. H., Timmons, S., and Olsen, B. R. (1994). Cloning of cDNA and genomic DNA encoding human type XVIII collagen and localization of the alpha 1(XVIII) collagen gene to mouse chromosome 10 and human chromosome 21. Genomics 19, 494–499. 59. Yamagata, M., Yamada, K. M., Yamada, S. S., Shinomura, T., Tanaka, H., Nishida, Y., Obara, M., and Kimata, K. (1991). The complete primary structure of type XII collagen shows a chimeric molecule with reiterated fibronectin type III motifs, von Willebrand factor A motifs, a domain homologous to a noncollagenous region of type IX collagen, and short collagenous domains with an Arg-Gly-Asp site. J. Cell Biol. 115, 209–221. 60. O’Reilly, M. S., Boehm, T., Shing, Y., Fukai, N., Vasios, G., Lane, W. S., Flynn, E., Birkhead, J. R., Olsen, B. R., and Folkman, J. (1997). Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 88, 277–285. 61. Hagg, P., Rehn, M., Huhtala, P., Vaisanen, T., Tamminen, M., and Pihlajaniemi, T. (1998). Type XIII collagen is identified as a plasma membrane protein. J. Biol. Chem. 273, 15590–15597. 62. Pihlajaniemi, T., Tamminen, M. (1990). The alpha1 chain of type XIII collagen consists of three collagenous and four non-collagenous domains, and its primary transcript undergoes complex alternative splicing. J. Biol. Chem. 265, 16922–16928. 63. Gordon, M. K., Fitch, J. M., Foley, J. W., Gerecke, D. R., Linsenmayer, C., Birk, D. E., and Linsenmayer, T. F. (1997). Type XVII collagen (BP 180) in the developing avian cornea. Invest. Ophthalmol. Vis. Sci. 38, 153–166. 64. Farjanel, J., Schurmann, G., and Bruckner, P. (2001). Contacts with fibrils containing collagen I, but not collagens II, IX, and XI, can destabilize the cartilage phenotype of chondrocytes. Osteoarthritis Cartilage 9, S55–S63. 65. Tulla, M., Pentikainen, O. T., Viitasalo, T., Kapyla, J., Impola, U., Nykvist, P., Nissinen, L., Johnson, M. S., and Heino, J. (2001). Selective binding of collagen subtypes by integrin alpha 1I, alpha 2I, and alpha 10I domains. J. Biol. Chem. 276, 48206–48212. 66. Tiger, C. F., Fougerousse, F., Grundstrom, G., Velling, T., and Gullberg, D. (2001). Alpha11beta1 integrin is a receptor for interstitial collagens involved in cell migration and collagen reorganization on mesenchymal nonmuscle cells. Dev. Biol. 237, 116–129. 67. Jokinen, J., Dadu, E., Nykvist, P., Kapyla, J., White, D. J., Ivaska, J., Vehvilainen, P., Reunanen, H., Larjava, H., Hakkinen, L., and Heino, J. (2004). Integrin-mediated cell adhesion to type I collagen fibrils. J. Biol. Chem. 279, 31956–31963. 68. Aszodi, A., Bateman, J. F., Gustafsson, E., Boot-Handford, R., and Fassler, R. (2000). Mammalian skeletogenesis and extracellular matrix: what can we learn from knockout mice? Cell. Struct. Funct. 25, 73–84. 69. Aszodi, A., Chan, D., Hunziker, E., Bateman, J. F., and Fassler, R. (1998). Collagen II is essential for the removal of the notochord and the formation of intervertebral discs [In Process Citation]. J. Cell Biol. 143, 1399–1412.
28 70. Hagg, R., Hedbom, E., Mollers, U., Aszodi, A., Fassler, R., and Bruckner, P. (1997). Absence of the alpha1(IX) chain leads to a functional knock-out of the entire collagen IX protein in mice. J. Biol. Chem. 272, 20650–20654. 71. Cheah, K. S., Lau, E. T., Au, P. K., and Tam, P. P. (1991). Expression of the mouse alpha 1(II) collagen gene is not restricted to cartilage during development. Development 111, 945–953. 72. von der Mark, K., von der Mark, H., Timpl, R., and Trelstad, R. L. (1977). Immunofluorescent localization of collagen type-1, type-2 and type-3 in embryonic chick eye. Dev. Biol. 59, 75–85. 73. Linsenmayer, T. F., Smith, G. N., and Hay, E. D. (1977). Synthesis of 2 collagen types by embryonic chick corneal epithelium in vitro. P. Nas. US 74, 39–43. 74. Hay, E. D. (1981). Collagen and embryonic development. In “Cell biology of extracellular matrix” (E. D. Hay, ed.) Plenum Press, N. Y., pp. 379–409. 75. von der Mark, H., von der Mark, K., and Gay, S. (1976). Study of differential collagen synthesis during development of the chick embryo by immunofluorescence. I. Preparation of collagen type I and type II specific antibodies and their application to early stages of the chick embryo. Develop. Biol. 48, 237–249. 76. Niyibizi, C., and Eyre, D. R. (1989). Bone type V collagen: chain composition and location of a trypsin cleavage site. Connect. Tissue Res. 20, 247–250. 77. Hulmes, D. J. (1992). The collagen superfamily – diverse structures and assemblies. Essays. Biochem. 27, 49–67. 78. Ramachandran, G. N. (1988). Stereochemistry of collagen. Int. J. Pept. Protein Res. 31, 1–16. 79. Adachi, E., and Hayashi, T. (1985). In vitro formation of fine fibrils with a D-periodic banding pattern from type V collagen. Collagen Rel. Res. 5, 225–232. 80. Birk, D. E. (2001). Type V collagen: heterotypic type I/V collagen interactions in the regulation of fibril assembly. Micron. 32, 223–237. 81. Glimcher, M. J., Shapiro, F., Ellis, R. D., and Eyre, D. R. (1980). Changes in tissue morphology and collagen composition during the repair of cortical bone in the adult chicken. J. Bone Am. V. 62, 964–973. 82. Lees, S. (2003). Mineralization of type I collagen. Biophys. J. 85, 204–207. 83. Giraud-Guille, M. M., Besseau, L., and Martin, R. (2003). Liquid crystalline assemblies of collagen in bone and in vitro systems. J. Biomech. 36, 1571–1579. 84. Tromp, G., Kuivaniemi, H., Stacey, A., Shikata, H., Baldwin, C. T., Jaenisch, R., and Prockop, D. J. (1988). Structure of a full-length cDNA clone for the prepro alpha 1(I) chain of human type I procollagen. Biochem. J. 253, 633–640. 85. Kuivaniemi, H., Tromp, G., Chu, M. L., and Prockop, D. J. (1988). Structure of a full-length cDNA clone for the prepro alpha 2(I) chain of human type I procollagen. Comparison with the chicken gene confirms unusual patterns of gene conservation. Biochem. J. 252, 633–640. 86. Gerstenfeld, L. C., Chipman, S. D., Kelly, C. M., Hodgens, K. J., Lee, D. D., and Landis, W. J. (1988). Collagen expression, ultrastructural assembly, and mineralization in cultures of chicken embryo osteoblasts. J. Cell Biol. 106, 979–989. 87. Kratochwil, K., Ghaffari-Tabrizi, N., Holzinger, I., and Harbers, K. (1993). Restricted expression of Mov13 mutant alpha 1(I) collagen gene in osteoblasts and its consequences for bone development. Dev. Dyn. 198, 273–283. 88. Fleischmajer, R., MacDonald, E. D., Perlish, J. S., Burgeson, R. E., and Fisher, L. W. (1990). Dermal collagen fibrils are hybrids of type I and type III collagen molecules. J. Struct. Biol. 105, 162–169. 89. Birk, D. E., Fitch, J. M., Babiarz, J. P., and Linsenmayer, T. F. (1988). Collagen type I and type V are present in the same fibril in the avian corneal stroma. J. Cell Biol. 106, 999–1008. 90. von der Mark, K. (1981). Localization of collagen types in tissues. Int. Rev. Conn. Tissue Res. 9, 265–324.
KLAUS VON DER MARK 91. Byers, P. H. (2000). Osteogenesis imperfecta: perspectives and opportunities. Curr. Opin. Pediatr. 12, 603–609. 92. Brenner, R. E., Vetter, U., Nerlich, A., Worsdorfer, O., Teller, W. M., and Muller, P. K. (1989). Biochemical analysis of callus tissue in osteogenesis imperfecta type IV. Evidence for transient overmodification in collagen types I and III. J. Clin. Invest. 84, 915–921. 93. Rao, V. H., Steinmann, B., de Wet, W., and Hollister, D. W. (1989). Decreased thermal denaturation temperature of osteogenesis imperfecta mutant collagen is independent of post-translational overmodifications of lysine and hydroxylysine. J. Biol. Chem. 264, 1793–1798. 94. Montesano, R., Schaller, C., and Orci, L. (1991). Induction of epithelial tubular morphogenesis in vitro by fibroblast-derived soluble factors. Cell 66, 697–711. 95. Emsley, J., Knight, C. G., Farndale, R. W., Barnes, M. J., and Liddington, R. C. (2000). Structural basis of collagen recognition by integrin alpha2beta1. Cell 101, 47–56. 96. Brakebusch, C., and Fassler, R. (2003). The integrin-actin connection, an eternal love affair. Embo. J. 22, 2324–2333. 97. DeMali, K. A., Wennerberg, K., and Burridge, K. (2003). Integrin signaling to the actin cytoskeleton. Curr. Opin. Cell Biol. 15, 572–582. 98. Giancotti, F. G., and Ruoslahti, E. (1999). Integrin signaling. Science 285, 1028–1032. 99. Velling, T., Risteli, J., Wennerberg, K., Mosher, D. F., and Johansson, S. (2002). Polymerization of type I and III collagens is dependent on fibronectin and enhanced by integrins alpha 11beta 1 and alpha 2beta 1. J. Biol. Chem. 277, 37377–37381. 100. Still, J., Glat, P., Silverstein, P., Griswold, J., and Mozingo, D. (2003). The use of a collagen sponge/living cell composite material to treat donor sites in burn patients. Burns 29, 837–841. 101. Ueda, H., Hong, L., Yamamoto, M., Shigeno, K., Inoue, M., Toba, T., Yoshitani, M., Nakamura, T., Tabata, Y., and Shimizu, Y. (2002). Use of collagen sponge incorporating transforming growth factor-beta1 to promote bone repair in skull defects in rabbits. Biomaterials 23, 1003–1010. 102. Yamanouchi, K., Satomura, K., Gotoh, Y., Kitaoka, E., Tobiume, S., Kume, K., and Nagayama, M. (2001). Bone formation by transplanted human osteoblasts cultured within collagen sponge with dexamethasone in vitro. J. Bone Miner. Res. 16, 857–867. 103. Jaenisch, R., Harbers, K., Schnieke, A., Lohler, J., Chumakov, I., Jahner, D., Grotkopp, D., and Hoffmann, E. (1983). Germline integration of moloney murine leukemia virus at the Mov13 locus leads to recessive lethal mutation and early embryonic death. Cell 32, 209–216. 104. Hartung, S., Jaenisch, R., and Breindl, M. (1986). Retrovirus insertion inactivates mouse alpha 1(I) collagen gene by blocking initiation of transcription. Nature 320, 365–367. 105. Kratochwil, K., Dziadek, M., Lohler, J., Harbers, K., and Jaenisch, R. (1986). Normal epithelial branching morphogenesis in the absence of collagen I. Dev. Biol. 117, 596–606. 106. Schwarz, M., Harbers, K., and Kratochwil, K. (1990). Transcription of a mutant collagen-I gene is a cell type and stage-specific marker for odontoblast and osteoblast differentiation. Development 108, 717–726. 107. Kratochwil, K., von der Mark, K., Kollar, E. J., Jaenisch, R., Mooslehner, K., Schwarz, M., Haase, K., Gmacht, I., and Harbers, K. (1992). Retrovirus-induced insertional mutation in Mov13 mice affects collagen-I expression in a tissue-specific manner. Cell 57, 807–816. 108. Burgeson, R. E., Hebda, P. A., Morris, N. P., and Hollister, D. (1982). Human cartilage collagens. Comparison of cartilage collagens with human type V collagen. J. Biol. Chem. 257, 7852–7856. 109. Burgeson, R. E., and Hollister, D. (1979). Collagen heterogeneity in human cartilage: identification of several new collagen chains. Biochem. Biophys. Res. Commun. 87, 1124–1131. 110. Zhidkova, N. I., Brewton, R. G., and Mayne, R. (1993). Molecular cloning of PARP (proline/arginine-rich protein) from human cartilage and subsequent demonstration that PARP is a fragment of the
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126. 127.
128.
129.
NH2-terminal domain of the collagen alpha 2(XI) chain. FEBS Lett. 326, 25–28. Fessler, L., Kumamoto, C., Meis, M., and Fessler, J. (1981). Assembly and processing of procollagen V (AB) in chick blood vessels and other tissues. J. Biol. Chem. 256, 9640–9645. Fessler, L. I., Brosh, S., Chapin, S., and Fessler, J. H. (1986). Tyrosine sulfation in precursors of collagen V. J. Biol. Chem. 261, 5034–5040. Wu, J. J., Lark, M. W., Chun, L. E., and Eyre, D. R. (1991). Sites of stromelysin cleavage in collagen types II, IX, X, and XI of cartilage. J. Biol. Chem. 266, 5625–5628. Haralson, M. A., Mitchell, W. M., Rhodes, R. K., Kresina, T. F., Gay, R., and Miller, E. J. (1980). Chinese hamster lung cells synthesize and confine to the cellular domain a collagen composed solely of B chains. Proc. Natl. Acad. Sci. USA 77, 5206–5210. Eyre, D., and Wu, J.-J. (1987). Type XI or 1alpha, 2alpha, 3alpha collagen. In: “Stucture and Function of Collagen Types” (R. Mayne and R.E. Burgeson, eds.), pp. 261–281. Niyibizi, C., and Eyre, D. R. (1989). Identification of the cartilage alpha 1(XI) chain in type V collagen from bovine bone. FEBS Lett. 242, 314–318. Mayne, R., Brewton, R. G., Mayne, P. M., and Baker, J. R. (1993). Isolation and characterization of the chains of type V/type XI collagen present in bovine vitreous. J. Biol. Chem. 268, 9381–9386. von der Mark, K., and Oecalan, M. (1982). Immunofluorescent localization of type V collagen in the chick embryo with monoclonal antibodies. Collagen Rel. Res. 2, 541–555. Linsenmayer, T. F., Fitch, J. M., Schmid, T. M., Zak, N. B., Gibney, E., Sanderson, R. D., and Mayne, R. (1983). Monoclonal antibodies against chicken type V collagen: Production, specificity, and use for immunocytochemical localization in embryonic cornea and other organs. J. Cell Biol. 96, 124–132. Mayne, R. (1989). Cartilage collagens – what is their function, and are they involved in articular disease. Arthritis Rheumatism 32, 241–246. von der Mark, K., and von der Mark, H. (1977). Role of 3 genetically distinct collagen types in endochondral ossification and calcification of cartilage. J. Bone Joint Res. 59, 458–464. Eyre, D. R., Brickley-Parsons, D. M., and Glimcher, M. J. (1978). Predominance of type I collagen at the surface of avian articular cartilage. FEBS Lett. 85, 259–263. Goldwasser, M., Astley, T., van der Rest, M., and Glorieux, F. H. (1982). Analysis of the type of collagen present in osteoarthritic human cartilage. Clin. Orthop. Rev. Res. 167, 296–302. Ronziere, M. C., Ricard-Blum, S., Tiollier, J., Hartmann, D. J., Garrone, R., and Herbage, D. (1990). Comparative analysis of collagens solubilized from human foetal, and normal and osteoarthritic adult articular cartilage, with emphasis on type VI collagen. Biochim. Biophys. Acta 1038, 222–230. Aigner, T., Bertling, W., Stöb, H., Weseloh, G., and von der Mark, K. (1993). Independent expression of fibril-forming collagens I, II, and III in chondrocytes in human osteoarthritic cartilage. J. Clin. Invest. 91, 829–837. Wotton, S. F., and Duance, V. C. (1994). Type III collagen in normal human articular cartilage. Histochem. J. 26, 412–416. Young, R. D., Lawrence, P. A., Duance, V. C., Aigner, T., and Monaghan, P. (2000). Immunolocalization of collagen types II and III in single fibrils of human articular cartilage. J. Histochem. Cytochem. 48, 423–432. Eerola, I., Salminen, H., Lammi, P., Lammi, M., von der Mark, K., Vuorio, E., and Saamanen, A. M. (1998). Type X collagen, a natural component of mouse articular cartilage: association with growth, aging, and osteoarthritis. Arthritis Rheum. 41, 1287–1295. Lammi, P. E., Lammi, M. J., Hyttinen, M. M., Panula, H., Kiviranta, I., and Helminen, H. J. (2002). Site-specific immunostaining for type X
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
29
collagen in noncalcified articular cartilage of canine stifle knee joint. Bone 31, 690–696. Rucklidge, G. J., Milne, G., and Robins, S. P. (1996). Collagen type X: a component of the surface of normal human, pig, and rat articular cartilage. Biochem. Biophys. Res. Commun. 224, 297–302. von der Mark, K., Kirsch, T., Nerlich, A., Kuss, A., Weseloh, G., Glueckert, K., and Stoess, H. (1992). Type X collagen synthesis in human osteoarthritic cartilage: Indication of chondrocyte hypertrophy. Arthritis Rheum. 35, 806–811. Aigner, T., Reichenberger, E., Bertling, W., Kirsch, T., Stöb, H., and von der Mark, K. (1993). Type X collagen expression in osteoarthritic and rheumatoid articular cartilage. Virchows Arch. 63, 205–211. Kwan, K. M., Pang, M. K., Zhou, S., Cowan, S. K., Kong, R. Y., Pfordte, T., Olsen, B. R., Sillence, D. O., Tam, P. P., and Cheah, K. S. (1997). Abnormal compartmentalization of cartilage matrix components in mice lacking collagen X: implications for function. J. Cell Biol. 136, 459–471. Jacenko, O., Chan, D., Franklin, A., Ito, S., Underhill, C. B., Bateman, J. F., and Campbell, M. R. (2001). A dominant interference collagen X mutation disrupts hypertrophic chondrocyte pericellular matrix and glycosaminoglycan and proteoglycan distribution in transgenic mice. Am. J. Pathol. 159, 2257–2269. Jacenko, O., Roberts, D. W., Campbell, M. R., McManus, P. M., Gress, C. J., and Tao, Z. (2002). Linking hematopoiesis to endochondral skeletogenesis through analysis of mice transgenic for collagen X. Am. J. Pathol. 160, 2019–2034. Topping, R. E., Bolander, M. E., and Balian, G. (1994). Type X collagen in fracture callus and the effects of experimental diabetes. Clin. Orthop. 308, 220–228. Trelstad, R. L., Kang, A. H., Igarashi, S., and Gross, J. (1970). Isolation of two distinct collagens from chick cartilage. Biochemistry 9, 4993–4998. Miller, E. J. (1971). Isolation and characterization of a collagen from chick cartilage containing three identical alpha chains. Biochemistry 10, 1652–1659. Ryan, M. C., and Sandell, L. J. (1990). Differential expression of a cysteine-rich domain in the amino- terminal propeptide of type II procollagen by alternative splicing of messenger-RNA. J. Biol. Chem. 265, 10334–10339. Sandell, L. J., Nalin, A. M., and Reife, R. A. (1994). Alternative splice form of type II procollagen mRNA (IIA) is predominant in skeletal precursors and non-cartilaginous tissues during early mouse development. Dev. Dyn. 199, 129–140. Madsen, K., Moskalewski, S., von der Mark, K., and Friberg, U. (1983). Synthesis of proteoglycans, collagen, and elastin by cultures of rabbit auricular chondrocytes – relation to age of the donor. Dev. Biol. 96, 63–73. Eyre, D. R., and Muir, H. (1976). Types I and II collagens in intervertebral disc. Interchanging radial distributions in annulus fibrosus. Biochem. J. 157, 267–270. Yang, C. L., Rui, H., Mosler, S., Notbohm, H., Sawaryn, A., and Muller, P. K. (1993). Collagen II from articular cartilage and annulus fibrosus. Structural and functional implication of tissue specific posttranslational modifications of collagen molecules. Eur. J. Biochem. 213, 1297–1302. Newsome, D. A., Linsenmayer, T. F., and Trelstad, R. L. (1976). Vitreous body collagen. Evidence for a dual origin from the neural retina and hyalocytes. J. Cell Biol. 71, 59–67. Trelstad, R. L., and Kang, A. H. (1974). Collagen heterogeneity in the avian eye: lens, vitreous body, cornea and sclera. Exp. Eye Res. 18, 395–406. Thorogood, P., Bee, J., and von der Mark, K. (1986). Transient expression of collagen type II at epitheliomesenchymal interfaces during morphogenesis of the cartilaginous neurocranium. Develop. Biol. 116, 497–509.
30 147. Wood, A., Ashhurst, D. E., Corbett, A., and Thorogood, P. (1991). The transient expression of type II collagen at tissue interfaces during mammalian craniofacial development. Development 111, 955–968. 148. Mundlos, S., and Olsen, B. R. (1997). Heritable diseases of the skeleton. Part II: Molecular insights into skeletal development-matrix components and their homeostasis. Faseb. J. 11, 227–233. 149. Spranger, J., Winterpacht, A., and Zabel, B. (1994). The type II collagenopathies: a spectrum of chondrodysplasias. Eur. J. Pediatr. 153, 56–65. 150. Byers, P. H. (1994). Molecular genetics of chondrodysplasias, including clues to development, structure, and function. Curr. Opin. Rheumatol. 6, 345–350. 151. Chan, D., Cole, W. G., Chow, C. W., Mundlos, S., and Bateman, J. F. (1995). COL2A1 mutation in achondrogenesis type II results in the replacement of type II collagen type I and III collagens in cartilage. J. Biol. Chem. 270, 1747–1753. 152. Mundlos, S., Chan, D., McGill, J., and Bateman, J. F. (1996). An alpha 1(II) Gly913 to Cys substitution prevents the matrix incorporation of type II collagen which is replaced with type I and III collagens in cartilage from a patient with hypochondrogenesis. Am. J. Med. Genet. 63, 129–136. 153. Aszodi, A., Hunziker, E. B., Olsen, B. R., and Fassler, R. (2001). The role of collagen II and cartilage fibril-associated molecules in skeletal development. Osteoarthritis Cartilage 9, Suppl A, S150–159. 154. Conrad, G. W., Kelly, P. T., von der Mark, K., and Edelhauser, H. F. (1981). A comparative study of elasmobranch corneal and scleral collagens. Expt. Eye Res. 32, 659–672. 155. Holmdahl, R., Malmstrom, V., and Vuorio, E. (1993). Autoimmune recognition of cartilage collagens. Ann. Med. 25, 251–264. 156. Stuart, J. M., Watson, W. C., and Kang, A. H. (1988). Collagen autoimmunity and arthritis. Faseb. J. 2, 2950–2956. 157. Watson, W. C., Cremer, M. A., Wooley, P. H., and Townes, A. S. (1986). Assessment of the potential pathogenicity of type II collagen autoantibodies in patients with rheumatoid arthritis. Evidence of restricted IgG3 subclass expression and activation of complement C5 to C5a. Arthritis Rheum. 29, 1316–1321. 158. Holmdahl, R., Andersson, M., Goldschmidt, T. J., Gustafsson, K., Jansson, L., and Mo, J. A. (1990). Type II collagen autoimmunity in animals and provocations leading to arthritis. Immunol. Rev. 118, 193–232. 159. Andersson, M., Cremer, M. A., Terato, K., Burkhardt, H., and Holmdahl, R. (1991). Analysis of type II collagen reactive T cells in the mouse. II. Different localization of immunodominant T cell epitopes on heterologous and autologous type II collagen. Scand. J. Immunol. 33, 505–510. 160. Schulte, S., Unger, C., Mo, J. A., Wendler, O., Bauer, E., Frischholz, S., von der Mark, K., Kalden, J. R., Holmdahl, R., and Burkhardt, H. (1998). Arthritis-related B cell epitopes in collagen II are conformationdependent and sterically privileged in accessible sites of cartilage collagen fibrils. J. Biol. Chem. 273, 1551–1561. 161. Burkhardt, H., Koller, T., Engstrom, A., Nandakumar, K. S., Turnay, J., Kraetsch, H. G., Kalden, J. R., and Holmdahl, R. (2002). Epitopespecific recognition of type II collagen by rheumatoid arthritis antibodies is shared with recognition by antibodies that are arthritogenic in collagen-induced arthritis in the mouse. Arthritis Rheum. 46, 2339–2348. 162. Backlund, J., Carlsen, S., Hoger, T., Holm, B., Fugger, L., Kihlberg, J., Burkhardt, H., and Holmdahl, R. (2002). Predominant selection of T cells specific for the glycosylated collagen type II epitope (263–270) in humanized transgenic mice and in rheumatoid arthritis. Proc. Natl. Acad. Sci. USA 99, 9960–9965. 163. Morris, N. P., and Baechinger, H. P. (1987). Type XI collagen is a heterotrimer with the composition (1alpha, 2alpha, 3alpha) retaining non-triple-helical domains. J. Biol. Chem. 262, 11345–11350.
KLAUS VON DER MARK 164. Wu, J. J., and Eyre, D. R. (1995). Structural analysis of cross-linking domains in cartilage type XI collagen. Insights on polymeric assembly. J. Biol. Chem. 270, 18865–18870. 165. Oxford, J. T., DeScala, J., Morris, N., Gregory, K., Medeck, R., Irwin, K., Oxford, R., Brown, R., Mercer, L., and Cusack, S. (2004). Interaction between amino propeptides of type XI procollagen alpha1 chains. J. Biol. Chem. 279, 10939–10945. 166. Davies, G. B., Oxford, J. T., Hausafus, L. C., Smoody, B. F., and Morris, N. P. (1998). Temporal and spatial expression of alternative splice-forms of the alpha1(XI) collagen gene in fetal rat cartilage. Dev. Dyn. 213, 12–26. 167. Lui, V. C. H., Ng, L. J., Sat, E. W. Y., Nicholls, J., and Cheah, K. S. E. (1996). Extensive alternative splicing within the amino-propeptide coding domain of alpha2(XI) procollagen mRNAs. Expression of transcripts encoding truncated pro-alpha chains. J. Biol. Chem. 271, 16945–16951. 168. Tsumaki, N., and Kimura, T. (1995). Differential expression of an acidic domain in the amino- terminal propeptide of mouse pro-alpha 2(XI) collagen by complex alternative splicing. J. Biol. Chem. 270, 2372–2378. 169. Neame, P. J., Young, C. N., and Treep, J. T. (1990). Isolation and primary structure of PARP, a 24-kDa proline- and arginine-rich protein from bovine cartilage closely related to the NH2-terminal domain in collagen alpha 1 (XI). J. Biol. Chem. 265, 20401–20408. 170. Mendler, M., Eich-Bender, S. G., Vaughan, L., Winterhalter, K. H., and Bruckner, P. (1989). Cartilage contains mixed fibrils of collagen types II, IX and XI. J. Cell Biol. 108, 191–197. 171. Blaschke, U. K., Eikenberry, E. F., Hulmes, D. J., Galla, H. J., and Bruckner, P. (2000). Collagen XI nucleates self-assembly and limits lateral growth of cartilage fibrils. J. Biol. Chem. 275, 10370–10378. 172. Gregory, K. E., Oxford, J. T., Chen, Y., Gambee, J. E., Gygi, S. P., Aebersold, R., Neame, P. J., Mechling, D. E., Bachinger, H. P., and Morris, N. P. (2000). Structural organization of distinct domains within the non-collagenous N-terminal region of collagen type XI. J. Biol. Chem. 275, 11498–11506. 173. Li, Y., Lacerda, D. A., Warman, M. L., Beier, D. R., Yoshioka, H., Ninomiya, Y., Oxford, J. T., Morris, N. P., Andrikopoulos, K., and Ramirez, F. (1995). A fibrillar collagen gene, Col11a1, is essential for skeletal morphogenesis. Cell 80, 423–430. 174. Vikkula, M., Mariman, E. C., Lui, V. C., Zhidkova, N. I., Tiller, G. E., Goldring, M. B., van Beersum, S. E., de Waal Malefijt, M. C., van den Hoogen, F. H., and Ropers, H. H. (1995). Autosomal dominant and recessive osteochondrodysplasias associated with the COL11A2 locus. Cell 80, 431–437. 175. Shimokomaki, M., Duance, V. C., and Bailey, A. J. (1980). Identification of a new disulphide bonded collagen from cartilage. FEBS Lett. 121, 51–54. 176. Reese, C. A., and Mayne, R. (1981). Minor collagens of chicken hyaline cartilage. Biochem. 20, 5443–5448. 177. Ninomiya, Y., and Olsen, B. R. (1984). Synthesis and characterization of cDNA encoding a cartilage-specific short collagen. Proc. Natl. Acad. Sci. USA 81, 3014–3018. 178. von der Mark, K., van Menxel, M., and Wiedemann, H. (1984). Isolation and characterization of a precursor form of M-collagen from embryonic chicken cartilage. Eur. J. Biochem. 138, 629–633. 179. Mayne, R., and von der Mark, K. (1983). Collagens of cartilage. Biochem. Cartilage I, 181–214. 180. van der Rest, M., Mayne, R., Ninomiya, Y., Seidah, N. G., Chretien, M., and Olsen, B. R. (1985). The structure of type IX collagen. J. Biol. Chem. 260, 220–225. 181. Vaughan, L., Mendler, M., Huber, S., Bruckner, P., Winterhalter, K. H., Irwin, M. I., and Mayne, R. (1988). D-periodic distribution of collagen type IX along cartilage fibrils. J. Cell Biol. 106, 998–997.
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone 182. van der Rest, M., Mayne, R. (1988). Type IX collagen proteoglycan from cartilage is covalently cross-linked to type II collagen. J. Biol. Chem. 263, 1615–1618. 183. Wu, J. J., Woods, P. E., and Eyre, D. R. (1992). Identification of crosslinking sites in bovine cartilage type IX collagen reveals an antiparallel type II-type IX molecular relationship and type IX to type IX bonding. J. Biol. Chem. 267, 23007–23014. 184. Vaughan, L., Winterhalter, K. H., and Bruckner, P. (1985). Proteoglycan Lt from chicken embryo sternum identified as type IX collagen. J. Biol. Chem. 260, 4758–4763. 185. Noro, A., Kimata, K., Oike, Y., Shinomura, T., Maeda, N., Yano, S., Takahashi, N., and Suzuki, S. (1983). Isolation and characterization of a third proteoglycan (PG-Lt) from chick embryo cartilage which contains disulfide-bonded collagenous polypeptide. J. Biol. Chem. 258, 9323–9331. 186. Huber, S., Winterhalter, K. H., and Vaughan, L. (1988). Isolation and sequence-analysis of the glycosaminoglycan attachment site of type-IX collagen. J. Biol. Chem. 263, 752–756. 187. Brewton, R. G., Wright, D. W., and Mayne, R. (1991). Structural and functional comparison of type IX collagen- proteoglycan from chicken cartilage and vitreous humor. J. Biol. Chem. 266, 4752–4757. 188. Nishimura, I., Muragaki,Y., and Olsen, B. R. (1989). Tissue-specific forms of type IX collagen-proteoglycan arise from the use of two widely separated promoters. J. Biol. Chem. 264, 20033–20041. 189. Muragaki, Y., Nishimura, I., Henney, A., Ninomiya, Y., and Olsen, B. R. (1990). The alpha 1(IX) collagen gene gives rise to two different transcripts in both mouse embryonic and human fetal RNA. Proc. Natl. Acad. Sci. USA 87, 2400–2404. 190. Fassler, R., Schnegelsberg, P. N., Dausman, J., Shinya, T., Muragaki, Y., McCarthy, M. T., Olsen, B. R., and Jaenisch, R. (1994). Mice lacking α1(IX) collagen develop non-inflammatory degenerative joint disease. Proc. Natl. Acad. Sci. USA 91, 5070–5074. 191. Watt, S. L., Lunstrum, G. P., McDonough, A. M., Keene, D. R., Burgeson, R. E., and Morris, N. P. (1992). Characterization of collagen types XII and XIV from fetal bovine cartilage. J. Biol. Chem. 267, 20093–20099. 192. Gordon, M. K., Gerecke, D. R., and Olsen, B. R. (1987). Type XII collagen: Distinct extracellular matrix component discovered by cRNA cloning. Proc. Natl. Acad. Sci. USA 84, 6040–6044. 193. Dublet, B., and van der Rest, M. (1991). Type XIV collagen, a new homotrimeric molecule extracted from bovine skin and tendon, with a triple helical disulfide-bonded domain homologous to type IX and type XII collagens. J. Biol. Chem. 266, 6853–6858. 194. Lunstrum, G. P., Morris, N. P., McDonough, A. M., Keene, D. R., and Burgeson, R. E. (1991). Identification and partial characterization of two type XII-like collagen molecules. J. Cell Biol. 113, 963–969. 195. Keene, D. R., Lunstrum, G. P., Morris, N. P., Stoddard, D. W., and Burgeson, R. E. (1991). Two type XII-like collagens localize to the surface of banded collagen fibrils. J. Cell Biol. 113, 971–978. 196. Lunstrum, G. P., McDonough, A. M., Marinkovich, M. P., Keene, D. R., Morris, N. P., and Burgeson, R. E. (1992). Identification and partial purification of a large, variant form of type XII collagen. J. Biol. Chem. 267, 20087–20092. 197. Gerecke, D. R., Olson, P. F., Koch, M., Knoll, J. H., Taylor, R., Hudson, D. L., Champliaud, M. F., Olsen, B. R., and Burgeson, R. E. (1997). Complete primary structure of two splice variants of collagen XII, and assignment of alpha 1(XII) collagen (COL12A1), alpha 1(IX) collagen (COL9A1), and alpha 1(XIX) collagen (COL19A1) to human chromosome 6q12- q13. Genomics 41, 236–242. 198. Aubert-Foucher, E., Font, B., Eichenberger, D., Goldschmidt, D., Lethias, C., and van der Rest, M. (1992). Purification and characterization of native type XIV collagen. J. Biol. Chem. 267, 15759–15764. 199. Schuppan, D., Cantaluppi, M. C., Becker, J., Veit, A., Bunte, T., Troyer, D., Schuppan, F., Schmid, M., Ackermann, R., and Hahn, E. G.
200.
201.
202.
203.
204.
205.
206.
207.
208.
209.
210.
211.
212.
213.
214.
215.
31
(1990). Undulin, an extracellular matrix glycoprotein associated with collagen fibrils. J. Biol. Chem. 265, 8823–8832. Walchli, C., Koch, M., Chiquet, M., Odermatt, B. F., and Trueb, B. (1994). Tissue-specific expression of the fibril-associated collagens XII and XIV. J. Cell Sci. 107, 669–681. Berthod, F., Germain, L., Guignard, R., Lethias, C., Garrone, R., Damour, O., van der Rest, M., and Auger, F. A. (1997). Differential expression of collagens XII and XIV in human skin and in reconstructed skin. J. Invest. Dermatol. 108, 737–742. Font, B., Aubert-Foucher, E., Goldschmidt, D., Eichenberger, D., and van der Rest, M. (1993). Binding of collagen XIV with the dermatan sulfate side chain of decorin. J. Biol. Chem. 268, 25015–25018. Ehnis, T., Dieterich, W., Bauer, M., Kresse, H., and Schuppan, D. (1997). Localization of a binding site for the proteoglycan decorin on collagen XIV (undulin). J. Biol. Chem. 272, 20414–20419. Kassner, A., Hansen, U., Miosge, N., Reinhardt, D. P., Aigner, T., Bruckner-Tuderman, L., Bruckner, P., and Grassel, S. (2003). Discrete integration of collagen XVI into tissue-specific collagen fibrils or beaded microfibrils. Matrix Biol. 22, 131–143. Kassner, A., Tiedemann, K., Notbohm, H., Ludwig, T., Morgelin, M., Reinhardt, D. P., Chu, M. L., Bruckner, P., and Grassel, S. (2004). Molecular structure and interaction of recombinant human type XVI collagen. J. Mol. Biol. 339, 835–853. Pan, T. C., Zhang, R. Z., Mattei, M. G., Timpl, R., and Chu, M. L. (1992). Cloning and chromosomal location of human alpha 1(XVI) collagen. Proc. Natl. Acad. Sci. USA 89, 6565–6569. Myers, J. C., Li, D., Amenta, P. S., Clark, C. C., Nagaswami, C., and Weisel, J. W. (2003). Type XIX collagen purified from human umbilical cord is characterized by multiple sharp kinks delineating collagenous subdomains and by intermolecular aggregates via globular, disulfidelinked, and heparin-binding amino termini. J. Biol. Chem. 278, 32047–32057. Sumiyoshi, H., Inoguchi, K., Khaleduzzaman, M., Ninomiya, Y., and Yoshioka, H. (1997). Ubiquitous expression of the alpha1(XIX) collagen gene (Col19a1) during mouse embryogenesis becomes restricted to a few tissues in the adult organism. J. Biol. Chem. 272, 17104–17111. Tuckwell, D. (2002). Identification and analysis of collagen alpha 1(XXI), a novel member of the FACIT collagen family. Matrix Biol. 21, 63–66. Timpl, R., and Engel, J. (1987). Type VI collagen. In: “Structure and Function of Collagen Types.” (R. Mayne and R. E. Burgeson, eds.), pp. 105–143, Academic Press. Chu, M. L., Pan, T. C., Conway, D., Kuo, H. J., Glanville, R. W., Timpl, R., Mann, K., and Deutzmann, R. (1989). Sequence analysis of alpha 1(VI) and alpha 2(VI) chains of human type VI collagen reveals internal triplication of globular domains similar to the A domains of von Willebrand factor and two alpha 2(VI) chain variants that differ in the carboxy terminus. EMBO J. 8, 1939–1946. Bonaldo, P., Russo, V., Bucciotti, F., Bressan, G. M., and Colombatti, A. (1989). Alpha 1 chain of chick type VI collagen. The complete cDNA sequence reveals a hybrid molecule made of one short collagen and three von Willebrand factor type A-like domains. J. Biol. Chem. 264, 5575–5580. Koller, E., Winterhalter, K. H., and Trueb, B. (1989). The globular domains of type VI collagen are related to the collagen-binding domains of cartilage matrix protein and von Willebrand factor. EMBO J. 8, 1073–1077. Ball, S., Bella, J., Kielty, C., and Shuttleworth, A. (2003). Structural basis of type VI collagen dimer formation. J. Biol. Chem. 278, 15326–15332. Doliana, R., Mucignat, M. T., Segat, D., Zanussi, S., Fabbro, C., Lakshmi, T. R., and Colombatti, A. (1998). Alternative splicing of VWFA modules generates variants of type VI collagen alpha 3 chain with a distinctive expression pattern in embryonic chicken tissues and potentially different adhesive function. Matrix Biol. 16, 427–442.
32 216. Saitta, B., Stokes, D. G., Vissing, H., Timpl, R., and Chu, M. L. (1990). Alternative splicing of the human alpha 2(VI) collagen gene generates multiple mRNA transcripts which predict three protein variants with distinct carboxyl termini. J. Biol. Chem. 265, 6473–6480. 217. Mayer, U., Poschl, E., Nischt, R., Specks, U., Pan, T. C., Chu, M. L., and Timpl, R. (1994). Recombinant expression and properties of the Kunitz-type protease-inhibitor module from human type VI collagen alpha 3(VI) chain. Eur. J. Biochem. 225, 573–580. 218. Heckmann, M., Aumailley, M., Hatamochi, A., Chu, M.-L., Timpl, R., and Krieg, T. (1989). Down-regulation of alpha3(VI) chain expression by gamma- interferon dereases synthesis and deposition of collagen type VI. Eur. J. Biochem. 182, 719–726. 219. Doane, K. J., Howell, S. J., and Birk, D. E. (1998). Identification and functional characterization of two type VI collagen receptors, alpha 3 beta 1 integrin and NG2, during avian corneal stromal development. Invest. Ophthalmol. Vis. Sci. 39, 263–275. 220. Kuo, H. J., Maslen, C. L., Keene, D. R., and Glanville, R. W. (1997). Type VI collagen anchors endothelial basement membranes by interacting with type IV collagen. J. Biol. Chem. 272, 26522–26529. 221. Furthmayr, H., Wiedemann, H., Timpl, R., Odermatt, E., and Engel, J. (1983). Electron-microscopical approach to a structural model of intima collagen. Biochem. J. 211, 303–311. 222. Bruns, R. R., Press, W., Engvall, E. T., and Gross, J. (1986). Type VI collagen in extracellular, 100-nm periodic filaments and fibrils: identification by immunoelectron microscopy. J. Cell Biol. 103, 393–404. 223. von der Mark, H., Aumailley, M., Wick, G., Fleischmajer, R., and Timpl, R. (1984). Immunochemistry, genuine size and tissue localization of collagen VI. Eur. J. Biochem. 142, 493–502. 224. Wiberg, C., Klatt, A. R., Wagener, R., Paulsson, M., Bateman, J. F., Heinegard, D., and Morgelin, M. (2003). Complexes of matrilin-1 and biglycan or decorin connect collagen VI microfibrils to both collagen II and aggrecan. J. Biol. Chem. 278, 37698–37704. 225. Wiberg, C., Heinegard, D., Wenglen, C., Timpl, R., and Morgelin, M. (2002). Biglycan organizes collagen VI into hexagonal-like networks resembling tissue structures. J. Biol. Chem. 277, 49120–49126. 226. Quarto, R., Dozin, B., Bonaldo, P., Cancedda, R., and Colombatti, A. (1993). Type VI collagen expression is upregulated in the early events of chondrocyte differentiation. Development 117, 245–251. 227. Swoboda, B., Pullig, O., Kirsch, T., Kladny, B., Steinhauser, B., and Weseloh, G. (1998). Increased content of type-VI collagen epitopes in human osteoarthritic cartilage: quantitation by inhibition ELISA. J. Orthop. Res. 16, 96–99. 228. Kielty, C. M., Cummings, C., Whittaker, S. P., Shuttleworth, C. A., and Grant, M. E. (1991). Isolation and ultrastructural analysis of microfibrillar structures from foetal bovine elastic tissues. Relative abundance and supramolecular architecture of type VI collagen assemblies and fibrillin. J. Cell. Sci. 99, 797–807. 229. Aumailley, M., Mann, K., von der Mark, H., and Timpl, R. (1989). Cell attachment properties of collagen type VI and Arg-Gly-Asp dependent binding to its alpha2(VI) and alpha3(VI) chains. Exp. Cell Res. 181, 463–474. 230. Gibson, G., Schor, S. L., and Grant, M. E. (1982). Effects of matrix molecules on chondrocyte gene expression. Synthesis of a low molecular weight collagen species by cells cultured within collagen gels. J. Cell Biol. 93, 767–774. 231. Capasso, O., Tajana, G., and Cancedda, R. (1984). Location of 64K collagen producing chondrocytes in developing chick embryo tibiae. Mol. Cell. Biol. 4, 1163–1168. 232. Schmid, T. M., and Linsenmayer, T. F. (1987). Type X collagen. In: “Structure and Function of Collagen types” (R. Mayne and R. E. Burgeson, eds.), Academic Press, pp. 223–259. 233. Aigner, T., Frischholz, S., Dertinger, S., Beier, F., Girkontaite, I., and von der Mark, K. (1997). Type X collagen expression and hypertrophic differentiation in chondrogenic neoplasias. Histochem. Cell Biol. 107, 435–440.
KLAUS VON DER MARK 234. Girkontaite, J., Frischholz, S., Lammi, P., Wagner, K., Swoboda, B., Aigner, T., and von der Mark, K. (1996). Immunolocalization of type X collagen in normal fetal and adult osteoarthritic cartilage with monoclonal antibodies. Matrix Biol. 15, 231–238. 235. Walker, G. D., Fischer, M., Gannon, J., Thompson, R. C., and Oegama, T. R. (1995). Expression of type X collagen in osteoarthritis. J. Orthopaedic Res. 13, 4–12. 236. Yamaguchi, N., Mayne, R., and Ninomiya, Y. (1991). The alpha 1 (VIII) collagen gene is homologous to the alpha 1 (X) collagen gene and contains a large exon encoding the entire triple helical and carboxyl-terminal non-triple helical domains of the alpha 1 (VIII) polypeptide. J. Biol. Chem. 266, 4508–4513. 237. Sawada, H., Konomi, H., and Hirosawa, K. (1990). Characterization of the collagen in the hexagonal lattice of Descemet’s membrane: its relation to type VIII collagen. J. Cell Biol. 110, 219–227. 238. Kwan, A. P. L., Cummings, C. E., Chapman, J. A., and Grant, M. E. (1991). Macromolecular organization of chicken type X collagen in vitro. J. Cell Biol. 114, 597–605. 239. Brass, A., Kadler, K. E., Thomas, J. T., Grant, M. E., and Boot-Handford, R. P. (1992). The fibrillar collagens, collagen VIII, collagen X and the C1q complement proteins share a similar domain in their C-terminal non-collagenous regions. FEBS Lett. 303, 126–128. 240. Marks, D. S., Gregory, C. A., Wallis, G. A., Brass, A., Kadler, K. E., and Boot-Handford, R. P. (1999). Metaphyseal chondrodysplasia type Schmid mutations are predicted to occur in two distinct threedimensional clusters within type X collagen NC1 domains that retain the ability to trimerize. J. Biol. Chem. 274, 3632–3641. 241. Bonaventure, J., Chaminade, F., and Maroteaux, P. (1995). Mutations in three subdomains of the carboxy-terminal region of collagen type X account for most of the Schmid metaphyseal dysplasias. Hum. Genet. 96, 58–64. 242. Chan, D., and Jacenko, O. (1998). Phenotypic and biochemical consequences of collagen X mutations in mice and humans. Matrix Biol. 17, 169–184. 243. Chan, D., Weng, Y. M., Golub, S., and Bateman, J. F. (1996). Type X collagen NC1 mutations produced by site-directed mutagenesis prevent in vitro assembly. Ann. Acad. Sci. N.Y. 785, 231–233. 244. Wallis, G. A., Rash, B., Sykes, B., Bonaventure, J., Maroteaux, P., Zabel, B., Wynne-Davies, R., Grant, M. E., and Boot-Handford, R. P. (1996). Mutations within the gene encoding the alpha 1 (X) chain of type X collagen (COL10A1) cause metaphyseal chondrodysplasia type Schmid but not several other forms of metaphyseal chondrodysplasia. J. Med. Genet. 33, 450–457. 245. Warman, M. L., Abbott, M., Apte, S. S., Hefferon, T., McIntosh, I., Cohn, D. H., Hecht, J. T., Olsen, B. R., and Francomano, C. A. (1993). A type X collagen mutation causes Schmid metaphyseal chondrodysplasia. Nat. Gen. 5, 79–82. 246. Dublet, B., Vernet, T., and van der Rest, M. (1999). Schmid’s metaphyseal chondrodysplasia mutations interfere with folding of the C-terminal domain of human collagen X expressed in Escherichia coli. J. Biol. Chem. 274, 18909–18915. 247. Jacenko, O., LuValle, P. A., and Olsen, B. R. (1993). Spondylometaphyseal dysplasia in mice carrying a dominant negative mutation in a matrix protein specific for cartilage to bone transition. Nature 365, 56–61. 248. Rosati, R., Horan, G. S., Pinero, G. J., Garofalo, S., Keene, D. R., Horton, W. A., Vuorio, E., deCrombrugghe, B., and Behringer, R. R. (1994). Normal long bone growth and development in type X collagennull mice. Nat. Gen. 8, 129–135. 249. Gress, C. J., and Jacenko, O. (2000). Growth plate compressions and altered hematopoiesis in collagen X null mice. J. Cell. Biol. 149, 983–993. 250. Hagg, P. M., Horelli-Kuitunen, N., Eklund, L., Palotie, A., and Pihlajaniemi, T. (1997). Cloning of mouse type XV collagen sequences and mapping of the corresponding gene to 4B1–3. Comparison of
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
251.
252.
253.
254.
255.
256.
257.
258.
259.
260.
261.
262.
263.
264.
265.
mouse and human alpha 1 (XV) collagen sequences indicates divergence in the number of small collagenous domains. Genomics 45, 31–41. Rehn, M., Hintikka, E., and Pihlajaniemi, T. (1996). Characterization of the mouse gene for the alpha 1 chain of type XVIII collagen (Col18a1) reveals that the three variant N- terminal polypeptide forms are transcribed from two widely separated promoters. Genomics 32, 436–446. Oh, S. P., Kamagata, Y., Muragaki, Y., Timmons, S., Ooshima, A., and Olsen, B. R. (1994). Isolation and sequencing of cDNAs for proteins with multiple domains of Gly-Xaa-Yaa repeats identify a distinct family of collagenous proteins. Proc. Natl. Acad. Sci. U.S.A. 91, 4229–4233. Van den Bergh, F., and Giudice, G. J. (2003). BP180 (type XVII collagen) and its role in cutaneous biology and disease. Adv. Dermatol. 19, 37–71. Ricard-Blum, S., Dublet, B. and van der Rest, M. (2000). Unconventional Collagens Types VI, VII, VIII, IX, X, XIV, XVI and XIX (S. RicardBlum, B. Dublet and M. van der Rest, eds.), Oxford University Press, p. 155. Tu, H., Sasaki, T., Snellman, A., Gohring, W., Pirila, P., Timpl, R., and Pihlajaniemi, T. (2002). The type XIII collagen ectodomain is a 150-nm rod and capable of binding to fibronectin, nidogen-2, perlecan, and heparin. J. Biol. Chem. 277, 23092–23099. Nykvist, P., Tu, H., Ivaska, J., Kapyla, J., Pihlajaniemi, T., and Heino, J. (2000). Distinct recognition of collagen subtypes by alpha(1)beta(1) and alpha(2)beta(1) integrins. Alpha(1)beta(1) mediates cell adhesion to type XIII collagen. J. Biol. Chem. 275, 8255–8261. Juvonen, M., and Pihlajaniemi, T. (1992) Characterization of the spectrum of alternative splicing of alpha 1 (XIII) collagen transcripts in HT-1080 cells and calvarial tissue resulted in identification of two previously unidentified alternatively spliced sequences, one previously unidentified exon, and nine new mRNA variants. J. Biol. Chem. 267, 24693–24699. Rehn, M., and Pihlajaniemi, T. (1994). Alpha 1(XVIII), a collagen chain with frequent interruptions in the collagenous sequence, a distinct tissue distribution, and homology with type XV collagen. Proc. Natl. Acad. Sci. U.S.A. 91, 4234–4238. Rehn, M., Hintikka, E., and Pihlajaniemi, T. (1994). Primary structure of the alpha 1 chain of mouse type XVIII collagen, partial structure of the corresponding gene, and comparison of the alpha 1(XVIII) chain with its homologue, the alpha 1(XV) collagen chain. J. Biol. Chem. 269, 13929–13935. Pufe, T., Petersen, W. J., Miosge, N., Goldring, M. B., Mentlein, R., Varoga, D. J., and Tillmann, B. N. (2004). Endostatin/collagen XVIIIan inhibitor of angiogenesis-is expressed in cartilage and fibrocartilage. Matrix Biol. 23, 267–276. Sasaki, T., Fukai, N., Mann, K., Gohring, W., Olsen, B. R., and Timpl, R. (1998). Structure, function and tissue forms of the C-terminal globular domain of collagen XVIII containing the angiogenesis inhibitor endostatin. EMBO J. 17, 4249–4256. Olsen, B. R. (1981). Collagen Biosynthesis. In “Cell Biology of Extracellular Matrix” (E. D. Hay, ed), pp. 139–177, Plenum Press, New York. Ryan, M. C., Sieraski, M., and Sandell, L. J. (1990). The human type II procollagen gene: identification of an additional protein-coding domain and location of potential regulatory sequences in the promoter and first intron. Genomics 8, 41–48. Peltonen, S., Rehn, M., and Pihlajaniemi, T. (1997). Alternative splicing of mouse alpha1(XIII) collagen RNAs results in at least 17 different transcripts, predicting alpha1(XIII) collagen chains with length varying between 651 and 710 amino acid residues. DNA Cell Biol. 16, 227–234. Chu, M. L., de Wet, W., Bernard, M., and Ramirez, F. (1985). Fine structural analysis of the human pro-alpha 1 (I) collagen gene.
266.
267.
268.
269. 270.
271.
272.
273.
274.
275.
276.
277. 278.
279.
280.
281.
282.
33
Promoter structure, AluI repeats, and polymorphic transcripts. J. Biol. Chem. 260, 2315–2320. Focht, R. J., and Adams, S. L. (1984). Tissue specificity of type I collagen gene expression is determined at both transcriptional and post-transcriptional levels. Mol. Cell. Biol. 4, 1843–1852. Mauch, C., Hatamochi, A., Scharffetter, K., and Krieg, T. (1988). Regulation of collagen synthesis in fibroblasts within a three- dimensional collagen gel. Exp. Cell Res. 178, 493–503. Eckes, B., Mauch, C., Huppe, G., and Krieg, T. (1993). Downregulation of collagen synthesis in fibroblasts within three- dimensional collagen lattices involves transcriptional and posttranscriptional mechanisms. FEBS Lett. 318, 129–133. Kivirikko, K. I., and Myllyharju, J. (1998). Prolyl 4-hydroxylases and their protein disulfide isomerase subunit. Matrix Biol. 16, 357–368. Kivirikko, K. I., and Myllyla, R. (1982). Posttranslational enzymes in the biosynthesis of collagen: intracellular enzymes. Methods Enzymol 82, 245–304. Kivirikko, K. I., and Pihlajaniemi, T. (1998). Collagen hydroxylases and the protein disulfide isomerase subunit of prolyl 4-hydroxylases. Adv. Enzymol. Relat. Areas Mol. Biol. 72, 325–398. Vranka, J. A., Sakai, L. Y., and Bachinger, H. P. (2004). Prolyl 3-hydroxylase 1, enzyme characterization and identification of a novel family of enzymes. J. Biol. Chem. 279, 23615–23621. Tryggvason, K., Majamaa, K., Risteli, J., and Kivirikko, K. I. (1979). Partial purification and characterization of chick-embryo prolyl 3-hydroxylase. Biochem. J. 183, 303–307. Helaakoski, T., Veijola, J., Vuori, K., Rehn, M., Chow, L. T., TaillonMiller, P., Kivirikko, K. I., and Pihlajaniemi, T. (1994). Structure and expression of the human gene for the alpha subunit of prolyl 4-hydroxylase. The two alternatively spliced types of mRNA correspond to two homologous exons the sequences of which are expressed in a variety of tissues. J. Biol. Chem. 269, 27847–27854. Annunen, P., Helaakoski, T., Myllyharju, J., Veijola, J., Pihlajaniemi, T., and Kivirikko, K. I. (1997). Cloning of the human prolyl 4-hydroxylase alpha subunit isoform alpha(II) and characterization of the type II enzyme tetramer. The alpha(I) and alpha(II) subunits do not form a mixed alpha(I)alpha(II)beta2 tetramer. J. Biol. Chem. 272, 17342–17348. Chvapil, M., and Ryan, J. N. (1972). Diverse effect of 1,10-phenanthroline and 2,2′-dipyridyl on the synthesis of collagen and noncollagenous proteins in in vivo and in vitro systems. Biochim. Biophys. Acta 273, 208–211. Bienkowski, R. S. (1989). Intracellular degradation of newly synthesized collagen. Revis. Biol. Celular. 21, 423–443. Hanauske-Abel, H. M. (1991). Prolyl 4-hydroxylase, a target enzyme for drug development. Design of suppressive agents and the in vitro effects of inhibitors and proinhibitors. J. Hepatol. 13 Suppl 3:S8–15; discussion S16. Tschank, G., Brocks, D. G., Engelbart, K., Mohr, J., Baader, E., Gunzler, V., and Hanauske-Abel, H. M. (1991). Inhibition of prolyl hydroxylation and procollagen processing in chick-embryo calvaria by a derivative of pyridine-2,4- dicarboxylate. Characterization of the diethyl ester as a proinhibitor. Biochem. J. 275, 469–476. Myllyla, R., Pihlajaniemi, T., Pajunen, L., Turpeenniemi-Hujanen, T., and Kivirikko, K. I. (1991). Molecular cloning of chick lysyl hydroxylase. Little homology in primary structure to the two types of subunit of prolyl 4- hydroxylase. J. Biol. Chem. 266, 2805–2810. Rautavuoma, K., Takaluoma, K., Passoja, K., Pirskanen, A., Kvist, A. P., Kivirikko, K. I., and Myllyharju, J. (2002). Characterization of three fragments that constitute the monomers of the human lysyl hydroxylase isoenzymes 1–3. The 30-kDa N-terminal fragment is not required for lysyl hydroxylase activity. J. Biol. Chem. 277, 23084–23091. Smith-Mungo, L. I., and Kagan, H. M. (1998). Lysyl oxidase: properties, regulation and multiple functions in biology. Matrix Biol. 16, 387–398.
34 283. Knott, L., and Bailey, A. J. (1998). Collagen cross-links in mineralizing tissues: a review of their chemistry, function, and clinical relevance. Bone 22, 181–187. 284. Blumenkrantz, N., Assad, R., and Peterkofsky, B. (1984). Characterization of collagen hydroxylysyl glycosyltransferases as mainly intramembranous microsomal enzymes. J. Biol. Chem. 259, 854–859. 285. Clark, C. C. (1979). The distribution and initial characterization of oligosaccharide units on the COOH-terminal propeptide extensions of the pro-alpha 1 and pro-alpha 2 chains of type I procollagen. J. Biol. Chem. 254, 10798–10802. 286. Doege, K. J., and Fessler, J. H. (1986). Folding of carboxyl domain and assembly of procollagen I. J. Biol. Chem. 261, 8924–8935. 287. Freedman, R. B., Hirst, T. R., and Tuite, M. F. (1994). Protein disulphide isomerase: building bridges in protein folding. Trends. Biochem. Sci. 19, 331–336. 288. Lang, K., Schmid, F. X., and Fischer, G. (1987). Catalysis of protein folding by prolyl isomerase. Nature 329, 268–270. 289. Bulleid, N. J., Dalley, J. A., and Lees, J. F. (1997). The C-propeptide domain of procollagen can be replaced with a transmembrane domain without affecting trimer formation or collagen triple helix folding during biosynthesis. EMBO J. 16, 6694–6701. 290. Steinmann, B., Bruckner, P., and Superti-Furga, A. (1991). Cyclosporin A slows collagen triple-helix formation in vivo: indirect evidence for a physiologic role of peptidyl-prolyl cis-trans-isomerase. J. Biol. Chem. 266, 1299–1303. 291. Nakai, A., Satoh, M., Hirayoshi, K., and Nagata, K. (1992). Involvement of the stress protein HSP47 in procollagen processing in the endoplasmic reticulum. J. Cell Biol. 117, 903–914. 292. Nagata, K., and Hosokawa, N. (1996). Regulation and function of collagen-specific molecular chaperone, HSP47. Cell Struct. Funct. 21, 425–430. 293. Nagata, K. (1996). Hsp47: a collagen-specific molecular chaperone. Trends. Biochem. Sci. 21, 22–26. 294. Natsume, T., Koide, T., Yokota, S., Hirayoshi, K., and Nagata, K. (1994). Interactions between collagen-binding stress protein HSP47 and collagen. Analysis of kinetic parameters by surface plasmon resonance biosensor. J. Biol. Chem. 269, 31224–31228. 295. Hattori, T., Von Der Mark, K., Kawaki, H., Yutani, Y., Kubota, S., Nakanishi, T., Eberspaecher, H., De Crombrugghe, B., and Takigawa, M. (2004). Downregulation of rheumatoid arthritis-related antigen RAA47 (HSP47/colligin-2) in chondrocytic cell lines induces apoptosis and cell-surface expression of RA-A47 in association with CD9. J. Cell Physiol. 30, 30. 296. Rawlings, N. D., and Barrett, A. J. (1995). Evolutionary families of metallopeptidases. Methods Enzymol. 248, 183–228. 297. Hojima, Y., McKenzie, J. A., van der Rest, M., and Prockop, D. J. (1989). Type I procollagen N-proteinase from chick embryo tendons. Purification of a new 500-kDa form of the enzyme and identification of the catalytically active polypeptides. J. Biol. Chem. 264, 11336–11345. 298. Colige, A., Li, S. W., Sieron, A. L., Nusgens, B. V., Prockop, D. J., and Lapiere, C. M. (1997). cDNA cloning and expression of bovine procollagen I N-proteinase: a new member of the superfamily of zincmetalloproteinases with binding sites for cells and other matrix components. Proc. Natl. Acad. Sci. U.S.A. 94, 2374–2379. 299. Tuderman, L., and Prockop, D. J. (1982). Procollagen N-proteinase. Properties of the enzyme purified from chick embryo tendons. Eur. J. Biochem. 125, 545–549. 300. Li, S. W., Sieron, A. L., Fertala, A., Hojima, Y., Arnold, W. V., and Prockop, D. J. (1996). The C-proteinase that processes procollagens to fibrillar collagens is identical to the protein previously identified as bone morphogenic protein-1. Proc. Natl. Acad. Sci. U.S.A. 93, 5127–5130. 301. Suzuki, N., Labosky, P. A., Furuta, Y., Hargett, L., Dunn, R., Fogo, A. B., Takahara, K., Peters, D. M., Greenspan, D. S., and Hogan, B. L. (1996).
KLAUS VON DER MARK
302.
303.
304.
305.
306.
307.
308.
309.
310.
311.
312.
313.
314.
315.
316.
317.
318.
319.
Failure of ventral body wall closure in mouse embryos lacking a procollagen C-proteinase encoded by Bmp1, a mammalian gene related to Drosophila tolloid. Development 122, 3587–3595. Helseth, D. L., Jr., and Veis, A. (1984). Cathepsin D-mediated processing of procollagen: lysosomal enzyme involvement in secretory processing of procollagen. Proc. Natl. Acad. Sci. U.S.A. 81, 3302–3306. Hamalainen, E. R., Kemppainen, R., Pihlajaniemi, T., and Kivirikko, K. I. (1993). Structure of the human lysyl oxidase gene. Genomics 17, 544–548. Ramirez, F., and diLiberto, M. (1990). Complex and diversified regulatory programs control the expression of vertebrate collagen genes. FASEB J. 4, 1616–1623. Finer, M. H., Gerstenfeld, L. C., Young, D., Doty, P., and Boedtker, H. (1985). Collagen expression in embryonic chicken chondrocytes treated with phorbol myristate acetate. Mol. Cell Biol. 5, 1415–1424. Bennett, V. D., and Adams, S. L. (1987). Characterization of the translational control mechanism preventing synthesis of alpha 2(I) collagen in chicken vertebral chondroblasts. J. Biol. Chem. 262, 14806–14814. Bennett, V. D., and Adams, S. L. (1990). Identification of a cartilagespecific promoter within intron-2 of the chick alpha-2(i) collagen gene. J. Biol. Chem. 265, 2223–2230. Penttinen, R. P., Kobayashi, S., and Bornstein, P. (1988). Transforming growth factor beta increases mRNA for matrix proteins both in the presence and in the absence of changes in mRNA stability. Proc. Natl. Acad. Sci. U.S.A. 85, 1105–1108. Speth, C., and Oberbaumer, I. (1993). Expression of basement membrane proteins: evidence for complex post-transcriptional control mechanisms. Exp. Cell Res. 204, 302–310. Rossi, P., and de Crombrugghe, B. (1987). Formation of a type I collagen RNA dimer by intermolecular base-pairing of a conserved sequence around the translation initiation site. Nucleic Acids Res. 15, 8935–8956. Bornstein, P., McKay, J., Devarayalu, S., and Cook, S. C. (1988). A highly conserved, 5′ untranslated, inverted repeat sequence is ineffective in translational control of the alpha 1(I) collagen gene [published erratum appears in Nucleic Acids Res 1988 Dec 9; 16(23):11399]. Nucleic. Acids. Res. 16, 9721–9736. Paglia, L., Wilczek, J., de Leon, L. D., Martin, G. R., Horlein, D., and Muller, P. (1979). Inhibition of procollagen cell-free synthesis by amino-terminal extension peptides. Biochemistry 18, 5030–5034. Paglia, L. M., Wiestner, M., Duchene, M., Ouellette, L. A., Horlein, D., Martin, G. R., and Muller, P. K. (1981). Effects of procollagen peptides on the translation of type II collagen messenger ribonucleic acid and on collagen biosynthesis in chondrocytes. Biochemistry 20, 3523–3527. Horlein, D., McPherson, J., Goh, S. H., and Bornstein, P. (1981). Regulation of protein synthesis: translational control by procollagenderived fragments. Proc. Natl. Acad. Sci. U.S.A. 78, 6163–6167. Schlumberger, W., Thie, M., Volmer, H., Rauterberg, J., and Robenek, H. (1988). Binding and uptake of Col 1(I), a peptide capable of inhibiting collagen synthesis in fibroblasts. Eur. J. Cell Biol. 46, 244–252. Fouser, L., Sage, E. H., Clark, J., and Bornstein, P. (1991). Feedback regulation of collagen gene expression: a Trojan horse approach. Proc. Natl. Acad. Sci. U.S.A. 88, 10158–10162. Wu, C. H., Donovan, C. B., and Wu, G. Y. (1986). Evidence for pretranslational regulation of collagen synthesis by procollagen propeptides. J. Biol. Chem. 261, 10482–10484. Aycock, R. S., Raghow, R., Stricklin, G. P., Seyer, J. M., and Kang, A. H. (1986). Post-transcriptional inhibition of collagen and fibronectin synthesis by a synthetic homolog of a portion of the carboxyl- terminal propeptide of human type I collagen. J. Biol. Chem. 261, 14355–14360. Mizuno, M., Fujisawa, R., and Kuboki, Y. (2000). The effect of carboxyl-terminal propeptide of type I collagen (c-propeptide) on
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone
320.
321. 322.
323.
324.
325.
326. 327.
328.
329.
330.
331.
332.
333.
334.
335.
collagen synthesis of preosteoblasts and osteoblasts. Calcif. Tissue Int. 67, 391–399. Nakata, K., Miyamoto, S., Bernier, S., Tanaka, M., Utani, A., Krebsbach, P., Rhodes, C., and Yamada, Y. (1996). The c-propeptide of type II procollagen binds to the enhancer region of the type II procollagen gene and regulates its transcription. Ann. Acad. Sci. N.Y. 785, 307–308. Rossert, J. A., and Garrett, L. A. (1995). Regulation of type I collagen synthesis. Kidney Int. Suppl. 49, S34–38. Bornstein, P. (1996). Regulation of expression of the alpha 1 (I) collagen gene: a critical appraisal of the role of the first intron. Matrix Biol. 15, 3–10. Frischauf, A. M., Lehrach, H., Rosner, C., and Boedtker, H. (1978). Procollagen complementary DNA, a probe for messenger RNA purification and the number of type I collagen genes. Biochemistry 17, 3243–3249. Boedtker, H., Frischauf, A. M., and Lehrach, H. (1976). Isolation and translation of calvaria procollagen messenger ribonucleic acids. Biochemistry 15, 4765–4770. Yamamoto, T., Sobel, M. E., Adams, S. L., Avvedimento, V. E., DiLauro, R., Pastan, I., de Crombrugghe, B., Showalter, A., Pesciotta, D., Fietzek, P., and Olsen, B. (1980). Construction of a recombinant bacterial plasmid containing pro-alpha 1(I) collagen DNA sequences. J. Biol. Chem. 255, 2612–2615. Vuorio, E., and de Crombrugghe, B. (1990). The family of collagen genes. Annu. Rev. Biochem. 59, 837–872. Chu, M. L., de Wet, W., Bernard, M., Ding, J. F., Morabito, M., Myers, J., Williams, C., and Ramirez, F. (1984). Human pro alpha 1(I) collagen gene structure reveals evolutionary conservation of a pattern of introns and exons. Nature 310, 337–340. Dewet, W., Bernard, M., Bensochanda, V., Chu, M. L., Dickson, L., Weil, D., and Ramirez, F. (1987). Organization of the human proalpha 2(I) collagen gene. J. Biol. Chem. 262, 16032–16036. Chuang, P. T., and McMahon, A. P. (1999). Vertebrate Hedgehog signalling modulated by induction of a Hedgehog-binding protein. Nature 397, 617–621. Sokolov, B. P., Mays, P. K., Khillan, J. S., and Prockop, D. J. (1993). Tissue- and development-specific expression in transgenic mice of a type I procollagen (COL1A1) minigene construct with 2.3 kb of the promoter region and 2 kb of the 3′-flanking region. Specificity is independent of the putative regulatory sequences in the first intron. Biochemistry 32, 9242–9249. Sokolov, B. P., Ala-Kokko, L., Dhulipala, R., Arita, M., Khillan, J. S., and Prockop, D. J. (1995). Tissue-specific expression of the gene for type I procollagen (COL1A1) in transgenic mice. Only 476 base pairs of the promoter are required if collagen genes are used as reporters. J. Biol. Chem. 270, 9622–9629. Rossert, J. A., Chen, S. S., Eberspaecher, H., Smith, C. N., and de Crombrugghe, B. (1996). Identification of a minimal sequence of the mouse pro-alpha 1(I) collagen promoter that confers high-level osteoblast expression in transgenic mice and that binds a protein selectively present in osteoblasts. Proc. Natl. Acad. Sci. U.S.A. 93, 1027–1031. Rossert, J., Eberspaecher, H., and de Crombrugghe, B. (1995). Separate cis-acting DNA elements of the mouse pro-alpha 1(I) collagen promoter direct expression of reporter genes to different type I collagen-producing cells in transgenic mice. J. Cell Biol. 129, 1421–1432. Bornstein, P., and McKay, J. (1988). The first intron of the alpha 1(I) collagen gene contains several transcriptional regulatory elements. J. Biol. Chem. 263, 1603–1606. Slack, J. L., Liska, D. J., and Bornstein, P. (1991). An upstream regulatory region mediates high-level, tissue- specific expression of the human alpha 1(I) collagen gene in transgenic mice. Mol. Cell Biol. 11, 2066–2074.
35
336. Liska, D. J., Robinson, V. R., and Bornstein, P. (1992). Elements in the first intron of the alpha 1(I) collagen gene interact with Sp1 to regulate gene expression. Gene Expr. 2, 379–389. 337. Kern, B., Shen, J., Starbuck, M., and Karsenty, G. (2001). Cbfa1 contributes to the osteoblast-specific expression of type I collagen genes. J. Biol. Chem. 276, 7101–7107. 338. Pavlin, D., Bedalov, A., Kronenberg, M. S., Kream, B. E., Rowe, D. W., Smith, C. L., Pike, J. W., and Lichtler, A. C. (1994). Analysis of regulatory regions in the COL1A1 gene responsible for 1,25- dihydroxyvitamin D3-mediated transcriptional repression in osteoblastic cells. J. Cell Biochem. 56, 490–501. 339. Meisler, N., Keefer, K. A., Ehrlich, H. P., Yager, D. R., Myers-Parrelli, J., and Cutroneo, K. R. (1997). Dexamethasone abrogates the fibrogenic effect of transforming growth factor-beta in rat granuloma and granulation tissue fibroblasts. J. Invest. Dermatol. 108, 285–289. 340. Meisler, N., Shull, S., Xie, R., Long, G. L., Absher, M., Connolly, J. P., and Cutroneo, K. R. (1995). Glucocorticoids coordinately regulate type I collagen pro alpha 1 promoter activity through both the glucocorticoid and transforming growth factor beta response elements: a novel mechanism of glucocorticoid regulation of eukaryotic genes. J. Cell Biochem. 59, 376–388. 341. Rossi, P., and deCrombrugghe, B. (1987). Identification of a cellspecific transcriptional enhancer in the first intron of the mouse alpha 2 (type I) collagen gene. Proc. Natl. Acad. Sci. USA 84, 5590–5594. 342. Ignotz, R. A., Endo, T., and Massague, J. (1987). Regulation of fibronectin and type I collagen mRNA levels by transforming growth factor-beta. J. Biol. Chem. 262, 6443–6446. 343. Ritzenthaler, J. D., Goldstein, R. H., Fine, A., and Smith, B. D. (1993). Regulation of the alpha 1(I) collagen promoter via a transforming growth factor-beta activation element. J. Biol. Chem. 268, 13625–13631. 344. D’Souza, R. N., Niederreither, K., and de Crombrugghe, B. (1993). Osteoblast-specific expression of the alpha 2(I) collagen promoter in transgenic mice: correlation with the distribution of TGF-beta 1. J. Bone Miner. Res. 8, 1127–1136. 345. Chung, K. Y., Agarwal, A., Uitto, J., and Mauviel, A. (1996). An AP-1 binding sequence is essential for regulation of the human alpha2(I) collagen (COL1A2) promoter activity by transforming growth factorbeta. J. Biol. Chem. 271, 3272–3278. 346. Greenwel, P., Inagaki, Y., Hu, W., Walsh, M., and Ramirez, F. (1997). Sp1 is required for the early response of alpha2(I) collagen to transforming growth factor-beta1. J. Biol. Chem. 272, 19738–19745. 347. Higashi, K., Kouba, D. J., Song, Y. J., Uitto, J., and Mauviel, A. (1998). A proximal element within the human alpha 2(I) collagen (COL1A2) promoter, distinct from the tumor necrosis factor-alpha response element, mediates transcriptional repression by interferongamma. Matrix Biol. 16, 447–456. 348. Bou-Gharios, G., Garrett, L. A., Rossert, J., Niederreither, K., Eberspaecher, H., Smith, C., Black, C., and Crombrugghe, B. (1996). A potent far-upstream enhancer in the mouse pro alpha 2(I) collagen gene regulates expression of reporter genes in transgenic mice. J. Cell Biol. 134, 1333–1344. 349. Horton, W., Miyashita, T., Kohno, K., Hassell, J. R., and Yamada, Y. (1987). Identification of a phenotype-specific enhancer in the first intron of the rat collagen II gene. Proc. Natl. Acad. Sci. USA 84, 8864–8868. 350. Zhou, G., Garofalo, S., Mukhopadhyay, K., Lefebvre, V., Smith, C. N., Eberspaecher, H., and de Crombrugghe, B. (1995). A 182 bp fragment of the mouse pro alpha 1(II) collagen gene is sufficient to direct chondrocyte expression in transgenic mice. J. Cell Sci. 108, 3677–3684. 351. Mukhopadhyay, K., Lefebvre, V., Zhou, G., Garofalo, S., Kimura, J. H., and de Crombrugghe, B. (1995). Use of a new rat chondrosarcoma cell line to delineate a 119- base pair chondrocyte-specific enhancer
36
352.
353.
354.
355.
356.
357.
358.
359. 360.
361.
362.
363.
364.
365.
366.
367.
KLAUS VON DER MARK element and to define active promoter segments in the mouse proalpha 1(II) collagen gene. J. Biol. Chem. 270, 27711–27719. Zhou, G., Lefebvre, V., Zhang, Z., Eberspaecher, H., and de Crombrugghe, B. (1998). Three high mobility group-like sequences within a 48-base pair enhancer of the Col2a1 gene are required for cartilage-specific expression in vivo. J. Biol. Chem. 273, 14989–14997. Lefebvre, V., Zhou, G., Mukhopadhyay, K., Smith, C. N., Zhang, Z., Eberspaecher, H., Zhou, X., Sinha, S., Maity, S. N., and de Crombrugghe, B. (1996). An 18-base-pair sequence in the mouse proalpha1(II) collagen gene is sufficient for expression in cartilage and binds nuclear proteins that are selectively expressed in chondrocytes. Mol. Cell Biol. 16, 4512–4523. Lefebvre, V., and de Crombrugghe, B. (1998). Toward understanding SOX9 function in chondrocyte differentiation. Matrix Biol. 16, 529–540. Savagner, P., Krebsbach, P. H., Hatano, O., Miyashita, T., Liebman, J., and Yamada, Y. (1995). Collagen II promoter and enhancer interact synergistically through Sp1 and distinct nuclear factors. DNA Cell Biol. 14, 501–510. Ruteshouser, E. C., and de Crombrugghe, B. (1992). Purification of BBF, a DNA-binding protein recognizing a positive cis-acting element in the mouse alpha 1(III) collagen promoter. J. Biol. Chem. 267, 14398–14404. Mudryj, M., and de Crombrugghe, B. (1988). Deletion analysis of the mouse alpha 1(III) collagen promoter. Nucleic. Acids. Res. 16, 7513–7526. Takahara, K., Hoffman, G. G., and Greenspan, D. S. (1995). Complete structural organization of the human alpha 1 (V) collagen gene (COL5A1): divergence from the conserved organization of other characterized fibrillar collagen genes. Genomics 29, 588–597. Lee, S., and Greenspan, D. S. (1995). Transcriptional promoter of the human alpha 1(V) collagen gene (COL5A1). Biochem. J. 310, 15–22. Truter, S., Di Liberto, M., Inagaki, Y., and Ramirez, F. (1992). Identification of an upstream regulatory region essential for cell typespecific transcription of the pro-alpha 2(V) collagen gene (COL5A2). J. Biol. Chem. 267, 25389–25395. Nagato, H., Matsuo, N., Sumiyoshi, H., Sakata-Takatani, K., Nasu, M., Yoshioka, H. (2004). The transcription factor CCAAT-binding -factor CBF/NF-Y and two repressors regulate the core promoter of the human Proa3(V) collagen gene(COL5A3). J. Biol. Chem. 279, in press. Yoshioka, H., Greenwel, P., Inoguchi, K., Truter, S., Inagaki, Y., Ninomiya, Y., and Ramirez, F. (1995). Structural and functional analysis of the promoter of the human alpha 1(XI) collagen gene. J. Biol. Chem. 270, 418–424. Yoshioka, H., Inoguchi, K., Khaleduzzaman, M., Ninomiya, Y., Andrikopoulos, K., and Ramirez, F. (1995). Coding sequence and alternative splicing of the mouse alpha 1(XI) collagen gene (Col11a1). Genomics 28, 337–340. Lui, V. C., Ng, L. J., Sat, E. W., and Cheah, K. S. (1996). The human alpha 2(XI) collagen gene (COL11A2): completion of coding information, identification of the promoter sequence, and precise localization within the major histocompatibility complex reveal overlap with the KE5 gene. Genomics 32, 401–412. Bridgewater, L. C., Lefebvre, V., and de Crombrugghe, B. (1998). Chondrocyte-specific enhancer elements in the Col11a2 gene resemble the Col2a1 tissue-specific enhancer. J. Biol. Chem. 273, 14998–15006. Hayman, A. R., Koppel, J., and Trueb, B. (1991). Complete structure of the chicken alpha 2(VI) collagen gene. Eur. J. Biochem. 197, 177–184. Trikka, D., Davis, T., Lapenta, V., Brahe, C., and Kessling, A. M. (1997). Human COL6A1: genomic characterization of the globular domains, structural and evolutionary comparison with COL6A2. Mamm. Genome. 8, 342–345.
368. Walchli, C., Marcionelli, R., Odermatt, B. F., Peltonen, J., Vuorio, E., and Trueb, B. (1996). Expression and distribution of two alternatively spliced transcripts from the chicken alpha 2 (VI) collagen gene. J. Cell Biochem. 63, 207–220. 369. Doliana, R., Bonaldo, P., and Colombatti, A. (1990). Multiple forms of chicken alpha 3(VI) collagen chain generated by alternative splicing in type A repeated domains. J. Cell Biol. 111, 2197–2205. 370. Braghetta, P., Fabbro, C., Piccolo, S., Marvulli, D., Bonaldo, P., Volpin, D., and Bressan, G. M. (1996). Distinct regions control transcriptional activation of the alpha1(VI) collagen promoter in different tissues of transgenic mice. J. Cell Biol. 135, 1163–1177. 371. Piccolo, S., Bonaldo, P., Vitale, P., Volpin, D., and Bressan, G. M. (1995). Transcriptional activation of the alpha 1(VI) collagen gene during myoblast differentiation is mediated by multiple GA boxes. J. Biol. Chem. 270, 19583–19590. 372. Fabbro, C., Braghetta, P., Girotto, D., Piccolo, S., Volpin, D., and Bressan, G. M. (1999). Cell type-specific transcription of the alpha1(VI) collagen gene. Role of the ap1 binding site and of the core promoter. J. Biol. Chem. 274, 1759–1766. 373. Saitta, B., and Chu, M. L. (1995). Characterization of the human alpha 1(VI) collagen promoter and its comparison with human alpha 2(VI) promoters. Eur. J. Biochem. 234, 542–549. 374. Braghetta, P., Vitale, P., Piccolo, S., Bonaldo, P., Fabbro, C., Girotto, D., Volpin, D., and Bressan, G. M. (1997). Tissue-specific expression of promoter regions of the alpha1(VI) collagen gene in cell cultures and transgenic mice. Eur. J. Biochem. 247, 200–208. 375. Willimann, T. E., Maier, R., and Trueb, B. (1995). A novel transcription factor and two members of the Sp 1 multigene family regulate the activity of the alpha 2 (VI) collagen promoter. Matrix Biol. 14, 653–663. 376. Hatamochi, A., Aumailley, M., Mauch, C., Chu, M.-L., Timpl, R., and Krieg, T. (1989). Regulation of collagen VI expression in fibroblasts. Effects of cell density, cell-matrix interactions, and chemical transformation. J. Biol. Chem. 264, 3494–3499. 377. Peltonen, J., Hsiao, L. L., Jaakkola, S., Sollberg, S., Aumailley, M., Timpl, R., Chu, M. L., and Uitto, J. (1991). Activation of collagen gene expression in keloids: co- localization of type I and VI collagen and transforming growth factor-beta 1 mRNA. J. Invest. Dermatol. 97, 240–248. 378. Harumiya, S., Gibson, M. A., and Koshihara, Y. (2002). Antisense suppression of collagen VI synthesis results in reduced expression of collagen I in normal human osteoblast-like cells. Biosci. Biotechnol. Biochem. 66, 2743–2747. 379. Heckmann, M., Aumailley, M., Chu, M. L., Timpl, R., and Krieg, T. (1992). Effect of transforming growth factor-beta on collagen type VI expression in human dermal fibroblasts. FEBS Lett. 310, 79–82. 380. Luvalle, P., Ninomiya, Y., and Rosenblum, N. D. (1988). The type X collagen gene. J. Biol. Chem. 263, 18378–18385. 381. Elima, K., Eerola, I., Rosati, R., Metsäranta, M., Garofalo, S., Perälä, M., deCrombrugghe, B., and Vuorio, E. (1993). The mouse collagen X gene: complete nucleotide sequence, exon structure and expression pattern. Biochem. J. 289, 247–253. 382. Kong, R. Y., Kwan, K. M., Lau, E. T., Thomas, J. T., BootHandford, R. P., Grant, M. E., and Cheah, K. S. (1993). Intron-exon structure, alternative use of promoter and expression of the mouse collagen X gene, COL10A-1. Eur. J. Biochem. 213, 99–111. 383. Apte, S. S., and Olsen, B. R. (1993). Characterization of the mouse type X collagen gene. Matrix 13, 165–179. 384. Thomas, J. T., Kwan, A. P. L., Grant, M. E., and Boot-Handford, R. P. (1991). Isolation of cDNAs encoding the complete sequence of bovine type X collagen. Biochem. J. 273, 141–148. 385. Thomas, J. T., Cresswell, C. J., Rash, B., Nicolai, H., Jones, T., Solomon, E., Grant, M. E., and Boot-Handford, R. P. (1991). The human collagen X gene. Biochem. J. 280, 617–623. 386. Reichenberger, E., Beier, F., Luvalle, P., Olsen, B. R., von der Mark, K., and Bertling, W. (1992). Genomic organisation and full length cDNA sequence of human collagen X. FEBS Lett. 311, 305–310.
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone 387. Beier, F., Eerola, I., Vuorio, E., Luvalle, P., Reichenberger, E., Bertling, W., von der Mark, K., and Lammi, M. J. (1996). Variability in the upstream promoter and intron sequences of the human, mouse and chick type X collagen genes. Matrix Biol. 15, 415–422. 388. Luvalle, P., Daniels, K., Hay, E. D., and Olsen, B. (1992). Type X collagen is transcriptionally activated and specifically localized during sternal cartilage maturation. Matrix 12, 404–413. 389. Schmid, I. M., and Linsenmayer, T. F. (1985). Immunohistochemical localization of short chain cartilage collagen (type-X) in avian tissues. J. Cell Biol. 100, 598–605. 390. Reichenberger, E., Aigner, T., von der Mark, K., Stoess, H., and Bertling, W. (1991). In situ hybridization studies on the expression of type X collagen in fetal human cartilage. Develop. Biol. 148, 562–572. 391. Luvalle, P., Iwamoto, M., Fanning, P., Pacifici, M., and Olsen, B. R. (1993). Multiple negative elements in a gene that codes for an extracellular matrix protein, collagen X, restrict expression to hypertrophic chondrocytes. J. Cell Biol. 121, 1173–1179. 392. Dourado, G., and Luvalle, P. (1998). Proximal DNA elements mediate repressor activity conferred by the distal portion of the chicken collagen X promoter. J. Cell Biochem. 70, 507–516. 393. Long, F., Sonenshein, G. E., and Linsenmayer, T. F. (1998). Multiple transcriptional elements in the avian type X collagen gene. Identification of Sp1 family proteins as regulators for high level expression in hypertrophic chondrocytes. J. Biol. Chem. 273, 6542–6549. 394. Volk, S. W., Luvalle, P., Leask, T., and Leboy, P. S. (1998). A BMP responsive transcriptional region in the chicken type X collagen gene. J. Bone Miner. Res. 13, 1521–1529. 395. Beier, F., Vornehm, S., Pöschl, E., von der Mark, K., and Lammi, M. J. (1997). Localization of silencer and enhancer elements in the human type X collagen gene. J. Cell Biochem. 66, 210–218. 396. Chambers, D., Young, D. A., Howard, C., Thomas, J. T., Boam, D. S., Grant, M. E., Wallis, G. A., and Boot-Handford, R. P. (2002). An enhancer complex confers both high-level and cell-specific expression of the human type X collagen gene. FEBS Lett. 531, 505–508. 397. Riemer, S., Gebhard, S., Beier, F., Poschl, E., and von der Mark, K. (2002). Role of c-fos in the regulation of type X collagen gene expression by PTH and PTHrP: localization of a PTH/PTHrP-responsive region in the human COL10A1 enhancer. J. Cell Biochem. 86, 688–699. 398. Gebhard, S., Poschl, E., Riemer, S., Bauer, E., Hattori, T., Eberspaecher, H., Zhang, Z., Lefebvre, V., de Crombrugghe, B., and von der Mark, K. (2004). A highly conserved enhancer in mammalian type X collagen genes drives high levels of tissue-specific expression in hypertrophic cartilage in vitro and in vivo. Matrix Biol. 23, 309–322. 399. O’Keefe, R. J., Loveys, L. S., Hicks, D. G., Reynolds, P. R., Crabb, I. D., Puzas, J. E., and Rosier, R. N. (1997). Differential regulation of type-II and type-X collagen synthesis by parathyroid hormone-related protein in chick growth-plate chondrocytes. J. Orthop. Res. 15, 162–174. 400. Harada, S., Sampath, T. K., and Rodan, G. A. (1996). Identification of an AP1-like response region for osteogenic protein-1 in type X collagen promoter. Ann. Acad. Sci. N.Y. 785, 259–260. 401. Zheng, Q., Zhou, G., Morello, R., Chen, Y., Garcia-Rojas, X., and Lee, B. (2003). Type X collagen gene regulation by Runx2 contributes directly to its hypertrophic chondrocyte-specific expression in vivo. J. Cell Biol. 162, 833–842. 402. Ducy, P., Zhang, R., Geoffroy, V., Ridall, A. L., and Karsenty, G. (1997). Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89, 747–754. 403. Otto, F., Thornell, A. P., Crompton, T., Denzel, A., Gilmour, K. C., Rosewell, I. R., Stamp, G. W., Beddington, R. S., Mundlos, S., Olsen, B. R., Selby, P. B., and Owen, M. J. (1997). Cbfa1, a candidate
404.
405.
406.
407.
408.
409.
410.
411.
412.
413. 414.
415.
416.
417.
37
gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89, 765–771. Komori, T., Yagi, H., Nomura, S., Yamaguchi, A., Sasaki, K., Deguchi, K., Shimizu, Y., Bronson, R. T., Gao, Y. H., Inada, M., Sato, M., Okamoto, R., Kitamura, Y., Yoshiki, S., and Kishimoto, T. (1997). Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89, 755–764. Nakashima, K., Zhou, X., Kunkel, G., Zhang, Z., Deng, J. M., Behringer, R. R., and de Crombrugghe, B. (2002). The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell 108, 17–29. Slavin, J., Unemori, E., Hunt, T. K., and Amento, E. (1994). Transforming growth factor beta (TGF-beta) and dexamethasone have direct opposing effects on collagen metabolism in low passage human dermal fibroblasts in vitro. Growth Factors. 11, 205–213. Murata, H., Zhou, L., Ochoa, S., Hasan, A., Badiavas, E., and Falanga, V. (1997). TGF-beta3 stimulates and regulates collagen synthesis through TGF-beta1- dependent and independent mechanisms. J. Invest. Dermatol. 108, 258–262. Inagaki, Y., Truter, S., and Ramirez, F. (1994). Transforming growth factor-beta stimulates alpha 2(I) collagen gene expression through a cis-acting element that contains an Sp1-binding site. J. Biol. Chem. 269, 14828–14834. Inagaki, Y., Nemoto, T., Nakao, A., Dijke, P., Kobayashi, K., Takehara, K., and Greenwel, P. (2001). Interaction between GC box binding factors and Smad proteins modulates cell lineage-specific alpha 2(I) collagen gene transcription. J. Biol. Chem. 276, 16573–16579. Chen, S. J., Yuan, W., Lo, S., Trojanowska, M., and Varga, J. (2000). Interaction of smad3 with a proximal smad-binding element of the human alpha2(I) procollagen gene promoter required for transcriptional activation by TGF-beta. J. Cell Physiol. 183, 381–392. Wang, E. A., Rosen, V., D’Alessando, J., Bauduy, M., Cordes, P., Harada, T., Israk, D. I., Hewick, R. M., Kern, K. M., LaDan, P., Luxenberg, D. P., McQuaid, D., Mountsatsos, J. K., Nove, J., and Wozney, J. M. (1990). Recombinant human bone morphogenetic protein induces bone formation. PNAS 87, 2220–2224. Sampath, T. K., Maliakal, J. C., Hauschka, P. V., Jones, W. K., Sasak, H., Tucker, R. F., White, K. H., Coughlin, J. E., Tucker, M. M., and Pang, R. H. (1992). Recombinant human osteogenic protein-1 (hOP-1) induces new bone formation in vivo with a specific activity comparable with natural bovine osteogenic protein and stimulates osteoblast proliferation and differentiation in vitro. J. Biol. Chem 267, 20352–20362. Reddi, A. H. (1998). Initiation of fracture repair by bone morphogenetic proteins. Clin. Orthop. S66–72. Fujii, M., Takeda, K., Imamura, T., Aoki, H., Sampath, T. K., Enomoto, S., Kawabata, M., Kato, M., Ichijo, H., and Miyazono, K. (1999). Roles of bone morphogenetic protein type I receptors and Smad proteins in osteoblast and chondroblast differentiation. Mol. Biol. Cell 10, 3801–3813. Takagi, M., Kamiya, N., Takahashi, T., Ito, S., Hasegawa, M., Suzuki, N., and Nakanishi, K. (2004). Effects of bone morphogenetic protein-2 and transforming growth factor beta1 on gene expression of transcription factors, AJ18 and Runx2 in cultured osteoblastic cells. J. Mol. Histol. 35, 81–90. Lepisto, J., Peltonen, J., Vaha-Kreula, M., Soderstrom, K., Niinikoski, J., and Laato, M. (1996). Selective modulation of collagen gene expression by different isoforms of platelet-derived growth factor in experimental wound healing. Cell Tissue Res. 286, 449–455. Hata, R., Sunada, H., Arai, K., Sato, T., Ninomiya, Y., Nagai, Y., and Senoo, H. (1988). Regulation of collagen metabolism and cell growth by epidermal growth factor and ascorbate in cultured human skin fibroblasts [published erratum appears in Eur J Biochem 1988 Aug 15;175(3): 679]. Eur. J. Biochem. 173, 261–267.
38 418. Rowe, D. W., and Kream, B. E. (1982). Regulation of collagen synthesis in fetal rat calvaria by 1,25- dihydroxyvitamin D3. J. Biol. Chem. 257, 8009–8015. 419. Kuroki, Y., Shiozawa, S., Kano, J., and Chihara, K. (1995). Competition between c-fos and 1,25(OH)2 vitamin D3 in the transcriptional control of type I collagen synthesis in MC3T3-E1 osteoblastic cells. J. Cell Physiol. 164, 459–464. 420. Kream, B. E., Lafrancis, D., Petersen, D. N., Woody, C., Clark, S., Rowe, D. W., and Lichtler, A. (1993). Parathyroid hormone represses alpha 1(I) collagen promoter activity in cultured calvariae from neonatal transgenic mice. Mol. Endocrinol. 7, 399–408. 421. Russell, S. B., Trupin, J. S., Myers, J. C., Broquist, A. H., Smith, J. C., Myles, M. E., and Russell, J. D. (1989). Differential glucocorticoid regulation of collagen mRNAs in human dermal fibroblasts. Keloidderived and fetal fibroblasts are refractory to down-regulation. J. Biol. Chem. 264, 13730–13735. 422. Oikarinen, A., Autio, P., Vuori, J., Vaananen, K., Risteli, L., Kiistala, U., and Risteli, J. (1992). Systemic glucocorticoid treatment decreases serum concentrations of carboxyterminal propeptide of type I procollagen and aminoterminal propeptide of type III procollagen. Br. J. Dermatol. 126, 172–178. 423. Silbermann, M. M. M. (1987). Dexamethasone impairs growth and collagen synthesis in condylar cartilage in vitro. Bone and Mineral 2, 87–106. 424. Advani, S., Lafrancis, D., Bogdanovic, E., Taxel, P., Raisz, L. G., and Kream, B. E. (1997). Dexamethasone suppresses in vivo levels of bone collagen synthesis in neonatal mice. Bone 20, 41–46. 425. Kream, B. E., Tetradis, S., Lafrancis, D., Fall, P. M., Feyen, J. H., and Raisz, L. G. (1997). Modulation of the effects of glucocorticoids on collagen synthesis in fetal rat calvariae by insulin-like growth factor binding protein-2. J. Bone Miner. Res. 12, 889–895. 426. Bird, J. L., and Tyler, J. A. (1994). Dexamethasone potentiates the stimulatory effect of insulin-like growth factor-I on collagen production in cultured human fibroblasts. J. Endocrinol. 142, 571–579. 427. Ireland, D. C., Bord, S., Beavan, S. R., and Compston, J. E. (2002). Effects of estrogen on collagen synthesis by cultured human osteoblasts depend on the rate of cellular differentiation. J. Cell Biochem. 86, 251–257. 428. Simeone, A., Acampora, D., Pannese, M., D’Esposito, M., Stornaiuolo, A., Gulisano, M., Mallamaci, A., Kastury, K., Druck, T., and Huebner, K. (1994). Cloning and characterization of two members of the vertebrate Dlx gene family. Proc. Natl Acad. Sci. USA 91, 2250–2254. 429. Tadic, T., Dodig, M., Erceg, I., Marijanovic, I., Mina, M., Kalajzic, Z., Velonis, D., Kronenberg, M. S., Kosher, R. A., Ferrari, D., and Lichtler, A. C. (2002). Overexpression of Dlx5 in chicken calvarial cells accelerates osteoblastic differentiation. J. Bone Miner. Res. 17, 1008–1014. 430. Ducy, P. (2000). Cbfa1: a molecular switch in osteoblast biology. Dev. Dyn. 219, 461–471. 431. Karsenty, G., Ducy, P., Starbuck, M., Priemel, M., Shen, J., Geoffroy, V., and Amling, M. (1999). Cbfa1 as a regulator of osteoblast differentiation and function. Bone 25, 107–108. 432. Komori, T. (2000). A fundamental transcription factor for bone and cartilage. Biochem. Biophys. Res. Commun. 276, 813–816. 433. Kim, I. S., Otto, F., Zabel, B., and Mundlos, S. (1999). Regulation of chondrocyte differentiation by Cbfa1. Mech. Dev. 80, 159–170. 434. Thunyakitpisal, P., Alvarez, M., Tokunaga, K., Onyia, J. E., Hock, J., Ohashi, N., Feister, H., Rhodes, S. J., and Bidwell, J. P. (2001). Cloning and functional analysis of a family of nuclear matrix transcription factors (NP/NMP4) that regulate type I collagen expression in osteoblasts. J. Bone Miner. Res. 16, 10–23. 435. Wagner, E. F. (2002). Functions of AP1 (Fos/Jun) in bone development. Ann. Rheum. Dis. 61, ii40–42.
KLAUS VON DER MARK 436. Grigoriadis, A. E., Wang, Z. Q., and Wagner, E. F. (1995). Fos and bone cell development: lessons from a nuclear oncogene. Trends Genet. 11, 436–441. 437. Kuroki, Y., Shiozawa, S., Sugimoto, T., and Fujita, T. (1992). Constitutive expression of c-fos gene inhibits type 1 collagen synthesis in transfected osteoblasts. Biochem. Biophys. Res. Commun. 182, 1389–1394. 438. Jinnin, M., Ihn, H., Yamane, K., and Tamaki, K. (2004). Interleukin13 stimulates the transcription of the human alpha2(I) collagen gene in human dermal fibroblasts. J. Biol. Chem. 279, 41783–41791. 439. Inagaki, Y., Nemoto, T., Kushida, M., Sheng, Y., Higashi, K., Ikeda, K., Kawada, N., Shirasaki, F., Takehara, K., Sugiyama, K., Fujii, M., Yamauchi, H., Nakao, A., de Crombrugghe, B., Watanabe, T., and Okazaki, I. (2003). Interferon alfa down-regulates collagen gene transcription and suppresses experimental hepatic fibrosis in mice. Hepatology 38, 890–899. 440. Rosenbloom, J., Feldman, G., Freundlich, B., and Jimenez, S. A. (1986). Inhibition of excessive scleroderma fibroblast collagen production by recombinant gamma-interferon. Association with a coordinate decrease in types I and III procollagen messenger RNA levels. Arthritis Rheum. 29, 851–856. 441. Scharffetter, K., Heckmann, M., Hatamochi, A., Mauch, C., Stein, B., Riethmuller, G., Ziegler-Heitbrock, H. W., and Krieg, T. (1989). Synergistic effect of tumor necrosis factor-alpha and interferongamma on collagen synthesis of human skin fibroblasts in vitro. Exp. Cell Res. 181, 409–419. 442. Kahari, V. M., Chen, Y. Q., Su, M. W., Ramirez, F., and Uitto, J. (1990). Tumor necrosis factor-alpha and interferon-gamma suppress the activation of human type I collagen gene expression by transforming growth factor-beta 1. Evidence for two distinct mechanisms of inhibition at the transcriptional and posttranscriptional levels. J. Clin. Invest. 86, 1489–1495. 443. Bird, J. L., and Tyler, J. A. (1995). Tumour necrosis factor alpha, interferon gamma and dexamethasone regulate IGF-I-maintained collagen production in cultured human fibroblasts. J. Endocrinol. 147, 167–176. 444. Xu, Y., Wang, L., Buttice, G., Sengupta, P. K., and Smith, B. D. (2003). Interferon gamma repression of collagen (COL1A2) transcription is mediated by the RFX5 complex. J. Biol. Chem. 278, 49134–49144. 445. Sengupta, P. K., Fargo, J., and Smith, B. D. (2002). The RFX family interacts at the collagen (COL1A2) start site and represses transcription. J. Biol. Chem. 277, 24926–24937. 446. Wright, E., Hargrave, M. R., Christiansen, J., Cooper, L., Kun, J., Evans, T., Gangadharan, U., Greenfield, A., and Koopman, P. (1995). The Sry-related gene Sox9 is expressed during chondrogenesis in mouse embryos. Nat. Genet. 9, 15–20. 447. Zhao, Q., Eberspaecher, H., Lefebvre, V., and de Crombrugghe, B. (1997). Parallel expression of Sox9 and Col2a1 in cells undergoing chondrogenesis. Dev. Dyn. 209, 377–386. 448. Ng, L. J., Wheatley, S., Muscat, G. E., Conway-Campbell, J., Bowles, J., Wright, E., Bell, D. M., Tam, P. P., Cheah, K. S., and Koopman, P. (1997). SOX9 binds DNA, activates transcription, and coexpresses with type II collagen during chondrogenesis in the mouse. Dev. Biol. 183, 108–121. 449. Foster, J. W. (1996). Mutations in SOX9 cause both autosomal sex reversal and campomelic dysplasia. Acta Paediatr. Jpn. 38, 405–411. 450. Giese, K., Pagel, J., and Grosschedl, R. (1994). Distinct DNA-binding properties of the high mobility group domain of murine and human SRY sex-determining factors. Proc. Natl Acad. Sci. USA 91, 3368–3372. 451. Lefebvre, V., Huang, W., Harley, V. R., Goodfellow, P. N., and de Crombrugghe, B. (1997). SOX9 is a potent activator of the chondrocyte-specific enhancer of the pro alpha1(II) collagen gene. Mol. Cell Biol. 17, 2336–2346.
CHAPTER 1 Structure, Biosynthesis and Gene Regulation of Collagens in Cartilage and Bone 452. Bell, D. M., Leung, K. K., Wheatley, S. C., Ng, L. J., Zhou, S., Ling, K. W., Sham, M. H., Koopman, P., Tam, P. P., and Cheah, K. S. (1997). SOX9 directly regulates the type-II collagen gene. Nat. Gen. 16, 174–178. 453. Huang, W., Zhou, X., Lefebvre, V., and de Crombrugghe, B. (2000). Phosphorylation of SOX9 by cyclic AMP-dependent protein kinase A enhances SOX9’s ability to transactivate a Col2a1 chondrocytespecific enhancer. Mol. Cell. Biol. 20, 4149–4158. 454. Bi, W., Huang, W., Whitworth, D. J., Deng, J. M., Zhang, Z., Behringer, R. R., and de Crombrugghe, B. (2001). Haploinsufficiency of Sox9 results in defective cartilage primordia and premature skeletal mineralization. Proc. Natl. Acad. Sci. USA 98, 6698–6703. 455. Bi, W., Deng, J. M., Zhang, Z., Behringer, R. R., and de Crombrugghe, B. (1999). Sox9 is required for cartilage formation. Nat. Genet. 22, 85–89. 456. Lefebvre, V., Behringer, R. R., and de Crombrugghe, B. (2001). L-Sox5, Sox6 and Sox9 control essential steps of the chondrocyte differentiation pathway. Osteoarthritis Cartilage 9 Suppl A, S69–75. 457. Joyce, M. E., Roberts, A. B., Sporn, M. B., and Bolander, M. E. (1990). Transforming growth factor-β and the initiation of chondrogenesis and osteogenesis in the rat femur. J. Cell Biol. 110, 2195–2207. 458. Redini, F., Galera, P., Mauviel, A., Loyau, G., and Pujol, J. P. (1988). Transforming growth factor β stimulates collagen and glycosaminoglycan biosynthesis in cultured rabbit articular chondrocytes. FEBS Lett. 234, 172–176. 459. Galera, P., Vivien, D., Pronost, S., Bonaventure, J., Redini, F., Loyau, G., and Pujol, J. P. (1992). Transforming growth factor-beta 1 (TGF-beta 1) up-regulation of collagen type II in primary cultures of rabbit articular chondrocytes (RAC) involves increased mRNA levels without affecting mRNA stability and procollagen processing. J. Cell Physiol. 153, 596–606. 460. Pujol, J. P., Galera, P., Pronost, S., Boumediene, K., Vivien, D., Macro, M., Min, W., Redini, F., Penfornis, H., and Daireaux, M. (1994). Transforming growth factor-beta (TGF-beta) and articular chondrocytes. Ann. Endocrinol. (Paris) 55, 109–120. 461. Chadjichristos, C., Ghayor, C., Herrouin, J. F., Ala-Kokko, L., Suske, G., Pujol, J. P., and Galera, P. (2002). Down-regulation of human type II collagen gene expression by transforming growth factor-beta 1 (TGFbeta 1) in articular chondrocytes involves SP3/SP1 ratio. J. Biol. Chem. 277, 43903–43917. 462. Bradham, D. M., in der Wiesche, B., Precht, P., Balakir, R., and Horton, W. (1994). Transrepression of type II collagen by TGF-beta and FGF is protein kinase C dependent and is mediated through regulatory sequences in the promoter and first intron. J. Cell Physiol. 158, 61–68. 463. Chen, P., Carrington, J. L., Hammonds, R. G., and Reddi, A. H. (1991). Stimulation of chondrogenesis in limb bud mesoderm cells by recombinant human bone morphogenetic protein 2B (BMP-2B) and modulation by transforming growth factor β1 and β2. Exp. Cell. Res. 195, 509–515. 464. Pizette, S., and Niswander, L. (2000). BMPs are required at two steps of limb chondrogenesis: formation of prechondrogenic condensations and their differentiation into chondrocytes. Dev. Biol. 219, 237–249. 465. Hanada, K., Solchaga, L. A., Caplan, A. I., Hering, T. M., Goldberg, V. M., Yoo, J. U., and Johnstone, B. (2001). BMP-2 induction and TGF-beta 1 modulation of rat periosteal cell chondrogenesis. J. Cell Biochem. 81, 284–294. 466. Shukunami, C., Ohta, Y., Sakuda, M., and Hiraki, Y. (1998). Sequential progression of the differentiation program by bone morphogenetic protein-2 in chondrogenic cell line ATDC5. Exp. Cell Res. 241, 1–11. 467. Valcourt, U., Ronziere, M. C., Winkler, P., Rosen, V., Herbage, D., and Mallein-Gerin, F. (1999). Different effects of bone morphogenetic proteins 2, 4, 12, and 13 on the expression of cartilage and
468.
469.
470.
471.
472.
473. 474.
475.
476.
477.
478. 479.
480.
481.
482.
483.
484.
485.
39
bone markers in the MC615 chondrocyte cell line. Exp. Cell Res. 251, 264–274. Ahrens, M., Ankenbauer, T., Schroder, D., Hollnagel, A., Mayer, H., and Gross, G. (1993). Expression of human bone morphogenetic proteins-2 or -4 in murine mesenchymal progenitor C3H10T1/2 cells induces differentiation into distinct mesenchymal cell lineages. DNA Cell Biol. 12, 871–880. Hatakeyama, Y., Nguyen, J., Wang, X., Nuckolls, G. H., and Shum, L. (2003). Smad signaling in mesenchymal and chondroprogenitor cells. J. Bone Joint Surg. Am. 85-A Suppl 3, 13–18. Leboy, P. S., Sullivan, T. A., Nooreyazdan, M., and Venezian, R. A. (1997). Rapid chondrocyte maturation by serum-free culture with BMP-2 and ascorbic acid. J. Cell Biochem. 66, 394–403. Volk, S. W., D’Angelo, M., Diefenderfer, D., and Leboy, P. S. (2000). Utilization of bone morphogenetic protein receptors during chondrocyte maturation. J. Bone Miner. Res. 15, 1630–1639. Drissi, M. H., Li, X., Sheu, T. J., Zuscik, M. J., Schwarz, E. M., Puzas, J. E., Rosier, R. N., and O’Keefe, R. J. (2003). Runx2/Cbfa1 stimulation by retinoic acid is potentiated by BMP2 signaling through interaction with Smad1 on the collagen X promoter in chondrocytes. J. Cell Biochem. 90, 1287–1298. Benya, P. D., and Shaffer, J. D. (1982). Dedifferentiated chondrocytes reexpress the differentiated collagen phenotype. Cell 30, 215–224. Dietz, U., Aigner, T., Bertling, W., and von der Mark, K. (1993). Alterations of collagen mRNA expression during retinoic acid induced chondrocyte modulation: Absence of untranslated alpha1(I) mRNA in hyaline chondrocytes. J. Cell Biochem. 52, 57–68. Benya, P., and Padilla, P. D. (1986). Modulation of the rabbit chondrocyte phenotype by retinoic acid terminates type II collagen synthesis without inducing type I collagen: The modulated phenotype differs from that produced by subculture. Develop. Biol. 118, 296–305. Paulsen, D. F., Langille, R. M., Dress, V., and Solursh, M. (1988). Selective stimulation of in vitro limb-bud chondrogenesis by retinoic acid. Differentiation 39, 123–130. Pacifici, M., Golden, E. B., Iwamoto, M., and Adams, S. L. (1991). Retinoic acid treatment induces type X collagen gene expression in cultured chick chondrocytes. Exp. Cell Res. 195, 38–46. Lee, G. S., Kochhar, D. M., and Collins, M. D. (2004). Retinoidinduced limb malformations. Curr. Pharm. Des. 10, 2657–2699. Cho, S. H., Oh, C. D., Kim, S. J., Kim, I. C., and Chun, J. S. (2003). Retinoic acid inhibits chondrogenesis of mesenchymal cells by sustaining expression of N-cadherin and its associated proteins. J. Cell Biochem. 89, 837–847. Luyten, F. P., Hascall, V. C., Nissley, S. P., Morales, T. I., and Reddi, A. H. (1988). Insulin-like growth factors maintain steady-state metabolism of proteoglycans in bovine articular cartilage explants. Arch Biochem. Biophys. 267, 416–425. Loeser, R. F., Todd, M. D., and Seely, B. L. (2003). Prolonged treatment of human osteoarthritic chondrocytes with insulin-like growth factor-I stimulates proteoglycan synthesis but not proteoglycan matrix accumulation in alginate cultures. J. Rheumatol. 30, 1565–1570. Sonal, D. (2001). Prevention of IGF-1 and TGFbeta stimulated type II collagen and decorin expression by bFGF and identification of IGF-1 mRNA transcripts in articular chondrocytes. Matrix Biol. 20, 233–242. Wroblewski, J., and Edwall-Arvidsson, C. (1995). Inhibitory effects of basic fibroblast growth factor on chondrocyte differentiation. J. Bone Miner. Res. 10, 735–742. Vortkamp, A., Lee, K., Lanske, B., Segre, G. V., Kronenberg, H. M., and Tabin, C. J. (1996). Regulation of rate of cartilage differentiation by Indian hedgehog and PTH-related protein. Science 273, 613–622. Vortkamp, A. (2000). The Indian hedgehog–PTHrP system in bone development. Ernst. Schering Res. Found Workshop, 191–209.
40 486. Weir, E. C., Philbrick, W. M., Amling, M., Neff, L. A., Baron, R., and Broadus, A. E. (1996). Targeted overexpression of parathyroid hormone-related peptide in chondrocytes causes chondrodysplasia and delayed endochondral bone formation. Proc. Natl Acad. Sci. U.S.A. 93, 10240–10245. 487. Amizuka, N., Warshawsky, H., Henderson, J. E., Goltzman, D., and Karaplis, A. C. (1994). Parathyroid hormone-related peptidedepleted mice show abnormal epiphyseal cartilage development and altered endochondral bone formation. J. Cell Biol. 126, 1611–1623. 488. Iwamoto, M., Jikko, A., Murakami, H., Shimazu, A., Nakashima, K., Takigawa, M., Baba, H., Suzuki, I., and Kato, Y. (1994). Changes in parathyroid hormone receptors during chondrocyte differentiation. J. Biol. Chem. 269, 17245–17251. 489. Crabb, I. D., O’Keefe, R. J., Puzas, J. E., and Rosier, R. N. (1992). Differential effects of parathyroid hormone on chick growth plate and articular chondrocytes. Calcif. Tissue Int. 50, 61–66.
KLAUS VON DER MARK 490. Goldring, M., Birkhead, J., Sandell, L., Kimura, T., and Krane, S. (1988). Interleukin 1 supresses expression of cartilage-specific types II and IX collagens and increases types I and III collagens in human chondrocytes. J. Clin. Inv. 82, 2026–2037. 491. Goldring, M. B., Fukuo, K., Birkhead, J. R., Dudek, E., and Sandell, L. J. (1994). Transcriptional suppression by interleukin-1 and interferon- gamma of type II collagen gene expression in human chondrocytes. J. Cell Biochem. 54, 85–99. 492. Goldring, M. B., Birkhead, J. R., Suen, L. F., Yamin, R., Mizuno, S., Glowacki, J., Arbiser, J. L., and Apperley, J. F. (1994). Interleukin-1 beta-modulated gene expression in immortalized human chondrocytes. J. Clin. Invest. 94, 2307–2316. 493. Pinnell, S. R. (1985). Regulation of collagen biosynthesis by ascorbic acid: a review. Yale J. Biol. Med. 58, 553–559. 494. Schwarz, R. I. (1985). Procollagen secretion meets the minimum requirements for the rate- controlling step in the ascorbate induction of procollagen synthesis. J. Biol. Chem. 260, 3045–3049.
Chapter 2
Fibrillogenesis and Maturation of Collagens Simon P. Robins
Matrix Biochemistry Group, Rowett Research Institute, Bucksburn, Aberdeen AB21 9SB
I. Introduction II. Fibrillogenesis III. Cross-linking
IV. Concluding Remarks References
I. INTRODUCTION
and, in addition to containing mixtures of fibrillar collagens, fibrils often incorporate a number of FACIT (Fibril Associated with Interrupted Triple-helix) collagens, particularly types IX, XII, XIV, and XIX. Such heterotypic fibrils are prevalent in cartilage but their structure and assembly will not be discussed here in detail (see Chapter 25). Proteoglycans, such as decorin, fibromodulin, and biglycan, together with other matrix constituents, have important influences on collagen fibril formation and, although the structure and biosynthesis of these components is discussed elsewhere in this volume (Chapters 4 and 5), their influence on fibrillogenesis will be reviewed briefly. A major factor in the functional capacity of collagen fibrils is the cross-linking process that imparts the tensile properties and resistance to enzymatic degradation. The initial cross-linking reactions and the spontaneous processes that lead to mature collagen fibrils will be discussed in detail, but changes associated with true aging and senescence that might be defined as being deleterious to function are beyond the scope of this review. Several reviews have been published on collagen fibrillogenesis [1, 2] and on cross-linking [3, 4].
The biophysical characteristics of bone and cartilage are governed to a large degree by the properties of their respective collagens, which constitute the majority of the protein content of these tissues. The properties of the collagens are, in turn, a consequence primarily of their fibrillar structure, and the characteristics of collagen fibrils dictate not only the biomechanical properties but also many other metabolic processes, including the susceptibility to degradation. The factors that control collagen fibrillogenesis and the subsequent stabilization of the fibrils are therefore crucial to an understanding of the dynamics of bone and cartilage metabolism. There are some 27 genetically distinct forms of collagen, but for the present discussion we shall be particularly concerned with those types that spontaneously self-assemble to form fibrils: collagens I, II, III, V, and XI. Of these fibrillar collagens, type I is the major component of bone and collagen II is the main collagenous component of cartilage. It has become increasingly clear, however, that collagen fibrils rarely comprise a single collagen type Dynamics of Bone and Cartilage Metabolism
41
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
42
Simon P. Robins
II. FIBRILLOGENESIS Most studies of fibril formation have been performed using collagen I and discussion here will be limited to this type of collagen. The main steps involved in fibrillogenesis are illustrated in Figure 1, although studies by Kadler et al. [5] have led to new concepts on the time-course and control of this process during embryonic growth, as will be described in more detail later in this section. As the collagen is synthesized and secreted as a precursor procollagen molecule with globular extensions at both the N- and C-terminal ends (see Chapter 1), the initial events are removal of the extension propeptides, which occurs through the action of discrete proteinases. One function of the propeptides is to prevent inappropriate intracellular fibril formation, since the released triplehelical collagen monomers have an intrinsic property to self assemble into fibrils arranged in a quarter-staggered array. This organization is mainly an entropy-driven process occurring through disruption of ordered solvent molecules at the monomer surface, and enhanced by specific ionic interactions between neighboring molecules to give a 4D overlap (see Fig. 1). This gap and overlap structure gives rise to the 67-nm banding pattern revealed in the electron microscope after negative staining with phosphotungstic acid where the excess stain enters the gap regions of the fibril. Removal of the N-propeptides intact is achieved by a family of proteases including a disintegrin and metalloprotease with thrombospondin motifs (ADAMTS)-2,
which is thought to act on procollagen I [6]. A related enzyme, ADAMTS-3, is prevalent in cartilage [7] and is likely to process procollagen II, although other members of this family of proteases, including ADAMTS-14, have been described [8]. C-proteinase activity, a property of all members of the tolloid family of metalloproteinases including bone morphogenetic protein-1 (BMP-1) [9], cleaves procollagens I, II, and III at a single locus releasing the propeptide intact. A procollagen C-proteinase enhancer (PCPE) has been isolated; this 55-kDa glycoprotein appears to facilitate cleavage by binding to sites either side of the procollagen cleavage site [10]. The N-proteinase binds to one of the FACIT collagens, type XIV [11], and the C-proteinase has been shown to bind the α1β1 and α2β1 integrins [12]. These proteinases may therefore have important biological functions in addition to their role in processing procollagens. Much of the information currently available on the mechanisms of fibril assembly has come from cell-free systems using purified proteinases to generate collagen monomers de novo [13]. Under these conditions, after a relatively short “nucleation” period, intermediate filaments with a blunt (β) end and a pointed (α) end are formed with growth occurring at the pointed end. Subsequently, growth of the fibril from the blunt end is initiated giving rise to a parabola-shaped intermediate fibril. Initial studies showed that the intermediate filaments were bipolar with the N-terminals oriented towards both of the pointed ends, thus necessitating a change in orientation at some
Figure 1 Fibril formation and structure: following secretion of procollagen, the propeptides are cleaved by an N-protease (PNP) and C-protease (BMP-1); intermediate fibrils assemble with collagen monomers in a regular, quarter-staggered array (see inset) that gives rise to 67 nm banding; longer fibrils are formed by fusion of intermediates which may also occur laterally. The assembly process is influenced by interactions of both procollagen and the fibrils with proteoglycans and other matrix molecules (not shown).
43
Chapter 2 Fibrillogenesis and Maturation of Collagens
point along the filament. This type of intermediate filament was subsequently shown to occur in vivo [14]. Fibrils produced from acid-extracted collagen monomers are unipolar and the initial presence of the C-propeptide during fibril formation appears to have a major effect [1]. Both longitudinal and lateral growth of the collagen fibril are believed to occur through fusion of these intermediates. Some studies have shown that only unipolar intermediates undergo fusion; indeed, regulation of the formation of these two types of intermediate may have important influences on fibril growth. Other studies of developing chick tendon support the concept of fibril growth by fusion of fibril segments [15], but they also found a more rapid growth in length with development so that, in an 18-day tendon, no small, intermediate fibril segments were observed [16]. At this stage, analyses of serial sections by transmission electron microscopy showed that fibril ends were detected randomly along the fibril. This observation indicates that single fibrils did not span the full length of tendon but, in contrast to earlier stages of development, no fibrils shorter than 60 µm were detected [16]. As will be discussed later, these results also emphasize the importance of inter- as well as intrafibril cross-linking, since the former are essential in stabilizing the fibrils within the tendon fibre. Studies in vivo of fibrillogenesis in embryonic chick tendon have led to novel concepts of how cells may orchestrate the process of tissue assembly [5]. Using sophisticated electron microscopy techniques, the authors showed that procollagen could be processed intracellularly and that initiation of fibrillogenesis occurred in Golgi to plasma membrane carriers that were targeted to protrusions on the cell surface, designated “fibripositors”. These structures appeared to deposit their pre-assembled material onto the growing tendon and maintained the alignment by forming channels between cells parallel to the axis of the tendon. Fibripositors were not observed in collagenous matrices formed in vitro suggesting that, unless novel methods can be devised to maintain the 3-D shape of the cell, culture systems are unable to provide a full model for studying the mechanisms of fibril alignment [5]. Fibripositors were also absent in vivo postnatally suggesting that, for larger fibrils, increases in diameter occur through the accretion of extracellular collagen, as discussed previously in this section.
A. Factors Affecting Fibril Formation The kinetics of removal of the propeptides clearly influences fibril formation [17], and transient retention of the N-propeptides has been suggested as a model to account for preferred fibril diameters at intervals of about
11 nm corresponding to cleavage of the propeptide at the surface [18, 19]. Once the propeptides have been removed, however, there are several other factors which affect the nature and rate of fibril formation. 1. Telopeptides
The short, nonhelical domains at each end of the molecule, the telopeptides, which contain the cross-linking sites (see Section III), are only 15–20 amino acids in length but their removal through enzyme treatment has a dramatic effect on the kinetics of fibril formation [20]. Experiments in which the interactions between collagen molecules were assessed by X-ray diffraction and osmotic stress indicated that collagen fibrils formed after enzymatic removal of telopeptides had a native-like, banded structure, suggesting that all information necessary for fibril assembly was contained within the helical portion [21]. The conclusion from these studies was that the telopeptides performed essentially a catalytic role during fibrillogenesis [21]. Further evidence for the role of telopeptides in fibrillogenesis came from molecular modeling studies which showed that the N-telopeptides adopt an ordered structure after interaction with its helical receptor site [22]. Extension of the models to include quasi-hexagonal packing of collagen molecules showed that each N-telopeptide region can form linkages with two adjacent, aligned helical receptor regions [22], consistent with the highly ordered structure of this domain observed by X-ray diffraction [23, 24]. 2. Other Fibrillar Collagens
Fibril assembly may be modified through the association of different fibrillar collagens to form heterotypic fibrils and the formation of macromolecular alloys may provide a mechanism to produce a diverse range of fibrillar composites, consistent with their varied functional requirements [25, 26]. Collagen I forms hybrid fibrils particularly with collagens III and V, and retention of the N-propeptides of these collagens appears to constitute an additional mechanism for the control of fibril assembly, as has been shown, for example, for collagen III in skin [27]. Studies of the tissues from mice containing a mutant collagen III gene have demonstrated the crucial role of this molecule for adequate collagen type I fibrillogenesis [28]. Collagen V is known to play a central role in regulating fibril assembly in the cornea [25, 29] and, because of the existence of tissue-specific isoforms, this collagen also plays an important role in collagen matrix assembly in many other tissues [25]. Collagen type II also forms heterotypic fibrils, particularly with collagen XI, although the aggregates formed between II/XI are very different from those between collagens I/XI [26]. This observation may also be associated with the much higher content of
44
Simon P. Robins
glycosylated hydroxylysine residues in collagen II, which affects directly formation and morphology of the fibrils as shown by experiments with recombinant collagens [30]. Interestingly, in the cartilage matrix produced by a chondrosarcoma cell line lacking N-propeptidase activity, normal covalent cross-linking was observed between collagen II and other collagens including type XI, despite the presence primarily of unprocessed N-procollagen [31]. The importance of heterotypic fibrils in cartilage collagen assembly is discussed in more detail in Chapter 25. 3. FACIT Collagens
The fibril-associated collagens with interrupted triple helix (FACIT) are particularly important in cartilage, but this topic will be dealt with in Chapter 25. For collagen type I, interactions with the FACIT types XII and XIV have major effects in modulating the properties of soft tissues, but these collagens are essentially absent in bone. Collagen types XII and XIV are relatively large molecules with extended globular domains. There is an increasing body of evidence to indicate that these collagens mediate interactions between fibrils and are important in modulating responses to biomechanical stimuli [32–34]. 4. Other Matrix Constituents
Of the many influences from other matrix constituents on collagen fibrillogenesis, that exerted by the family of small, leucine-rich proteoglycans, which includes decorin, fibromodulin, and biglycan (see Chapter 5), has probably received the most attention. Decorin appears to play a major role in collagen fibrillogenesis, as mice with targeted decorin gene disruption were shown to exhibit skin fragility with abnormal morphology characterized by uncontrolled lateral fusion of collagen fibrils [35]. There are, however, conflicting reports on the effects of decorin on fibrillogenesis in vitro, with one model proposing that binding of archshaped decorin molecules to the collagen helix potentially limits lateral accretion of fibrils [36], whereas other studies suggested that the presence of decorin increased fibril diameter [37]. More detailed studies of decorin structure [38] may help to resolve the mechanism of these interactions.
III. CROSS-LINKING The formation of intermolecular cross-links within newly formed collagen fibrils conveys the structural stability necessary for tissue function. The main processes involved in the formation initially of intermediate, chemically reducible cross-links, followed by maturation to more stable, nonreducible bonds, have been established over several years. A number of questions remain, however,
concerning the nature of the mature forms of bonds and the mechanisms of tissue-specific cross-link formation. It is clear that lysine-derived cross-links account for almost all cross-linking in young tissue and that their formation is driven by a single enzymatic process – the action of lysyl oxidase.
A. Lysyl Oxidase Lysyl oxidase (LOX) is known to be one member of a family of enzymes including four isoforms termed lysyl oxidase-like proteins (LOXL1–4), many of which have copper-dependent amine oxidase activity [39]. For collagen, lysyl oxidases catalyze the conversion of lysine or hydroxylysine residues in the N- and C-telopeptides to aldehyde residues. For LOX, lysine tyrosyl quinone, formed by the cross-linking of two amino acid side chains within the enzyme, acts a cofactor necessary for activity [40]. Early studies showed that lysyl oxidase acts extracellularly and requires as a substrate native collagen fibrils in a quarter-staggered array [41]. This observation led to suggestions that amino acid residues at the overlap sites in the helix of adjacent molecules in the fibril participate in the reaction: these sites are known to have highly conserved amino acid sequences [42]. The presence of multiple LOX isoforms suggests an addition level of complexity for the regulation of collagen cross-linking. Studies of MC3T3-E1 osteoblastic cells indicated that LOX and LOX-like protein expression is highly regulated during differentiation, with all isoforms except LOXL2 being expressed during growth and mineralization but with different quantitative differences at each stage [43]. The results of other studies involving lysyl oxidase inhibitors in osteoblastic cells are also consistent with the view that expression of this enzyme may be an important regulator of collagen deposition [44]. LOX is synthesized as a precursor which undergoes post-translational glycosylation and cleavage of the propeptide portion by procollagen C-proteinase/BMP-1 [45]. LOXL1 is processed similarly by BMP-1 but there are no known proteolytically processed forms of LOXL2, LOXL3, or LOXL4 [39]. Expression of LOX is regulated by several cytokines including transforming growth factor-β [46]. Early studies suggested that LOX may also have important functions in tumor suppression [47]. This activity was subsequently shown to reside in the 18–20 kDa, N-propeptide portion of LOX, which is highly basic, possibly facilitating heparin sulfate-mediated re-uptake by cells to accomplish their role [48]. The N-propeptide domains are not well conserved in the LOX-like proteins and the latter are, therefore, unlikely to share tumor suppressor activity [48].
45
Chapter 2 Fibrillogenesis and Maturation of Collagens
B. Tissue Specificity of Cross-linking As shown in Figure 2, there are two major pathways of cross-linking depending on whether the residue in the telopeptide is lysine or hydroxylysine. As will become evident later in this section, the presence of hydroxylysine in the telopeptide has a profound effect on the stability of the cross-links initially formed as well as on the final products. The types of cross-link formed are related to the tissue rather than to the collagen type, and the tissue specificity is governed primarily by enzymatic hydroxylation reactions in the telopeptides. Thus, in skin collagen, virtually no hydroxylysine is present in the telopeptides so that cross-linking proceeds through the aldimine route (Fig. 2) and ultimately to histidine-containing compounds, whereas, in cartilage and bone, a large proportion of the telopeptide lysines are hydroxylated giving rise mainly to keto-imine forms of cross-link, which can lead to the formation of pyridinium and pyrrolic cross-links on maturation. Hydroxylation of lysine residues in the telopeptides clearly has crucial effects on cross-linking. Following early studies indicating that the enzyme which hydroxylates lysine residues in the telopeptides is distinct from that which acts on lysine residues destined to be in the helix [49], the enzyme designated telopeptide lysyl hydroxylase was characterized as the longer splice variant of one isotype of lysyl hydroxylase 2 (LH2) [50, 51]. Three isotypes of lysyl hydroxylase (LH1–LH3) have been described [52], but only LH2 has tissue-specific expression of alternatively
spliced forms [53]. Studies of a rare form of osteogenesis imperfecta, Bruck syndrome, indicated that the cause was a lack of telopeptide lysyl hydroxylase activity: this was initially mapped to chromosome 17 [54] but later studies of two more cases showed that the mutation was in LH2 located on chromosome 3 [55], and the earlier report probably, therefore, reflects the heterogeneity of this disorder. Studies in vitro have confirmed that alternatively spliced forms of LH2 direct the pathway of collagen cross-linking [56–58]. It is clear, therefore, that tissue differences in the patterns of collagen cross-linking are due to variations in the expression and activities of telopeptide and helical lysyl hydroxylases. Some hydroxylysine residues in central parts of fibrillar collagens destined to become the helix are further modified enzymatically during synthesis by addition of galactosyl or both galactosyl and glucosyl residues to the hydroxyl group (see Chapter 1). These hydroxylysine glycosides may participate in cross-linking giving rise to glycosylated forms of both the difunctional cross-links and their maturation products.
C. Mechanisms of Cross-linking 1. Intermediate, Difunctional Cross-links
Following the action of lysyl oxidase, the lysine aldehyde (allysine) or hydroxylysine aldehyde (hydroxyallysine) residues undergo spontaneous reactions according
Figure 2 Collagen cross-linking is tissue specific. Intracellular hydroxylation of lysine in the telopeptide gives rise to two pathways of extracellular cross-linking, the lysine aldehyde (allysine) pathway and the hydroxyallysine pathway; the latter results in mainly pyridinium and pyrrolic cross-links in bone and cartilage.
46
Simon P. Robins
to the separate pathways indicated in Figure 3. For allysine, reaction with a hydroxylysine or lysine residue in the helix results in the formation of the aldimines, dehydro-hydroxylysinonorleucine (∆-HLNL) or dehydrolysinonorleucine (∆-LNL), respectively. These aldimine bonds provide the main structural stability in newly formed dermal tissue, but are readily cleaved by heat or dilute acids, rendering such tissue completely soluble in dilute acetic acid. Another consequence of the chemical lability of the aldimine bonds is that reduction with borohydride is essential to stabilize the compounds to the hydrolytic processes necessary for their isolation: the use of radiolabeled borohydride was helpful initially in identifying the reduced compounds as HLNL and LNL [59]. The corresponding difunctional cross-links formed through reactions of hydroxyallysine with helical hydroxylysine or lysine residues are stabilized by a spontaneous Amadori rearrangement to produce hydroxylysino-5-ketonorleucine (HLKNL) or lysino-5-keto-norleucine (LKNL), respectively (Fig. 3). This rearrangement renders these compounds stable to the effects of mild acid treatment and peptides containing these cross-links have been isolated without borohydride reduction. After reduction, HLKNL and LKNL are converted to dihydroxy-lysinonorleucine (DHLNL) and hydroxylysinonorleucine (HLNL). The latter compound can therefore be derived from both the allysine and hydroxyallysine pathways but these can be distinguished by identifying the products of a Smith degradation (periodate cleavage followed by borohydride
Figure 3
reduction): in bone about 75% of HLNL is derived from hydroxyallysine [60]. Another difference between the lysine aldehyde and hydroxylysine aldehyde pathways of cross-linking is that the former compound can participate in an “aldol” condensation reaction to give an intramolecular crosslink between the α-chains of a single molecule. The aldol condensation product (ACP) undergoes further reactions that have been detected in tissues after reduction with borohydride (see Fig 3). Hydroxyallysine does not form intramolecular aldols and no corresponding products from this pathway have been detected. The main product derived from the aldol which was identified in borohydride-reduced tissue was the compound histidino-hydroxymerodesmosine (HHMD), involving the addition of hydroxylysine and histidine residues to the ACP [61]. Although evidence was presented that this compound was an artefact of the borohydride reduction process and that the nonreduced form did not constitute a cross-link in vivo [62], contradictory results were reported subsequently [63] and this question has still not been fully resolved. 2. Maturation of Intermediate Cross-links
The reducible, intermediate cross-links decrease in concentration during maturation and aging of the tissues [59]. Initial studies to define the way in which these bonds are modified during maturation focused on a reduction in vivo to give the same cross-links as those produced by borohydride, but these observations have been discounted: similarly, a proposed oxidative pathway of modification
Inter- and intramolecular cross-link formation occurs in different ways depending on the aldehyde involved. Telopeptide lysine aldehyde reacts with lysine or hydroxylysine residues in the helix to form aldimine (Schiff base) cross-links that give rise to LNL and HLNL on reduction with borohydride. When hydroxylysine aldehyde is present in the telopeptide, the products initially formed undergo Amadori rearrangement to form the more stable keto-imines, LKNL and HLKNL, which on reduction give HLNL and DHLNL. Lysine aldehyde also forms an intramolecular cross-link, the Aldol condensation product, which on reduction can produce either a reduced aldol or the tetrafunctional cross-link, HHMD.
47
Chapter 2 Fibrillogenesis and Maturation of Collagens
also has not been substantiated. Firm evidence has been obtained that the major maturation products are pyridinium and pyrrolic cross-links in skeletal tissues (Fig. 4) and a histidine adduct in soft tissues, although mechanisms of the maturation processes are not entirely clear, particularly with respect to kinetics and stoichiometry. In addition, immunochemical studies of mature human bone indicated that, at the C-terminal end at least, there appeared to be as yet unidentified cross-links [64]. a. Pyridinium cross-links A fluorescent, 3-hydroxypyridinium compound in hydrolysates of bovine tendon was identified by Fujimoto and colleagues [65] as a trifunctional cross-link, termed pyridinoline, and an analog, named deoxypyridinoline, was discovered in bone [66]. Because the latter cross-link involved reaction of a lysine residue from the helix, it has been termed lysyl-pyridinoline, or LP, to distinguish this compound from pyridinoline which is derived from helical hydroxylysine, and is referred to as hydroxylysylpyridinoline or HP [67]. Two mechanisms for formation of the pyridinium cross-links have been proposed: one involving interaction between two intermediate keto-imine cross-links [68] and the other a reaction between the ketoimine and hydroxylysine aldehyde [69]. The chemistry of these two mechanisms is in fact very similar but there are important functional implications as will be discussed in Section III.E. b. Pyrrolic cross-links The initial suggestion for the presence in collagen of pyrrolic cross-links came from work by Scott and
colleagues [70], based on the formation of colored reaction products of collagen hydrolysates with Ehrlich’s reagent (p-dimethylaminobenzaldehyde in acid solution). Later studies identified pyrrole-containing peptides from tendon and led to a proposed mechanism of cross-link formation [71]. This proposal suggests that in tissues where not all of the telopeptide lysine residues are hydroxylated, lysine aldehyde may interact with the intermediate keto-imine to form a pyrrolic, trifunctional cross-link. According to this mechanism, therefore, pyrrole cross-links should be absent from dermal tissue because of the lack of keto-imines and also absent from cartilage because the telopeptides are fully hydroxylated: this tissue distribution has been verified experimentally [4]. Characterization of pyrrole cross-links was difficult mainly because of their instability during the hydrolytic and chromatographic procedures necessary to isolate them. By targeting the pyrrole with novel Ehrlich’s reagent derivatives followed by mass spectral analysis, direct confirmation for the structure and location of pyrrole cross-links in bone was obtained [72]. These studies showed that, similar to the pyridinium compounds, pyrrole cross-link analogs from both helical hydroxylysyl and lysyl residues were present [72]: consistent with a previous report [73], pyrrole cross-linked peptides were derived mainly from the N-terminal end of collagen. The methods developed to date do not provide precise quantitative data for pyrrole cross-links and further studies are required to provide a comprehensive analysis of the changes in pyrrole cross-linking during development and with age.
Figure 4 Pyridinium and pyrrolic cross-links are the best characterized products of maturation. The difunctional cross-links, hydroxylysino-5-ketonorleucine (HLKNL) and lysino-5-ketonorleucine (LKNL), where R = –OH or –H, respectively, are converted to the different trifunctional cross-links depending on whether the telopeptide aldehydes are hydroxylated.
48
Simon P. Robins
c. Kinetics of cross-link maturation There have been relatively few studies of the kinetics of cross-link maturation and most of these have involved administration of radioactively labeled lysine to rats. From these types of study, the half-life of the reducible, difunctional cross-links has been estimated as a few weeks [74]. Several groups have shown that the concentrations of reducible, intermediate cross-links remain much higher in bone than in other tissues, including skin and cartilage. The continual turnover of bone matrix, in contrast to the other tissues, undoubtedly contributes to this observation and results in a higher proportion of recently synthesized collagen being present in bone.
D. Mineralization and Cross-linking The structural features of collagen type I fibrils in bone which allow mineralization to occur have not been fully elucidated, and conflicting reports have been published. It is generally agreed, however, that the types and location of cross-linking are different in bone compared with soft tissues containing collagen type I, such as skin or tendon. Studies of C-terminal cross-linking have suggested that mineralization itself induces structural alterations in the fibril [75], although some of the observed changes may be due to the conditions of tissue processing [76]. Because the concentrations of pyridinium cross-links in mammalian bone are relatively low throughout life in comparison with other tissues, such as cartilage [77], it has been suggested that the presence of mineral arrests maturation of difunctional cross-links into pyridinoline and deoxypyridinoline. Arguing against this contention is the fact that “aging” in vitro of mineralized bone (incubation in physiological saline at 37°C) converts precursor crosslinks to their mature forms [78]; however, earlier studies indicated that there may be differences in the rate of reaction between mineralized and nonmineralized bone [79]. The role of pyrrole cross-links in mammalian bone, although suggested to be functionally the more important mature cross-link in avian bone [4], has not yet been fully evaluated. The natural mineralization of turkey leg tendon at around 12–14 weeks of age has been widely used as a model system for the study of structural changes in collagen type I during the addition of mineral [80]. These changes appear to be strain-induced and are associated with neovascularization [81, 82]. Yamauchi and colleagues showed that the nonmineralized part of the tendon contained primarily pyridinoline cross-links whereas the mineralized portion also contained a high proportion of
deoxypyridinoline [83]. As this altered cross-linking pattern appears to involve changes in the intracellular hydroxylation of lysine residues, these results inferred that mineralization is accompanied by the production of a new matrix with modified post-translational modifications. Similar changes in lysine hydroxylation and cross-link patterns were found to be associated with changes in thermal characteristics of the collagenous matrix [76]. Studies of calcifying callus in dogs also revealed decreases in the ratio of pyridinoline to deoxypyridinoline associated with increased mineralization within the collagen fibrils [84]. These experiments serve to emphasize the fact that collagen type I capable of mineralization exhibits specific patterns of post-translational hydroxylation and fibril structure, which in turn are associated with particular cross-linking patterns, of which the presence of deoxypyridinoline is one characteristic.
E. Cross-linking in Relation to Molecular Packing A large number of studies of the three-dimensional structure of fibrillar collagen over many years culminated in establishing a quasi-hexagonal lattice model as the most appropriate to fit the available data [24]. This model not only satisfies the limitations imposed by the chemistry and location of the intermolecular cross-links [85], but also is consistent with the packing density and crytallinity changes associated with the overlap segments compared with other regions of the fibrils [86]. In conjunction with molecular modeling studies [22], the data reveal a high degree of interconnectivity between microfibrils [87], with a corrugated arrangement of fibrils cross-linked between segment 1 and 5 (Fig. 5). The difunctional, intermediate cross-links are known to stabilize the quarter-staggered overlap (between segments 1 and 5 in Fig. 5). During their maturation to trifunctional pyridinium or pyrrole cross-links, reactions can occur either with an aldehyde from another chain of the same molecule, giving an intrafibril cross-link, or with a reactive group from an adjacent microfibril, giving an interfibrillar bond linking three molecules. Distinguishing between these possibilities biochemically is difficult, but these arrangements will have major effects on the functional properties of the tissue. For mineralized tissues, electron microscopy, combined with computed tomography and three-dimensional image reconstruction methods, have provided evidence that some mineral crystals span the hole regions of adjacent microfibrils [88], a configuration that requires all microfibrils to be in register with their hole regions aligned.
49
Chapter 2 Fibrillogenesis and Maturation of Collagens
discovered as the toxic agent released in vivo after chronic ingestion of the sweet pea, Lathyrus oderatus, is now known to be an irreversible inhibitor of the cross-link initiating enzyme, lysyl oxidase, and leads to severe malformations in the skeleton as well as many other abnormalities. Following the discovery of this agent in 1959 [94], βAPN has been used extensively for both in vivo and in vitro experiments to provide noncross-linked matrix proteins for research purposes. In addition to the developmental problems associated with a severe lack of cross-linking, the quantity and nature of cross-link has profound effects on the susceptibility to proteolysis and the mechanical characteristics of collagen fibrils. 1. Susceptibility to Proteolysis Figure 5
Inter- and intrafibrillar cross-links are shown by this schematic representation of quarter-staggered collagen molecules, 4.4D periods in length, where each segment is numbered (upper). One possible arrangement of compressed microfibrils (pentafilaments denoted by the shaded areas) is shown (lower) with a corrugated arrangement of both intra- and interfibrillar cross-link between segments 1 and 5.
F. In situ Analysis of Collagen Cross-linking With improvements in instrumentation and software, Fourier transform infrared imaging (FTIRI) has found applications in the analysis of both mineral and matrix constituents of bone [89–91]. In addition to providing information on the mineral to matrix ratios and mineral crystallinity, spectral analysis can reveal specific data on the environment of different cross-link residues: whilst this information does not directly identify particular cross-links, analysis of isolated peptides with known cross-links has allowed distinctions to be made between difunctional, reducible bonds and pyridinium compounds [89]. These analyses are performed in an array so that an image of cross-link types can be built up in sections of bone. The technique not only provides a novel way to monitor maturational changes in collagen cross-linking in bone but also facilitates studies of the changes that occur in diseases and during therapy [91–93].
G. Biological Consequences of Intermolecular Collagen Cross-linking Inhibition of collagen cross-linking via the dietary lathyrogen, β-amino-proprionitrile (βAPN) provides graphic evidence of the importance of the intermolecular bonds for proper tissue function. The lathyrogen, originally
The triple helical structure of individual monomers within the collagen fibril is resistant to most proteolytic enzymes except specific collagenases or matrix metalloproteinases (see Chapter 11) which cleave through all three chains simultaneously at a point three-quarters distance from the N-terminal end. Other enzymes, such as cathepsins and elastase cleave the telopeptide portions of the molecule between the cross-linking site and the helix, thus depolymerizing the fibril. The presence of the lysinederived cross-links has, however, been shown to affect dramatically the susceptibility to proteolysis of the fibrils. Thus, for fibrils that were formed in vitro from soluble chick bone collagen, the susceptibility to proteolysis was related directly to the concentrations of difunctional crosslinks in the fibrils introduced by the actions of purified lysyl oxidase [95]. Although these studies appeared to show that as little as one cross-link per ten collagen molecules was sufficient to impart a two- to three-fold resistance to collagenase digestion, this may be an underestimate, as the cross-link values represent a figure for the whole fibril but the methods used will have introduced higher concentrations of cross-links at the surface of the fibrils. Maturation of fibrils leads to a lower susceptibility to enzyme degradation and the formation of multifunctional cross-links that link more than two molecules (see Section III.E) is crucial in giving rise to these properties. Distinct from maturational changes, the introduction of cross-links between the helices of adjacent molecules through aging processes, including advanced glycation end-product (AGE) reactions, leads to major changes in the properties of the fibrils [3]. The presence of a fibrillar network cross-linked at the telopeptides means that once cleavage of a peripheral molecule by collagenase has occurred the cleaved chains will remain attached for a period of time to the fibril through the cross-links. At body temperature, the cleaved
50
Simon P. Robins
fragments denature and can be detected by specific antibodies that react only with sequence-dependent epitopes, thus providing a means to assess immunocytochemically patterns of degradation in tissues [96]. Similar techniques have been used to detect the degradation patterns in cartilage [97]. 2. Mechanical Properties
There is no doubt that cross-linking in collagen is primarily responsible for the strength of the fibrils but the relative contributions of the different types of cross-link are less clear. Studies of the effects of partial lathyrism on the mechanical strength of embryonic avian bone indicated that a 10% decrease in the concentration of the difunctional keto-imine cross-link resulted in a 15% decrease in bone strength [78]. Similar studies have shown that the concentrations of pyridinoline are also related to the strength of adult bone in the rabbit [98] and rat [99]. It has been suggested that the pyrrole cross-links are more important than the pyridinium cross-links in determining the mechanical properties of bone [78], but this conclusion is based on experiments in avian bone which contain much smaller proportions of the pyridinium cross-links than mammalian bone. In bovine tendon, the concentrations of the pyrrolic and pyridinium cross-links were found to have similar positive correlations with thermal stability [100], suggesting that both types of cross-link were performing a similar function. There appear to be differences in the relative proportions of cross-links involving the N- and C-terminal telopeptides, with about 85% of the pyrroles being present at the N-terminus whereas the pyridinium cross-links were equally distributed [73]. It is possible that the N-telopeptides are involved in more intermicrofibrillar bonding in comparison to the C-terminal cross-links (see Section III.E), a factor that may have effects on biomechanical properties of the tissue. Studies of the age-related changes in bone toughness revealed very little change in the concentrations of pyridinium cross-links, but there were significant increases in pentosidine, a marker for nonenzymatic glycation, associated with decreased bone stiffness and strength [101]. These observations of the effects of in vivo glycation contrast with those for in vitro glycation by ribose that induced increased stiffness of bone together with increased fragility [102].
IV. CONCLUDING REMARKS Intracellular, post-translational modifications of collagen are recognized as key to many of the properties of the fibrillar collagens and hydroxylation of lysine residues, in particular, has far-reaching effects on cross-linking,
tissue strength and the propensity to mineralize. The application of new techniques for microscopy and spectroscopy has contributed to advances in our understanding of collagen fibril formation, mineralization and cross-link maturation. There has been relatively little consideration in this chapter of age-related changes in collagen but these may have important implications for the mechanical properties of the tissue in relation to diseases such as osteoporosis. Of the many biochemical changes that occur, attention has focused on increased hydroxylation of lysine residues and its effects on cross-linking and hydroxylysine glycoside concentrations [103]. The latter may give rise to the production of smaller collagen fibrils [30] that have modified, less robust mechanical properties consistent with the changes seen in osteoporosis. Another age-related change in collagen structure is an aspartyl to isoaspartyl transformation in both the C-terminal [104] and N-terminal [105] telopeptides. The C-telopeptide has proved useful as a biochemical marker (see Chapters 24 and 35) and alterations in the relative amounts of these modified residues have been associated with increased skeletal fragility [106]. With the realization that bone mineral density measurements have a limited ability to reflect fracture risk, much emphasis has been placed on attempting to define biochemical parameters that may provide assessments of “bone quality” in terms of fragility. The ratios of aspartyl/ isoaspartyl residues in the telopeptides have been proposed as one such parameter, as have the maturation rates of collagen cross-links detected by Fourier transform infrared microscopy. One of the challenges of future research is to fully validate these newer measures and to establish real links between the mechanical integrity of bone and biochemical characteristics of the collagen matrix.
Acknowledgments I am grateful to the Scottish Executive Environmental and Rural Affairs Department for support.
References 1. Kadler, K. E., Holmes, D. F., Trotter, J. A., and Chapman, J. A. (1996). Collagen fibril formation. Biochem. J. 316, 1–11. 2. Canty, E. G., and Kadler, K. E. (2002). Collagen fibril biosynthesis in tendon: a review and recent insights. Comp. Biochem. Physiol. A. Mol. Integr. Physiol. 133, 979–985. 3. Reiser, K., McCormick, R. J., and Rucker, R. B. (1992). Enzymatic and nonenzymatic cross-linking of collagen and elastin. FASEB J. 6, 2439–2449. 4. Knott, L., and Bailey, A. J. (1998). Collagen cross-links in mineralizing tissues: a review of their chemistry, function, and clinical relevance. Bone 22, 181–187.
Chapter 2 Fibrillogenesis and Maturation of Collagens 5. Canty, E. G., Lu, Y., Meadows, R. S., Shaw, M. K., Holmes, D. F., and Kadler, K. E. (2004). Coalignment of plasma membrane channels and protrusions (fibripositors) specifies the parallelism of tendon. J. Cell. Biol. 165, 553–563. 6. Colige, A., Sieron, A. L., Li, S. W., Schwarze, U., Petty, E., Wertelecki, W., Wilcox, W., Krakow, D., Cohn, D. H., Reardon, W., Byers, P. H., Lapiere, C. M., Prockop, D. J., and Nusgens, B. V. (1999). Human Ehlers-Danlos syndrome type VII C and bovine dermatosparaxis are caused by mutations in the procollagen I N-proteinase gene. Am. J. Hum. Genet. 65, 308–317. 7. Fernandes, R. J., Hirohata, S., Engle, J. M., Colige, A., Cohn, D. H., Eyre, D. R., and Apte, S. S. (2001). Procollagen II amino propeptide processing by ADAMTS-3. Insights on dermatosparaxis. J. Biol. Chem. 276, 31502–31509. 8. Colige, A., Vandenberghe, I., Thiry, M., Lambert, C. A., Van Beeumen, J., Li, S. W., Prockop, D. J., Lapiere, C. M., and Nusgens, B. V. (2002). Cloning and characterization of ADAMTS-14, a novel ADAMTS displaying high homology with ADAMTS-2 and ADAMTS-3. J. Biol. Chem. 277, 5756–5766. 9. Scott, I. C., Blitz, I. L., Pappano, W. N., Imamura, Y., Clark, T. G., Steiglitz, B. M., Thomas, C. L., Maas, S. A., Takahara, K., Cho, K. W., and Greenspan, D. S. (1999). Mammalian BMP-1/Tolloid-related metalloproteinases, including novel family member mammalian Tolloid-like 2, have differential enzymatic activities and distributions of expression relevant to patterning and skeletogenesis. Dev. Biol. 213, 283–300. 10. Ricard-Blum, S., Bernocco, S., Font, B., Moali, C., Eichenberger, D., Farjanel, J., Burchardt, E. R., van der Rest, M., Kessler, E., and Hulmes, D. J. (2002). Interaction properties of the procollagen C-proteinase enhancer protein shed light on the mechanism of stimulation of BMP-1. J. Biol. Chem. 277, 33864–33869. 11. Colige, A., Beschin, A., Samyn, B., Goebels, Y., Vanbeeumen, J., Nusgens, B., and Lapiere, C. (1995). Characterization and partial amino acid sequencing of a 107- kDa procollagen I N-proteinase purified by affinity chromatography on immobilized type XIV collagen. J. Biol. Chem. 270, 16724–16730. 12. Davies, D., Tuckwell, D. S., Calderwood, D. A., Weston, S. A., Takigawa, M., and Humphries, M. J. (1997). Molecular characterisation of integrin-procollagen C-propeptide interactions. Eur. J. Biochem. 246, 274–282. 13. Kadler, K. E., Hojima, Y., and Prockop, D. J. (1987). Assembly of collagen fibrils de novo by cleavage of the type I pC-collagen with procollagen C-proteinase. Assay of critical concentration demonstrates that collagen self-assembly is a classical example of an entropy-driven process. J. Biol. Chem. 262, 15696–15701. 14. Holmes, D. F. , Lowe, M. P., and Chapman, J. A. (1994). Vertebrate (chick) collagen fibrils formed in vivo can exhibit a reversal in molecular polarity. J. Mol. Biol. 235, 80–83. 15. Birk, D. E., Nurminskaya, M. V., and Zycband, E. I. (1995). Collagen fibrillogenesis in situ: fibril segments undergo post-depositional modifications resulting in linear and lateral growth during matrix development. Dev. Dyn. 202, 229–243. 16. Birk, D. E., Zycband, E. I., Woodruff, S., Winkelmann, D. A., and Trelstad, R. L. (1997). Collagen fibrillogenesis in situ: fibril segments become long fibrils as the developing tendon matures. Dev. Dyn. 208, 291–298. 17. Mellor, S. J., Atkins, G. L., and Hulmes, D. J. S. (1991). Developmental changes in the type I procollagen processing pathway in chick-embryo cornea. Biochem. J. 276, 777–784. 18. Hulmes, D. J. (1983). A possible mechanism for the regulation of collagen fibril diameter in vivo. Coll. Relat. Res. 3, 317–321. 19. Chapman, J. A. (1989). The regulation of size and form in the assembly of collagen fibrils in vivo. Biopolymers 28, 1367–1382. 20. Helseth, D. L., and Veis, A. (1981). Collagen self-assembly in vitro. Differentiating specific telopeptide-dependent interactions using selective enzyme modification and the addition of free amino telopeptide. J. Biol. Chem. 256, 7118–7128.
51 21. Kuznetsova, N., and Leikin, S. (1999). Does the triple helical domain of type I collagen encode molecular recognition and fiber assembly while telopeptides serve as catalytic domains? Effect of proteolytic cleavage on fibrillogenesis and on collagen-collagen interaction in fibers. J. Biol. Chem. 274, 36083–36088. 22. Malone, J. P., George, A., and Veis, A. (2004). Type I collagen N-telopeptides adopt an ordered structure when docked to their helix receptor during fibrillogenesis. Proteins 54, 206–215. 23. Hulmes, D. J. S., Wess, T. J., Prockop, D. J., and Fratzl, P. (1995). Radial packing, order, and disorder in collagen fibrils. Biophysi. J. 68, 1661–1670. 24. Orgel, J. P., Miller, A., Irving, T. C., Fischetti, R. F., Hammersley, A. P., and Wess, T. J. (2001). The in situ supermolecular structure of type I collagen. Structure (Camb) 9, 1061–1069. 25. Birk, D. E. (2001). Type V collagen: heterotypic type I/V collagen interactions in the regulation of fibril assembly. Micron. 32, 223–237. 26. Hansen, U., and Bruckner, P. (2003). Macromolecular specificity of collagen fibrillogenesis: fibrils of collagens I and XI contain a heterotypic alloyed core and a collagen I sheath. J. Biol. Chem. 278, 37352–37359. 27. Fleischmajer, R., MacDonald, E. D., Perlish, J. S., Burgeson, R. E., and Fisher, L. W. (1990). Dermal collagen fibrils are hybrids of type I and type III collagen molecules. J. Struct. Biol. 105, 162–169. 28. Liu, X., Wu, H., Bryne, M., Krane, S., and Jaenisch, R. (1997). Type III collagen is crucial for collagen I fibrillogenesis and for normal cardiovascular development. Proc. Natl Acad. Sci. USA 94, 1852–1856. 29. Linsenmayer, T., Gibney, E., Igoe, F., Gordon, M., Fitch, J., Fessler, L., and Birk, D. (1993). Type V collagen: molecular structure and fibrillar organization of the chicken alpha 1(V) NH2-terminal domain, a putative regulator of corneal fibrillogenesis. J. Cell. Biol. 121, 1181–1189. 30. Notbohm, H., Nokelainen, M., Myllyharju, J., Fietzek, P. P., Muller, P. K., and Kivirikko, K. I. (1999). Recombinant human type II collagens with low and high levels of hydroxylysine and its glycosylated forms show marked differences in fibrillogenesis in vitro. J. Biol. Chem. 274, 8988–8992. 31. Fernandes, R. J., Schmid, T. M., and Eyre, D. R. (2003). Assembly of collagen types II, IX and XI into nascent hetero-fibrils by a rat chondrocyte cell line. Eur. J. Biochem. 270, 3243–3250. 32. Nishiyama, T., McDonough, A. M., Bruns, R. R., and Burgeson, R. E. (1994). Type XII and XIV collagens mediate interactions between banded collagen fibers in vitro and may modulate extracellular matrix deformability. J. Biol. Chem. 269, 28193–28199. 33. Chiquet, M., Renedo, A. S., Huber, F., and Fluck, M. (2003). How do fibroblasts translate mechanical signals into changes in extracellular matrix production? Matrix Biol. 22, 73–80. 34. Jin, X., Iwasa, S., Okada, K., Ooi, A., Mitsui, K., and Mitsumata, M. (2003). Shear stress-induced collagen XII expression is associated with atherogenesis. Biochem. Biophys. Res. Commun. 308, 152–158. 35. Danielson, K. G., Baribault, H., Holmes, D. F., Graham, H., Kadler, K. E., and Iozzo, R. V. (1997). Targeted disruption of decorin leads to abnormal collagen fibril morphology and skin fragility. J. Cell Biol. 136, 729–743. 36. Weber, I. T., Harrison, R. W., and Iozzo, R. V. (1996). Model structure of decorin and implications for collagen fibrillogenesis. J. Biol. Chem. 271, 31767–31770. 37. Kuc, I. M., and Scott, P. G. (1997). Increased diameters of collagen fibrils precipitated in vitro in the presence of decorin from various connective tissues. Connect. Tissue. Res. 36, 287–296. 38. Scott, P. G., McEwan, P. A., Dodd, C. M., Bergmann, E. M., Bishop, P. N., and Bella, J. (2004). Crystal structure of the dimeric protein core of decorin, the archetypal small leucine-rich repeat proteoglycan. Proc. Natl Acad. Sci. USA 101, 15633–15638. 39. Molnar, J., Fong, K. S., He, Q. P., Hayashi, K., Kim, Y., Fong, S. F., Fogelgren, B., Szauter, K. M., Mink, M., and Csiszar, K. (2003). Structural and functional diversity of lysyl oxidase and the LOX-like proteins. Biochim. Biophys. Acta. 1647, 220–224.
52 40. Wang, S. X., Mure, M., Medzihradszky, K. F., Burlingame, A. L., Brown, D. E., Dooley, D. M., Smith, A. J., Kagan, H. M., and Klinman, J. P. (1996). A crosslinked cofactor in lysyl oxidase: redox function for amino acid side chains. Science 273, 1078–1084. 41. Siegel, R. C. (1979). Lysyl oxidase. Int. Rev. Connect. Tissue Res. 8, 73–118. 42. Fietzek, P. P., Allmann, H., Rauterberg, J., and Wachter, E. (1977). Ordering of cyanogen bromide peptides of type III collagen based on their homology to type I collagen: preservation of sites for crosslink formation during evolution. Proc. Natl Acad. Sci. USA 74, 84–86. 43. Atsawasuwan, P., Mochida, Y., Parisuthiman, D., and Yamauchi, M. (2005). Expression of lysyl oxidase isoforms in MC3T3-E1 osteoblastic cells. Biochem. Biophys. Res. Commun. 327, 1042–1046. 44. Hong, H. H., Pischon, N., Santana, R. B., Palamakumbura, A. H., Chase, H. B., Gantz, D., Guo, Y., Uzel, M. I., Ma, D., and Trackman, P. C. (2004). A role for lysyl oxidase regulation in the control of normal collagen deposition in differentiating osteoblast cultures. J. Cell. Physiol. 200, 53–62. 45. Trackman, P., Bedell, H. D., Tang, J., and Kagan, H. (1992). Posttranslational glycosylation and proteolytic processing of a lysyl oxidase precursor. J. Biol. Chem. 267, 8666–8671. 46. Boak, A. M., Roy, R., Berk, J., Taylor, L., Polgar, P., Goldstein, R. H., and Kagan, H. M. (1994). Regulation of lysyl oxidase expression in lung fibroblasts by transforming growth factor-beta 1 and prostaglandin E2. Am. J. Respir. Cell. Mol. Biol. 11, 751–755. 47. Kenyon, K., Contente, S., Trackman, P., Tang, J., Kagan, H., and Friedman, R. (1991). Lysyl Oxidase and rrg Messenger RNA. Science 253, 802. 48. Palamakumbura, A. H., Jeay, S., Guo, Y., Pischon, N., Sommer, P., Sonenshein, G. E., and Trackman, P. C. (2004). The propeptide domain of lysyl oxidase induces phenotypic reversion of ras-transformed cells. J. Biol. Chem. 279, 40593–40600. 49. Royce, P. M., and Barnes, M. J. (1985). Failure of highly purified lysyl hydroxylase to hydroxylate lysyl residues in the non-helical regions of collagen. Biochem. J. 230, 475–480. 50. Mercer, D. K., Nicol, P. F., Kimbembe, C., and Robins, S. P. (2003). Identification, expression, and tissue distribution of the three rat lysyl hydroxylase isoforms. Biochem. Biophys. Res. Commun. 307, 803–809. 51. van der Slot, A. J., Zuurmond, A. M., Bardoel, A. F., Wijmenga, C., Pruijs, H. E., Sillence, D. O., Brinckmann, J., Abraham, D. J., Black, C. M., Verzijl, N., De Groot, J., Hanemaaijer, R., Tekoppele, J. M., Huizinga, T. W., and Bank, R. A. (2003). Identification of PLOD2 as telopeptide lysyl hydroxylase, an important enzyme in fibrosis. J. Biol. Chem. 278, 40967–40972. 52. Valtavaara, M., Szpirer, C., Szpirer, J., and Myllyla, R. (1998). Primary structure, tissue distribution, and chromosomal localization of a novel isoform of lysyl hydroxylase (lysyl hydroxylase 3). J. Biol. Chem. 273, 12881–12886. 53. Yeowell, H. N., and Walker, L. C. (1999). Tissue specificity of a new splice form of the human lysyl hydroxylase 2 gene. Matrix Biol. 18, 179–187. 54. Bank, R. A., Robins, S. P., Wijmenga, C., Breslau-Siderius, L. J., Bardoel, A. F., van der Sluijs, H. A., Pruijs, H. E., and TeKoppele, J. M. (1999). Defective collagen crosslinking in bone, but not in ligament or cartilage, in Bruck syndrome: indications for a bone-specific telopeptide lysyl hydroxylase on chromosome 17. Proc. Natl Acad. Sci. USA 96, 1054–1058. 55. Ha-Vinh, R., Alanay, Y., Bank, R. A., Campos-Xavier, A. B., Zankl, A., Superti-Furga, A., and Bonafe, L. (2004). Phenotypic and molecular characterization of Bruck syndrome (osteogenesis imperfecta with contractures of the large joints) caused by a recessive mutation in PLOD2. Am. J. Med. Genet. A 131A, 115–120. 56. Pornprasertsuk, S., Duarte, W. R., Mochida, Y., and Yamauchi, M. (2004). Lysyl hydroxylase-2b directs collagen cross-linking pathways in MC3T3-E1 cells. J. Bone Miner. Res. 19, 1349–1355.
Simon P. Robins 57. Walker, L. C., Overstreet, M. A., and Yeowell, H. N. (2005). Tissuespecific expression and regulation of the alternatively-spliced forms of lysyl hydroxylase 2 (LH2) in human kidney cells and skin fibroblasts. Matrix Biol. 23, 515–523. 58. Pornprasertsuk, S., Duarte, W. R., Mochida, Y., and Yamauchi, M. (2005). Overexpression of lysyl hydroxylase-2b leads to defective collagen fibrillogenesis and matrix mineralization. J. Bone Miner. Res. 20, 81–87. 59. Bailey, A. J., Robins, S. P., and Balian, G. (1974). Biological significance of the intermolecular crosslinks of collagen. Nature 251, 105–109. 60. Robins, S. P., and Bailey, A. J. (1975). The mechanism of stabilization of the reducible intermediate crosslinks. Biochem. J. 149, 381–385. 61. Tanzer, M. L., Housley, T., Berube, L., Fairweather, R., Franzblau, C., and Gallop, P. M. (1973). Structure of two histidine-containing crosslinks from collagen. J. Biol. Chem. 248, 393–402. 62. Robins, S. P., and Bailey, A. J. (1973). The characterisation of Fraction C, a possible artifact produced during the reduction of collagen fibres with borohydride. Biochem. J. 135, 657–665. 63. Bernstein, P. H., and Mechanic, G. L. (1980). A natural histidine-based imminium cross-link in collagen and its location. J. Biol. Chem. 255, 10414–10422. 64. Eriksen, H. A., Sharp, C. A., Robins, S. P., Sassi, M. L., Risteli, L., and Risteli, J. (2004). Differently cross-linked and uncross-linked carboxyterminal telopeptides of type I collagen in human mineralised bone. Bone 34, 720–727. 65. Fujimoto, D., Moriguchi, T., Ishida, T., and Hayashi, H. (1978). The structure of pyridinoline, a collagen crosslink. Biochem. Biophys. Res. Commun. 84, 52–57. 66. Ogawa, T., Ono, T., Tsuda, M., and Kawanashi, Y. (1982). A novel fluor in insoluble collagen: a crosslinking molecule in collagen molecule. Biochem. Biophys. Res. Commun. 107, 1252–1257. 67. Eyre, D. R. (1984). Cross-linking in collagen and elastin. Ann. Rev. Biochem. 53, 717–748. 68. Eyre, D. R., and Oguchi, H. (1980). Hydroxypyridinium crosslinks of skeletal collagens: their measurement, properties and a proposed pathway of formation. Biochem. Biophys. Res. Commun. 92, 403–410. 69. Robins, S. P. (1983). Crosslinking of collagen: isolation, structural characterization and glycosylation of pyridinoline. Biochem. J. 215, 167–173. 70. Scott, J. E., Hughes, E. W., and Shuttleworth, A. (1981). A collagenassociated Ehrlich chromogen: a pyrrolic cross-link? Biosci. Rep. 1, 611–618. 71. Kuypers, R., Tyler, M., Kurth, L. B., Jenkins, I. D., and Horgan, D. J. (1992). Identification of the loci of the collagen-associated Ehrlich chromogen in type I collagen confirms its role as a trivalent cross-link. Biochem. J. 283, 129–136. 72. Brady, J. D., and Robins, S. P. (2001). Structural characterization of pyrrolic cross-links in collagen using a biotinylated Ehrlich’s reagent. J. Biol. Chem. 276, 18812–18818. 73. Hanson, D., and Eyre, D. (1996). Molecular site specificity of pyridinoline and pyrrole cross- links in type I collagen of human bone. J. Biol. Chem. 271, 26508–26516. 74. Reiser, K. M., and Last, J. A. (1986). Biosynthesis of collagen crosslinks: in vivo labelling of neonatal skin, tendon, and bone in rats. Connect. Tissue Res. 14, 293–306. 75. Otsubo, K., Katz, E. P., Mechanic, G. L., and Yamauchi, M. (1992). Cross-linking connectivity in bone collagen fibrils: the COOH-terminal locus of free aldehyde. Biochemistry (Mosc) 31, 396–402. 76. Knott, L., Tarlton, J. F., and Bailey, A. J. (1997). Chemistry of collagen cross-linking: biochemical changes in collagen during the partial mineralization of turkey leg tendon. Biochem. J. 322, 535–542. 77. Eyre, D. R., Dickson, I. R., and VanNess, K. P. (1988). Collagen crosslinking in human bone and cartilage: age-related changes in the content of mature hydroxypyridinium residues. Biochem. J. 252, 495–500.
Chapter 2 Fibrillogenesis and Maturation of Collagens 78. Knott, L., Whitehead, C. C., Fleming, R. H., and Bailey, A. J. (1995). Biochemical changes in the collagenous matrix of osteoporotic avian bone. Biochem. J. 310, 1045–1051. 79. Eyre, D. R. (1981). Crosslink maturation in bone collagen. Dev. Biochem. 22, 51–55. 80. Traub, W., Arad, T., and Weiner, S. (1989). Three-dimensional ordered distribution of crystals in turkey tendon collagen fibers. Proc. Natl Acad. Sci. USA 86, 9822–9826. 81. Landis, W. J., and Silver, F. H. (2002). The structure and function of normally mineralizing avian tendons. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 133, 1135–1157. 82. Landis, W. J., Kraus, B. L., and Kirker-Head, C. A. (2002). Vascularmineral spatial correlation in the calcifying turkey leg tendon. Connect. Tissue Res. 43, 595–605. 83. Yamauchi, M., and Katz, E. (1993). The post-translational chemistry and molecular packing of mineralizing tendon collagens. Connect. Tissue Res. 29, 81–98. 84. Wassen, M. H., Lammens, J., Tekoppele, J. M., Sakkers, R. J., Liu, Z., Verbout, A. J., and Bank, R. A. (2000). Collagen structure regulates fibril mineralization in osteogenesis as revealed by cross-link patterns in calcifying callus. J. Bone Miner. Res. 15, 1776–1785. 85. Bailey, A. J., Light, N. D., and Atkins, E. D. (1980). Chemical crosslinking restrictions on models for the molecular organization of the collagen fibre. Nature 288, 408–410. 86. Hulmes, D. J., Holmes, D. F., and Cummings, C. (1985). Crystalline regions in collagen fibrils. J. Mol. Biol. 184, 473–477. 87. Wess, T. J. (2005). Collagen fibril form and function. Adv. Prot. Chem. 70, 341–374. 88. Landis, W., Hodgens, K., Song, M., Arena, J., Kiyonaga, S., Marko, M., Owen, C., and McEwen, B. (1996). Mineralization of collagen may occur on fibril surfaces: Evidence from conventional and high-voltage electron microscopy and three-dimensional imaging. J. Struct. Biol. 117, 24–35. 89. Paschalis, E. P., Verdelis, K., Doty, S. B., Boskey, A. L., Mendelsohn, R., and Yamauchi, M. (2001). Spectroscopic characterization of collagen cross-links in bone. J. Bone Miner. Res. 16, 1821–1828. 90. Blank, R. D., Baldini, T. H., Kaufman, M., Bailey, S., Gupta, R., Yershov, Y., Boskey, A. L., Coppersmith, S. N., Demant, P., and Paschalis, E. P. (2003). Spectroscopically determined collagen Pyr/deH-DHLNL cross-link ratio and crystallinity indices differ markedly in recombinant congenic mice with divergent calculated bone tissue strength. Connect. Tissue Res. 44, 134–142. 91. Huang, R. Y., Miller, L. M., Carlson, C. S., and Chance, M. R. (2003). In situ chemistry of osteoporosis revealed by synchrotron infrared microspectroscopy. Bone 33, 514–521. 92. Paschalis, E. P., Recker, R., DiCarlo, E., Doty, S. B., Atti, E., and Boskey, A. L. (2003). Distribution of collagen cross-links in normal human trabecular bone. J. Bone Miner. Res. 18, 1942–1946. 93. Paschalis, E. P., Shane, E., Lyritis, G., Skarantavos, G., Mendelsohn, R., and Boskey, A. L. (2004). Bone fragility and collagen cross-links. J. Bone Miner. Res. 19, 2000–2004.
53 94. Levene, C. I., and Gross, J. (1959). Alterations in state of molecular aggregation of collagen induced in chick embryos by beta-aminopropionitrile (lathyrus factor). J. Exp. Med. 110, 771–790. 95. Vater, C. A., Harris, E. D. J., and Siegel, R. C. (1979). Native crosslinks in collagen fibrils induce resistance to human synovial collagenase. Biochem. J. 181, 639–645. 96. Rucklidge, G. J., Riddoch, G. I., and Robins, S. P. (1986). Immunocytochemical staining of rat tissues with antibodies to denatured type I collagen: a technique for localizing areas of collagen degradation. Collagen Rel. Res. 6, 185–193. 97. Dodge, G. R., and Poole, A. (1989). Immunohistochemical detection and immunochemical analysis of type II collagen degradation in human normal, rheumatoid, and osteoarthritic articular cartilages and in explants of bovine articular cartilage cultured with interleukin 1. J. Clin. Investi. 83, 647–661. 98. Lees, S., Eyre, D. R., and Barnard, S. M. (1990). BAPN dose dependence of mature crosslinking in bone matrix collagen of rabbit compact bone: corresponding sonic velocity and equatorial diffraction spacing. Connect. Tissue Res. 24, 95–105. 99. Oxlund, H., Barckman, M., Ortoft, G., and Andreassen, T. T. (1995). Reduced concentrations of collagen cross-links are associated with reduced strength of bone. Bone 17, 365S–371S. 100. Horgan, D. J., King, N. L., Kurth, L. B., and Kuypers, R. (1990). Collagen crosslinks and their relationship to the thermal properties of calf tendons. Arch. Biochem. Biophys. 281, 21–26. 101. Wang, X., Shen, X., Li, X., and Agrawal, C. M. (2002). Age-related changes in the collagen network and toughness of bone. Bone 31, 1–7. 102. Vashishth, D., Gibson, G. J., Khoury, J. I., Schaffler, M. B., Kimura, J., and Fyhrie, D. P. (2001). Influence of nonenzymatic glycation on biomechanical properties of cortical bone. Bone 28, 195–201. 103. Bätge, B., Winter, C., Notbohm, H., Acil, Y., Brinckmann, J., and Muller, P. K. (1997). Glycosylation of human bone collagen in relation to lysylhydroxylation and fibril diameter. J. Biochem. (Tokyo) 122, 109–115. 104. Fledelius, C., Johnsen, A. H., Cloos, P. A. C., Bonde, M., and Qvist, P. (1997). Characterization of urinary degradation products derived from type I collagen. Identification of a beta-isomerized Asp-Gly sequence within the C-terminal telopeptide (α1) region. J. Biol. Chem. 272, 9755–9763. 105. Brady, J. D., Ju, J., and Robins, S. P. (1999). Isoaspartyl bond formation within n-terminal sequences of collagen type I: implications for their use as markers of collagen degradation. Clin. Sci. 96, 209–215. 106. Garnero, P., Cloos, P., Sornay-Rendu, E., Qvist, P., and Delmas, P. D. (2002). Type I collagen racemization and isomerization and the risk of fracture in postmenopausal women: the OFELY prospective study. J. Bone Miner. Res. 17, 826–833.
This Page Intentionally Left Blank
Chapter 3
Vitamin K Dependent Proteins of Bone and Cartilage Caren M. Gundberg Ph.D.
Department of Orthopaedics and Rehabilitation, Yale University School of Medicine, New Haven, Connecticut 06510
Satoru K. Nishimoto Ph.D.
Department of Molecular Sciences, The University of Tennessee College of Medicine, Memphis, Tennesse 38163
V. Gas6 VI. Vitamin K/Warfarin References
I. Abstract II. Introduction III. Osteocalcin IV. Matrix Gla Protein
I. ABSTRACT
and cellular functions. The formation of gammacarboxyglutamic acid (Gla) occurs via a unique posttranslational modification of specific peptide-bound glutamate residues in selective proteins. This reaction, which is required for the biological activities of all the vitamin K-dependent proteins, is inhibited by warfarin (Fig. 1). Beside the vitamin K-dependent coagulation factors, prothrombin, and factors VII, IX, X, Gla-containing proteins include gamma-carboxylase, Protein C and Protein S anticoagulation factors and Gas6 which is involved in cell proliferation. Gla-containing proteins of unknown function include Protein Z, a plasma glycoprotein, and Proline Rich Gla Protein 1, Proline Rich Gla Protein 2, Transmembrane Gla Protein 3, and Transmembrane Gla Protein 4 [1–5]. The last four proteins were identified as transmembrane Gla Proteins and are widely expressed in many tissue types [6, 7]. In addition, either carboxylase activity or vitamin K-dependent formation of Gla residues have been observed in a wide variety of other tissues, including bone, kidney, intestine, placenta, pancreas, skin,
The importance of vitamin K for the normal functioning of blood coagulation factors is well known. However, this vitamin is also involved in the physiological activation of various proteins that are not involved in hemostasis. Here we focus on the biochemistry and physiology of osteocalcin and Matrix Gla Protein (MGP), which are abundant in bone and cartilage, respectively. While osteocalcin biosynthesis is restricted to bone and dentin, MGP is synthesized in a variety of tissues and cell types. Two other vitamin K-dependent proteins, Gas6, which is involved in the regulation of cell proliferation and Protein S, an inhibitor of coagulation, have been shown to have a potential role in bone metabolism.
II. INTRODUCTION The family of vitamin K-dependent gammacarboxylated proteins are important in a variety of tissue Dynamics of Bone and Cartilage Metabolism
55
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
56
Caren M. Gundberg and Satoru K. Nishimoto
Figure 1 Gamma-carboxyglutamic acid (Gla). Vitamin K and CO2 are required for the carboxylation of specific glutamic acid residues in osteocalcin. This reaction is inhibited by warfarin. The adjacent carboxyl groups of Gla are binding sites for Ca2+.
spleen, lung, heart, blood vessels, spinal cord, and testis. Accumulation of Gla-containing proteins in sites of pathological ectopic calcification have been observed [2–4]. Bone and cartilage Gla proteins are osteocalcin and Matrix Gla Protein (MGP), which are small extracellular Ca2+-binding proteins. Osteocalcin is exclusive to bone and dentin. MGP is found in highest concentrations in calcified cartilage, but MGP occurs in a broad variety of tissues [8–10]. MGP plays an important role in regulating cartilage and vascular calcification [11–14]. Osteocalcin and MGP have been grouped in the same gene family because they share significant sequence similarity. In addition two other vitamin K-dependent proteins, Gas6 and Protein S, have been identified in bone [15, 16].
III. OSTEOCALCIN A. Structure and Biosynthesis Osteocalcin is a small protein (49 amino acids) synthesized by mature osteoblasts, odontoblasts, and hypertrophic chondrocytes. Osteocalcin is characterized by the presence of three residues of Gla and in all species the Gla-containing portion of the molecule is strongly
conserved [17, 18]. In human bone, however, the first potential Gla residue at position 17 is only partially carboxylated (55–89%) while residues 21 and 24 are greater than 90% gamma-carboxylated, likely due to partial carboxylation of the protein during synthesis [19]. In all species, the protein is synthesized as an 11-kDa molecule consisting of a 23 residue hydrophobic signal peptide, a 26 residue propeptide, and the 49 residue mature protein. After the hydrophobic region is cleaved by a signal peptidase, pro-osteocalcin is gamma-carboxylated. The pro-region contains a gamma-carboxylation recognition site homologous to corresponding regions in the vitamin K-dependent clotting factors (Fig. 2). Arg at position -1 and Phe at -16 are strictly conserved and appear to be critical for the binding of the vitamin K-dependent carboxylase enzyme to its substrates [19–22]. However, recent studies show that the carboxylase has a 2 × 105-fold weaker affinity for the osteocalcin propeptide compared to most of the other propeptide factors and an attached propeptide is not required for carboxylation of osteocalcin. Rather, osteocalcin contains a high-affinity carboxylase recognition site within its primary sequence [23]. After carboxylation, the propeptide is removed and the mature protein is secreted [24]. Combined chemical, immunochemical, and spectral investigations have provided a model of osteocalcin structure [25]. A detailed three-dimensional structure for osteocalcin has been determined by 1H 2D NMR and X-ray crystallography (Fig. 3). Apo-osteocalcin is relatively unstructured except for a disulfide bond between Cys23–Cys29. Calcium addition induces a folded structure containing several helical regions, a C-terminal hydrophobic core and an unstructured N-terminus from residues 1–15. The structured region spans residues 16–49 and contains three regions of helical secondary structure. All three Gla residues are found in the first helical region and a disulfide bond between Cys23 and Cys29 separates this and the second helical region (residues 27–35).
Figure 2 Comparison of the primary sequences of human osteocalcin (bottom) and Human MGP (top). Bold symbols indicate conserved regions; E’ indicate γ-carboxyglutamic acid; italics indicate phosphoserines. Processing of the 84 amino acid protein yields the 1–77 form isolated from human bone. A γ-carboxylation recognition site is homologous to corresponding regions in the vitamin K-dependent clotting factors. Arg (at position −1 in OC and +30 in MGP) and Phe (at −16 in OC and +15 in MGP) are strictly conserved and appear to be critical for the binding of the vitamin K-dependent carboxylase enzyme to its substrates. The ala-gly substitution at −10 osteocalcin results in reduced affinity for the vitamin K-carboxylase enzyme.
Chapter 3 Vitamin K Dependent Proteins of Bone and Cartilage
Figure 3 Native osteocalcin contain several helical regions, a C-terminal hydrophobic core and an unstructured N-terminus from residues 1–15. The structured region spans residues 16–49 and contains three regions of helical secondary structure. All three Gla residues are found in the first helical region and a disulfide bond between Cys23 and Cys29 separates this and the second helical region (residues 27–35). Residues 41–44 comprise the third helical region. Hydrogen bonds within these three regions serve to stabilize the helical regions forming a tight globular structure.
Residues 41–44 comprise the third helical region. Hydrogen bonds within these three regions serve to stabilize the helical regions forming a tight globular structure. The Gla residues which coordinate calcium are on the same face of the helical region and are surface exposed. These can interact with the intercalcium spacings in the hydroxyapatite lattice. The C-terminus extends outward and would be accessible to neighboring cells as well as endogenous proteinases [26, 27]. This structure is consistent with many reports of osteocalcin C-terminal peptides having chemotactic activity to osteoclast precursor cells [28–30]. Likewise, consistent with an unstructured N-terminus, the serine proteases, cathepsin D, L, and H degrade osteocalcin at the 7–8 bond [31].
B. Function Although osteocalcin is one of the most abundant noncollagenous proteins in bone, the biological function of osteocalcin has not been precisely defined. Because of its specific interaction with hydroxyapatite, osteocalcin was thought to affect the growth or maturation of Ca2+-phosphate mineral phases. Osteocalcin first appears coincident with the onset of mineralization and an increase in synthesis of the protein occurs in concert with hydroxyapatite deposition during skeletal growth [32, 33]. Immunolocalization studies show that the protein is distributed throughout the mineralized regions of bone matrix, dentin, and calcified cartilage [34–36] . However, an accumulation of evidence indicates that osteocalcin is not related to events which allow mineral deposition to occur but rather that it participates in the regulation of mineralization or bone turnover. Early studies showed that the
57
protein could regulate growth of hydroxyapatite crystals in solution [37]. In vitro studies demonstrate that osteocalcin is a marker of late osteoblast differentiation [38–40]. When osteoblasts are grown in culture they produce a dense collagenous extracellular matrix [41, 42]. Alkaline phosphatase activity and MGP expression increase post-proliferatively coincident with the onset of mineralization. As the extracellular matrix accumulates and mineralizes, synthesis of osteocalcin and other calcium-binding proteins, e.g. osteopontin and bone sialoprotein, is initiated. Alkaline phosphatase activity subsequently decreases, but osteocalcin synthesis remains high throughout the life of the culture [43, 44]. In cultures in which mineralization is delayed, expression of osteocalcin is also delayed [38]. Osteocalcin-null mice are characterized by a progressive increase in bone mass, leading to bone of better biomechanical quality by 6–9 months. These mice exhibit an accelerated rate of bone formation without changes in osteoclast or osteoblast number [45]. The gradual appearance of this phenotype is consistent with the fact that osteocalcin content of bone is low in newborn rats (less than 4% of the adult level) and increases gradually with age [46]. No changes in mineral content of the bones of osteocalcindepleted mice were detectable by von Kossa staining or histomorphometry. However, a more sensitive assay of mineralization, Fourier transform infrared microspectroscopy, revealed differences in the size and perfection of the crystallite [47]. In wild-type animals the crystals were larger and more perfect in the cortical bone than in trabecular bone. In contrast, in the osteocalcin knockout animals, the crystal size and perfection were the same in both the trabecular and cortical bone and resembled that of the wild-type trabecular bone. These findings are consistent with impaired mineral maturation in the osteocalcin-deficient bone and implies the presence of newer (less remodeled) mineral. Similar findings were observed when ovariectomized wild-type and knockout mice were compared. Based on these findings, it appears that osteocalcin plays some as yet undefined role in regulating bone mineral turnover. Other in vivo and in vitro studies support a role for osteocalcin in bone remodeling. First, disrupted collagen fibrillogenesis in the cloned mouse calvarial cell line MCT3T3-E1 results in increased turnover of the collagenous matrix, a decrease in alkaline phosphatase but a fivefold increase in osteocalcin biosynthesis [48]. Second, the pattern of osteocalcin distribution in human osteons changes with gender and age, and localized reductions of osteocalcin in the extracellular matrix are associated with reduced cortical remodeling [49]. Several earlier studies have suggested that osteocalcin is involved in
58
Caren M. Gundberg and Satoru K. Nishimoto
recruitment and activation of bone resorbing cells. The protein is a chemoattractant for peripheral mononuclear cells and osteoclast-like cells of giant cell tumors [28–30]. Incorporation of osteocalcin into synthetic hydroxyapatite particles placed onto chick chorioallantoic membrane enhanced osteoclast formation [50]. Subcutaneously implanted bone particles which were osteocalcin-deficient show a decrease in both progenitor cell recruitment and resorption of the bone particles [51]. It is possible that osteocalcin may act in combination with other hydroxyapatite binding proteins as a signal for bone remodeling. Osteopontin, which potentiates osteoclast adhesion to mineral surfaces, forms a complex with osteocalcin in vitro [52, 53]. Altered expression of both osteocalcin and osteopontin have been reported in rat models of osteopetrosis in which osteoclastic resorption is impaired. Further studies in animals depleted of combinations of matrix proteins should provide clues to the function of osteocalcin and other bone-specific proteins.
C. Gene Structure and Regulation The human osteocalcin gene is a single-copy gene located at the distal long arm of chromosome 1 [54]. The structure and sequence of the gene in various species are similar [55] but multiple copies of the gene occur in the rat and mouse [56, 57]. In the rat, one or more copies of the gene have been found, depending upon the strain. The mouse osteocalcin cluster contains three genes, two of which encode osteocalcin and are expressed only in bone (OG1 and OG2), while a third gene, the osteocalcinrelated gene (ORG) is found in low levels in brain, lung, and kidney, but not in bone. ORG is expressed at day 10 in embryonic development, earlier than OG1 and OG2, which are expressed at day 17, the beginning of osteogenesis. The coding sequence of ORG carries five amino acid substitutions, one at the propeptide cleavage site. ORG also contains an additional exon that is not translated and a 3-kb insertion separates the ORG coding sequence from a bone-specific promoter. The insertion sequence has the structure of a typical retrovirus, an attribute which leads to the down-regulation of transcription, possibly explaining the low levels of expression of ORG in nonosseous tissues [57]. Analysis of the OG2 promoter revealed that a 147-bp fragment contains two osteoblast cis-acting elements, OSE1 and OSE2. The OSE2-binding protein is Runx2, a master regulator of osteoblast differentiation [58]. The OSE1-binding protein is ATF4, also known as cAMPresponse element protein 2, which belongs to the subfamily of cAMP-response element-binding protein/ATF
basic leucine zipper proteins [59]. ATF4, like Runx2 and Osterix, another known osteoblast-specific transcription factor, can induce osteoblast-specific gene expression in nonosteoblastic cells and all three appear to be important factors involved in the regulation of temporal expression of osteoblastic genes [60]. Osteocalcin expression is regulated by various hormones and growth factors [reviewed in reference 61]. A specific vitamin D regulatory element has been identified in the osteocalcin promoter [62–65]. In most species 1,25(OH)2D up-regulates osteocalcin biosynthesis; however, in the mouse osteocalcin synthesis is suppressed by 1,25(OH)2D [66, 67]. Stimulation of osteocalcin by 1,25(OH)2D in chicks is increased when cultures are derived from 12-dayold embryos, but decreased in 17-day-old embryos [68]. There is a significant nucleotide difference in the VDRE of the mouse gene and the lack of inducibility of mouse osteocalcin is likely attributable to the inability of the VDR to bind and transactivate the mouse promoter [69]. This vitamin D response element is also the mediator of positive regulation in response to retinoic acid [70]. Separate promoter regions are responsible for repression by glucocorticoid and stimulation by TGFβ. On the other hand, cAMP influences osteocalcin gene expression by modulating mRNA stability [71]. Some effectors, e.g. γ-interferon, IGF I or II, TNFα, and some metal ions indirectly regulate osteocalcin through vitamin D responsiveness [72]. Osteocalcin mRNA levels are also altered by estrogen, thyroxine, calcitonin, prostaglandin E2, or BMPs, but the specific mechanisms have not been defined and may be indirect [61]. Osteocalcin mRNA has also been detected in bone marrow megakaryocytes and peripheral blood platelets, liver, lung, and brain by reverse transcription-polymerase chain reaction but translation into protein has not been verified [73, 74]. Like the mouse ORG gene, the mRNA levels observed in these tissues are three orders of magnitude lower than found in bone. Furthermore, they are not vitamin D-responsive, suggesting that control of expression is under a nonskeletal promoter. To date, bone and dentine are the only known sites of production of osteocalcin.
D. Catabolism While osteocalcin is primarily deposited in the extracellular matrix of bone, a small amount enters the blood. The circulating levels of osteocalcin are taken as a specific marker of osteoblastic activity [75; see also Chapter 34]. The catabolic fate of osteocalcin is of interest because of the use of circulating osteocalcin as a marker of bone formation in metabolic bone disease. Various fragments
59
Chapter 3 Vitamin K Dependent Proteins of Bone and Cartilage
of osteocalcin are known to circulate and these have been detected by antibodies originally made against the intact molecule. It is thought that the majority of circulating osteocalcin comprises the intact molecule and a large N-terminal mid-molecule fragment [76, 77]. Although the large N-terminal mid-molecule fragment has not been directly sequenced, it is thought to encompass residues 1–43. This is based on the fact that osteocalcin contains a tryptic site at residue 43, and the two antibodies used to characterize this fragment are specific for residues 5–13 and for the 25–37 region. It has been suggested that the large N-terminal midmolecule fragment is generated by proteolysis in the circulation or during sample processing and storage. This is primarily based on the findings that: (1) the fragment is detected immediately after blood sampling; and (2) osteocalcin levels decrease with incubation at room temperature when measured by conventional RIA or by intact assays, but values are stable with an assay that recognizes both the intact and large N-terminal mid-molecule fragment [76, 78]. However, the large N-terminal mid-molecule fragment was also detected in conditioned media from human osteoblast-like cells [76]. Whether this fragment was derived from proteolysis in the media or by intracellular processing has not been established. Studies by Taylor et al. [79] suggest that a unique fragment found in sera from patients with Paget’s disease may have been derived from altered osteoblastic synthesis. It is possible that osteoblastic degradation of osteocalcin may serve as a mechanism to regulate its own concentration. Recent data indicate that there are in fact other smaller (80 members) of calcium-dependent proteins that play prominent roles in morphogenesis and the maintenance of adhesive contacts in solid tissues. They are divided into five subsets of receptors: the classical cadherin types I and II, the latter
133
Chapter 8 Integrins and Other Adhesion Molecules
directly linked to the actin cytoskeleton, desmocollins and desmogleins, protocadherins, and a number of other, more distantly related cadherins [8, 9]. Cadherins are calciumdependent single-chain single-pass transmembrane glycoproteins that share “cadherin domains”, which contain specific amino acid motifs, which have essential roles in calcium binding and dimerization of the receptors [8, 10]. In the context of this review the classical cadherins are the most important subgroup. The type I cadherins (members E-, N-, M-, P-, and R-cadherin) all share a common structure and have highest homology in the short cytoplasmic domain. They have an extracellular domain containing five repeats of around 110 amino acids which contain the negatively charged, calcium-binding motifs and conserved cysteine residues in the fifth repeat. They have molecular weights of around 100–130 kDa. The ligand-binding site of type I cadherins is the conserved HAV motif located at the N-terminal region of the molecule in the first conserved extracellular repeat. Type II cadherins (cadherin 5–12) share the basic cadherin structure, but have lower amino acid homology with type I cadherins. They also lack the cell adhesion HAV motif [8]. Cadherins are mediators of cell–cell adhesion and bind ligand mainly in a homophilic manner, although heterophilic binding between different cadherin molecules is possible, but appears restricted to cadherins from the same class [9]. As with integrins, cells often express a repertoire of different cadherins simultaneously, although a particular feature of cadherins is their restricted expression at specific stages of embryonic and cellular development. On the cell surface, cadherins tend to be concentrated at cell– cell junctions. The cytoplasmic tail of cadherins binds to catenins and this complex is important in gene transcription as well as regulation of adhesion. β Catenin is also the key player in the Wnt signaling pathway, a pathway that has recently been found to be important in the regulation of bone mass [11] and current work is trying to unravel interconnections between Wnt signaling and cadherinmediated cell adhesion through regulation of free β catenin levels in cells [12]. Signal transduction through cadherins is complex, dependent upon cell type and, in addition to phosphorylation of β-catenin, can include activation of small GTPases and tyrosine kinase pathways [13].
C. Immunoglobulin Superfamily The immunoglobulin (Ig) family of receptors all share a basic motif consisting of an Ig fold of between 70 and 110 amino acids (reviewed in [1, 14]) organized into two antiparallel β sheets which seem to serve as a scaffold on which unique determinants can be displayed. There is
considerable variation in the primary structure of the members of this family and hence in molecular weights, but their tertiary structure is well conserved. There are now over 700 human genes known which share Ig motifs, making this one of the largest superfamilies in the human genome [15]. Many of these are splice variants of cell adhesion molecules and Ig genes, but overall the cell adhesion molecules constitute a large part of this superfamily. Here we are concerned only with the cell adhesion molecules. Their functions are wide ranging with some members functioning as true signal-transducing receptors, whereas others have predominantly adhesive functions. Ligands for Ig family members include other Ig family members (identical, as well as nonidentical members), such as NCAM binding to itself, but also members of the integrin family (such as for the ICAMs, which bind β2 integrins) and components of the extracellular matrix, for example collagen binding for myelin-associated glycoprotein, a neuron-expressed Ig family member [1]. Signaling pathways activated by Ig family members include MAP kinase pathways [1, 16].
D. Syndecans Syndecans are a family of cell surface proteoglycans, varying in size from 20–45kDa. They are type I transmembrane proteins characterized by a shared heparan sulfate attachment sequence on the N-terminus of their single polypeptide chain. There is little homology in the rest of their extracellular domain, but their single transmembrane domain and short cytoplasmic domains are highly conserved. The cytoplasmic domain contains four conserved tyrosine residues, suggesting that phosphorylation of this domain could occur and be involved in signal transduction events. However, syndecans are thought to function predominantly as co-receptors for other receptors such as integrins, and members of the fibroblast growth factor family and vascular endothelial cell growth factor, which need heparan sulfate for signaling. In such situations the signaling is thought to occur via the cytoplasmic domains of associated receptors rather than the syndecan molecule [17]. There are four mammalian syndecans known, with syndecan-1 best studied so far. Syndecan-1 is the major syndecan of epithelia and can function as a cell–matrix receptor binding various matrix proteins (type I collagen, fibronectin, tenascin), and in addition can bind members of the FGF family. It appears that in different cell types syndecan-1 can have different patterns of glycosaminoglycans attached to its core protein and this influences the ligand-binding capabilities. Thus, where in one cell type syndecan-1 may contain heparan sulfate, as well as
134
M. H. HELFRICH AND M. A. HORTON
chondroitin sulfate side chains and bind collagen, this may not be the case in another cell type in which it has heparan sulfate side chains only. Syndecan-2 is the main form in mesenchymal cells and is present in neuronal tissues alongside syndecan-3. Syndecan-4 is present in many cells that form stable adhesions and in vitro is consistently present in focal adhesions [18]. The functions of syndecans are varied with roles in cell–matrix interaction and cell proliferation. Signal transduction, in addition to associated tyrosine kinsases, may also include cytoskeletal proteins through association with actin and tubulin.
E. CD44 CD44, also known as the hyaluronate receptor, is a family of transmembrane glycoproteins with molecular weights of 85–250 kDa. They share an N-terminal region that is related to the cartilage proteoglycan core and link proteins. Alternative splicing of ten exons and extensive post-translational modification such as glycosylation and addition of chondroitin sulfate produces the wide variety of CD44 proteins. Chondroitin sulfate containing variants can bind fibronectin, laminin, and collagen in addition to hyaluronate, and CD44 also binds to osteopontin [19]. Finally, CD44 may also bind homotypically. CD44 therefore functions in a variety of ways including in cell–cell interaction (homing of lymphocytes or cell clustering), but also in cell–matrix adhesion. Malignant transformation of cells leads to up-regulation of CD44 expression and metastatic tumors often express an altered repertoire of CD44 variants. More recently evidence has pointed to a more complex role of CD44 than simply as a cell–cell or cell–matrix adhesion molecule. It is becoming apparent that CD44 can be enzymatically cleaved and that functional fragments from the cytoplasmic domain can act as transcriptional regulators, whereas functional fragments from the ectodomain circulate in body fluids or can become incorporated in the matrix and thereby modulate cellular behavior [20].
74–240 kDa, with differently glycosylated forms expressed in different cell types. In general, the function of selectins is in leukocyte trafficking. They are thought to be involved in the earliest stages of leukocyte extravasation, where binding of the selectin ligand on the leukocyte to selectins expressed on the endothelial surface results in weak intercellular interactions and “rolling” of leukocytes over the endothelial surface. These functions have now been confirmed in selectin knockout mice for L- and P-selectin. In a double knockout for E- and P-selectin, it appeared that E-selectin also contributes to leukocyte homeostasis (reviewed in references [14, 21]). Other functions for selectins are in β2 integrin activation and O2− production by leukocytes. Selectin ligands are specific oligosaccharide sequences in sialated and often in sulfated glycans, such as sialyl-Lewisx. There remains some uncertainty about the natural ligands for selectins. Many have been identified in in vitro assays, using recombinant selectins presented in multimeric form or in clustered arrays that increase affinity for ligand, but these have no known in vivo equivalent (reviewed in reference [22]). However, some molecules seem to fulfill most of the criteria for true selectin ligands (i.e. with confirmation of a biological role), such as CD24 and P-selectin glycoprotein ligand-1 (PSGL-1), and when expressed in the neutrophil surface in a properly glycosylated and tyrosine sulfated form can bind to P-selectin in vascular endothelium and to L-selectin on other neutrophils to enable “rolling” over either surface. CD34, GlyCAM-1, and MAdCAM-1 are other candidate ligands for L-selectin and E-selectin ligand-1 for E-selectin (reviewed in reference [23]). The signal transduction pathways linked to selectins are only partially elucidated [24]. As with the other classes of adhesion molecules discussed above they include tyrosine phosphorylation cascades and increases in intracellular calcium.
IV. ADHESION MOLECULES IN CELLS OF THE OSTEOBLAST LINEAGE A. Integrins
F. Selectins Selectins are a family of three closely related glycoproteins (P- and E-selectin expressed in endothelial cells and L-selectin expressed in leukocytes). Their common structure consists of an N-terminal Ca2+-dependent lectintype domain, an EGF domain, and variable numbers of short repeats homologous to complement-binding sequences, a single transmembrane region and a short cytoplasmic domain. Their molecular weights range from
In vivo osteoblasts in normal human bone have been reported to express β1 and β5 integrins [25–33] and most recently reviewed by Bennett et al. [34], but there remain controversies regarding the α subunits associated with β1 and concerning expression of αvβ3 (Table I). β2 and β4 integrins have not been reported in osteoblastic cells. Recently cultured human primary osteoblasts were for the first time reported to also express αvβ6 and αvβ8 integrin [35]. There are considerable differences between various reports on the integrin repertoire expressed by
Chapter 8 Integrins and Other Adhesion Molecules
osteoblasts in situ, by human osteoblast cultures, or by murine osteoblastic cell lines and these are most likely due to the heterogeneity of the cell population under study and the different culture conditions employed. Moreover, it is becoming clear that osteoblast activation status (e.g. “real” osteoblasts, synthesizing bone matrix versus the synthetically inactive bone-lining cell), or anatomical site (e.g. osteoblast phenotype in endochondral versus intramembranous bone), or the presence of bone disease, influences integrin expression, but no definitive studies have been published. For example, it has been clearly demonstrated that in vitro, the substrate on which cells are cultured directly influences the pattern of integrin expression by osteoblasts [36, 37]. Generally, cultured osteoblasts express a wider integrin repertoire than osteoblasts in situ [25, 32] and in particular increased expression of the vitronectin receptor αvβ3 is seen. It is still unclear whether altered expression of integrins has an equivalent in bone pathology, since osteoblast integrin expression in bone disease is relatively unexplored, but bone diseases such as Paget’s disease and osteopetrosis are accompanied by major changes in the numbers of “real” osteoblasts, i.e. osteoid synthesizing cells [38], and hence differences in integrin expression in the overall osteoblastic population are likely. Also, recent studies are beginning to examine the effects of mechanical stimuli on integrin expression and suggest that external stimuli may influence the integrin expression in the osteoblastic population. Integrin expression in osteocytes has not been studied extensively, but there are some data from immunocytochemical analysis of human bone sections and from functional studies in the chick, where isolation procedures for osteocytes have been developed. Chick osteocytes and osteoblasts bind a comprehensive range of extracellular matrix proteins in vitro in a β1- and partially RGDdependent way [39]. The exact receptor involved was not determined, because of lack of α chain-specific antibodies for avian integrins. Expression of β1 integrin has been confirmed in mammalian osteocytes [31, 40], including the osteocytic cell line MLO-Y4 [41]. However, there are few definitive reports where integrin expression has been followed throughout osteoblastic differentiation to osteocytes within the same species. Difficulties in interpretation of immunocytochemical staining in bone sections, in particular for cells embedded within matrix, have been reported by many authors and this may well have contributed to the continuing controversies in integrin phenotype of osteoblasts/osteocytes. From the limited reports on isolated cells at different stages of differentiation, it has become clear that, at least in the mouse, α5β1 is the most abundantly expressed integrin throughout osteoblastic differentiation and that, maybe surprisingly,
135 α2β1 is expressed at much lower levels, in particular in more differentiated cells [41]. Expression of α1 and α6 integrins has been reported in mesenchymal stem cells [42] and Stewart et al. [43] recently showed that selection of α1 integrin-positive bone marrow cells markedly enriched the population of CFU-F, clonogenic precursors with osteogenic potential, confirming the importance of α1 early in the osteoblast lineage. Human osteogenic stem cells also express β1 integrin [44]. Current interest in differentiation of osteogenic cells from stem cells [44, 45] may reveal additional information on expression of integrins and other cell adhesion molecules during differentiation, an area which is still poorly understood (see also reference [34]). This will require comprehensive studies in well-defined osteoblastic populations (both in developmental stage and in synthetic activity) combining immunocytochemical, biochemical, and molecular techniques. Functional studies on osteoblast integrins initially concentrated on their role in adhesion. In keeping with their extensive repertoire of integrins, rodent and human osteoblasts were found to adhere to osteopontin, bone sialoprotein, vitronectin, and fibronectin in an RGD-dependent way (with fibronectin requiring much higher concentrations of peptide for inhibition), whereas binding to type I collagen and thrombospondin was less inhibited by RGD peptides [25, 28, 46]. In organ cultures of mineralizing fetal rat parietal bone, RGD peptides decreased bone formation accompanied by a decrease in α2 and β1 expression and disruption of the organization in the osteoblast layer [47]. Down-regulation of α2 and β1 integrin expression in osteoblasts by glucocorticoids has also been noted with similar disruption in osteoblast organization, whereas IGF-1 increased β1 expression in osteoblasts and increased calcified bone formation [40, 48]. Equally, it was shown that, in vitro at least, αvβ3 and αvβ5 integrins, which regulate adhesion to vitronectin and osteopontin, are down-regulated by long-term glucocorticoid exposure [49], a phenomenon that, together with the negative effect on β1 integrin, might help to explain the bone loss associated with long-term glucocorticoid usage. More recently it has become clear that the role of osteoblast integrins extends beyond adhesion. BMP-2, a potent stimulator of bone formation, increases expression of the whole repertoire of osteoblast αv integrins and the BMP-2 receptor colocalizes with integrins. The finding that blocking of the function of αv inhibited the BMP-2 effects in osteoblast, suggests that integrins may regulate BMP-2 effects in bone [35] and suggests that there is potentially a wider role for integrins in facilitating growth factor signaling in bone. Functional studies on osteoblast integrins also include reports on their role in osteoblast differentiation.
136 Moursi et al. [50] demonstrated that osteoblast–fibronectin interaction was a critical event in the differentiation of fetal rat osteoblasts in an in vitro model and the importance of the central cell-binding domain of fibronectin in this effect suggested that the α5β1 fibronectin receptor was implicated. Later studies by this group also suggested a role for α3β1 in osteoblast differentiation [51]. In addition to fibronectin, collagen is also implicated in osteoblast differentiation. Ganta et al. [29] showed that ascorbic acid deficiency, which leads to under-hydroxylation of type I collagen, resulted in down-regulation of α2β1 in osteoblasts and dysregulation of differentiation and mineralization in cultures of rat calvaria. Jikko et al. [52] showed that α1 and α2 integrins mediate differentiation signals, such as from BMP-2 in early osteoblastic cells. Minzuno et al. [53] demonstrated a role for α2β1 and type I collagen in osteoblast differentiation from early progenitors and Xiao et al. [54–56] provided evidence that binding of α2β1 to ligand leads to expression of differentiation-associated genes such as Runx2, alkaline phosphatase, and osteocalcin. These data correlated well with the higher levels of α1 and α2 collagen-binding integrins in early differentiation stages of the lineage and reduced expression of these integrins in more differentiated cells as reported above. Since β1 integrins have a critical role in embryogenesis, not much information has been gained from studies of β1 knockout mice, which are embryonic lethal at a time well before skeletal development begins [57, 58]. Zimmerman et al. [59], therefore generated a transgenic mouse in which a dominant negative β1 integrin is expressed under the control of the osteoblast-specific osteocalcin promoter, resulting in expression in mature osteoblasts and in osteocytes. In vitro, osteoblasts from these animals did not properly adhere to matrix, whereas in vivo, osteoblast and osteocyte morphology, polarity and matrix secretion was severely affected, resulting in decreased bone mass, especially in females. Anatomical differences were found in older animals [60]. The tibias of the transgenic animals were straighter, consistent with the possibility that in absence of functional β1 integrin, osteoblastic cells are not able to sense and/or respond adequately to load bearing, which in wild-type animals results in the characteristic curvature of this long bone. Viable knockouts have been generated for α1 and α2 integrins. Surprisingly, neither has a prominent bone phenotype [61–63], although detailed bone histology has not been reported. It is possible that compensation occurs in these single knockouts for collagenbinding integrins, since overall the absence of gross abnormalities in any tissue in the α2 null was largely unexpected [64]. The α1 null demonstrates abnormalities
M. H. HELFRICH AND M. A. HORTON
during fracture healing, which are, however, related mainly to slow cartilage proliferation, rather than abnormalities in osteoblasts [63]. The α3/α6 double knockout has severe developmental deficiencies in limb formation, but no data exist for a role in adult mice (see reference [65] for a review on all integrin deletion phenotypes). Single knockouts for β3 and β5 and αv and a double knockout for β3/β5 have been made (see reference [66]) and none of these appear to have osteoblast malfunction. The relatively subtle defects seen in the skeleton of the β3 knockout are caused by osteoclast malfunction and are discussed further below. An area of increasing interest is that of the possible role of adhesion molecules, especially integrins, in mechanosensing in bone. Osteocytes and osteoblasts are known to be highly responsive to mechanical effects on bone. In other cell systems it has been demonstrated that twisting or turning of integrin molecules directly affects gene transcription [67]. Stretching of cells, i.e. change of cell shape and dimension, has also been shown to affect gene transcription and cell survival [68] and cell shape is largely controlled by extracellular matrix [69]. It is therefore tempting to speculate that positive mechanical effects on bone cells resulting in bone formation may be mediated indirectly via adhesion receptors and/or cellular shape changes resulting from the bone deformation. In contrast, absence of mechanical stimuli which leads to a decrease in bone mass may be the result of lack of “cellular stretch” in particular in osteocytes, which may lead to apoptotic cell death [68, 70], a cellular phenomenon associated with induction of bone resorption [71]. Despite these plausible hypotheses, the mechanisms whereby bone cells sense and respond to mechanical strain remain largely unresolved, as it remains technically challenging to subject bone cells, in particular osteocytes to defined levels of strain and measure single cell responses. A further complication is that osteocytes/osteoblasts in vivo exist as a syncytium and these cell–cell interactions are not readily re-established in vitro. Despite these difficulties some progress has been made recently. Charras and Horton used atomic force microscopy to estimate the cellular strain necessary to elicit cellular responses in bone cells [72] and then went on, using finite element modeling, to determine the strain exerted on individual cells by a number of common techniques used for mechanical stimulation of cells in vitro [73], such as the magnetic bead twisting and stretch experiments described above. Interestingly, fluid shear stress, the mechanisms by which osteocytes have been thought to detect strain in bone [74], gives rise to the lowest cellular strain (deformation) and it is possible that cells may have different mechanisms for detecting different magnitudes
Chapter 8 Integrins and Other Adhesion Molecules
of strain [73]. Using paramagnetic beads coated with specific integrin antibodies, Pommerenke et al. [75] were able to show that drag forces (shear) applied to α2 and β1 integrin subunits resulted in a cellular response, in this case intracellular calcium increase. Other studies (further discussed below) indicate that cadherins [76] are also directly involved in the response to stain and there is strong evidence for integrin-mediated mechanosensing in chondrocytes (see reference [77]). Although further information is clearly needed, especially in osteocytes, the principle that integrins and other classes of adhesion molecules are involved in mechanosensing appears well established. Taken together, the available evidence from in vivo and in vitro studies confirms that β1 integrins are the main integrins in the osteoblastic lineage and have a variety of functions (Table IV). In addition to providing an anchor to the bone matrix, they are important in the differentiation of osteoblasts, in the expression of differentiation-associated genes, in the response to growth factors and ultimately in the maintenance of bone mass [78], possibly through a role in mechanosensing. Unsurprisingly, there is great interest in the orthopedics/biomaterials field in identification of bioactive surface coatings to enhance osteoblast attachment and matrix (reviewed in reference [79]). Coating of biomaterials with fibronectin has been shown to give the best adhesion of osteoblastic cells, which is completely in agreement with all data presented above. Many studies integrating adhesive proteins with micro- and nano-topography on biomaterials are underway and will increase our understanding of integrin–matrix interactions and applications in tissue engineering.
B. Cadherins Cell–cell interactions between osteoblastic cells themselves and osteoblasts and hemopoietic cells within the bone marrow compartment are known to be important in bone metabolism. Such interactions are likely to be governed by adhesion molecules such as cadherins. Indeed, osteoblasts have been found to express a number of cadherins, in particular E-cadherin, cadherin-4, cadherin-11 (OB-cadherin), and N-cadherin [80–82] and, in culture, osteoblasts form tight cellular junctions containing cadherins [83]. Some differences have been found between different species, skeletal site and differentiation stage of the osteoblasts, but it is clear from a range of immunocytochemical and molecular studies that N-cadherin and cadherin-11 are abundantly expressed in osteoblastic cells in many mammalian species (reviewed in reference [84]). Cadherin expression has been studied during the
137 differentiation of early mesenchymal cells into mature osteogenic cells both in vivo and in vitro. N-cadherin appears to be expressed widely in all mesenchymal lineage cells and remains expressed at all stages of bone formation with especially high levels of expression at late stages of differentiation [85]. Cadherin-11 seems to be more specifically associated with the osteoblast lineage [86]. Cadherin expression appears to be lost in osteocytes [86, 87]. It has been suggested that this is linked to osteoblast apoptosis at the end of a formation cycle and is necessary to allow a proportion of mature osteoblasts to enter into the osteoid and become osteocytes. However, it is somewhat surprising that cadherins might not contribute to maintaining the cellular contacts between osteocytes and surface osteoblasts and more information is required on the molecules that form the cell–cell contacts in this network. There is good functional evidence that cadherins are important for bone cell function (Table IV). HAV peptides inhibit cell–cell contacts in osteoblast cultures and matrix formation in vitro, suggesting a role for class I cadherins in osteoblast synthetic activity [85, 88]. Likewise, antibodies to E-cadherin inhibit cell–cell adhesion [89]. Expression of a dominant negative N-cadherin in vitro in committed pre-osteoblastic cells as well as more differentiated osteoblasts inhibited expression of genes associated with bone formation and mineralization [85, 90]. Recently, cadherins were implicated in the response to mechanical stimulation in cultured osteoblasts by the finding that fluid shear stress increased translocation of β-catenin to the nucleus and regulated COX-2 expression in osteoblasts [76]. In mice in which the gene for cadherin 11 is deleted, a reduction in bone density is seen, strongly implying a role for cadherin-11 in osteogenesis [91]. At present there is no in vivo information about the bone phenotype in absence of N-cadherin since the N-cadherin-null mice die at day E10, before mature osteoblasts are present [92]. However, using a dominant negative approach, Castro et al. [93] showed that a reduction in functional N-cadherin in osteoblasts in vivo leads to a reduction in osteoblast number and thus in peak bone mass, whilst the number of adipocytes was increased possibly via changes in mesenchymal cell lineage commitment. Osteoblast cadherins are up-regulated, through as yet unknown mechanisms, by a variety of cytokines and hormones including BMP-2, FGF-2, PTH, and downregulated by IL-1 and TNFα (reviewed in references [84, 94]), and it is plausible that some of the well-known effects of these hormones are mediated in part through modulation of cadherin function. Application of stretch to osteoblasts in vitro selectively up-regulated expression
138
M. H. HELFRICH AND M. A. HORTON
of N-cadherin [95], leading to increased cell–cell adhesion. The signaling pathways controlling cell–cell adhesion through osteoblast cadherins and the ways they regulate osteoblastic gene expression are as yet unclear. Finally, osteoblast cadherins are implicated in malignancy: N-cadherin expression is reduced and cadherin-11 is not normally displayed on the cell surface in osteosarcoma [96] and β-catenin mutations are found in malignant bone tumors [97]. Up-regulation of E- and N-cadherin is found in Apert syndrome (discussed in reference [84]), and since cadherin expression is differentially regulated in osteoblastic versus adipocytic cell populations [93, 98] cadherins may play a role in development of osteoporosis and osteoarthritis, diseases associated with changes in the differentiation of mesenchymal cells to osteogenic versus adipocytic cells.
C. Ig Family Members A number of Ig family members are expressed in osteoblasts. During skeletal development transient expression of NCAM is seen [99–101]. In mature osteoblasts expression of ICAM-1, and ICAM–2, VCAM-1, and LFA-3 have been reported [26, 99, 102–108] and VCAM-1 is also seen in early osteogenic cells [44]. There is circumstantial evidence for the expression of NCAM in human osteoblasts from studies in multiple myeloma, where NCAM antibodies blocked IL-6 production in osteoblast–myeloma co-cultures [109]. Similarly, crosslinking of ICAM-1 or VCAM-1 resulted in the production of bone resorbing cytokines by osteoblasts [105], suggesting that, in general, expression of Ig family members by osteoblasts may lead to direct cellular interaction with immune cells, which express their ligands (Table IV) and subsequently to activation of production of bone-resorbing cytokines. In addition, Ig family adhesion molecules expressed on osteoblasts are firmly implicated in osteoclast formation. VCAM-1 has been shown to be involved in the development of osteoclasts in vitro, an effect that could be mediated via osteoblastic cells [110]. Harada et al. [107] found evidence for involvement of ICAM-1 and its ligand LFA-1 in osteoclastogenesis, suggesting a role not only for osteoblast–osteoclast interaction, but also for osteoclast precursor fusion, enabled by transient expression of LFA-1 in pre-osteoclasts. The role of ICAM-1 has been studied in more detail recently. Tanaka et al. [103] found that ICAM-1 expression characterized a population of osteoblasts, induced by pro-inflammatory cytokines, that have arrested in G0 /G1 and are uniquely equipped to sustain osteoclast differentiation. Interestingly, Everts et al.
described that the only cells positive for ICAM-1 on the bone surface in resorbing mouse calvaria are bone-lining cells [111]. Linking roles for integrins and Ig family members, Nakayamada et al. [26] reported how, following β1-dependent osteoblast adhesion, signaling through FAK results in up-regulation of ICAM-1 and RANKL leading to osteoclast formation. Further molecules that have been found to induce ICAM-1 expression in rodent osteoblasts in vitro, leading to increased osteoclast formation, are PTH, IL1, TNFα, and 1,25-D3 [106, 112]. The same stimuli also independently induce expression of RANKL and it is becoming clear that efficient stimulation of osteoclast precursors by RANKL is dependent upon high-affinity adhesion between osteoblast and osteoclast precursors, a process coined as “juxtacrine stimulation” [106]. Furthermore, crosslinking of CD44 in osteoblasts results in strong up-regulation of ICAM-1 and VCAM-1 [104], adding yet another mechanism by which bone metabolism and in particular osteoclastogenesis may be regulated, in this case through a CD44 ligand such as, for example, hyaluronan or osteopontin. ICAM-1 expression on osteoblasts in vivo in pathologies has yet to be studied. Expression of human primary osteoblasts in vitro is highly variable, but has been associated with pathological status: increased expression was seen in cells derived from patients with osteoporosis and to a lesser extent patients with osteoarthritis, compared to normal individuals, and increased expression was related in particular to the presence of elevated levels of IL6 and PGE2 [113]. These data fit well with the established role of ICAM-1 in inflammatory diseases in general and especially in rheumatoid synovium, where levels are considerably increased during active periods of disease and where anti-ICAM-1 therapy has been considered [114, 115]. Taken together with the experimental data it seems likely that high osteoblastic ICAM-1 levels are involved in the mechanism of diseases associated with bone loss by increasing osteoclast recruitment and formation.
D. Syndecans The expression of these molecules in bone is best studied during differentiation. Syndecan-3 (also known as N-syndecan) expression is found in cartilage (see below) and periosteum during endochondral bone formation, where it interacts with tenascin to set boundaries in tissues [116], but low levels are still seen in osteoblasts and osteocytes close to periosteal surfaces [117] and higher levels are found in osteoblasts and precursors in areas of bone regeneration after damage [118]. Messenger RNA for the three cloned human forms of
139
Chapter 8 Integrins and Other Adhesion Molecules
syndecan-1, -2, and -4 are found in primary human osteoblast cultures and a number of osteoblast cell lines [119, 120], where expression levels are regulated by cytokine-induced tissue regeneration [119]. Culture of osteoblastic cells under conditions that induce differentiation results in a decrease in syndecan-1 levels [120]. Syndecan-2 has been implicated in the mitogenic effect of GM-CSF in osteoblasts [121]. A role of syndecans in presentation of growth factors to their receptors has been suggested as their most prominent function in bone but this falls outside the scope of this chapter. In addition, syndecan-1 could also be an adhesion receptorregulating interaction of osteoblastic cells with the extracellular matrix. The heparin-binding growth-associated molecule (HB-GAM, or pleiotrophin), which is expressed by osteoblasts and osteocytes [118, 122], indeed seems to act in such a way by attracting N-syndecan-expressing osteoblasts, resulting in enhanced bone deposition [118], a process which may be important both during development as well as in response to bone injury.
E. CD44 CD44 has been found in humans and rodents to be expressed in osteoblasts and osteocytes at all stages of maturation; however, there appears to be a clear increase in expression levels with increasing maturation in this cell lineage with low levels expressed in pre-osteoblastic cells and very high levels expressed in osteocytes [123, 124]. Detailed localization studies have shown that CD44 expression in mature osteoblasts is confined to cytoplasmic processes only [125]. By a variety of immunological and molecular techniques only the standard form of CD44, CD44H, was found to be expressed in osteoblasts and osteocytes in rat bone [124, 126]; however, several isoforms of CD44 were described in the human osteoblastic cell line MG63 [127]. Expression of CD44 in cells of the osteoblastic lineage correlates well with the presence of hyaluronate in surrounding tissues [126], suggesting that the receptor may function as a hyaluronate-binding protein. In vitro, osteoblastic cells have indeed been shown to be able to bind to, and degrade, hyaluronate in the transition zone from cartilage to bone in the growth plate by a CD44dependent mechanism [126]. However, a variety of other known ligands for CD44, e.g. type I collagen, fibronectin, laminin, and osteopontin, are also produced by osteoblasts (see Chapters 1–4) and osteocytes [128] and co-localize with the receptor, indicating that a much wider range of functions for this molecule may exist. As described above, crosslinking of CD44 on osteoblasts
leads to up-regulation of ICAM-1 and VCAM-1 [104], which in turn can lead to increased osteoclast formation [106, 107].
F. Selectins There are as yet no published data on expression of selectins in osteoblasts or their precursors. As discussed for osteoclasts below, they may well play a role in extravasation of the mesenchymal precursors of osteoblasts. In the absence of definitive markers for osteoblast precursors, however, such information is as yet unavailable.
V. ADHESION MOLECULES IN OSTEOCLASTS A. Integrins The integrin repertoire expressed in mature osteoclasts is well defined and has been the subject of several reviews (see e.g. reference [129]). Immunocytochemical, biochemical, and functional assays have been used to confirm the expression of three major integrin receptors in human osteoclasts: αvβ3, α2β1, and αvβ1, e.g. [129–131], whereas expression of other subunits has been largely excluded (Tables II and IV). αvβ5 integrin is expressed in osteoclast precursors such as immature bone marrow macrophages, but no longer present in mature osteoclasts [132, 133]. Detailed localization studies have tried to answer the question whether a member of the integrin family is present in, and possibly responsible for, maintenance of the tight sealing zone in resorbing osteoclasts, but the data so far are conflicting. Although some studies have reported the presence of αv and/or β3 subunits at this site [134, 135], others have excluded the presence of αvβ3 (reviewed in references [136, 137]) and α2β1 [138] in the sealing zone, and recent data suggest that this structure is not as impermeable as originally thought [139]. It is clear that none of the integrin receptors are enriched in the clear zone of resorbing osteoclasts and that they are in fact most abundant on the basolateral membrane, where they may act as true receptors coupled to signal transduction pathways [138, 140, 141]. There is also high expression of integrins on the ruffled border membrane in resorbing osteoclasts [138, 140], suggesting that cell–matrix interactions at that site may be important in the formation of the ruffled border by facilitating penetration of the osteoclast membrane deep into the bone and possibly also in the cessation of bone resorption, since elevated levels of calcium in the
140 resorption lacunae may bind to cation-binding sites of αvβ3 integrin and inhibit adhesion (see reference [3]). The role of the vitronectin receptor αvβ3 in osteoclast biology is well established. In vitro mammalian osteoclasts adhere to a wide range of extracellular matrix proteins known to be expressed in bone and bone marrow using αvβ3 (Table IV) and an RGD peptide sequence forms the recognition site for the receptor in all these ligands. Possibly surprisingly, the question of which protein constitutes the natural ligand of osteoclasts in bone remains to be determined. Osteopontin is a candidate [142], since it was found to be enriched underneath the clear zones of resorbing osteoclasts. However, since osteoclasts actively synthesize osteopontin [143], this finding should be interpreted with some reservation. RGD-containing peptides and snake venoms (echistatin and kistrin) have been shown to inhibit osteoclast polarization, activity and formation in vitro [144–147] and they, as well as vitronectin receptor antibodies, also block bone resorption in various rodent models in vivo [148–150]. In vitro, occupation of the vitronectin receptor by antibodies or RGD peptides causes osteoclasts to retract and detach from matrix, similar to the shape changes observed after administration of the potent antiresorptive peptide hormone calcitonin. In mammalian osteoclasts this effect is preceded by a rise in intracellular calcium, localized predominantly to the nucleus [151, 152]. The impressive effects of nonpeptide RGD analogs developed by the pharmaceutical industry [153] on bone resorption (both in in vitro assays and more recently in in vivo animal models (see reference [129]) has led to their introduction into clinical trails for bone diseases associated with excessive bone resorption such as osteoporosis [261a]. Gene-knockout studies where β3 integrins have been deleted in the mouse have underscored the central role of αvβ3 integrin in osteoclast biology [154]. β3-null mice are relatively normal at birth, but develop a progressive, but mild, osteosclerosis by adulthood; this is associated with in vitro evidence of abnormal osteoclast adhesion and bone resorption. They are protected against bone loss after ovariectomy [155], a finding that, together with the reported down-regulation of β3 expression by estradiol in osteoclasts in vitro [156], suggests that the bone-sparing effects of estrogen may be in part through direct effects on osteoclasts. The critical role of β3 integrin in osteoclasts is also underscored by its role in promoting osteoclast survival and in regulating apoptosis [157]. αv knockout mice die perinatally due to vascular defects [158] and osteoclast defects have therefore not been studied extensively. However, mice from both deletions have shown normal skeletal development, contrary to the prediction
M. H. HELFRICH AND M. A. HORTON
that bone modeling would be significantly affected, suggesting that other integrins compensate in the process of bone recognition by osteoclasts. Indeed, this has been shown to be true in an equivalent study in humans using peripheral blood cultures from patients with Glanzmann thrombasthenia carrying an integrin β3-null mutation [159]; in the face of absent αvβ3 there is an up-regulation of α2β1 collagen-binding receptor in these osteoclasts that enables bone resorption to proceed, albeit to a reduced extent. αvβ5-null mice have a relatively normal phenotype [160]. However, αvβ5-null mice generate more osteoclast precursors and, in the absence of estrogen, osteoclast formation and bone resorption in vivo is markedly increased [161], indicating that αvβ5 is an inhibitor of osteoclast formation, contrary to the role of αvβ3 as described above. The function of α2β1 in mammalian osteoclasts is predominantly as a receptor for type I collagen. In contrast to other cell types, adhesion of osteoclasts to collagen appears to be RGD dependent [138, 144], although this could possibly be explained as a dominant-negative effect of RGD occupation of the abundant αvβ3 receptors on osteoclasts. Antibodies to α2 and β1 integrin inhibit resorption by human osteoclasts in vitro, but not to the same extent as antivitronectin receptor antibodies. Avian osteoclasts express abundant β1 integrin, but do not adhere to collagen [162] and probably use β1 integrin predominantly in association with α5 as a fibronectin-binding receptor. The role of αvβ1 on osteoclasts has not been explored in functional assays since no receptor complex-specific antibodies are available at present. This receptor is far less abundant than αvβ3 and α2β1 in osteoclasts [131]. By analogy with other cell types it is likely that αvβ1functions as a receptor for collagen or fibronectin in osteoclasts. Several groups have investigated the downstream signaling pathways associated with osteoclast polarization, which follows integrin-mediated adhesion to matrix and is essential for bone resorption. A number of candidate molecules, such as c-src, integrin-linked kinase (ILK), phosphatidylinositol 3-kinase (PI-3 kinase), FAK and Pyk2, are expressed at high levels in osteoclasts (reviewed in references [163, 164]). c-Src knockout mice have severe osteopetrosis due to lack of ruffled border formation in osteoclasts, clearly implicating this molecule in signaling cascade leading to polarization of osteoclasts [165]. Both PI-3 kinase and ILK have been shown to be associated with the cytoskeleton in osteoclasts when attached to bone matrix and ILK was also found associated with the β3 subunit in bone-adherent osteoclasts [153, 166], suggesting
141
Chapter 8 Integrins and Other Adhesion Molecules
that these molecules may also play a role. The importance of FAK in osteoclast function has been queried since no osteoclast abnormalities are seen in the FAK knockout mouse [167]. However, Pyk2, another molecule of the FAK family with high homology to FAK and a similar role as adaptor protein orchestrating cytoskeletal architecture [168], is highly expressed in osteoclasts and seems to fulfill the role that FAK plays in less-motile cell types [169]. Pyk2 has been shown to interact with gelsolin, an actinbinding protein involved in the formation of adhesive structures in osteoclasts and in bone resorption [170, 171]. Following engagement of αvβ3 with ligand, clustering of the receptor occurs and association of its cytoplasmic domain with a complex containing Pyk2 and p130CAS. Autophosphorylation of Pyk2 then leads to recruitment of c-Src and ultimately to cytoskeletal reorganization and bone resorption [163, 172]. Activated c-Src also recruits and phosphorylates c-Cbl, which binds to the SH3 domain of c-Src and negatively regulates Src kinase activity. It is suggested that this may enable transient adhesion and allow for cell migration, a process critically dependent on turnover of adhesive structures [173]. In addition, the Src-dependent phosphorylation of c-Cbl has been shown to be critical for osteoclastic resorption [174]. In c-Srcnull osteoclasts, Pyk-2 has been shown to associate with PLC-γ, a process which can be induced not only by adhesion but also by M-CSF, and can lead to recruitment of integrin to adhesion contacts and cytoskeletal reorganization [175].
showed prominent staining in the clear zone, suggesting a possible role for cadherin in creating a close contact with bone matrix [178].
C. Immunoglobulin Family Members There is good evidence for expression of receptors of the Ig family in osteoclast precursors. Functional evidence implies ICAM-1 and VCAM-1 in osteoclast development in vitro [26, 107, 110] and as discussed in the osteoblast section, osteoclast precursors, i.e. monocytes, not only express ICAM-1 but also its ligand LFA-1, allowing for cell–cell adhesion and possibly cell fusion [107]. Now that osteoclasts can be generated in the absence of osteoblasts in defined cultures with RANKL and M-CSF, it will be possible to study in more detail the expression and functional role of this important class of adhesion molecule during osteoclast differentiation and in mature osteoclasts.
D. Syndecans There is no information about the expression of syndecans in mature osteoclasts or for a role during osteoclast development.
E. CD44 B. Cadherins There is limited information about cadherin expression in osteoclasts. Mbalaviele et al. [176] reported expression of E-cadherin, and absence of P- and N-cadherin, in mature human and mouse osteoclasts. This study also suggests that in the mouse in an in vitro system osteoclast development and in particular osteoclast fusion requires expression of E-cadherin, although it remains to be firmly established whether this is at the level of the osteoclast precursor, or whether E-cadherin expression is required in the accessory cells (osteoblasts and/or stromal cells, which are known to be crucially important in formation of multinucleated osteoclasts and express cadherins). In support of the former a recent study suggested a role for cadherinmediated interactions in early hemopoietic development [177]. In support of a role for cadherins in mature osteoclasts, Ilvesaro et al. [178] demonstrated that a cadherin dimer disrupting HAV peptide inhibited bone resorption. Interestingly pan-cadherin immunostaining in osteoclasts
CD44, for which several possible ligands are expressed in bone (Table IV), is highly expressed in osteoclasts in vivo [123, 125, 130] and generated in vitro [179]. Detailed studies of sites of expression have so far been confined to rodent osteoclasts, where expression is at the basolateral membrane, rather than the clear zone or ruffled border of resorbing osteoclasts [125]. In a functional study aimed at addressing the role of CD44 in mature osteoclasts and during osteoclast development, Kania et al. [180] described that mouse osteoclast formation in vitro was inhibited by CD44 antibodies, whereas the resorptive capacity of mature osteoclasts was not affected. More recently, these in vitro data were confirmed when it was demonstrated that the interaction between CD44 and osteopontin is involved in osteoclast migration, fusion, and bone resorption [181, 182]. However, two independently generated CD44-null mice show no differences in osteoclast formation, fusion, or bone resorption and only small differences in bone mass and anatomy [183], suggesting that
142
M. H. HELFRICH AND M. A. HORTON
there is an increased level of complexity and redundancy in the actions of CD44 in vivo. This has also been highlighted by the recent finding that the role of CD44 in osteoclast formation in vitro is dependent on the culture substrate and that it plays no role when cells are grown on bone [184].
F. Selectins So far there are no reports describing expression of selectins on mature osteoclasts. Selectins are likely, however, to have a role during osteoclast development. Osteoclast precursors are present in the circulation and mature osteoclasts can be generated in vitro from mononuclear cells in blood. It remains unclear how osteoclast precursors in vivo reach the sites in bone where they differentiate into fully active osteoclasts, but this must at some point include extravasation, a process that, for all lymphoid cells at least, initially involves selectin-mediated interaction with endothelial cells, followed by integrinmediated processes. Osteoclast–endothelial interactions have been studied [185, 186], but so far the nature of the adhesive molecular interactions between these cells remains unclear. This issue deserves further study, because it might offer therapeutic potential in diseases where osteoclast formation is increased.
VI. ADHESION MOLECULES IN CHONDROCYTES A. Integrins As for cells of the osteoblast lineage, the reported integrin phenotype of chondrocytes is complex (Table III), with additional inconsistency between publications (e.g. see references [187–192] and also reviewed in reference [193]). A synthesis of the literature suggests that human chondrocytes express the β1 integrins α1, α2, α3, α5, and α6, but not α4. β2, β4 and β6 are absent; analysis of β7–9 and CD11 has not been reported. Some studies have shown high expression of αv integrin. As in osteoblasts, this is mainly as αvβ5 and not the αvβ3 dimer; however, a subpopulation of superficial articular chondrocytes was found to be αvβ3 positive [191]. The most abundant chondrocyte integrin appears to be α5β1. These complexities could well relate in part to variation in sampling site, use of fetal versus adult material, species differences, artefacts induced in culture, or influences of disease on phenotypes. Indeed the first possibility is borne out by the study of Salter et al. [190] where the distribution of
integrin clearly differs by site (human articular, epiphyseal, and growth plate chondrocytes were studied) and variation in expression has been observed in cartilage when comparing fetal and adult samples but not during the endocrine-driven pubertal growth [194]. Likewise, changes have been reported in cultured chondrocytes [195–197]. Two relatively newly discovered integrins, α10 [198, 199] and α11 [200, 201], have been shown to be collagen receptors that are expressed in cartilage. α10 has a cellular distribution that differs from α1 and α2 and was dominant during embryonic development [199]. No knockout mice are available to study their cartilage phenotype. Some integrin receptors have been found in the cartilage matrix, in addition to expression in chondrocytes [202]. The role of cell adhesion molecules in cartilage is thus relatively unclear (though the current state of knowledge regarding the function of some cell adhesion molecules is summarized in Table IV). From first principles, these could include roles in cartilage differentiation during fetal development; response to mechanical forces (for example in articular cartilage, menisci); maintenance of tissue architecture and integrity including by matrix synthesis and assembly (e.g. via matrix integrins) and cell adhesion, and regulation of chondrocyte gene expression. Additionally, there is likely to be a role of cell adhesion molecules in responses in cartilage to injury and disease [203, 204]. For example, up-regulation of α2, α4, and β2 subunit expression in osteoarthritic versus normal and cartilage were reported by Ostergaard et al. [205], and altered physiological responses to mechanical stress mediated by α5β1 integrin in osteoarthritic cartilage [206]. The changes in distribution of both integrin and matrix proteins [190] in different zones of cartilage suggest a role in chondrocyte differentiation from mesenchymal precursors [207, 208] and in chondrocyte function. Recently, in vivo studies have underscored the important role of β1 integrin in chondrocytes where Aszódi and colleagues inactivated β1 selectively in chondrocytes [209]. The resulting mice had chondrodysplasia, failed to align chondrocytes into columns in the growth plate. Underlying these abnormalities was an inability of the cells to adhere to fibronectin and to progress normally through mitosis and cytokinesis, due to reduction in cyclin D1 expression. Evidence is accruing for a specialized function for integrins in responses of chondrocytes to mechanical stresses [77, 210–213] and most recently reviewed in reference [77]. Extensive studies with monolayers of isolated articular chondrocytes show that integrin, especially α5β1 interacting with fibronectin, is critical in the hyperpolarization response to mechanical load. Adhesion via integrin leads to formation of focal adhesions in chondrocytes and,
143
Chapter 8 Integrins and Other Adhesion Molecules
following mechanical stimulus, phosphorylation of focal adhesion components, including signaling molecules in the MAP kinase pathway, is seen. Synthesis of autocrine/ paracrine factors known to be involved in transducing the biological effect from mechanosensor to effector cells (similar to osteocyte–osteoblast interactions described above) is stimulated by mechanical load on chondrocytes. In this context, IL-4 and substance P appear especially important, in addition to nitric oxide and prostaglandins. Extensive studies have though been performed to address the role of α5β1 in chondrocyte interactions with fibronectin [187, 189, 214–219]. Function-blocking antibodies and RGD peptides have been shown to inhibit cell adhesion to fibronectin and its fragments, thus modifying chondrocyte behavior and cartilage function. Likewise, chondrocyte recognition of collagen (including types I, II, and VI collagen) has been studied in vitro [187, 189, 214, 216, 219–221] and shown to be mediated via β1 integrins: α1β1 for adhesion to type I [196, 222] and type VI [222] and α3β1 [196, 197, 222] or α2β1 [221] for adhesion to type II. There is also increasing evidence for a connection between chondrocyte adhesion to extracellular matrix proteins, especially fibronectin, and chondrocyte–synovial cell interaction [223], chondrocyte cell signaling [224], and chondrocyte survival [225–227]. Regulation of integrin expression and function by cytokines such as IL1, TGFβ, and IGF1 has been linked to the release of matrix metalloproteinases [228] and hence cartilage breakdown [215, 218, 222, 229, 230]. Such events are likely to be involved in the pathogenesis of the cartilage destruction seen in osteoarthritis and rheumatoid arthritis. The key role for downstream signaling following integrin interaction with matrix ligands is underscored by the phenotype of integrin-linked kinase (ILK) knockout mice [231, 232]. ILK acts as a pleiotropic adapter interfacing β integrin cytoplasmic domains into the PKB/Akt signaling pathway; mice with ILK deletion in the chondrocyte lineage have chondrodysplasia, show dwarfism, and isolated chondrocytes have matrix adhesion defects [231, 232]. As discussed above for osteogenic cells, there is an increasing interest in articular cartilage engineering and adhesion processes, especially integrin–matrix interactions, which will need to be considered when designing appropriate scaffolds [233].
B. Cadherins Cadherin-11 is expressed in mesenchymal cells migrating from the neuroectodermal ridge which form presumptive cartilage (and in other morphogenetic events)
in the developing mouse [234]. Likewise, N-cadherin is expressed in prechondrocytic cells in avian and mammalian limb buds, but not mature cartilage [235–237]. In vitro culture of mesenchymal cells suggests a functional role for N-cadherin in early chondrogenesis [235, 236, 238] and its function may be regulated by calciotropic factors such as 1,25D3 and TGFβ [239].
C. Ig Family Members NCAM is similarly distributed to N-cadherin in early cartilage development [239, 240], with N-cadherin expression temporally preceding NCAM. It is therefore suggested that the role of NCAM is to stabilize cell–cell adhesions formed initially by N-cadherin [236]. Primary human articular chondrocytes constitutively express VCAM-1 and ICAM-1 and this expression is increased by cytokines such as IL1β and TNFα and reduced by TGFβ and γ interferon [241–243]. In vitro VCAM-1 and ICAM-1 contribute to T-cell adhesive processes, suggesting that in vivo these molecules may be important players in mediating T cell–chondrocyte interactions at sites of inflammatory joint destruction [242, 244, 245].
D. Syndecans Syndecan-3 expression has been analyzed in developing avian limb buds; there are little data in mammals or for other syndecan forms. Syndecan-3 is highly expressed in proliferating chondrocytes, below the tenascin C-rich layer of articular chondrocytes; decreased levels are found in hypertrophic cartilage [246]. High levels are also found in forming perichondrium (and later in periosteum) in the developing avian limb [247] and it has been suggested that, with tenascin C, syndecan-3 is involved in establishing or maintaining boundaries during skeletogenesis [248]. More recent data support a role for syndecan-3 in the regulation of chondrocyte proliferation [249, 250].
E. CD44 CD44 is expressed by cartilage and has been studied for a variety of sites and species [123, 126, 251], reviewed most recently in [252]. The predominant isoform detected is the standard CD44H variant; epithelial CD44E is not found [253]. There is some evidence from the use of function-blocking antibodies that CD44 is involved in chondrocyte pericellular matrix assembly [254, 255]. The full range of extracellular matrix molecules recognized
144
M. H. HELFRICH AND M. A. HORTON
by CD44 in cartilage is unclear, but interaction with hyaluronate clearly occurs [256]. CD44 is up-regulated during cartilage catabolism (for example on IL1α treatment of bovine articular cartilage) and chondrocytes have been shown to actively take up hyaluronate via CD44mediated endocytosis; thus it is reasonable to speculate that this molecule plays a regulatory role in cartilage matrix turnover in health and disease [257, 258]. In support of this, CD44 has been found to be up-regulated in chondrocytes from rheumatoid arthritis cartilage [259]. Recent evidence points to the presence of soluble CD44 in many tissues and especially in serum. Although mainly studied in the context of tumor biology, there is evidence for elevated levels of soluble CD44 in inflammation such as in rheumatoid arthritis [20].
References
F. Selectins There is no information on selectin expression in chondrocytes.
VII. CONCLUSION Bone and cartilage cells express a wide variety of adhesion molecules. Integrin expression has been studied extensively, and there is increasing information on expression of other adhesion molecule family members. There is still limited information on the expression and function of adhesion molecules of all classes during osteoblast and osteoclast development, but, with the ever-increasing knowledge of stage-specific markers, this information should become available. Adhesion molecules fulfill many functions in skeletal cells and are linked to a variety of intracellular signaling pathways, which are increasingly being elucidated. Although no unique osteoblast or osteoclast adhesion molecule has been identified to date, therapeutic strategies based on selectively inhibiting highly expressed receptors, such as αvβ3 in osteoclasts, have proved to be successful in the inhibition of excessive bone resorption. Better knowledge of the expression and roles of adhesion molecules in bone pathology may help in diagnosis, is essential to design appropriate biomaterials for use with skeletal tissue engineering, and may lead to therapeutic strategies for other bone cell types and for specific bone disorders.
Acknowledgments The authors acknowledge their longstanding support from the Arthritis Research Campaign, UK (MHH) and the Wellcome Trust (MAH) for their work in this area.
1. Isacke, C., and Horton, M. A. (2000). “The Adhesion Molecule Facts Book.” 2nd ed. Academic Press, San Diego. 2. Hynes, R. O. (2002). Integrins: bidirectional, allosteric signaling machines. Cell 110, 673–687. 3. Plow, E. F., Haas, T. A., Zhang, L., Loftus, J., and Smith, J. W. (2000). Ligand binding to integrins. J. Biol. Chem. 275, 21785–21788. 4. Nermut, M. V., Green, N. M., Eason, P., Yamada, S. S., and Yamada, K. M. (1988). Electron microscopy and structural model of human fibronectin receptor. EMBO J. 7, 4093–4099. 5. Xiong, J. P., Stehle, T., Diefenbach, B., Zhang, R., Dunker, R., Scott, D. L., Joachimiak, A., Goodman, S. L., and Arnaout, M. A. (2001). Crystal structure of the extracellular segment of integrin alpha Vbeta3. Science 294, 339–345. 6. Xiong, J. P., Stehle, T., Zhang, R., Joachimiak, A., Frech, M., Goodman, S. L., and Arnaout, M. A. (2002). Crystal structure of the extracellular segment of integrin alphaVbeta3 in complex with an Arg-Gly-Asp ligand. Science 296, 151–155. 7. van der Flier, A., and Sonnenberg, A. (2001). Function and interactions of integrins. Cell Tissue Res. 305, 285–298. 8. Potter, E., Bergwitz, C., and Brabant, G. (1999). The cadherin-catenin system: implications for growth and differentiation of endocrine tissues. Endocr. Rev. 20, 207–239. 9. Patel, S. D., Chen, C. P., Bahna, F., Honig, B., and Shapiro, L. (2003). Cadherin-mediated cell-cell adhesion: sticking together as a family. Curr. Opin. Struct. Biol. 13, 690–698. 10. Yagi, T., and Takeichi, M. (2000). Cadherin superfamily genes: functions, genomic organization, and neurologic diversity. Genes Dev. 14, 1169–1180. 11. Westendorf, J. J., Kahler, R. A., and Schroeder, T. M. (2004). Wnt signaling in osteoblasts and bone diseases. Gene 341, 19–39. 12. Nelson, W. J., and Nusse, R. (2004). Convergence of Wnt, beta-catenin, and cadherin pathways. Science 303, 1483–1487. 13. Wheelock, M. J., and Johnson, K. R. (2003). Cadherin-mediated cellular signaling. Curr. Opin. Cell Biol. 15, 509–514. 14. Barclay, A. N., Brown, M. H., Law, S. K. A., McKnight, A. J., Tomlinson, M. G., and van der Merwe, P. A. (1997). The Leukocyte Antigen Factsbook, 2nd ed. Academic Press, San Diego. 15. Brummendorf, T., and Lemmon, V. (2001). Immunoglobulin superfamily receptors: cis-interactions, intracellular adapters and alternative splicing regulate adhesion. Curr. Opin. Cell Biol. 13, 611–618. 16. Hubbard, A. K., and Rothlein, R. (2000). Intercellular adhesion molecule-1 (ICAM-1) expression and cell signaling cascades. Free Radic. Biol. Med. 28, 1379–1386. 17. Elenius, K., and Jalkanen, M. (1994). Function of the syndecans – a family of cell surface proteoglycans. J. Cell Sci. 107, 2975–2982. 18. Russell, R. G., Smith, R., Preston, C., Walton, R. J., and Woods, C. G. (1974). Diphosphonates in Paget’s disease. Lancet 1, 894–898. 19. Weber, G. F., Ashkar, S., Glimcher, M. J., and Cantor, H. (1996). Receptor-ligand interaction between CD44 and osteopontin (Eta-1). Science 271, 509–512. 20. Cichy, J., and Pure, E. (2003). The liberation of CD44. J Cell Biol. 161, 839–843. 21. Frenette, P. S., and Wagner, D. D. (1997). Insights into selectin function from knockout mice. Thromb. Haemost. 78, 60–64. 22. Varki, A. (1997). Selectin ligands: will the real ones please stand up? J. Clin. Invest. 99, 158–162. 23. Ehrhardt, C., Kneuer, C., and Bakowsky, U. (2004). Selectins – an emerging target for drug delivery. Adv. Drug Deliv. Rev. 56, 527–549. 24. Crockett-Torabi, E. (1998). Selectins and mechanisms of signal transduction. J. Leukocyte Biol. 63, 1–14.
Chapter 8 Integrins and Other Adhesion Molecules 25. Grzesik, W. J., and Robey, P. G. (1994). Bone matrix RGD glycoproteins: immunolocalization and interaction with human primary osteoblastic bone cells in vitro. J. Bone Miner. Res. 9, 487–496. 26. Nakayamada, S., Okada, Y., Saito, K., Tamura, M., and Tanaka, Y. (2003). Beta1 integrin/focal adhesion kinase-mediated signaling induces intercellular adhesion molecule 1 and receptor activator of nuclear factor kappaB ligand on osteoblasts and osteoclast maturation. J. Biol. Chem. 278, 45368–45374. 27. Brighton, C. T., and Albelda, S. M. (1992). Identification of integrin cell-substratum adhesion receptors on cultured rat bone cells. J. Orthop. Res. 10, 766–773. 28. Majeska, R. J., Port, M., and Einhorn, T. A. (1993). Attachment to extracellular matrix molecules by cells differing in the expression of osteoblastic traits. J. Bone Miner. Res. 8, 277–289. 29. Ganta, D. R., McCarthy, M. B., and Gronowicz, G. A. (1997). Ascorbic acid alters collagen integrins in bone culture. Endocrinology 138, 3606–3612. 30. Clover, J., and Gowen, M. (1994). Are MG-63 and HOS TE85 human osteosarcoma cell lines representative models of the osteoblastic phenotype? Bone 15, 585–591. 31. Hughes, D. E., Salter, D. M., Dedhar, S., and Simpson, R. (1993). Integrin expression in human bone. J. Bone Miner. Res. 8, 527–533. 32. Clover, J., Dodds, R. A., and Gowen, M. (1992). Integrin subunit expression by human osteoblasts and osteoclasts in situ and in culture. J. Cell Sci. 103, 267–271. 33. Bennett, J. H., Carter, D. H., Alavi, A. L., Beresford, J. N., and Walsh, S. (2001). Patterns of integrin expression in human mandibular explant model of osteoblast differentiation. Arch. Oral Biol. 46, 229–238. 34. Bennett, J. H., Moffatt, S., and Horton, M. (2001). Cell adhesion molecules in human osteoblasts: structure and function. Histol. Histopathol. 16, 603–611. 35. Lai, C. F., and Cheng, S. L. (2005). AlphaVbeta integrins play an essential role in BMP-2 induction of osteoblast differentiation. J Bone Miner. Res. 20, 330–340. 36. Sinha, R. K., and Tuan, R. S. (1996). Regulation of human osteoblast integrin expression by orthopedic implant materials. Bone 18, 451–457. 37. Gronowicz, G., and McCarthy, M. B. (1996). Response of human osteoblasts to implant materials: integrin-mediated adhesion. J. Orthop. Res. 14, 878–887. 38. Helfrich, M. H. (2003). Osteoclast diseases. Microsc. Res. Tech. 61, 514–532. 39. Aarden, E. M., Nijweide, P. J., van der Plas, A., Alblas, M. J., Mackie, E. J., Horton, M. A., and Helfrich, M. H. (1996). Adhesive properties of isolated chick osteocytes in vitro. Bone 18, 305–313. 40. Gohel, A. R., Hand, A. R., and Gronowicz, G. A. (1995). Immunogold localization of β1-integrin in bone: effect of glucocorticoids and insulin-like growth factor I on integrins and osteocyte formation. J. Histochem. Cytochem. 43, 1085–1096. 41. Helfrich, M., van’t Hof, R., and MacPhee, J. (2004). Expression and localization of integrins, NOS and caveolin-1 in osteoblasts and osteocytes. Calcif. Tissue Int. 74, S64–S65. 42. Bruder, S. P., Jaiswal, N., Ricalton, N. S., Mosca, J. D., Kraus, K. H., and Kadiyala, S. (1998). Mesenchymal stem cells in osteobiology and applied bone regeneration. Clin. Orthopaed. Relat. Res. 355(S), 247–256. 43. Stewart, K., Monk, P., Walsh, S., Jefferiss, C. M., Letchford, J., and Beresford, J. N. (2003). STRO-1, HOP-26 (CD63), CD49a and SB-10 (CD166) as markers of primitive human marrow stromal cells and their more differentiated progeny: a comparative investigation in vitro. Cell Tissue Res. 313, 281–290. 44. Kuznetsov, S. A., Mankani, M. H., Gronthos, S., Satomura, K., Bianco, P., and Robey, P. G. (2001). Circulating skeletal stem cells. J Cell Biol. 153, 1133–1140.
145 45. Caplan, A. I., and Bruder, S. P. (2001). Mesenchymal stem cells: building blocks for molecular medicine in the 21st century. Trends Mol. Med. 7, 259–264. 46. Puleo, D. A., and Bizios, R. (1991). RGDS tetrapeptide binds to osteoblasts and inhibits fibronectin-mediated adhesion. Bone 12, 271–276. 47. Gronowicz, G. A., and DeRome, M. E. (1994). Synthetic peptide containing Arg-Gly-Asp inhibits bone formation and resorption in a mineralizing organ culture system of fetal rat parietal bones. J. Bone Miner. Res. 9, 193–201. 48. Doherty, W. J., DeRome, M. E., McCarthy, M. B., and Gronowicz, G. A. (1995). The effect of glucocorticoids on osteoblast function. The effect of corticosterone on osteoblast expression of β1 integrins. J. Bone Joint Surg. Am. 77, 396–404. 49. Cheng, S. L., Lai, C. F., Fausto, A., Chellaiah, M., Feng, X., McHugh, K. P., Teitelbaum, S. L., Civitelli, R., Hruska, K. A., Ross, F. P., and Avioli, L. V. (2000). Regulation of alphaVbeta3 and alphaVbeta5 integrins by dexamethasone in normal human osteoblastic cells. J. Cell Biochem. 77, 265–276. 50. Moursi, A. M., Damsky, C. H., Lull, J., Zimmerman, D., Doty, S. B., Aota, S., and Globus, R. K. (1996). Fibronectin regulates calvarial osteoblast differentiation. J. Cell Sci. 109, 1369–1380. 51. Moursi, A. M., Globus, R. K., and Damsky, C. H. (1997). Interactions between integrin receptors and fibronectin are required for calvarial osteoblast differentiation in vitro. J. Cell Sci. 110, 2187–2196. 52. Jikko, A., Harris, S. E., Chen, D., Mendrick, D. L., and Damsky, C. H. (1999). Collagen integrin receptors regulate early osteoblast differentiation induced by BMP-2. J. Bone Miner. Res. 14, 1075–1083. 53. Mizuno, M., Fujisawa, R., and Kuboki, Y. (2000). Type I collageninduced osteoblastic differentiation of bone-marrow cells mediated by collagen-alpha2beta1 integrin interaction. J. Cell Physiol. 184, 207–213. 54. Xiao, G., Cui, Y., Ducy, P., Karsenty, G., and Franceschi, R. T. (1997). Ascorbic acid-dependent activation of the osteocalcin promoter in MC3T3-E1 preosteoblasts: requirement for collagen matrix synthesis and the presence of an intact OSE2 sequence. Mol. Endocrinol. 11, 1103–1113. 55. Xiao, G., Wang, D., Benson, M. D., Karsenty, G., and Franceschi, R. T. (1998). Role of the alpha2-integrin in osteoblast-specific gene expression and activation of the Osf2 transcription factor. J Biol. Chem. 273, 32988–32994. 56. Xiao, G., Gopalakrishnan, R., Jiang, D., Reith, E., Benson, M. D., and Franceschi, R. T. (2002). Bone morphogenetic proteins, extracellular matrix, and mitogen-activated protein kinase signaling pathways are required for osteoblast-specific gene expression and differentiation in MC3T3-E1 cells. J. Bone Miner. Res. 17, 101–110. 57. Fässler, R., and Meyer, M. (1995). Consequences of lack of beta 1 integrin gene expression in mice. Genes Dev. 9, 1896–1908. 58. Stephens, L. E., Sutherland, A. E., Klimanskaya, I. V., Andrieux, A., Meneses, J., Pedersen, R. A., and Damsky, C. H. (1995). Deletion of beta 1 integrins in mice results in inner cell mass failure and periimplantation lethality. Genes Dev. 9, 1883–1895. 59. Zimmerman, D., Jin, F., Leboy, P., Hardy, S., and Damsky, C. (2000). Impaired bone formation in transgenic mice resulting from altered integrin function in osteoblasts. Dev. Biol. 220, 2–15. 60. Globus, R. K., Amblard, D., Nishimura, Y., Iwaniec, U. T., Kim, J. B., Almeida, E. A., Damsky, C. D., Wronski, T. J., and van der Meulen, M. C. (2004). Skeletal phenotype of growing transgenic mice that express a function-perturbing form of beta1 Integrin in Osteoblasts. Calcif. Tissue Int. 76, 39–49. 61. Chen, J., Diacovo, T. G., Grenache, D. G., Santoro, S. A., and Zutter, M. M. (2002). The alpha(2) integrin subunit-deficient mouse: a multifaceted phenotype including defects of branching morphogenesis and hemostasis. Am. J. Pathol. 161, 337–344.
146 62. Holtkötter, O., Nieswandt, B., Smyth, N., Müller, W., Hafner, M., Schulte, V., Krieg, T., and Eckes, B. (2002). Integrin alpha 2-deficient mice develop normally, are fertile, but display partially defective platelet interaction with collagen. J Biol. Chem. 277, 10789–10794. 63. Ekholm, E., Hankenson, K. D., Uusitalo, H., Hiltunen, A., Gardner, H., Heino, J., and Penttinen, R. (2002). Diminished callus size and cartilage synthesis in alpha 1 beta 1 integrin-deficient mice during bone fracture healing. Am. J. Pathol. 160, 1779–1785. 64. Mercurio, A. M. (2002). Lessons from the alpha2 integrin knockout mouse. Am. J. Pathol. 161, 3–6. 65. Bouvard, D., Brakebusch, C., Gustafsson, E., Aszodi, A., Bengtsson, T., Berna, A., and Fassler, R. (2001). Functional consequences of integrin gene mutations in mice. Circ. Res. 89, 211–223. 66. Sheppard, D. (2000). In vivo functions of integrins: lessons from null mutations in mice. Matrix Biol. 19, 203–209. 67. Wang, N., and Ingber, D. E. (1994). Control of cytoskeletal mechanics by extracellular matrix, cell shape and mechanical tension. Biophys. J. 66, 2181–2189. 68. Ruoslahti, E. (1997). Stretching is good for a cell. Science 276, 1345–1346. 69. Chen, C. S., Mrksich, M., Huang, S., Whitesides, G. M., and Ingber, D. E. (1997). Geometric control of cell life and death. Science 276, 1425–1428. 70. Noble, B. S., and Reeve, J. (2000). Osteocyte function, osteocyte death and bone fracture resistance. Mol. Cell Endocrinol. 159, 7–13. 71. Bonewald, L. F. (2004). Osteocyte biology: its implications for osteoporosis. J. Musculoskelet. Neuronal. Interact. 4, 101–104. 72. Charras, G. T., and Horton, M. A. (2002). Single cell mechanotransduction and its modulation analyzed by atomic force microscope indentation. Biophys. J. 82, 2970–2981. 73. Charras, G. T., and Horton, M. A. (2002). Determination of cellular strains by combined atomic force microscopy and finite element modeling. Biophys. J. 83, 858–879. 74. Burger, E. H., and Klein-Nulend, J. (1999). Mechanotransduction in bone–role of the lacuno-canalicular network. FASEB J. 13 Suppl, S101–S112. 75. Pommerenke, H., Schmidt, C., Durr, F., Nebe, B., Luthen, F., Muller, P., and Rychly, J. (2002). The mode of mechanical integrin stressing controls intracellular signaling in osteoblasts. J. Bone Miner. Res. 17, 603–611. 76. Norvell, S. M., Alvarez, M., Bidwell, J. P., and Pavalko, F. M. (2004). Fluid shear stress induces beta-catenin signaling in osteoblasts. Calcif. Tissue Int. 75, 396–404. 77. Millward-Sadler, S. J., and Salter, D. M. (2004). Integrin-dependent signal cascades in chondrocyte mechanotransduction. Ann. Biomed. Eng. 32, 435–446. 78. Damsky, C. H. (1999). Extracellular matrix-integrin interactions in osteoblast function and tissue remodeling. Bone 25, 95–96. 79. Siebers, M. C., ter Brugge, P. J., Walboomers, X. F., and Jansen, J. A. (2005). Integrins as linker proteins between osteoblasts and bone replacing materials. A critical review. Biomaterials 26, 137–146. 80. Hay, E., Lemonnier, J., Modrowski, D., Lomri, A., Lasmoles, F., and Marie, P. J. (2000). N- and E-cadherin mediate early human calvaria osteoblast differentiation promoted by bone morphogenetic protein-2. J. Cell Physiol. 183, 117–128. 81. Chen, S.-L., Zhang, S.-F., Zhang, L.-M., Warlow, P. M., Davidson, M. K., and Civitelli, R. (1995). Changing pattern of cadherin expression during human osteoblast differentiation. J. Bone Miner. Res. 10, S316. 82. Okazaki, M., Takeshita, S., Kawai, S., Kikuno, R., Tsujimura, A., Kudo, A., and Amann, E. (1994). Molecular cloning and characterization of OB-cadherin, a new member of cadherin family expressed in osteoblasts. J. Biol. Chem. 269, 12092–12098. 83. Prêle, C. M., Horton, M. A., Caterina, P., and Stenbeck, G. (2003). Identification of the molecular mechanisms contributing to polarized trafficking in osteoblasts. Exp. Cell Res. 282, 24–34.
M. H. HELFRICH AND M. A. HORTON 84. Marie, P. J. (2002). Role of N-cadherin in bone formation. J. Cell Physiol. 190, 297–305. 85. Ferrari, S. L., Traianedes, K., Thorne, M., Lafage-Proust, M. H., Genever, P., Cecchini, M. G., Behar, V., Bisello, A., Chorev, M., Rosenblatt, M., and Suva, L. J. (2000). A role for N-cadherin in the development of the differentiated osteoblastic phenotype. J. Bone Miner. Res. 15, 198–208. 86. Kawaguchi, J., Kii, I., Sugiyama, Y., Takeshita, S., and Kudo, A. (2001). The transition of cadherin expression in osteoblast differentiation from mesenchymal cells: consistent expression of cadherin-11 in osteoblast lineage. J. Bone Miner. Res. 16, 260–269. 87. Monaghan, H., Bubb, V. J., Sirimujalin, R., Millward-Sadler, S. J., and Salter, D. M. (2001). Adenomatous polyposis coli (APC), betacatenin, and cadherin are expressed in human bone and cartilage. Histopathology 39, 611–619. 88. Cheng, S. L., Lecanda, F., Davidson, M. K., Warlow, M. P., Zhang, S. F., Zhang, L., Suzuki, S., St. John, T., and Civitelli, R. (1998). Human osteoblasts express a repertoire of cadherins, which are critical for BMP-2 induced osteogenic differentiation. J. Bone Miner. Res. 13, 633–644. 89. Babich, M., and Foti, L. R. P. (1994). E-Cadherins identified in osteoblastic cells effects of parathyroid hormone and extracellular calcium on localization. Life Sci. 54, 201–208. 90. Cheng, S. L., Shin, C. S., Towler, D. A., and Civitelli, R. (2000). A dominant negative cadherin inhibits osteoblast differentiation. J. Bone Miner. Res. 15, 2362–2370. 91. Kawaguchi, J., Azuma, Y., Hoshi, K., Kii, I., Takeshita, S., Ohta, T., Ozawa, H., Takeichi, M., Chisaka, O., and Kudo, A. (2001). Targeted disruption of cadherin-11 leads to a reduction in bone density in calvaria and long bone metaphyses. J. Bone Miner. Res. 16, 1265–1271. 92. Radice, G. L., Rayburn, H., Matsunami, H., Knudsen, K. A., Takeichi, M., and Hynes, R. O. (1997). Developmental defects in mouse embryos lacking N-cadherin. Dev. Biol. 181, 64–78. 93. Castro, C. H., Shin, C. S., Stains, J. P., Cheng, S. L., Sheikh, S., Mbalaviele, G., Szejnfeld, V. L., and Civitelli, R. (2004). Targeted expression of a dominant-negative N-cadherin in vivo delays peak bone mass and increases adipogenesis. J. Cell Sci. 117, 2853–2864. 94. Civitelli, R., Lecanda, F., Jørgensen, N. R., and Steinberg, T. H. (2002). Intercellular junctions and cell-cell communication in bone. In “Principles of Bone Biology”, 2nd ed (J. P. Bilezikan, L. R. Raisz and G. A. Rodan, Eds), Vol I, pp 287–302. Academic Press, San Diego. 95. Di Palma, F., Chamson, A., Lafage-Proust, M. H., Jouffray, P., Sabido, O., Peyroche, S., Vico, L., and Rattner, A. (2004). Physiological strains remodel extracellular matrix and cell-cell adhesion in osteoblastic cells cultured on alumina-coated titanium alloy. Biomaterials 25, 2565–2575. 96. Kashima, T., Kawaguchi, J., Takeshita, S., Kuroda, M., Takanashi, M., Horiuchi, H., Imamura, T., Ishikawa, Y., Ishida, T., Mori, S., Machinami, R., and Kudo, A. (1999). Anomalous cadherin expression in osteosarcoma. Am. J. Pathol. 155, 1549–1555. 97. Iwao, K., Miyoshi, Y., Nawa, G., Yoshikawa, H., Ochi, T., and Nakamura, Y. (1999). Frequent β catenin abnormalities in bone and soft tissue tumours. Jap. J. Cancer Res. 90, 205–209. 98. Shin, C. S., Lecanda, F., Sheikh, S., Weitzmann, L., Cheng, S. L., and Civitelli, R. (2000). Relative abundance of different cadherins defines differentiation of mesenchymal precursors into osteogenic, myogenic, or adipogenic pathways. J. Cell Biochem. 78, 566–577. 99. Lee, Y. S., and Chuong, C. M. (1992). Adhesion molecules in skeletogenesis: I. Transient expression of neural cell adhesion molecules (NCAM) in osteoblasts during endochondral and intramembranous ossification. J. Bone Miner. Res. 7, 1435–1446. 100. Lackie, P. M., Zuber, C., and Roth, J. (1994). Polysialic acid of the neural cell adhesion molecule (N-CAM) is widely expressed during organogenesis in mesodermal and endodermal derivatives. Differentiation 57, 119–131.
Chapter 8 Integrins and Other Adhesion Molecules 101. Chimal-Monroy, J., and Diaz, d. L. (1999). Expression of N-cadherin, N-CAM, fibronectin and tenascin is stimulated by TGF-beta1, beta2, beta3 and beta5 during the formation of precartilage condensations. Int. J. Dev. Biol. 43, 59–67. 102. Reyes-Botella, C., Montes, M. J., Vallecillo-Capilla, M. F., Olivares, E. G., and Ruiz, C. (2002). Antigenic phenotype of cultured human osteoblast-like cells. Cell Physiol. Biochem. 12, 359–364. 103. Tanaka, Y., Maruo, A., Fujii, K., Nomi, M., Nakamura, T., Eto, S., and Minami, Y. (2000). Intercellular adhesion molecule 1 discriminates functionally different populations of human osteoblasts: characteristic involvement of cell cycle regulators. J. Bone Miner. Res. 15, 1912–1923. 104. Fujii, Y., Fujii, K., Nakano, K., and Tanaka, Y. (2003). Crosslinking of CD44 on human osteoblastic cells upregulates ICAM-1 and VCAM-1. FEBS Lett. 539, 45–50. 105. Tanaka, Y., Morimoto, I., Nakano, Y., Okada, Y., Hirota, S., Nomura, S., Nakamura, T., and Eto, S. (1995). Osteoblasts are regulated by the cellular adhesion through ICAM-1 and VCAM-1. J. Bone Miner. Res. 10, 1462–1469. 106. Okada, Y., Morimoto, I., Ura, K., Watanabe, K., Eto, S., Kumegawa, M., Raisz, L., Pilbeam, C., and Tanaka, Y. (2002). Cell-to-Cell adhesion via intercellular adhesion molecule-1 and leukocyte function-associated antigen-1 pathway is involved in 1alpha,25(OH)2D3, PTH and IL-1alpha-induced osteoclast differentiation and bone resorption. Endocr. J. 49, 483–495. 107. Harada, H., Kukita, T., Kukita, A., Iwamoto, Y., and Iijima, T. (1998). Involvement of lymphocyte function-associated antigen-1 and intercellular adhesion molecule-1 in osteoclastogenesis: a possible role in direct interaction between osteoclast precursors. Endocrinology 139, 3967–3975. 108. Kurachi, T., Morita, I., and Murota, S. (1993). Involvement of adhesion molecules LFA-1 and ICAM-1 in osteoclast development. Biochim. Biophys. Acta 1178, 259–266. 109. Barillé, S., Collette, M., Bataille, R., and Amiot, M. (1995). Myeloma cells upregulate interleukin-6 secretion in osteoblastic cells through cell-to-cell contact but downregulate osteocalcin. Blood 86, 3151–3159. 110. Feuerbach, D., and Feyen, J. H. (1997). Expression of the cell-adhesion molecule VCAM-1 by stromal cells is necessary for osteoclastogenesis. FEBS Lett. 402, 21–24. 111. Everts, V., Delaisse, J. M., Korper, W., Jansen, D. C., Tigchelaar-Gutter, W., Saftig, P., and Beertsen, W. (2002). The bone lining cell: its role in cleaning Howship’s lacunae and initiating bone formation. J. Bone Miner. Res. 17, 77–90. 112. Kurokouchi, K., Kambe, F., Yasukawa, K., Izumi, R., Ishiguro, N., Iwata, H., and Seo, H. (1998). TNF-alpha increases expression of IL-6 and ICAM-1 genes through activation of NF-kappaB in osteoblast-like ROS17/2.8 cells. J. Bone Miner. Res. 13, 1290–1299. 113. Lavigne, P., Benderdour, M., Lajeunesse, D., Shi, Q., and Fernandes, J. C. (2004). Expression of ICAM-1 by osteoblasts in healthy individuals and in patients suffering from osteoarthritis and osteoporosis. Bone 35, 463–470. 114. Kavanaugh, A. F., Schulze-Koops, H., Davis, L. S., and Lipsky, P. E. (1997). Repeat treatment of rheumatoid arthritis patients with a murine anti-intercellular adhesion molecule 1 monoclonal antibody. Arthritis Rheum. 40, 849–853. 115. Kavanaugh, A. F., Davis, L. S., Jain, R. I., Nichols, L. A., Norris, S. H., and Lipsky, P. E. (1996). A phase I/II open label study of the safety and efficacy of an anti-ICAM-1 (intercellular adhesion molecule-1; CD54) monoclonal antibody in early rheumatoid arthritis. J. Rheumatol. 23, 1338–1344. 116. Hall, B. K., and Miyake, T. (2000). All for one and one for all: condensations and the initiation of skeletal development. Bioessays 22, 138–147.
147 117. Koyama, E., Shimazu, A., Leatherman, J. L., Golden, E. B., Nah, H. D., and Pacifici, M. (1996). Expression of syndecan-3 and tenascin-C: possible involvement in periosteum development. J. Orthop. Res. 14, 403–412. 118. Imai, S., Kaksonen, M., Raulo, E., Kinnunen, T., Fages, C., Meng, X., Lakso, M., and Rauvala, H. (1998). Osteoblast recruitment and bone formation enhanced by cell matrix-associated heparin-binding growth-associated molecule (HB-GAM). J. Cell Biol. 143, 1113–1128. 119. Worapamorn, W., Tam, S. P., Li, H., Haase, H. R., and Bartold, P. M. (2002). Cytokine regulation of syndecan-1 and -2 gene expression in human periodontal fibroblasts and osteoblasts. J. Periodontal Res. 37, 273–278. 120. Birch, M. A., and Skerry, T. M. (1999). Differential regulation of syndecan expression by osteosarcoma cell lines in response to cytokines but not osteotropic hormones. Bone 24, 571–578. 121. Modrowski, D., Basle, M., Lomri, A., and Marie, P. J. (2000). Syndecan-2 is involved in the mitogenic activity and signaling of granulocyte-macrophage colony-stimulating factor in osteoblasts. J. Biol. Chem. 275, 9178–9185. 122. Liedert, A., Augat, P., Ignatius, A., Hausser, H. J., and Claes, L. (2004). Mechanical regulation of HB-GAM expression in bone cells. Biochem. Biophys. Res. Commun. 319, 951–958. 123. Hughes, D. E., Salter, D. M., and Simpson, R. (1994). CD44 expression in human bone: a novel marker of osteocytic differentiation. J. Bone Miner. Res. 9, 39–44. 124. Jamal, H. H., and Aubin, J. E. (1996). CD44 expression in fetal rat bone: in vivo and in vitro analysis. Exp. Cell Res. 223, 467–477. 125. Nakamura, H., Kenmotsu, S., Sakai, H., and Ozawa, H. (1995). Localization of CD44, the hyaluronate receptor, on the plasma membrane of osteocytes and osteoclasts in rat tibiae. Cell Tissue Res. 280, 225–233. 126. Noonan, K. J., Stevens, J. W., Tammi, R., Tammi, M., Hernandez, J. A., and Midura, R. J. (1996). Spatial distribution of CD44 and hyaluronan in the proximal tibia of the growing rat. J. Orthop. Res. 14, 573–581. 127. Pavasant, P., Shizari, T. M., and Underhill, C. B. (1994). Distribution of hyaluronan in the epiphysial growth plate: turnover by CD44expressing osteoprogenitor cells. J. Cell Sci. 107, 2669–2677. 128. Aarden, E. M., Wassenaar, A. M., Alblas, M. J., and Nijweide, P. J. (1996). Immunocytochemical demonstration of extracellular matrix proteins in isolated osteocytes. Histochem. Cell Biol. 106, 495–501. 129. Horton, M. A., Nesbitt, S. A., Bennett, J. H., and Stenbeck, G. (2002). Integrins and other cell surface attachment molecules of bone cells. In “Principles of Bone Biology”, 2nd ed (J. P. Bilezikan, L. R. Raisz and G. A. Rodan, Eds), Vol I, pp 265–286. Academic Press, San Diego. 130. Athanasou, N. A., and Quinn, J. (1990). Immunophenotypic differences between osteoclasts and macrophage polykaryons: immunohistological distinction and implications for osteoclast ontogeny and function. J. Clin. Pathol. 43, 997–1003. 131. Nesbitt, S., Nesbit, A., Helfrich, M., and Horton, M. (1993). Biochemical characterization of human osteoclast integrins. Osteoclasts express αvβ3, α2β1, and αvβ1 integrins. J. Biol. Chem. 268, 16737–16745. 132. Inoue, M., Teitelbaum, S. L., Hurter, L., Hruska, K., Seftor, E., Hendrix, M., and Ross, F. P. (1995). GM-CSF regulates expression of the functional integrins αvβ3 and αvβ5 in a reciprocal manner during osteoclastogenesis. J. Bone Miner. Res. 10, S163. 133. Inoue, M., Ross, F. P., Erdmann, J. M., Abu-Amer, Y., Wei, S., and Teitelbaum, S. L. (2000). Tumor necrosis factor alpha regulates alpha(v)beta5 integrin expression by osteoclast precursors in vitro and in vivo. Endocrinology 141, 284–290. 134. Nakamura, I., Gailit, J., and Sasaki, T. (1996). Osteoclast integrin αvβ3is present in the clear zone and contributes to cellular polarization. Cell Tissue Res. 286, 507–515.
148 135. Hultenby, K., Reinholt, F. P., and Heinegård, D. (1993). Distribution of integrin subunits on rat metaphyseal osteoclasts and osteoblasts. Eur. J. Cell Biol. 62, 86–93. 136. Väänänen, H. K., and Horton, M. (1995). The osteoclast clear zone is a specialized cell-extracellular matrix adhesion structure. J. Cell Sci. 108, 2729–2732. 137. Stenbeck, G. (2002). Formation and function of the ruffled border in osteoclasts. Semin. Cell Dev. Biol. 13, 285–292. 138. Helfrich, M. H., Nesbitt, S. A., Lakkakorpi, P. T., Barnes, M. J., Bodary, S. C., Shankar, G., Mason, W. T., Mendrick, D. L., Väänänen, H. K., and Horton, M. A. (1996). Beta 1 integrins and osteoclast function: involvement in collagen recognition and bone resorption. Bone 19, 317–328. 139. Stenbeck, G., and Horton, M. A. (2000). A new specialized cell-matrix interaction in actively resorbing osteoclasts. J. Cell Sci. 113 (Pt 9), 1577–1587. 140. Lakkakorpi, P. T., Helfrich, M. H., Horton, M. A., and Väänänen, H. K. (1993). Spatial organization of microfilaments and vitronectin receptor, αvβ3, in osteoclasts. A study using confocal laser scanning microscopy. J. Cell Sci. 104, 663–670. 141. Lakkakorpi, P. T., Horton, M. A., Helfrich, M. H., Karhukorpi, E. K., and Väänänen, H. K. (1991). Vitronectin receptor has a role in bone resorption but does not mediate tight sealing zone attachment of osteoclasts to the bone surface. J. Cell Biol. 115, 1179–1186. 142. Reinholt, F. P., Hultenby, K., Oldberg, Å., and Heinegård, D. (1990). Osteopontin—a possible anchor of osteoclasts to bone. Proc. Natl Acad. Sci. USA 87, 4473–4475. 143. Dodds, R. A., Connor, J. R., James, I. E., Rykaczewski, E. L., Appelbaum, E., Dul, E., and Gowen, M. (1995). Human osteoclasts, not osteoblasts, deposit osteopontin onto resorption surfaces: an in vitro and ex vivo study of remodeling bone. J. Bone Miner. Res. 10, 1666–1680. 144. Nakamura, I., Takahashi, N., Sasaki, T., Jimi, E., Kurokawa, T., and Suda, T. (1996). Chemical and physical properties of the extracellular matrix are required for the actin ring formation in osteoclasts. J. Bone Miner. Res. 11, 1873–1879. 145. Horton, M. A., Taylor, M. L., Arnett, T. R., and Helfrich, M. H. (1991). Arg-Gly-Asp (RGD) peptides and the anti-vitronectin receptor antibody 23C6 inhibit dentine resorption and cell spreading by osteoclasts. Exp. Cell Res. 195, 368–375. 146. van der Pluijm, G., Mouthaan, H., Baas, C., de Groot, H., Papapoulos, S., Löwik, C. (1994). Integrins and osteoclastic resorption in three bone organ cultures: differential sensitivity to synthetic Arg-Gly-Asp peptides during osteoclast formation. J. Bone Miner. Res. 9, 1021–1028. 147. Sato, M., Sardana, M. K., Grasser, W. A., Garsky, V. M., Murray, J. M., and Gould, R. J. (1990). Echistatin is a potent inhibitor of bone resorption in culture. J. Cell Biol. 111, 1713–1723. 148. Fisher, J. E., Caulfield, M. P., Sato, M., Quartuccio, H. A., Gould, R. J., Garsky, V. M., Rodan, G. A., and Rosenblatt, M. (1993). Inhibition of osteoclastic bone resorption in vivo by echistatin, an “arginylglycyl-aspartyl” (RGD)-containing protein. Endocrinology 132, 1411–1413. 149. Crippes, B. A., Engleman, V. W., Settle, S. L., Delarco, J., Ornberg, R. L., Helfrich, M.H., Horton, M. A., and Nickols, G. A. (1996). Antibody to β3 integrin inhibits osteoclast-mediated bone resorption in the thyroparathyroidectomized rat. Endocrinology 137, 918–924. 150. Engleman, V. W., Nickols, G. A., Ross, F. P., Horton, M. A., Griggs, D. W., Settle, S. L., Ruminski, P. G., and Teitelbaum, S. L. (1997). A peptidomimetic antagonist of the αvβ3 integrin inhibits bone resorption in vitro and prevents osteoporosis in vivo. J. Clin. Invest. 99, 2284–2292. 151. Shankar, G., Davison, I., Helfrich, M. H., Mason, W. T., and Horton, M. A. (1993). Integrin receptor-mediated mobilisation of intranuclear calcium in rat osteoclasts. J. Cell Sci. 105, 61–68.
M. H. HELFRICH AND M. A. HORTON 152. Zimolo, Z., Wesolowski, G., Tanaka, H., Hyman, J. L., Hoyer, J. R., and Rodan, G. A. (1994). Soluble αvβ3-integrin ligands raise [Ca2+]i in rat osteoclasts and mouse-derived osteoclast-like cells. Am. J. Physiol. 266, C376–C381. 153. Lakkakorpi, P. T., Wesolowski, G., Zimolo, Z., Rodan, G. A., and Rodan, S. B. (1996). Phosphatidylinositol-3 kinase (PI3-kinase) involvement in osteoclast attachment and spreading on bone. J. Bone Miner. Res. 11, S287. 154. McHugh, K. P., Hodivala-Dilke, K., Zheng, M. H., Namba, N., Lam, J., Novack, D., Feng, X., Ross, F. P., Hynes, R. O., and Teitelbaum, S. L. (2000). Mice lacking beta3 integrins are osteosclerotic because of dysfunctional osteoclasts. J. Clin. Invest. 105, 433–440. 155. Zhao, H., Ross, F. P., Teitelbaum, S. L., and Novack, D. V. (2002). Prevention of ovariectomy-induced bone loss in β3 knockout mice. J. Bone Miner. Res. 17 suppl 1, S133. 156. Saintier, D., Burde, M. A., Rey, J. M., Maudelonde, T., de Vernejoul, M. C., and Cohen-Solal, M. E. (2004). 17beta-estradiol downregulates beta3-integrin expression in differentiating and mature human osteoclasts. J. Cell Physiol. 198, 269–276. 157. Zhao, H., Ross, F. P., and Teitelbaum, S. L. (2004). Unoccupied αvβ3 integrin regulates osteoclast apoptosis by transmitting a positive death signal. Mol. Endocrinol. 19, 771–780. 158. Bader, B. L., Rayburn, H., Crowley, D., and Hynes, R. O. (1998). Extensive vasculogenesis, angiogenesis, and organogenesis precede lethality in mice lacking all alpha v integrins. Cell 95, 507–519. 159. Horton, M. A., Massey, H. M., Rosenberg, N., Nicholls, B., Seligsohn, U., and Flanagan, A. M. (2003). Upregulation of osteoclast alpha2beta1 integrin compensates for lack of alphavbeta3 vitronectin receptor in Iraqi-Jewish-type Glanzmann thrombasthenia. Br. J. Haematol. 122, 950–957. 160. Huang, X., Griffiths, M., Wu, J., Farese, R. V., Jr., and Sheppard, D. (2000). Normal development, wound healing, and adenovirus susceptibility in beta5-deficient mice. Mol. Cell Biol. 20, 755–759. 161. Lane, N. E., Yao, W., Nakamura, M. C., Humphrey, M. B., Kimmel, D., Huang, X., Sheppard, D., Ross, F. P., and Teitelbaum, S. L. (2005). Mice lacking the integrin beta5 subunit have accelerated osteoclast maturation and increased activity in the estrogen-deficient state. J. Bone Miner. Res. 20, 58–66. 162. Ross, F. P., Chappel, J., Alvarez, J. I., Sander, D., Butler, W. T., Farach-Carson, M. C., Mintz, K. A., Robey, P. G., Teitelbaum, S. L., and Cheresh, D. A. (1993). Interactions between the bone matrix proteins osteopontin and bone sialoprotein and the osteoclast integrin αvβ3 potentiate bone resorption. J. Biol. Chem. 268, 9901–9907. 163. Duong, L. T., Lakkakorpi, P., Nakamura, I., and Rodan, G. A. (2000). Integrins and signaling in osteoclast function. Matrix Biol. 19, 97–105. 164. Golden, L. H., and Insogna, K. L. (2004). The expanding role of PI3kinase in bone. Bone 34, 3–12. 165. Soriano, P., Montgomery, C., Geske, R., and Bradley, A. (1991). Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64, 693–702. 166. Lakkakorpi, P. T., Dedhar, S., Mostachfi-Vasseghi, H., Rodan, G. A., and Rodan, S. B. (1996). Association of integrin-linked kinase (ILK) with 3-integrin and its activation and translocation to the cytoskeleton upon osteoclast attachment to bone. J. Bone Miner. Res. 11, S287. 167. Ilic, D., Furuta, Y., Kanazawa, S., Takeda, N., Sobue, K., Nakatsuji, N., Nomura, S., Fujimoto, J., Okada, M., and Yamamoto, T. (1995). Reduced cell motility and enhanced focal adhesion contact formation in cells from FAK-deficient mice. Nature 377, 539–544. 168. Xiong, W. C., and Feng, X. (2003). PYK2 and FAK in osteoclasts. Front Biosci. 8, d1219–d1226. 169. Duong, L. T., Lakkakorpi, P. T., Nakamura, I., Machwate, M., Nagy, R. M., and Rodan, G. A. (1998). PYK2 in osteoclasts is an adhesion kinase, localized in the sealing zone, activated by ligation of
Chapter 8 Integrins and Other Adhesion Molecules
170.
171.
172.
173.
174.
175.
176.
177.
178.
179.
180. 181.
182.
183.
184.
185.
alpha(v)beta3 integrin, and phosphorylated by src kinase. J. Clin. Invest. 102, 881–892. Wang, Q., Xie, Y., Du, Q. S., Wu, X. J., Feng, X., Mei, L., McDonald, J. M., and Xiong, W. C. (2003). Regulation of the formation of osteoclastic actin rings by proline-rich tyrosine kinase 2 interacting with gelsolin. J. Cell Biol. 160, 565–575. Chellaiah, M., Kizer, N., Silva, M., Alvarez, U., Kwiatkowski, D., and Hruska, K. A. (2000). Gelsolin deficiency blocks podosome assembly and produces increased bone mass and strength. J. Cell Biol. 148, 665–678. Lakkakorpi, P. T., Bett, A. J., Lipfert, L., Rodan, G. A., and Duong, l. T. (2003). PYK2 autophosphorylation, but not kinase activity, is necessary for adhesion-induced association with c-Src, osteoclast spreading, and bone resorption. J. Biol. Chem. 278, 11502–11512. Sanjay, A., Houghton, A., Neff, L., DiDomenico, E., Bardelay, C., Antoine, E., Levy, J., Gailit, J., Bowtell, D., Horne, W. C., and Baron, R. (2001). Cbl associates with Pyk2 and Src to regulate Src kinase activity, alpha(v)beta(3) integrin-mediated signaling, cell adhesion, and osteoclast motility. J. Cell Biol. 152, 181–195. Miyazaki, T., Sanjay, A., Neff, L., Tanaka, S., Horne, W. C., and Baron, R. (2004). Src kinase activity is essential for osteoclast function. J. Biol. Chem. 279, 17660–17666. Nakamura, I., Lipfert, L., Rodan, G. A., and Le, T. D. (2001). Convergence of alpha(v)beta(3) integrin- and macrophage colony stimulating factor-mediated signals on phospholipase Cgamma in prefusion osteoclasts. J. Cell Biol. 152, 361–373. Mbalaviele, G., Chen, H., Boyce, B. F., Mundy, G., and Yoneda, T. (1995). The role of cadherin in the generation of multinucleated osteoclast from mononuclear precursors in murine marrow. J. Clin. Invest. 95, 2757–2765. Puch, S., Armeanu, S., Kibler, C., Johnson, K. R., Muller, C. A., Wheelock, M. J., and Klein, G. (2001). N-cadherin is developmentally regulated and functionally involved in early hematopoietic cell differentiation. J. Cell Sci. 114, 1567–1577. Ilvesaro, J. M., Lakkakorpi, P. T., and Vänäänen, H. K. (1998). Inhibition of bone resorption in vitro by a peptide containing the cadherin cell adhesion recognition sequence HAV is due to prevention of sealing zone formation. Exp. Cell Res. 242, 75–83. Flanagan, A. M., Sarma, U., Steward, C. G., Vellodi, A., and Horton, M. A. (2000). Study of the nonresorptive phenotype of osteoclast-like cells from patients with malignant osteopetrosis: a new approach to investigating pathogenesis. J. Bone Miner. Res. 15, 352–360. Kania, J. R., Kehat-Stadler, T., and Kupfer, S. R. (1997). CD44 antibodies inhibit osteoclast formation. J. Bone Miner. Res. 12, 1155–1164. Chellaiah, M. A., Kizer, N., Biswas, R., Alvarez, U., StraussSchoenberger, J., Rifas, L., Rittling, S. R., Denhardt, D. T., and Hruska, K. A. (2003). Osteopontin deficiency produces osteoclast dysfunction due to reduced CD44 surface expression. Mol. Biol. Cell 14, 173–189. Suzuki, K., Zhu, B., Rittling, S. R., Denhardt, D. T., Goldberg, H. A., McCulloch, C. A., and Sodek, J. (2002). Colocalization of intracellular osteopontin with CD44 is associated with migration, cell fusion, and resorption in osteoclasts. J. Bone Miner. Res. 17, 1486–1497. Cao, J. J., Singleton, P. A., Majumdar, S., Boudignon, B., Burghardt, A., Kurimoto, P., Wronski, T. J., Bourguignon, L. Y., and Halloran, B. P. (2005). Hyaluronan increases RANKL expression in bone marrow stromal cells through CD44. J. Bone Miner. Res. 20, 30–40. de Vries, T. J., Schoenmaker, T., Beertsen, W., van der, N. R., and Everts, V. (2004). Effect of CD44 deficiency on in vitro and in vivo osteoclast formation. J. Cell Biochem. 94, 954–966. Collin-Osdoby, P., Rothe, L., Anderson, F., Nelson, M., Maloney, W., and Osdoby, P. (2001). Receptor activator of NF-kappa B and osteoprotegerin expression by human microvascular endothelial cells, regulation by inflammatory cytokines, and role in human osteoclastogenesis. J. Biol. Chem. 276, 20659–20672.
149 186. McGowan, N. W., Walker, E. J., MacPherson, H., Ralston, S. H., and Helfrich, M. H. (2001). Cytokine-activated endothelium recruits osteoclast precursors. Endocrinology 142, 1678–1681. 187. Enomoto-Iwamoto, M., Iwamoto, M., Nakashima, K., Mukudai, Y., Boettinger, D., Pacifici, M., Kurisu, K., and Suzuki, F. (1997). Involvement of α5β1 integrin in matrix interactions and proliferation of chondrocytes. J. Bone Miner. Res. 12, 1124–1132. 188. Wang, H., and Kandel, R. A. (2004). Chondrocytes attach to hyaline or calcified cartilage and bone. Osteoarthritis. Cartilage. 12, 56–64. 189. Durr, J., Goodman, S., Potocnik, A., von der Mark, H., and von der Mark, K. (1993). Localization of β1-integrins in human cartilage and their role in chondrocyte adhesion to collagen and fibronectin. Exp. Cell Res. 207, 235–244. 190. Salter, D. M., Godolphin, J. L., and Gourlay, M. S. (1995). Chondrocyte heterogeneity: immunohistologically defined variation of integrin expression at different sites in human fetal knees. J. Histochem. Cytochem. 43, 447–457. 191. Woods, V. L., Jr., Schreck, P. J., Gesink, D. S., Pacheco, H. O., Amiel, D., Akeson, W. H., and Lotz, M. (1994). Integrin expression by human articular chondrocytes. Arthritis. Rheum. 37, 537–544. 192. Salter, D. M., Hughes, D. E., Simpson, R., and Gardner, D. L. (1992). Integrin expression by human articular chondrocytes. Brit. J. Rheumatol. 31, 231–234. 193. Loeser, R. F. (2000). Chondrocyte integrin expression and function. Biorheology 37, 109–116. 194. Hausler, G., Helmreich, M., Marlovits, S., and Egerbacher, M. (2002). Integrins and extracellular matrix proteins in the human childhood and adolescent growth plate. Calcif. Tissue Int. 71, 212–218. 195. Loeser, R. F., Carlson, C. S., and McGee, M. P. (1995). Expression of beta 1 integrins by cultured articular chondrocytes and in osteoarthritic cartilage. Exp. Cell Res. 217, 248–257. 196. Shakibaei, M. (1995). Integrin expression on epiphyseal mouse chondrocytes in monolayer culture. Histol. Histopathol. 10, 339–349. 197. Shakibaei, M., Abou-Rebyeh, H., and Merker, H. J. (1993). Integrins in ageing cartilage tissue in vitro. Histol. Histopathol. 8, 715–723. 198. Camper, L., Hellman, U., and Lundgren-Akerlund, E. (1998). Isolation, cloning, and sequence analysis of the integrin subunit alpha10, a beta1associated collagen binding integrin expressed on chondrocytes. J. Biol. Chem. 273, 20383–20389. 199. Camper, L., Holmvall, K., Wangnerud, C., Aszodi, A., and LundgrenAkerlund, E. (2001). Distribution of the collagen-binding integrin alpha10beta1 during mouse development. Cell Tissue Res. 306, 107–116. 200. Tiger, C. F., Fougerousse, F., Grundstrom, G., Velling, T., and Gullberg, D. (2001). alpha11beta1 integrin is a receptor for interstitial collagens involved in cell migration and collagen reorganization on mesenchymal nonmuscle cells. Dev. Biol. 237, 116–129. 201. Lehnert, K., Ni, J., Leung, E., Gough, S. M., Weaver, A., Yao, W. P., Liu, D., Wang, S. X., Morris, C. M., and Krissansen, G. W. (1999). Cloning, sequence analysis, and chromosomal localization of the novel human integrin alpha11 subunit (ITGA11). Genomics 60, 179–187. 202. Shakibaei, M., and Merker, H. J. (1999). Beta1-integrins in the cartilage matrix. Cell Tissue Res. 296, 565–573. 203. Lapadula, G., Iannone, F., Zuccaro, C., Grattagliano, V., Covelli, M., Patella, V., Lo Bianco, G., and Pipitone, V. (1997). Integrin expression on chondrocytes: correlations with the degree of cartilage damage in human osteoarthritis. Clin. Exp. Rheumatol. 15, 247–254. 204. Forster, C., Kociok, K., Shakibaei, M., Merker, H. J., Vormann, J., Gunther, T., and Stahlmann, R. (1996). Integrins on joint cartilage chondrocytes and alterations by ofloxacin or magnesium deficiency in immature rats. Arch. Toxicol. 70, 261–270. 205. Ostergaard, K., Salter, D. M., Petersen, J., Bendtzen, K., Hvolris, J., and Andersen, C. B. (1998). Expression of alpha and beta subunits of the integrin superfamily in articular cartilage from macroscopically
150
206.
207.
208.
209.
210.
211.
212.
213.
214.
215.
216.
217.
218.
219.
220.
221.
M. H. HELFRICH AND M. A. HORTON normal and osteoarthritic human femoral heads. Ann. Rheum. Dis. 57, 303–308. Millward-Sadler, S. J., Wright, M. O., Lee, H., Caldwell, H., Nuki, G., and Salter, D. M. (2000). Altered electrophysiological responses to mechanical stimulation and abnormal signalling through alpha5beta1 integrin in chondrocytes from osteoarthritic cartilage. Osteoarthritis. Cartilage. 8, 272–278. Hirsch, M. S., and Svoboda, K. K. (1996). Beta 1 integrin antibodies inhibit chondrocyte terminal differentiation in whole sterna. Ann. N. Y. Acad. Sci. 785, 267–270. Tavella, S., Bellesse, G., Castagnola, P., Martin, I., Piccini, D., Doliana, R., Colombatti, A., Cancedda, R., and Tacchetti, C. (1997). Regulated expression of fibronectin, laminin and related integrin receptors during the early chondrocyte differentiation. J. Cell Sci. 110, 2270–3361. Aszódi, A., Hunziker, E. B., Brakebusch, C., and Fassler, R. (2003). Beta1 integrins regulate chondrocyte rotation, G1 progression, and cytokinesis. Genes Dev. 17, 2465–2479. Lucchinetti, E., Bhargava, M. M., and Torzilli, P. A. (2004). The effect of mechanical load on integrin subunits alpha5 and beta1 in chondrocytes from mature and immature cartilage explants. Cell Tissue Res. 315, 385–391. Lee, H. S., Millward-Sadler, S. J., Wright, M. O., Nuki, G., and Salter, D. M. (2000). Integrin and mechanosensitive ion channeldependent tyrosine phosphorylation of focal adhesion proteins and beta-catenin in human articular chondrocytes after mechanical stimulation. J. Bone Miner. Res. 15, 1501–1509. Millward-Sadler, S. J., Wright, M. O., Lee, H., Nishida, K., Caldwell, H., Nuki, G., and Salter, D. M. (1999). Integrin-regulated secretion of interleukin 4: A novel pathway of mechanotransduction in human articular chondrocytes. J. Cell Biol. 145, 183–189. Wright, M. O., Nishida, K., Bavington, C., Godolphin, J. L., Dunne, E., Walmsley, S., Jobanputra, P., Nuki, G., and Salter, D. M. (1997). Hyperpolarisation of cultured human chondrocytes following cyclical pressure-induced strain: evidence of a role for alpha 5 beta 1 integrin as a chondrocyte mechanoreceptor. J. Orthop. Res. 15, 742–747. Enomoto, M., Leboy, P. S., Menko, A. S., and Boettiger, D. (1993). Beta 1 integrins mediate chondrocyte interaction with type I collagen, type II collagen, and fibronectin. Exp. Cell Res. 205, 276–285. Clancy, R. M., Rediske, J., Tang, X., Nijher, N., Frenkel, S., Philips, M., and Abrahamson, S. B. (1997). Outside-in signaling in the chondrocyte. Nitric oxide disrupts fibronectin-induced assembly of a subplasmalemmal actin/rho A/focal adhesion kinase signaling complex. J. Clin. Invest. 100, 1789–1796. Shimizu, M., Minakuchi, K., Kaji, S., and Koga, J. (1997). Chondrocyte migration to fibronectin, type I collagen, and type II collagen. Cell Struct. Funct. 22, 309–315. Homandberg, G. A., and Hui, F. (1994). Arg-Gly-Asp-Ser peptide analogs suppress cartilage chondrolytic activities of integrin-binding and nonbinding fibronectin fragments. Arch. Biochem. Biophys. 310, 40–48. Xie, D., and Homandberg, G. A. (1993). Fibronectin fragments bind to and penetrate cartilage tissue resulting in proteinase expression and cartilage damage. Biochem. Biophys. Acta 1182, 189–196. Loeser, R. F. (1993). Integrin-mediated attachment of articular chondrocytes to extracellular matrix proteins. Arthritis. Rheum. 36, 1103–1110. Camper, L., Heinegård, D., and Lundgren-Akerlund, E. (1997). Integrin α2β1 is a receptor for the cartilage matrix protein chondroadherin. J. Cell Biol. 138, 1159–1167. Holmvall, K., Camper, L., Johansson, S., Kimura, J. H., and Lundgren-Akerlund, E. (1995). Chondrocyte and chondrosarcoma cell integrins with affinity for collagen type II and their response to mechanical stress. Exp. Cell Res. 221, 496–503.
222. Loeser, R. F. (1997). Growth factor regulation of chondrocyte integrins. Differential effects of insulin-like growth factor 1 and transforming growth factor beta on α1β1 integrin expression and chondrocyte adhesion to type VI collagen. Arthritis. Rheum. 40, 270–276. 223. Ramachandrula, A., Tiku, K., and Tiku, M. L. (1992). Tripeptide RGD-dependent adhesion of articular chondrocytes to synovial fibroblasts. J. Cell Sci. 101, 859–871. 224. Loeser, R. F. (2002). Integrins and cell signaling in chondrocytes. Biorheology 39, 119–124. 225. Pulai, J. I., Del, C. M., Jr., and Loeser, R. F. (2002). The alpha5beta1 integrin provides matrix survival signals for normal and osteoarthritic human articular chondrocytes in vitro. Arthritis Rheum. 46, 1528–1535. 226. Cao, L., Lee, V., Adams, M. E., Kiani, C., Zhang, Y., Hu, W., and Yang, B. B. (1999). Beta-1 integrin-collagen interaction reduces chondrocyte apoptosis. Matrix Biol. 18, 343–355. 227. Hirsch, M. S., Lunsford, L. E., Trinkaus-Randall, V., and Svoboda, K. K. (1997). Chondrocyte survival and differentiation in situ are integrin mediated. Dev. Dyn. 210, 249–263. 228. Arner, E. C., and Tortorella, M. D. (1995). Signal transduction through chondrocyte integrin receptors induces matrix metalloproteinase synthesis and synergizes with interleukin-1. Arthritis Rheum. 38, 1304–1314. 229. Yonezawa, I., Kato, K., Yagita, H., Yamauchi, Y., and Okumura, K. (1996). VLA-5-mediated interaction with fibronectin induces cytokine production by human chondrocytes. Biochem. Biophys. Res. Commun. 219, 261–265. 230. Loeser, R. F. (1994). Modulation of integrin-mediated attachment of chondrocytes to extracellular matrix proteins by cations, retinoic acid, and transforming growth factor beta. Exp. Cell Res. 211, 17–23. 231. Grashoff, C., Aszodi, A., Sakai, T., Hunziker, E. B., and Fassler, R. (2003). Integrin-linked kinase regulates chondrocyte shape and proliferation. EMBO Rep. 4, 432–438. 232. Terpstra, L., Prud’homme, J., Arabian, A., Takeda, S., Karsenty, G., Dedhar, S., and St. Arnaud, R. (2003). Reduced chondrocyte proliferation and chondrodysplasia in mice lacking the integrin-linked kinase in chondrocytes. J. Cell Biol. 162, 139–148. 233. van der Kraan, P. M., Buma, P., van Kuppevelt, T., and van den Berg, W. B. (2002). Interaction of chondrocytes, extracellular matrix and growth factors: relevance for articular cartilage tissue engineering. Osteoarthritis. Cartilage. 10, 631–637. 234. Simonneau, L., Kitagawa, M., Suzuki, S., and Thiery, J. P. (1995). Cadherin 11 expression marks the mesenchymal phenotype: towards new functions for cadherins? Cell Adh. Commun. 3, 115–130. 235. Oberlender, S. A., and Tuan, R. S. (1994). Expression and functional involvement of N-cadherin in embryonic limb chondrogenesis. Development 120, 177–187. 236. Tavella, S., Raffo, P., Tacchetti, C., Cancedda, R., and Castagnola, P. (1994). N-CAM and N-cadherin expression during in vitro chondrogenesis. Exp. Cell Res. 215, 354–362. 237. DeLise, A. M., and Tuan, R. S. (2002). Alterations in the spatiotemporal expression pattern and function of N-cadherin inhibit cellular condensation and chondrogenesis of limb mesenchymal cells in vitro. J. Cell Biochem. 87, 342–359. 238. DeLise, A. M., and Tuan, R. S. (2002). Analysis of N-cadherin function in limb mesenchymal chondrogenesis in vitro. Dev. Dyn. 225, 195–204. 239. Tsonis, P. A., Del Rio-Tsonis, K., Millan, J. L., and Wheelock, M. J. (1994). Expression of N-cadherin and alkaline phosphatase in chick limb bud mesenchymal cells: regulation by 1,25-dihydroxyvitamin D3 or TGF-beta 1. Exp. Cell Res. 213, 433–437. 240. Hitselberger Kanitz, M. H., Ng, Y. K., and Iannaccone, P. M. (1993). Distribution of expression of cell adhesion molecules in the mid to late gestational mouse fetus. Pathobiol. 61, 13–18. 241. Bujia, J., Behrends, U., Rotter, N., Pitzke, P., Wilmes, E., and Hammer, C. (1996). Expression of ICAM-1 on intact cartilage and
Chapter 8 Integrins and Other Adhesion Molecules
242.
243.
244.
245.
246.
247.
248.
249.
250.
251.
isolated chondrocytes. In Vitro Cell. Develop. Biol. Animal. 32, 116–122. Kienzle, G., and von Kempis, J. (1998). Vascular cell adhesion molecule 1 (CD106) on primary human articular chondrocytes: functional regulation of expression by cytokines and comparison with intercellular adhesion molecule 1 (CD54) and very late activation antigen 2. Arthritis Rheum. 41, 1296–1305. Davies, M. E., Dingle, J. T., Pigott, R., Power, C., and Sharma, H. (1991). Expression of intercellular adhesion molecule 1 (ICAM-1) on human articular cartilage chondrocytes. Connect Tissue Res. 26, 207–216. Seidel, M. F., Keck, R., and Vetter, H. (1997). ICAM-1/LFA-1 expression in acute osteodestructive joint lesions in collagen-induced arthritis in rats. J. Histochem. Cytochem. 45, 1247–1253. Horner, A., Davies, M. E., and Franz, B. (1995). Chondrocyte-peripheral blood mononuclear cell interactions: the role of ICAM-1. Immunol. 86, 584–590. Shimazu, A., Nah, H. D., Kirsch, T., Koyama, E., Leatherman, J. L., Golden, E. B., Kosher, R. A., and Pacifici, M. (1996). Syndecan-3 and the control of chondrocyte proliferation during endochondral ossification. Exp. Cell Res. 229, 126–136. Seghatoleslami, M. R., and Kosher, R. A. (1996). Inhibition of in vitro limb cartilage differentiation by syndecan-3 antibodies. Dev. Dyn. 207, 114–119. Koyama, E., Leatherman, J. L., Shimazu, A., Nah, H. D., and Pacifici, M. (1995). Syndecan-3, tenascin-C, and the development of cartilaginous skeletal elements and joints in chick limbs. Dev. Dyn. 203, 152–162. Kirsch, T., Koyama, E., Liu, M., Golub, E. E., and Pacifici, M. (2002). Syndecan-3 is a selective regulator of chondrocyte proliferation. J. Biol. Chem. 277, 42171–42177. Pfander, D., Swoboda, B., and Kirsch, T. (2001). Expression of early and late differentiation markers (proliferating cell nuclear antigen, syndecan-3, annexin VI, and alkaline phosphatase) by human osteoarthritic chondrocytes. Am. J. Pathol. 159, 1777–1783. Stevens, J. W., Noonan, K. J., Bosch, P. P., Rapp, T. B., Martin, J. A., Kurriger, G. L., Maynard, J. A., Daniels, K. J., Solursh, M., Tammi, R., Tammi, M., and Midura, R. J. (1996). CD44 in growing normal and neoplastic rat cartilage. Ann. N. Y. Acad. Sci. 785, 333–336.
151 252. Knudson, W., and Loeser, R. F. (2002). CD44 and integrin matrix receptors participate in cartilage homeostasis. Cell Mol. Life Sci. 59, 36–44. 253. Salter, D. M., Godolphin, J. L., Gourlay, M. S., Lawson, M. F., Hughes, D. E., and Dunne, E. (1996). Analysis of human articular chondrocyte CD44 isoform expression and function in health and disease. J. Pathol. 179, 396–402. 254. Knudson, C. B. (1993). Hyaluronan receptor-directed assembly of chondrocyte pericellular matrix. J. Cell Biol. 120, 825–834. 255. Knudson, W., Aguiar, D. J., Hua, Q., and Knudson, C. B. (1996). CD44-anchored hyaluronan-rich pericellular matrices: an ultrastructural and biochemical analysis. Exp. Cell Res. 228, 216–228. 256. Knudson, C. B., and Knudson, W. (2004). Hyaluronan and CD44: modulators of chondrocyte metabolism. Clin. Orthop. S152–S162. 257. Chow, G., Knudson, C. B., Homandberg, G., and Knudson, W. (1995). Increased expression of CD44 in bovine articular chondrocytes by catabolic cellular mediators. J. Biol. Chem. 270, 27734–27741. 258. Hua, Q., Knudson, C. B., and Knudson, W. (1993). Internalization of hyaluronan by chondrocytes occurs via receptor-mediated endocytosis. J. Cell Sci. 106, 365–375. 259. Takagi, T., Okamoto, R., Suzuki, K., Hayashi, T., Sato, M., Sato, M., Kurosaka, N., and Koshino, T. (2001). Up-regulation of CD44 in rheumatoid chondrocytes. Scand. J. Rheumatol. 30, 110–113. 260. Helfrich, M. H., Nesbitt, S. A., Dorey, E. L., and Horton, M. A. (1992). Rat osteoclasts adhere to a wide range of RGD (Arg-Gly-Asp) peptide-containing proteins, including the bone sialoproteins and fibronectin, via a β3 integrin. J. Bone Miner. Res. 7, 335–343. 261. Nakamura, I., Pilkington, M. F., Lakkakorpi, P. T., Lipfert, L., Sims, S. M., Dixon, S. J., Rodan, G. A., and Duong, L. T. (1999). Role of alpha(v)beta(3) integrin in osteoclast migration and formation of the sealing zone. J. Cell Sci. 112, 3985–3993. 261a.Murphy, M. G., Cerchio, K., Stoch, S. A., Gottesdiener, K., Wu, M., Recker, R., and L-000845704 Study Group. (2005). Effect of L-000845704, an alphaVbeta3 integrin antagonist, on markers of bone turnover and bone mineral density in postmenopausal osteoporotic women. J. Clin. Endocrinol. Metab. 90, 2022–2228.
This Page Intentionally Left Blank
Chapter 9
Alkaline Phosphatases José Luis Millán
The Burnham Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037-1005, USA
I. Introduction II. Structure and Regulation of the TNAP Gene III. Protein Structure
IV. Function of TNAP V. Clinical Use References
I. INTRODUCTION
completion of the mouse genome project has revealed the presence of a yet uncharacterized mouse AP gene, which this author has named Akp6. The mouse Akp-ps1, Akp5, Akp6, and Akp3 loci, in this spatial order, are clustered in the same region of mouse chromosome 1 [1]. The mRNA molecules for all AP isozymes, human or mouse, are in the order of 2.5 kb in length and encode peptides ranging from 518 to 535 amino acids. While little new data have become available concerning the regulation of the AP genes in the last 5 years, a wealth of new information has been published regarding the protein structure and the function of mammalian APs. This chapter will focus particularly on the structure and function of TNAP, making reference to these new studies and mentioning briefly studies on other AP isozymes when there is paucity of data for TNAP or to exemplify structural features relevant to TNAP activity and function.
The alkaline phosphatase (E.C.3.1.3.1; AP) isozyme expressed in bone is one of several members of the mammalian AP gene family. In humans, APs are encoded by four genes traditionally named after the tissues where they are predominantly expressed, although the gene nomenclature is now gaining wider use (Table I). The tissuenonspecific AP (TNAP) gene (ALPL), located on chromosome 1, is expressed at its highest levels in liver, bone, and kidney (hence the alternative name “L/B/K” AP), in the placenta during the 1st trimester of pregnancy, and at lower levels in numerous other tissues [1]. The other three isozymes, i.e., placental (PLAP), placental-like or germ cell (GCAP), and intestinal AP (IAP), show a much more restricted tissue expression; hence, the general term tissuespecific APs. These isozymes are encoded by three genes (ALPP, ALPP2, and ALPI, respectively) clustered on human chromosome 2, bands q34–q37 [2, 3] and are closely related to one another, showing 90% and 87% identical nucleotide and amino acid sequences, respectively. The orthologous TNAP gene in mice is called Akp2 and is located on mouse chromosome 4 (Table I). The mouse Akp3 gene encodes the IAP isozyme and the mouse Akp5 gene encodes the embryonic AP (EAP) isozyme, that appears to be related to both the human PLAP and GCAP isozymes [4]. In addition to the existence of the inactive pseudoAP gene Akp-ps1 previously described [4], the recent Dynamics of Bone and Cartilage Metabolism
II. STRUCTURE AND REGULATION OF THE TNAP GENE The human TNAP gene (ALPL; NM_000478) is located on the distal short arm of chromosome 1, band 1p36.12 specifically at position chr1: 21581175-21650208, thus occupying a length of 69034 bp. The gene is composed of 13 exons; the first two exons (1a and 1b) are noncoding and are separated from one another and from the exon 153
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
154
JOSÉ LUIS MILLA´ N Table I.
Genes Human genes ALPL
Protein names TNAP
ALPP
PLAP
ALPP2
GCAP
ALPI Mouse genes Akp2 Akp3 Akp5 Akp-ps1 Akp6*
Nomenclature of the Human and Mouse Alkaline Phosphatase Genes and Isozymes
IAP TNAP IAP EAP N/a uprIAP*
Common names
Chromosomal location
Tissue-nonspecific alkaline phosphatase; TNSALP; “liver-bone-kidney type” AP Placental alkaline phosphatase; PLALP; “Regan isozyme” Germ cell alkaline phosphatase, GCALP; “placentallike”; “Nagao isozyme” Intestinal alkaline phosphatase, IALP Tissue-nonspecific alkaline phosphatase; TNSALP; “liver-bone-kidney type” AP Intestinal alkaline phosphatase, IALP Embryonic alkaline phosphatase AP Pseudogene, pseudoAP RIKEN sequence, New AP locus
Accession numbers
chr1:21581174-21650208
NM_001631
chr2:233068964-233073097
NM_001632
Chr2:233097057-233100922
NM_031313
Chr2:233146369-233150245
NM_001631
chr4:136199753-136254338
NM_007431
chr1:87031694-87555136 chr1:86990248-86993641 chr1:86968828-86972484 chr1:87002298-87005230
NM_007432 NM_007433 NG 001240 AK008000
*Names tentatively assigned by the author.
that contains the ATG translation initiation site (exon 2) by relatively large introns, as first demonstrated in the rat [5, 6]. The last exon contains the termination codon and the 3′ untranslated region of the mRNA. The TNAP genes and promoter regions have been cloned and sequenced from human, rat, and mouse [5–10]. Exon 1a in humans and rats shares approximately 66% identical bases [5, 6], while sequence homology cannot be detected between the 1b exon in these two species. The major transcription start site and surrounding sequences of the 5′-most promoter (1a) upstream from the 1a exon of the TNAP gene has been determined for the human [5–7, 11] and mouse [9] TNAP gene. The 1b exon and its promoter (1b) were first recognized in rat [5, 6] and have also been identified and sequenced in the human and mouse genes [8, 9, 12–14]. Importantly, exons 1a and 1b are incorporated into the mRNA in a mutually exclusive fashion that results from the fact that each exon has its own promoter sequence. This results in two types of mRNA, each encoding an identical polypeptide, but having different 5′ untranslated sequences [5, 6, 10, 12, 15]. The promoter usage has primarily been inferred by examination of the structure of TNAP cDNAs isolated from various tissues. In humans and rats, the upstream promoter (1a) is preferentially utilized in osteoblasts and the downstream promoter (1b) preferred in kidney and liver. However, in mice, the upstream promoter is used in all tissues expressing significant amounts of TNAP [9, 10]. Although the 5′ flanking regions of human, rat, and mouse TNAP have been analyzed in transient transfection assays [16, 17], more work is needed to elucidate the mechanism(s) of tissue-specific regulation. Studies using agents that increase TNAP activity have provided evidence for both transcriptional and post-transcriptional regulation
of the enzyme [18, 19]. Retinoic acid, dexamethasone, and granulocyte colony-stimulating factor all promote transcription of TNAP from the upstream 1a promoter and there is evidence that TNAP transcription in nonexpressing tissues of mice is repressed in vivo by methylation at the 1a promoter [11, 20–22]. Regulation of TNAP expression by the forkhead transcription factor (FKHR) via a forkhead response element in the TNAP promoter has been demonstrated [23]. The TNAP promoter appears to be repressed by Smad-interacting protein 1 [24], while Smad3 promotes TNAP expression and mineralization in osteoblastic cell lines [25]. There is also evidence for regulation of TNAP at translational or post-translational stages [10, 19, 26–29]. Since the expression of TNAP in osteoblasts is linked to differentiation, some agents which cause elevated enzyme levels in these cells may do so by increasing expression of osteoblast-specific transcription factors, or relieving osteogenic suppression, rather than acting directly on the TNAP gene. Osteosarcoma/fibroblast cell hybrids regain elevated TNAP when treated with a combination of 1,25-dihydroxyvitamin D3 and TGF, suggesting that this combination diminishes repression of the TNAP gene [30]. Other reagents, such as ascorbic acid (vitamin C) and the bone morphogenetic proteins, may also act indirectly by stimulating osteogenic pathways. For example, ascorbic acid has been shown to increase TNAP mRNA in MC3T3-E1 cells by a slow mechanism requiring increased extracellular collagen synthesis which, in turn, promotes increased osteoblast differentiation [31]. It has been suggested that this mechanism involves activation of a promoter element by the osteogenic transcription factor OSF-2/CBFA-1 [32]. It is clear that detailed analyses of TNAP promoter function using cell transfection and
Chapter 9 Alkaline Phosphatases
trangenic techniques will be necessary to define the regulatory sequences controlling TNAP expression in bone and other tissues.
III. PROTEIN STRUCTURE APs catalyze the hydrolysis of monoesters of phosphoric acid and also catalyze a transphosphorylation reaction in the presence of large concentrations of phosphate acceptors [1]. APs occur widely in nature, and are found in many organisms from bacteria to man. Irrespective of their origin, APs are homodimeric enzymes and each catalytic site contains three metal ions, i.e., two Zn and one Mg, necessary for enzymatic activity. While the main features of the catalytic mechanism are conserved between bacterial and mammalian APs, mammalian APs have higher specific activity and Km values; have a more alkaline pH optimum; are membrane-bound; are inhibited by L-amino acids and peptides through an uncompetitive mechanism; and display lower heat stability. These properties, however, differ noticeably among isozymes. For many years E. coli AP [33] was the only source of structural information on APs, but now the threedimensional structure of the first mammalian AP, i.e., human PLAP (1EW2; http://pdbbeta.rcsb.org/pdb/; [34]) has been solved. The availability of the PLAP structure facilitated modeling the human GCAP, IAP, and TNAP isozymes, revealing that all the novel features discovered in PLAP are conserved in those human isozymes as well [35]. As had been predicted from sequence comparisons, the central core of PLAP, consisting of an extended β-sheet and flanking α-helices, is very similar to that of the E. coli enzyme. The same is true in the immediate vicinity of the three catalytic ions. However, a number of distinctive features, including a different positioning of the aminoterminal segment of the molecule and the expanded top-loop or “crown” domain are now very apparent (Fig. 1). Isozyme-specific properties, such as the characteristic uncompetitive inhibition [36–40], their variable heatstability [41], and their allosteric behavior [42], have been attributed to residues located on the top crown domain unique to mammalian APs (Fig. 1, upper panel). This domain is also responsible for the binding of TNAP to collagen [41, 43], and it may be involved in the function of this isozyme during bone mineralization. Both the amino terminal arm and the crown domain take part in stabilizing the AP dimeric structure. Modeling of the structure of TNAP, GCAP, and IAP allowed the examination of the residues that constitute the monomer–monomer interphase [35]. The overall surface buried at the interface varies between 4134 and 4244 Å2 per monomer, which
155 corresponds to about 25% of the overall protein surface and comprises about 90 residues per monomer. The comparison of the interfaces reveals high conservation in the tissuespecific APs compatible with the fact that PLAP/GCAP or PLAP/IAP heterodimers form readily in nature [44, 45]. In TNAP, however, a number of charged substitutions lead to repulsive forces at the interface, incompatible with the formation of heterodimers between TNAP and any of the tissue-specific APs [35]. An additional noncatalytic metal-binding site, not present in E. coli AP, was revealed in the PLAP structure that appears to be occupied by calcium [34, 46]. The ligands to the calcium site in PLAP include residues W248, F269, E270, D285, and Trp248 that interacts with the calcium through a very stable water molecule. This fourth metal site is conserved in all human and mouse APs and presumably represents a novel feature common to all mammalian APs. However, the structural and functional significance of this new metal site remains to be established. Recent structure-function studies comparing PLAP and the E. coli AP structure have found a conserved function for those residues that stabilize the active site Zn and Mg metal ions [39]. Mutations at residues that coordinate the Zn2 or Mg ions, i.e., D42, S155, E311, D357, and H358, had similar effects in PLAP and E. coli AP, but the environment of the Zn1 ion, i.e., D316, H320, and H432, in PLAP is less affected by substitutions than in E. coli AP. This is due to the identity and location of residue 429 (E429 in PLAP) that covers the entrance to the active site and prevents diffusion of the metal ions despite the introduction of destabilizing mutations at the coordinating residues (Fig. 1, bottom left panel). As will be seen below, residue 429 has crucial significance for the uncompetitive inhibition of other mammalian APs, including TNAP [36–40]. All mammalian APs have five cysteine residues (C101, C121, C183, C467, and C474 in PLAP) per subunit, not homologous to any of the four cysteines in E. coli AP. Disrupting the disulfide bond between C121 and C183 completely prevents the formation of the active enzyme while the C-terminally located C467–C474 bond plays a lesser structural role. The substitution of the free C101 does not significantly affect the properties of the enzyme [39]. Several isozyme-specific inhibitors of APs have been reported. They include L-aminoacids, such as L-phenylalanine, L-tryptophan, L-leucine, L-homoarginine [47, 48], as well as some nonrelated compounds, such as levamisole, the L-stereoisomer of tetramisole [49] and theophylline (a 1,3-dimethyl derivative of xanthine) [50]. The inhibition is of a rare uncompetitive type [51] and, while the biological implications of this inhibition are not known, the inhibitors have proven to be useful in the
156
JOSÉ LUIS MILLA´ N
Figure 1 Three-dimensional structure of mammalian APs. Top panel: Modeled structure of human TNAP dimer [35] based on the 3D crystallographic coordinates of PLAP [34]. One subunit is colored magenta and the other white, both in backbone representation. The location of the top “crown domain” is indicated. The site of GP1 anchor attachment, the amino terminal arm and the active site, are only indicated for one subunit but are present in both. The active site metals and phosphate (CPK colors) are indicated. Bottom left panel: Active site environment of PLAP in the vicinity of the Zn1 and Zn2 metal sites and their ligands. Water molecules are shown as red spheres. Green dotted lines denote metal–ligand interactions and hydrogen bonds. Bottom right panel: Top view of the entrance to the PLAP active site, showing the position of the Y367 residue from one subunit (wireframe representation) in the immediate vicinity of the E429 residue of the other subunit (space filling representation). The lower left and right panels were initially published in [39] and are reproduced here with permission from the Journal of Biological Chemistry.
differential determination of AP isozymes in clinical chemistry [52, 53]. The selective uncompetitive inhibition mechanism of PLAP and GCAP by L-Phe and L-Leu is extraordinarily dependent on residue 429. A single E429G substitution in GCAP accounts for the sensitivity of GCAP for L-Leu and the reduced heat stability and decreased Km of this AP isozyme [36–38]. The inhibition occurs through three interaction points of the inhibitor, i.e., the carboxylic group of L-Phe or L-Leu attacks the
active site Arg166 during catalysis, the amino group of the inhibitor interacts with Zn1 in the active site, while the side chain of the inhibitor is stabilized by the loop containing residue 429 (Fig. 1, lower left panel) [36–39]. Another interesting novel structural feature of mammalian APs, also relevant to our understanding of AP inhibition, is Y367. This residue is part of the subunit interface in the PLAP dimer, where it protrudes from one subunit and is positioned within 5.6 Å of the catalytic Zn1
157
Chapter 9 Alkaline Phosphatases
ion in the active site of the other subunit (Fig. 1, lower right panel). Mutagenesis of this residue leads to PLAP mutants with significantly compromised heat and inhibition properties [39]. Recently, Kozlenkov et al. [40] clarified which amino acid residues are responsible for the marked differences in inhibition selectivity between TNAP and PLAP. In total, six TNAP residues were identified that cluster into two groups; the first group includes residues 433 and 434, and the second group includes residues 108, 109, 120, and 166 (using TNAP numbers). Residue 108 in TNAP largely determines the specificity of inhibition by L-homoarginine, while the conserved Tyr-371 (equivalent to Y367 in PLAP) contributes to a lesser degree to L-homoarginine positioning. In contrast, the binding of levamisole is mostly dependent on His-434 and Tyr-371, but not on residues 108 or 109, while the main determinant of sensitivity to theophylline is His-434 [40]. These studies open the door to drug design efforts aimed at developing more specific inhibitors of TNAP that could be used therapeutically to treat conditions of ectopic calcification where TNAP is known to play a role (see Section IV). As already mentioned, TNAP is also known as liver/ bone/kidney AP since these are tissues known to express high levels of TNAP. These different isoforms differ in their post-translational modifications of the carbohydrate side-chains. The amino acid sequence of the TNAP gene suggests five putative N-glycosylation sites in TNAP, i.e., N123, N213, N254, N286, and N413 [54]. Nosjean et al. [55] have confirmed that both the bone and liver TNAP isoforms are N-glycosylated; however, they did not determine the number of N-glycosylation sites that actually are linked to oligosaccharides. Their data further suggest that the main difference between the bone and liver isoforms is due to a difference in O-glycosylation, i.e., bone-derived TNAP has some O-linked oligosaccharides, whereas liverderived TNAP does not, but that the different immunoreactivity between the bone and liver isoforms is mainly due to differences in N-glycosylation [55]. It has also been reported that bone-derived TNAP contains more fucose [56] and sialic acid residues [55] compared with the liver isoform. The structures of the N-linked and O-linked oligosaccharides and other oligosaccharide differences of bone and liver TNAP have not been reported. Finally, while E. coli AP is located in the periplasmic space of the bacterium, mammalian APs are ectoenzymes bound to the plasma membrane via a glycosylphosphatidylinositol (GPI) anchor [57]. Asp484 was identified as the point of attachment of this anchor in PLAP [58]. Since Asp484 is perfectly conserved in GCAP and IAP we can predict the same attachment site for these human isozymes. Experimental evidence also indicates that the bovine IAPs may be attached via residue 480 [59].
However, the corresponding anchor residue for TNAP has not yet been identified (Fig. 1, upper panel). Pertinent to our discussion of skeletal mineralization, the bone isoform of TNAP functions as an ectoenzyme attached to the osteoblast cell membrane via its GPI anchor [60-63]. In vitro studies have demonstrated that bone AP is released from human osteoblast-line cells in an anchor-intact (insoluble) form attached to matrix vesicles [64, 65], where it participates in the initiation of bone matrix mineralization [66–69]. In vivo, bone AP circulates in an anchordepleted (soluble) homodimeric form, indicating the conversion of the insoluble to the soluble form found in serum by the actions of two endogenous circulating phospholipases, GPI-specific phospholipase C and GPI-specific phospholipase D. Both of these cleavage enzymes are present in humans; but phospholipase C is found in lower levels in serum compared with phospholipase D, which is abundant in the circulation [70–74].
IV. FUNCTION OF TNAP The identification and study of inborn-errors-ofmetabolism often provides invaluable clues as to the in vivo function of a molecule. In humans, only deficiencies in the ALPL gene have been observed. There are no reported cases of deficiencies in the ALPP, ALPP2, or ALPI genes, so the in vivo functions of the PLAP, GCAP, and IAP isozymes remain unclear. However, some clues may be derived from gene-targeting experiments in mice. Functional deletion of the Akp3 gene leads to an accelerated transport of fat through the intestinal epithelium suggesting that IAP may be involved in a rate-limiting step during fat absorption [75]. Deletion of the Akp5 gene leads to a 12-hour delay in pre-implantation development and increased embryo lethality in utero [76]. Whether similar functions could be attributable to the human orthologs remains to be determined. However, missense mutations in the human ALPL gene lead to the inborn-error-of metabolism known as hypophosphatasia [77] and studies of this disease, both in humans and in mouse models, have provided compelling evidence for an important role for TNAP during the development and mineralization of the human skeleton. Hypophosphatasia represents a rare form of rickets and osteomalacia in which neither calcium nor inorganic phosphate levels in serum are subnormal [78]. In fact hypercalcemia and hyperphosphatemia may exist and hypercalciuria is common in infantile hypophosphatasia. The clinical severity in hypophosphatasia patients varies widely. The different syndromes, listed from the most severe to the mildest forms, are: perinatal hypophosphatasia, infantile hypophosphatasia,
158 childhood hypophosphatasia, adult hypophosphatasia, odontohypophosphatasia, and pseudohypophosphatasia [78]. These phenotypes range from complete absence of bone mineralization and stillbirth to spontaneous fractures and loss of decidual teeth in adult life. The severity and expressivity of hypophosphatasia depends on the nature of the ALPL mutation [79–81]. The mapping of hypophosphatasia mutations to specific three-dimensional locations on the TNAP-molecule has provided clues as to the structural significance of these areas for enzyme structure and function [46]. In fact most mutations causing severe hypophosphatasia were mapped to five crucial regions on the TNAP-modeled structure, namely the active site and its vicinity, the active site valley, the homodimer interface, the crown domain, and the metal-binding site (see Section III). Inactivation of the mouse TNAP gene (Akp2) phenocopies the severe “infantile hypophosphatasia” phenotype. The Akp2−/− mice display rickets and osteomalacia at about 6–10 days after birth; they develop extensive epileptic seizures and suffer from apnea, increased apoptosis in the thymus, and abnormal lumbar nerve roots [82] and die around postnatal days 12–15. Biochemically, these animals show increased levels of inorganic pyrophosphate (PPi), pyridoxal-5′-phosphate (PLP, a hydropholic form of vitamin B6), and phosphoethanolamine (PEA) [82–84], molecules that have been proposed as natural substrates of TNAP [77]. Administration of pyridoxal (a hydrophobic form of vitamin B6 that can easily traverse biological membranes), temporarily suppresses the epileptic seizures [83], and reverses the apoptosis in the thymus and the morphology of the lumbar nerve roots [85]. However, the Akp2−/− mice still show a 100% mortality rate before weaning and their demise is often preceded by epileptic seizures. Vitamin B6 is an important coenzyme in several biochemical reactions, including the biosynthesis of the neurotransmitters γ-aminobutyric acid (GABA), dopamine, and serotonin, and is likely to be important for the normal perinatal development of the central nervous system [86]. It is known that vitamin B6 depletion causes epilepsy in rats accompanied by reduced PLP levels in the brain [87]. In general, the epilepsy is believed to involve the GABAergic synapses. Glutamic acid decarboxylase (GAD) catalyzes the synthesis of GABA requiring PLP as a cofactor, and knockout mice lacking one of the GAD isozymes, GAD65, developed spontaneous epilepsy in the adult stage [88]. Waymire et al. [83] showed that the Akp2−/− mice had reduced brain levels of GABA and hypothesized that the observed epileptic seizures result from GAD dysfunction due to shortage of PLP. Vitamin B6 deficiency in suckling rats has been found to result in a decrease in brain sphingolipids since sphingosine synthesis is also a PLP dependent reaction [89].
JOSÉ LUIS MILLA´ N
We observed abnormal morphology of the lumbar nerve roots in both Akp2−/− and vitamin B6-deficient mice during the suckling stage when myelination is rapidly occurring [82, 85]. It is possible that impaired myelination of nerve cells in the brain and nerve roots descending within the dura may be a contributing factor to the development of epileptic attacks in these mice. While abnormalities in the metabolism of PLP explain many of the abnormalities of infantile hypophosphatasia, it is not the basis of the abnormal mineralization that characterizes this disease. In bone, TNAP is confined to the cell surface of osteoblasts and chondrocytes, including the membranes of their shed matrix vesicles (MVs) [90, 91]. In fact, by an unknown mechanism, MVs are markedly enriched in TNAP compared to both whole cells and the plasma membrane [91]. It has been proposed that the role of TNAP in the bone matrix is to generate the inorganic phosphate needed for hydroxyapatite crystallization [92–94]. However, TNAP has also been hypothesized to hydrolyze the mineralization inhibitor PPi to facilitate mineral precipitation and growth [95, 96]. Electron microscopy observations revealed that TNAP-deficient MVs contain apatite-like mineral crystals, but that extravesicular crystal propagation is retarded [97]. This growth retardation could be due to either the lack of TNAP’s pyrophosphatase function or the lack of inorganic phosphate generation. Recent studies have provided compelling proof that the function of TNAP in bone tissue consists in hydrolyzing PPi to maintain a proper concentration of this mineralization inhibitor to ensure normal bone mineralization [98–100]. The nucleotidetriphosphate pyrophosphohydrolase activity of NPP1 and the transmembrane PPi-channeling protein ANK are responsible for supplying the larger amount of PPi to the extracellular spaces. Mice deficient in NPP1 (Enpp1−/−) or ANK (ank/ank), have decreased levels of extracellular PPi and display soft tissue ossification. Enpp1−/− mice develop features essentially identical to the previously described phenotype of the tiptoe walking (ttw/ttw) mice [101, 102]. These include the development of hyperostosis, starting at approximately 3 weeks of age, in a progressive process that culminates in ossific intervertebral fusion and peripheral joint ankylosis, as well as Achilles tendon calcification. The ank/ank mice have also been characterized as a model of ankylosis [103, 104]. We surmised that affecting the function of either NPP1 or ANK would have beneficial consequences on hypophosphatasia by reducing the amounts of extracellular PPi in the Akp2−/− mice and, conversely, that affecting the function of TNAP should also ameliorate the soft tissue ossification in the Enpp1−/− and ank/ank mutant animals. Indeed, experiments to rescue these abnormalities by combining
159
Chapter 9 Alkaline Phosphatases
two mutations, i.e., [Akp2−/−; Enpp1−/−] and [Akp2−/−; ank/ank] revealed that both double mutants led to improvement of hypophosphatasia as well as of the ossification of the vertebral apophyses. We found that even carriers (heterozygotes) of Akp2, Enpp1, and Ank mutations were affected to some degree, demonstrating gene dosage effects for all three genes [99, 100]. Interestingly, the expression of yet another mineralization inhibitor, i.e., osteopontin (OPN), was highly elevated in Akp2−/− mice while it was decreased in both the Enpp1−/− and the ank/ank osteoblasts [102]. Importantly, both PPi and OPN levels were corrected in [Akp2−/−; Enpp1−/−] and [Akp2−/−; ank/ank] double knockout mice (Fig. 2). In vitro experiments on wt osteoblasts treated with exogenous PPi revealed an increase in OPN expression and decreased NPP1 and ANK expression [100]. These studies provided evidence for a direct regulation of OPN expression by NPP1 and ANK, mediated by PPi and enabled us to propose a model for the concerted action of the molecules that produce, transport, and degrade PPi. We proposed that under normal conditions the concerted action of TNAP, NPP1, and ANK regulate PPi and OPN levels and thus control hydroxyapatite deposition. Hypophosphatasia in the Akp2−/− mice arises from deficits in TNAP activity, resulting in an increase in PPi levels and a concomitant increase in OPN levels; the combined inhibitory effect of these molecules leads to hypomineralization. In contrast, an NPP1 or ANK deficiency leads to a decrease in the extracellular PPi and OPN
pools, thereby enabling ectopic soft tissue ossification (Fig. 3). The hypomineralization defects in Akp2−/− mice, along with elevated PPi and OPN levels are normalized by ablation of either the NPP1 or ANK gene. Conversely, ablating the function of TNAP causes normalization of the abnormalities in the Enpp1−/− and ank/ank mutant mice via the resulting increase in the concentrations of two inhibitors of mineralization, i.e., PPi and OPN. Thus, NPP1 and ANK represent possible therapeutic targets to correct abnormally elevated levels of PPi, such as those
Figure 3
Figure 2 Correlation between serum PPi and OPN levels. The elevated levels of PPi in Akp2−/− mice cause a secondary increase in OPN, whereas the decreased PPi concentrations in Enpp1−/− and ank/ank mice result in depressed OPN levels. Therefore there is a strong correlation between PPi concentration and serum OPN levels. The double knockout [Akp2−/−; Enpp1−/−] and [Akp2−/−; ank/ank] mice both show normalized PPi levels that also result in correction of OPN levels.
Concerted action of TNAP, NPP1, and ANK in regulating extracellular PPi and OPN levels. Both NPP1 and ANK raise extracellular levels of PPi while TNAP is required for depletion of the PPi pool. Both TNAP and NPP1 are functional in matrix vesicles whereas ANK is not; therefore, NPP1 plays a more crucial role in PPi production than ANK. As a result, the absence of NPP1 in Enpp1−/− mice results in a more severe phenotype than in ank/ank mice. A negative feedback loop exists in which PPi, produced by NPP1 and transported by the channeling action of ANK, inhibits expression of the Enpp1 and Ank genes. In addition, PPi induces expression of the Opn gene and production of OPN, which further inhibits mineralization. In the absence of TNAP, high levels of PPi inhibit mineral deposition directly and also via its induction of OPN expression. The combined action of increased concentrations of PPi and OPN causes hypomineralization. In the absence of NPP1 or ANK, low levels of PPi, in addition to a decrease in OPN levels, leads to hypermineralization. This model clearly points to NPP1 and ANK as therapeutic targets for the treatment of hypophosphatasia. Similarly, targeting TNAP function can be useful in the treatment of hypermineralization abnormalities caused by altered PPi metabolism. This figure was originally published in [100] and is reproduced here with permission from the American Journal of Pathology.
160 found in hypophosphatasia, and TNAP appears as a rational target molecule for the treatment of soft tissue ossification abnormalities due to decreased levels of extracellular PPi, including ankylosis, osteoarthritis, and arterial calcification, since the same molecules implicated in skeletal mineralization appear to also be involved in the pathological calcification of the arteries [105, 106]. It is clear that PLP and PPi are physiological substrates of TNAP and that abnormalities in their metabolism cause the epileptic seizures and hypomineralization, respectively, in hypophosphatasia. Interestingly, several hypophosphatasia mutations were found to affect PLP catalysis to a larger extent than PPi catalysis [81], providing a clue as to one mechanism to explain phenotypic variability in hypophosphatasia, i.e., some individuals suffer from epileptic seizures while displaying mild bone abnormalities, while others have the reverse manifestations [78]. While pyridoxal administration improves the lifespan of the Akp2−/− mice, they still succumb to the disease at about day 25–30. Even [Akp2−/−; Enpp1−/−] mice, that display corrected skeletal abnormalities, die at about this same time point. The ultimate cause of death in the mice has not yet been identified. At the time of death the animals have considerable respiratory distress and episodes of apnea. As discussed above it is possible that anatomic disturbances in the central nervous system resulting from lack of TNAP activity during embryonic development that are not correctable by vitamin B6 supplementation account for the late epilepsy and death [85]. Alternatively, a recent report has pointed to the potential role of TNAP in dephosphorylating AMP to produce adenosine, needed for proper mucociliary clearance and inflammatory responses in the airways that help prevent lung infections [107]. It is possible that abnormalities in adenosine production in the lungs of Akp2−/− mice contribute to their demise and this possibility should be tested experimentally. There is also some anecdotal evidence that PEA may be epileptogenic in hypophosphatasia patients [108]. The especially great elevations of endogenous PEA levels in the Akp2−/− mice could cause this complication as the mice age, which would then not be correctable by vitamin B6 administration. Meanwhile, the reasons for the increased excretion of PEA in hypophosphatasia remain unclear. One possibility is that indeed PEA is a natural substrate of TNAP [109], but the possible metabolic pathway has not been elucidated. An alternative explanation may relate to abnormalities in the function of O-phosphorylethanolamine phosphoylase (PEA-P-lyase) [110, 111]. PEA-P-lyase is an enzyme reported to require PLP as cofactor and, since TNAP is crucial for the normal metabolism of PLP [85], it is possible that it is the insufficient amount of PLP inside the cells that leads to suboptimal activity of
JOSÉ LUIS MILLA´ N
PEA-P-lyase, which in turn leads to increased excretion of PEA in hypophosphatasia. While one study has addressed this issue [112] and did not find support for this hypothesis, in this author’s opinion there is merit in further exploring this possible metabolic pathway as being affected in hypophosphatasia.
V. CLINICAL USE Measurement of bone-derived TNAP activity in serum can provide an index of the rate of osteoblastic bone formation [50, 113] (see also Chapter 34). It has, however, been difficult to analyze the bone-specific fraction of AP in serum due to the co-existing liver-specific fraction of AP, which has similar physical and biochemical characteristics, e.g., the same primary structure. A vast number of techniques and methods have been developed for separation and quantification of the bone and liver isoforms, such as heat inactivation [114], various chemical inhibitors (e.g., L-homoarginine and levamisole) [49, 115], wheatgerm lectin precipitation [116], electrophoresis [117], isoelectric focusing [118], HPLC [119], and different immunoassays [120]. The development of commercial immunoassays for routine assessment of the bone-specific TNAP isoform [121–125] has indeed increased the availability and use of bone AP, particularly as a biochemical marker of bone turnover for monitoring therapy in osteoporotic patients [126–128]. In healthy adults, the bone and liver TNAP isoforms constitute approximately 95% of the total serum AP activity with similar quantitative levels [129]. At least six different AP isoform peaks can be separated and quantified by weak anion-exchange high-performance liquid chromatography (HPLC) in serum from healthy adults: three bone isoform peaks (B/I, B1, and B2) and three liver isoform peaks (L1, L2, and L3) isoforms [117, 128]. In healthy adults, the three bone isoforms, B/I, B1, and B2, account on average for 4, 16, and 37%, respectively, of the total serum AP activity [128]. In serum, the minor fraction B/I is not a pure bone isoform as it co-elutes with the IAP isozyme and is composed, on average, of 70% bone and 30% IAP. The circulating levels of these bone isoform peaks can vary independently during the pubertal growth spurt and in metabolic bone diseases [130, 131]. Some anatomical differences in the skeletal content of the bonespecific isoforms have also been reported; cortical bone had approximately two-fold higher activity of B1 compared with B2 and, conversely, B2 was approximately two-fold higher in trabecular bone compared with B1. However, the significance of these finding in terms of function and clinical utility remain to be determined.
Chapter 9 Alkaline Phosphatases
Acknowledgments This work was supported by grants AR47908 and DE12889 from the National Institutes of Health, USA.
References 1. Millán, J. L. (2006). Mammalian alkaline phosphatases. From biology to applications in medicine and bio-technology. Wiley-VCH Verlag GmbH & Co., Weinheim, Germany. 2. Smith, M., Weiss, M. J., Griffin, C. A., Murray, J. C., Buetow, K. H., Emanuel, B. S., Henthorn, P. S., and Harris, H. (1988). Regional assignment of the gene for human liver/bone/kidney alkaline phosphatase to chromosome 1p36.1-p34. Genomics 2, 139–143. 3. Martin, D., Tucker, D. F., Gorman, P. A., Sheer, D., Spurr, N. K., and Trowsdale, J. (1987). The human placental alkaline phosphatase gene and related sequences map to chromosome 2 band q37. Ann. Hum. Gen. 51, 145–152. 4. Manes, T., Glade, K., Ziomek, C. A., and Millán, J. L. (1990). Genomic structure and comparison of mouse tissue-specific alkaline phosphatase genes. Genomics 8, 541–554. 5. Toh, Y., Yamamoto, M., Endo, H., Fujiyata, A., Misumi, Y., and Ikehara, Y. (1989a). Sequence divergence of 5’ extremities in rat liver alkaline phosphatase mRNAs. J. Biochem. 105, 61–65. 6. Toh, Y., Yamamoto, M., Endo, H., Misumi, Y., and Ikehara, Y. (1989b). Isolation and characterization of a rat liver alkaline phosphatase gene: a single gene with two promoters. Eur. J. Biochem. 182, 231–237. 7. Weiss, M. J., Ray, K., Henthorn, P. S., Lamb, B., Kadesch, T., and Harris, H. (1988). Structure of the human liver/bone/kidney alkaline phosphatase gene. J. Biol. Chem. 263, 12002–12010. 8. Matsuura, S., Kishi, F., and Kajii, T. (1990). Characterization of a 5’flanking region of the human liver/bone/kidney alkaline phosphatase gene: two kinds of mRNA from a single gene. Biochem. Biophys. Res. Comm. 168, 993–1000. 9. Terao, M., Studer, M., Giannì, M., and Garattini, E. (1990). Isolation and characterization of the mouse liver/bone/kidney-type alkaline phosphatase gene. Biochem. J. 268, 641–648. 10. Studer, M., Terao, M., Giannì, M., and Garattini, E. (1991). Characterization of a second promoter for the mouse liver/bone/kidneytype alkaline phosphatase gene: Cell and tissue specific expression. Biochem. Biophys. Res. Comm. 179, 1352–1360. 11. Zernik, J., Thiede, M. A., Twarog, K., Stover, M. L., Rodan, G. A., Upholt, W. B., and Rowe, D. W. (1990). Cloning and analysis of the 5’ region of the rat bone/liver/kidney/placenta alkaline phosphatase gene – a dual function promoter. Matrix 10, 38–47. 12. Kishi, F., Matsuura, S., and Kaji, T. (1989). Nucleotide sequence of the human liver-type alkaline phosphatase cDNA. Nucl. Acids Res. 17, 2129. 13. Uhler, M.D., and Stofko, R.E. (1990). Induction of alkaline phosphatase in mouse L cells by overexpression of the catalytic subunit of cAMP-dependent protein kinase. J. Biol. Chem. 265, 13181–13189. 14. Terao, M., and Mintz, B. (1987). Cloning and characterization of a cDNA coding for mouose placental alkaline phosphatase. Proc. Natl Acad. Sci. USA 84, 7051–7055. 15. Zernik, J., Kream, B., and Twarog, K. (1991). Tissue-specific and dexamethasone-inducible expression of alkaline phosphatase from alternative promoters of the rat bone/liver/kidney/placenta gene. Biochem. Biophys. Res. Commun. 176, 1149–1156. 16. Kiledjian, M., and Kadesch, T. (1990). Analysis of the human liver/bone/kidney alkaline phosphatase promoter in vivo and in vitro. Nucl. Acids Res. 18, 957–961. 17. Kiledjian, M., and Kadesch, T. (1991). Post-transcriptional regulation of the human liver/bone/kidney alkaline phosphatase gene. J. Biol. Chem. 266, 4207–4213.
161 18. Rodan, G.A., and Rodan, S.B. (1983). Expression of the osteoblastic phenotype. In: Bone and Mineral Research, ed. W.A. Peck, Vol 2, pp. 244–285. Elsevier, Amsterdam. 19. Sato, N., Takahashi, Y., and Asano, S. (1994). Preferential usage of the bone-type leader sequence for the transcripts of liver/bone/kidney-type alkaline phosphatase gene in neutrophilic granulocytes. Blood 83, 1093–1101. 20. Heath, J. K., Suva, L. J., Yoon, K., Kiledjian, M., Martin, T. J., and Rodan, G. A. (1992). Retinoic acid stimulates transcriptional activity from the alkaline phosphatase promoter in the immortalized rat calvarial cell line, RCT-1. Mol. Endo. 6, 636–646. 21. Weiss, M. J., Ray, K., Fallon, M. D., Whyte, M. P., Fedde, K. N., Lafferty, M. A., Mulivor, R., and Harris, H. (1989). Analysis of liver/ bone/kidney alkaline phosphatase mRNA, DNA, and enzymatic activity in cultured skin fibroblasts from 14 unrelated patients with severe hypophosphatasia. Am. J. Hum. Genet. 44, 686–694. 22. Escalante-Alcalde, D., Recillas-Targa, F., Hernández-García, D., Castro-Obregón, S., Terao, M., Garattini, E., and Covarrubias, L. (1996). Retinoic acid and methylation cis-regulatory elements control the mouse tissue non-specific alkaline phosphatase gene expression. Mech. Dev. 57, 21–32. 23. Hatta, M., Daitoku, H., Matsuzaki, H., Deyama, Y., Yoshimura, Y., Suzuki, K., Matsumoto, A., and Fukamizu, A. (2002). Regulation of alkaline phosphatase promoter by forkhead transcription factor FKHR. Int. J. Mol. Med. 9, 147–152. 24. Tylzanowski, P., Verschueren, K., Huylebroeck, D., and Luyten, F. P. (2001). Smad-interacting protein 1 is a repressor of liver/bone/kidney alkaline phosphatase transcription in bone morphogenetic proteininduced osteogenic differentiation of C2C12 cells. J. Biol. Chem. 276, 40001–40007. 25. Sowa, H., Kaji, H., Yamaguchi, T., Sugimoto, T., and Chihara, K. (2002). Smad3 promotes alkaline phosphatase activity and mineralization of osteoblastic MC3T3-E1 cells. J. Bone Min. Res. 17, 1190–1199. 26. Leboy, P. S., Beresford, J. N., Devlin, C., and Owen, M. E. (1991). Dexamethasone induction of osteoblast mRNAs in rat marrow stromal cell cultures. J. Cell. Physiol. 146, 370–378. 27. Kyeyune-Nyombi, E., Nicolas, V., Strong, D. D., and Farley, J. (1995). Paradoxical effects of phosphate to directly regulate the level of skeletal alkaline phosphatase activity in human osteosarcoma (SaOS-2) cells and inversely regulate the level of skeletal alkaline phosphatase mRNA. Calcif. Tissue Int. 56, 154–159. 28. Kobayashi, T., and Robinson, J. M. (1991). A novel intracellular compartment with unusual secretory properties in human neutrophils. J. Cell. Biol. 113, 743–756. 29. Cain, T. J., Liu, Y., Takizawa, T., and Robinson, J. M. (1995). Solubilization of glycosyl-phosphatidylinositol-anchored proteins in quiescent and stimulated neutrophils. Biochim. Biophys. Acta 1235, 69–78. 30. Johnson-Pais, T. L., and Leach, R. J. (1996). 1,25-dihydroxyvitamin D3 and transforming growth factor-β act synergistically to override extinction of liver/bone/kidney alkaline phosphatase in osteosarcoma hybrid cells. Exp. Cell Res. 226, 67–74. 31. Franceschi, R. T., Iyer, B. S., and Cui, Y. (1994). Effects of ascorbic acid on collagen matrix formation and osteoblast differentiation in murine MC3T3-E1 cells. J. Bone Min. Res. 9, 843–854. 32. Xiao, G., Cui, Y., Ducy, P., Karsenty, G., and Franceschi, R. T. (1997). Ascorbic-acid-dependent activation of the osteocalcin promoter in MC3T3-E1 preosteoblasts: Requirement for collagen matrix synthesis and the presence of an intact OSE2 sequence. Mol. Endo. 11, 1103–1113. 33. Kim, E. E., and Wyckoff, H. W. (1991). Reaction mechanism of alkaline phosphatase based on crystal structures: two metal ions catalysis. J. Mol. Biol. 218, 449–464. 34. Le Du, M. H., Stigbrand, T., Taussig, M. J., Menez, A., and Stura, E. A. (2001). Crystal structure of alkaline phosphatase from human placenta
162
35.
36.
37.
38.
39.
40.
41.
42. 43.
44.
45.
46.
47.
48.
49.
50. 51. 52. 53.
54.
at 1.8 A resolution. Implication for a substrate specificity. J. Biol. Chem. 276, 9158–9165. Le Du, M-H., and Millán, J. L. (2002). Structural evidence of functional divergence in human alkaline phosphatases. J. Biol. Chem. 277, 49808–49814. Hummer, C., and Millán, J. L. (1991). Gly429 is the major determinant of uncompetitive inhibition of human germ cell alkaline phosphatase by L-leucine. Biochem. J. 274, 91–95. Hoylaerts, M. F., and Millán, J. L. (1991). Site-directed mutagenesis and epitope-mapped monoclonal antibodies define a catallytically important conformational difference between human placental and germ cell alkaline phosphatase. Eur. J. Biochem. 202, 605–616. Hoylaerts, M. F., Manes, T., and Millan, J. L. (1992). Molecular mechanism of uncompetitive inhibition of human placental and germ cell alkaline phosphatase. Biochem. J. 286, 23–30. Kozlenkov, A., Manes, T., Hoylaerts, M. F., and Millán, J. L. (2002). Function assignment to conserved residues in mammalian alkaline phosphatases. J. Biol. Chem. 277, 22992–22999. Kozlenkov, A., Hoylaerts, M. F., Ny, T., Le Du, M-H., and Millán, J. L. (2004). Residues determining the binding specificity of uncompetitive inhibitors to tissue-nonspecific alkaline phosphatase. J. Bone Min. Res. 19, 1862–1872. Bossi, M., Hoylaerts, M. F., and Millán, J. L. (1993). Modifications in a flexible surface loop modulate the isozyme-specific properties of mammalian alkaline phosphatases. J. Biol. Chem. 268, 25409–25416. Hoylaerts, M. F., Manes, T., and Millán, J. L. (1997). Mammalian alkaline phosphatases are allosteric enzymes. J. Biol. Chem. 272, 22781–22787. Wu, L. N. Y., Genge, B. R., Lloyd, G. C., and Wuthier, R. E. (1991). Collagen binding proteins in collagenase-released matrix vesicles from cartilage. J. Biol. Chem. 266, 1195–1203. Behrens, C. M., Enns, C. A., and Sussman, H. H. (1983). Characterization of human foetal intestinal alkaline phosphatase. Comparison with the isoenzymes from the adult intestine and human tumour cell lines. Biochem. J. 211, 553–558. Imanishi, H., Hada, T., Muratani, K., Hirano, K., and Higashino, K. (1990). Expression of a hybrid form of alkaline phosphatase isoenzyme in a newly established cell line (HuG-1) from a gastric cancer patient. Cancer Res. 50, 3408–3412. Mornet, E., Stura, E., Lia-Baldini, A. S., Stigbrand, T., Menez, A., and Le Du, M. H. (2001). Structural evidence for a functional role of human tissue-nonspecific alkaline phosphatase in bone mineralization. J. Biol. Chem. 276, 31171–31178. Fishman, W. H., and Sie, H.-G. (1971). Organ-specific inhibition of human alkaline phosphatase isozymes of liver, bone, intestine, and placenta: L-phenylalanine, L-tryptophan and L-homoarginine. Enzymologia 41, 140–167. Doellgast, G. J., and Fishman, W. H. (1976). L-Leucine, a specific inhibitor of a rare human placental alkaline phosphatase phenotype. Nature 259, 49–51. Van Belle, H. (1976). Alkaline phosphatase. I. Kinetics and inhibition by levamisole of purified isoenzymes from humans. Clin. Chem. 22, 972–976. Farley, J. R., and Baylink, D. J. (1986). Skeletal alkaline phosphatase activity as a bone formation index in vitro. Metabolism 35, 563–571. Cornish-Bowden, A. (1995). Fundamentals of Enzyme Kinetics, revised Edition, Chap. 5, p 100. Portland Press Ltd, London. Fishman, W. H. (1974). Perspectives on alkaline phosphatase isoenzymes. Am. J. Med. 56, 617–650. Mulivor, R. A., Plotkin, L. I., and Harris, H. (1978). Differential inhibition of the products of the human alkaline phosphatase loci. Ann. Hum. Genet. 42, 1–13. Weiss, M. J., Henthorn, P. S., Lafferty, M. A., Slaughter, C., Raducha, M., and Harris, H. (1986). Isolation and characterization of a cDNA encoding a human liver/bone/kidney-type alkaline phosphatase. Proc. Natl Acad. Sci. USA 83, 7182–7186.
JOSÉ LUIS MILLA´ N 55. Nosjean, O., Koyama, I., Goseki, M., Roux, B., and Komoda, T. (1997). Human tissue non-specific alkaline phosphatases: sugar-moietyinduced enzymic and antigenic modulations and genetic aspects. Biochem. J. 321, 297–303. 56. Kuwana, T., Sugita, O., and Yakata, M. (1991). Sugar chain heterogeneity of bone and liver alkaline phosphatase in serum. Enzyme 45, 63–66. 57. Low, M. G., and Saltiel, A. R. (1988). Structural and functional roles of glycosylphosphatidylinositol in membranes. Science 239, 268–275. 58. Micanovic, R., Bailey, C. A., Brink, L., Gerber, L., Pan, Y. C. E., Hulmes, J. D., and Udenfriend, S. (1988). Aspartic acid-484 of nascent placental alkaline phosphatase condenses with a phosphatidylinositol glycan to become the carboxyl terminus of the mature enzyme. Proc. Natl Acad. Sci. USA 85, 1398–1402. 59. Manes, T., Hoylaerts, M. F., Muller, R., Lottspeich, F., Holke, W., and Millán, J. L. (1998). Genetic complexity, structure, and characterization of highly active bovine intestinal alkaline phosphatases. J. Biol. Chem. 273, 23353–23360. 60. Hawrylak, K., and Stinson, R. A. (1988). The solubilization of tetrameric alkaline phosphatase from human liver and its conversion into various forms by phosphatidylinositol phospholipase C or proteolysis. J. Biol. Chem. 263, 14368–14373. 61. Hamilton, B. A., McPhee, J. L., Hawrylak, K., and Stinson, R.A. (1990). Alkaline phosphatase releasing activity in human tissues. Clin. Chim. Acta. 186, 249–254. 62. Fedde, K. N., Lane, C. C., and Whyte, M. P. (1988). Alkaline phosphatase is an ectoenzyme that acts on micromolar concentrations of natural substrates at physiologic pH in human osteosarcoma cells. Arch. Biochem. Biophys. 264, 400–409. 63. Hooper, N. M. (1997). Glycosyl-phosphatidylinositol anchored membrane enzymes. Clin. Chim. Acta. 266, 3–12. 64. Fedde, K. N. (1992). Human osteosarcoma cells spontaneously release matrix-vesicle-like structures with the capacity to mineralize. Bone Min. 17, 145–151. 65. Anh, D. J., Dimai, H. P., Hall, S. L., and Farley, J. R. (1998). Skeletal alkaline phosphatase activity is primarily released from human osteoblasts in an insoluble form, and the net release is inhibited by calcium and skeletal growth factors. Calcif. Tissue Int. 62, 332–340. 66. Tenenbaum, H. C. (1987). Levamisole and inorganic pyrophosphate inhibit β-glycerophosphate induced mineralization of bone formed in vitro. Bone Min. 3, 13–26. 67. Bellows, C. G., Aubin, J. E., and Heersche, J. N. M. (1991). Initiation and progression of mineralization of bone nodules formed in vitro: the role of alkaline phosphatase and organic phosphate. Bone Min. 14, 27–40. 68. Barling, P. M., Gupta, D. K., and Lim, C. E. L. (1999). Involvement of phosphodiesterase I in mineralization: histochemical studies using antler from red deer (Cervus elaphus) as a model. Calcif. Tissue Int. 65, 384–389. 69. Wennberg, C., Hessle, L., Lundberg, P., Mauro, S., Narisawa, S., Lerner, U. H., and Millán, J. L. (2000). Functional characterization of osteoblasts and osteoclasts from alkaline phosphatase knockout mice. J. Bone Min. Res. 15, 1879–1888. 70. Low, M. G., Carroll, R. C., and Cox, A. C. (1986). Characterization of multiple forms of phosphoinositide-specific phospholipase C purified from human platelets. Biochem. J. 237, 139–145. 71. Davitz, M. A., Hereld, D., Shak, S., Krakow, J., Englund, P. T., and Nussenzweig, V. (1987). A glycan-phosphatidylinositol-specific phospholipase D in human serum. Science 238, 81–84. 72. Low, M. G., and Prasad, A. R. S. (1988). A phospholipase D specific for the phosphatidylinositol anchor of cell-surface proteins is abundant in plasma. Proc. Natl Acad. Sci. USA 85, 980–984. 73. Miller, J. L. (1998). Release and extracellular transit of glycosylphosphatidylinositol proteins. J. Lab. Clin. Med. 131, 115–123.
Chapter 9 Alkaline Phosphatases 74. Anh, D. J., Eden, A., and Farley, J. R. (2001). Quantitation of soluble and skeletal alkaline phosphatase, and insoluble alkaline phosphatase anchor-hydrolase activities in human serum. Clin. Chim. Acta. 311, 137–148. 75. Narisawa, S., Huang, L., Iwasaki, A., Hasegawa, H., Alpers, D. H., and Millán, J. L. (2003). Accelerated fat absorption in intestinal alkaline phosphatase knockout mice. Mol. Cell. Biol. 23, 7525–7530. 76. Dehghani, H., Narisawa, S., Millán, J. L., and Hahnel, A. C. (2000). Effects of disruption of the embryonic alkaline phosphatase gene on the preimplantation development of the mouse. Dev. Dyn. 217, 440–448. 77. Whyte, M. P. (1994). Hypophosphatasia and the role of alkaline phosphatase in skeletal mineralization. Endo. Rev. 15, 439–461. 78. Whyte, M. P. (1995). Hypophosphatasia. In: C. R. Scriver, A. L. Beaudet, W. S. Sly, D.Valle (eds), The Metabolic and Molecular Bases of Inherited Disease, 7th ed. McGraw-Hill, New York, pp. 4095–4112. 79. Henthorn, P. S., and Whyte, M. P. (1992). Missense mutations of the tissue-nonspecific alkaline phosphatase gene in hypophosphatasia. Clin. Chem. 38, 2501–2505. 80. Zurutuza, L., Muller, F., Gibrat, J. F., Taillandier, A., Simon-Buoy, B., Serre, J. L., and Mornet, E. (1999). Correlations of genotype and phenotype in hypophosphatasia. Hum. Mol. Genet. 8, 1039–1046. 81. Di Mauro, S., Manes, T., Hessle, L., Kozlenkov, A., Pizauro, J. M., Hoylaerts, M. F., and Millán, J. L. (2002). Kinetic characterization of hypophosphatasia mutations with physiological substrates. J. Bone Min. Res. 17, 1383–1391. 82. Narisawa, S., Fröhlander, N., and Millán, J. L. (1997). Inactivation of two mouse alkaline phosphatase genes and establishment of a model of infantile hypophosphatasia. Dev. Dyn. 208, 432–446. 83. Waymire, K. G., Mahuren, J. D., Jaje, J. M., Guilarte, T. R., Coburn, S. P., and MacGregor, G. R. (1995). Mice lacking tissue non-specific alkaline phosphatase die from seizures due to defective metabolism of vitamin B-6. Nat. Genet. 11, 45–51. 84. Fedde, K. N., Blair, L., Silverstein, J., Coburn, S. P., Ryan, L. M., Weinstein, R. S., Waymire, K., Narisawa, S., Millán, J. L., MacGregor, G. R., and Whyte, M. P. (1999). Alkaline phosphatase knock-out mice recapitulate the metabolic and skeletal defects of infantile hypophosphatasia. J. Bone Min. Res. 14, 2015–2026. 85. Narisawa, S., Wennberg, C., and Millán, J. L. (2001). Abnormal vitamin B6 metabolism causes multiple abnormalities, but not impaired bone mineralization in alkaline phosphatase knockout mice. J. Pathol. 193, 125–133. 86. Spector, R. (1978). Vitamin B6 transport in the central nervous system: in vivo studies. J. Neurochem. 30, 881–887. 87. Stephens, M. C., Havlicek, V., and Dakshinamurti, K. (1971). Pyridoxine deficiency and development of the central nervous system in the rat. J. Neurochem. 18, 2407–2416. 88. Kash, S. F., Johnson, R. S., Tecott, L. H., Noebels, J. L., Mayfield, R. D., Hanahan, D., and Baekkeskov, S. (1997). Epilepsy in mice deficient in the 65-kDa isoform of glutamic acid decarboxylase. Proc. Natl Acad. Sci. USA 94, 14060–14065. 89. Stephens, M. C., and Dakshinamurti, K. (1975). Brain lipids in pyridoxine-deficient young rats. Neurobiology 5, 262–269. 90. Bernard, G. W. (1978). Ultrastructural localization of alkaline phosphatase in initial membranous osteogenesis. Clin. Orthop. 135, 218–225. 91. Morris, D. C., Masuhara, K., Takaoka, K., Ono, K., and Anderson, H. C. (1992). Immunolocalization of alkaline phosphatase in osteoblasts and matrix vesicles of human fetal bone. Bone Min. 19, 287–298. 92. Robison, R. (1923). The possible significance of hexosephosphoric esters in ossification. Biochem. J. 17, 286–293. 93. Majeska, R. J., and Wuthier, R. E. (1975). Studies on matrix vesicles isolated from chick epiphyseal cartilage. Association of pyrophosphatase and ATPase activities with alkaline phosphatase. Biochim. Biophys. Acta. 391, 51–60.
163 94. Fallon, M. D., Whyte, M. P., and Teitelbaum, S. L. (1980). Stereospecific inhibition of alkaline phosphatase by L-tetramisole prevents in vitro cartilage calcification. Lab. Invest. 43, 489–494. 95. Moss, D. W., Eaton, R. H., Smith, J. K., and Whitby, L. G. (1967). Association of inorganic-pyrophosphatase activity with human alkaline-phosphatase preparations. Biochem. J. 102, 53–57. 96. Rezende, A., Pizauro, J., Ciancaglini, P., and Leone, F. (1994). Phosphodiesterase activity is a novel property of alkaline phosphatase from osseous plate. Biochem. J. 301, 517–522. 97. Anderson, H. C., Hsu, H. H., Morris, D. C., Fedde, K. N., and Whyte, P. W. (1997). Matrix vesicles in osteomalacic hypophosphatasia bone contain apatite-like mineral crystals. Am. J. Path. 151, 1555–1561. 98. Johnson, K. A., Hessle, L., Vaingankar, S., Wennberg, C., Mauro, S., Narisawa, S., Goding, J. W., Sano, K., Millán, J. L., and Terkeltaub, R. (2000). Osteoblast tissue-nonspecific alkaline phosphatase antagonizes and regulates PC-1. Am. J. Physiol. Reg. Integr. Comp. Physiol. 279, R1365–R1377. 99. Hessle, L., Johnsson, K. A., Anderson, H. C., Narisawa, S., Sali, A., Goding, J. W., Terkeltaub, R., and Millán, J. L. (2002). Tissuenonspecific alkaline phosphatase and plasma cell membrane glycoprotein-1 are central antagonistic regulators of bone mineralization. Proc. Natl Acad. Sci. USA 99, 9445–9449. 100. Harmey, D., Hessle, L., Narisawa, S., Johnson, K., Terkeltaub, R., and Millán, J. L. (2004). Concerted regulation of inorganic pyrophosphate and osteopontin by Akp2, Enpp1 and Ank. An integrated model of the pathogenesis of mineralization disorders. Am. J. Pathol. 164, 1199–1209. 101. Okawa, A., Nakamura, I., Goto, S., Moriya, H., Nakamura, Y., and Ikegawa, S. (1998). Mutation in Npps in a mouse model of ossification of the posterior longitudinal ligament of the spine. Nat. Genet. 19, 271–273. 102. Johnson, K., Goding, J., Van Etten, D, Sali, A., Hu, S.-I., Farley, D., Krug, H., Hessle, L., Millán, J. L., and Terkeltaub, R. (2003). Linked deficiencies in extracellular inorganic pyrophosphate (ppi) and osteopontin expression mediate pathologic ossification in PC-1 null mice. J. Bone Min. Res. 18, 994–1004. 103. Ho, A. M., Johnson, M. D., and Kingsley, D. M. (2000). Role of the mouse ank gene in control of tissue calcification and arthritis. Science 289, 265–270. 104. Nürnberg, P., Theile, H., Chandler, D., Höhne, W., Cunningham, M. L., and Ritter, H. (2001). Heterozygous mutations in ANKH, the human ortholog of the mouse progressive ankylosis gene, result in craniometaphyseal dysplasia. Nat. Genet. 28, 37–41. 105. Bostrom, K. (2001). Insights into the mechanism of vascular calcification. Am. J. Cardiol. 88, 20E–22E. 106. Bostrom, K., and Demer, L. L. (2000). Regulatory mechanisms in vascular calcification. Crit. Rev. Eukaryotic Gene Expression 10, 151–158. 107. Picher, M., Burch, L. H., Hirsh, A. J., Spychala J., and Boucher, R. C. (2003). Ecto 5′-nucleotidase and non-specific alkaline phosphatase: two AMP-hydrolyzing ectoenzymes with distinct roles in human airways. J. Biol. Chem. 278, 13468–13479. 108. Takahashi, T., Iwantanti, T., Mizuno, A. S., Morishita, Y., Nishio, H., Kodama, S., and Matsuo, T. (1984). The relationship between phosphoethanolamine level in serum and intractable seizure on hypophosphatasia infantile form. In D.V. Cohn, T. Fugita, J.T. Potts, Jr., R.V. Talmage (eds.), Endocrine Control of Bone and Calcium Metabolism, Vol 8B. Excerpta Medica, Amsterdam, The Netherlands, pp. 93–94. 109. Whyte, M. P., Landt, M., Ryan, L. M., Mulivor, R. A., Henthorn, P. S., Fedde, K. N., Mahuren, J. D., and Coburn, S. P. (1995). Alkaline phosphatase: placental and tissue-nonspecific isoenzymes hydrolyze phosphoethanolamine, inorganic pyrophosphate, and pyridoxal 5’-phosphate. Substrate accumulation in carriers of hypophosphatasia corrects during pregnancy. J. Clin. Invest. 95, 1440–1445.
164 110. Fleshood, H. L., and Pitot, H. C. (1969). O-phosphorylethanolamine ammonia lyase, a new pyridoxal phosphate-dependent enzyme. Biochem. Biophys. Res. Comm. 36, 110–118. 111. Fleshood, H. L., and Pitot, H. C. (1970). The metabolism of O-phosphorylethanolamine in animal tissues. II. Metabolic regulation of O-phosphorylethanolamine phospho-lyase in vivo. Arch. Biochem. Biophysics. 141, 423–429. 112. Grøn, I. H. (1978). Mammalian O-phosphorylethanolamine phospho-lyase activity and its inhibition. Scand. J. Clin. Lab. Invest. 38, 107–112. 113. Van Straalen, J. P., Sanders, E., Prummel, M. F., and Sanders, G. T. B. (1991). Bone-alkaline phosphatase as indicator of bone formation. Clin. Chim. Acta. 201, 27–34. 114. Moss, D. W., and Whitby, L. G. (1975). A simplified heat-inactivation method for investigating alkaline phosphatase isoenzymes in serum. Clin. Chim. Acta. 61, 63–71. 115. Lin, C. W., and Fishman, W. H. (1972). L-Homoarginine. An organspecific, uncompetitive inhibitor of human liver and bone alkaline phosphohydrolases. J. Biol. Chem. 247, 3082–3087. 116. Rosalki, S. B., and Foo, A. Y. (1984). Two new methods for separating and quantifying bone and liver alkaline phosphatase isoenzymes in plasma. Clin. Chem. 30, 1182–1186. 117. Van Hoof, V. O., Lepoutre, L. G., Hoylaerts, M. F., Chevigné, R., and De Broe, M. E. (1988). Improved agarose electrophoretic method for separating alkaline phosphatase isoenzymes in serum. Clin. Chem. 34, 1857–1862. 118. Griffiths, J., and Black, J. (1987). Separation and identification of alkaline phosphatase isoenzymes and isoforms in serum of healthy persons by isoelectric focusing. Clin. Chem. 33, 2171–2177. 119. Magnusson, P., Löfman, O., and Larsson, L. (1992). Determination of alkaline phosphatase isoenzymes in serum by high-performance liquid chromatography with post-column reaction detection. J. Chromat. 576, 79–86. 120. Bailyes, E. M., Seabrook, R. N., Calvin, J., Maguire, G. A., Price, C. P., Siddle, K., and Luzio, J. P. (1987). The preparation of monoclonal antibodies to human bone and liver alkaline phosphatase and their use in immunoaffinity purification and in studying these enzymes when present in serum. Biochem. J. 244, 725–733. 121. Hill, C. S., and Wolfert, R. L. (1990). The preparation of monoclonal antibodies which react preferentially with human bone alkaline
JOSÉ LUIS MILLA´ N
122.
123.
124.
125.
126. 127.
128.
129.
130.
131.
phosphatase and not liver alkaline phosphatase. Clin. Chim. Acta. 186, 315–320. Panigrahi, K., Delmas, P. D., Singer, F., Ryan, W., Reiss, O., Fisher, R., Miller, P. D., Mizrahi, I., Darte, C., Kress, B. C., and Christenson, R. H. (1994). Characteristics of a two-site immunoradiometric assay for human skeletal alkaline phosphatase in serum. Clin. Chem. 40, 822–828. Gomez, Jr., B., Ardakani, S., Ju, J., Jenkins, D., Cerelli, M. J., Daniloff, G. Y., and Kung, V. T. (1995). Monoclonal antibody assay for measuring bone-specific alkaline phosphatase activity in serum. Clin. Chem. 41, 1560–1566. Broyles, D. L., Nielsen, R. G., Bussett, E. M., Lu, W. D., Mizrahi, I. A., Nunnelly, P. A., Ngo, T. A., Noell, J., Christenson, R. H., and Kress, B. C. (1998). Analytical and clinical performance characteristics of Tandem-MP Ostase, a new immunoassay for serum bone alkaline phosphatase. Clin. Chem. 44, 2139–2147. Magnusson, P., Ärlestig, L., Paus, E., Di Mauro, S., Testa, M. P., Stigbrand, T., Farley, J. R., Knustad, K., and Millán, J. L. (2002). Monoclonal antibodies against bone alkaline phosphatase: The ISOBM TD-9 workshop. Tumor Biol. 23, 228–248. Kleerekoper, M. (1996). Biochemical markers of bone remodeling. Am. J. Med. Sci. 312, 270–277. Miller, P. D., Baran, D. T., Bilezikian, J. P., Greenspan, S. L., Lindsay, R., Riggs, B. L., and Watts, N. B. (1999). Practical clinical application of biochemical markers of bone turnover. J. Clin. Densitometry 2, 323–342. Riggs, B.L. (2000). Are biochemical markers for bone turnover clinically useful for monitoring therapy in individual osteoporotic patients? Bone 26, 551–552. Magnusson, P., Larsson, L., Magnusson, M., Davie, M. W. J., and Sharp, C. A. (1999). Isoforms of bone alkaline phosphatase: characterization and origin in human trabecular and cortical bone. J. Bone Min. Res. 14, 1926–1933. Magnusson, P., Larsson, L., Englund, G., Larsson, B., Strang, P., and Selin-Sjögren, L. (1998). Differences of bone alkaline phosphatase isoforms in metastatic bone disease and discrepant effects of clodronate on different skeletal sites indicated by the location of pain. Clin. Chem. 44, 1621–1628. Magnusson, P., Sharp, C. A., Magnusson, M., Risteli, J., Davie, M. W. J., and Larsson, L. (2001). Effect of chronic renal failure on bone turnover and bone alkaline phosphatase isoforms. Kidney Int. 60, 257–265.
Chapter 10
Acid Phosphatases Helena Kaija, and Lila O.T. Patrikainen Sari L. Alatalo H. Kalervo Väänänen Pirkko T. Vihko
I. Acid Phosphatases II. Tartrate-Resistant Acid Phosphatase (TRACP)
Finnish Red Cross Blood Service, Helsinki, Finland Institute of Biomedicine, Department of Anatomy, University of Turku, Turku, Finland Department of Biological and Environmental Sciences, Division of Biochemistry, University of Helsinki, Finland
III. Prostatic Acid Phosphatase (PAP) References
I. ACID PHOSPHATASES
for catalysis. One obvious difference among phosphatases is the presence or absence of metal ion cofactors. Acid phosphatases catalyze the hydrolysis of phosphate ester bond in the following reaction at an optimal pH below 7:
The phosphate ester bond functions as an extremely important linkage within the living cell. It participates in storage and transfer of the genetic information, carries chemical energy, and regulates the activity of enzymes and signaling molecules in the cell. Enzymes capable of acting on ester bonds (Enzyme Commission classification number EC 3.1) and catalyzing the cleavage of phosphate esters (EC 3.1.3) constitute the subclass of phosphohydrolases, i.e. phosphatases. Phosphatases can be classified according to several frameworks [1]. They can be divided into groups based on their substrate type. Nonspecific phosphatases catalyze the hydrolysis of almost any phosphate ester, whereas protein phosphatases prefer phosphoproteins or phosphopeptides as substrates. Nonspecific phosphatases can be divided into alkaline phosphatases (EC 3.1.3.1) and acid phosphatases (EC 3.1.3.2) based on their optimal pH Dynamics of Bone and Cartilage Metabolism
Research Center for Molecular Endocrinology, University of Oulu, Finland
orthophosphoric monoester + H2O = alcohol + phosphate Acid phosphatases have rather wide substrate specificity and some of them catalyze transphosphorylations between phosphoesters and alcohols. Nonspecific acid phosphatases recycle phosphate in metabolic reactions [2]. Protein phosphatases are a structurally miscellaneous group of enzymes that remove phosphate groups that have been attached to amino acid residues of proteins by protein kinases. Protein phosphatases have different mechanisms of action, subcellular localization, and substrate specificity with varied demands for optimal pH. Protein phosphatases comprise two groups based upon their substrate specificity. Protein-tyrosine phosphatases (PTPases, EC 3.1.3.48) 165
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
166
HELENA KAIJA ET AL.
prefer to remove phosphate from tyrosine residue and serine/threonine protein phosphatases (PPases, EC 3.1.3.16) from serine or threonine residue: phosphoprotein + H2O = protein + phosphate The PTPase group includes dual-specificity phosphatases, which are capable of hydrolyzing phosphorylated tyrosine, serine, and threonine, and phosphorylated lipids. Protein phosphatases regulate signal-transduction pathways [2]. Approximately 30% of intracellular proteins are subject to reversible protein phosphorylation [3].
A. Mammalian Acid Phosphatases Li and co-workers demonstrated, in 1970, seven different acid phosphatase isoenzymes in human leukocytes by acrylamide gel electrophoresis and ion exchange chromatography [4]. The heterogeneity of acid phosphatases within tissues had been reported previously, but the relationship between acid phosphatase isoenzymes and tissues and different cells in tissues were not clarified until later due to technical limitations. Based on the differences at the structural level of the gene, acid phosphatases can be divided into five isoenzymes, namely prostatic (AcPP, PAP, human chromosomal location 3q21–q23), lysosomal (AcP2, LAP, human chromosomal location 11p12–p11), erythrocytic (AcP1, soluble acid phosphatase, human chromosomal location 2p25), testicular (AcPT, human chromosomal localization 19q13), and macrophagic acid phosphatases (AcP5, tartrate-resistant acid phosphatase, chromosomal location 19p13.3–p13.1). Acid phosphatases can also be distinguished based on their molecular weight into low-molecular-weight acid phosphatases and high-molecular-weight acid phosphatases, or on the basis of their resistance to inhibition by tartrate into tartrate-sensitive and tartrate-resistant acid phosphatases. Both protein tyrosine phosphatase group and protein serine/threonine phosphatase group include phosphatases with acidic optimal pH.
B. Biochemical and Physiological Significance of Phosphatase Activity The ubiquity of phosphatases in nature makes them versatile regulators of metabolic reactions and cell signaling. Phosphoregulation is involved in many biological events and often occurs as a network-like cascade, in which activity of one phosphatase or kinase is dependent on the upstream activity of another. Many protein kinases are
regulated by phosphorylation constituting an important substrate group for protein phosphatases. Regulation of enzyme activities is most frequently performed by sequential cycles of phosphorylations and dephosphorylation. There seem to be fewer phosphatases than kinases, and phosphatases are not as well characterized as kinases. Activators, inhibitors as well as substrates, affect the rate at which the enzyme is phosphorylated/ dephosphorylated. Regulation of glucose and lipid metabolism at the cellular as well as molecular level is mainly carried out by alternating phosphorylations and dephosphorylations. Activation of the insulin receptor requires the tyrosine autophosphorylation [5]. Dephosphorylation of any one of the activating phosphotyrosine residues dramatically reduces the receptor kinase activity, thus functioning as negative regulators of the insulin signaling cascade [6]. Alterations in the extracellular environment through hormones, growth factors, or cytokines, induce signal transduction pathways to target transcription factors and transcriptional coregulators leading to their phosphorylation or dephosphorylation. Regulation of transcriptional activity by kinases and phosphatases is a rapid and readily reversible mechanism with great versatility and flexibility [7]. Phosphorylation/dephosphorylation plays also a central role in cell-cycle progression by regulating the activity of CDK/cyclin complexes. Although the concentration of human acid phosphatases is normally low, significant changes in their expression occur in certain diseases combined with specific tissue location. This observation suggests that acid phosphatases may be usable as cell-organelle markers as well as diagnostic markers.
II. TARTRATE-RESISTANT ACID PHOSPHATASE (TRACP) Tartrate-resistant acid phosphatase (TRACP) belongs to the acid phosphatases (EC 3.1.3.2), sharing catalytic activity towards phosphoesters in an acidic environment. In 1970, Li and colleagues discovered at least seven distinct isoenzymes 0, 1, 2, 3a, 3b, 4, and 5 from human leukocytes, numbered in order of their increasing electrophoretic mobility towards the cathode [4, 8]. “Band 5” acid phosphatase (AcP 5) or type-5 AcP was the only resistant to L(+)-tartrate inhibition, from which the name tartrateresistant acid phosphatase derives. Also, platelets and red blood cells contain tartrate-resistant AcPs, but these are clearly different from type-5 AcP based on their molecular weights, substrate specificities, electrophoretic mobilities, and antigenic properties. Hence, TRACP specifically refers
167
Chapter 10 Acid Phosphatases
to type-5 AcP demonstrated on acidic acrylamide gels. Previously, the commonly used abbreviation for type-5 AcP was TRAP, but following a recommendation from the nomenclature committee an additional ‘C’ was added to TRACP as a separation from alkaline phosphatases (APs). The members of TRACP family are also known as purple acid phosphatases because of the bound metal ions in their active sites giving an intense purple color.
A. Isolation of Mammalian TRACP Enzymes TRACP enzymes have been purified from many mammalian sources, including pig uterine fluid, bovine bone [9], the spleens of patients affected with hairy cell leukemia [10, 11], and rat [12, 13] bone. When TRACP enzymes were purified from distinct sources they were shown to exist either in monomeric, approximately 30–35 kDa proteins, or in two subunits, sized approximately 16 and 20–25 kDa, connected by a covalent bond. For a long time this discrepancy remained enigmatic, until in the 1990s several research groups were able to demonstrate an exposed, highly antigenic, protease-sensitive loop structure in the sequence of TRACP [14, 15]. After cleavage of this loop, the enzyme was turned into two subunits connected by a disulfide bridge between two conserved cysteine residues.
B. Expression of TRACP Physiologically, TRACP expression is found in cells of the mononuclear phagocyte system, most abundantly in bone-resorbing osteoclasts, alveolar macrophages, and dendritic cells. High TRACP activity has been demonstrated in bone, spleen, liver, lung, and placenta, with lower amounts in kidney, skin, colon, stomach, ileum, testis, and brain. TRACP activity in other than bone tissue is primarily due to the wide distribution of TRACP-positive dendritic cells and activated macrophages. Unstimulated monocytes in bone marrow or peripheral blood are TRACP negative. Pathologically, TRACP expression is elevated in diseases such as hairy cell leukemia (HCL) and Gaucher’s disease [16]; hence, TRACP is clinically used as a marker enzyme for these diseases. Both hairy cells in HCL and spleen cells of patients with Gaucher’s disease stain strongly for TRACP. Elevated TRACP activities in serum have been demonstrated in situations with increased osteoclastic activity such as hyperparathyroidism, Paget’s disease, inflamed synovial joints, postmenopausal women, osteoporosis, osteoclastoma, and various cancers with bone metastases.
C. Subcellular Localization of TRACP TRACP activity in osteoclasts was originally detected adjacent to the bone-resorbing organ, ruffled border membrane, with nonspecific substrates such as β-glycerophosphate and adenosine triphosphate (ATP). However, TRACP is not able to hydrolyze β-glycerophosphate, and the activity detected with this substrate was most probably tartrate-sensitive lysosomal AcP (LAP) activity. Also, ATP is a better substrate for vacuolar-type proton pump, V-ATPase, located in the ruffled border membrane than for TRACP [17]. Immunohistochemical detection revealed more reliable results, showing intracellular vesicular staining for TRACP with no positive reaction at the ruffled border region [18, 19]. However, contradictory data have been observed using an immunogold technique demonstrating intensive TRACP staining both at the ruffled border membrane and intracellularly in vesicular structures [20]. In studies using electron microscopy or separation of subcellular granules by density gradient centrifugation, followed by cytochemical or immunocytochemical staining, TRACP was localized to lysosomes, lysosome-like organelles of the Golgi complex, and microsomes in spleen cells of HCL, rat liver macrophages, rat bone osteoclasts, and bovine spleen histiocytes. Within the lysosome-like organelles, TRACP activity was restricted to the inner membrane surface [19], which is consistent with the sequence data, suggesting a putative N-terminal lysosomal leader sequence in the TRACP gene. Nevertheless, confocal microscopy studies revealed no or only partial colocalization with TRACP and known lysosomal markers, lysosome-associated membrane protein 1 and the cationindependent mannose-6-phosphate receptor in osteoclasts and in alveolar macrophages [21]. Instead, TRACP was shown to colocalize with bone degradation products at the transcytotic route in resorbing osteoclasts [19] and with internalized S. aureus in alveolar macrophages [21]. The transcytotic route of osteoclasts is analogous to the antigen presentation route of activated macrophages, as both are a transport route from a late endosomal/lysosomal compartment to a cell membrane. These data indicate that TRACP may be localized into some new, still unidentified, vesicular compartment both in osteoclasts and in macrophages. Some researchers have also found TRACP in other bone cells, namely osteoblasts and osteocytes, but the intensity of the staining has been lower than in osteoclasts. However, in other studies osteoblasts and osteocytes have been totally negative for TRACP [18, 20]. It has been speculated that osteoblasts would endocytose TRACP secreted by osteoclasts, and TRACP would be as a coupling factor between bone resorption and formation [22].
168
HELENA KAIJA ET AL.
Nevertheless, researchers are in agreement that TRACP is mainly produced by osteoclasts and secreted from them during bone resorption [23]. Active enzyme can be detected in the serum of mammals as a complex with α2-macroglobulin [24] and used as a marker of bone resorption [25, 26].
D. Protein Structure Purple acid phosphatase from the red kidney bean (Protein Data Bank code, 1KBP) was the first member of the TRACP family for which the crystal structure became available at 2.9 Å resolution in 1995 [27]. In 1999, the three-dimensional structures for three mammalian TRACP enzymes were published almost simultaneously: for pig TRACP (1UTE) at 1.55 Å resolution [28], for rat TRACP from bone (1QHW] at 2.2 Å resolution [29], and for recombinant rat TRACP (1QFC) at 2.7 Å resolution [30]. Despite less than 15% nucleotide sequence homology between distinct mammalian TRACP enzymes, the amino acid sequences exhibit over 80% identity and the protein
folds are very similar. The structure of mammalian TRACP enzymes is approximately spherical with dimensions of 45 × 45 × 40 Å. The enzymes are formed of two sandwiched β-sheets flanked by α-helical segments facing the solvent. The enzymes show internal symmetry, with the metal ions bound at the interface between the two halves comprising the active site (Fig. 1). Mammalian enzymes are approximately 35 kDa monomeric proteins, whereas plant enzymes are typically dimers of 55 kDa subunits. Mammalian TRACPs contain two iron atoms in their active site, while plant enzymes have one iron atom and either zinc or manganese [27]. The antiferromagnetically spin-coupled binuclear iron center of the mammalian TRACP enzymes exists in two stable interconvertible states: pink, reduced, and enzymatically active, with a mixedvalent Fe3+–Fe2+ cluster; and purple, oxidized, and catalytically inactive, with a binuclear pair as Fe3+–Fe3+ [31]. Another of the iron atoms in the active site of mammalian TRACP enzymes is stabilized into the ferric (Fe3+) form by a tyrosine residue (Tyr55; numbering according to human sequence), accounting for the characteristic purple color of TRACPs. A histidine (His223) and an
Figure 1 The three-dimensional structure of rat bone TRACP determined at 2.2 Å resolution (PDB code 1QHW). The core of the enzyme was formed by two seven-stranded β-sheets which, packed together, making up a β-sandwich. The β-sandwich was surrounded by three solvent-exposed α-helices on both sides. The di-iron center was located on the other edge of the β-sandwich. Two N-acetylglucosamine (NAG) molecules were linked to the side chain of Asn118. One zinc and one sulfate ion were found in the active site and one further zinc and sulfate ion were involved in crystal contact. The protease sensitive loop, seen at the upper left corner of the structure, is located close to the active site [29].
169
Chapter 10 Acid Phosphatases
aspartate (Asp14) residue coordinate the ferric iron among tyrosine residue. A µ-hydroxo bridge and an aspartate carboxylate (Asp52) connect the ferric iron to the redox active iron (Fe2+/3+), which is in turn coordinated by two histidines (His186, His221) and an asparagine (Asn91) residue [27]. The residues coordinating the binuclear active site of TRACP enzymes are essentially identical to those of plant TRACPs and the regulatory Ser/Thr protein phosphatases (PPs) [32]. However, both plant TRACPs and PPs have a ferric iron together with a zinc or manganese in their active site instead of two iron atoms. Additionally, PPs differ essentially from the TRACP family by having a water molecule instead of the tyrosine residue coordinating the ferric iron, accounting for the loss of purple color [27]. Near the active site pocket is an exposed, proteasesensitive loop. The loop structure of mammalian enzymes can be cleaved by trypsin or cathepsins, yielding two fragments connected by a disulfide bridge. Proteolytic cleavage shifts the pH optimum approximately from 5 to 6 and activates the acid phosphatase activity of TRACP by 3–8fold, also causing changes in the EPR spectrum [32, 15]. Enzymatic activation was shown to be not due to the removal of steric hindrance or conformational changes, but rather due to changes in the molecular interaction between residue Asp146 of the loop region and residue Asn91 coordinating the redox active iron in the active site [15]. From the two putative N-linked glycosylation sites Asn97 and Asn128, the previous one was shown to be glycosylated [29, 30]. In uteroferrin, the carbohydrate side chain consists of phosphorylated, high-mannose-type oligosaccharide composed of six mannose residues and two N-acetylglucosamine residues [34]. Similar to uteroferrin, human bone TRACP also contains only N-linked highmannose carbohydrate [11]. Rat, mouse, and human TRACP, but not uteroferrin, contain a consensus motif for protein tyrosine phosphatases (PTPase); Cys(X)5Arg [35]. The conserved cysteine residue in the motif for PTPases forms a covalent bond with phosphate during catalysis and is essential for enzyme activity. However, in TRACPs the conserved cysteine forms a disulfide bridge, being unable to participate in dephosphorylation.
shown to be due to additional sialic acid in the carbohydrate side chain of TRACP 5a. After sialidase treatment TRACP 5a is converted to TRACP 5b. TRACP 5a has a pH optimum of 5.2 and relatively low specific activity, whereas TRACP 5b has a pH optimum of 5.8 with substantially higher specific activity. In addition to a different pH optimum and carbohydrate content, TRACP 5a and 5b are also distinguishable by antigenic properties (Table I) [36]. Based on current knowledge, osteoclasts contain and secrete only TRACP 5b isoform, while activated macrophages express both TRACP 5a and 5b, but secrete only TRACP 5a into the circulation [37]. The osteoclastic origin of TRACP 5b makes it a specific marker for osteoclasts and bone resorption [38, 39].
F. TRACP Gene The mammalian TRACP genes have been cloned from several species. The exon–intron structure of the TRACP gene is highly conserved, consisting of five exons with the translation initial signal (A+1TG) at the beginning of exon 2. The TRACP gene in mammals is 975–1020 bp long, and it encodes approximately 1.5 kb mRNA. Translation will encode a protein with 323–325 amino acids, including a signal peptide of 19 residues and two potential sites for N-glycosylation. Very recently multiple tissue-specific promoters were identified [40], solving the problem with contradictory results with several transcription start points (tsp). Walsh and colleagues were able to identify three distinct mRNAs with differential 5′UTRs, but similar 3′ ends of the mRNA from the first base of exon 2. The novel 5′-UTRs represent alternative first exons located upstream of the known 5′UTR [41]. This genomic structure is conserved at least between mouse and human. Expression of the most distal 5′-UTR (exon 1A) in mouse is restricted to adult bone and spleen tissue, while exon 1B is expressed primarily in tissues containing TRACP-positive nonhematopoietic cells. The known 5′-UTR (exon 1C) is expressed in cells
Table I.
E. Isoforms 5a and 5b Soon after the discovery of TRACP, Lam and co-workers demonstrated that TRACP exists in two distinct isoforms in human serum [36]. Two tartrate-resistant bands with slightly different molecular weights were observed and named according to their electrophoretic mobility as TRACP 5a and TRACP 5b. The difference in size was
Properties of TRACP 5a and 5b TRACP 5a
TRACP 5b
Sialic acid pH optimum for AcP Specific activity of AcP Polypeptide chain
Present 4.9–5.1 Low Intact ∼35 kDa
Origin
Macrophages and dendritic cells
Not present 5.7–5.9 High Cleaved 16 kDa + 20–25 kDa Osteoclasts
170
HELENA KAIJA ET AL.
originating from a myeloid lineage in osteoclasts and macrophages [40]. Although the genomic structure of the 5′-flanking region of the human and mouse TRACP genes is conserved, the sequence identity is limited. Studies on the 5′-flanking region of the TRACP gene have indicated a complex transcriptional regulation. The most conserved region of the TRACP 5′-flanking region lies within the 2-kb segment from the tsp A+1TG, and this area has been under extensive study. The possibility that TRACP might be involved in iron transport or storage has led several groups to examine the regulatory influence of intracellular iron on TRACP expression. Simmen and co-workers found a putative iron-response element in the 5′-flanking region of the uteroferrin gene in 1989 [42], and a few years later Alcantara and colleagues were able to show that expression of TRACP was regulated by iron and the iron responsive element was located between base pairs –1846 and –1240 in the mouse promoter [43]. This iron regulation is very complex and occurs at the transcriptional level. Two iron-delivering agents, hemin (ferric protoporphyrin IX) and transferrin have opposite effects on TRACP expression, hemin decreases and transferrin increases [43, 44] TRACP expression. Two heminresponsive elements (HRE) were found within the human 5′-flanking region, representing binding sites for an HRE-binding protein (HRE-BP), and both are needed for the hemin-induced regulation of the human TRACP gene expression [45]. Monocyte-associated transcription factor PU.1 acts synergistically with osteoclast commitment factor MiTF (microphthalmia transcription factor) regulating the TRACP expression during terminal differentiation of osteoclasts [46]. TRACP expression is stimulated by RANKL, and at least upstream stimulatory factors (USF) 1 and 2 are involved in RANKL-mediated activation [47]. In addition to these, the promoter area of TRACP contains numerous candidate transcription factor binding sequences, including those for Sp1, AP1, GT-1, and H-APF-1 [41].
G. Functions of TRACP The biological function(s) as well as natural substrate(s) of TRACP are still unknown despite intensive studies within recent decades. However, the highly conserved structure of TRACP enzymes throughout the animal kingdom provides strong evidence for an important biological role for the enzyme. Location in distinct tissues and cells, as well as functions shown in vitro, suggest that TRACP may have several functions in different cells and cell compartments.
1. G.1. AcP
TRACP is primarily an AcP catalyzing the hydrolysis of a wide range of phosphate monoesters such as β-umbelliferyl phosphate, p-nitrophenyl phosphate, α-naphthyl phosphate and phosphotyrosine. Also, phosphoanhydrides such as pyrophosphate and nucleoside tri- and diphosphates are hydrolyzed by TRACP [12], but aliphatic phosphoesters such as monophosphates, β-glycerophosphate, mannose-6-phosphate, phosphoserine, and phosphothreonine are not. However, TRACP can catalyze the release of phosphate from certain phosphoproteins that carry phosphoserine residue such as osteopontin and osteonectin [49]. Nevertheless, the AcP activity of TRACP prefers phosphotyrosine-containing proteins over other phosphoproteins, suggesting a specific protein tyrosine phosphatase (PTPase) activity for TRACP [9]. Phosphorylation of specific tyrosine residues by kinases plays an important role in various cell events such as signal transduction, activation, proliferation, and differentiation. As a PTPase, TRACP could regulate protein-tyrosine kinases and thereby influence various cellular events [50]. The AcP activity of TRACP is competitively inhibited by inorganic phosphate and its analogs, e.g. vanadate, arsenate, and molybdate. Fluoride, tungstate, copper, and zinc cause noncompetitive inhibition of AcP activity [51]. Reducing agents such as β-mercaptoethanol, dithiothreitol, ascorbic acid, cysteine, and glutathione, activate the AcP activity of TRACP by reducing the ferric iron into the ferrous form and simultaneously changing the purple color to pink [51]. The reverse reaction from pink to purple (Fe2+ → Fe3+) causes inactivation and is achieved by oxidizing agents such as H2O2 and ferricyanide. Dithionite inactivates TRACP irreversibly by removing the metal ions from the active site, thereby bleaching its color [48, 51]. In contrast, common inhibitors for AcPs such as cyanide, azide, tartrate, and p-nitrophenol show no inhibition of the AcP activity of TRACP [51]. Combined structural and mechanistic studies provide a model for the catalytic mechanism of TRACP, differing from the mechanism of PTPases. In the first step, the phosphate group of the substrate is coordinated to the divalent metal ion. Formation of the enzyme–substrate complex is followed by a nucleophilic attack on the phosphorus, leading to the release of the product alcohol. Three possible candidates have been proposed for the attacking group: (1) a terminal Fe3+-bound hydroxide [52]; (2) a bridging hydroxide [53]; or (3) a hydroxide residing in the second coordination sphere of Fe3+ [54]. In the final step of catalysis, the metal-bound phosphate group is released. TRACP has been hypothesized to participate in bone resorption by dephosphorylating bone matrix phosphoproteins osteopontin, osteonectin, and bone sialoprotein [49].
171
Chapter 10 Acid Phosphatases
Osteopontin has been shown to bind to osteoclast cellsurface integrins via the RGD motif, thereby mediating substrate adhesion, at least in vitro [55]. After dephosphorylation osteopontin is no longer able to support osteoclast binding to the substrate, suggesting that TRACP could regulate osteoclast attachment [49]. Dephosphorylation of bone matrix phosphoproteins suggests a catabolic function for TRACP in bone resorption.
H. TRACP-Deficient and Transgenic Mice
2. G.2. Generator of Reactive Oxygen Species (ROS)
Three individual research groups have been able to demonstrate by different methods that TRACP is capable of generating ROS via Fenton’s reaction [19, 56, 57]. In an enzymatically active reduced state, TRACP has a mixedvalent Fe3+–Fe2+ active site, in which the ferrous iron (Fe2+) is able to react with hydrogen peroxide and produce ferric iron (Fe3+) and hydroxyl radical (•OH) (Equation 1). The newly formed ferric iron is still able to react with hydrogen peroxide to form superoxide anion (•O2−) and a ferrous iron (Equation 2). Hydroxyl radical formation: TRACP-Fe2+ + H2O2 → TRACP-Fe3+ + •OH + OH−
two activities are functionally independent [61]. Beck and co-workers were able to demonstrate that the AcP activity of TRACP is inhibited by the substrate of ROS-generating activity, H2O2, and vice versa, inorganic phosphate blocks H2O2 inhibition, confirming the use of the same active site [62].
(1)
TRACP regeneration and superoxide formation: TRACP-Fe3+ + H2O2 → TRACP-Fe2+ + •O2− + 2H+ (2) The generation of these destructive oxygen radicals can continue as long as H2O2 is available. When hydroxyl radicals alone attack proteins, they cause protein aggregation, but in combination with superoxide anions they cause protein fragmentation without formation of aggregates [58]. Halleen and co-workers were able to demonstrate in vitro that TRACP is able to destroy the major component of the bone matrix, type I collagen, without disruption of the enzyme itself [19]. They also showed that TRACP is localized in transcytotic vesicles of osteoclasts together with bone degradation products, hypothesizing that TRACP would participate in the final degradation of bone matrix molecules intracellularly in the transcytotic vesicles [19]. In general, ROS have been shown to stimulate osteoclastic bone resorption both in vivo and in vitro, and they also enhance the recruitment of osteoclasts [59]. Based on studies with TRACP-deficient or TRACP overexpressing macrophages, TRACP may also have a role in antigen processing by generating destructive ROS [21, 60]. Both the ROS generation and the AcP activity of TRACP use the redox active iron for the catalysis. However, the acid phosphatase activity has an acidic pH-optimum, whereas ROS are generated in a neutral pH [61]. Also, single amino acid TRACP mutants that are completely inactive as phosphatase are still able to produce ROS, suggesting that the
The role of TRACP in bone metabolism and also in other tissues has been elucidated by the TRACP knockout [63] and TRACP over-expressing mouse models [64]. Both models exhibit a bone phenotype; mice lacking TRACP have a mild osteopetrosis resulting from defective bone resorption [63], whereas mice over-expressing TRACP have increased bone resorption that is accompanied by increased bone formation, yielding a net increase in bone turnover [64]. The mechanism by which TRACP increases bone resorption in transgenic mice remains unclear, but nevertheless mice exhibit mild osteoporosis [64]. TRACP-deficient mice suffer from progressive foreshortening and deformity of the long bones and axial skeleton [63]. Tooth eruption and skull plate development is normal in these mice, indicating a role for TRACP in endochondral ossification [65]. Furthermore, the formation of osteoclasts is normal in TRACP-deficient mice, suggesting that TRACP is not essential during osteoclastogenesis [65]. However, osteoclasts isolated from TRACP knockout mice demonstrate defective bone resorption also in vitro determined by a conventional resorption pit assay [63]. Ultrastructural examination of young TRACP null mice revealed that osteoclasts exhibit an increased relative area of ruffled border and accumulation of cytoplasmic vesicles [65] containing filamentous material [66]. The accumulation of intracellular vesicles is probably not due to the impaired secretion, since the location of cathepsin K is normal. Some explanations have been offered for the accumulated vesicles; TRACP-deficient osteoclasts are incapable of totally digesting the internalized bone degradation products, and vacuoles containing filamentous material accumulate into the cytoplasm instead of normal transcytosis and secretion through the FSD [19]; or TRACP may have a function in modulating intracellular vesicular transport by dephosphorylating phosphoproteins [65]. A possible role for TRACP in the immune system has been studied both with TRACP-deficient and TRACP overexpressing macrophages [60]. TRACP-null macrophages have disordered inflammatory responses, such as enhanced superoxide and nitrite production, and increased secretion of proinflammatory cytokines TNFα, IL-1β, and IL-12 [60]. Additionally, TRACP knockout mice exhibit delayed
172
HELENA KAIJA ET AL.
clearance of Staphylococcus aureus in vivo [60]. Vice versa, phorbol 12-myristate 13-acetate (PMA) activated macrophages over-expressing TRACP showed increased overall production of ROS and a tendency to higher capacity for bacterial killing in vitro [67]. The biological roles and functional relationship between two acid phosphatases, TRACP and LAP, have remained unknown. Single TRACP [63] or LAP [68] knockouts have distinct, but rather mild phenotypes, suggesting that the phosphatases may in part substitute each other. The double knockout mouse model of both TRACP and LAP demonstrates a more severe phenotype than the sum of the single knockouts, thus strengthening the hypothesis [66]. Mice deficient for both phosphatases showed massive lysosomal storage, for example in Kupffer cells, in bone marrow macrophages and in osteoclasts. Also, the bone phenotype, which was more evident in TRACP- than in LAP-deficient mice, was even more severe in double-deficient mice. The dephosphorylation of osteopontin was severely affected in double-knockout mice, and the authors suggest that TRACP represents the major enzyme implicated in this process [66]. Nevertheless, no data are available from dephosphorylation of osteopontin in single-knockout mice to verify this hypothesis. Furthermore, no attention has been addressed to the ROS generation activity of TRACP, which is not possible to overcome by LAP.
I. TRACP as a Clinical Tool TRACP was considered as a specific histochemical marker for osteoclasts more than 20 years ago [23]. Since then TRACP has been widely accepted as a histological marker of osteoclasts both in tissue sections and in in vitro cell cultures. Without knowing the exact biological role of TRACP in bone metabolism it has become one of the best-known markers for bone-resorbing osteoclasts and their precursors. At present there are up to 1000 references in the database of National Library of Medicine with keywords “TRACP and osteoclast”. The first assays for analytical purposes were kinetic assays to measure the total TRACP activity in serum or plasma. The main problem with these assays was the lack of specificity for TRACP, since they also detected tartrateresistant AcPs from erythrocytes and platelets, which are not related to type-5 TRACP. Kinetic assays were improved by inhibiting non-TRACP AcPs by using fluoride or incubating the samples at 37°C for an hour before measurement [9]. However, these assays still measured total TRACP activity, and lacked the specificity for osteoclastderived TRACP 5b. An important finding was the substrate specificity of distinct AcPs and also TRACP 5a
and 5b. The most common substrate used to detect AcP activity is a 4-nitrophenylphosphate (4-NPP), which is hydrolyzed well by nontype 5 AcPs, as well as TRACP 5a and 5b. Instead, α-naphthyl phosphate and naphtholASBI phosphate have been shown to be more selective substrates and are hydrolyzed effectively by TRACP 5b, but only poorly by TRACP 5a and not at all by other nontype 5 AcPs [69]. Another approach for TRACP 5b-specific kinetic assay was developed using fluoride to inhibit nontype 5 AcPs and heparin to inhibit TRACP 5a activity during measurement [70]. The production of TRACP antibodies made it possible to develop immunoassays to specifically detect TRACP activity or TRACP protein concentration from serum or plasma. Most probably these assays are not selective for TRACP 5a or TRACP 5b, since the TRACP isoforms show very high immunological identity. Until this millennium, the lack of absolute specificity for bone-specific TRACP 5b had held back progress on the use of biochemical assay of TRACP 5b activity or protein concentration as a marker of bone resorption. Both kinetic assays and immunoassays for TRACP showed that individuals with normally or pathologically high rates of bone turnover have increased levels of serum total TRACP activity or protein concentration. Children with physiologically active bone growth and postmenopausal women were shown to have significantly elevated serum TRACP levels compared to healthy premenopausal adults [36, 71]. Marker levels were also elevated in cases where bone resorption is known to be increased, such as osteoporosis, Paget’s disease, humoral hypercalcemia of malignancy, multiple myeloma, osteomalacia, immobilization, and chronic renal failure. An early finding of elevated serum TRACP levels in patients with metastatic bone diseases was later confirmed by several research groups. Also, patients with primary hyperparathyroidism had markedly elevated TRACP levels, which after removal of parathyroid adenoma decreased to normal levels within 2 weeks [72]. In contrast, individuals with hypoparathyroidism, who were expected to have decreased rates of bone resorption, had appropriately less TRACP in serum than healthy controls. The association between serum TRACP and bone resorption has also been confirmed by investigating bone density simultaneously with serum TRACP. A significant inverse correlation has been observed between serum TRACP and bone mineral content or bone mineral density. Similarly, bone density assessed by quantitative ultrasound (QUS) from calcaneus showed a significant negative association with serum TRACP [73]. Furthermore, a significant correlation was observed between histomorphometric parameters of bone resorption and serum TRACP 5b in uremic patients [74]. Additionally, therapies known to
173
Chapter 10 Acid Phosphatases
decrease bone resorption, such as hormone replacement therapy, also decreased serum TRACP, confirming the close relation of the enzyme to bone degradation [11]. Similar to other bone metabolic markers, levels of serum TRACP are also strongly influenced by age, therefore age-matched reference material is essential for proper experimentation [73]. In healthy individuals, the activity of total serum TRACP measured at the pH optimum 5.5 comprises, in approximately equal amounts, the isoforms TRACP 5a and 5b [39]. However, total serum TRACP protein measured by two-site immunoassays contains almost 90% of nonosteoclastic, low-activity TRACP 5a isoform, the remaining 10% being osteoclastic, highly active TRACP 5b [39]. Thus, neither total TRACP activity nor total TRACP protein provides a sensitive and specific tool for assessing the rate of bone resorption.
J. Serum TRACP 5b as a Bone Resorption Marker TRACP 5b expression is strictly limited to boneresorbing osteoclasts, activated macrophages and dendritic cells, which in principle could all be the sources of circulating active TRACP 5b. However, it has been shown in in vitro studies that TRACP 5b is retained almost entirely intracellularly in macrophages and dendritic cells, which secrete exclusively TRACP 5a [37]. Thus, serum TRACP 5b is exclusively secreted by bone-resorbing osteoclasts. Indeed, the existing data demonstrate that osteoclastic TRACP has identical biochemical and physical properties as serum TRACP 5b. Therefore, serum TRACP 5b provides an excellent tool for assessing the rate of bone resorption. Elevated levels of serum TRACP 5b have been observed in conditions with increased bone resorption, such as healthy postmenopausal women and individuals with osteopenia, osteoporosis, or Paget’s disease [26]. Also, children have significantly elevated levels of serum TRACP 5b, which are even further elevated in individuals with vitamin D deficiency [75]. In contrast, antiresorptive treatment such as HRT and alendronate treatment decrease serum TRACP 5b levels in postmenopausal women, as expected [26]. Breast cancer patients with bone metastases have significantly elevated serum TRACP 5b levels [38], and the enzyme levels are closely associated with the metastatic burden. Bisphosphonates are normally used in pain relief and improvement of symptoms of breast cancer patients with bone lesions. Studies revealed that serum TRACP 5b levels specifically reflect the efficacy of bisphosphonate treatment. Similar to breast cancer, TRACP 5b was proven to be associated with the severity of bone disease in multiple myeloma, and specifically monitors the efficacy of antimyeloma treatment [76]. The progression of
the disease was also reflected by serum TRACP 5b, suggesting that TRACP 5b could be used as a predictive tool for bone disease in multiple myeloma. The value of markers for bone metabolism is not only to detect the rate of high bone turnover, but also to predict future bone loss or fractures. A large prospective study with over 1000 postmenopausal women showed that serum TRACP 5b can predict fracture and, in particular, fractures that engage the trabecular bone, such as vertebral fractures [77]. For further discussion of the applications of TRACP as a bone resorption marker, see Chapter 35.
K. Clinical Significance of Serum TRACP 5a Based on the recent data, it seems that serum TRACP 5a has its own specific biological significance in some specific diseases such as rheumatoid arthritis (RA) [78]. Serum total TRACP concentration, but not activity, was shown to be significantly elevated in patients with RA [78]. The elevated levels of TRACP concentration did not correlate with bone metabolic markers bone ALP or NTX, suggesting that the TRACP elevation was not osteoclastic in origin and may not be related to bone turnover. Instead, the levels of TRACP concentration in RA were significantly correlated with C-reactive protein, an acute-phase protein marker of inflammation and indicator of disease activity in RA [37]. Since as much as 90% of the total serum TRACP protein in RA is TRACP 5a, the total TRACP concentration gives a good estimate of TRACP 5a concentration [39]. Based on these results, the authors conclude that increased TRACP protein is a low-activity TRACP 5a isoform from a source other than osteoclasts, probably inflammatory macrophages and dendritic cells abundant in the synovial tissues of affected joints [37]. Further investigations are needed to confirm the clinical significance of serum TRACP 5a concentrations in RA and other chronic inflammatory conditions.
III. PROSTATIC ACID PHOSPHATASE (PAP) Human prostatic acid phosphatase (hPAP) is a glycoprotein synthesized in the epithelial cells of the prostate gland [79–82], and present in spermatic ejeculate [83–85]. This enzyme hydrolyzes a wide range of alkyl and aryl orthophosphate monoesters, including phosphotyrosine [86, 87] and nucleotides [88] and has also been found to dephosphorylate macromolecules, such as phosphopeptides and phosphoproteins [89, 90]. hPAP is categorized as an acid phosphatase, since in has an optimum pH of 4–6.
174
HELENA KAIJA ET AL.
A. Gene Structure of hPAP The gene encoding hPAP is located at chromosome 3q21-qter [91]. cDNA encodes a 354-residue protein with a calculated molecular mass of 41 ± 126 Da. The hPAP gene contains ten exons of 170, 96, 87, 153, 99, 93, 133, 83, 104, and 2098 bp, respectively. A 32 amino acid signal sequence and the first eight amino acids of the protein are encoded by exon 1, and the rest of the coding region and the 3′-untranslated region are covered by the exons 2–10 and 10, respectively [92]. When prostatic carcinoma or benign hyperplasia tissue is used as a source of RNA, a major transcription start site can be found 50 nt upstream of the ATG codon of the gene [92, 93]. The length of the 3′ noncoding region in hPAP cDNA varies between 646 and 1913 bp, although, in Northern blot analysis, a 3.3-kb mRNA species is usually observed [94]. The heterogeneity in the length of cDNA is explained by the multiple polyadenylation signals (AAATAA) following two copies of alu-type repetitive sequences in the 3′ noncoding region.
B. Protein Structure of hPAP Figure 2
The hPAP protein is a homodimer with subunits related by a twofold axis [95]. Each subunit comprises two domains. The larger, α/β type domain is composed of a central seven-stranded mixed β-sheet with helices on both sides, while the smaller α domain contains six α-helices and is formed mostly of a long-chain excursion from the first domain. The hPAP active site, which contains an essential histidine (His12) residue [96], is located in a large open cleft between the two domains. This allows the enzyme to accept a large variety of substrates [95]. The rate-limiting step is the breakdown of the covalent phosphoenzyme intermediate by the attack of water on the phosphoroamidate, resulting in the formation of a noncovalent enzyme–inorganic phosphate complex (Fig. 2). The active site was shown to contain two arginines probably involved in the binding of the negatively charged phosphate group of the substrate [97]. The presence of carboxylic acid residues, Asp or Glu, at the active site was indicated by Saini and van Etten [98] and van Etten [99]. Site-directed mutagenesis of hPAP shows His12 acts as an acceptor of the phospho group, Asp258 is a proton donor for the substrate-leaving group, and His257 may participate in substrate-binding or may facilitate the breakdown of the phosphoenzyme complex [100]. These results are consistent with the crystal structure of rPAP [95]. Most of the active site residues of rPAP come from the loops after strands 1 and 4 of the large domain.
Minimal reaction mechanism of rPAP. Modified from
[101a].
Residues Arg11, His12 and Arg15 are part of a sequence motif RHGXRXP characteristic for acid phosphatase [101]. These residues are part of a cluster of conserved amino acid residues in the center of the active site, consisting of residues of Arg11, His12, Arg15, Arg79, His257, and Asp258. The active sites of hPAP and rPAP are very similar in three-dimensional structure, except for the conformation of the Arg15 side chain [102]. A recent study [103] shows that Trp174 was at the active site of the human enzyme, because it was protected by the competitive inhibitor tartrate in the DNPS-Cl modification studies. This is also consistent with the location of a homologous residue in the structure of the rat enzyme. Denaturation–renaturation and subunit reassociation studies show that hPAP activity depends on dimer formation [104]. Oligomerization of rPAP using site-directed mutagenesis found that mutants W106E and H112D, as well as the double mutant W106E/H112D, are monomers without catalytic activity or the ability to bind tartrate [105]. The His112 side-chain forms a hydrogen bond with the Asp76 side-chain between two subunits, and the sidechains of Trp106 are stacked on top of each other across the twofold axis relating the two subunits [95]. Since the PAP active site is located far from the subunits’ interface,
175
Chapter 10 Acid Phosphatases
these results indicate that the formation of dimers induces structural changes which propagate to the active site. Human PAP from native sources possesses molecular heterogeneity. Isoelectric focusing gives up to 30 variants of hPAP from normal, malignant, or hyperplastic prostatic tissue and sera from patients with prostate cancer and seminal fluid [83, 84, 106]. Variation in the amount of sialic acid and other sugar residues and post-translational deamination of glutamine and asparagine account for part of the heterogeneity of hPAP [101, 107]. Furthermore, the carboxyterminal amino acid of hPAP can be threonine, glutamic acid, or aspartic acid [101]. The heterogeneity of hPAP could also be due to the existence of true isoenzymes consisting of different polypeptide chains.
progression [127]. The binding of phosphotyrosine to PAP has been characterized by theoretical modeling and experimental work with peptides derived from the residue sequences of EGFR- and ErbB-2-containing autophosphorysable tyrosine [128, 129] and altogether results confirm that PAP is tyrosine phosphatase as suggested earlier [90]. PAP might function as a dual-specificity protein tyrosine phosphatase as it has been shown to possess affinity also for phosphoserine in proteins in vitro [130]. Binding of PAP to the main high-density lipoprotein apolipoprotein A-I (apoA-I) has also been demonstrated [131]. ApoA-I is lipidated during its progress through intracellular vesicle traffic from the cell surface into early endosomes and further into late endosomes and finally back to the cell surface as a nascent HDL particle [132].
C. Function of hPAP D. Expression of hPAP in Prostate Human PAP is one of the prostate epithelium-specific differential antigens [108–110]. There are two forms of hPAP: one is intracellular and the other secreted [111]. The physiological substrate of hPAP is unknown. However, hPAP is present in large amounts in seminal ejeculate [85], suggesting a physiological role in fertility [112]. The serum activity of the enzyme is frequently elevated in patients with prostate carcinoma and correlates with tumor progression [113–116]. In vitro, hPAP hydrolyzes phosphorycholine, which is found in semen [98], and phosphocreatine, an intracellular high-energy compound present in seminal plasma [117]. Suggested substrates include seminal phosphocholine [105, 118], semenogelins [119], EGFR/ErbB-2 [120], and lysophosphatidic acid [121]. It also has amidolytic activity on semenogelins [122]. hPAP possesses intrinsic protein tyrosine phosphatase (PTP) activity [90, 105, 123], and several lines of evidence indicate that cellular hPAP functions as a neutral PTP, although it shares little sequence homology with other typical PTPs [101, 124, 125]. hPAP also has amidolytic activity on a seminal vesicle protein semenogelin I. The enzyme is able to cleave both peptide substrates derived from the semenogelin sequence and native semenogelin I. The main cleavage sites are at Tyr 292 and Ser 170 [122]. In addition, hPAP interacts with ErbB-2, a member of the erbB receptor tyrosine kinase family, in the androgenpromoted growth of human prostate cancer cells. Androgen-stimulated cell growth concurs with downregulation of cellular hPAP, an elevated p-Tyr level of ErbB-2 and the activation of mitogen-activated protein kinases [126]. Cellular hPAP can down-regulate prostate cancer cell growth, at least partially, by dephosphorylating c-ErbB-2. Therefore, decreased cellular hPAP expression in cancer cells may be involved in prostate cancer
Immunohistochemistry using specific antibodies against PAP has demonstrated that the enzyme is located in the columnar, secretory epithelial cells of the prostate [82, 133–137]. In situ hybridization analysis has shown that hPAP mRNA is confined to the glandular and ductal epithelial cells of the prostate, and that stromal cells are devoid of this mRNA [82]. No hPAP mRNA has been detected in human liver, lung, pancreatic cancer tissue, placenta, breast cancer cells, mononuclear blood cells, or acute promyelocytic leukemia cells, nor in spleen, thymus, testis, ovary, small intestine, colon, or peripheral blood leukocytes [138]. Little is known about the mechanism of tissue-specific regulation of the hPAP gene at the molecular level. Zeliavinski et al. [139] have shown that, in addition to the basic promoter, the region between –1258/−779 is able to enhance the PAP promoter activity in PC-3 and DU-145 human prostate cancer cells, but not in nonprostate cancer cells, indicating that this region is involved in governing the cell type-specific expression of the hPAP gene. Shan et al. [140] showed that a construct containing the sequence of hPAP between the nucleotides –734 and +467 in front of the CAT reporter gene was significantly expressed in the prostate of transgenic mice, while the proximal promoter –734/+50 alone achieved low levels of CAT mRNA in all tissues analyzed. This suggests the involvement of regulatory elements in the intron area and/or the possible interactions of the transcription factors binding along the whole DNA area. It has been shown previously that a 12-bp sequence, GAAAATATGATA is involved in prostatespecific and androgen receptor-dependent gene regulation of the rat probasin gene [141]. In hPAP, five homologs for this previously identified prostate-specific DNA-binding
176
HELENA KAIJA ET AL.
site were found between the nucleotides –734 and +467 (Fig. 3). Deletions of the different binding sites of the prostatic transcription factor resulted in bidirectional effects depending on the location of the sites and the hormone status in LNCaP cells. Site C at the proximal promoter of hPAP acts as a positive regulatory element in prostatic cells in an androgen-dependent manner, while site E or sites D and E act as negative regulatory element(s) in the androgendepleted environment [140].
E. Androgen Regulation of hPAP In cell culture models, both up- and down-regulation by androgens have been reported for hPAP. The amount of hPAP released into LNCaP cell culture medium was decreased to 26% of the control level in 7 days when a synthetic androgen, R1881, was present in charcoalstripped serum [142]. Accordingly, 5α-dihydrotestosterone (DHT) treatment was found to decrease the activity of hPAP in these cells. A stimulatory effect of androgen on hPAP secretion has been confirmed by Horozewicz et al. [143] and Lin et al. [144]. hPAP mRNA levels are also increased when tissue slides derived from benign prostatic hyperplasia are treated with DHT and fibroblastic growth factor (bFGF) [145]. A biphasic pattern of the effect of androgen on LNCaP cells has been reported: stimulation of growth and inhibition of hPAP secretion was detected at less than 1 nM concentrations of androgen, while an opposite effect was observed at higher concentrations [146, 147]. Shan et al. [148] showed that reporter constructs of hPAP promoter covering the region –734/+467 were functional in both prostatic and nonprostatic cell lines in transient transfections. This region contained two putative AREs, which have been shown to have androgen receptor-binding ability in vitro, but the promoter could not be induced with androgen, glucocorticoid, or progesterone, indicating that steroids cannot directly regulate hPAP gene expression via receptor binding to these AREs.
Figure 3
Location of the prostate-specific DNA-binding site GAAAATATGATA-related sequences and potential androgen response elements in the schematic representation of the regulatory region of the hPAP gene. Binding capacities of the elements for the prostatic protein in vitro are also marked (−, +, +++). In vitro androgen receptor-binding capacities are indicated by + or ++ above AREs.
Androgen may regulate hPAP expression differently in diverse physiological or pathological conditions. Both up- and down-regulation of hPAP by 12-otetradecanoyl phorbol-13-acetate (TPA), a protein kinase C (PKC) activator, has been reported [142, 149]. Lin et al. [149] reported that TPA is able to increase the secretion of hPAP in a dose- and time-dependent fashion in the androgen-responsive LNCaP cell line. This TPA stimulation of hPAP secretion was more potent than the conventional stimulating agent DHT at the same concentration. Furthermore, the action of TPA and DHT on hPAP secretion was blocked by five different PKC inhibitors. DHT, as well as TPA, could rapidly modulate PKC activity. Therefore, PKC can regulate hPAP secretion and may also be involved in the DHT action on hPAP secretion. So far, very little is known about physiological substrates of acid phosphatases, but increasing interest in kinases/ phosphatases is obviously opening a new insight into this research area.
References 1. Vincent, J. B., Crowder, M. W., and Averill, B. A. (1992). Hydrolysis of phosphate monoesters: a biological problem with multiple chemical solutions. Trends. Biochem. Sci. 17, 105–110. 2. Lohse, D. L., Denu, J. M., and Dixon, J. E. (1995). Insights derived from the structures of the Ser/Thr phosphatases calcineurin and protein phosphatase 1. Structure 3, 987–990. 3. Barford, D., Das, A. K., and Egloff, M. P. (1998). The structure and mechanism of protein phosphatases: insights into catalysis and regulation. Annu. Rev. Biophys. Biomol. Struct. 27, 133–164. 4. Li, C. Y., Yam, L. T., and Lam, K. W. (1970). Studies of acid phosphatase isoenzymes in human leukocytes demonstration of isoenzyme cell specificity. J. Histochem. Cytochem. 18, 901–910. 5. Saltiel, A. R., and Kahn, C. R. (2001). Insulin signalling and the regulation of glucose and lipid metabolism. Nature 414, 799–806. 6. Hashimoto, N. Feener, E. P., Zhang, W. R., and Goldstein, B. J. (1992). J. Biol. Chem. 267, 13811–138142. 7. Jackson, S. P. (1992). Insulin receptor protein-tyrosine phosphatases. Leukocyte common antigen-related phosphatase rapidly deactivates the insulin receptor kinase by preferential dephosphorylation of the receptor regulatory domain. Trends. Cell. Biol. 2, 104–108. 8. Li, C. Y., Yam, L. T., and Lam, K. W. (1970). Acid phosphatase isoenzyme in human leukocytes in normal and pathologic conditions. J. Histochem. Cytochem. 18 (7), 473–481. 9. Lau, K. H., Freeman, T. K., and Baylink, D. J. (1987a). Purification and characterization of an acid phosphatase that displays phosphotyrosylprotein phosphatase activity from bovine cortical bone matrix. J. Biol. Chem. 262 (3), 1389–1397. 10. Lam, K. W., and Yam, L. T. (1977). Biochemical characterization of the tartrate-resistant acid phosphatase of human spleen with leukemic reticuloendotheliosis as a pyrophosphatase. Clin. Chem. 23 (1), 89–94. 11. Halleen, J., Hentunen, T. A., Hellman, J., and Väänänen, H. K. (1996). Tartrate-resistant acid phosphatase from human bone: purification and development of an immunoassay. J. Bone. Miner. Res. 11 (10), 1444–1452. 12. Anderson, T. R., and Toverud, S. U. (1986). Purification and characterization of purple acid phosphatase from developing rat bone. Arch. Biochem. Biophys. 247 (1), 131–139.
Chapter 10 Acid Phosphatases 13. Ek-Rylander, B., Bergman, T., and Andersson, G. (1991). Characterization of a tartrate-resistant acid phosphatase (ATPase) from rat bone: hydrodynamic properties and N-terminal amino acid sequence. J. Bone. Miner. Res. 6 (4), 365–373. 14. Ljusberg, J., Ek-Rylander, B., and Andersson, G. (1999). Tartrate-resistant purple acid phosphatase is synthesized as a latent proenzyme and activated by cysteine proteinases. Biochem. J. 343 Pt 1, 63–69. 15. Funhoff, E. G., Ljusberg, J., Wang, Y., Andersson, G., and Averill, B. A. (2001). Mutational analysis of the interaction between active site residues and the loop region in mammalian purple acid phosphatases. Biochemistry 40 (38), 11614–11622. 16. Yam, L. T., Li, C. Y., and Lam, K. W. (1971). Tartrate-resistant acid phosphatase isoenzyme in the reticulum cells of leukemic reticuloendotheliosis. N. Engl. J. Med. 284 (7), 357–360. 17. Väänänen, H. K., Karhukorpi, E. K., Sundquist, K., Wallmark, B., Roininen, I., Hentunen, T., Tuukkanen, J., and Lakkakorpi, P. (1990). Evidence for the presence of a proton pump of the vacuolar H(+)-ATPase type in the ruffled borders of osteoclasts. J. Cell Biol. 111 (3), 1305–1311. 18. Clark, S. A., Ambrose, W. W., Anderson, T. R., Terrell, R. S., and Toverud, S. U. (1989). Ultrastructural localization of tartrate-resistant, purple acid phosphatase in rat osteoclasts by histochemistry and immunocytochemistry. J. Bone Miner. Res. 4 (3), 399–405. 19. Halleen, J. M., Räisänen, S., Salo, J. J., Reddy, S. V., Roodman, G. D., Hentunen, T. A., Lehenkari, P. P., Kaija, H., Vihko, P., and Väänänen, H. K. (1999b). Intracellular fragmentation of bone resorption products by reactive oxygen species generated by osteoclastic tartrateresistant acid phosphatase. J. Biol. Chem. 274 (33), 22907–22910. 20. Reinholt, F. P., Widholm, S. M., Ek-Rylander, B., and Andersson, G. (1990). Ultrastructural localization of a tartrate-resistant acid ATPase in bone. J. Bone Miner. Res. 5 (10), 1055–1061. 21. Räisänen, S. R., Halleen, J., Parikka, V., and Väänänen, H. K. (2001). Tartrate-resistant acid phosphatase facilitates hydroxyl radical formation and colocalizes with phagocytosed Staphylococcus aureus in alveolar macrophages. Biochem. Biophys. Res. Commun. 288 (1), 142–150. 22. Sheu, T. J., Schwarz, E. M., Martinez, D. A., O’Keefe, R. J., Rosier, R. N., Zuscik, M. J., and Puzas, J. E. (2003). A phage display technique identifies a novel regulator of cell differentiation. J. Biol. Chem. 278 (1), 438–443. 23. Minkin, C. (1982). Bone acid phosphatase: tartrate-resistant acid phosphatase as a marker of osteoclast function. Calcif. Tissue Int. 34 (3), 285–290. 24. Brehme, C. S., Roman, S., Shaffer, J., and Wolfert, R. (1999). Tartrateresistant acid phosphatase forms complexes with alpha2-macroglobulin in serum. J. Bone Miner. Res. 14 (2), 311–318. 25. Lau, K. H., Onishi, T., Wergedal, J. E., Singer, F. R., and Baylink, D. J. (1987b). Characterization and assay of tartrate-resistant acid phosphatase activity in serum: potential use to assess bone resorption. Clin. Chem. 33 (4), 458–462. 26. Halleen, J. M., Alatalo, S. L., Suominen, H., Cheng, S., Janckila, A. J., Väänänen, H. K. (2000). Tartrate-resistant acid phosphatase 5b: a novel serum marker of bone resorption. J. Bone Miner. Res. 15 (7), 1337–1345. 27. Sträter, N., Klabunde, T., Tucker, P., Witzel, H., and Krebs, B. (1995). Crystal structure of a purple acid phosphatase containing a dinuclear Fe(III)-Zn(II) active site. Science 268 (5216), 1489–1492. 28. Guddat, L. W., McAlpine, A. S., Hume, D., Hamilton, S., de Jersey, J., and Martin, J. L. (1999). Crystal structure of mammalian purple acid phosphatase. Arch. Biochem. Biophys. 352 (1), 97–102. 29. Lindqvist, Y., Johansson, E., Kaija, H., Vihko, P., and Schneider, G. (1999). Three-dimensional structure of a mammalian purple acid phosphatase at 2.2 A resolution with a mu-(hydr)oxo bridged di-iron center. J. Mol. Biol. 291 (1), 135–147. 30. Uppenberg, J., Lindqvist, F., Svensson, C., Ek-Rylander, B., and Andersson, G. (1999). Crystal structure of a mammalian purple acid phosphatase. J. Mol. Biol. 290 (1), 201–211.
177 31. Davis, J. C., and Averill, B. A. (1982). Evidence for a spin-coupled binuclear iron unit at the active site of the purple acid phosphatase from beef spleen. Proc. Natl Acad. Sci, USA 79 (15), 4623–4627. 32. Vincent, J. B., and Averill, B. A. (1990b). Sequence homology between purple acid phosphatases and phosphoprotein phosphatases. Are phosphoprotein phosphatases metalloproteins containing oxide-bridged dinuclear metal centers? FEBS Lett 263 (2), 265–268. 33. Kaija, H., Jia, J., Lindqvist, Y., Andersson, G. and Vihko, P. (1999). Tartrate-resistant bone acid phosphatase: large-scale production and purification of the recombinant enzyme, characterization, and crystallization. J. Bone Miner. Res. 14 (3), 424–430. 34. Baumbach, G. A., Saunders, P. T., Bazer, F. W., and Roberts, R. M. (1984). Uteroferrin has N-asparagine-linked high-mannose-type oligosaccharides that contain mannose 6-phosphate. Proc. Natl Acad. Sci. USA 81 (10), 2985–2989. 35. Zhang, Z. Y., Wang, Y., Wu, L., Fauman, E. B., Stuckey, J. A., Schubert, H. L., Saper, M. A., and Dixon, J. E. (1994). The Cys(X)5Arg catalytic motif in phosphoester hydrolysis. Biochemistry 33 (51), 15266–15270. 36. Lam, W. K., Eastlund, D. T., Li, C. Y., and Yam, L. T. (1978a). Biochemical properties of tartrate-resistant acid phosphatase in serum of adults and children. Clin. Chem. 24 (7), 1105–1108. 37. Janckila, A. J., Neustadt, D. H., Nakasato, Y. R., Halleen, J. M., Hentunen, T. and Yam, L. T. (2002a). Serum tartrate-resistant acid phosphatase isoforms in rheumatoid arthritis. Clin Chim Acta 320 (1–2), 49–58. 38. Halleen, J. M., Alatalo, S. L., Janckila, A. J., Woitge, H. W., Seibel, M. J., and Väänänen, H. K. (2001). Serum tartrate-resistant acid phosphatase 5b is a specific and sensitive marker of bone resorption. Clin. Chem. 47 (3), 597–600. 39. Janckila, A. J., Takahashi, K., Sun, S. Z., and Yam, L. T. (2001b). Tartrate-resistant acid phosphatase isoform 5b as serum marker for osteoclastic activity. Clin. Chem. 47 (1), 74–80. 40. Walsh, N. C., Cahill, M., Carninci, P., Kawai, J., Okazaki, Y., Hayashizaki, Y., Hume, D. A., and Cassady, A. I. (2003). Multiple tissuespecific promoters control expression of the murine tartrate-resistant acid phosphatase gene. Gene 307, 111–123. 41. Reddy, S. V., Scarcez, T., Windle, J. J., Leach, R. J., Hundley, J. E., Chirgwin, J. M., Chou, J. Y., and Roodman, G. D. (1993). Cloning and characterization of the 5′-flanking region of the mouse tartrateresistant acid phosphatase gene. J. Bone. Miner. Res. 8 (10), 1263–1270. 42. Simmen, R. C., Srinivas, V., and Roberts, R. M. (1989). CDNA sequence, gene organization, and progesterone induction of mRNA for uteroferrin, a porcine uterine iron transport protein. DNA 8 (8), 543–554. 43. Alcantara, O., Reddy, S. V., Roodman, G. D., and Boldt, D. H. (1996). Inhibition of tartrate-resistant acid phosphatase gene expression by hemin and protoporhyrin IX. Identification of a hemin-responsive inhibitor of transcription. Blood 88 (6), 2288–2297. 44. Reddy, S. V., Alcantara, O., Roodman, G. D., and Boldt, D. H. (1996). Inhibition of tartrate-resistant acid phosphatase gene expression by hemin and protoporphyrin IX. Identification of a hemin-responsive inhibitor of transcription. Blood 88 (6), 2288–2297. 45. Fleckenstein, E. C., Dirks, W. G., and Drexler, H. G. (2000). The human tartrate-resistant acid phosphatase (TRAP): involvement of the hemin responsive elements (HRE) in transcriptional regulation. Leuk Lymphoma 36 (5–6): 603–612. 46. Luchin, A., Suchting, S., Merson, T., Rosol, T. J., Hume, D. A., Cassady, A. I., and Ostrowski, M. C. (2001). Genetic and physical interactions between Microphthalmia transcription factor and PU.1 are necessary for osteoclast gene expression and differentiation. J. Biol. Chem. 276 (39), 36703–36710. 47. Liu, Y., Shi, Z., Silveira, A., Liu, J., Sawadogo, M., Yang, H., and Feng, X. (2003). Involvement of upstream stimulatory factors 1 and 2 in RANKL-induced transcription of tartrate-resistant acid phosphatase
178
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
gene during osteoclast differentiation. J. Biol. Chem. 278 (23), 20603–20611. Halleen, J. M., Kaija, H., Stepan, J. J., Vihko, P., and Väänänen, H. K. (1998b). Studies on the protein tyrosine phosphatase activity of tartrateresistant acid phosphatase. Arch. Biochem. Biophys. 352 (1), 97–102 Ek-Rylander, B., Flores, M., Wendel, M., Heinegard, D., and Andersson, G. (1994). Dephosphorylation of osteopontin and bone sialoprotein by osteoclastic tartrate-resistant acid phosphatase. Modulation of osteoclast adhesion in vitro. J. Biol. Chem. 269 (21), 14853–14856. Janckila, A. J., Woodford, T. A., Lam, K. W., Li, C. Y., and Yam, L. (1992). Protein-tyrosine phosphatase activity of hairy cell tartrateresistant acid phosphatase. Leukemia 6 (3), 199–203. Schlosnagle, D. C., Sander, E. G., Bazer, F. W., and Roberts, R. M. (1976). Requirement of an essential thiol group and ferric iron for the activity of the progesterone-induced porcine uterine purple phosphatase. J. Biol. Chem. 251 (15), 4680–4685. Klabünde, T., Sträter, N., Frohlich, R., Witzel, H., and Krebs, B. (1996). Mechanism of Fe(III)-Zn(II) purple acid phosphatase based on crystal structures. J. Mol. Biol. 259 (4), 737–748. Wang, X., Randall, C. R., True, A. E., and Que, L., Jr. (1996). X-ray absorption spectroscopic studies of the FeZn derivative of uteroferrin. Biochemistry 35 (44), 13946–13954. Merkx, M., Pinkse, M. W., and Averill, B. A. (1999). Evidence for nonbridged coordination of p-nitrophenyl phosphate to the dinuclear Fe(III)-M(II) center in bovine spleen purple acid phosphatase during enzymatic turnover. Biochemistry 38 (31), 9914–9925. Flores, M. E., Norgard, M., Heinegard, D., Reinholt, F. P., and Andersson, G. (1992). RGD-directed attachment of isolated rat osteoclasts to osteopontin, bone sialoprotein, and fibronectin. Exp. Cell. Res. 201 (2), 526–530. Sibille, J. C., Doi, K., and Aisen, P. (1987). Hydroxyl radical formation and iron-binding proteins. Stimulation by the purple acid phosphatases. J. Biol. Chem. 262 (1), 59–62. Hayman, A. R., and Cox, T. M. (1994). Purple acid phosphatase of the human macrophage and osteoclast. Characterization, molecular properties, and crystallization of the recombinant di-iron-oxo protein secreted by baculovirus-infected insect cells. J. Biol. Chem. 269 (2), 1294–1300. Davies, K. J., and Delsignore, M. E. (1987). Protein damage and degradation by oxygen radicals. III. Modification of secondary and tertiary structure. J. Biol. Chem. 262 (20), 9908–9913. Garrett, I. R., Boyce, B. F., Oreffo, R. O., Bonewald, L., Poser, J., and Mundy, G. R. (1990). Oxygen-derived free radicals stimulate osteoclastic bone resorption in rodent bone in vitro and in vivo. J. Clin. Invest. 85 (3), 632–639. Bune, A. J., Hayman, A. R., Evans, M. J., and Cox, T. M. (2001). Mice lacking tartrate-resistant acid phosphatase (Acp 5) have disordered macrophage inflammatory responses and reduced clearance of the pathogen, Staphylococcus aureus. Immunology 102 (1), 103–113. Kaija, H. Alatalo, S. L., Halleen, J. M., Lindqvist, Y., Schneider, G., Väänänen, H. K., and Vihko, P. (2002). Phosphatase and oxygen radicalgenerating activities of mammalian purple acid phosphatase are functionally independent. Biochem. Biophys. Res. Commun. 292 (1), 128–132. Beck, J. L., Durack, M. C., Hamilton, S. E., and de Jersey, J. (1999). Irreversible inactivation of purple acid phosphatase by hydrogen peroxide and ascorbate. J. Inorg. Biochem. 73 (4), 245–252. Hayman, A. R., Jones, S. J., Boyde, A., Foster, D., Colledge, W. H., Carlton, M. B., Evans, M. J., and Cox, T. M. (1996). Mice lacking tartrate-resistant acid phosphatase (Acp 5) have disrupted endochondral ossification and mild osteopetrosis. Development 122 (10), 3151–3162. Angel, N. Z., Walsh, N., Forwood, M. R., Ostrowski, M. C., Cassady, A. I., and Hume, D. A. (2000). Transgenic mice overexpressing tartrate-resistant acid phosphatase exhibit an increased rate of bone turnover. J. Bone Miner. Res. 15 (1), 103–110.
HELENA KAIJA ET AL. 65. Hollberg, K., Hultenby, K., Hayman, A., Cox, T., and Andersson, G. (2002). Osteoclasts from mice deficient in tartrate-resistant acid phosphatase have altered ruffled borders and disturbed intracellular vesicular transport. Exp. Cell. Res. 279 (2), 227–238. 66. Suter, A., Everts, V., Boyde, A., Jones, S. J., Lullmann-Rauch, R., Hartmann, D., Hayman, A. R., Cox, T. M., Evans, M. J., Meister, T., von Figura, K., and Saftig, P. (2001). Overlapping functions of lysosomal acid phosphatase (LAP) and tartrate-resistant acid phosphatase (Acp5) revealed by doubly deficient mice. Development 128 (23), 4899–4910. 67. Räisänen, S. (2004). Differential effects of carbonic anhydrase III and tartrate-resistant acid phosphatase on intracellular radical balance. Institute of Biomedicine, Department of Anatomy, Turku, University of Turku: 1–93 [Thesis]. 68. Saftig, P., Hartmann, D., Lullmann-Rauch, R., Wolff, J., Evers, M., Koster, A., Hetman, M., von Figura, K., and Peters, C. (1997). Mice deficient in lysosomal acid phosphatase develop lysosomal storage in the kidney and central nervous system. J. Biol. Chem. 272 (30), 18628–18635. 69. Janckila, A. J., Takahashi, K., Sun, S. Z., and Yam, L. T. (2001a). Naphthol-ASBI phosphate as a preferred substrate for tartrate-resistant acid phosphatase isoform 5b. J. Bone. Miner. Res. 16 (4), 788–793. 70. Nakanishi, M., Yoh, K., Miura, T., Ohasi, T., Rai, S. K., and Uchida, K. (2000). Development of a kinetic assay for band 5b tartrate-resistant acid phosphatase activity in serum. Clin. Chem. 46 (4), 469–473. 71. Halleen, J. M., Karp, M., Viloma, S., Laaksonen, P., Hellman, J., Käkönen, S. M., Stepan, J. J., Holmes, S., Väänänen, H., and Pettersson, K. (1999a). Two-site immunoassays for osteoclastic tartrate-resistant acid phosphatase based on characterization of six monoclonal antibodies. J. Bone. Miner. Res. 14 (3), 464–469. 72. Stepan, J. J., Silinkova-Malkova, E., Havranek, T., Formankova, J., Zichova, M., Lachmanova, J., Strakova, M., Broulik, P., and Pacovsky, V. (1983). Relationship of plasma tartrate resistant acid phosphatase to the bone isoenzyme of serum alkaline phosphatase in hyperparathyroidism. Clin. Chim. Acta. 133 (2), 189–200. 73. Nakanishi, M., Yoh, K., Uchida, K., Maruo, S., Rai, S. K., and Matsuoka, A. (1999) Clinical usefulness of serum tartrate-resistant fluoride-sensitive acid phosphatase activity in evaluating bone turnover. J. Bone Miner. Metab. 17 (2), 125–130. 74. Chu, P., Chao, T. Y., Lin, Y. F., Janckila, A. J., and Yam, L. T. (2003). Correlation between histomorphometric parameters of bone resorption and serum type 5b tartrate-resistant acid phosphatase in uremic patients on maintenance hemodialysis. Am. J. Kidney Dis. 41 (5), 1052–1059. 75. Cheng, S., Tylavsky, F., Kröger, H., Kärkkäinen, M., Lyytikäinen, A., Koistinen, A., Mahonen, A., Alen, M., Halleen, J., Väänänen, K., and Lamberg-Allardt, C. (2003). Association of low 25-hydroxyvitamin D concentrations with elevated parathyroid hormone concentrations and low cortical bone density in early pubertal and prepubertal Finnish girls. Am. J. Clin. Nutr. 78 (3), 485–492. 76. Terpos, E., de la Fuente, J., Szydlo, R., Hatjiharissi, E., Viniou, N., Meletis, J., Yataganas, X., Goldman, J. M., and Rahemtulla, A. (2003). Tartrate-resistant acid phosphatase isoform 5b: a novel serum marker for monitoring bone disease in multiple myeloma. Int. J. Cancer. 106 (3), 455–457. 77. Gerdhem, P., Ivaska, K. K., Alatalo, S. L., Halleen, J. M., Hellman, J., Isaksson, A., Pettersson, K., Väänänen, H. K., Åkesson, K., and Obrant, K. J. (2004). Biochemical markers of bone metabolism and prediction of fracture in elderly women. J. Bone Miner. Res. 19 (3), 386–393. 78. Nakasato, Y. R., Janckila, A. J., Halleen, J. M., Väänänen, H. K., Walton, S. P., and Yam, L. T. (1999). Clinical significance of immunoassays for type-5 tartrate-resistant acid phosphatase. Clin. Chem. 45 (12), 2150–2157. 79. Derechin, M., Ostrowski, W., Galka, M., and Barnard, E. A. (1971). Acid phosphomonoesterase of human prostate: molecular weight dissociation and chemical composition. Biochim. Biophys. Acta. 250, 143–154.
Chapter 10 Acid Phosphatases 80. Ostrowski, W., Bhargava, A. K., Dziembor, E., Gizler, M., Gryszkiewicz, J.T. and Barnard, E. A. (1976). Acid phosphomonoesterase of human prostate. Carbohydrate content and optical properties. Biochim. Biophys. Acat. 453, 262–269. 81. Risley, M. J., and van Etten, R. L. (1987). Structures of the carbohydrate moieties of human prostatic acid phosphatase elucidated by 1H nuclear magnetic resonance spectroscopy. Arch. Biochem. Biophys. 258, 404–412. 82. Hakalahti, L., Vihko, P., Henttu, P., Autio-Harmainen, H., Soini, Y., and Vihko, R. (1993). Evaluation of PAP and PSA gene expression in prostatic hyperplasia and prostatic carcinoma using Northern blot analysis, in situ hybridization and immunochemical stainings with monoclonal and bispecific antibodies. Int. J. Cancer 55, 590–597. 83. Vihko, P. (1978). Characterization of the principal human prostatic acid phosphatase isoenzyme, purified by affinity chromatography and isolectric focusing. Part II. Clin. Chem. 24, 1783–1787. 84. Vihko, P., Kontturi, M., and Korhonen, L. K. (1978). Purification of human prostatic acid phosphatase by affinity chromatography and isoelectric focusing. Part I. Clin. Chem. 24, 466–470. 85. Rönnberg, L., Vihko, P., Sajanti, E., and Vihko, R. (1981). Clomiphene citrate administration to normogonadotropic subfertile men: Blood hormone changes and activation of acid phosphatase in seminal fluid. Int. J. Androl. 4, 372–378. 86. Apostol, I., Kuciel, R., Wasylewska, E., and Ostrowski, W. S. (1985). Phosphotyrosine as a substrate of acid and alkaline phosphatases. Acta. Biochem. Pol. 32, 187–197. 87. Vihko, P., Kurkela, R., Porvari, K., Herrala, A., Lindfors, A., Lindqvist, Y., and Schneider, G. (1993). Rat acid phosphatase: Overexression of active secreted enzyme by recombinant baculovirus-infected insect cells, molecular properties and crystallization. Proc. Natl Acad. Sci. USA. 90, 799–803. 88. Dziembor-Gryszkiewicz, E., Fikus, M., Kazimierczuk, Z., and Ostrowski, W. (1978). Activity of human prostatic acid phosphatase toward purine 5′-phosphonucleosides. Bull. Acad. Pol. Sci. Biol. 26, 815–821. 89. Wasylewska, E., Czubak, J., and Ostrowski, W. S. (1983). Phosphoprotein phosphatase activity of human prostate acid phosphatase. Acta. Biochim. Pol. 30, 175–184. 90. Li, H. X., Chernoff, J., Chen, L. B., and Kirschonbaum, A. (1984). A phosphotyrosyl-protein activity associated with acid phosphatase from human prostate gland. Eur. J. Biochem. 138, 45–51. 91. Winqvist, R., Virkkunen, P., Grzeschik, K. H., and Vihko, P. (1989). Chromosomal localization to 3q21→qter and two TaqI RFLPs of the human prostae-specific acid phosphatase gene (ACCP). Cytogenet. Cell. Genet. 52, 68–71. 92. Virkkunen, P., Hedberg, P., Palvimo, J., Birr, E., Porvari, K., Ruokonen, M., Taavitsainen, P., Jänne, O., and Vihko, P. (1994). Structure comparison of human and rat prostate-specific acid phosphatase genes and their promoters: identification of putative androgen response elements. Biochem. Biophys. Res. Comm. 202, 49–57. 93. Sharief, F. S., and Li, S. S. L. (1992). Structure of human prostatic acid phosphatase gene. Biochem. Biophys. Res. Comm. 184, 1468–1476. 94. Vihko, P., Virkkunen, P., Henttu, P., Roiko, K., Solin, T., and Huhtala, M. L. (1988). Molecular cloning and sequence analysis of cDNA encoding human prostatic acid phosphatase. FEBS Lett 236, 275–281. 95. Schneider, G., Lindqvist, Y., and Vihko, P. (1993). Three-dimensional structure of rat acid phosphatase. EMBO. J. 12, 2609–2615. 96. Davidson, R. E. (1990). Structural aspects of human prostatic acid phosphatase and fast atom bombardment mass spectrometric analysis of E. coli L-threonine deaminase. Ph.D. thesis, Purdue University, West Lafayette, IN. 97. Mctigue, J. J., and van Etten, R. L. (1978). An essential arginine residue in human prostatic acid phosphatase. Biochim. Biophys. Acta. 523, 422–429.
179 98. Saini, M. S., and van Etten, R. L. (1979). An essential carboxylic acid group in human prostate acid phosphatase. Biochim. Biophys. Acta. 568, 370–376. 99. van Etten, R. L. (1982). Human prostatic acid phosphatase: a histidine phosphatase. Ann. N. Y. Acad. Sci. 390, 27–51. 100. Ostanin, K., Saeed, A., and van Etten, R. L. (1994). Heterologous expression of human prostatic acid phosphatase and site-directed mutagenesis of the enzyme active site. J. Biol. Chem. 269, 9680–9684. 101. Van Etten, R. L., Davidson, R., Stevis, P. E., MacArthur, H., and Moore, D. L. (1991). Covalent structure, disulfide bonding, and identification reactive surface and active site residues of human prostatic acid phosphatase. J. Biol. Chem. 266, 2313–2319. 101a.Lindqvist, Y., Schneider, G., and Vihko, P. (1994). Crystal structure of rat acid phosphatase complexed with the transition-state analogs vanadate and molybdate. Implications for the reaction mechanism. Eur. J. Biochem. 221, 139–142. 102. Lindqvist, Y., Schneider, G., and Vihko, P. (1993). Three-dimensional structure of rat acid phosphatase in complex with L(+)-tartrate. J. Biol. Chem. 268, 20744–20746. 103. Zhang, Z., Ostanin, K., and van Etten, R. L. (1997). Covalent modification and site-directed mutagenesis of an active site tryptophan on human prostatic acid phosphatase. Acta. Biochim. Pol. 44, 659–672. 104. Kuciel, R., Bakalova, A., Mazurkiewicz, A., Bilska, A., and Ostrowski, W. (1990). Is the subunit of prostatic acid phosphatase active? Reversible denaturation of prostatic acid phosphatase. Biochem. Int. 22, 329–334. 105. Porvari, K. S., Herrala, A. M., Kurkela, R. M., Taavitsainen, P. A., Lindqvist, Y., Schneider, G., and Vihko, P. T. (1994). Site-directed mutagenesis of prostatic acid phosphatase: catalytically important aspartic acid 258, substrate specificity, and oligomerization. J. Biol. Chem. 269, 22642–22646. 106. Chu, T. M., Wang, M. C., Merrin, C., Valenzuela, L., and Murphy, G. P. (1978). Isoenzymes of human prostatic acid phosphatase. Ann, N. Y. Acad. Sci. 390, 16–26. 107. Smith, J. K., and Whitby, L. B. (1968). The heterogeneity of prostatic acid phosphatase. Biochim. Biophys. Acta. 151, 607–618. 108. Yam, L. T. (1974). Clinical significance of the human prostatic acid phosphatase. Am. J. Med. 56, 604–616. 109. Lam, K. W., Li, C. Y., Yam, L. T., Smith, R. S., and Hacker, B. (1982). Comparison of prostatic and nonprostatic acid phosphatase. Ann. N. Y. Acad. Sci. 390, 1–15. 110. Kamoshida, S., and Tsutsumi, Y. (1990). Extraprostatic localization of prostatic acid phosphatase and prostate-specific antigen: distribution in cloacogenic glandular epithelium and sex-dependent expression in human anal gland. Hum. Pathol. 21, 1108–1111. 111. Vihko, P. (1979). Human prostatic acid phosphatase. Purification of a minor enzyme and comparisons of the enzymes. Invest. Urol. 16, 349–352. 112. Coffey, D. S. and Pienta, K. J. (1987). New concepts of study in the control of normal and cancer growth of the prostate. In, D. S. Coffey, N. Bruchovsky, W. A, Gardner, M. I. Resnick, and J. P. Karr (eds), “Current Concepts and Approaches to the Study of Prostate Cancer.” Alan R Liss, Inc., New York, p. 1–73. 113. Gutman, A. B., and Gutman, E. N. (1938). An acid phosphatase occurring in the serum of patients with metastasizing carcinoma of prostate gland. J. Clin. Invest. 17, 473–478. 114. Choe, B. K., Pontes, E. J., Dong, M. K. and Rose, N. R. (1980). Double-antibody immunoassay for human prostatic acid phosphatase. Clin. Chem. 26, 1854–1959. 115. Griffiths, J. C. (1989). Prostate-specific acid phosphatase: re-evaluation of radioimmunoassay in diagnosing prostatic disease. Clin. Chem. 26, 433–436.
180 116. Vihko, P., Kostama, A., Jänne, O., Sajanti, E. and Vihko, R. (1980). Rapid radioimmunoassay for prostate-specific acid phosphatase in human serum. Clin. Chem. 26, 1544–1547. 117. Lee, H. J., Fillers, W. S., and Iyyengar, M. R. (1988). Phosphocreatine, an intracellular high-energy compound, is found in the extracellular fluid of the seminal vesicles in mice and rats. Proc. Natl Acad. Sci. USA 85, 7265–7269. 118. Saini, M. S., and Van Etten, L. R. (1981). A clinical assay for prostatic acid phosphatase using choline phosphate as a substrate: comparison with thymolphthalein phosphate. Prostate 2, 359–368. 119. Ek, P., Malm, J., Lilja, H., Carlsson, L., and Ronquist, G. (2002). Exogenous protein kinases, A and C, but not endogenous prostasomeassociated protein kinase, phosphorylate semenogelins I and II from human semen. J. Androl. 23, 806–814. 120. Lin, M.-F., and Clinton, G. M. (1988). The epidermal growth factor receptor from prostate cells is dephosphorylated by a prostate-specific phosphotyrosyl phosphatase. Mol. Cell Biol. 8, 5477–5485. 121. Tanaka, M., Kishi, Y., Takanezawa, Y., Kakehi, Y., Aoki, J., and Arai, H. (2004). Prostatic acid phosphatase degrades lysophosphatidic acid in seminal plasma. FEBS Lett. 571, 197–204. 122. Brillard-Bourdet, M., Rehault, S., Juliano, L., Ferrer, M., Moreau, T. and Gauthier, F. (2002). Amidolytic activity of prostatic acid phosphatase on human semenogelins and semenogelin-derived synthetic substrates. Eur. J. Biochem. 269, 390–395. 123. Boissonneault, M., Chapdelaine, A., and Chevalier, S. (1995). The enhancement by pervanadate of tyrosine phosphorylation of prostatic proteins occurs through the inhibition of membrane-associated tyrosine phosphatase. Mol. Cell Biochem. 153, 139–144. 124. Guan, K., Haun, R. S., Watson, S. J., Geahlen, R. L., and Dixon, J. E. (1990). Cloning and expression of a protein-tyrosine-phosphatase. Proc. Natl Acad. Sci. USA 87, 1501–1505. 125. Chernoff, J., Schievella, A. R., Jost, C. A., Erikson, R. L., Neel, B. G. (1990). Cloning of a cDNA for a major human protein-tyrosinephosphatase. Proc. Natl Acad. Sci. USA 87, 2735–2739. 126. Meng, T. C., Lee, M. S., and Lin, M. F. (2000). Interaction between protein tyrosine phosphatase and protein tyrosine kinase is involved in androgen-promoted growth of human prostate cancer cells. Oncogene 19, 2664–2677. 127. Lin, M. F., Lee, M. S., Zhou, X. W., Andressen, J. C., Meng, T. C., Johansson, S. L., West, W. W., Taylor, R. J., Anderson, J. R., and Lin, F. F. (2001). Decreased expression of cellular prostatic acid phosphatase increases tumorigenicity of human prostate cancer cells. J. Urol. 166, 1943–1950. 128. Vihko, P. (1993). 129. Sharma, S., Pirilä, P., Kaija, H., Porvari, K., Vihko, P., and Juffer, A. H. Theoretical investigations of prostatic acid phosphatase. Proteins – structure, function and genetics (in press). 130. Lin, M. F., and Clinton, G. M. (1986). Human prostatic acid phosphatase has phosphotyrosyl protein phosphatase activity. Biochem. J. 235, 351–357. 131. Vihko, P., Wahlberg, L., Ehnholm, C., Lukka, M., and Vihko, R. (1986). Acid phosphatases bind to the main high densitylipoprotein apolipoprotein A-I. FEBS Lett. 202, 309–313. 132. Neufeld, E. B., Stonik, J. A., Demosky, S. J., Jr., Knapper, C. L., Combs, C. A., Cooney, A., Comly, M., Dwyer, N., BlanchetteMackie, J., Remaley, A. T., et al. (2004). The ABCA1 transporter modulates late endocytic trafficking: insights from the correction of the genetic defect in Tangier disease. J. Biol. Chem. 279, 15571–15578. 133. Aumüller, G., and Seitz, J. (1985). Cytochemistry and biochemistry of acid phosphatases. VI: Immunoelectron microscopic studies on
HELENA KAIJA ET AL.
134.
135. 136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
human prostatic and leucocytic acid phosphatases. Prostate 17, 31–40. Mori, K., and Wakasugi, C. (1985). Immunocytochemical demonstration of prostatic acid phosphatase: different secretion kinetics between normal, hyperplastic and neoplastic prostate. J. Urol. 133, 877–883. Lilja, H., and Abrahamsson, P. A. (1988). Three predominant proteins secreted by the human prostate gland. Prostate 12, 29–38. Sinha, A. A., Gleason, D. F., Wilson, M. J., Wick, M. R., Reddy, P. K., and Blackard, C. E. (1988). Relationship of prostatic acid phosphatase localization in human prostate by a monoclonal antibody with the Gleason grading system. Prostate 13, 1–15. Lam, K. W., Li Cy., Yam, L. T., Sun, T., Lee, G., and Ziesmer, S. (1989). Improved immunohistochemical detection of prostatic acid phosphatase by a monoclonal antibody. Prostate 15, 13–21. Zeliavinski, S., Comeau, D., and Lin, M. F. (1998). Cloning and analysis of the promoter activity of the human prostatic acid phosphatase gene. Biochem. Biophys. Res. Commun. 245, 108–112. Zeliavinski, S., Igawa, T., Lim, S., Taylor, R., and L M.-F. (2002). Identification and characterization of regulatory elements of the human prostatic acid phosphatase promoter. Oncogene 21, 3696–3705. Shan, J. D., Porvari, K., Kivinen, A., Patrikainen, L., Halmetkytö, M., Jänne, J., and Vihko, P. (2003). Tissue-specific expression of the prostatic acid phosphatase promoter constructs. Biochem. Biophys. Res. Comm. 311, 864–869. Patrikainen, L., Shan, J., Porvari, K., and Vihko, P. (1999). Identification of the DNA-binding site of a regulatory protein involved in prostate-specific and androgen receptor-dependent gene expression. Endocrinology 140, 2063–2070. Henttu, P., and Vihko, P. (1996). Effect of tumor-promoting phorbol ester on androgen-regulated gene expression in prostatic LNCaP cells. “In vitro Biology of Sex Steroid Hormone Action.” Churchill Livingstone, Tokyo, pp. 384–395. Horozewicz, J., Leong, S., Kawinski, E., Karr, J., Rosenthal, H., Chu, T., Mirand, E., and Murphy, G. (1983). LNCaP model of human prostatic carcinoma. Cancer Res. 43, 1809–1818. Lin, M. F., Garcia-Arenas, R., Chao, Y. C., Lai, M., Patel, P. C., and Xia, X. Z. (1993). Regulation of prostatic acid phosphatase expression and secretion by androgen in LNCaP human prostate carcinoma cells. Arch. Biochem. Biophys. 300, 384–390. Dulinska, J., Laidler, P., and Labrfz, M. (2002). Comparative analysis of prostatic acid phosphatase and prostate-specific antigen mRNA levels in hyperplastic prostate stimulated with steroid hormones and growth factors. Acta. Biochim. Pol. 49, 357–368. Henttu, P., Liao, S., and Vihko, P. (1992). Androgens up-regulate the human prostate-specific antigen mRNA but down-regulate the prostatic acid phosphatase mRNA in the LNCaP cell line. Endocrinology 130, 766–772. Langler, E. G., van Uffelen, C. J., Blankenstein, M. A., van Steenbrugge, G. J., and Mulder, E. (1993). Effect of culture conditions on androgen sensitivity of the human prostatic cancer cell line LNCaP. Prostate 23, 213–223. Shan, J. D., Porvari, K., Ruokonen, M., Karhu, A., Launonen, V., Hedberg, P., Oikarinen, J., and Vihko, P. (1997). Steroid-involved transcriptional regulation of human genes encoding prostatic acid phosphatase, prostate-specific antigen, and prostate-specific glandular kallikrein. Endocrinology 138, 3764–3770. Lin, M. F., Zhang, X. Q., Dean, J., and Lin, F-F. (2001). Protein kinase C pathway is involved in regulating the secretion of prostatic acid phosphatase in human prostate cancer cells. Cell Biol. Int. 25(11), 1139–1148.
Chapter 11
Matrix Proteinases Ian M. Clark Gillian Murphy
School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, UK Dept of Oncology, Cambridge University, Cambridge CB2 2XY, UK
I. Introduction II. Aspartic Proteinases III. Cysteine Proteinases
IV. Serine Proteinases V. Metalloproteinases References
I. INTRODUCTION
II. ASPARTIC PROTEINASES
Proteolysis of the extracellular matrix (ECM) of tissues is a feature of the normal remodeling associated with physiological processes such as morphogenesis, growth, and wound healing. Degradation of the ECM is also associated with many pathologies including arthritic diseases, osteoporosis, and periodontal disease. More recently, the concept of the role of proteinases in the modulation of cell–cell and cell–ECM interactions as a means of signaling to the cell and the determination of cell phenotype has emerged. This more subtle activity is clearly of importance not only in developmental processes, but also in pathology. Endopeptidases are thought to be the key enzymes in ECM degradation and in vitro evidence is available that members of all four major classes (metallo-, serine, cysteine, and aspartate at the active site) can cleave individual components of the ECM (Table IA, B). Although the emphasis on different proteinase activities varies according to the situation, the metalloproteinases appear to have the most ubiquitous role in matrix turnover. In this review, we will summarize the role of other proteinase types and describe the metalloproteinases in rather more detail.
Cathepsin D, a lysosomal enzyme [1], is the major aspartic proteinase involved in matrix degradation. It has optimal proteolytic activity at acid pH between 3 and 5, showing little activity at neutral pH. It can degrade proteoglycan core protein, gelatin, and collagen telopeptides, but not native triple helical collagen. A recent study showed that cathepsin D cleaves aggrecan within both interglobular and chondroitin sulfate domain over a pH range of 5.2–6.5 [2]. Although early reports indicated that cathepsin D might be involved in the extracellular degradation of matrix molecules, later studies have demonstrated that its role is confined to the intracellular breakdown of phagocytosed matrix molecules. For example, pepstatin, an inhibitor of cathepsin D, and anticathepsin D antiserum do not inhibit vitamin A-induced cartilage breakdown in vitro [3]; in rat osteoclasts, cathepsin D was immunolocalized in granules or vacuoles, but negligible staining was seen along resorption lacunae and in eroded bone matrix, in contrast to the staining pattern seen for other cathepsins [4].
Dynamics of Bone and Cartilage Metabolism
181
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
182
Ian M. Clark and Gillian Murphy Table IA.
Proteolysis of the Extracellular Matrix
Proteinase class range Serine Plasmin Tissue plasminogen activator Urokinase-type plasminogen activator Neutrophil elastase Proteinase 3 Cathepsin G Plasma kallikrein Tissue kallikrein Tryptase Chymase Granzymes Cysteine Cathepsin B Cathepsin L Cathepsin S Cathepsin K Calpains Aspartate Cathepsin D
Matrix substrates
Inhibitors
pH
Fibrin, fibronectin, aggrecan pro MMPs Plasminogen Plasminogen Most ECM components pro MMPs Similar to elastase Aggrecan, elastin collagen II, pro MMPs pro UPA, pro MMP1, MMP3 pro MMP8 Collagen VI, fibronectin pro MMP3 Many ECM components, pro MMP1, MMP3 Aggrecan
α2 anti-plasmin PAI-1, PAI-2 PAI-1, PAI-2, PN-1 α1-PI α1-PI, elafin α1-PI Aprotinin Kallistatin Trypstatin α1-PI
Neutral Neutral Neutral Neutral Neutral Neutral Neutral Neutral Neutral Neutral
Collagen telopeptides, collagen IX, XI, aggrecan As cathepsin B, elastin As cathepsin B, elastin Collagen telopeptides, collagen, elastin Proteoglycan, fibronectin, vitronectin
Cystatins Cystatin Cystatin Cystatin Calpastatin
5.0–6.5 4.0–6.5 6.0–6.5 Neutral Neutral
Many phagocytosed ECM components
Pepstatin
3.0–6.0
III. CYSTEINE PROTEINASES The cysteine proteinases which degrade extracellular matrix components include cathepsins B, L, S, K, and calpains I and II. Cathepsins B, L, and S are lysosomal enzymes with acid pH optima; ultrastructural studies using the electron microscope have demonstrated their involvement in intracellular degradation of phagocytosed matrix molecules. However, they may also function as extracellular proteinases at pH closer to neutral. The lysosomal active form of cathepsin B is unstable at neutral pH, but both pro-cathepsin B and a higher Mr active form (which may be a noncovalent complex between cleaved propeptide and active enzyme) are secreted extracellularly from stimulated connective tissue cells and activated macrophages; these are more stable and possibly active at near neutral pH. A study of cathepsin B activity in synovial fluid demonstrated a higher level of activity in RA fluid than OA fluid, with 42 kDa pro-enzyme and 29 kDa active enzyme being present [5]. These data were recently replicated with the demonstration that the majority of cathepsin activity in synovial fluid was from cathepsin B [6]. Human osteoblasts are reported to secrete a form of cathepsin B which is stable at neutral pH and can be induced with IL-1 and PTH [7]. At acid pH in vitro, cathepsins B, L, and S cleave the telopeptides of types I and II collagen, causing depolymerization; they also cleave proteoglycan core protein, gelatin, and nonhelical regions of types IX and XI collagen [8, 9]. Cathepsins L and S are both potently elastolytic
and, at neutral pH, cathepsin S has elastolytic activity comparable with that of neutrophil elastase [9]. Cathepsin S is stable over a broader pH range than cathepsin B or L. If cathepsins B and L are added to cartilaginous matrix at near neutral pH, then proteoglycan and collagen are released [10]. Cathepsin B has been shown to be the major cysteine proteinase activity in OA chondrocytes and cartilage [11]. Indeed, cathepsin B has been shown to be increased in lysosomes of OA chondrocytes compared to normal, and also extracellularly at sites of cartilage degradation in OA [12]. A lipophilic cathepsin B inhibitor can block proteoglycan release induced by interleukin 1 in an explant model and this can also be blocked by MMP inhibitors, perhaps revealing a role for the cathepsin in the activation of proMMP [13]. In endochondral bone formation, it has been proposed that the complete degradation of type X collagen requires both the action of MMP-13 from the chondrocyte and then cathepsin B from the osteoclast [14]. In bone resorption by osteoclasts, studies with specific inhibitors in a murine calvarial assay suggest that cathepsins L and S might act both intracellularly and extracellularly, but that cathepsin B only acts in the intracellular compartment, possibly via activation of other proteinases [15]. In rat osteoclasts, cathepsin B and L immunostained along lacunae with strong extracellular staining of both on collagen fibrils and the bone matrix under ruffled border, with a stronger signal for cathepsin L than cathepsin B [4]. A further ultrastructural study using specific
183
Chapter 11 Matrix Proteinases Table IB. Proteolysis of the Extracellular Matrix MMP No
Mr enzyme
Mr latent
MMP-1
Interstitial collagenase-1
55 000
45 000
MMP-8
Neutrophil collagenase
75 000
58 000
MMP-13
Collagenase-3
60 000
48 000
MMP-18 MMP-3
Xenopus collagenase Stromelysin-1
55 000 57 000
? 45 000
MMP-10
Stromelysin-2
57 000
44 000
MMP-11 MMP-19
Stromelysin-3 RASI
51 000 ?
44 000 ?
MMP-20 MMP-2
Enamelysin gelatinase A
72 000
66 000
MMP-9
gelatinase B
92000
86 000
MMP-7
Matrilysin (PUMP-1)
28 000
19 000
MMP-26
Matrilysin 2
MMP-12
Macrophage
54 00
45 000/22 000
MMP-14
MT-1 MMP-66000
MMP-15
MT-2 MMP
72 000
MMP-16
MT-3MMP
64 000
MMP-17 MMP-24 MMP-25 MMP23A MMP23B MMP27 MMP28
MT-4MMPMT5 MMP MT6 MMP
Epilysin
Active
56 000
52 000
Known substrates Collagens I, II, III, VII, VIII, X; gelatin: aggrecan, versican, proteoglycan link protein; laminin; perlecan; nidogen, α1-proteinase inhibitor, α2- macroglobulin (α2M); fibrin; fibrinogen; pregnancy zone protein; ovostatin; myelin basic protein (MBP); IGFBP; proTNF; CXCL12; proIL1β; stromal cell-derived factor, SDF; L-Selectin; proMMP-2; proMMP-9 Collagens I, II, III, V, VII, VIII, X; gelatin; aggrecan; brevican; α2-M; α1-proteinase inhibitor; α2-antiplasmin; fibronectin; CXCL5; IGFBP; ADAM-TS-1; C1q Collagens I, II, III, IV, VI, IX, X, XIV; gelatin; aggrecan; perlecan; brevican; fibrillin; fibronectin tenascin; α2-M; plasminogen activator inhibitor-2; CXCL12; SDF; proMMP-9 Collagen; gelatin Collagens III, IV, V, VII, IX, X, XI; gelatin; aggrecan; versican, perlecan; proteoglycan link protein; fibronectin; laminin; elastin; nidogen; decorin; osteonectin, osteopontin; tenascin; vitronectin; fibrin; fibrinogen; MBP; L-selectin; E-cadherin; antithrombin-III; proteoglycan link protein; α2-M ovostatin; α1- proteinase inhibitor; proHB-EGF; proTNF; pro MMP-1; proMMP-7; proMMP-8; proMMP-9; proMMP-13; uPA Collagens III, IV, V; gelatin; casein; aggrecan; brevican; elastin; proteoglycan link protein; fibronectin; fibrinogen; α2-M; proMMP-1; proMMP-8; proMMP-9 α1-proteinase inhibitor; IGFBP; α2-M Collagens I, IV; gelatin, aggrecan, COMP, nidogen; fibronectin; laminin; tenascin Amelogenin; aggrecan; COMP; Collagens I, IV, V, VII, X, XI, XIV; gelatin; elastin; fibronectin; aggrecan; versican; proteoglycan link protein; laminin; nidogen; fibulins; vitronectin; osteonectin; fibrillin; MBP; α2-M; α1-proteinase inhibitor; C1q; CCL7; CXCL12; proTNF; latent TGFβ; FGFR1; proIL1β; proMMP-9; proMMP-13; ADAM TS1; plasminogen Collagens IV, V, VII, X, XIV; gelatin; elastin; aggrecan; versican; decorin; nidogen; proteoglycan link protein; fibronectin; vitronectin; osteonectin; MBP; α2-M; α1-proteinase inhibitor; C1q; CXCL1; CXCL4; CXCL7; CXCL 12; endothelin, galectin-3; IL8; proTNF; IL1β; IL-2Rα; latent TGFβ; plasminogen Collagens IV, X; gelatin; aggrecan; brevican; decorin; elastin; fibulins; osteonectin; osteopontin; tenascin; vitronectin; proteoglycan link protein; fibronectin; laminin; nidogen; MBP; fibrinogen; E-cadherin; FASligand; α1-proteinase inhibitor; proHB-EGF; proTNF; β4integrin; pro MMP-1; pro MMP-2; proMMP-9 Collagen IV; gelatin; fibronectin; fibrinogen; vitronectin; α1-proteinase inhibitor; proMMP9 Collagen IV; gelatin; aggrecan; elastin; fibronectin; fibrillin; nidogen; metalloelastase vitronectin; laminin; MBP; α1-proteinase inhibitor; proTNF; plasminogen Collagens I, II, III gelatin; aggrecan; nidogen; fibrillin; perlecan; fibrinogen; elastin; fibronectin; laminin B chain; vitronectin; tenascin; α2-M; α1-proteinase inhibitor; CD44; CXCL12; αVintegrin; tissue transglutaminase; proMMP-2; pro MMP-13; proTNFα Aggrecan, nidogen; fibronectin; laminin; perlecan; tenascin; transglutaminase; proMMP-2; ADAM TS Collagen III; gelatin; fibronectin; laminin; perlecan; tenascin; tissue transglutaminase; proMMP-2; ADAM TS1 Gelatin; fibrin; fibrinogen; tissue transglutaminase; proMMP2; proTNFα Gelatin; fibronectin; proMMP2 Collagen IV; gelatin; fibronectin; fibrinogen; fibrin; proMMP2
184
Ian M. Clark and Gillian Murphy
inhibitors also suggested that cathepsin L played a key role in bone resorption by cultured osteoclasts [16]. More recently a new cysteine proteinase, cathepsin K (previously called cathepsin O or O2), has been discovered in osteoclasts. Cathepsin K has 56% sequence homology with cathepsin S; it has a pH optimum of 6–6.5 similar to cathepsin S, but has activity over a broader range, remaining largely active at neutral pH. It degrades elastin, collagen telopeptides, and gelatin better than the other cysteine proteinase cathepsins [9, 17]. Aside from its collagen telopeptidase activity, cathepsin K also cleaves both types I and II collagen within the triple helix [18, 19]. Cathepsin K is stabilized by interactions with chondroitin sulfates [20] and requires complex formation with chondroitin sulfate for collagenase activity [21, 22]. Inhibitors of cathepsin K have been shown to reduce bone resorption in vivo or in vitro [23] and an antisense approach blocks osteoclastic bone resorption in vitro [24]. It has been suggested that only cathepsin K, and not cathepsins B, L, or S, is abundantly expressed in human osteoclasts, and that this enzyme is responsible for osteoclastic bone resorption [25]. Interestingly, pycnodyostosis, an osteochondrodysplasia is caused by mutation in the cathepsin K gene [26]. Cathepsin K is also expressed in synovial fibroblasts and chondrocytes and is raised in both OA and RA [27–29]. A family of cysteine proteinase inhibitors, the cystatins, has also been described [30]. Their role in controlling the turnover of cartilage and bone is largely unknown; however, cystatin C is reported to inhibit bone resorption in different in vitro systems due to inhibition of osteoclastic proteolytic enzymes and to be produced by bone cells [31]. Calpains are cytosolic, Ca2+-dependent, cysteine proteinases with neutral pH optima. Calpains 1 and 2 require micromolar and millimolar Ca2+ respectively for activation. These enzymes can degrade proteoglycan, fibronectin, and vitronectin [32]. Calpains have been shown to be raised in RA and OA synovial fluid and tissues [33, 34] and in some animal models of arthritis [35]. Recently, Calpain 2 has been shown to cleave aggrecan at a distinct C-terminal site consistent with the presence of such cleaved species in cartilage [36]. Calpastatin, a specific cellular inhibitor of calpains has been identified as an autoantigen in some RA patients [37].
IV. SERINE PROTEINASES A. Leukocyte Enzymes Neutrophil elastase, cathepsin G, and proteinase 3 are found in the azurophil granules of polymorphonuclear leukocytes. They are synthesized as precursors in
promyelocytes of bone marrow during development [38]. Granule contents may be discharged into the extracellular space during inflammation where they have the opportunity to degrade matrix. These enzymes have pH optima between 7 and 9 and can degrade many components of the matrix. Neutrophil elastase and proteinase 3 are both potently elastolytic; neutrophil elastase and cathepsin G degrade the telopeptides of fibrillar collagens and type IV collagen; all three enzymes can degrade other matrix glycoproteins, gelatin, and proteoglycan core protein [39–41]. Human neutrophil elastase has also been shown to cleave within the collagen triple helix [42]. Proteinase 3 is an autoantigen in Wegener’s granulomatosis [43]. The role of these enzymes in cartilage degradation may only be in specialized situations where they can escape inhibition by synovial fluid inhibitors (see below), for example during joint sepsis. Beige mice lack both PMN elastase and cathepsin G, but still undergo severe cartilage loss during antigen-induced arthritis [44]; rabbits depleted of neutrophils by nitrogen mustard show the same rate of cartilage degradation in arthritis models as the nontreated animals [45]. A number of inhibitors of these enzymes exist, including the general proteinase inhibitor α2-macroglobulin. α1 proteinase inhibitor is the major specific serum inhibitor of PMN elastase and also inhibits proteinase 3 [46]; α1 antichymotrypsin and secreted leukocyte proteinase inhibitor (SLPI) inhibit both PMN elastase and cathepsin G; SLPI has been isolated from cartilage [47]. Thrombospondin 1, a matrix protein, is also a potent inhibitor of PMN elastase and cathepsin G; this molecule is found to be induced in RA synovium proportionally to the numbers of surrounding leukocytes [48]. Cross-talk may exist between proteinase families via action on proteinase inhibitors, e.g. serpins may be inactivated by MMPs and neutrophil elastase may inactivate TIMPs [49, 50].
B. Plasminogen Activators and Plasmin The conversion of plasminogen (Pg, 90 kDa) to plasmin by plasminogen activators (PA) is a key step in connective tissue remodeling since plasmin can degrade a number of matrix macromolecules (aside from fibrin, plasmin can degrade the core protein of proteoglycan, fibronectin, laminin, and type IV collagen) [51] and also take part in the activation of proMMPs (e.g. proMMP-1 and -3). There are two types of PA, tissue-type (tPA) and urokinase-type (uPA). tPA (70 kDa) is mainly associated with endothelial cells, but is also produced by fibroblasts and chondrocytes; it also binds to extracellular matrix components
185
Chapter 11 Matrix Proteinases
such as laminin, fibronectin, and thrombospondin [52]. tPA is the major activator of Pg in fibrinolysis. uPA (54 kDa, two subunits of 30 kDa and 24 kDa which are disulfide bonded) is produced by many connective tissue cells. It binds a specific cell surface receptor (uPAR) via a growth factor domain in uPA, and this enhances activation of cellassociated Pg and protects the uPA from inhibition; uPA may also have mitogenic effects via this receptor [53]. Inhibitors of plasmin and PA come from the serpin superfamily [46]. α2-antiplasmin is the major plasmin inhibitor in plasma. α1-proteinase inhibitor and α2-antichymotrypsin are also active against plasmin [54]. PA inhibitors (PAIs) control the rate of plasmin production in the local extracellular environment. PAI-1, originally isolated from endothelial cells, is produced by a number of cell types and is present in platelets; it inhibits both tPA and uPA, but forms more stable complexes with the former [55]. PAI-1 binds to the pericellular matrix which enhances its extracellular stability and therefore may inhibit proteolysis in this environment. PAI-2 is a better inhibitor of uPA than tPA; it is present in plasma and is expressed by a number of cell types including macrophages [56]. PAI-3 is also an inhibitor of activated protein C, and is less well characterized [57]. Finally, protease nexin 1 (PN-1) inhibits uPA and tPA with the PN–1 proteinase complex binding a specific cell surface receptor via the inhibitor, followed by internalization and degradation [58]. In synovial fluid, uPA, PAI-1, and uPAR are all raised in RA compared to OA, and in OA compared to normal, though there was no direct association with clinical parameters [59]. There is also evidence for local production of uPA and PAI-1, with synovial fluid levels raised compared to plasma. At the invasive pannus front in RA synovium, uPA, uPAR, PAI-1, and PAI-2 are all induced compared to normal tissue [60, 61], whilst tPA is repressed [61]. In human OA and animal models of OA where enhanced bone remodeling may trigger cartilage damage, the uPA/plasmin system is up-regulated [62]. Plasminogen added to chondrocyte cultures causes an increase in interleukin-1-induced matrix degradation, implicating PA in this process; indeed, IL-1 induces uPA and PAI-1 in these cells. Similarly, plasminogen added to IL-1-stimulated cartilage explants potentiates increased collagen and proteoglycan loss, and this can be blocked with PA inhibitors or MMP inhibitors. Hence, there may be a cascade of enzyme activation with PA activating plasminogen, and plasmin activating MMPs [63–65]. Furthermore, intra-articular injection of protease nexin-1 blocks IL-1 or FGF-induced proteoglycan loss from rabbit knee and tranexamic acid (TEA), another antiplasmin agent, used orally, has the same effect [66]. However, a
recent pilot-scale double-blind placebo-controlled trial of TEA in RA patients showed no therapeutic efficacy [67]. In a murine model of mild arthritis (antigen-induced arthritis), uPA- or tPA-null mice have increased disease severity [68] whilst PAI-1 null mice have decreased disease severity [69]; disease severity also correlates with fibrin deposition. However, in a systemic and more aggressive model (collagen-induced arthritis), uPA-null mice develop less severe disease whilst tPA-null mice develop more severe disease. Disease severity still correlates with fibrin deposition, with the uPA-null mice also having defects in T-cell responses [70]. Osteoblasts also produce uPA, tPA, PAI-1, and uPAR, and their synthesis is regulated by osteotropic hormones [71, 72]; it has been postulated that they function as activators of proMMPs or latent growth factors in bone remodeling, or that uPA may be acting as a mitogen via its growth factor domain. Similarly, the direct resorptive activity of osteoclasts does not require tPA or uPA, but the latter may be involved in migration of the pre-osteoclasts to the mineral surface [73]. It has been proposed that this enzyme system is key to the initiation of bone resorption [74]. In mice lacking plasminogen activators increased bone formation is observed [75].
C. Mast Cell Proteinases Mast cell granules contain tryptase, chymase, and cathepsin G. Tryptase has a trypsin-like specificity and degrades collagen type VI and fibronectin. Chymase has a chymotrypsin-like specificity and degrades a number of matrix substrates. Both enzymes have been implicated in the activation of proMMPs [76]. Activated, degranulated mast cells have been localized to regions of cartilage erosion in RA [77].
D. Granzymes Granzymes A and B are serine proteinases which are stored in the granules of activated cytotoxic T-cells and NK cells. Granzyme B has been reported to degrade aggrecan in the extracellular matrix synthesized by chondrocytes [78, 79] and recently this enzyme has been reported to be expressed by chondrocytes themselves [80]. Granzymes A- and B-expressing lymphocytes have been localized in the synovium of patients with RA and OA, and may be increased in RA tissue [81]. Granzymes A and B are elevated in the plasma and synovial fluid of patients with RA compared to OA or reactive arthritis [82].
186
Ian M. Clark and Gillian Murphy
V. METALLOPROTEINASES At least three classes of zinc-dependent endopeptidase activity have been implicated in the degradation of cellsurface molecules and the ECM [83]: the matrix metalloproteinase (MMP) family, the reprolysin-related proteinases (ADAMs or MDCs) that contain both a metalloproteinase and a disintegrin domain, and the astacin-related proteinases. The astacins discovered to date include the enzyme that cleaves the carboxy-terminal propeptides of the fibrillar collagens during collagen assembly (bone morphogenetic protein-1; [84]) and others involved in cell differentiation and pattern formation during development. Recent studies have provided direct morphological and functional links between its proteolytic activity and the ECM [85]. The ADAM family and a related ADAMTS family are discussed in more detail below.
A. ADAM Family The ADAMs (a disintegrin and metalloproteinase) are a family of 21 human proteinases involved in numerous physiological and developmental processes. They each contain an amino-terminal signal sequence, a pro-domain, a metalloproteinase domain, a disintegrin domain with homology to the snake venom disintegrins, a cysteinerich domain, an EGF-like region, and transmembrane and cytoplasmic domains. Not all of the ADAMs contain a complete metalloproteinase active site consensus, suggesting that only 13 are catalytically active. The ADAMs were initially described as having roles in fertilization, but have now been implicated in cell migration, differentiation, activation of signaling pathways, and shedding of cell-surface molecules [86]. Levels of soluble TNFα are significantly elevated in RA, largely due to the action of ADAM-17, TNFα-converting enzyme (TACE). Expression of TACE is predominantly localized to the synovium in RA [87]. Both TNFα and TACE can also be expressed in cartilage chondrocytes [87–89]. There have been several reports of mRNA for ADAM 9, ADAM 10, ADAM 12, and ADAM 15 being expressed in chondrocytes, with ADAM 10 and ADAM 15 being up-regulated in OA cartilage compared to normal [90–94]. Inoue et al. [95] have shown that ADAMs 11, 12, 15, and 19 are expressed by osteoblasts and ADAMs 11 and 15 by osteoclasts. Other studies suggest a function for ADAM 10 and ADAM 12S in the formation of osteoclasts from human peripheral blood mononuclear cells and subsequent resorption activity [96]. These proteins have proposed functions in the proteolytic processing of cell-surface
ectodomains, in membrane fusion events, and in cell–cell and cell–matrix interactions [97].
B. ADAMTS Family The ADAMTS family are a recently described group of secreted proteinases which, like ADAMs, contain a signal sequence, pro-domain, metalloproteinase domain and disintegrin-like region, but this is followed by a thrombospondin-1 repeat, then a cysteine-rich domain, one or more additional thrombospondin-1 repeats and in some cases an additional C-terminal domain often containing a PLAC motif [98]. There are currently 19 human family members including six enzymes reported to degrade aggrecan (ADAMTS 1, ADAMTS 4, ADAMTS 5, ADAMTS 8, ADAMTS 9, ADAMTS 15) and these are termed aggrecanases [98, 99]. Cleavage of aggrecan in the G1–G2 interglobular domain is believed to be a key event in aggrecan degradation and this region contains cleavage sites for both MMPs (Asn341-Phe342) and aggrecanases (Glu373–Ala374). The major aggrecan fragments released from resorbing cartilage are not products of MMP action [100]. Using neoepitope antibodies to study cleavage products of both enzyme families in cartilage, it appears that aggrecanases may be active early in the arthritic disease process with subsequent increases in MMP activity [101–103]. ADAMTS 4 and ADAMTS 5 also cleave the chondroitin sulfate proteoglycans brevican and versican [104, 105]. The most studied aggrecanase enzymes are ADAMTS 4 and ADAMTS 5. Western blot analysis has demonstrated expression of ADAMTS 4 and ADAMTS 5 in normal bovine cartilage explants, human osteoarthritic cartilage, and rheumatoid, osteoarthritic and normal synovium. In cartilage explant cultures aggrecanase activity can be induced by the addition of catabolic cytokines. Levels of ADAMTS 4 are induced in cartilage in response to IL-1 and TNFα and in fibroblast-like synoviocytes in response to TGFβ [106–109]. ADAMTS5 expression tends to be less responsive to stimulation, though this is dependent on cell type [107, 109]. Immunodepletion of ADAMTS4 from bovine articular cartilage cultures following IL-1 stimulation resulted in a 75% reduction of aggrecanase activity in the culture medium, demonstrating that ADAMTS4 is the major aggrecanase in this model, presuming the antibody is monospecific [106]. In addition to cleavage at the Glu373–Ala374 bond, at least ADAMTS4 and ADAMTS5 cleave aggrecan at four other sites within the chondroitin sulfate-rich region of the core protein between G2 and G3. Cleavage of these sites in vitro appears to be more efficient than cleavage
Chapter 11 Matrix Proteinases
within the G1–G2 IGD [106, 110, 111]. Using recombinant ADAMTS4, the full-length active form of the enzyme was an effective aggrecanase, but exhibited little activity against the Glu373–Ala374 bond, preferentially cleaving at Glu1480–Gly1481. However, deletion of the spacer region led to a reduction in aggrecanase activity, but a markedly increased cleavage of Glu373–Ala374; this C-terminally truncated form of the enzyme also diminishes binding to extracellular matrix and may be the primary form found in, for example, interleukin-1-stimulated cartilage [112]. In the cell, C-terminal processing of ADAMTS4 appears to be mediated by MT4-MMP [113]. The presence of the TSP1 domain in ADAMTS4 and the GAG substitutions in aggrecan are also important for substrate recognition [111, 114]. Despite a growing understanding of the biochemistry and expression of the known aggrecanases, the relative contribution of each enzyme in cartilage destruction remains to be determined.
C. Matrix Metalloproteinases The MMPs are key ECM-degrading proteinases and have been divided into subgroups according to their structure and function. To date 23 individual human enzymes have been described (Table IB). The MMPs have a common domain structure with a signal peptide, a propeptide, a catalytic domain, a hinge region and a hemopexinlike C-terminal domain. Individual MMPs have variations on this general structure: the MT-MMPs have a transmembrane domain and cytoplasmic tail (MMP14–16 and 24) at the C-terminal end or are GPI-anchored (MMP17 and 25) and, in common with MMP11, 21, 23, and 28, have a sequence containing a potential furin-cleavage site at the N-terminal end of the catalytic domain. Matrilysins (MMP7 and 26) lack the C-terminal domain and hinge region, the gelatinases A and B (MMP2 and 9) have an insert of three fibronectin type II repeats in the catalytic domain; and MMP9 has a collagen-like sequence at the C-terminal end of the catalytic domain. MMP23 occurs in two forms, A and B, encoded by separate genes. These are type II transmembrane proteins containing a unique cysteine array and an immunoglobulin-like domain at the C-terminus [115]. The collagenases-1, -2, and -3 (MMP1, MMP8, and MMP13, respectively) have the specific ability to degrade fibrillar collagens within the native triple helix. They also have limited ability to cleave other proteins, MMP13 being the most active, with significant activity against type IV collagen and denatured collagens (gelatins). The catalytic domains of collagenases alone retain proteolytic activity
187 but they do not digest native triple helical collagen. MMP2 and MMP9 actively degrade denatured collagens and MMP2 has low but significant activity on soluble native fibrillar collagens. They also degrade soluble type IV collagen but activity on more native preparations is extremely weak [116]. Both enzymes degrade type V collagen, elastin, aggrecan, and other matrix proteins to a limited extent. Some of the MT MMPs have collagenolytic activity (see below). The stromelysin subgroup, MMP3 and MMP10 cleave a number of matrix components, notably aggrecan, fibronectin, laminin, and type IV collagen. MMP3 has limited activity against collagens III, IX, and X and the telopeptides of collagens I, II, and XI [117]. Stromelysin 3, MMP11, has some of the substrate specificity of the other stromelysins but its activity is so weak that its role as a proteinase may be against a specific, as yet unidentified, substrate. Matrilysin, MMP7, and metalloelastase, MMP12, are also somewhat similar to the stromelysins, but are rather more active. The more recently discovered MT MMPs, MT 1–3 (MMP14–16), are also known to cleave a number of ECM proteins, including fibrillar collagens. The other newer MMPs also degrade some matrix proteins (Table IB). The catalytic domains of the MMPs are spherical or ellipsoidal and have an active site cleft containing a catalytic zinc ion at the bottom. The structure is stabilized by a second zinc ion and a calcium ion. Crystal structures have been obtained for the catalytic domains of human MMP1 [118–120], human MMP3 [121], human MMP8 [122–124], and human MMP9 [125] complexed with synthetic inhibitors. The three-dimensional structures of the catalytic domain of several human MMPs have also been determined by NMR and used in the design of lowmolecular-weight inhibitors [126]. Some of the differences in substrate specificity can be accounted for by variations in the six specificity subsites in the active site cleft and in sequences around the entrance to the cleft. The S1′ pocket (the first specificity subsite on the carboxy-terminal side of the substrate scissile bond) is deeper in MMP3 [121, 127] and larger in MMP8 [122, 124] than in MMP1. The type II fibronectin-like repeats that form the gelatin-binding domain of MMP2 occur as a three pronged “fish hook” to the side of the active site cleft, conferring specificity on the catalytic domain [128]. The hinge region varies in length between the different MMPs. The crystal structure of porcine MMP1 showed the hinge to be highly exposed and with no secondary structure [129]. In the collagenases this region is prolinerich, and has been suggested to mimic the triple helix of collagen and to contribute to the binding of MMP8 to type I collagen [130]. It has been shown that collagenases
188 bind to and locally unwind triple helical collagens. MMP1 preferentially interacts with the α2(I) chain of type I collagen, by both catalytic domain and hemopexin domain interactions and cleaves the three α chains in succession and the flexible hinge may be critical to this mechanism, coordinating the binding of the domains to the collagen [131]. The crystal structures of full-length porcine MMP1 [129], the C-terminal domain of human MMP13 [132], and human proMMP2 [128, 133] showed their hemopexinlike domains to comprise four similar units that form a four-bladed β-propeller structure. This propeller is stabilized by a disulfide bridge between the first and fourth units and has cations, postulated to be calcium, and chloride ions at the core. The whole structure has a disk-like shape with a funnel-like tunnel running through the short axis [132, 133]. On MMP2 there is a surface patch of positive residues extending from the exit side of the tunnel across propeller blade III that has been proposed as an anchoring point for TIMP-2 [133] and this was confirmed by the proMMP2-TIMP-2 crystal structure [134]. The hemopexin-like domain contributes towards the specificity of the collagenases. Studies on the isolated N-terminal domain of MMP1 [135], on truncated mutants of human collagenases and on collagenase/stromelysin-1 chimeric enzymes, have shown that the ability of collagenase-1 [136] and MMP8 [137] to cleave native type I collagen is influenced by the C-terminal domain. This domain is also involved in the binding of active MMPs to TIMPs [135, 136, 138, 139] and in the binding of latent and active MMP3 [136] and MMP8 [140] to extracellular matrix molecules. The C-terminal domains of MMPs 2 and 9 allow them to form complexes with certain of the TIMPs when latent. This is the basis for the cellular activation of proMMP2 by MT-1 MMP, involving the formation of a multiprotein cell-surface cluster where TIMP-2 tethers proMMP2 by its hemopexin domain to MT1 MMP, facilitating proteolysis of the propeptide by free MT1 MMP and active MMP2 [141]. The genes encoding for example MMP1, MMP3, MMP7, MMP9, and MMP10, have a TATA box and a TPA-responsive element (TRE), a sequence that binds to Fos and Jun (AP1), which is thought to mediate the induction of these MMPs by inflammatory cytokines such as IL-1 and TNF-α. MMP2 appears to be constitutively produced by cells in culture and is not generally significantly modulated by cytokines. The 5′-flanking region has no TATA box or identifiable TRE sequence, but does contain a p53-sensitive sequence and an adenovirus ElAresponsive element resembling the AP2-binding site and two “silencer” regions [142]. Glucocorticoids, retinoids, and TGFβ repress the expression of at least MMP1 and MMP3 in connective tissue fibroblasts [143, 144].
Ian M. Clark and Gillian Murphy
Agents may induce the expression of one MMP, whilst repressing expression of another. Furthermore, the effect of any one agent may differ with cell type; this makes it difficult to predict the effects of blocking cytokine/growth factor action. Table II shows some of the factors which modulate expression of the MMP-1 gene. The extracellular regulation of MMP activity is a key feature of their function. The regulation of zymogen activation is possibly the most critical level of control of enzyme activity. The MMPs have latency conferred on them by their propeptide domain, which effectively blocks the active site in the catalytic domain. In the latent enzyme, a cysteine residue in the N-terminal propeptide PRCGVPDV sequence co-ordinates with the catalytic zinc ion, preventing water from occupying this site. In the active enzyme, water co-ordinated with this zinc ion is important in the mechanism of peptide bond hydrolysis. Upon enzymic cleavage in the prodomain the Zn2+–cysteine complex becomes destabilized, and a conformational
Table II. Gene MMP-1
TIMP-1
TIMP-2 TIMP-3
Factors that Modulate Expression of MMP and TIMP Genes Inducing agent
Interleukin-1 Tumor necrosis factor α Epidermal growth factor Basic fibroblast growth factor Platelet-derived growth factor Lipopolysaccharide Phorbol ester Colchicine Cytochalasin B Concanavalin A Calcium pyrophosphate Calcium ionophore A23187 UV irradiation Transforming growth factor β Interleukin-6 Interleukin-11 Oncostatin M Leukemia inhibitory factor Basic fibroblast growth factor Epidermal growth factor Interleukin-1 Progesterone Oestrogen Retinoic acid Phorbol ester Lipopolysaccharide Progesterone Transforming growth factor β Epidermal growth factor Platelet-derived growth factor Phorbol ester Dexamethasone
Repressing agent Transforming growth factor β Retinoic acid Estrogen Progesterone Interferon γ Dexamethasone
Dexamethasone Concanavalin A
Transforming growth factor β Lipopolysaccharide Oncostatin M
189
Chapter 11 Matrix Proteinases
change results in autoproteolysis of the entire propeptide [145]. Most of the MMPs are secreted in their latent forms and activation takes place outside the cell. The MMPs with a furin cleavage site in the propeptide (MMP11, 21, 23, 28) are exceptions to this, and are activated intracellularly by furin, or other proprotein convertases, prior to secretion [146]. The MT-MMPs have a similar furincleavage site (RXKR) and are also likely to be activated before leaving the cell. In vitro the MMPs can be activated by chemical agents that modify cysteine residues, such as the organomercurial 4-aminophenymercuric acetate (APMA). Activation may also occur when the conformation of the proenzyme is perturbed, as is evident from the SDS-activation of proMMP2 and proMMP9 during zymography. In vivo, however, activation probably takes place upon the enzymic cleavage of the propeptide. This cleavage may be an autolytic step, as occurs with proMMP2 at high concentrations in vitro [147], but more usually requires a second proteolytic activity. As can be seen from Table III, many of the MMPs share common activators and many are activated by other MMPs. This has given rise to speculation that the MMPs are involved in activation cascades [148]. An important cascade is initiated by the action of plasminogen activator (tPA or uPA) on plasminogen, which results in the active enzyme plasmin. This enzyme can activate several MMPs, as depicted in Figure 1. MMP2 is not involved in the plasmin cascade, and appears to be regulated differently to the plasmin-activated MMPs. It is not cleaved by trypsin, chymotrypsin, plasmin, thrombin, Table III. Human MMP
MMP No.
Collagenase-1 Collagenase-2 Collagenase-3 Stromelysins-1, -2 Stromelysin-3 Matrilysin Metalloelastase Gelatinase A
1 8 13 3,10 11 7 12 2
Gelatinase B MT1-MMP MT2-MMP MT3-MMP MT4-MMP MT5-MMP MT6-MMP MMP-19 MMP-20 MMP-23 MMP-26 MMP-28
9 14 15 16 17 24 25 19 20
elastase, cathepsin G, plasma kallikrein, or MMP3 [149, 150]. The mechanism involving MT MMPs described above appears to be the most likely pathway for MMP2 activation and may lead to the activation of MMP13 and MMP9 [151].
D. Tissue Inhibitors of Metalloproteinases There is a group of specific inhibitors of the MMPs called the TIMPs (tissue inhibitors of metalloproteinases; [152]), Four TIMPs have been described to date; TIMP-1 [153], a 28-kDa glycosylated protein; TIMP-2 [154], a 22-kDa nonglycosylated protein; TIMP-3 [155–161], a 21–27-kDa protein that binds to the extracellular matrix [144,148]; and TIMP-4 [162, 163], the human form of which was cloned from a cDNA library and expressed as a 22-kDa recombinant protein [162]. TIMP-4 is more closely related to TIMPs-2 and -3 than to TIMP-1 [163]. The TIMPs have 12 conserved cysteines, giving a protein structure of six loops which can be divided into two subdomains [164]. TIMP-1 has the shape of an elongated wedge in which the long edge, consisting of five different chains, occupies the active site cleft of MMP3 in the complex and are responsible for around 75% of the interactions between TIMP and MMP3. The central disulfide-linked segments Cysl–Val4 and Ser68–Val69 bind either side of the MMP catalytic zinc, with ligation of the zinc by the carbonyl oxygen and α amino groups of Cys1 [115]. Although they are able to inhibit most
Activation of MMPs
Exogenous activators Plasmin, kallikrein, chymase, MMP-3 Plasmin, MMP-3, MMP-10 Plasmin, MMP-2, MMP-3, MMP-14 Plasmin, kallikrein, chymase, tryptase, elastase, cathepsin G Furin Plasmin, MMP-3 Plasmin MMP-1, MMP-2, MMP-7, MMP-9, MMP-13, MMP-14, MMP-15, MMP-16, MMP-17, MMP-24, MMP-25, elastase, cathepsin G Plasmin, neutrophil elastase, MMP-3, MMP-2, MMP-13, MMP-26 Furin Probably a proprotein convertase Probably a proprotein convertase Probably a proprotein convertase Probably a proprotein convertase Probably a proprotein convertase Not known MMP-14 Probably a proprotein convertase MMP-26 Probably a proprotein convertase
Activating MMP-2 Not known MMP-2, MMP-9 MMP-1, MMP-8, MMP-9, MMP-13 Not known MMP-1, MMP-2, MMP-3, MMP-9 Not known MMP-1, MMP-2, MMP-9, MMP-13 Not known MMP-2, MMP-13 MMP-2 MMP-2 ADAMTS-4 (C-terminal processing) MMP-2 MMP-2 Not known Not known Not known MMP-9 Not known
190
Ian M. Clark and Gillian Murphy
Figure 1
A scheme for the cellular matrix metalloproteinase activation cascades that occur pericellularly. The activation of proMMPs is largely limited to the pericellular environment where cell-associated proteinases can function in a privileged environment away from an excess of proteinase inhibitors. Key initiators of the MMP activation cascades are thought to be MT-MMPs and plasmin-mediated proteolysis. The generation of partially active or active MMPs allows a cascade of cleavages to generate fully active enzymes. The efficiency of these interactions is dependent upon mechanisms for the concentration of MMPs at the cell surface or on the extracellular matrix. (Courtesy of Dr V. Knäuper, Dept. of Biology, University of York, UK.)
active MMPs, TIMP-1 binds very poorly to many MT MMPs and MMP 19 [152]. In addition to their ability to bind to active MMPs the TIMPs can bind to certain latent MMPs. TIMP-1 and -3 bind to pro-MMP9 [165, 166] and TIMPs-2, -3, and -4 bind to pro-MMP2 [138, 167–172]. The C-terminal loops of TIMP-1 are involved in the binding of this inhibitor to proMMP9 [170]. Similarly, the C-terminal domain of TIMP-2 is essential for binding to pro-MMP2 which occurs through interactions between a negatively charged “tail” on TIMP-2 [172] and a positively charged area on the hemopexin domain of MMP2 [134]. Since pro-MMP2 complexed to TIMP-2 is less prone to autoactivation, it was suggested that TIMP-2 acts to stabilize the proenzyme [173–175]; however, more recent work has implicated TIMP-2 involvement in the activation of pro-MMP2 by cellular MT1 MMP [141, 176]. Pro-MMP9 complexed to TIMP-1 resists activation by MMP3 [166] but can be activated by organomercurial compounds, albeit more slowly than noncomplexed pro-MMP9 [165, 177, 178]. TIMP-3, which appears to be sequestered in the ECM, is a good inhibitor of ADAM17 and may also regulate other ADAMs [152]. It is also the most effective inhibitor of ADAMTS enzymes [179], and is likely to be an endogenous inhibitor of aggrecanases.
TIMP-1 is produced by many cell types and can also be up-regulated by a variety of factors such as serum, bFGF, EGF, TGFβ, IL-6 family members, retinoids, progesterone, and phorbol ester; expression is repressed by dexamethasone. In murine embryonic development, TIMP-1 is expressed at sites of osteogenesis in the limbs, ribs, digits, skull, and vertebrae [180]. TIMP-2 is often constitutively produced by cells and does not generally show a response to actions of cytokines, however, where expression is regulated, it is often in the opposite direction to TIMP-1 and -3, e.g. expression of TIMP-2 is repressed by TGFβ in various cell types. TIMP-3 has been shown to be induced by factors such as EGF, PDGF, TGFβ, phorbol ester, and dexamethasone [180]. TIMP-4 exhibits the most tissuerestricted expression pattern of the TIMPs, being expressed predominantly in heart and brain [152]. Table II shows some of the factors which regulate TIMP gene expression. The TIMP-1, -2, and -3 promoters have potential Sp1binding sites and share other features which are suggestive of housekeeping genes; however, TIMP-1 and -3 are known to be highly stimulus-responsive. The TIMP-1 promoter has an AP-1 and PEA3 motif in close proximity reminiscent of the inducible MMP genes described above; this gene also appears to have important regulatory sequences
Chapter 11 Matrix Proteinases
downstream of the transcription start sites. The murine TIMP-3 gene has six upstream AP-1 sites, which may be involved in basal expression of the gene, although data from the murine and human genes are in conflict here: the murine gene also has a putative p53-binding site, but this appears to be nonfunctional. The TIMP-2 gene has a promoter proximal AP-1 motif, but this appears not to have a major role in basal expression of the gene [180]. The TIMP-4 gene promoter has at least a functional CCAAT and Sp1 site [181].
E. RECK MMP activity can also be inhibited by a recently identified GPI-anchored inhibitor termed RECK (reversioninducing cysteine-rich protein with Kazal motifs) [182]. RECK was shown to inhibit the activity of MMP-2, -9, and -14, though it is unclear how this occurs at the molecular level.
F. α2-Macroglobulin α2-macroglobulin (α2M) is a plasma proteinase inhibitor which is able to inhibit almost all proteinases including MMPs. Human α2M is a 720 kDa tetrameric protein consisting of four 180 kDa subunits. Each subunit contains an exposed “bait” region, a sequence which can be cleaved by virtually all proteinases. Human MMP1 cleaves human α2-macroglobulin at the Gly–Leu bond in the bait region sequence –Gly–Pro–Glu–Gly679–Leu–Arg–Val–Gly–, which strongly resembles the collagenase-1 cleavage site in calf skin collagen α1(I) and α2(I) chains [183]. This cleavage leads to a conformational change in the α2M which traps the proteinase to the exclusion of large substrates; small (peptide) substrates may still be cleaved. ADAMTS4 and ADAMTS5 cleave α2M at Met690–Gly691, which represents a novel bond cleaved for these enzymes [184]. Inhibition by α2-macroglobulin is irreversible. α2M–proteinase complexes are cleared from the circulation by a specific receptor which is a low-density lipoprotein receptor-related protein. Aside from its function as a proteinase inhibitor, α2M has been shown to bind a variety of cytokines such as PDGF, TGF , bFGF, and IL-1β whence it may act as a carrier protein [185, 186]. Interestingly, α2-macroglobulin can inhibit urokinase-type plasminogen activator (uPA) when it is in solution but not when it is bound to uPA receptors at the cell surface, probably due to steric hindrance preventing the engulfing of the uPA by the α2-macroglobulin [187]. The presence of this inhibitor thus localizes uPA activity to the cell surface.
191
G. Metalloproteinases and TIMPs in Cartilage and Bone Extracts of cartilage from osteoarthritic patients show increased metalloproteinase activity compared to normal cartilage [188]. An imbalance of MMP and TIMP activity is implicated in cartilage loss. Recent expression-profiling studies have examined the expression of all MMPs, ADAMTSs, and TIMPs at the mRNA level in cartilage [189, 190]. These studies have unequivocally shown an increase in MMP13 expression in osteoarthritic cartilage. Interestingly, MMP3 expression appears to be strongly reduced in osteoarthritic cartilage compared to normal tissue. Aggrecanases from the ADAMTS family also tend to show lower expression in this end-stage OA cartilage but this may reflect their potential role early in the disease process [189, 190]. MMPs are raised in the synovium and cartilage of RA patients [191]. In the septic joint, high levels of MMP–TIMP complexes have been found in the synovial fluid, with active MMPs present, but no free TIMPs; rapid cartilage destruction is seen in these patients [192]. Control over MMP activity in vivo occurs at different levels and involves factors such as regulation of gene expression, activation of zymogens, and inhibition of active enzymes by specific inhibitors. The importance of activation of proMMPs in cartilage destruction has recently been underlined by the demonstration that cartilage collagen release in an explant model does not take place until activation occurs, even if the expression of procollagenases is induced [65]; furthermore, the overexpression of interleukin-4 prevents MMP-mediated cartilage erosion in a murine model by interfering with proMMP activation [193]. This pivotal role for activation of proMMPs may explain data from both murine immune complex arthritis, and also antigen-induced arthritis where MMP-3-deficient animals show no difference in proteoglycan depletion to wild-type, but fail to degrade collagen (and hence do not develop severe cartilage erosions) with evidence of failure to activate procollagenases [194, 195]. Growth factors such as IL-1, TNFα, bFGF, PDGF, OSM, and ATRA stimulate MMP expression in cartilage and also induce cartilage resorption; factors such as TGFβ, IGF-I, IL-4, IL-10, and IL-13 may oppose these effects. TIMP-1 and -2 have been shown to prevent collagen release in an ex vivo cartilage resorption model, underlining the involvement of collagenase MMPs in this event [196]. TIMP-3 (but not TIMP-1 or TIMP-2) blocks proteoglycan release in a similar model, again highlighting the role of aggrecanase ADAMTSs in this explant system [197]. MMPs have been implicated in bone turnover, either via removal of osteoid to allow osteoclastic bone resorption or more directly as osteoclast enzymes.
192
Ian M. Clark and Gillian Murphy
In fetal rat calvarial osteoblasts, a variety of growth factors have been shown to alter expression of collagenase (MMP-13) and also to modulate bone resorption in calvarial explant assays; inducing agents include PTH, bFGF, ATRA, PDGF, and IL-6; suppressing factors include IGF-I and -II, TGFβ, and BMP-2. Some of these factors (bFGF, TGFβ, and BMP-2) also induce expression of TIMP-1 and -3 [198]. Human osteoblasts have been reported to express at least MMP-1, -2, -3, -9, -10, -13, and -14 [199]. Synthetic MMP inhibitors have been demonstrated to suppress bone resorption [200]. In human development, MMP-13 is expressed in mineralizing skeletal tissue, in hypertrophic chondrocytes and osteoblasts involved in ossification; in osteoblasts, MMP14 and MMP-2 were co-localized with MMP-13. In postnatal tissues, MMP-13 is expressed at sites of remodeling such as bone cysts and ectopic bone and cartilage formation, whilst in RA patients strong expression is seen in cartilage. MMP-13 was therefore proposed to function in the degradation of type II collagen in primary ossification, skeletal remodeling and joint disease [201]. Similar patterns of expression are seen in developing rat and mouse bone; rat osteoblasts possess a scavenger receptor which removes MMP-13 from the extracellular space [198]. Expression of MMPs has also been noted in osteoclasts, particularly MMP-9, but also MMP-1, -2, -3, -13, and -14 [202]; indeed, some inhibition of bone resorption by MMP inhibitors has been shown even on osteoid-free bone [200].
References 1. Dunn, B. M. (1990). Structure and function of the aspartic proteinases. Genetics, structures and mechanisms. Adv. Exp. Med. Biol. 306, 1–585. 2. Handley, C. J., Mok, M. T., Ilic, M. Z., Adcocks, C., Buttle, D. J., and Robinson, H. C. (2001). Cathepsin D cleaves aggrecan at unique sites within the interglobular domain and chondroitin sulfate attachment regions that are also cleaved when cartilage is maintained at acid pH. Matrix Biol. 20, 543–553. 3. Hembry, R. M., Knight, C. G., Dingle, J. T., and Barrett, A. J. (1982). Evidence that extracellular cathepsin D is not responsible for the resorption of cartilage matrix in culture. Biochim. Biophys. Acta 714, 307–312. 4. Goto, T., Kiyoshima, T., Moroi, R., Tsukuba, T., Nishimura, Y., Himeno, M., Yamamoto, K., and Tnaka, T. (1994). Localisation of cathepsins B, cathepsins D and cathepsins L in rat osteoclast by immunolight and immunoelectron microscopy. Histochemistry 101, 33–40. 5. Duffy, J. M., Walker, B., Guthrie, D., Grimshaw, J., McNally, G., Grimshaw, J. T., Spedding, P. L., and Mollan, R. A. B. (1994). The detection, quantification and partial characterisation of cathepsin B-like activity in human pathological synovial fluids. Eur. J. Clin. Chem. Clin. Biochem. 32, 429–434. 6. Ikeda, Y., Ikata, T., Mishiro, T., Nakano, S., Ikebe, M., and Yasuoka, S. (2000). Cathepsins B and L in synovial fluids from patients with rheumatoid arthritis and the effect of cathepsin B on the activation of pro-urokinase. J. Med. Invest. 47, 61–75
7. Aisa, M. C., Rahman, S., Senin, U., Maggio, D., and Russell, R. G. G. (1996). Cathepsin B activity in normal human osteoblast-like cells and human osteoblastic osteosarcoma cells (MG-63) – regulation by interluekin-1-beta and parathyroid hormone. Biochim. Biophys. Acta 1290, 29–36. 8. Maciewicz, R. A., Wotton, S. F., Etherington, D. J., and Duance, V. C. (1990). Susceptibility of the cartilage collagens type II, type IX and type XI to degradation by the cysteine proteinases, cathepsin B and cathepsin L. FEBS Lett. 269, 189–193. 9. Brömme, D., Okamoto, K., Wang, B. B., and Biroc, S. (1996). Human cathepsin O2, a matrix protein-degrading cysteine protease expressed in osteoclasts J. Biol. Chem. 271, 2126–2132. 10. Maciewicz, R. A., and Wotton, S. F. (1991). Degradation of cartilage matrix components by the cysteine proteinases, cathepsin B and cathepsin L. Biomed. Biochim. Acta 50, 561–564. 11. Lang, A., Horler, D., and Baici, A. (2000). The relative importance of cysteine peptidases in osteoarthritis. J. Rheumatol. 27, 1970–1979. 12. Baici, A., Lang, A., Horler, D., Kissling, R., and Merlin, C. (1995). Cathepsin B in osteoarthritis – cytochemical and histochemical analysis of human femoral head cartilage. Ann. Rheum. Dis. 54, 289–297. 13. Buttle, D. J., Handley, C. J., Ilic, M. Z., Saklatvala, J., Murata, M., and Barrett, A. J. (1993). Inhibition of cartilage proteoglycan release by a specific inactivator of cathepsin B and an inhibitor of matrix metalloproteinases – evidence for 2 converging pathways of chondrocytemediated proteoglycan degradation Arth. Rheum. 36, 1709–1717. 14. Sires, U. I., Schmid, T. M., Fliszar, C. J., Wang, Z. Q., Gluck, S. L., and Welgus, H. G. (1995). Complete degradation of type X collagen requires the combined action of interstitial collagenase and osteoclastderived cathepsin B. J. Clin. Invest. 95, 2089–2095. 15. Hill, P. A., Buttle, D. J., Jones, S. J., Boyde, A., Murata, M., Reynolds, J. J., and Meikle, M. C. (1994). Inhibition of bone resorption by selective inactivators of cysteine proteinases. J. Cell. Biochem. 56, 118–130. 16. Debari, K., Sasaki, T., Udagawa, N., and Rifkin, B. R. (1995). An ultrastructural evaluation of the effects of cysteine proteinases inhibitors on osteoclastic resorptive functions. Calcif. Tiss. Int. 56, 566–570. 17. Bossard, M. J., Thaddeus, T. A., Thompson, S. K., Amegadzie, B. Y., Hannings, C. R., Jones, C., Kurdyla, J. T., McNulty, D. E., Drake, F. H., Gowen, M., and Levy, M. A. (1996). Proteolytic activity of human osteoclast cathepsin K. J. Biol. Chem. 271, 12517–12524. 18. Garnero, P., Borel, O., Byrjalsen, I., Ferreras, M., Drake, F. H., McQueney, M. S., Foged, N. T., Delmas, P. D., and Delaisse, J. M. (1998). The collagenolytic activity of cathepsin K is unique among mammalian proteinases. J. Biol. Chem. 273, 32347–32352. 19. Kafienah, W., Bromme, D., Buttle, D. J., Croucher, L. J., and Hollander, A. P. (1998). Human cathepsin K cleaves native type I and II collagens at the N-terminal end of the triple helix. Biochem. J. 331, 727–732. 20. Li, Z., Hou, W. S., and Bromme, D. (2000). Collagenolytic activity of cathepsin K is specifically modulated by cartilage-resident chondroitin sulfates. Biochemistry 39, 529–536. 21. Li, Z., Hou, W. S., Escalante-Torres, C. R., Gelb, B. D., and Bromme, D. (2002). Collagenase activity of cathepsin K depends on complex formation with chondroitin sulfate. J. Biol. Chem. 277, 28669–28676. 22. Li, Z., Yasuda, Y., Li, W., Bogyo, M., Katz, N., Gordon, R. E., Fields, G. B., and Bromme, D. (2004). Regulation of collagenase activities of human cathepsins by glycosaminoglycans. J. Biol. Chem. 279, 5470–5479. 23. Votta, B. J., Levy, M. A., Badger, A., Bradbeer, J., Dodds, R. A., James, I. E., Thompson, S., Bossard, M. J., Carr, T., Connor, J. R., Tomaszek, T. A., Szewczuk, L., Drake, F. H., Veber, D. F., and Gowen, M. (1997). Peptide aldehyde inhibitors of cathepsin K inhibit bone resorption both in vitro and in vivo. J. Bone Min. Res. 12, 1396–1406. 24. Inui, T., Ishibashi, O., Inaoka, T., Origane, Y., Kumegawa, M., Kokubo, T., and Yamamura, T. (1997). Cathepsin K antisense
Chapter 11 Matrix Proteinases
25.
26.
27.
28.
29.
30. 31.
32. 33.
34.
35.
36.
37.
38.
39.
40.
41.
oligodeoxynucleotide inhibits osteoclastic bone resorption. J. Biol. Chem. 272, 8109–8112. Drake, F. H., Dodds, R. A., James, I. E., Connor, J. R., Debouck, C., Richardson, S., Leerykaczewski, E., Coleman, L., Rieman, D., Barthlow, R., Hastings, G., and Gowen, M. (1996). Cathepsin K, but not cathepsin B, cathepsin L or cathepsin S, is abundantly expressed in human osteoclasts. J. Biol. Chem. 271, 12511–12516. Gelb, B. D., Shi, G. P., Chapman, H. A., and Desnick, R. J. (1996). Pycnodyostosis, a lysosomal disease caused by cathepsin K deficiency. Science 273, 1236–1238. Hou, W. S., Li, W., Keyszer, G., Weber, E., Levy, R., Klein, M. J., Gravallese, E. M., Goldring, S. R., and Bromme, D. (2002). Comparison of cathepsins K and S expression within the rheumatoid and osteoarthritic synovium. Arthritis Rheum. 46, 663–674. Hou, W. S., Li, Z., Buttner, F. H., Bartnik, E., and Bromme, D. (2003). Cleavage site specificity of cathepsin K toward cartilage proteoglycans and protease complex formation. Biol. Chem. 384, 891–897. Konttinen, Y. T., Mandelin, J., Li, T. F., Salo, J., Lassus, J., Liljestrom, M., Hukkanen, M., Takagi, M., Virtanen, I., and Santavirta, S. (2002). Acidic cysteine endoproteinase cathepsin K in the degeneration of the superficial articular hyaline cartilage in osteoarthritis. Arthritis Rheum. 46, 953–960. Barrett, A. J. (1987). The cystatins: a new class of peptidase inhibitors. Trends Biochem. Sci. 12, 193–196. Lerner, U. H., Johansson, L., Ransjo, M., Rosenquist, J. B., Reinholt, F. P., and Grubb, A. (1997). Cystatin C, an inhibitor of bone resorption produced by osteoblasts. Acta Phys. Scand. 161, 81–92. Ménard, H. A., and El-Amine, M. (1996). The calpain-calpastatin system in rheumatoid arthritis. Immunol. Today 17, 545–547. Yamamoto, S., Shimizu, K., Niibayashi, H., Yasuda, T., and Yamamuro, T. (1994). Immunocytochemical demonstration of calpain in synovial cells in human arthritic synovial joints. Biomed. Res. 15, 77–88. Szomor, Z., Shimizu, K., Yamamoto, S., Yasuda, T., Ishikawa, H., and Nakamura, T. (1999). Externalization of calpain (calcium-dependent neutral cysteine proteinase) in human arthritic cartilage. Clin. Exp. Rheumatol. 17, 569–554. Szomor, Z., Shimizu, K., Fujimori, Y., Yamamoto, S., and Yamamuro, T. (1995). Appearance of calpain correlates with arthritis and cartilage destruction in collagen-induced arthritic knee joints of mice. Ann. Rheum. Dis. 54, 477–483. Oshita, H., Sandy, J. D., Suzuki, K., Akaike, A., Bai, Y., Sasaki, T., and Shimizu, K. (2004) Mature bovine articular cartilage contains abundant aggrecan that is C-terminally truncated at Ala719-Ala720, a site which is readily cleaved by m-calpain. Biochem. J. 382, 253–259. Despres, N., Talbot, G., Plouffe, B., Boire, G., and Menard, H. A. (1995). Detection and expression of a cDNA clone that encodes a polypeptide containing 2 inhibitory domains of human calpastatin and its recognition by rheumatoid arthritis sera. J. Clin. Invest. 95, 1891–1896. Salvesen, G., and Enghild, J. J. (1991). Zymogen activation, specificity and genomic structures of human neutrophil elastase and cathepsin G reveal a new branch of the chymotrypsinogen superfamily of serine proteinases. Biomed. Biochim. Acta 50, 665–671. Starkey, P. M., Barrett, A. J., and Burleigh, M. C. (1977). The degradation of articular collagen by neutrophil proteinases. Biochim. Biophys. Acta 483, 386–397. Rao, N. V., Wehner, N. G., Marshall, B. C., Gray, W. R., Gray, B. H., and Hoidal, J. R. (1991). Characterisation of proteinase-3 (PR-3), a neutrophil serine proteinase. J. Biol. Chem. 266, 9540–9548. McDonnell, J., Lobner, J. M., Knight, W. B., Lark, M. W., Green, B., Poe, M., and Moore, V. L. (1993). Comparison of the proteoglycanolytic activities of human leukocyte elastase and human cathepsin G in vitro and in vivo. Connect. Tissue Res. 30, 1–9.
193 42. Kafienah, W., Buttle, D. J., Burnett, D., and Hollander, A. P. (1998). Cleavage of native type I collagen by human neutrophil elastase. Biochem. J. 330, 897–902. 43. Campanelli, D., Melchior, M., Fu, Y. P., Nakata, M., Shuman, H., Nathan, C., and Gabay, J. E. (1990). Cloning of a cDNA for proteinase 3: A serine protease, antibiotic, and autoantigen from human neutrophils. J. Exp. Med. 172, 1709–1715. 44. Schalkwijk, J., Joosten, L. A. B., Van den Berg, W. B., and Van de Putte, L. B. A. (1990). Antigen-induced arthritis in beige (Chediak-Higashi) mice. Ann. Rheum. Dis. 49, 607–610. 45. Pettipher, E. R., Henderson, B., Hardingham, T. E., and Ratcliffe, A. (1989). Cartilage proteoglycan depletion in acute and chronic antigeninduced arthritis. Arth. Rheum. 32, 601–607. 46. Potempa, J., Korzus, E., and Travis, J. (1994). The serpin superfamily of proteinase inhibitors: structure function and regulation. J. Biol. Chem. 269, 15957–15960. 47. Thompson, R. C., and Ohlsson, K. (1986). Isolation, properties and complete amino acid sequence of human secretory leukocyte protease inhibitor, a potent inhibitor of leukocyte elastase Proc. Natl Acad. Sci. USA 83, 6692–6696. 48. Gotis-Graham, I., Hogg, P. J., and McNeil, H. P. (1997). Significant correlation between thrombospondin 1 and serine proteinase expression in rheumatoid synovium. Arth. Rheum. 40, 1780–1787. 49. Mast, A. E., Enghild, J. J., Nagase, H., Suzuki, K., Pizzo, S. V., and Salvesen, G. (1991). Kinetics and physiological relevance of the inactivation of α1-proteinase inhibitor, α1-antichymotrypsin and antithrombin III by matrix metalloproteinases-1 (tissue collagenase), -2 (72-kDa gelatinase / type IV collagenase), and -3 (stromelysin) J. Biol. Chem. 266, 15810–15816. 50. Okada, Y., Watanabe, S., Nakanishi, I., Kishi, J., Hayakawa, T., Watorek, W., Travis, J., and Nagase, H. (1988). Inactivation of tissue inhibitor of metalloproteinases by neutrophil elastase and other serine proteinases. FEBS Lett. 229, 157–160. 51. Dano, K., Andreasen, P. A., Grondahl-Hansen, J., Kristensen, P., Nielsen, L. S., and Skriver, L. (1985). Plasminogen activators, tissue degradation and cancer. Adv. Cancer Res. 44, 139–266. 52. Saksela, O., and Rifkin, D. B. (1988). Cell-associated plasminogen activation: regulation and physiological functions. Ann. Rev. Cell Biol. 4, 93–126. 53. Ellis, V., Behrendt, N., and Dano, K. (1991). Plasminogen activation by receptor-bound urokinase: a kinetic study with both cell-associated and isolated receptor. J. Biol. Chem. 266, 12752–12758. 54. Andreasen, P. A., Georg, B., Lund, L. R., Riccio, A., and Stacey, S. N. (1990). Plasminogen activator inhibitors: hormonally regulated serpins. Mol. Cell Endocrinol. 68, 1–19. 55. Blasi, F., Vassalli, J-D., and Dano, K. (1987). Urokinase-type plasminogen activator: proenzyme, receptor and inhibitors. J. Cell Biol. 104, 801–804. 56. Chapman, H. A., and Stone, O. L. (1985). Characterisation of a macrophage-derived plasminogen-activator inhibitor. Similarities with placental urokinase inhibitor Biochem. J. 230, 109–116. 57. Kruithof, E. K. O (1988). Plasminogen activator inhibitors – a review. Enzyme 40, 113–121 58. Conese, M., Olson, D., and Blasi, F. (1994). Protease nexin-1-urokinase complexes are internalised and degraded through a mechanism that requires both urokinase receptor and α2-macroglobulin receptor. J. Biol. Chem. 269, 17886–17892. 59. Belcher, C., Fawthrop, F., Bunning, R., and Doherty, M. (1996). Plasminogen activators and their inhibitors in synovial fluids from normal, osteoarthritis, and rheumatoid arthritis knees. Ann. Rheum. Dis. 55, 230–236. 60. Ronday, H. K., Smits, H. H., Van Muijen, G. N. P., Pruszczynski, M. S. M., Van Langelaan, E. J., Breedveld, F. C., and Verheijen, J. H. (1996). Difference in expression of the plasminogen activator system in synovial tissue of patients with rheumatoid arthritis and osteoarthritis. Brit. J. Rheumatol. 35, 416–423.
194 61. Busso, N., Peclat, V., So, A., and Sappino, A. P. (1997). Plasminogen activation in synovial tissues: differences between normal, osteoarthritis, and rheumatoid arthritis joints. Ann. Rheum. Dis. 56, 550–557. 62. Lajeunesse, D. (2004). The role of bone in the treatment of osteoarthritis. Osteoarthritis Cartilage 12 Suppl A, S34–38. 63. Oleksyszyn, J., and Augustine, A. J. (1996). Plasminogen modulation of IL-1-stimulated degradation in bovine and human articular cartilage explants – the role of the endogenous inhibitors PAI-1, α2-antiplasmin, α1-PI, α2-macroglobulin and TIMP. Inflammation Res. 45, 464–472. 64. Saito, S., Katoh, M., Matsumoto, M., Matsumoto, S., and Masuho, Y. (1997). Collagen degradation induced by the combination of IL-1α and plasminogen in rabbit articular cartilage explant culture. J. Biochem. 122, 49–54. 65. Milner, J. M., Elliott, S. F., and Cawston, T. E. (2001). Activation of procollagenases is a key control point in cartilage collagen degradation: interaction of serine and metalloproteinase pathways. Arthritis Rheum. 44, 2084–2096. 66. Vignon, E., Mathieu, P., Ejui, J., Descotes, J., Hartmann, D., Patricot, L. M., and Richard, M. (1991). Study of an inhibitor of plasminogen activator (tranexamic acid) in the treatment of experimental osteoarthritis. J. Rheumatol. 18 (Suppl 27) 131–133. 67. van der Laan, W. H., Ronday, H. K., TeKoppele, J. M., Breedveld, F. C., Huizinga, T. W., and Verheijen, J. H. (2003). No therapeutic effect of plasmin antagonist tranexamic acid in rheumatoid arthritis. A doubleblind placebo-controlled pilot study. Clin. Exp. Rheumatol. 21, 359–362. 68. Yang, Y. H., Carmeliet, P., and Hamilton, J. A. (2001) Tissue-type plasminogen activator deficiency exacerbates arthritis. J. Immunol. 167, 1047–1052. 69. Van Ness, K., Chobaz-Peclat, V., Castellucci, M., So, A., and Busso, N. (2002). Plasminogen activator inhibitor type-1 deficiency attenuates murine antigen-induced arthritis. Rheumatology 41, 136–141. 70. Cook, A. D., Braine, E. L., Campbell, I. K., and Hamilton, J. A. (2002). Differing roles for urokinase and tissue-type plasminogen activator in collagen-induced arthritis. Am. J. Pathol. 160, 917–926. 71. Martin, T. J., Allan, E. H., and Fukumoto, S. (1993). The plasminogen activator and inhibitor system in bone remodeling. Growth Regulation 3, 209–214. 72. Hoekman, K., Lowik, C. W. G. M., Van Derruit, M., Bijvoet, O. L. M., Verheijen, J. H., and Papapoulos, S. E. (1991). Regulation of the production of plasminogen activators by bone resorption-enhancing and inhibiting factors in 3 types of osteoblast-like cells. Bone Min. 14, 189–204. 73. Leloup, G., Lemoine, P., Carmeliet, P., and Vaes, G. (1996). Bone resorption and response to calcium regulating hormones in the absence of tissue or urokinase plasminogen activator or of their type 1 inhibitor. J. Bone Min. Res. 11, 1146–1157. 74. Tumber, A., Papaioannou, S., Breckon, J., Meikle, M. C., Reynolds, J. J., and Hill, P. A. (2003). The effects of serine proteinase inhibitors on bone resorption in vitro. J. Endocrinol. 178, 437–447. 75. Daci, E., Everts, V., Torrekens, S., Van Herck, E., Tigchelaar-Gutterr, W., Bouillon, R., and Carmeliet, G. (2003). Increased bone formation in mice lacking plasminogen activators. J. Bone. Miner. Res. 18, 1167–1176. 76. Suzuki, K., Lees, M., Newlands, G. F. J., Nagase, H., and Woolley, D. E. (1995). Activation of precursors for matrix metalloproteinases 1 (interstitial collagenase) and 3 (stromelysin) by rat mast cell proteinases I and II. Biochem. J. 305, 301–306. 77. Tetlow, L. C., and Woolley, D. E. (1995). Distribution, activation and tryptase chymase phenotype of mast cells in the rheumatoid lesion. Ann. Rheum. Dis. 54, 549–555. 78. Froelich, C. J., Zhang, X. L., Turbov, J., Hudig, D., Winkler, U., and Hanna, W. L. (1993). Human granzyme B degrades aggrecan proteoglycan in matrix synthesized by chondrocytes. J. Immunol. 151, 7161–7171. 79. Ronday, H. K., van der Laan, W. H., Tak, P. P., de Roos, J. A., Bank, R. A., TeKoppele, J. M., Froelich, C. J., Hack, C. E., Hogendoorn, P. C.,
Ian M. Clark and Gillian Murphy
80.
81.
82.
83. 84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
Breedveld, F. C., and Verheijen, J. H. (2001). Human granzyme B mediates cartilage proteoglycan degradation and is expressed at the invasive front of the synovium in rheumatoid arthritis. Rheumatology 40, 55–61. Horiuchi, K., Saito, S., Sasaki, R., Tomatsu, T., and Toyama, Y. (2003). Expression of granzyme B in human articular chondrocytes. J. Rheumatol. 30, 1799–1810. Tak, P. P., Kumer, J. A., Hack, C. E., Daha, M. R., Smeets, T. J. M., and Erkelens, G. W. (1994). Granzyme-positive cytotoxic cells are specifically increased in early rheumatoid synovial tissue. Arth. Rheum. 12, 1735–1743. Tak, P. P., Spaeny-Dekking, L., Kraan, M. C., Breedveld, F. C., Froelich, C. J., and Hack, C. E. (1999). The levels of soluble granzyme A and B are elevated in plasma and synovial fluid of patients with rheumatoid arthritis (RA). Clin. Exp. Immunol. 116, 366–370. Werb, Z. (1997). ECM and cell surface proteolysis: regulating cellular ecology. Cell 91, 439–442. Kessler, E., Takahara, K., Biniaminov, L., Brusel, M., and Greenspan, D. S. (1996). Bone morphogenetic protein-1: The type I procollagen C- proteinase. Science 271, 360–362. Sarras, M. P., Jr. (1996). BMP-1 and the astacin family of metalloproteinases: A potential link between the extracellular matrix, growth factors and pattern formation. BioEssays 18, 439–442. Moss, M. L., White, J. M., Lambert, J. H., and Andrews, R. C. (2001). TACE and other ADAM proteases as targets for drug discovery. Drug Discov. Today 6, 417–426. Ohta, S., Harigai, M., Tanaka, M., Kawaguchi, Y., Sugiura, T., Takagi, K., Fukasawa, C., Hara, M., and Kamatani, N. (2001). Tumour necrosis factor-alpha (TNF-alpha) converting enzyme contributes to production of TNF-alpha in synovial tissues from patients with rheumatoid arthritis. J. Rheumatol. 28, 1756–1763. Patel, I. R., Attur, M. G., Patel, R. N., Stuchin, S. A., Abagyan, R. A., Abramson, S. B., and Amin, A. R. (1998). TNF-alpha convertase enzyme from human arthritis-affected cartilage: isolation of cDNA by differential display, expression of the active enzyme, and regulation of TNF-alpha. J. Immunol. 160, 4570–4579. Amin, A. R. (1999). Regulation of tumour necrosis factor-alpha and tumour necrosis factor converting enzyme in human osteoarthritis. Osteoarthritis Cartilage 7, 392–394. Bohm, B. B., Aigner, T., Gehrsitz, A., Blobel, C. P., Kalden, J. R., and Burkhardt, H. (1999). Up regulation of MDC15 (metargidin) messenger RNA in human osteoarthritic cartilage. Arthritis Rheum. 42, 1946–1950. Chubinskaya, S., Cs-Szabo, G., and Kuettner, K. E. (1998). ADAM-10 message is expressed in human articular cartilage. J. Histochem. Cytochem 46, 723–729. Chubinskaya, S., Mikhail, R., Deutsch, A., and Tindal, M. H. (2001). ADAM-10 protein is present in human articular cartilage primarily in the membrane-bound form and is upregulated in response to IL-1alpha in bovine nasal cartilage. J. Histochem. Cytochem. 49, 1165–1176. McKie, N., Edwards, T., Dallas, D. J., Houghton, A., Stringer, B., Graham, R., Russell, G., and Croucher, P. I. (1997). Expression of members of a novel membrane linked metalloproteinase family (ADAM) in human articular chondrocytes. Biochem. Biophys. Res. Commun. 230, 335–339. Flannery, C. R., Little, C. B., Caterson, B., and Hughes, C. E. (1999). Effects of culture conditions and exposure to catabolic stimulators (IL-1 and retinoic acid) on the expression of matrix metalloproteinases (MMPs) and disintegrin metalloproteinases (ADAMs) by articular cartilage chondrocytes. Matrix Biol. 18, 225–237. Inoue, D., Reid, M., Lum, L., Kratzschmar, J., Weskamp, G., Myung, Y. M., Baron, R., and Blobel, C. P. (1998). Cloning and initial characterization of mouse meltrin beta and analysis of the expression of four metalloprotease-disintegrins in bone cells. J. Biol. Chem. 273, 4180–4187.
Chapter 11 Matrix Proteinases 96. Verrier, S., Hogan, A., McKie, N., and Horton, M. (2004). ADAM gene expression and regulation during human osteoclast formation. Bone 35, 34–46. 97. Wolfsberg, T. G., Primakoff, P., Myles, D. G., and White, J. M. (1995). ADAMs, membrane proteins containing a disintegrin and metalloprotease domain – multipotential roles in cell-cell and cellmatrix interaction. J. Cell Biol. 131, 275–278. 98. Apte, S. S. (2004). A disintegrin-like and metalloprotease (reprolysin type) with thrombospondin type 1 motifs: the ADAMTS family. Int. J. Biochem. Cell Biol. 36, 981–985. 99. Collins-Racie, L. A., Flannery, C. R., Zeng, W., Corcoran, C., Annis-Freeman, B., Agostino, M. J., Arai, M., DiBlasio-Smith, E., Dorner, A. J., Georgiadis, K. E., Jin, M., Tan, X. Y., Morris, E. A., and LaVallie, E. R. (2004). ADAMTS-8 exhibits aggrecanase activity and is expressed in human articular cartilage. Matrix Biol. 23, 219–230. 100. Sandy, J. D., Flannery, C. R., Neame, P. J., and Lohmander, L. S. (1992). The structure of aggrecan fragments in human synovial fluid. Evidence for the involvement in osteoarthritis of a novel proteinase which cleaves the Glu373 – Ala374 bond of the interglobular domain. J. Clin. Invest. 89, 1512–1516. 101. Lark, M. W., Bayne, E. K., Flanagan, J., Harper, C. F., Hoerrner, L. A., Hutchinson, N. I., Singer, I. I., Donatelli, S. A., Weidner, J. R., Williams, H. R., Mumford, R. A., and Lohmander, L. S. (1997). Aggrecan degradation in human cartilage. Evidence for both matrix metalloproteinase and aggrecanase activity in normal, osteoarthritic and rheumatoid joints. J. Clin. Invest. 100, 93–106. 102. Singer, I. I., Scott, S., Kawka, D. W., Bayne, E. K., Weidner, J. R., Williams, H. R., Mumford, R. A., Lark, M. W., McDonnell, J., Christen, A. J., Moore, V. L., Mudgett, J. S., and Visco, D. M. (1997). Aggrecanase and metalloproteinase-specific aggrecan neo-epitopes are induced in the articular cartilage of mice with collagen II-induced arthritis. Osteoarthritis Cartilage 5, 407–418. 103. Van Meurs, J. B., van Lent, P. L., Holthuysen, A. E., Singer, I. I., Bayne, E. K., and van den Berg, W. B. (1999). Kinetics of aggrecanase- and metalloproteinase-induced neoepitopes in various stages of cartilage destruction in murine arthritis. Arthritis Rheum. 42, 1128–1139. 104. Matthews, R. T., Gary, S. C., Zerillo, C., Pratta, M., Solomon, K., Arner, E. C., and Hockfield, S. (2000). Brain-enriched hyaluronan binding (BEHAB)/brevican cleavage in a glioma cell line is mediated by a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS) family member. J. Biol. Chem. 275, 22695–22703. 105. Sandy, J. D., Westling, J., Kenagy, R. D., Iruela-Arispe, M. L., Verscharen, C., Rodriguez-Mazaneque, J. C., Zimmermann, D. R., Lemire, J. M., Fischer, J. W., Wight, T. N., and Clowes, A. W. (2001). Versican V1 proteolysis in human aorta in vivo occurs at the Glu441Ala442 bond, a site that is cleaved by recombinant ADAMTS-1 and ADAMTS-4. J. Biol. Chem. 276, 13372–13378. 106. Tortorella, M. D., Malfait, A. M., Deccico, C., and Arner, E. (2001). The role of ADAM-TS4 (aggrecanase-1) and ADAM-TS5 (aggrecanase-2) in a model of cartilage degradation. Osteoarthritis Cartilage 9, 539–552. 107. Malfait, A. M., Liu, R.Q, Ijiri, K., Komiya, S., and Tortorella, M. D. (2002). Inhibition of ADAM-TS4 and ADAM-TS5 prevents aggrecan degradation in osteoarthritic cartilage. J. Biol. Chem. 277, 22201–22208. 108. Yamanishi, Y., Boyle, D. L., Clark, M., Maki, R. A., Tortorella, M. D., Arner, E. C., and Firestein, G. S. (2002). Expression and regulation of aggrecanase in arthritis: the role of TGF-beta. J. Immunol. 168, 1405–1412. 109. Vankemmelbeke, M. N., Holen, I., Wilson, A. G., Ilic, M. Z., Handley, C. J., Kelner, G. S., Clark, M., Liu, C., Maki, R. A., Burnett, D., and Buttle, D. J. (2001). Expression and activity of ADAM-TS5 in synovium. Eur. J. Biochem. 268, 1259–1268.
195 110. Sugimoto, K., Takahashi, M., Yamamoto, Y., Shimada, K., and Tanzawa, K. (1999). Identification of aggrecanase activity in medium of cartilage culture. J. Biochem. 126, 449–455. 111. Tortorella, M. D., Pratta, M., Liu, R. Q., Austin, J., Ross, O. H., Abbaszade, I., Burn, T., and Arner, E. (2000). Sites of aggrecan cleavage by recombinant human aggrecanase-1 (ADAMTS-4). J. Biol. Chem. 275, 18566–18573. 112. Kashiwagi, M., Enghild, J. J., Gendron, C., Hughes, C., Caterson, B., Itoh, Y., and Nagase, H. (2004). Altered proteolytic activities of ADAMTS-4 expressed by C-terminal processing. J. Biol. Chem. 279, 10109–10119. 113. Gao, G., Plaas, A., Thompson, V. P., Jin, S., Zuo, F., and Sandy, J. D. (2004). ADAMTS4 (aggrecanase-1) activation on the cell surface involves C-terminal cleavage by glycosylphosphatidyl inositolanchored membrane type 4-matrix metalloproteinase and binding of the activated proteinase to chondroitin sulfate and heparan sulfate on syndecan-1. J. Biol. Chem. 279, 10042–10051. 114. Pratta, M. A., Tortorella, M. D., and Arner, E. (2000). Age-related changes in aggrecan glycosylation affect cleavage by aggrecanase. J. Biol. Chem. 275, 39096–39102. 115. Brinckerhoff, C. E., and Matrisian, L. M. (2002). Matrix metalloproteinases: a tail of a frog that became a prince. Nature Rev. Mol. Cell Biol. 3, 207–214. 116. Eble, J. A., Ries, A., Lichy, A., Mann, K., Stanton, H., Gavrilovic, J., Murphy, G., and Kühn, K. (1996). The recognition sites of the integrins α1β1 and α2β1 within collagen IV are protected against gelatinase A attack in the native protein. J. Biol. Chem. 271, 30964–30970. 117. Nagase, H. (1995). Human stromelysins 1 and 2. Methods Enzymol. 248, 449–470. 118. Borkakoti, N., Winkler, F. K., Williams, D. H., D’Arcy, A., Broadhurst, M. J., Brown, P. S., Johnson, W. H., and Murray, E. J. (1994). Structure of the catalytic domain of human fibroblast collagenase complexed with an inhibitor. Nat. Struct. Biol. 1, 106–110. 119. Spurlino, J. C., Smallwood, A. M., Carlton, D. D., Banks, T. M., Vavra, K. J., Johnson, J. S., Cook, E. R., Falvo, J., Wahl, R. C., Pulvino, T. A., Wendoloski, J. J., and Smith, D. L. (1994). 1.56 A structure of mature truncated human fibroblast collagenase. Proteins 19, 98–109. 120. Lovejoy, B., Cleasby, A., Hassell, A. M., Longley, K., Luther, M. A., Weigl, D., McGeehan, G., McElroy, A. B., Drewry, D., Lambert, M. H., and Jordan, S. R. (1994). Structure of the catalytic domain of fibroblast collagenase complexed with an inhibitor. Science 263, 375–377. 121. Dhanaraj, V., Ye, Q. Z., Johnson, L. L., Hupe, D. J., Ortwine, D. F., Dubar, J. B., Jr., Rubin, J. R., Pavlovsky, A., Humblet, C., and Blundell, T. L. (1996). X-ray structure of a hydroxamate inhibitor complex of stromelysin catalytic domain and its comparison with members of the zinc metalloproteinase superfamily. Structure 4, 375–386. 122. Stams, T., Spurlino, J. C., Smith, D. L., Wahl, R. C., Ho, T. F., Qoronfleh, M. W., Banks, T. M., and Rubin, B. (1994). Structure of human neutrophil collagenase reveals large S1′ specificity pocket. Nat. Struct. Biol. 1, 119–123. 123. Bode, W., Reinemer, P., Huber, R., Kleine, T., Schnierer, S., and Tschesche, H. (1994). The X-ray crystal structure of the catalytic domain of human neutrophil collagenase inhibited by a substrate analogue reveals the essentials for catalysis and specificity. EMBO J. 13, 1263–1269. 124. Grams, F., Reinemer, P., Powers, J. C., Kleine, T., Pieper, M., Tschesche, H., Huber, R., and Bode, W. (1995). X-ray structures of human neutrophil collagenase complexed with peptide hydroxamate and peptide thiol inhibitors. Implications for substrate binding and rational drug design. Eur. J. Biochem. 228, 830–841. 125. Rowsell, S., Hawtin, P., Minshull, C. A., Jepson, H., Brockbank, S. M., Barratt, D. G., Slater, A. M., McPheat, W. L., Waterson, D., Henney, A. M., and Pauptit, R. A. (2002). Crystal structure of human
196
126. 127.
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
Ian M. Clark and Gillian Murphy MMP9 in complex with a reverse hydroxamate inhibitor. J. Mol. Biol. 319, 173–181. Borkakoti, N. (2004). Matrix metalloprotease inhibitors: design from structure. Biochem. Soc. Trans. 32, 17–20. Gooley, P. R., O’Connell, J. F., Marcy, A. I., Cuca, G. C., Salowe, S. P., Bush, B. L., Hermes, J. D., Esser, C. K., Hagmann, W. K., Springer, J. P., and Johnson, B. A. (1994). The NMR structure of the inhibited catalytic domain of human stromelysin-1. Nat. Struct. Biol. 1, 111–118. Morgunova, E., Tuuttila, A., Bergmann, U., Isupov, M., Lindqvist, Y., Schneider, G., and Tryggvason, K. (1999). Structure of human pro-matrix metalloproteinase-2: activation mechanism revealed. Science 284, 1667–1670. Li, J., Brick, P., O’Hare, M. C., Skarzynski, T., Lloyd, L. F., Curry, V. A., Clark, I. M., Bigg, H. F., Hazleman, B. L., Cawston, T. E., and Blow, D. M. (1995). Structure of full-length porcine synovial collagenase reveals a C-terminal domain containing a calcium-linked, four-bladed -propeller. Structure 3, 541–549. De Souza, S. J., and Brentani, R. R. (1994). On the structure/function relationship of polymorphonuclear-leucocyte collagenase. Biochem. J. 300, 605–605. Chung, L., Dinakarpandian, D., Yoshida, N., Lauer-Fields, J. L., Fields, G. B., Visse, R., and Nagase, H. (2004). Collagenase unwinds triple-helical collagen prior to peptide bond hydrolysis. EMBO J. 23, 3020–3030. Gomis-Rüth, F. X., Gohlke, U., Betz, M., Knäuper, V., Murphy, G., López-Otín, C., and Bode, W. (1996). The helping hand of collagenase-3 (MMP-13): 2.7 Å crystal structure of its C-terminal haemopexin-like domain. J. Mol. Biol. 264, 556–566. Gohlke, U., Gomis-Rüth, F. X., Crabbe, T., Murphy, G., Docherty, A. J. P., and Bode, W. (1996). The C-terminal (haemopexinlike) domain structure of human gelatinase A (MMP2): Structural implications for its function. FEBS Lett. 378, 126–130. Morgunova, E., Tuuttila, A., Bergmann, U., and Tryggvason, K. (2002). Structural insight into the complex formation of latent matrix metalloproteinase 2 with tissue inhibitor of metalloproteinase 2. Proc. Natl. Acad. Sci. USA 99, 7414–7419. Clark, I. M., and Cawston, T. E. (1989). Fragments of human fibroblast collagenase: purification and characterisation. Biochem. J. 263, 201–206. Murphy, G., Allan, J. A., Willenbrock, F., Cockett, M. I., O’Connell, J. P., and Docherty, A. J. P (1992). The role of the C-terminal domain in collagenase and stromelysin specificity. J. Biol. Chem. 267, 9612–9618. Hirose, T., Patterson, C., Pourmotabbed, T., Mainardi, C. L., and Hasty, K. A. (1993). Structure-function relationship of human neutrophil collagenase: identification of regions responsible for substrate specificity and general proteinase activity. Proc .Natl. Acad. Sci. USA 90, 2569–2573. Murphy, G., Willenbrock, F., Ward, R. V., Cockett, M. I., Eaton, D., and Docherty, A. J. P (1992). The C-terminal domain of 72 kDa gelatinase A is not required for catalysis, but is essential for membrane activation and modulates interactions with tissue inhibitors of metalloproteinases. Biochem. J. 283, 637–641. Baragi, V. M., Fliszar, C. J., Conroy, M. C., Ye, Q.-Z., Shipley, J. M., and Welgus, H. G. (1994). Contribution of the C-terminal domain of metalloproteinases to binding by tissue inhibitor of metalloproteinases. C-terminal truncated stromelysin and matrilysin exhibit equally compromised binding affinities as compared to full-length stromelysin. J. Biol. Chem. 269, 12692–12697. Knäuper, V., Osthues, A., DeClerck, Y. A., Langley, K. E., Blaser, J., and Tschesche, H. (1993). Fragmentation of human polymorphonuclear leukocyte collagenase. Biochem. J. 291, 847–854. Butler, G. S., Butler, M. J., Atkinson, S. J., Will, H., Tamura, T., van Westrum, S. S., Crabbe, T., Clements, J., d’Ortho, M. P., and
142.
143.
144.
145.
146. 147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
Murphy, G. (1998). The TIMP2 membrane type 1 metalloproteinase “receptor” regulates the concentration and efficient activation of progelatinase A. A kinetic study. J. Biol. Chem. 273, 871–880. Frisch, S. M., and Morisaki, J. H. (1990). Positive and negative transcriptional elements of the type IV collagenase gene. Mol. Cell. Biol. 10, 6524–6532. Jonat, C., Rahmsdorf, H. J., Park, K. K., Cato, A. C. B., Gebel, S., Ponta, H., and Herrlich, P. (1990). Anti-tumor promotion and anti-inflammation – down modulation of AP-1 (Fos, Jun) activity by glucocorticoid hormone. Cell 62, 1189–1204. Nicholson, R. C., Mader, S., Nagpal, S., Leid, M., Rochette-Egly, C., and Chambon, P. (1990). Negative regulation of the rat stromelysin gene promoter by retinoic acid is mediated by an AP1 binding site. EMBO J. 9, 4443–4454. Springman, E. B., Angleton, E. L., Birkedal-Hansen, H., and Van Wart, H. E. (1990). Multiple modes of activation of latent human fibroblast collagenase – evidence for the role of a cys-73 active site zinc complex in latency and a cysteine switch mechanism for activation. Proc. Natl Acad. Sci. USA 87, 364–368. Pei, D., and Weiss, S. J. (1995). Furin-dependent intracellular activation of the human stromelysin-3 zymogen. Nature 375, 244–247. Crabbe, T., Ioannou, C., and Docherty, A. J. P (1993). Human progelatinase A can be activated by autolysis at a rate that is concentrationdependent and enhanced by heparin bound to the C-terminal domain. Eur. J. Biochem. 218, 431–438. Murphy, G., and Knäuper, V. (1997). Relating matrix metalloproteinase structure to function: Why the “hemopexin” domain. Matrix Biol. 15, 511–518. Okada, Y., Morodomi, T., Enghild, J. J., Suzuki, K., Yasui, A., Nakanishi, I., Salvesen, G., and Nagase, H. (1990). Matrix metalloproteinase 2 from human rheumatoid synovial fibroblasts. Purification and activation of the precursor and enzymic properties. Eur. J. Biochem. 194, 721–730. Hipps, D. S., Hembry, R. M., Docherty, A. J. P., Reynolds, J. J., and Murphy, G. (1991). Purification and characterisation of human 72-kDa gelatinase (type IV) collagenase. Use of immunolocalisation to demonstrate the non-coordinated regulation of the 72-kDa and 95-kDa gelatinases by human fibroblasts. Biol. Chem. Hoppe-Seyler 372, 287–296. Cowell, S.,. Knäuper, V., Stewart, M. L., d’Ortho, M. P., Stanton, H., Hembry, R. M., Lopez-Otin, C., Reynolds, J. J., and Murphy, G. (1998). Induction of matrix metalloproteinase activation cascades based on membrane-type 1 matrix metalloproteinase: associated activation of gelatinase A, gelatinase B and collagenase 3. Biochem. J. 331, 453–458. Baker, A. H., Edwards, D. R., and Murphy, G. (2002) Metalloproteinase inhibitors: biological actions and therapeutic opportunities. J. Cell Sci. 115, 3719–3727. Docherty, A. J. P., Lyons, A., Smith, B. J., Wright, E. M., Stephens, P. E., Harris, T. J. R., Murphy, G., and Reynolds, J. J. (1985). Sequence of human tissue inhibitor of metalloproteinases and its identity to erythroid-potentiating activity. Nature 318, 66–69. Boone, T. C., Johnson, M. J., De Clerck, Y. A., and Langley, K. E. (1990). cDNA cloning and expression of a metalloproteinase inhibitor related to tissue inhibitor of metalloproteinases. Proc. Natl Acad. Sci. USA 87, 2800–2804. Pavloff, N., Staskus, P. W., Kishnani, N. S., and Hawkes, S. P. (1992). A new inhibitor of metalloproteinases from chicken – ChIMP-3 - a 3rd member of the TIMP family. J. Biol. Chem. 267, 17321–17326. Apte, S. S., Hayashi, K., Seldin, M. F., Mattei, M.-G., Hayashi, M., and Olsen, B. R. (1994). Gene encoding a novel murine tissue inhibitor of metalloproteinases (TIMP), TIMP-3, is expressed in developing mouse epithelia, cartilage, and muscle, and is located on mouse chromosome 10. Dev. Dyn. 200, 177–197. Apte, S. S., Mattei, M.-G., and Olsen, B. R. (1994). Cloning of the cDNA encoding human tissue inhibitor of metalloproteinases-3
197
Chapter 11 Matrix Proteinases
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
168.
169.
170.
171.
172.
(TIMP-3) and mapping of the TIMP3 gene to chromosome 22. Genomics 19, 86–90. Apte, S. S., Olsen, B. R., and Murphy, G. (1995). The gene structure of tissue inhibitor of metalloproteinases (TIMP)-3 and its inhibitory activities define the distinct TIMP gene family. J. Biol. Chem. 270, 14313–14318. Leco, K. J., Khokha, R., Pavloff, N., Hawkes, S. P., and Edwards, D. R. (1994). Tissue inhibitor of metalloproteinases-3 (TIMP-3) is an extracellular matrix-associated protein with a distinctive pattern of expression in mouse cells and tissues. J. Biol. Chem. 269, 9352–9360. Silbiger, S. M., Jacobsen, V. L., Cupples, R. L., and Koski, R. A. (1994). Cloning of cDNAs encoding human TIMP-3, a novel member of the tissue inhibitor of metalloproteinase family. Gene 141, 293–297. Uria, J. A., Ferrando, A. A., Velasco, G., Freije, J. M. P., and Lopez-Otin, C. (1994). Structure and expression in breast tumors of human TIMP-3, a new member of the metalloproteinase inhibitor family. Cancer Res. 54, 2091–2094. Greene, J., Wang, M., Liu, Y. E., Raymond, L. A., Rosen, C., and Shi, Y. E. (1996). Molecular cloning and characterization of human tissue inhibitor of metalloproteinase 4. J. Biol. Chem. 271, 30375–30380. Leco, K. J., Apte, S. S., Taniguchi, G. T., Hawkes, S. P., Khokha, R., Schultz, G. A., and Edwards, D. R. (1997). Murine tissue inhibitor of metalloproteinases-4 (Timp-4): cDNA isolation and expression in adult mouse tissues. FEBS Lett. 401, 213–217. Gomis-Ruth, F. X., Maskos, K., Betz, M., Bergner, A., Huber, R., Suzuki, K., Yoshida, N., Nagase, H., Brew, K., Bourenkov, G. P., Bartunik, H., and Bode, W. (1997). Mechanism of inhibition of the human matrix metalloproteinase stromelysin-1 by TIMP-1. Nature, 389, 77–81. Wilhelm, S. M., Collier, I. E., Marmer, B. L., Eisen, A. Z., Grant, G. A., and Goldberg, G. I. (1989). SV40-transformed human lung fibroblasts secrete a 92-kDa type IV collagenase which is identical to that secreted by normal human macrophages. J. Biol. Chem. 264, 17213–17221. Goldberg, G. I., Strongin, A., Collier, I. E., Genrich, L. T., and Marmer, B. L. (1992). Interaction of 92-kDa type IV collagenase with the tissue inhibitor of metalloproteinases prevents dimerization, complex formation with interstitial collagenase, and activation of the proenzyme with stromelysin. J. Biol. Chem. 267, 4583–4591. Goldberg, G. I., Marmer, B. L., Grant, G. A., Eisen, A. Z., Wilhelm, S., and He, C. (1989). Human 72-kilodalton type IV collagenase forms a complex with a tissue inhibitor of metalloproteases designated TIMP-2. Proc. Natl Acad. Sci. USA 86, 8207–8211. Stetler-Stevenson, W. G., Krutzsch, H. C., Wacher, M. P., Margulies, I. M. K., and Liotta, L. A. (1989). The activation of human type IV collagenase proenzyme. Sequence identification of the major conversion product following organomercurial activation. J. Biol. Chem. 264, 1353–1356. Howard, E. W., Bullen, E. C., and Banda, M. J. (1991). Regulation of the autoactivation of human 72-kDa progelatinase by tissue inhibitor of metalloproteinases-2. J. Biol. Chem. 266, 13064–13069. Murphy, G., Houbrechts, A., Cockett, M. I., Williamson, R. A., O’Shea, M., and Docherty, A. J. P (1991). The N-terminal domain of tissue inhibitor of metalloproteinases retains metalloproteinase inhibitory activity. Biochemistry 30, 8097–8101. Fridman, R., Fuerst, R. R., Bird, R. E., Hoyhtya, M., Oelkuct, M., Kraus, S., Komarek, D., Liotta, L. A., Berman, M. L., and Stetler-Stevenson, W. G. (1992). Domain structure of human 72-kDa gelatinase/type IV collagenase. Characterization of proteolytic activity and identification of the tissue inhibitor of metalloproteinases-2 (TIMP-2) binding regions. J. Biol. Chem. 267, 15398–15405. Willenbrock, F., Crabbe, T., Slocombe, P. M., Sutton, C. W., Docherty, A. J. P., Cockett, M. I., O’Shea, M., Brocklehurst, K., Phillips, I. R., and Murphy, G. (1993). The activity of the tissue inhibitors of metalloproteinases is regulated by C-terminal domain
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
184.
185. 186.
187.
188.
interactions: a kinetic analysis of the inhibition of gelatinase A. Biochemistry 32, 4330–4337. Howard, E. W., Bullen, E. C., and Banda, M. (1991). Preferential inhibition of 72-kDa and 92-kDa gelatinases by tissue inhibitor of metalloproteinases-2. J. Biol. Chem. 266, 13070–13075. Howard, E. W., and Banda, M. J. (1991). Binding of tissue inhibitor of metalloproteinases 2 to two distinct sites on human 72-kDa gelatinase. Identification of a stabilization site. J. Biol. Chem. 266, 17972–17977. Kleiner, D. E., Tuuttila, A., Tryggvason, K., and Stetler-Stevenson, W. G. (1993). Stability analysis of latent and active 72-kDa type IV collagenase: the role of tissue inhibitor of metalloproteinases-2 (TIMP-2). Biochemistry 32, 1583–1592. Strongin, A. Y., Collier, I., Bannikov, G., Marmer, B. L., Grant, G. A., and Goldberg, G. I. (1995). Mechanism of cell surface activation of 72-kDa type IV collagenase. Isolation of the activated form of the membrane metalloprotease. J. Biol. Chem. 270, 5331–5338. Murphy, G., Ward, R., Hembry, R. M., Reynolds, J. J., Kühn, K., and Tryggvason, K. (1989). Characterization of gelatinase from pig polymorphonuclear leucocytes. A metalloproteinase resembling tumour type IV collagenase. Biochem. J. 258, 463–472. Ward, R. V., Hembry, R. M., Reynolds, J. J., and Murphy, G. (1991). The purification of tissue inhibitor of metalloproteinases-2 from its 72 kDa progelatinase complex – demonstration of the biochemical similarities of tissue inhibitor of metalloproteinases-2 and tissue inhibitor of metalloproteinases-1. Biochem. J. 278, 179-187. Kashiwagi, M., Tortorella, M., Nagase, H., and Brew, K. (2001). TIMP-3 is a potent inhibitor of aggrecanase 1 (ADAM-TS4) and aggrecanase 2 (ADAM-TS5). J. Biol. Chem. 276, 12501–12504. Edwards, D. R., Leco, K. J., Leco, P. A., Lim, M. S., Phillips, B. W., Raja, J., and Sharma, R. (2000). Regulation of TIMP gene expression. In “Inhibitors of Metalloproteinases in Development and Disease”, S. P. Hawkes, D. R. Edwards, R. Khokha (eds), Harwood Academic Press, Lausanne, Switzerland, pp. 13–25. Young, D. A., Phillips, B. W., Lundy, C., Nuttall, R. K., Hogan, A., Schultz, G. A., Leco, K. J., Clark, I. M., and Edwards, D. R. (2002). Identification of an initiator-like element essential for the expression of the tissue inhibitor of metalloproteinases-4 (Timp-4) gene. Biochem. J. 364, 89–99. Takahashi, C., Sheng, Z., Horan, T. P., Kitayama, H., Maki, M., Hitomi, K., Kitaura, Y., Takai, S., Sasahara, R. M., Horimoto, A., Ikawa, Y., Ratzkin, B. J., Arakawa, T., and Noda, M. (1998). Regulation of matrix metalloproteinase-9 and inhibition of tumor invasion by the membrane-anchored glycoprotein RECK. Proc. Natl Acad. Sci. USA. 95, 13221–13226. Sottrup-Jensen, L., and Birkedal-Hansen, H. (1989). Human fibroblast collagenase-α2-macroglobulin interactions. Localization of cleavage sites in the bait regions of five mammalian a-macroglobulins. J. Biol. Chem. 264, 393–401. Tortorella, M. D., Arner, E. C., Hills, R., Easton, A., Korte-Sarfaty, J., Fok, K., Wittwer, A. J., Liu, R. Q., and Malfait, A. M. (2004). Alpha2macroglobulin is a novel substrate for ADAMTS-4 and ADAMTS-5 and represents an endogenous inhibitor of these enzymes. J. Biol. Chem. 279, 17554–17561. Borth, W. (1992). α2-macroglobulin, a multifunctional binding protein with targeting characteristics. FASEB J. 6, 3345-3353. Borth, W. (1994). α2-macroglobulin, a multifunctional binding and targeting protein with possible roles in immunity and autoimmunity. Ann. N. Y. Acad. Sci. 737, 267–272. Stephens, R. W., Tapiovaara, H., Reisberg, T., Bizik, J., and Vaheri, A. (1991). Alpha-2-macroglobulin restricts plasminogen activation to the surface of RC2A leukemia cells. Cell Regul. 2, 1057–1065. Dean, D. D., Martel-Pelletier, J., Pelletier, J-P., Howell, D. S., and Woessner, J. F. (1989). Evidence for metalloproteinase and metalloproteinase inhibitor imbalance in human osteoarthritic cartilage J. Clin. Invest. 84, 678–685.
198 189. Kevorkian, L., Young, D. A., Darrah, C., Donell, S. T., Shepstone, L., Porter, S., Brockbank, S. M., Edwards, D. R., Parker, A. E., and Clark, I. M. (2004). Expression profiling of metalloproteinases and their inhibitors in cartilage. Arthritis Rheum. 50, 131–141. 190. Bau, B., Gebhard, P. M., Haag, J., Knorr, T., Bartnik, E., and Aigner, T. (2002). Relative messenger RNA expression profiling of collagenases and aggrecanases in human articular chondrocytes in vivo and in vitro. Arthritis Rheum. 46, 2648–2657. 191. Martel-Pelletier, J., McCollum, R., Fujimoto, N., Obata, K., Cloutier, J-M., and Pelletier, J-P. (1994). Excess of metalloproteases over tissue inhibitor of metalloproteases may contribute to cartilage degradation in osteoarthritis and rheumatoid arthritis. Lab. Invest. 70, 807–815. 192. Cawston, T., McLaughlan, P., Coughlan, R., Kyle, V., and Hazleman, B. (1990). Synovial fluids from infected joints contain metalloproteinasetissue inhibitor of metalloproteinase (TIMP) complexes. Biochim. Biophys. Acta 1033, 96–102. 193. van Lent, P. L., Holthuysen, A. E., Sloetjes, A., Lubberts, E., and van den Berg, W.B. (2002). Local overexpression of adeno-viral IL-4 protects cartilage from metallo proteinase-induced destruction during immune complex-mediated arthritis by preventing activation of pro-MMPs. Osteoarthritis Cartilage 10, 234–243. 194. van Meurs, J., van Lent, P., Holthuysen, A., Lambrou, D., Bayne, E., Singer, I., and van den Berg, W. (1999). Active matrix metalloproteinases are present in cartilage during immune complex-mediated arthritis: a pivotal role for stromelysin-1 in cartilage destruction. J. Immunol. 163, 5633–5639. 195. van Meurs, J., van Lent, P., Stoop, R., Holthuysen, A., Singer, I., Bayne, E., Mudgett, J., Poole, R., Billinghurst, C., van der Kraan, P., Buma, P., and van den Berg, W. (1999). Cleavage of aggrecan at the Asn341-Phe342 site coincides with the initiation of collagen damage
Ian M. Clark and Gillian Murphy
196.
197.
198.
199.
200.
201.
202.
in murine antigen-induced arthritis: a pivotal role for stromelysin 1 in matrix metalloproteinase activity. Arth. Rheum. 42, 2074–2084. Ellis, A. J., Powell, L. K., Curry, V. A., and Cawston, T. E. (1994). TIMP-1 and TIMP-2 prevent the release of collagen fragments in cartilage stimulated with IL-1. Biochem. Biophys. Res. Commun. 201, 94–101. Gendron, C., Kashiwagi, M., Hughes, C., Caterson, B., and Nagase, H. (2003). TIMP-3 inhibits aggrecanase-mediated glycosaminoglycan release from cartilage explants stimulated by catabolic factors. FEBS Lett. 555, 431–436. Partridge, N. C., Walling, H. W., Bloch, S. R., Omura, T. H., Chan, P. T., Pearman, A. T., and Chou, W-Y. (1996). The regulation and regulatory role of collagenase in bone. Crit. Rev. in Euk. Gene Exp. 6, 15–27. Bord, S., Horner, A., Hembry, R. M., Reynolds, J. J., and Compston, J. E. (1997). Distribution of matrix metalloproteinases and their inhibitor, TIMP-1, in developing human osteophytic bone. J. Anatomy 191, 39–48. Hill, P. A., Docherty, A. J. P., Bottomley, K. M., O’Connell, J. P., Morphy, J. R., Reynolds, J. J., and Meikle, M. C. (1995). Inhibition of bone resorption in vitro by selective inhibitors of gelatinase and collagenase. Biochem. J. 308, 167–175. Stahle-Backdahl, M., Sandstedt, B., Bruce, K., Lindahl, A., Jimenez, M. G., Vega, J. A., and Lopez-Otin, C. (1997). Collagenase-3 (MMP-13) is expressed during human fetal ossification and re-expressed in postnatal bone remodeling and in rheumatoid arthritis. Lab. Invest. 76, 717–728. Okada, Y., Naka, K., Kawamura, K., Matsumoto, T., Nakanishi, I., Fujimoto, N., Sato, H., and Seiki, M. (1995). Localization of matrix metalloproteinase-9 (92-kilodalton gelatinase/type IV collagenase, gelatinase B) in osteoclasts – implications for bone resorption. Lab. Invest. 72, 311–322.
Part II
Structure and Metabolism of the Extracellular Matrix of Bone and Cartilage
This Page Intentionally Left Blank
Chapter 12
Mineralization, Structure and Function of Bone Adele L. Boskey, Ph.D.
Starr Chair in Mineralized Tissue Research, Hospital for Special Surgery, New York, NY 10021 and Weill Medical College and Graduate School of Medical Sciences of Cornell University, New York, NY 10021
IV. Bone Modeling and Remodeling References
I. Abstract II. The Structure and Function of Bone III. Bone Mineralization
I. ABSTRACT
the locomotion process. In addition to these mechanical functions, bone provides a protective housing for bloodforming marrow, and the bone serves as a reservoir for mineral ions (Ca+2, Mg+2, PO4−3). There are two histologically or radiologically defined bone types [1], dense (also known as compact) bone, and spongy or cancellous (also known as trabecular) bone. When bone is newly formed, as during development or fracture healing, the matrix is loose or woven. With maturation, the bone becomes denser, better organized, and better able to serve its mechanical function. Mature bone is described as compact (dense) or trabecular (meaning composed of little beams). The general features of both compact and trabecular bones are similar. Both are solid mineralized matrices with small canals (canaliculi), and spaces (lacunae), and bone cells. (For discussion of bone cells, see Chapter 14.) In cancellous bone, the matrix, lacunae, and mineral-encased cells (osteocytes) are organized in the form of thin interconnecting spicules. In cortical bone the tissue is organized in Haversian systems or osteons. The Haversian systems consist of a canal containing a blood vessel surrounded by concentric and interstitial lamellae.
The development of bone is a complex process that involves cellular, extracellular, and physicochemical events. Here, the overall process of bone formation is reviewed from a biologic and chemical viewpoint. The factors regulating the initial mineralization of osteoid, and those regulating the entire process of mineralization are the main subjects of this review. Attention is paid to new data on the role of matrix proteins in these processes and to the interaction between collagen and matrix proteins, and matrix proteins and cells in regulating bone mineralization.
II. THE STRUCTURE AND FUNCTION OF BONE Bones are dynamic tissues, replaced as they age or are damaged with newly deposited bone. These dynamic tissues serve numerous essential functions in man and other vertebrates. Bone provides mechanical protection for internal organs, allows the animal to direct motion, and facilitates Dynamics of Bone and Cartilage Metabolism
201
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
202
ADELE L. BOSKEY
In addition to descriptions based on architecture (spicules or lamellae) bone may also be described based on its mechanism of formation during development. Bones that develop by the replacement of a cartilage model (endochondral ossification) are often distinguished from those which form directly (intramembranous ossification). Details of these events will be presented later.
A. The Composition of Bone Bone is a composite material consisting of mineral, collagen, water, noncollagenous proteins, lipids, vascular elements, and cells. The absolute amounts of these constituents vary with animal age, sex, tissue site, health, and dietary status. The arrangement of the bone tissues, as described in the classic treatise by Wolff [2], is such that bones are organized optimally to resist loads imposed by functional activities. Thus with growth and development the tissue must be constantly reshaped and remodeled to maintain this maximization, and maintain a form appropriate to its mechanical function. As used here, modeling refers to formation of bone on sites it has not been before, whereas remodeling refers to its formation on surfaces previously containing bone. Both these processes are required to shape bone. The mineral found in bone is an analog of the naturally occurring geologic mineral hydroxyapatite, Ca10(PO4)6(OH)2 [3]. Bone apatite is a calcium- and hydroxide-deficient apatite containing numerous impurities, the most abundant of which is carbonate [4]. Sodium, potassium, fluoride, and citrate are also common substituents [3–5]. In general, in animals of the same age, the proportion of mineral is greatest in the bones of the ear, and lowest in the ribs. Woven and lamellar bones have a lower mineral content than compact bone. These bones will be described below. Bone mineral apatite crystals are relatively small (~9 nm × 6 nm × 2 nm) in young bone as revealed by atomic force microscopy and electron microscopy [6, 7]. This is in contrast with the slightly larger apatite found in tooth enamel or the much larger geologic apatites. It is the small size of bone mineral crystals that facilitates the incorporation and adsorption of foreign ions, and enables bone mineral to dissolve in the acidic milieu created by the osteoclast during bone remodeling. The small crystal size also enables the crystals to provide the flexible collagen fibrils of the mineral–collagen composite with structural rigidity. Bone mineral crystals, while appearing like needles in some species and plates in others [7] have been shown to be thin curved plates [8], which agglomerate, or fuse, as the bone matures [9].
The variation in the amount of mineral in an osteon was first appreciated by the examination of backscattered electron images [10] and has also been quantified by Fourier transform infrared microspectroscopy (FTIRM) and microscopic imaging (FTIRI) techniques in, which an infrared spectrophotometer, coupled to a light microscope, is used to measure mineral content, composition, and parameters related to mineral crystal size and perfection at 6–20 µm spatial resolution [11]. Figure 1A shows a light micrograph of one such osteon from healthy adult bone, and Figure 1B shows typical FTIR spectral maps from this region. Figure 1C shows the mineral content, calculated from these spectra as the ratio of absorptions of the phosphate (900–1200 cm−1) and amide I (1585–1720 cm−1) peaks respectively in four lines orthogonal to the center of the Haversian canal. Figure 1D shows the variation in carbonate to phosphate ratio along these same directions, and Figure 1E shows the changes in crystal maturity estimated from analyses of the underlying sub-bands in the complex phosphate spectra. These data are presented to illustrate that significant changes in mineral properties occur as the tissue ages, and to emphasize that bone is a dynamic tissue. The carbonate content of bone mineral increases with maturity [12], and there are also variations in the nature of the carbonate substitution. Carbonate can replace hydroxide ions (A-type carbonate) or phosphate ions (B-type carbonate), or can adsorb onto the surface of the apatite (labile carbonate). With increasing bone age the proportion of labile carbonate decreases, and in the most mature bone apatites there is a preponderance of B-substituted carbonate. Less is known about the age and site variation of the other ions that accumulate in bone, and there is some debate as to whether ions such as Mg and Sr are incorporated into the apatite lattice or only reside on the surface of mineral crystals [3]. The second most abundant component of bone is collagen, predominantly type I (for discussion of collagen structure and function see Chapters 1 and 2). Collagen provides bone with elasticity and flexibility and directs the organization of the matrix. The collagen fibers in bone are organized in sheets (periosteal bone), or circumferentially (osteonal bone). They are oriented spirally in the lamellae, and all lie in the same direction in a single lamellae, but in adjacent lamellae they are oriented in a different direction. In physiologically mineralized tissues the long axes of the apatite crystals are always aligned parallel to the axis of the collagen fibrils [9]. This orientation allows the crystals to contribute to the strength of bone. Illustrations of the importance of collagen for the mechanical strength of bone are provided by man and mice with osteogenesis imperfecta. This “brittle” bone disease due
203
Chapter 12 Mineralization, Structure and Function of Bone
A
B
C
D
E Figure 1 Mineral property changes in osteonal bone. (A) Light micrograph of an osteon in adult human bone. (B) Fourier transform infrared spectra going from the center of the osteon (0) to 60 µm from the origin at 10 µm spatial resolution. (C) Mineral-to-matrix ratio in the osteon calculated from spectra obtained in four orthogonal directions increases gradually from the center of the Haversian canal. (D) Carbonate-to-phosphate ratio calculated from the spectra increases as mineral content increases in the four orthogonal directions from the center of the Haversian canal. (E) The mineral crystal maturity (size and perfection) as calculated from curve-fit spectra increases with distance from the center of the Haversian canal. Reprinted from Paschalis et al. [11] FTIR microspectroscopic analysis of human osteonal bone.” Calcif. Tissue Int. 59, 480–487, with kind permission of Springer Science and Business Media.
204
ADELE L. BOSKEY
to genetic abnormalities in collagen structure is associated with frequent fractures, even in the neonate [13, 14]. A second illustration is the observation that when bone is decalcified (by soaking in acid or decalcifying agents) it becomes soft and rubber-like. Water accounts for 5–10% of the weight of bone tissue. Tissue hydration is needed both for nutrition and function of cells, and also for mechanical function of the mineral–collagen composite [15]. Hydrogen bonds between water and collagen contribute to the stabilization of the collagen fibril, and there have been suggestions that dehydration of the bone collagen, i.e., replacement of water by mineral, may take place during the mineralization process. Noncollagenous proteins, reviewed in Chapters 3-5, make up approximately 5% of the dry weight of bone. They are important for a variety of processes described elsewhere [15] as well as in this book. The functions of noncollagenous proteins in mineralization will be reviewed in detail below. Lipids, which account for 2–5 % of the organic matrix of bone, are components of cell membranes, and as such are essential for cell function. They also appear to play an important role in initial mineralization and bone [15].
B. The Cell Biology of Bone Formation The matrix of bone changes with growth and development. To appreciate these changes, it is useful to briefly review the embryological stages of development. More details can be found in a recent article by Reddi [16]. Initial long bone development begins by the vascular invasion of a cartilaginous model. During embryogenesis, a cartilage matrix is formed in the center of the bone surrounded by the periosteum /perichondrium, which throughout development defines the boundary between bone and the soft connective tissues. As the cartilage cells become enlarged, the matrix calcifies, while the cells in the periosteum form a thin layer of bone. The calcified cartilage is removed as a vascular bud and tissue from the periosteum invades. Concurrently, the cartilage cells on either side of this cavity continue to proliferate, causing the bone to grow in length and width. The periosteum continues to form, the cartilage cells closest to the cavity hypertrophy (swell), and a calcified matrix forms. Bone forms on the spicules of calcified cartilage (endochondral bone) and eventually is remodeled, as marrow cells fill in the cavity. As bone starts to form surrounding a marrow cavity, growth continues as the proximal and distal epiphyses move further and further apart. The networks of growth factors, cytokines, hormones, and gene regulatory elements that regulate these processes have been discussed elsewhere
in this book (Chapter 6). For this review, it is only important to note that, during development, calcification of cartilage occurs first, and calcified cartilage is replaced by bone. The calcifying cartilage matrix is enriched in type X collagen [17], extracellular membrane-bound bodies known as matrix vesicles [18], and metalloproteinases [19], all of which are important for mineralization. The matrix vesicles contain Ca-transport proteins known as annexins, complexed acidic phospholipids, and other associated proteins, as part of their “nucleational core” [20]. As endochondral ossification starts, the calcifying cartilage matrix is depleted to some extent of the large aggregating proteoglycans [21]. All of these are important for the calcification process. During intramembranous ossification, as in endochondral ossification, bone is first laid down as a network of spicules. In connective tissue, cells differentiate directly, not osteoblasts, but similar to the accumulation of osteoblasts on the calcified cartilage spicules. These cells aggregate and deposit osteoid, which then becomes mineralized. Like calcifying and uncalcified cartilage, the unmineralized and mineralized osteoid have different composition. During initial osteoid formation, the matrix is rich in cell-binding proteins such as laminin, and fibronectin. Later, bone sialoprotein and osteopontin, also cell-binding proteins, are added [22]. Type I collagen and the small proteoglycans decorin, biglycan, and osteodidherin [23], which regulate collagen fibrollogenesis [23, 24], are laid down upon this initial network. As the tissue matures the collagen fibrils form inter- and intrafibrillar cross-links; the number of such cross-links increases with tissue age [25, 26]. Osteocalcin, the most abundant of the noncollagenous matrix proteins in humans [27], does not accumulate in the osteoid, but only is expressed and appears in the bones as it mineralizes [22, 28]. Several other proteins that are concentrated in the mineralized matrix are listed in Table I, some of these are synthesized by osteoblasts, but some from serum accumulate in bone because of their affinity for hydroxyapatite [15]. As reviewed elsewhere in this volume, the factors regulating the expression and post-translational modification of most of these proteins are known. But how their expression and post-translational modification is linked with the onset of mineralization is less clear.
III. BONE MINERALIZATION Physiologic bone mineralization in mammals involves the ordered deposition of apatite on a type I collagen matrix. The bone apatite crystals are always deposited such that their longest dimension lies parallel to the axis of the collagen fibril (Fig. 2). The cascade of events
Chapter 12 Mineralization, Structure and Function of Bone Table I.
The Extracellular Matrix Proteins of Bone
Collagens Type I Type XI Type V Type III Type XII
Proteoglycans Versican Syndecan CS-decorin CS-biglycan Lumican Osteoadherin Serum proteins Albumin Fetuin (α 2HS-glycoprotein) IgG IgE
Phosphoproteins BAG-75 Bone sialoprotein Dentin matrix protein-1 Dentin sialoprotein MEPE (matrix extracellular phosphoglycoprotein) Osteopontin Osteonectin Gamma carboxy-glutamate containing proteins Osteocalcin Matrix Gla protein
Other Thrombospondin Fibromodulin Proteolipids
includes the formation of that matrix and the oriented deposition of these crystals. Our current knowledge of the process of bone mineralization is derived from a combination of cell and organ culture studies, analysis of bones from normal, mutant, transgenic, and diseased species, analogies drawn from biomineralization in other species and cell free studies in solution [15]. While mineral phases other than apatite have been proposed to be formed as precursors, X-ray diffraction [3], NMR [29], and other studies of newly formed embryonic bone [30–34], have failed to show the presence of other
Figure 2
Electron micrograph of the newly formed mineral in a developing rat bone showing the alignment of the long axis of the electron dense mineral crystals along the collagen fibrils. Courtesy of Dr. S. B. Doty.
205 calcium phosphate phases such as brushite, octacalcin phosphate, or noncrystalline (amorphous) tri-calcium phosphate in the youngest bone. This does not exclude the possibility that these phases may form transiently during the initial steps of bone formation prior to apatite formation. Some nonvertebrates form amorphous mineral phases which may be transient or persistent [35]. Calcium phosphate mineral deposition in bone is, in part, a physical chemical process. Precipitation, in general, takes place from solution when the ion activity product in solution exceeds the solubility product of the precipitating phase. As a first approximation, in the case of hydroxyapatite, that would mean [Ca+2]5 [PO4−3]3 [OH] ≥ 10−57 [36] and most body fluids would be undersaturated. When the concentration of any of these ions is increased, as may be the case during the formation of urinary or salivary stones, crystal deposition may occur. However, even in such cases an energy requiring nucleation process must occur. During nucleation, ions or ion clusters must associate to form a stable configuration containing the building block(s) of the crystal in question. Once this structure (the “critical nucleus”) is large enough to persist in solution, it is easy to continue to add ions or ion clusters to it during the crystal growth process [15]. Secondary nucleation occurs as new crystal starts to form on the initial crystal in a fashion analogous to glycogen branching. The new crystal branches separate and provide additional nuclei. Nucleation, as described above, may occur de novo, or may occur on foreign substances (heterogeneous nucleation). If the foreign substance surface resembles the surface of the nucleus, the process of “epitaxial nucleation” occurs. In biological systems, there are numerous heterogeneous and epitaxial nucleators that facilitate mineralization in a variety of ways. Furthermore, in bone, it is apparent that there are many such nucleation sites, since initial crystals start to form concurrently at numerous separated loci [3]. Examination of newly mineralized collagen shows the first crystals forming in the “e-bands” throughout the collagen matrix [6, 37]. There are two not necessarily mutually exclusive theories concerning how mineral deposition starts at discrete sites in bone. During endochondral ossification, cartilage calcification is widely believed to start within extracellular membrane-bound bodies, known as matrix vesicles (ECMVs) [18, 38]. ECMVs (Fig. 3) released from the chondrocyte, are seen as the site where mineral first appears in the extraterritorial matrix adjacent to the hypertrophic cells of the calcifying cartilage. ECMVs provide protected areas for the accumulation of mineral ions away from the space-filling proteoglycan aggregates that keep the cartilage matrix hydrated, and by virtue of their negative charge and bulk, are effective inhibitors of mineral nucleation
206
Figure 3 Electron micrograph showing extracellular matrix vesicles in calcifying cartilage of a young rat. Courtesy of Dr. S. B. Doty.
and growth [39–42]. ECMVs contain a nucleational core [20] consisting of acidic phospholipids, calcium, inorganic phosphate, and a Ca-transport protein, annexin. This nucleational core is believed to be responsible for membraneassociated mineral formation within the ECMVs. The ECMVs also contain enzymes that can modulate inhibitors of mineral formation found in the extracellular matrix [43, 44]. Isolated ECMVs often have matrix proteins such as type X collagen associated with them [38]. This small chain collagen is important to the organization of the matrix and cell produced by hypertrophic chondrocytes [45]. The ECMVs are always adjacent to collagen (Fig. 3), and there is some controversy concerning how the mineral formed within matrix vesicles gets to the collagen matrix upon which the bone mineral is deposited. High-resolution 3D tomographic imaging of the mineralizing turkey tendon, a highly organized model system frequently used to study processes in bone, has demonstrated mineral crystals growing across from mineralizing vesicles to the collagen fibers [46]. These studies also showed that there were mineralizing sites on the tendon collagen not associated with ECMVs, suggesting that ECMV and collagen mineralizing could occur independently of binding [45]. Mineralization of osteoid, as opposed to cartilage and tendon mineralization, is rarely associated with ECMVs;
ADELE L. BOSKEY
however, ECMVs do occur where bone mineralization is impaired, for example in hypocalcemia [47]. It is not known whether the delayed mineralization of the osteoid enables the observation of ECMVs or whether ECMVs are formed specifically to promote mineral formation when other mechanisms are hindered. It is known that purified type I collagen aggregated into fibrils can support apatite formation from metastable calcium phosphate solutions [48]. However, when extracted collagen was implanted in the animal, mineral deposition was relatively slow [49]. Termine showed that, after noncollagenous proteins were extracted from bone collagen with chaotropic solvents and protease inhibitors, the collagen could not be re-mineralized in vitro [50]. Likewise, Endo and Glimcher showed that dephosphorylation of bone collagen retarded its in vitro mineralization [51]. From this and other evidence it is now generally accepted that there are noncollagenous proteins associated with the collagen that can serve both as nucleators of initial mineral formation and regulators of crystal size and shape [3, 15]. The challenge has been in determining which proteins are essential for these processes. The approach to this question has included identifying which proteins are expressed at the boundary between mineralized bone and osteoid (the mineralization front) using molecular [15, 52, 53, 54] and histochemical [22, 55] techniques. These studies have revealed inter alia that a family of phosphoproteins, the so-called SIBLING proteins [56] for small integrin binding ligand N-glycosylated, are all expressed at the start of mineralization. The SIBLING proteins are all expressed on human chromosome 4q. They include osteopontin, bone sialoprotein, matrix extracellular phosphoglycoprotein (MEPE), dentin matrix protein 1 (DMP1), and dentin sialophosphoprotein (DSPP). All these proteins also bind to complement factor H [57] and several of those studied to date activate metalloproteases [58], showing they have multiple functions as signaling molecules and potential regulators of mineralization. Table II lists additional matrix proteins which have been noted to change in distribution or post-translational modification at the mineralization front. Even though a protein may have an altered distribution or composition at the mineralization front, it may not be directly involved in causing mineral crystals to deposit. It may simply be there because it has a high affinity for mineral. To determine which proteins can act as nucleators, investigators have studied the effects of individual proteins, either isolated from bone or prepared using recombinant technology, on in vitro mineralization [15]. Studies of solution-medicated mineralization have included analysis carried out in media containing only calcium and phosphate salts, media made to mimic the composition of
207
Chapter 12 Mineralization, Structure and Function of Bone Table II. Modified gene
Mutant Animals with Genetically Altered Matrix Proteins and Bone Phenotypes Phenotype
Mineral properties
Reference
Knockouts Biglycan Matrix glaprotein Ostoecalcin DMP-1
Vanishing bone Osteoporosis-like Vascular calcification Thickened bones Altered bone structure
Osteonectin
Thinning bones
Osteopontin
Thickened bones
Type X collagen MEPE Collagen I
Transgenics Disrupted epiphyses Increased bone mass Brittle bone disease (young animals)
body fluids (synthetic lymph [59]), media in which the calcium, phosphate, and hydroxide ion concentrations are kept fixed (constant composition [60]), and others in which the concentrations of these ions are allowed to vary. Some studies examine growth of calcium phosphates added as “seed crystals” to these solutions, others evaluate nucleators in solutions free of dust and other potential heterogeneous nucleators, while still others look for the effects in agar, silica, or gelatin gels containing the proteins in question [61]. Despite the variation in these conditions, in general, there is consistency as to which proteins are found to act as nucleators and which can bind to nuclei or crystals and inhibit or regulate crystal proliferation and growth. These effects, of course, are all concentration dependent. Only a few bone matrix proteins, and the nucleational core discussed above [20], act as in vitro nucleators. The most effective of the proteins analyzed to date is bone sialoprotein [62]. This bone-specific protein (BSP) causes apatite formation in a variety of in vitro conditions [3, 63, 64]. Both the native and the recombinant form of BSP can inhibit mineral crystal growth as well [15, 65]. BSP is distinguished from the other phosphorylated sialoproteins made by osteoblasts in that it is rich in poly-glutamate repeats, whereas another SIBLING protein, osteopontin, contains poly-aspartate sequence, and BAG-75 has a mixture of poly-glutamate and poly-aspartate repeats. BSP, BAG-75, and alkaline phosphatase form spherical structures in osteoblast cultures that are the foci for initial calcification [64]. It has been suggested based on studies with synthetic poly-peptides and peptidomimetics that this poly-glutamate repeat stabilizes the apatite nuclei [66, 67]. This would explain why BSP at high concentrations could coat apatite crystals and retard their growth [65].
Larger but fewer crystals
Less mineral Larger crystals Increased mineral content Increased crystal size Increased mineral content Increased crystal size
No mineral change ? Decreased mineral content Decreased crystallinity
85 87 88, 89 90 86 86
91 101 102
A variety of other bone and dentin matrix components seem to have the ability to act as nucleators at low concentrations and regulate the extent of proliferation of mineral crystals at higher concentrations. These include the lipids of the “nucleational core” of ECMVs [59], biglycan, the small CS-containing leucine-rich proteoglycan of bone [68], and several phosphoproteins including dentin matrix protein-1, a component of bone as well as dentin [69], and dentin phosphophoryn [70]. Although the precise mechanisms of action of these proteins may differ, it is likely that they can interact with specific faces on the apatite crystal, stabilizing the critical nucleus, and blocking growth in certain directions [15]. There is also evidence that these proteins’ ability to nucleate apatite depends on enzymatic and post-translational modifications [69]. Other extracellular matrix proteins isolated from bone, such as osteonectin, osteopontin, and osteocalcin do not act as nucleators but bind to apatite with high affinities [15] and act as effective inhibitors of apatite proliferation [71, 72]. It is interesting to note that a highly phosphorylated form of osteopontin can act as a nucleator, while a small peptide from MEPE is an inhibitor of cell-medicated mineralization [73]. It is thus likely that many of these proteins are modified at sites of mineralization to either enhance or modulate the calcification process. Although it is known that the proteins shown to be nucleators in solution are associated with collagen [3], to determine their action in situ investigators have used in vitro studies of cell and organ systems [e.g., 15, 64, 74]. These studies provide data on the sequence of expression of the matrix proteins [23, 27, 75] and the effects of altered matrix protein expression on mineralization [76–79]. For example, blocking the synthesis of large proteoglycans, and thereby preventing their sulfation, or accelerating the
208
ADELE L. BOSKEY
degradation of large proteoglycans, enhanced the rate of mineralization in calcifying cartilage cultures [75]. Blocking phosphoprotein phosphorylation in the same culture system blocked mineralization [76]. These cell culture studies support the postulated role of proteoglycans and phosphoproteins in bone mineralization. The development of embryonic stem cell technology by which proteins synthesized in ovo could be modified, ablated (“knocked out”), or inserted (“knocked in”) provides an additional tool to probe the functions of the bone matrix proteins. Knocking out these proteins does not always produce a bone phenotype, because many of the essential processes, such as mineralization, are regulated in redundant ways. However, proof of function of several proteins has become available through the studies of both transgenic over-expression (expressing an altered gene protein) and/or knockout (gene ablation) animal models. The first example of a transgenic mouse relevant to bone disease was the model of human osteogenesis imperfecta (brittle bone disease), produced by the random insertion of a viral gene sequence into the mouse genome [80]. These transgenic mice showed the typical bone fragility of the human disease. Other models of human bone disease are
seen in the MEPE knockout which forms hypermineralized bones [81], the brittle bone collagen knock-in [82], the dwarf mice created by deletion of the proteoglycan aggregates link protein [83], the rachitic mice created by deletion of vitamin D hydroxylase gene [84], and the osteoporotic-like condition of biglycan knockout mice [85], to name a few. The effect of the general loss of bone matrix proteins or tissue specific loss of these proteins is indicated in Table II. The numerous mutants in which knockout of cytokines and hormones or hormone receptors produce a bone phenotype are not discussed here. The bones of the knockout and other mutant animals are usually described based on histology, radiography, and changes in mechanical testing. We have used FTIR microspectroscopy and imaging to further characterize these bones (Fig. 4) [85–91]. Changes in bone strength and mineral properties have validated most of the predictions made from in vitro data. Thus, for example, biglycan is an in vitro nucleator [68] and when it is absent in the mouse there is less mineral present but whatsoever crystals are present are larger [85]. There must be many nucleators present in bone, but, when one is absent, the crystals formed initially on other nucleating materials are sparse
Figure 4 Infrared imaging of the bone mineral in two representative knockout animals. Top: the DMP1 knockout at 4 weeks has decreased mineral:matrix ratio (mineral content) in its cortical bone, and increased mineral crystal size/perfection (crystallinity) relative to age matched control. Bottom: the osteonectin (ON) knockout mouse has decreased amounts of trabecular bone at 11 weeks of age relative to wild-type controls, and both the mineral content and the crystallinity in the knockout are decreased. Note the scales for mineral:matrix shown in the top and bottom figures are not the same as the trabecular bone has a high mineral:matrix content at this age in the wild-type.
209
Chapter 12 Mineralization, Structure and Function of Bone
and tend to grow to greater size. Similarly DMP-1 in vitro is both an inhibitor and a nucleator [69]; when it is absent in the knockout mouse there are fewer crystals (supporting the role as a nucleator), but these crystals become excessively large, implying a role as an inhibitor. In contrast, osteopontin is an in vitro [71], and the osteopontin knockout has increased mineral content and increased crystal size [86], both changes being associated with the absence of an inhibitor of crystal growth and proliferation. Osteopontin functions both as a matrix protein and a signaling molecule, and has an important role in bone remodeling [92, 93].
IV. BONE MODELING AND REMODELING Following initial mineralization of calcified cartilage or osteoid, the mineralized matrix remains in a dynamic state. It must be reshaped as the bones grow, and as the loads applied to the bone vary. The periosteum, where new bone formation starts, is very responsive to mechanical load, enabling the modeling to occur. Mineral crystals grow, agglomerate, dissolve, and thus change in composition during remodeling [3, 8, 9, 11]. During endochondral ossification the calcified cartilage matrix serves as a site for the deposition of mineral in the form of the primary spongiosa. Mineral crystals in the woven bone are smaller and contain more impurities than those in compact bone [30, 94] and larger than those in calcified cartilage [95]. This combination of calcified cartilage and woven bone is removed and replaced by woven bone alone. As the epiphyses move further apart and the bone grows longer, the proportions of the bone also change. This modeling is accomplished by the resorption of bone on the endosteal surface and the formation of periosteal bone. Modeling of the structures formed by intramembranous ossification are similarly governed by the periosteum. Remodeling, as contrasted to modeling, refers to the continuous shaping of the bones in response to repair in which the events of endochondral ossification are recapitulated. Remodeling is the skeletal process that allows mineral ion homeostasis. It is regulated by a number of hormones such as vitamin D, parathyroid hormone, calcitonin, and estrogen [96]. The remodeling process in the healthy bone involves the coupled actions of bone-forming and bone-resorbing cells. Sequentially, the remodeling process involves recruitment of osteoclasts to a site on the bone surface, an osteopontin-mediated process [97], and the removal of bone mineral and matrix, creating a resorption pit. As the osteoclast moves away, osteoblasts move in, fill in the pit with osteoid, which is then mineralized. It is this coupling that is defective in osteoporosis, thus, with age, the bones become more and more porous and less able
to perform their mechanical function. These processes are regulated by growth factors, hormones, and cytokines [98]. Among the recently recognized factors that regulate the cross-talk between the bone-forming ostoeblasts and the bone-resorbing osteoclasts is the TNF receptor superfamily [99, 100]. RANK ligand (made by osteoblasts), also called OPGL and TRANCE, and its receptor RANK and the decoy receptor OPG (osteoprotegrin) control osteoclastogenesis. When RANKL binds to RANK, osteoclast development is triggered, but this can be blocked by another osteoblast product, OPG, which binds RANKL and thus blocks osteoclast formation. While many other signaling processes are involved, this is a clear example of cross-talk between these cells. The ways in which mechanical and other signals are conveyed to the osteoclast and osteoblast are still being probed. It is likely that, as these signals become better understood, more will also be learned about the fundamental aspects of bone formation.
Acknowledgments Dr. Boskey’s data described in this review was supported by NIH grants DE04141, AR 037661, and AR 041325.
References 1. Reith, E. J., and Ross, M. J. (1970). “Atlas of Descriptive Histology,” Second Edition. Harper and Row, Publishers, Hagerstown MD, pp. 23–40. 2. Wolff, J. (1986). “The Law of Bone Remodelling”. Translated from the original 1892 edition, by P. Maquet and R. Furlong. Springer-Verlag, Berlin. 3. Glimcher, M. J. (1998). The nature of the mineral phase in bone: biological and clinical implications. In “Metabolic Bone Disease and Clinically Related Disorders” (L. V. Avioli and S. M. Krane, eds.), Academic Press, San Diego, pp. 23–50. 4. Holden, J. L., Clement, J. G., and Phakey P. P. (1995). Age and temperature related changes to the ultrastructure and composition of human bone mineral. J. Bone Miner. Res. 10, 1400–1409. 5. Loong, C. K., Rey, C., Kuhn, L. T., Combes, C., Wu, Y., Chen, S., and Glimcher, M. J. (2000). Evidence of hydroxyl-ion deficiency in bone apatites: an inelastic neutron-scattering study. Bone 26, 599–602. 6. Tong, W., Glimcher, M. J., Katz, J. L., Kuhn, L., and Eppell, S. J. (2003). Size and shape of mineralites in young bovine bone measured by atomic force microscopy. Calcif. Tissue Int. 72, 592–598. 7. Fratzl, P., Schreiber, S., and Klaushofer, K. (1996). Bone mineralization as studied by small-angle x-ray scattering. Connect. Tissue Res. 34, 247–254. 8. Kim, H-M., Rey, C., and Glimcher, M. J. (1995). Isolation of calciumphosphate crystals of bone by nonaqueous methods of low temperature. J. Bone Miner. Res. 10, 1589–1660. 9. Landis, W. J., Hodgens, K. J., Arena, J., Song, M. J., and McEwen, B. F. (1996). Structural relations between collagen and mineral in bone as determined by high voltage electron microscopic tomography. Microsc. Res. Tech. 33, 192–202.
210 10. Boyde, A., and Jones, S. J. (1996). Scanning electron microscopy of bone: instrument, specimen, and issues. Microsc. Res. Tech. 33, 92–120. 11. Paschalis, E. P., DiCarlo, E., Betts, F., Sherman, P., Mendelsohn, R., and Boskey, A. L. (1996). FTIR microspectroscopic analysis of human osteonal bone. Calcif. Tissue Int. 59, 480–487. 12. Tarnowski, C. P., Ignelzi, M. A. Jr., and Morris, M. D. (2002). Mineralization of developing mouse calvaria as revealed by Raman microspectroscopy. J. Bone Miner. Res. 17, 1118–1126. 13. Roughley, P. J., Rauch, F., and Glorieux, F. (2003). Osteogenesis imperfecta-clinical and molecular diversity. Eur. Cell Mater. 5, 41–47. 14. Camacho, N. P., Raggio, C. L., Doty, S. B., Root, L., Zraick,V., Ilg, W.A., Toledano, T. R., and Boskey, A. L. (2001). A controlled study of the effects of alendronate in a growing mouse model of osteogenesis imperfecta. Calcif. Tissue Int. 69, 94–101. 15. Gokhale, J., Robey, P. G., and Boskey, A. L. (2001). “The Biochemistry of Bone”. In “Osteoporosis”, Second Edition (R. Marcus, D. Feldman and J. Kelsey, eds.), Academic Press, San Diego, Volume 1, pp. 107–189. 16. Reddi, A. H. (2000). Morphogenesis and tissue engineering of bone and cartilage: inductive signals, stem cells, and biomimetic biomaterials. Tissue Eng. 6, 351–349. 17. Schmid, T. M., Bonen, D. K., Luchene, L., and Linsenmayer, T. F. (1991). Late events in chondrocyte differentiation: hypertrophy, type X collagen synthesis and matrix calcification. In Vivo. 5, 533–540. 18. Anderson, H. C. (2003). Matrix vesicles and calcification. Curr. Rheumatol. Rep. 5, 222–226. 19. Ortega, N., Behonick, D., Stickens, D., and Werb, Z. (2003). How proteases regulate bone morphogenesis. Ann. N. Y. Acad. Sci. 995, 109–116. 20. Wu, L. N., Genge, B. R., Dunkelberger, D. G., LeGeros, R. Z., Concannon, B., and Wuthier, R. E. (1997). Physicochemical characterization of the nucleational core of matrix vesicles. J. Biol. Chem. 272, 4404–4411. 21. Buckwalter, J. A., Rosenberg, L. C. (1983). Structural changes during development in bovine fetal epiphyseal cartilage. Collagen Relat. Res. 3, 489–504. 22. Cowles, E. A., DeRome, M. E., Pastizzo, G., Brailey, L. L., and Gronowicz, G. A. (1998). Mineralization and the expression of matrix proteins during in vivo bone development. Calif. Tissues Int. 62, 74–82. 23. Ramstad, V. E., Franzen, A., Heinegard, D., Wendel, M., and Reinholt, F. P. (2003). Ultrastructural distribution of osteoadherin in rat bone shows a pattern similar to that of bone sialoprotein. Calcif. Tissue Int. 72, 57–64. 24. Vogel, K. G., Paulsson, M., Heinegard, D. (1984). Specific inhibition of type I and type II collagen fibrillogenesis by the small proteoglycan of tendon. Biochem. J. 223, 587–597. 25. Kuboki, Y., Kudo, A., Mizuno, M., and Kawamura, M. (1992). Timedependent changes of collagen cross-links and their precursors in culture of osteogeneic cells. Calcif. Tissue Int. 50, 473–480. 26. Eyre, D. R., Dickson, I. R., and Van Ness, K. (1988). Collagen crosslinking in human bone and articular cartilage. Age–related changes in the content of mature hydroxypyridinium residues. Biochem. J. 252, 495–500. 27. Gerstenfeld, L. C., Chipman, S. D., Glowacki, J., and Lian, J. B. (1986). Expression to differentiate function by mineralizing cultures of chicken osteoblasts. Dev. Biol. 122, 49–60. 28. Hauschka, P. V., and Wians, F. H., Jr. (1989). Osteocalcin–hydroxyapatite interaction in the extracellular organic matrix of bone. Anat. Rec. 224, 180–188. 29. Wu, Y., Ackerman, J. L., Strawich, E. S., Rey, C., Kim, H. M., and Glimcher, M. J. (2003). Phosphate ions in bone: identification of a calcium-organic phosphate complex by 31P solid-state NMR spectroscopy at early stages of mineralization. Calcif. Tissue Int. 72, 610–626.
ADELE L. BOSKEY 30. Mbuyi–Muamba, J. M., Dequeker, J., and Gevers, G. (1989). Collagen and noncollagenous proteins in different mineralization stages of human femur. Acta Anat. (Basel). 134, 265–268. 31. Lane, J. M., Betts, F., Posner, A. S., and Yue, D. W. (1984). Mineral parameters in early fracture repair. J. Bone Joint Surg. [Am] 66, 1289–1293. 32. Boskey, A. L., and Marks, S. C. (1985). Mineral and matrix alterations in the bones of incisors–absent (ia/ia). Osteopetrotic rat. Calcif. Tissue Int. 37, 287–292. 33. Kuhn, L. T., Wu, Y., Rey, C., Gerstenfeld, L. C., Grynpas, M. D., Ackerman, J. L., Kim, H. M., and Glimcher, M. J. (2000). Structure, composition, and maturation of newly deposited calcium-phosphate crystals in chicken osteoblast cell cultures. J. Bone Miner. Res. 15, 1301–1309. 34. Mendelsohn, R., Hassenkhani, A., DiCarlo, E., and Boskey, A. L. (1989). FT-IR microscopy of endochondral ossification at 20µ spatial resolution. Calcif. Tissue Int. 44, 20–24. 35. Weiss, I. M., Tuross, N., Addadi, L., and Weiner, S. (2002). Mollusc larval shell formation: amorphous calcium carbonate is a precursor phase for aragonite. J. Exp. Zool. 293, 478–491. 36. Driessens, F. C. M., and Verbeeck, R. M. H. (1990). “Biominerals,” CRC Press, Boca Raton, FL., Ch 3, pp. 37–61. 37. Traub, W., Arad, T., and Weiner, S. (1992). Origin of mineral crystal growth in collagen fibrils. Matrix. 12, 251–255. 38. Garimella, R., Bi, X., Camacho, N., Sipe, J. B., and Anderson, H. C. (2004). Primary culture of rat growth plate chondrocytes: an in vitro model of growth plate histotype, matrix vesicle biogenesis and mineralization. Bone 34, 961–970. 39. Eanes, E. D., Hailer, A. W., Midura, R. J., and Hascall, V. C. (1992). Proteoglycan inhibition of calcium phosphate precipitation in liposomal suspensions. Glycobiology 2, 571–578. 40. Boskey, A. L., Maresca, M., Armstrong, A. L., and Ehrlich, M. G. (1992). Treatment of proteoglycan aggregates with growth plate neutral metalloprotease reduces their ability to inhibit hydroxyapatite proliferation in a gelatin gel. J. Orthop. Res. 10, 313–319. 41. Embery, G., Hall, R., Waddington, R., Septier, D., and Goldberg, M. (2001). Proteoglycans in dentinogenesis. Crit. Rev. Oral Biol. Med. 12, 331–349. 42. Cornelissen, M., and de Ridder, L. (1993). Electron microscopic study of proteoglycan aggregates synthesized by embryonic chick sternal chondrocytes in vitro. Eur. J. Morphol. 31, 17–20. 43. Dean, D. D., Boyan, B. D., Muniz, O. E., Howell, D. S., and Schwartz, Z. (1996). Vitamin D metabolites regulate matrix vesicle metalloproteinase content in a cell maturation–dependent manner. Calcif. Tissue Int. 59, 109–116. 44. Anderson, H. C., Sipe, J. B., Hessle, L., Dhanyamraju, R., Atti, E., Camacho, N. P., and Millan, J. L. (2004). Impaired calcification around matrix vesicles of growth plate and bone in alkaline phosphatasedeficient mice. Am. J. Pathol. 164, 841–847. 45. Luckman, S. P., Rees, E., and Kwan, A. P. (2003). Partial characterization of cell-type X collagen interactions. Biochem. J. 372, 485–493. 46. Landis, W. J., Hodgens, K. J., McKee, M. D., Nanci, A., Song, M. J., Kiyonaga, S., Arena, J., and McEwen, B. (1992). Extracellular vesicles of calcifying turkey leg tendon characterized by immunocytochemistry and high voltage electron microscopic tomography and 3–D graphic image reconstruction. Bone Miner. 17, 237–241. 47. Mocetti, P., Ballanti, P., Zalzal, S., Silvestrini, G., Bonucci, E., and Nanci, A. (2000). A histomorphometric, structural, and immunocytochemical study of the effects of diet-induced hypocalcemia on bone in growing rats. J. Histochem. Cytochem. 48, 1059–1078. 48. Glimcher, M. J., Hodge, A. J., and Schmitt, F. O. (1957). Macromolecular aggregation states in relation to mineralization: the collagen hydroxyapatite system as studied in vitro. Proc. Natl Acad. Sci. U.S.A. 43, 860–967.
211
Chapter 12 Mineralization, Structure and Function of Bone 49. Mergenhagen, S. E., Martin, G. R., Rizzo, A., Wright, D. N., and Scott, D. B. (1960). Calcification in vivo of implanted collagens. Biochim. Biophys. Acta. 43, 563–565. 50. Termine, J. D., Kleinman, H. K., Whitson, S. W., Conn, K. M., McGarvey, M. L., and Martin, G. R. (1981). Osteonectin, a bonespecific protein linking mineral to collagen. Cell. 26, 99–105. 51. Endo, A., and Glimcher, M. J. (1989). The effect of complexing phosphoproteins to decalcified collagen on in vitro calcification. Connect. Tissue Res. 21, 179–190. 52. de Bri, E., Reinholt, F. P., Heinegard, D., Mengarelli-Widholm, S., Norgard, M., and Svensson, O. (1996). Bone sialoprotein and osteopontin distribution at the osteocartilaginous interface. Clin. Orthop. Relat. Res. 330, 251–260. 53. Stein, G. S., and Lian, J. B. (1993). Molecular mechanisms mediating proliferation/differentiation interrelationships during progressive development of the osteoblast phenotype. Endocr. Rev. 14, 424–442. 54. Young, M. F., Bi, Y., Ameye, L., and Chen, X. D. (2002). Biglycan knockout mice: new models for musculoskeletal diseases. Glycoconj. J. 19, 257–262. 55. McKee, M. D., and Nanci, A. (1995). Postembedding colloidal-gold immunocytochemistry of noncollagenous extracellular matrix proteins in mineralized tissues. Microsc. Res. Tech. 31, 44–62. 56. Fisher, L. W., and Fedarko, N. S. (2003). Six genes expressed in bones and teeth encode the current members of the SIBLING family of proteins. Connect. Tissue Res. 44, 33–40. 57. Jain, A., Karadag, A., Fohr, B., Fisher, L. W., and Fedarko, N. S. (2002). Three SIBLINGs (small integrin-binding ligand, N-linked glycoproteins) enhance factor H’s cofactor activity enabling MCP-like cellular evasion of complement-mediated attack. J. Biol. Chem. 277, 13700–13708. 58. Fedarko, N. S., Jain, A., Karadag, A., and Fisher, L. W. (2004). Three small integrin binding ligand N-linked glycoproteins (SIBLINGs) bind and activate specific matrix metalloproteinases. FASEB J. 18, 734–746. 59. Wu, L. N., Sauer, G. R., Genge, B. R., Valhmu, W. B., and Wuthier, R. E. (2003). Effects of analogues of inorganic phosphate and sodium ion on mineralization of matrix vesicles isolated from growth plate cartilage of normal rapidly growing chickens. J. Inorg. Biochem. 94, 221–235. 60. Pampena, D. A., Robertson, K. A., Litvinova, O., Lajoie, G., Goldberg, H. A., and Hunter, G. K. (2004). Inhibition of hydroxyapatite formation by osteopontin phosphopeptides. Biochem. J. 378, 1083–1087. 61. Silverman, L., and Boskey, A. L. (2004). Diffusion systems for evaluation of biomineralization. Calcified Tissue Int. 75, 494–501. 62. Ganss, B., Kim, R. H., and Sodek, J. (1999). Bone sialoprotein. Crit. Rev. Oral Biol. Med. 10, 79–98. 63. Hunter, G. K., and Goldberg, H. A. (1993). Nucleation of hydroxyapatite by bone sialoprotein. Proc. Natl Acad. Sci. USA 90, 8562–8565. 64. Midura, R. J., Wang, A., Lovitch, D., Law, D., Powell, K., and Gorski, J. P. (2004). Bone acidic glycoprotein-75 delineates the extracellular sites of future bone sialoprotein accumulation and apatite nucleation in osteoblastic cultures. J. Biol. Chem. 279, 25464–25473. 65. Stubbs, J. T. 3rd, Mintz, K. P., Eanes, E. D., Torchia, D. A., and Fisher, L. W. (1997). Characterization of native and recombinant bone sialoprotein: delineation of the mineral-binding and cell adhesion domains and structural analysis of the RGD domain. J. Bone Miner. Res. 12, 1210–1222. 66. Fujisawa, R., Wada, Y., Nodasaka, Y., and Kuboki, Y. (1996). Acidic amino acid–rich sequences as binding sites of osteonectin to hydroxyapatite crystals. Biochim. Biophys. Acta. 1292, 53–60. 67. Tye, C. E., Rattray, K. R., Warner, K. J., Gordon, J. A., Sodek, J., Hunter, G. K., and Goldberg, H. A. (2003). Delineation of the hydroxyapatite-nucleating domains of bone sialoprotein. J. Biol. Chem. 278, 7949–7955. 68. Boskey, A. L., Spevak, L., Doty, S. B., and Rosenberg, L. (1997). Effects of bone CS-proteoglycans, DS-decorin, and DS-biglycan on
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
hydroxyapatite formation in a gelatin gel. Calcif. Tissue Int. 61, 298-305. Tartaix, P. H., Doulaverakis, M., George, A., Fisher, L. W., Butler, W. T., Qin, C., Salih, E., Tan, M., Fujimoto, Y., Spevak, L., and Boskey, A. L. (2004). Dentin matrix protein-1: In vitro effects on hydroxyapatite formation provide insights into in vivo functions. J. Biol. Chem. 279, 18115–18120. Boskey, A. L., Maresca, M., Doty, S., Sabsay, B., and Veis, A. (1990). Concentration dependent effects of dentin-phosphophoryn in the regulation of in vitro hydroxyapatite formation and growth. Bone Mineral. 11, 55–65. Boskey, A. L., Maresca, M., Ulrich, W., Doty, S. B., Butler, W. T., and Prince, C. W. (1993). Osteopontin-hydroxyapatite interactions in vitro: Inhibition of hydroxyapatite formation and growth in a gelatin-gel. Bone Miner. 22, 147–159. Hunter, G. K., Kyle, C. L., and Goldberg, H. A. (1994). Modulation of crystal formation by bone phosphoproteins: structural specificity of the osteopontin-mediated inhibition of hydroxyapatite formation. Biochem. J. 300, 723–728. Rowe, P. S., Kumagai, Y., Gutierrez, G., Garrett, I. R., Blacher, R., Rosen, D., Cundy, J., Navvab, S., Chen, D., Drezner, M. K., Quarles, L. D., and Mundy, G. R. (2004). MEPE has the properties of an osteoblastic phosphatonin and inhibin. Bone 34, 303–319. Prince, M., Banerjee, C., Javed, A., Green, J., Lian, J. B., Stein, G. S., Bodine, P. V., and Komm, B. S. (2001). Expression and regulation of Runx2/Cbfa1 and osteoblast phenotypic markers during the growth and differentiation of human osteoblasts. J. Cell. Biochem. 80, 424–440. Boskey, A. L., Stiner, D., Binderman, I., and Doty, S. B. (1997). Effects of proteoglycan modification on mineral formation in a differentiating chick limb-bud mesenchymal cell culture system. J. Cell Biochem. 64, 632–643. Boskey, A. L., Guidon, P., Doty, S. B., Stiner, D., Leboy, P., and Binderman, I. (1996). The mechanism of ß-glycerophosphate action in mineralizing chick limb-bud mesenchymal cells. J. Bone Min. Res. 11, 1694–1702. Luppen, C. A., Leclerc, N., Noh, T., Barski, A., Khokhar, A., Boskey, A. L., Smith, E., and Frenkel, B. (2003). Brief bone morphogenetic protein 2 treatment of lucocorticoid-inhibited MC3T3-E1 osteoblasts rescues commitment-associated cell cycle and mineralization without alteration of Runx2. J. Biol. Chem. 278, 44995– 45003. Thomas, G., Moffatt, P., Salois, P., Gaumond, M. H., Gingras, R., Godin, E., Miao, D., Goltzman, D., and Lanctot, C. (2003). Osteocrin, a novel bone-specific secreted protein that modulates the osteoblast phenotype. J. Biol. Chem. 278, 50563–50571. Schinke, T., Amendt, C., Trindl, A., Poschke, O., Muller-Esterl, W., and Jahnen-Dechent, W. (1996). The serum protein alpha2-HS glycoprotein/fetuin inhibits apatite formation in vitro and in mineralizing calvaria cells. A possible role in mineralization and calcium homeostasis. J. Biol. Chem. 271, 20789–20796. Stacey, A., Bateman, J., Choi, T., Mascara, T., Cole, W., and Jaenisch, R. (1988). Perinatal lethal osteogenesis imperfecta in transgenic mice bearing an engineered mutant pro-alpha 1(i) collagen gene. Nature 332, 131–136. Gowen, L. C., Petersen, D. N., Mansolf, A. L., Qi, H., Stock, J. L., Tkalcevic, G. T., Simmons, H. A., Crawford, D. T., Chidsey-Frink, K. L., Ke, H. Z., McNeish, J. D., and Brown, T. A. (2003). Targeted disruption of the osteoblast/osteocyte factor 45 gene (OF45) results in increased bone formation and bone mass. J. Biol. Chem. 278, 1998–2007. Forlino, A., Porter, F. D., Lee, E. J., Westphal, H., and Marini, J. C. (1999). Use of the Cre/lox recombination system to develop a nonlethal knock-in murine model for osteogenesis imperfecta with an alpha1(I) G349C substitution. Variability in phenotype in BrtlIV mice. J. Biol. Chem. 274, 37923–37931.
212 83. Watanabe, H., and Yamada, Y. (1999). Mice lacking link protein develop dwarfism and craniofacial abnormalities. Nat. Genet. 21, 225–229. 84. Panda, D. K., Miao, D., Tremblay, M. L., Sirois, J., Farookhi, R., Hendy, G. N., and Goltzman, D. (2001) Targeted ablation of the 25-hydroxyvitamin D 1alpha -hydroxylase enzyme:evidence for skeletal, reproductive, and immune dysfunction. Proc. Natl Acad. Sci. USA 98, 7498–7503. 85. Xu, T., Bianco, P., Fisher, L. W., Longenecker, G., Smith, E., Goldstein, S., Bonadio, J., Boskey, A., Heegaard, A. M., Sommer, B., Satomura, K., Dominguez, P., Zhao, C., Kulkarni, A. B., Robey, P. G., and Young, M. F. (1998). Targeted disruption of the biglycan gene leads to an osteoporosis-like phenotype in mice. Nat. Genet. 20, 78–82. 86. Boskey, A. L., Spevak, L., Paschalis, E., Doty, S. B., McKee, M. D. (2002). Osteopontin deficiency increases mineral content and mineral crystallinity in mouse bone. Calcif. Tissue Int. 71, 145–154. 87. Boskey, A. L., Karsenty, G., and McKee, M. D. (2000). Mineral characterization of bones and soft tissues in matrix gla protein deficient mice. In “Chemistry and Biology of Mineralized Tissues” (M. Goldberg, A. Boskey, C. Robinson, eds.), Am. Acad. Orthopaed. Surgeons. Chicago, IL, pp. 63–67. 88. Ducy, P., Desbois, C., Boyce, B., Pinero, G., Story, B., Dunstan, C., Smith, E., Bonadio, J., Goldstein, S., Gundberg, C., Bradley, A., and Karsenty, G. (1996). Increased bone formation in osteocalcin– deficient mice. Nature 382, 448–452. 89. Boskey, A. L., Gadaleta, S., Gundberg, C., Doty, S. B., Ducy, P., and Karsenty, G. (1998). FT-IR microspectroscopic analysis of bones of osteocalcin-deficient mice provided insight into the function of osteocalcin. Bone 23, 187–196. 90. Boskey, A. L., Moore, D. J., Amling, M., Canalis, E., Delany, A. M. (2003). Infrared analysis of the mineral and matrix in bones of osteonectin-null mice and their wildtype controls. J. Bone Miner. Res. 18, 1005–1011. 91. Paschalis, E. P., Jacenko, O., Olsen, B., Mendelsohn, R., and Boskey, A. L. (1996). Fourier transform infrared microspectroscopic analysis identifies alterations in mineral properties in bones from mice transgenic for type X collagen. Bone 19, 151–156
ADELE L. BOSKEY 92. Giachelli, C. M., and Steitz, S. (2000). Osteopontin: a versatile regulator of inflammation and biomineralization. Matrix Biol. 19, 615–622. 93. Denhardt, D. T., Noda, M., O’Regan, A. W., Pavlin, D., Berman, J.S. (2001). Osteopontin as a means to cope with environmental insults: regulation of inflammation, tissue remodeling, and cell survival. J. Clin. Invest. 107(9); 1055–1061. 94. Dequeker, J., and Mundy, G. R. (1998). Bone structure and function. In “Rheumatology,” Second Edition (J. H. Klippel and P. A. Dieppe, eds), Mosby, London, pp. 34:1–12. 95. Bigi, A., Cojazzi, G., Panzavolta, S., Ripamonti, A., Roveri, N., Romanello, M., Noris Suarez, K., and Moro, L. (1997). Chemical and structural characterization of the mineral phase from cortical and trabecular bone. J. Inorg. Biochem. 68, 45–51. 96. Jilka, R. L. (2003). Biology of the basic multicellular unit and the pathophysiology of osteoporosis. Med. Pediatr. Oncol. 41, 182–185. 97. Yoshitake, H., Rittling, S. R., Denhardt, D. T., and Noda, M. (1999). Osteopontin-deficient mice are resistant to ovariectomy-induced bone resorption. Proc. Natl Acad. Sci. USA 96, 8156–8160. 98. Raisz, L. G. (1999). Physiology and pathophysiology of bone remodeling. Clin. Chem. 45, 1353–1358. 99. Cheng, X., Kinosaki, M., Murali, R., and Greene, M. I. (2003). The TNF receptor superfamily: role in immune inflammation and bone formation. Immunol. Res. 27, 287–294. 100. Boyle, W. J., Simonet, W. S., and Lacey, D. L. (2003). Osteoclast differentiation and activation. Nature 423, 337–342. 101. Hashimoto-Gotoh, T., Ohnishi, H., Tsujimura, A., Tsunezuka, H., Imai, K., Masuda, H., and Nakamura, T. (2004). Bone mass increase specific to the female in a line of transgenic mice overexpressing human osteoblast stimulating factor-1. J. Bone Miner. Metab. 22, 278–282. 102. Kozloff, K. M., Carden, A., Bergwitz, C., Forlino, A., Uveges, T. E., Morris, M. D., Marini, J. C., and Goldstein, S. A. (2004). Brittle IV mouse model for osteogenesis imperfecta IV demonstrates postpubertal adaptations to improve whole bone strength. J. Bone Miner. Res. 19, 614–622.
Chapter 13
Bone Structure and Strength Ego Seeman BSc MBBS FRACP MD
V. Strength Maintenance VI. Conclusion References
I. Introduction II. Gravity and the Need for Stiffness, Flexibility, Lightness and Speed III. The Material Composition and Structural Design of Bone IV. Bone Modeling and Remodeling – The Mechanism of Bone’s Construction during Growth and Decay with Advancing Age
I. INTRODUCTION
its structure. A negative bone balance between the volumes of bone resorbed and formed in each remodeling transaction is the morphological basis of irreversible bone loss, a process that compromises structure. Understanding the mechanisms of adaptation and challenges to it requires attention to the individual components of bone’s material composition and structural design so that the regulators and co-regulators of this machinery can be identified. With this approach, rational approaches to intervention can be designed to prevent and to restore structural compromise and ensuring fragility fractures.
Structural failure, another way of referring to skeletal fracture, is a problem in biomechanics. Understanding structural failure, therefore, requires the study of structure. Currently available noninvasive techniques provide a measure of bone mineral mass in a region of interest but they provide little information about bone’s material composition and structural design, two critical determinants of strength [1]. It is self-evident that structure determines tolerable loads but it is less evident as loads also determine structure. Bone has the ability to modify its material composition and structural design to accommodate prevailing loads [2]. This is achieved by the cellular machinery of adaptive modeling and remodeling. Adaptation is most successful when bones are in their growth phase, but with maturation of the skeleton maintaining bone strength is a greater challenge, particularly when aging leads to sex hormone deficiency, secondary hyperparathyroidism, reduced physical activity, loss of muscle mass, and other factors. Age-related abnormalities in the cellular machinery of bone itself can also contribute to the compromised ability of bone to adapt. Alterations in the rate of remodeling can compromise material properties of bone and also indirectly Dynamics of Bone and Cartilage Metabolism
Dept of Endocrinology and Medicine, Austin Hospital, University of Melbourne, Melbourne, Australia.
II. GRAVITY AND THE NEED FOR STIFFNESS, FLEXIBILITY, LIGHTNESS AND SPEED Land-dwelling mammals must be able to move to catch dinner, and move quickly to avoid becoming it. Propulsion against gravity requires levers. Bones are levers and levers must be stiff. They must resist deformation but they must also be light enough to allow rapidity of movement. Impact loading imparts energy to bone. As energy cannot be destroyed, it must be absorbed by bone, a process 213
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
214
EGO SEEMAN
requiring flexibility. Bone must be a spring, able to change shape, deform without cracking, shorten and widen in compression, lengthen and narrow in tension. The elastic properties of bone allow it to absorb energy by deforming reversibly when loaded [2]. If the load imposed exceeds bone’s ability to deform elastically, it can deform further, yielding by changing shape permanently in plastic deformation. The permanent change in shape is associated with microcracks that allow energy release, a compromise of structure that is a final defense against the alternative – complete fracture. Provided these microcracks are small and do not extend in length, bone can remain intact. If both the elastic and plastic zones are exceeded, the only way the energy can be released is fracture (producing two pieces of bone). Bone achieves its strength in each of these contradictory properties – stiffness yet flexibility, lightness yet strength, by its material composition and structural design.
III. THE MATERIAL COMPOSITION AND STRUCTURAL DESIGN OF BONE The right balance between material stiffness and flexibility is achieved by varying the degree of mineral content of
the bone tissue [1]. The greater the mineral content, the greater the material stiffness but the lower its flexibility (Fig. 1). Through millions of years, nature has selected characteristics that are most suited to the particular function bone usually performs. Ossicles of the ear are densely mineralized to be stiff and vibrate like tuning forks without loss of energy in deformation. In contrast, deer antlers are less densely mineralized and so can deform like springs during head butting in the mating season [2] (Fig. 2). The fabric of bone is configured as a structure. Structural stiffness and flexibility (as opposed to material stiffness and flexibility), and lightness, are achieved by geometry and architectural design – not just its mass. Indeed, for tubular bones, larger and smaller bones within a species do not differ by the amount of material used to construct them. Although counterintuitive, there is no correlation between the external volume of long bones and bone mass. The same amount of material is used to build a wider and longer tubular bone by fashioning it with a thinner cortex [3]. This is an example of minimizing mass for optimal function. Nature can fashion a bigger bone using the same material by creating a bigger hole in the middle of the bone. Larger bones have a larger medullary canal, constructed during intrauterine and postnatal life by excavation of this
Figure 1 Increasing tissue mineral content increases material stiffness but decreases the bone material’s ability to absorb energy without cracking (toughness) (data provided by courtesy of JD Currey).
Chapter 13 Bone Structure and Strength
215
Figure 2
Bones requiring stiffness like auditory ossicles are selected for high mineral content, they are very brittle. Bones requiring flexibility like antlers are selected for lower mineral content (adapted from JD Currey [2]).
cavity by bone resorption. Larger bones have a lower apparent volumetric BMD than smaller bones. The further displacement of the slightly thinner cortex confers greater resistance to bending (structural stiffness) because bending strength is proportional to the fourth power of the radius [4]. In long bones, structures that are effectively levers, stiffness is favored over flexibility.
Figure 3 During growth before puberty periosteal apposition increases long bone diameter and exceeds endocortical resorption, so the bone develops a thicker and thicker cortex. Cortical thickness is similar by sex because reduced endocortical resorption or increased endocortical apposition during puberty contribute to final cortical thickness in females. After epiphyseal closure, endocortical resorption now exceeds the slow periosteal apposition, producing cortical thinning.
A. Building Levers As long bones grow in length, periosteal apposition increases its diameter while concurrent endocortical resorption excavates the marrow cavity (Fig. 3). As periosteal apposition exceeds net endocortical resorption the lengthening bone develops a wider and wider cortex. The enlarging medullary canal effectively shifts the thickening cortex further and further from the long bone neutral axis, producing the structural stiffness needed for lever function. In females, earlier completion of longitudinal growth and earlier inhibition of periosteal apposition produces a smaller bone. However, cortical thickness is similar across both sexes because a reduction in endocortical resorption and perhaps net endocortical apposition contributes to final cortical thickness [5, 6]. What differs among the sexes, therefore, is the position of the cortex in relationship to the long axis of the long bone. Likewise, between races, the main difference is the position of the cortex in relationship to the long axis of the long bone, not the cortical thickness per se. The absolute and relative growth of periosteal and endocortical surfaces varies by sex, pubertal stage, the type of bone, and varies in degree, in fact, at every position and surface along the bone. At each level, these traits are
determined by the interaction between bone formation on the periosteum, endocortical bone resorption, and formation adjacent to marrow. Long bones are not drinking straws with the same diameter throughout and the same thickness of cortex throughout. The complex and irregular periosteal and endocortical perimeters of long bone shafts create elliptical and triangular structures, adapted locally to the loads they are exposed to and adapted to tolerate. These loads then determine extent and degrees of local modeling and remodeling in adjacent regions. The process of adaptation involves using the same material but fashioning and refashioning the same amount of bone into the shape and contours best suited to tolerated prevailing loads. This notion is important and can be seen in the complex shape of the femoral neck. Near the shaft, where bending moments are greatest the shape is elliptical and the cortex is thicker inferiorly than superiorly. The mass of mineralized bone is mainly cortical. Moving proximally along the femoral neck the amount of bone material remains constant but the shape of the femoral neck is more circular, the proportion of trabecular bone increases and the cortical thickness is similar superiorly and inferiorly, features adapted to the loading pattern present which is more shear
216
EGO SEEMAN
and compressive than bending [3]. The loading circumstances are changing geometry not bone mass. While this process is successful during growth, it is not very successful after epiphyseal closure has occurred.
B. Building Springs In the axial skeleton (e.g. vertebral body of the spine), bone is constructed differently but also demonstrates the effective use of void space to achieve lightness with a minimal amount of material. The vertebral bodies serve more as springs or shock absorbers than levers. The structure is light yet “strong” in a different sense of the word; these bones are “stronger” in their ability to deform without cracking. While tubular bones can deform by only 1–2% of their original length, vertebral bodies can deform more substantially by absorbing energy but they sacrifice peak stresses tolerated (load per unit area) in favor of greater peak strain (change in length/original length). The ability to deform facilitates flexion, extension, and rotation of the whole vertebral skeleton of the upper body (Fig. 4). For the vertebrae, increasing bone size by periosteal apposition builds a wider vertebral body in males than females and in some races than others. The number of trabeculae, established at the growth plates, does not increase with age [7]. At puberty, trabecular BMD increases by increasing the size and thickness of the trabeculae plates and sheets to a similar degree in boys and girls so that males and females have the same vertebral body trabecular BMD (number and thickness of trabeculae) during prepubertal and pubertal growth, and at peak young
Figure 4
Cortical bone tolerates greater loads (“stronger” in load bearing) at the price of less flexibility while trabecular bone is able to deform more (“stronger” in being more flexible) but sacrifices peak loading ability.
adulthood [8, 9]. Thus, growth does not build a “denser” skeleton in males than females, it builds a bigger skeleton. Strength of the vertebral body is greater in young males than females due to sex differences in size, not BMD.
IV. BONE MODELING AND REMODELING – THE MECHANISM OF BONE’S CONSTRUCTION DURING GROWTH AND DECAY WITH ADVANCING AGE That structure determines the loads that can be tolerated is obvious. The shape of bone is contained within the ancestral genetic code selected for survival in an anticipated environment; fetal lower limb buds removed and grown in vitro, form the shape of the proximal femur [10]. The concept in reverse, namely that loads determine structure, is less obvious but central to understanding the pathogenesis of bone fragility. Bone accommodates load in given circumstances, by adaptive modeling and remodeling. This process occurs throughout the whole of life, but its purpose differs during growth (when the epiphyses are open) and during adulthood (when the epiphyses are closed). During growth, through the processes of bone modeling and remodeling, remarkable adaptations to loading can be achieved. A wealth of literature on skeletal morphology in elite athletes supports this premise [11]. Variability in loading and the forces associated with it help to account for the complex shape of bone described above. For example, in animal studies, collagen abnormalities in the MOV13 mutant mouse result in reduced bone strength but vigorous adaptive modeling by periosteal apposition results in bone strength above that seen in the wild-type [12]. Likewise, in other mouse models of osteogenesis, imperfecta adaptations take place in material composition rather than its structure [13]. The process of adaptive modeling and remodeling can compensate for abnormalities in one trait that reduce whole bone strength to modify another trait to preserve whole bone strength. The compensatory effects are limited. If the severity of the abnormality is too great [14], or the compensatory mechanism is itself abnormal, then adaptation is incomplete, resulting in more fragile bones. The emergence of bone fragility during aging can be viewed in this way. Over many years, accumulating abnormalities due to disease, hormonal deficiency and/or excess, exposure to risk factors such as tobacco and excess alcohol, as well as abnormalities in the machinery of bone modeling and remodeling itself, produce changes in the material composition of bone and its structure that together form the
217
Chapter 13 Bone Structure and Strength
basis of bone fragility. Viewed in this way, bone fragility, or osteoporosis, can be regarded as a disorder of failed adaptation.
V. STRENGTH MAINTENANCE After completion of longitudinal growth (closure of the epiphyses), bone modeling and remodeling help to maintain bone strength, rather than to increase it. After bone size has been achieved during the growth period, periosteal apposition ceases to be vigorous and the remodeling rate, reflected in the levels of the biochemical measures of bone remodeling, declines. The remodeling rate is rapid during growth because each remodeling event deposits only a small amount of bone as bone is constructed [15]. Whether there is a relationship between the rate of remodeling and the degree of positive BMU balance is not known. It is possible that smaller positive balances can be compensated for by higher rates of remodeling or vice versa. As growth nears its “programmed” completion rapid remodeling is no longer needed and a positive balance in each remodeling unit is also no longer needed. The remodeling rate slows down and there may be a progressive lessening of the degree of positive balance at the level of the BMU. The first “abnormality” signaling the onset of a change in bone structure may be a reduction in bone formation at the cellular level. Information regarding this is controversial and inconsistent [16–18]. Evidence for a decline in the volume of bone formed is documented at midlife (around 50 years of age), but there is evidence for a decline in peak bone mineral mass and trabecular bone volume well before the menopause [19–21]. After epiphyseal closure, rapid periosteal apposition ceases to the point where bone enlarges no more than a few millimeters over the next 60 years [22–24]. Whereas during growth periosteal apposition is greater than net endocortical resorption, producing an increasingly thicker cortex, during the aging process, endocortical resorption exceeds periosteal apposition. The result of this switch in relative activities of periosteal apposition, declining, and endocortical resorption, maintained or increasing, is a thinner cortex with slight increases in bone diameter. Negative bone balance within each BMU is the metabolic basis of bone loss. On the endocortical surface, this process produces cortical thinning while bone loss on the intracortical surface produces intracortical porosity. The increasing porosity of cortical bone effectively trabecularizes the cortex [25, 26]. The surface or volume ratio increases so that remodeling continues vigorously in cortical bone, producing cortical fragility. The same loads on bone are imposed on a structure diminished
Figure 5
Trabecular bone loss by loss of trabecular numbers and so loss of connectivity (as occurs in women) produces a greater loss of strength than trabecular bone loss by thinning (as occurs in men). Adapted from Ref [28].
in cross-sectional diameter so that the stresses on bone (load per unit area) increase predisposing to buckling, microdamage and ultimately fracture. The amount of trabecular bone lost during aging in women and men is similar, or only slightly less in men than women [4, 5]. However, trabecular bone loss occurs mainly by thinning in men and mainly by loss of connectivity in women [27] (Fig. 5). Strength of the vertebrae decreases more when there is loss of connectivity rather than thinning predisposing to fractures [28]. Loss of connectivity is the result of the accelerated loss of bone that occurs in midlife in women due to estrogen deficiency. Estrogen deficiency is associated with increased remodeling on the endosteal surface. In addition, imbalance in the BMU increases as estrogen deficiency increases the lifespan of osteoclasts and reduces the lifespan of osteoblasts. Increased numbers of BMUs, coupled with increased resorption depth, is responsible for the loss of connectivity seen in women [29]. The contribution of trabecular bone loss to overall bone loss decreases as trabecular plates perforate and disappear because there is less trabecular surface available for remodeling. In men, there is no midlife acceleration of bone remodeling because men do not go through an abrupt period of sex steroid deficiency, but bone loss, nevertheless, occurs due to reduced bone formation and thinning of trabeculae. Relatively greater maintenance of connectivity results in persistence of the trabecular surfaces available for remodeling so that trabecular bone loss probably continues longer in men than in women.
218
EGO SEEMAN
That is why final trabecular bone volume is similar in men and women. The amount of trabecular surface available for bone remodeling in old age appears to be greater in elderly men than women [27].
A. Periosteal Bone Formation During Aging The periosteal bone formation during aging offsets bone loss from the endosteal surfaces in both men and women. Periosteal apposition may be greater at some sites in men than in women so that the similar loss of bone from the endosteal surface is more greatly offset in men than in women [30, 31]. Thus, the smaller decline in BMD at the spine in men than women is due to the smaller reduction in cortical bone in men than in women. However, the better maintenance of cortical mass in men is the result of greater periosteal bone formation, not less resorptive removal of bone from the endosteal surface [Fig. 6].
B. Why Fewer Men than Women Sustain Fragility Fractures The larger skeleton in men produces a bone that tolerates larger absolute loads. However, the load per unit area (stress) imposed on the vertebral body is no different in young men and women, because the larger bone is subjected to correspondingly larger loads. There is scaling in nature – a bigger bone has a bigger muscle, a relationship that is likely to be largely determined by genes regulating size, not necessarily the amount of exercise undertaken. Fragility fractures are uncommon in young men and women because loads are less than the ability of the bone to withstand them. Structural failure emerges during aging because of the changing relationship between the imposed load and the bone’s ability to tolerate that load. Periosteal apposition increases the cross-sectional diameter of the bone more in men, so that load imposed per unit area decreases more than in women. Greater periosteal
Figure 6 Subperiosteal area increases more in men than in women due to greater periosteal apposition in men. Medullary expansion with age is similar in men and women. The net effect is better maintenance of cortical area and strength in men than in women (adapted from Ruff and Hayes [4]).
219
Chapter 13 Bone Structure and Strength
apposition adds more bone to the outer perimeter of the bone in men than women, offsetting the age-related endosteal bone loss more in men than women so that the fall in volumetric BMD is less in men than in women. During aging, the stress on bone decreases more in men, and bone strength decreases less, so that a lower proportion of elderly men than elderly women have bone size and architectural and material properties such as microdamage, tissue mineral density, loss of connectivity, porosity, trabecular and cortical thinning below a critical level (or fracture threshold) where the loads on the bone are greater than the bone’s ability to tolerate them. Structural failure occurs less in men than women because the relationship between load and bone strength is better maintained in men than in women.
4. 5.
6.
7.
8.
9.
10.
VI. CONCLUSION
11. 12.
Bones must be stiff so that they do not bend when loaded. Bones must also be flexible, able to absorb energy by elastic and plastic deformation. Structural failure may occur if bones deform too little or too much. Age-related and menopause-related abnormalities in bone remodeling produce loss of the material and structural properties that no longer keep bone “just right”. High remodeling reduces the mineral content of bone tissue resulting in loss of stiffness. Sex hormone deficiency increases the volume of bone resorbed and reduces the volume of bone formed in each BMU. The contributions made by differences in material composition (tissue mineral content, collagen type, and cross linking) and structure (bone size, cortical thickness and porosity, trabecular number, thickness, connectivity), and other factors (micro-damage burden, osteocyte density) to sex and racial difference bone fragility remain poorly defined. We do not know why and how bones break. The challenge is to measure these specific material and structural determinants of bone strength. Whether a combination of these material and structural properties will more accurately identify women likely to sustain fractures, or improve approaches to drug therapy, is unknown, but it is likely.
13.
14.
15.
16.
17.
18. 19.
20.
21.
References 1. Seeman, E. (1997). From density to structure: growing up and growing old on the surfaces of bone. J. Bone Min. Res. 12, 1–13. 2. Currey, J. D. (2002). Bones. In “Structure and Mechanics”. Princeton, UP, New Jersey, pp. 1– 380. 3. Zebaze, R. M., Jones, A., Welsh, F., Knackstedt, M., and Seeman, E. (2005). Femoral neck shape and the spatial distribution of its mineral
22.
23. 24.
mass varies with its size: Clinical and biomechanical implications. Bone 37(2), 243–252. Ruff, C. B., and Hayes, W. C. (1998). Sex differences in age-related remodeling of the femur and tibia. J. Orthop. Res. 6, 886–896. Duan, Y., Wang, X. F., Evans, A., and Seeman, E. (2005). Structural and biomechanical basis of racial and sex differences in vertebral fragility in Chinese and Caucasians. Bone 36, 987–998. Wang, X. F., Duan, Y., Beck, T., and Seeman, E. R. (2005). Varying contributions of growth and aging to racial and sex differences in femoral neck structure and strength in old age. Bone 36, 978–986. Parfitt, A. M., Travers, R., Rauch, F., and Glorieux, F. H. (2000). Structural and cellular changes during bone growth in healthy children. Bone 27, 487–494. Gilsanz, V., Roe, T. F., Stefano, M., Costen, G. and Goodman, W. G. (1991). Changes in vertebral bone density in black girls and white girls during childhood and puberty. New Engl. J. Med. 325, 1597–1600. Gilsanz, V., Gibbens, D. T., Roe, T. F., Carlson, M., and Senac, M. O. (1988). Vertebral bone density in children: effect of puberty. Radiology 166, 847–850. Murray, P. D. F., and Huxley, J. S. (1925). Self-differentiation in the grafted limb bud of the chick. J. Anatomy 59, 379–384. Seeman, E. (2002). An exercise in geometry. J. Bone Min. Res. 17, 373–380. Bonadio, J., Jepsen, K. J., Mansoura, M. K., Jaenisch, R., Kuhn, J. L., and Goldstein, S. A. (1993). A murine skeletal adaptation that significantly increases cortical bone mechanical properties. Implications for human skeletal fragility. J. Clin. Invest. 92, 1697–1705. Kozloff, K. M., Carden, A., Bergwitz, C., Forlino, A., Uveges, T. E., Morris, M. D., Marini, J. C., and Goldstein, S. A. Brittle IV (2004) mouse model for osteogenesis imperfect IV demonstrates postpubertal adaptations to improve whole bone strength. J. Bone Min. Res. 19, 614–622. McBride, D. J. Jr., Shapiro, J. R., and Dunn, M. G. (1998). Bone geometry and strength measurements in aging mice with the oim mutation. Calcif. Tissue Int. 62,172–176. Szulc, P., Seeman, E., and Delmas, P. D. (2000). Biochemical measurements of bone turnover in children and adolescents. Osteoporosis Int. 11, 281–294. Nishida, S., Endo, N., Yamagiwa, H., Tanizawa, T., and Takahashi, H. E. (1999). Number of osteoprogenitor cells in human bone marrow markedly decreases after skeletal maturation. J. Bone Miner. Metab. 17, 171–177. Stenderup, K., Justesen, J., Eriksen, E. F., Rattan, S. I., and Kassem, M. (2001). Number and proliferative capacity of osteogenic stem cells are maintained during aging and in patients with osteoporosis. J. Bone Miner. Res. 16, 1120–1129. Oreffo, R. O., Bord, S., and Triffitt, J. T. (1998). Skeletal progenitor cells and aging human populations. Clin. Sci. 94, 549–555. Lips, P., Courpron, P., and Meunier, P. J. (1978). Mean wall thickness of trabecular bone packets in the human iliac crest: changes with age. Calcif. Tissue Res. 10, 13–17. Vedi, S., Compston, J. E., Webb, A., and Tighe, J. R. (1984). Histomorphometric analysis of dynamic parameters of trabecular bone formation in the iliac crest of normal British subjects. Metabolic Bone Dis. Relat. Res. 5, 69–74. Gilsanz, V., Gibbens, D. T., Carlson, M., Boechat, I., Cann, C. E., and Schulz, E.S. (1987). Peak trabecular bone density: a comparison of adolescent and adult. Calcif. Tissue Int. 43, 260–262. Balena, R., Shih, M–S., and Parfitt, A. M. (1992). Bone resorption and formation on the periosteal envelope of the ilium: A histomorphometric study in healthy women. J. Bone Miner. Res. 7, 1475–1482. Seeman, E. (2003). Periosteal bone formation – a neglected determinant of bone strength. N. Eng. J. Med. 349, 320–323. Ahlborg, H. G., Johnell, O., Turner, C. H., Rannevik, G., and Karlsson, M. K. (2003). Bone loss and bone size after the menopause. N. Engl. J. Med. 349, 327–334.
220 25. Brown, J. P., Delmas, P. D., Arlot, M., and Meunier, P. J. (1987). Active bone turnover of the cortico-endosteal envelope in postmenopausal osteoporosis. J. Clin. Endocrinol. Metab. 64, 954–959. 26. Foldes, J., Parfitt, A. M., Shih, M–S., Rao, D. S., and Kleerekoper, M. (1991). Structural and geometric changes in iliac bone: relationship to normal aging and osteoporosis. J. Bone Miner. Res. 6, 759–766. 27. Aaron, J. E., Makins, N. B., and Sagreiy, K. (1987). The microanatomy of trabecular bone loss in normal aging men and women. Clin. Orth. RR. 215, 260–271. 28. Van der Linden, J.C., Homminga, J., Verhaar, J. A. N., and Weinans, H. (2001). Mechanical consequences of bone loss in cancellous bone. J. Bone Miner. Res. 16, 457–465.
EGO SEEMAN 29. Manolagas, S. C. (2000). Birth and death of bone cells: basic regulatory mechanisms and implications for the pathogenesis and treatment of osteoporosis. Endocr. Rev. 21, 115–137. 30. Duan, Y., Turner, C. H., Kim, B. T., and Seeman, E. (2001). Sexual dimorphism in vertebral fragility is more the results of gender differences in bone gain than bone loss. J. Bone Miner. Res. 16, 2267–2275. 31. Duan, Y., Beck, T. J., Wang, X–F., and Seeman, E. (2003). Structural and biomechanical basis of sexual dimorphism in femoral neck fragility has its origins in growth and aging. J. Bone Miner. Res. 18, 1766–1774.
Chapter 14
The Cells of Bone Jane B. Lian and Gary S. Stein
University of Massachusetts Medical School, Department of Cell Biology, 55 Lake Avenue North, Worcester, MA 01655, USA
V. The Osteoclast: A Functionally Unique Cell for Physiologically Regulated Resorption of Bone Mineral VI. Perspectives References
I. Abstract II. Introduction III. Developmental Signals for Cartilage and Bone Tissue Formation IV. Osteogenic Lineage Cells
I. ABSTRACT
that accommodate requirements for its central role in mineral homeostasis. The structural and metabolic properties of bone tissue arise from the organization and interactions of functionally distinct cell populations. The principal cells which develop the various bone structures and mediate the specialized functions of the mammalian skeleton are: mesenchymal stem cells recruited for the development and expansion of skeletal tissues, chondrocyte lineage cells which form the template for endochondral bone development and long bone growth, osteoblasts which synthesize the bone matrix, osteocytes throughout the mineralized bone matrix which support bone structure, the protective bone surface lining cells, and the multinucleated osteoclasts which are responsible for the resorption of calcified tissue. Throughout life, bone tissue undergoes remodeling; a continual process of resorption and renewal. Fidelity of bone formation and remodeling necessitates exchange of regulatory signals among these cell populations and their progenitors. The basis for our current understanding of bone development and homeostasis has been and continues to be principally derived from the properties of skeletal cells, characterization of expressed genes and identification of their functional roles through genetic analyses (mouse models and human diseases). Two distinct progenitor stem cell lineages contribute to formation of the skeleton: mesenchymal stem cells for
Bone formation requires an exquisite interplay among distinct populations of cells and tissue compartments to support the metabolic and structural functions of the skeleton. In this chapter, the properties of pluripotential skeletal progenitor cells, chondrocyte, osteoblast, and osteoclast cell lineage populations are described. Emphasis is placed on recent advances in understanding the cellular and molecular mechanism regulating the progression of stem cells to differentiated phenotypes during endochondral and intramembranous bone formation and the responsiveness of cells to physiologic regulation and mechanical forces. The dynamic integration of signaling pathways and novel transcriptional regulators of cell growth and differentiation that are coordinated for development of the skeleton and the control of bone turnover has led to new dimensions for improving the diagnosis and treatment of skeletal diseases.
II. INTRODUCTION The skeleton functions both as a connective tissue in response to mechanical forces to adapt its structural features and as an organ in response to physiologic signals Dynamics of Bone and Cartilage Metabolism
221
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
222
JANE B. LIAN AND GARY S. STEIN
chondrogenic and osteogenic cells and the hematopoietic stem cells which give rise to osteoclasts. It is becoming increasingly evident that both chondrocyte, osteoblast, and osteoclast differentiation require a multistep series of events modulated by an integrated cascade of gene expression that initially supports proliferation and the sequential expression of genes associated with each component of bone formation and resorption. Equally significant are the growth factor, mechanical stimuli, and hormone-responsive regulatory signals, which mediate developmental competency for expression of genes associated with skeletal cell proliferation, differentiation, and repair. In this chapter, the current concepts in understanding molecular and cellular mechanisms regulating the progression of osteoblast and osteoclast differentiation and the functional activities of distinct cell populations will be explored. Within this context, a basis can be provided for improved diagnosis of skeletal disease and treatment that is targeted to specific cells in bone tissue.
III. DEVELOPMENTAL SIGNALS FOR CARTILAGE AND BONE TISSUE FORMATION The complexity of the multiple signaling pathways that converge to form the skeleton is first appreciated by the formation of skeletal elements through two distinct developmental pathways. Intramembranous bone formation results from initial condensation of mesenchyme with direct differentiation of the cells into osteoblasts and gives rise to flat bones, for example, that comprise the cranium. Growth of this tissue continues from progenitor cells within the periosteum covering the bone surface or the cranial sutures. The process of endochondral bone formation (EBF) for the development and growth of long bones and vertebrae involves initial formation of a cartilage tissue from condensing mesenchyme. Perichondrium develops around a cartilage anlagen and provides a future source of progenitor cells for differentiation to chondrocytes. In the core of the anlagen, chondrocytes become hypertrophic producing a calcified cartilage matrix which subsequently is resorbed and replaced by bone tissue. Vascular invasion in hypertrophic cartilage leads to formation of the growth plates with clearly defined zones of proliferating and maturing chondrocytes. The growth plates spatially define growing ends of the bone and the marrow environment which houses hematopoietic tissue for blood cell differentiation and stromal cell progenitors for bone growth. Osteogenic cells are recruited to the cores of calcified cartilage for formation of trabecular bone spicules, concomitant
with the resorption of calcified cartilage. With induction of ossification, a periosteum develops over cortical bone to provide a continual source of progenitors for appositional growth to complement the endochondral long bone growth.
A. Inductive Events for Skeletal Development: Fibroblast Growth Factors, Bone Morphogenetic Proteins, and Wnt Signaling Different regions of the skeleton arise from three distinct embryonic zones, influenced by diffusible factors which induce the necessary cellular phenotypes for tissue formation [1, 2]. The major signaling proteins involved in segmentation of the vertebrate body are Hox genes which define the positions where bone structures will develop [3, 4]. Migration of neural crest cells from the ectoderm to the cephalic mesoderm provides progenitor cells for development of cranial and facial skeletal structures. Limb formation is initiated by proliferation of mesenchymal cells in the lateral plate mesodermal layer to form the appendicular skeleton. Cells of the paraxial mesoderm undergo condensation and segmentation to form somites under the regulation of the cell surface receptor Notch1 and its ligands, delta and serrate [5]. Sclerotome cells from somites are then induced to form cartilage by the cytokine sonic hedgehog (Shh) secreted by the notochord, producing the axial skeleton (spine, sternum, and ribs) [6, 7]. Importantly, Shh was shown to regulate mesenchymal cell recruitment into the osteogenic lineage in vitro [8]. Considering these different embryonic developmental programs of the mesoderm to form either flat bones directly (e.g., calvarium) or cartilage anlagen that will become bone (e.g., limbs and vertebrae), it should be recognized that a distinct osteoprogenitor may divert from a stem cell at these skeletal sites. The formation of skeletal tissues from the condensations of the mesenchymal progenitor cells at a specific site is determined by epithelial–mesenchymal interactions that control shape and size of the limbs through secretion of regulatory factors and target transcription factors. A group of epithelial cells, the apical ectodermal ridge (AER), caps the limb buds and secretes growth factors that pattern the limb. The clustered Dlx gene family encoding homeodomain proteins, for example, is expressed in the AER of the limb bud regulating the proximal–distal pattern of outgrowth as a regulatory network. These genes also specify the spatial and temporal formation of the craniofacial skeleton [9–11]. It is now being recognized that the regulatory factors which induce and control development of skeletal structures remain as key signals of bone renewal and repair. These include members of the TGFβ/BMP superfamily that induce
Chapter 14 The Cells of Bone
223
mesenchyme condensation for bone formation, Wnt proteins which contribute to the formation of the bone axes and fibroblast growth factors (FGFs) that are essential for the earlier stages of limb bud outgrowth [12–17]. 1. Fibroblast Growth Factor Receptors and Ligands
The FGF receptors, FGFR1, FGFR2, and FGFR3 and the FGF ligands are expressed throughout skeletal cell populations and contribute to the regulation of progenitor and differentiated populations of cells during both intramembranous and endochondral bone formation [18–21]. FGFR1 is expressed in limb mesenchyme and in osteoprogenitor cells at the osteogenic front separating the nonosseous suture tissue between the ossification plates of the calvarial bone tissue. An FGFR1 activating mutation leads to premature fusion of craniofacial structure (craniosynostoses), while dominant-negative forms of FGFR1 will inhibit calvarial suture fusion [22, 23]. Mutations in FGFR2 are responsible for Apert syndrome, severe craniosynostoses [22], and Pfeiffer and Crouzon disorders [24]. FGFR2 is expressed as two variants, FGFR2b is required for limb outgrowth, while FGFR2c is required for osteoblast maturation [25]. Interestingly, different point mutations result in distinct phenotypic alterations in gene expression of the osteoblasts conveying the mutation [26, 27]. FGFR3 expression is initiated as chondrocytes differentiate long bone growth and functions as an important negative regulator of endochondral bone formation. Various knockin mutations of this gene lead to severe dwarfism [28]. Together with FGFR3, PTHrP signals coordinate cartilage and bone formation [29]. Complete ablation of FGFR3 leads to embryonic skeletal overgrowth [30], but also an osteopenia phenotype in the post-natal mouse [31]. Specific FGF ligands control limb outgrowth [14, 21, 32, 33]. Beginning in the lateral plate mesoderm, FGF10 in the presumptive limb regions induces FGF4 and FGF8 in ectodermal layers to become the AER. FGF2, FGF4, FGF8, and FGF9 all contribute to increasing proliferation of mesenchymal cells, but FGF2 and FGF18 appear to be important negative regulators of chondrocyte proliferation. Mice lacking FGF18 have defects in both chondrogenesis and osteogenesis [34, 35]. FGF2 was initially isolated from the cartilage matrix [36] and later identified in periosteal cells and osteoblasts [37]. FGF2 supports osteoblast growth and differentiation [38] and contributes to osteoblast survival [39, 40]. Mice lacking FGF2 have decreased bone mass [41]. These examples illustrate how FGF signaling through multiple receptors and ligands and with specific activities on different cell populations control expansion of embryonic cartilaginous tissue for endochondral bone formation (illustrated in Figure 1).
Figure 1 Regulatory factors controlling chondrogenesis and endochondral bone formation. The condensation of mesenchymal cells committed to chondrocyte differentiation is characterized by expression of the transcription factors Sox5 and Sox6. PTHrP expressed in the perichondrium stimulates chondrocyte proliferation. The increase in growth rate delays maturation of the hypertrophic chondrocytes which secrete IHH, thereby negatively regulating IHH. IHH, which diffuses into the epiphysis, stimulates PTHrP activity, but also increases bone collar formation in conjunction with BMPs. In this manner, growth plate events are coordinated with bone formation during development. Runx2 which is expressed in MSCs, must be down-regulated, mediated by Nkx3.2 for recruitment of progenitors into the chondrogenic lineage. Runx2 is then up-regulated later during endochondral bone formation in hypertrophic chondrocytes. Runx2 transcriptional activity is tightly regulated by co-regulatory protein interactions, such as the negative regulator histone deacetylase-4 (HDAC4). This interaction may be required to attenuate Runx2 activity and expression of its target genes (VEGF and MMP9) during FGF. RZ, resting zone; PZ, proliferating zone; HZ, hypertophic zone.
2. The BMP Family of Proteins and Skeletal Development
Bone morphogenetic proteins (BMP) are multifunctional growth and differentiation factors (GDFs) that support the development of many tissues including cartilage and bone. BMP signaling is required at an early stage of skeletal development for formation of the AER and dorsal–ventral patterning of the limb [42]. Their specific roles in the skeleton and regulation of BMP activity through antagonists have been recently reviewed [13, 43]. Briefly, BMPs are members of the TGFβ superfamily which function through specific receptors that are distinct but have overlapping functions [16, 44]. Genetic studies that have modified BMP receptors 1A and 1B have identified distinct activities. For example, constitutively active forms of BMP1A and BMP1B promote chondrogenesis and osteogenesis, but only the dominant-negative form of BMPR1B inhibits these events [45]. Complete null mutation of the BMPR1B Type I receptor revealed defects mainly in the
224 appendicular skeleton, with marked reduction in proliferation of the prechondrocytic population and subsequent chondrocyte differentiation [46]. The receptor signal is transduced by phosphorylating intracellular BMP- and TGFβ-specific receptor Smad proteins (R-Smads), which results in dissociation from the receptor, followed by formation of heterodimers with a DNA-binding Smad4 (a common-mediator Smad) that enters the nucleus and is competent to either target gene promoters or form multimeric complexes on gene promoters with other transcription factors. Antagonistic or inhibitory Smads (I-Smads) contribute to regulating this pathway (reviewed in reference [47]). BMP ligands are expressed in mesenchymal condensations functioning in chondrogenesis and later are expressed in specific cartilaginous regions of the limb (reviewed in reference [46]). BMP2, 4, and 7 are expressed in the perichondrium, while BMP6 is found in the prehypertrophic and hypertrophic zones of the growth plate. Characterizing specific BMP activities has been challenging due to early embryonic lethality as shown for BMP2 and BMP4. However, conditional knockouts are being characterized. BMP7 (also designated OP-1, osteogenic protein-1) is essential for appendicular skeletal development and is a strong anabolic factor for cartilage [46, 48]. GDF-5 is essential for normal joint formation and development of the appendicular skeleton [49]. In the postnatal skeleton, several BMPs continue to function in stimulating chondrocyte proliferation (e.g., BMP7), but will inhibit terminal differentiation and require regulation by BMP antagonists [50]. BMP activity is regulated by several antagonists, one being Noggin which is highly expressed in cartilage tissue. Disruption of this control results in early skeletal malformations in the mouse [51, 52]. BMP2 and BMP6 can stimulate cartilage differentiation to the hypertrophic phenotype [53]. BMP2 and BMP4 are potent osteoinductive growth factors, a property mediated by the induction of the Dlx and Runx transcription factors as immediate early targets of BMP2 [54–57]. A significant finding relevant to the osteogenic effects of BMPs is the interaction of Smad complexes with Runx2 (Cbfa1/AML3), a transcription factor essential for bone formation [58–60]. Recent studies indicate an important positive regulatory loop between BMP2 and the Runx2 transcription factor. Runx2 is induced within a few hours of BMP2 treatment [54]. BMP2 and Runx2 together then have synergistic effects in promoting osteogenesis [61, 62]. Runx2 is expressed in early embryogenesis [63, 64], followed by an up-regulation of Runx2 in late stages of bone development [65, 66], suggesting that this factor may be important in early specification of the phenotype, as well as having a significant role for sustaining osteoblast differentiation
JANE B. LIAN AND GARY S. STEIN
(see Section IV). How crosstalk between the BMP and Runx2 signals is regulated remains a question, but such regulation provides an example of “feed forward” loops necessary to support development of the skeleton. BMPs induce a spectrum of other transcription factors essential for differentiation of skeletal tissues, including Sox genes which are required for chondrogenesis [67, 68], homeodomain transcription factors which pattern the skeleton and promote bone formation [54, 69, 70], and Osterix, which, like Runx2, is essential for bone formation [71]. Thus, the chondrogenic and the osteogenic activity of BMPs are related to induction of specialized transcriptional regulators of cell differentiation. 3. Wnt Signaling and Skeletal Development
The Wnt family of 19 secreted cysteine-rich glycoproteins regulate numerous developmental processes including cell polarity, cell differentiation, and migration [72]. The Wnt signaling pathway is a recently appreciated key regulator of early skeletal development. Wnt proteins bind to and activate receptor complexes consisting of the Frizzled family of G protein-coupled receptors and the low-density lipoprotein (LDL) receptor-related proteins. Wnt signaling is transduced through three pathways [72]. Activation of the canonical pathway results in stabilization of β-catenin by inhibiting its phosphorylation involving casein kinase 1 and glycogen synthase kinase 3 within a protein complex, which prevents the targeting of β-catenin for ubiquitination/ proteasome degradation. Subsequently, β-catenin is translocated into the nucleus to form heterodimers with the TCF1 or LEF transcription factors for expression of Wntresponsive genes. In the absence of nuclear β-catenin, TCF/LEF is associated with transcriptional corepressors and suppresses Wnt target genes. β-catenin is thus a critical regulatory step of canonical Wnt signaling and for embryogenesis. The other two pathways, the planar cell polarity and the calcium pathways are not as well defined in the mammalian tissues in the skeleton. The Wnt signaling pathway is regulated by several antagonists. DKK (Dickkopf) interacts with the frizzled receptor complex preventing transduction of the Wnt signal, while a class of secreted frizzled related proteins (SFRPs) interacts with Wnt proteins sequestering them from interaction with frizzled receptors, as does WIF (Wnt inhibitory factor-1) and Cerberus [73]. While specific Wnt proteins that activate either the canonical β-catenin pathway or noncanonical pathways have been identified, additional studies are necessary to clarify the specific roles of Wnt factors that may operate through multiple pathways. Both gain- and loss-of-function mutations in specific Wnt proteins, as well as forced expression of Wnt factors in limb bud cultures, have revealed the significance
Chapter 14 The Cells of Bone
of Wnt signaling in regulating skeletal development [17, 72, 74]. Wnt10a misexpression in the developing chick limb identified its importance for AER formation. Wnt3a knockout mice have a skeletal phenotype as Wnt3a is required for somite formation [75]. Recently, a rare human genetic disorder tetra-amaelia, characterized by absence of all the limbs has been linked to mutation in Wnt3 [76]. The expression of several Wnts (Wnt4, 5a, 5b, 6, 11 and 14) during limb development in the chick suggests key functions in initial stages of skeletal development [77, 78]. Misexpression of Wnt14 identified its role in induction of joint interzone [79]. Wnt5a and Wnt5b coordinate the pace and transitions between chondrocyte zones [80–82]. Wnt5b, which is expressed in the prehypertrophic chondrocyte zone, as well as in joints and perichondrium, delays hypertrophy, while Wnt4 blocks initiation of chondrogenesis, but accelerates hypertrophy [83–85]. Therefore, the developing limb and endochondral bone formation which both rely on the regulation of the proliferation, maturation, and spatial organization of chondrocytes, are processes highly dependent on Wnt signals. The importance of canonical Wnt signaling in skeletal development and postnatal bone formation is now established. By conditional ablation in mice, β-catenin has been identified as a key regulator of formation of the apical ectodermal ridge (AER) and of the dorsal–ventral axis of the limb [86]. In a series of genetic studies, inactivation of β-catenin in mesenchymal lineage cells results in severe loss of bone from inhibited osteoblast maturation and increases osteoclast differentiation [87–89]. By increasing Wnt signaling, e.g., by ectopic or by expressing a stabilized form of β-catenin, produces enhanced ossification and suppression of chondrogenesis [88–90]. More direct effects on formation and turnover of the mature skeleton have been revealed by an activating mutation (gain-of-function) in the Wnt co-receptor LRP5 (Low-density lipoprotein Related Protein 5), resulting in the high bone mass trait in humans [91, 92], a phenotype reproduced in the mouse model [93, 94]. Consistent with this phenotype, the LRP5 loss-of-function mutation leads to an osteopenia accompanied by fractures in humans causing osteoporosis pseudoglioma syndrome (OPPG) [95, 96]. This low bone mass phenotype is recapitulated in the mouse [97]. Inactivation of the Wnt antagonist sFRP1 in mouse, enabling increased Wnt signaling, also resulted in a high bone mass in the mouse adult skeleton with a delay in bone loss in ages 7–9 months [98]. These findings underscore the significance of Wnt signaling in maintaining bone mass in the adult skeleton. Expression studies suggest direct roles of several Wnts in bone formation and maintaining bone mass. Expression of Wnt10b in mouse marrow results in increased metaphyseal bone volume, and bone that is
225 mechanically stronger. In addition, the female mice are protected from bone loss due to estrogen deficiency [99]. Consistent with this observation, the null mutation of Wnt10b revealed decreased trabecular bone in the Wnt10b−/− mouse. Wnt10b is expressed in pre-adipocytes and the mechanism of increased osteogenesis by Wnt10b relates to the inhibition in development of fat tissue in vitro and in vivo in the transgenic mice, which had 50% less white fat and virtually no brown fat that generates body heat [100]. These studies suggest that Wnt signaling is not only necessary for bone development, but supports maintenance of bone tissue functions in the adult skeleton through several mechanisms. We have discussed above three key signaling centers for development of the skeleton. It is important to recognize that the molecular mechanisms underlying the signaling pathways induced by secreted factors for skeletal development point to specific transcriptional regulators that control pattern formation and/or guide the mesenchymal cell to the chondrogenic versus the osteoblast lineage (selected key transcription factors are shown in Table I). Furthermore, the importance of coordinated activities among these pathways is being realized and will be the focus of new investigations. Interactions between FGF, BMP, and Wnt signals occur primarily at the level of regulating secreted factors for coordinating the timing of developmental events.
B. Regulation of Endochondral Bone Formation The coordination of chondrogenic differentiation with bone formation events is tightly regulated through multiple signaling pathways as discussed above. Both selective expression of secreted factors in chondrogenic subpopulations and feedback loops control the proliferation and maturation of chondrocytes and the pace of endochondral ossification [101]. Two essential regulatory proteins include Indian hedgehog (Ihh), a mammalian homolog of the Drosphila hedgehog secreted factor and parathyroid hormone-related peptide (PTHrP) which is secreted by many tissues, but with specialized autocrine and paracrine activities for regulating endochondral bone formation. PTHrP is abundantly expressed in the periarticular perichondrium and diffuses towards the prehypertrophic zones. PTHrP functions through the PTH(PTHrP) G protein-coupled receptor which is expressed at high levels in pre- and hypertrophic chondrocyte zones. These events are illustrated in Figure 1. Human mutations in this receptor characterize several chondrodysplasias. In mice deficient in PTHrP, chondrocyte maturation is accelerated while overexpression of PTHrP
226
JANE B. LIAN AND GARY S. STEIN Table I.
Transcription factor Pax1, Pax 2 Nkx3.2
T-box genes (Tbx4, Tbx5) Sox5, Sox6 HIF-1α Sox9
Transcriptional Regulators of Early Skeletal Development Role
• • • • • • • • • •
Reference
Normal digit formation Sclerotome formation Chondrogenesis Axial skeletal development Limb bud outgrowth Growth plate organization Notocord ECM sheath formation and nucleus pulposus development Required for epiphyseal cartilage ECM synthesis Chondrogenic lineage determination in the cranial neural crest Essential for cartilage function
Foxc1 Homeodomain Proteins (Msx1, Msx2; Dlx1–6) Gli factors
• Progression of intramembranous bone formation • Patterning and cell differentiation
Runx2
• • • •
TCF/LEF-1
• Mediator of Shh for sclerotome development • Target of Ihh and PTHrP signal • Required for normal endochondral bone formation Present in pre-mesenchymal condensations, but function unknown Expressed in chondrocytes at the growth plate Mediator of canonical Wnt signaling Essential for differentiation of hypertrophic chondrocytes and osteoblast
stimulates chondrocyte proliferation, thereby inhibiting cartilage differentiation and delaying bone formation [102–104]. These studies identify PTHrP as a negative regulator of the hypertrophic chondrocyte phenotype (i.e., terminal differentiation). Ihh secretion in the hypertrophic zones serves multiple functions in coordinating the events of expansion of the proliferative zone and regulating osteoprogenitor differentiation for maturation of the bone collar [105]. Ihh increases PTHrP expression and consequently chondrocyte proliferation. Ihh therefore maintains the chondrocyte phenotype in the resting and proliferative zones [101, 104, 106, 107]. PTHrP at the growth plate by inhibiting the differentiation of proliferating chondrocytes to hypertrophic chondrocytes inhibits Ihh synthesis [108]. This negative loop maintains a steady rate of chondrocyte maturation and ensures a synchronous timing of long bone growth. In addition to PTHrP, prostaglandin, thyroid hormone, retinoic acid, and growth hormone influence chondrocyte proliferation and maturation to the hypertrophic cell. Reserve zone and proliferating chondrocytes appear to be the target of T3 in vivo. T3 inhibits clonal expression of chondrocytes in vitro, but induces markers of hypertrophic resting chondrocytes [109, 110]. The prostaglandin, PGE2, also inhibits chondrocyte differentiation [111]. Retinoic acid (RA) is required for normal development [112, 113] and, interestingly, for repeated regeneration of deer antler tissue [114]. However, nonphysiologic levels of RA are detrimental to endochondral bone formation as a result of
Lu et al., 1999 [492] Herbrand et al., 2002 [493] Murtaugh et al., 2001 [121] Tribioli & Lufkin, 1999 [117] Takeuchi et al., 2003 [494] Smits et al., 2004 [495] Smits & Lefebvre, 2003 [496] Pfander et al., 2003 [497] Mori-Akiyama et al., 2003 [68] Akiyama et al., 2002 [124] Bi et al., 1999 [498] Rice et al., 2003 [499] Bendall & Abate-Shen, 2000 [500] Depew et al., 2002 [9] Miao et al., 2004 [501] Buttitta et al., 2003 [502] Vortkamp et al., 1996 [503] Jemtland et al., 2003 [504] Smith et al., 2005 [340] Stricker et al., 2002 [130] Logan and Nusse, 2004 [72] Hsu et al., 1998 [505]
RA inhibition of chondrogenesis [115]. In contrast, growth hormone enhances chondrogenesis and osteogenesis, but oversecretion will result in tissue with inferior mechanical properties [116]. Thus, EBF requires not only developmental signals, but also hormones and cytokines that promote transcriptional control of target genes for regulation of growth plate events. Skeletal development and progression of endochondral bone growth at the hypertrophic stage of cartilage maturation producing a calcified matrix involve coupled positive and negative transcriptional control of gene expression (Table I). This is illustrated by two transcription factors expressed in undifferentiated mesenchymal cells, Nkx3.2, the gene associated with the bagpipe mutation in mouse [117] and Runx2/Cbfa1 [118]. Runx2, the transcription factor essential for osteoblast differentiation, is expressed in prechondrogenic mesenchymes [66, 119]. Nkx3.2 is a strong repressor transcription factor and one of the earliest mediators of chondrocyte commitment [120, 121]. Repression of Nkx3.2 is critical for development of the axial skeleton and position of the jaw joint [117, 122, 123]. Through a series of derepression events, Nkx3.2 allows for activation of Sox 9, a requirement for chondrocyte differentiation [124]. As the mesenchymal cells are recruited into the chondrogenic lineage by Nkx3.2, Runx2 becomes down-regulated through direct transcriptional control mediated by an Nkx3.2 response element in the Runx2 gene [125]. Thus Nkx3.2 initiates a cascade of events for both suppressing osteogenesis and activating chondrogenesis.
Chapter 14 The Cells of Bone
In situ hybridization studies in mouse models have led Helms and colleagues to propose that cell fates for cartilage and bone are specified by a hierarchy of Sox9 and Runx2 expression [126]. Runx2 is then re-expressed in the hypertrophic zone of the growth plate following the down-regulation of Nkx3.2. Runx2 activates vascular endothelial growth factor and matrix metalloproteinase 9, both essential factors for vascular invasion and recruitment of both resorbing cells to remove the cartilage matrix and osteoprogenitors [119, 127–129]. These findings suggest Runx2 may coordinate calcification with recruitment blood vessels and resorption/degradation of the calcified cartilage matrix. Forced expression of Runx2 in chondrocytes in transgenic mouse models shows that Runx2 supports hypertrophic chondrocyte differentiation [130–134], thus confirming the skeletal abnormalities of Runx2 null models as well as mutations in the Runx2 DNA-binding partner, CBFβ [63, 64, 66, 135, 136]. Evidence is accumulating for a key regulatory role of Runx2 to support chondrocyte maturation, perichondral invasion, and osteogenesis in cartilaginous tissue ([137] and reviewed in [138]). Synthesis of a cartilage extracellular matrix competent to calcify is a pivotal stage of long bone growth. Hypertrophic chondrocytes have unique properties defined by morphology, function, and by the expression of proteins with specific consequences restricted to this cell for production of a calcified matrix, such as type X collagen, small proteoglycans, and several noncollagenous proteins that are abundant in bone. The mineralized calcified cartilage matrix is requisite for resorption by chondroclasts/osteoclasts. Hypertrophic chondrocytes have a limited life span and are destined for apoptosis. These phenotypic changes in the chondrocytes are necessitated by the requirements for tissue organization, where the mineralized matrix at the growth plate eventually recruits the bone marrow stromal cells which will produce trabecular bone to accommodate lengthening of the bone during growth.
IV. OSTEOGENIC LINEAGE CELLS A. Mesenchymal Stem Cells: The Promise for Treating Skeletal Disorders 1. Stem Cell Populations and Properties
Advances in stem cell- and bone marrow-derived mesenchymal stem cell biology are increasing utilization of these cells for correcting genetic disorders and tissue repair.
227 Embryonic stem cells (ESC) are derived from the inner cell mass of blastula-stage embryos and express genes characteristic of the early blastocyst [139]. ESCs are defined by their infinite self-renewal capacity and will give rise to nearly every tissue and cell type in the body dependent on the appropriate in vivo environment and stimulus. Intact embryoid body tissue or micromass culture of ESCs will undergo chondrogenesis [140] and ESCs can be directed into osteogenic cells in the presence of ascorbate, 1,25(OH)2D3, β-glycerol phosphate [140–143]. Procedures have been detailed in the Methods in Enzymology series [144–146]. In the adult, stem celllike populations are being isolated from various tissues. Bone marrow contains the hematopoietic stem cell, a wellcharacterized self-renewing population that gives rise to every blood cell type. Within the bone marrow is a rare and rather elusive population of multipotential adult progenitor cells (MAPC) that can form endodermal, mesodermal, and ectodermally derived cell types; for example, endothelial cells, hepatocytes, and neurons in vitro [147, 148]. When these cells are injected into blastocytes, they contribute to tissues that include brain, long muscle, kidney, intestine, blood, skin, retina, spleen, and bone marrow in mice [149]. A side population (SP) of cells is characterized by their ability to extrude Hoechst dye and are enriched for hematopoietic stem cells ( HSC) and will differentiate into chondrocytes and osteoblasts [150]. Another population of stem-like cells of mesenchymal origin persists in the marrow. This is the mesenchymal stromal cell (MSC) which is an adherent cell with a limited lifespan (no more than 50 doublings) and has the ability to differentiate into multiple lineages, including adipocytes, chondrocytes, and osteoblasts. Osteogenic progenitor cells can also be derived from the periosteum. The stem cell-like properties of cells isolated from adult bone marrow and tissues are a question of recent debate as there is evidence for their plasticity (that is, the ability to acquire a mature phenotype by cell–cell fusion) or to transdifferentiate when exposed to a different microenvironment [151]. Progression of the most primitive pluripotent cell to the undifferentiated multipotential mesenchymal cell is not understood. Assurance of their stem cell activity must be defined by transplantation to a recipient animal examining their differentiation long term [152]. Using characterization of hematopoietic stem cells as a paradigm, several cell surface markers as well as antibodies generated to bone marrow stromal cell populations are used to identify presumptive stem cells from bone marrow. These reagents, including for example Stro-1, Sca−/+, SB10, have the potential for both recognition and purification of skeletal-related progenitor cells [153–158].
228
JANE B. LIAN AND GARY S. STEIN
Maintaining the stem cell-like properties in vitro has been challenging. Stem cells by their nature are generally in a noncycling (G0) stage of the cell cycle. ES cells can be propagated in culture on a feeder layer of mouse embryonic fibroblasts or without feeders in the presence of leukemia inhibitory factor (LIF), required for maintenance of mouse, but not human, ES cells [139]. Oct4 (a POU homeodomain protein) and nanog (a new homeodomain protein) are also requirements for a self-renewal of ES cells [159, 160]. Progenitor cells are responsive to a broad spectrum of regulatory signals that both maintain their pluripotent properties and mediate commitment to osteoblast differentiation and progression of phenotype development. When suspensions of marrow cells are plated in vitro, clonal colonies of adherent fibroblasts are formed; each derived from the single cell which has been designated as the colony-forming fibroblastic unit or CFU/F formation [161]. A proportion of these cells have high proliferative and differentiation capacity and exhibit characteristics of stem cells when transplanted in the closed environment of a diffusion chamber [162]. The CFUs are readily differentiated to osteogenic colonies in appropriate medium. The osteoprogenitor appears to have limited selfrenewal capacity and also, in contrast to the stem cell, has a capacity for extensive proliferation. Formation of CFUs requires the presence of hematopoietic cells [163] and recent studies have raised provocative implications of a direct influence of immune cells in contributing to osteogenic differentiation [164]. 2. From Stem Cell to Osteoprogenitor
Commitment of colony-forming units to the osteoprogenitor phenotype is regulated by several signaling proteins that include cytokines, growth factors, and hormones which influence growth and differentiation. Leukemia inhibitory factor, which maintains stem cell populations by inhibiting their differentiation, has also been reported to have osteogenic activity by enhancing differentiation of preosteoblasts [165, 166]. Platelet-derived growth factor and epidermal growth factor have been identified as important in stimulating expansion of the CFU/F [163, 167] and contribute to osteoprogenitor cell survival [168]. The fibroblast growth factor (bFGF) and transforming growth factor β1 (TGF-β1) are potent mitogens for periosteal osteoprogenitors and marrow stromal cells [169–171] and are expressed and produced by osteoblast lineage cells. These growth factors are stored in the bone extracellular matrix and thus provide a local mechanism for stimulating proliferation of progenitors in the bone microenvironment [172, 173]. The osteoinductive effects of the bone morphogenetic proteins (BMPs)
are complex and dependent on the specific BMP, concentration dependency, and the progenitor cell phenotype [43]. The activity of BMPs is regulated by production of inhibitors. Noggin, for example, is a glycoprotein that binds selectively to BMPs and inhibits stromal cell differentiation in vitro [174]. Tob proteins also inhibit BMP activity, but through a different mechanism, by enhancing the inhibitory Smad–receptor interactions to repress BMP signaling [175, 176]. BMP-2 rapidly induces osteoblast differentiation in marrow stromal cells [177], a number of pluripotent cell lines [178], and is competent to transdifferentiate the mouse myogenic cell line C2C12 into osteoblasts [179]. BMP2 is most widely used as an inducer of osteogenesis and is currently in clinical trials to augment fracture repair [180]. Among the hormones most influencing early osteogenesis are parathyroid hormone (PTH) which stimulates growth of osteoprogenitor populations [181], and glucocorticoids that have been shown to stimulate the growth and differentiation of CFUs and osteoprogenitors from human and rat, but not the mouse [182]. However, glucocorticoids have a marked inhibitory effect on bone formation by decreasing the number of mature osteoblasts and their activity through inhibition of insulin-like growth factor I [183, 184]. Increased bone formation by intermittent PTH is due to the stimulation of proliferation and differentiation of osteoprogenitor cells in bone marrow (reviewed in reference [185]). Recently it was shown that PTH strongly up-regulated amphiregulin, an epidermal growth factor family member [186]. Amphiregulin is a potent growth factor for preosteoblasts, although this growth factor inhibits osteoblast differentiation. In addition to growth-stimulating effects, PTH arrests cell cycle progression [187]; thereby, PTH promotes osteoblast maturation. Thus, PTH contributes to progenitor cell expansion in part through induction of the growth factors and further functions as an anabolic bone agent countering the inhibitory effects of growth factors on osteoblast differentiation. Parathyroid hormone-related peptide (PTHrP) that also binds to the PTH receptor, functions as a cellular cytokine regulating osteoblast growth for differentiation in development (reviewed in reference [188]). The significance of many of these growth factors and morphogens in regulating progression of the pluripotent stem cell and multipotential mesenchymal cell to the committed osteoprogenitor and finally to recognizable osteoblasts, is appreciated from mouse models, in which the PTHrP gene and the PTH/PTHrP receptor for these proteins have been ablated with severe consequences on formation of the skeleton [104, 189, 190]. Recent studies revealed PTHR-1R as a signature for definitive osteoprogenitors [191].
229
Chapter 14 The Cells of Bone 3. Molecular Mechanisms Regulating Allocation to Osteoblastic Phenotype
Much attention has been given to lineage allocation of mesenchymal stem cells in marrow between the osteoblast and adipocyte bone. While debate still occurs as to whether the inherent osteogenic potential of the MSC osteoprogenitor number in marrow declines with aging, factors influencing lineage direction have been clearly defined. Notch, an important signaling receptor that controls cell fate determination during development, has opposing effects on mesenchymal cells, suppressing chondrogenic differentiation [192], but stimulating osteoblast differentiation [193]. Menin, a product of the multiple endocrine neoplasia type 1 (MEN1) gene, was recently identified as a requirement for MSC commitment to osteoblasts in the null mouse [194]. PTHrP interacts with the BMP2 by enhancing BMP1A receptor expression pathway for osteoblastogenesis, but decreases adipogenesis [195]. Commitment of a stem cell to a phenotype is regulated by cell shape and cytoskeleton changes which involve Rho GTPase activity. A dominant-negative RhoA promotes a round shape leading to adipocyte differentiation,
Figure 2
while a constitutively active RhoA induced the osteogenic phenotype independent of cell shape [196, 197]. Physical forces on the MSC appear to be a significant component for osteoblast allocation as microgravity inhibits osteoblast colony formation of human MSCs and increases adipocytes [198, 199]. Finally, transcriptional regulators of gene expression have potent and direct effects in modifying cellular phenotypes. A number of key studies have defined master genes that direct a pluripotent cell to different lineages. Adipogenesis is promoted through the activities of PPARγ and CEBPα [200, 201], chondrogenesis requires Sox9 [202] and osteoblast maturation requires Runx2 [65, 66] and Osterix [71, 203] (Fig. 2). Inhibitory transcription factors, such as GILZ or retinoic acid, can block adipogenesis [204], thereby increasing a pool of progenitors for osteoblast differentiation. The potency of Runx2 in directing osteogenic commitment is provided by numerous studies that show Runx2 expression can activate bone phenotypic genes in pluripotent cells and redirect a committed premuscle cell into the osteoblast lineage [57, 205] or inhibit the adipogenic phenotype [206]. Conversely, activation of PPARγ in osteoblasts will down-regulate
Regulation of osteoblastogenesis. (A) Selected stages of osteogenic lineage cells are illustrated indicating secreted factors (BMPs, TGFβ, FGFs, and Wnt proteins) and hormones most influencing progression of the osteocyte from the earliest stem cell. In the lower panel of (A), transcriptional control of lineage allocation of the mesenchymal stem cell to nonosseous phenotypes is shown. (B) Markers frequently used to characterize stages of maturation include collagen type I, alkaline phosphatase, bone sialoprotein, osteopontin, and osteocalcin. (C) Examples of transcription factors regulating progression of osteoblast differentiation are indicated by direction of the arrows.
230
JANE B. LIAN AND GARY S. STEIN
Runx2-mediated transcription of bone phenotypic genes [207]. Even more significant, PPARγ-deficient ES cells not only failed to become adipocytes, but spontaneously differentiated to osteoblasts [208]. The Runx2 transcription factor is expressed in the early pluripotent mesenchymal cell prior to expression of the osteoblast phenotype, and increases during osteoblast differentiation, consistent with its genetic role in maturation of the osteoblast phenotype [63, 65]. Runx2 protein is also retained during mitosis and may provide a mechanism for regulating genes that retain the osteogenic lineage properties of dividing cells [209]. The in vivo significance of Runx2 in early commitment to the osteoblast lineage is indicated by the evidence that mesenchymal progenitor cells from Runx2-null mice differentiate more towards chondrocytes and adipocytes [210]. Importantly, Runx2 may function as an inhibitor of proliferation of progenitors, thus providing a mechanism for regulating the transition from growth to a post-proliferative stage as a component of cellular commitment to osteogenic lineage cells [211]. We are now beginning to reach an understanding of the complexity of factors required to support expansion of a progenitor cell and the signals which must be initiated for stem cells to acquire an osteogenic property. With the onset of new discoveries, considerations for how the different regulatory proteins can be applied for a therapeutic strategy must take into account their effects on a spectrum of diverse activities within different pathways.
B. Osteoblasts: The Bone-forming Cells and Gatekeepers of Hormonal Activity The mature surface osteoblast is defined by the biosynthesis and organization of the bone extracellular matrix. The primary function of osteoblasts on active bone-forming surfaces is the production of a bone matrix, contributing to expansion of bone volume by laying down osteoid and secreting factors that facilitate mineral deposition. An equally important function of osteoblasts and the preosteoblasts which lie in close proximity to the bone surface is their responsiveness to endocrine factors and the production of paracrine and autocrine factors for the recruitment of osteoprogenitors, the growth of preosteoblasts, and the regulation of osteoclastic resorption of the mineralized bone matrix. Osteoblasts and osteocytes have receptors for Key regulators of bone turnover, which include cytokines, parathyroid hormone, 1,25(OH)2D3, and sex steroids. These features provide mechanisms for mediating the coupling of osteoblast and osteoclast activities (see Section V).
The in vitro study of primary osteoblasts from several species has allowed molecular analysis of expressed genes and transcription factors regulating the differentiation of osteoblasts in relation to progressive formation of the bone matrix [212]. These studies, carried out over a period of several weeks, provide a definition of discrete stages of maturation of the osteoblasts (Fig. 2). The proliferating preosteoblast stage is engaged in synthesis of growth factors and matrix components of woven bone. When the osteoprogenitor or preosteoblast ceases to proliferate, a key signaling event occurs for development of the large cuboidal surface osteoblast from the spindle-shaped osteoprogenitor. The osteoblast vectorially secretes type I collagen and specialized bone matrix proteins towards the mineralizing front of the tissue. The matrix maturation stage marks the postproliferative osteoblast expressing maximal levels of alkaline phosphatase to render the matrix competent for mineralization. Type I collagen continues to accumulate, together with the specialized classes of calcium-binding proteins, osteopontin, bone sialoprotein, and osteocalcin. The proteins are up-regulated at the mineralization stage with initial deposition of hydroxyapatite in the ECM. Both in vivo and in vitro osteoblasts at sites of bone formation and in the mineralized nodule undergo apoptosis, reflecting a biological mechanism for triaging those osteoblasts recruited to produce matrix but cannot all be accommodated in the mineralized matrix as osteocytes [213–217]. The increase in collagenase detected in cells at the mineralization stage in vitro may reflect the reorganization of the extracellular matrix necessary for maturation of osteoblasts to the osteocytes. Guiding the transitions from the osteoprogenitor to the osteoblast and osteocyte are the cell–matrix and cell–cell interactions, presumed to be important for progression of differentiation. Functional studies establishing requirement of the type I collagen and other ECM components in promoting osteoblast/osteocyte differentiation have been carried out by modifying production of osteoblast-secreted products or culturing osteoblasts on various matrices [218–221]. Indeed, the ECM composition changes during maturation of the osteoprogenitor cells. Osteoprogenitor cells synthesize collagen type III and large proteoglycans, versican and hyaluronan; whereas in preosteoblasts and osteoblasts, these proteins are replaced by collagen types I and V and small chrondroitin sulfate proteoglycans, as decorin and biglycan [222]. ECM-integrin signaling is an important pathway for activation of osteoblast gene expression. Osteoblasts appear to use β1 integrins to adhere to the full range of RGD-containing bone matrix proteins [223]. The intracellular kinase pathways that become activated through integrin signaling can modify a spectrum of factors
Chapter 14 The Cells of Bone
promoting osteoblast differentiation. For example, the Runx2 transcription factor essential for bone formation becomes phosphorylated in response to ECM-kinase signaling, enhancing its functional activity [224]. Cell–cell adhesion proteins also characterize osteoblast maturation. Several members of the cadherin family are expressed in osteoblasts including cadherin-11, cadherin-4, N-cadherin, and OB-cadherin [225]. N-cadherin is present on proliferative preosteoblastic cells, but is lost as they become osteocytic [226]. In contrast, OB-cadherin is barely detected in osteoprogenitor cells and is up-regulated in alkaline phosphatase-expressing cells [227]. The activated leukocyte cell-adhesion molecule (ALCAM; CD166) is highly expressed on the surface of MSCs and becomes down-regulated in concert with changes in morphology and detection of alkaline phosphatase activity as periosteal progenitors migrate and develop into an osteoblast [153, 228]. Signaling pathways from the extracellular matrix through the cytoskeleton and finally to the nucleus, which allow expression and up-regulation of bone-specific and bone-related genes, are being investigated. For example, β-catenin, the intracellular regulator of the canonical Wnt pathway, also colocalizes and coprecipitates with cadherins [225]. Another potential candidate is CD44, the hyaluronate receptor which is a nonintegrin adhesion receptor that is linked to the cytoskeleton. CD44 is enriched during osteocyte differentiation [229, 230] and has become a useful marker for this stage [231]. The temporal expression of proteins and enzymes involved in cell adherence and the initial production and modifications of the extracellular matrix, provides competency for mineral deposition and the final stage of osteocyte phenotype development in the mineralization period. The induced expression of subclasses of osteoblastic genes (shown in Fig. 2, osteocalcin, osteopontin, bone sialoprotein) are used as markers of the stages of osteoblast maturation, but may reflect selective functions of osteoblasts dependent on their location in bone. This concept is supported by the analysis of these expressed genes and protein levels at single-cell levels in osteoblast cultures developing a tissue organization (the bone nodule) and in bone sections [191, 232, 233]. The variations in levels of expressed gene products reveal subtle yet important differences in functional properties that relate to the establishment and maintenance of bone structure. Although there are likely to be more than the major stages in development of a mature osteoblast (depicted in Fig. 2), this model of differentiation provides a better understanding of cellular properties and selective responses to growth factors [234–236] and hormones [182, 237–240]. For example, TGFβ stimulates the replication of progenitor cells and directly stimulates collagen
231 synthesis, but the mitogenic effects of TGFβ are not apparent on mature postproliferative osteoblasts. The fibroblast growth factors (FGF-1, FGF-2) also stimulate bone cell replication in cells capable of collagen synthesis. However, FGF2 does not directly affect the differentiated functions, Indeed, in mature osteoblasts, FGF both inhibits type I collagen synthesis and increases collagenase expression [235]. The insulin-like growth factors (IGF-1 and IGF-2), which are synthesized by osteoblasts, are key regulators of osteoblast anabolic activities [241]. IGF-1 directly increases type I collagen synthesis and complementarily decreases the activity of the collagenase 3 (MMP-13) enzyme, reflecting its role in maintenance of the bone matrix. Anabolic effects of IGF-1 and IGF-2 in bone remodeling are further underscored by regulation of these proteins by hormones [242]). Their essential requirement for bone mineralization is established by the osteoblastspecific mutation of the IGF receptor [243]. Osteoblast differentiation and activity is strongly influenced by the steroid hormones glucocorticoid [182, 244], 1,25(OH)2D3 [237, 238, 245], estrogen [239], and androgens [246]. All have selective effects at specific stages of osteogenesis. Glucocorticoids and 1,25(OH)2D3 have antiproliferative effects to promote differentiation of the cells at early stages of maturation, but inhibit anabolic activities and promote resorptive properties of the osteoblast at later stages. Estrogens exert both antiapoptotic effects on bone-forming cells mediated by activation of the Src/Shc signal transduction pathway [247] and proapoptotic effects on mesenchymal cells and osteoclasts through classical ER regulation of target genes [248, 249]. The thyroid hormone receptor is expressed in all osteogenic lineage cells and affects bone formation and resorption [250]. Thyroid hormone (T3) is essential for linear growth and maintenance of bone mass (reviewed in reference [251]). Mice deficient in thyroid hormone receptors exhibit decreased bone mineral density and increased adipogenesis [252]. Differential regulation of specific gene expression as a function of differentiation by polypeptide hormones has also been documented. For example, PTH increases the resorption promoting RANKL cytokine to a greater extent in mature osteoblasts, but inhibits expression of the antiresorptive osteoprotegerin factor at all stages of osteoblast differentiation [253]. PTHrP has bidirectional effects on phosphorylation of ERK between mesenchymal cells and differentiated mouse osteoblast lines implicating MAPK signaling in regulating the selective effects of this hormone [254]. Taken together, our current understanding is one of selective and combinatorial responses of osteogenic cells to growth factors, hormones, and physical forces dependent on the structural and metabolic requirements of the cells in a particular tissue environment.
232
JANE B. LIAN AND GARY S. STEIN
C. The Osteocyte: Differentiation Supports Bone Structure, Viability and Physiologic Responsiveness The final stage of osteoblastogenesis begins with deposition of mineral in the ECM by several mechanisms (reviewed in reference [255, 256]). Osteoblasts on boneforming surfaces have several fates. As mineralization of the matrix envelopes the osteoblast, morphologic changes are induced in the cell when the osteoblast progresses to the fully differentiated osteocyte. Alternatively, on a quiescent bone surface, the osteoblast develops into a flattened bone-lining cell of a single layer forming the endosteum against the marrow and underlying the periosteum directly on the mineralized surfaces. This cell layer lining protects the bone from the extracellular fluid space. Lining cells function to (1) maintain the bone’s own environment through synthesis and resorptive-like activities, (2) protect themselves from cell death, and (3) respond to mechanical forces on the skeleton. Osteoblasts receive the majority of systemic and local signals and can transmit these to osteocytes, while the osteocytes perceive mechanical forces on the bone and transmit the regulatory information to the surface-lining cells. Osteocytes represent the most abundant cell in the skeleton. They are in direct communication with the bone-lining cell and with each other within the mineralized matrix through cellular processes that lie within and are tethered to the canaliculi channels [257]. Thus, the osteocytes and surface-lining cells form a continuum, or syncytium, by connection of their cell processes through gap junctions. The osteocyte phenotype develops during embedment of the osteoblast in mineralized bone, although the signaling mechanisms for how the cuboidal osteoblast transitions to the stellate osteocyte remain unknown. Osteoblasts that do not differentiate to the osteocyte or bone-lining cell undergo programmed cell death [216, 258]. However, TGFβ activation has been reported to maintain osteoblast survival during their differentiation into osteocytes [259]. Intermittent PTH (as opposed to continuous PTH treatment) also prevents osteoblast apoptosis increasing the lifespan of mature osteoblasts [260, 261]. The osteocyte is a terminally differentiated cell and when isolated directly from bone tissue, retains its morphologic features in vitro. In culture, the cell does not divide and upon attachment to surface, forms characteristic cytoplasmic extrusions in all directions [262]. The characterization of osteocytic cell lines has recently advanced our understanding of their phenotypic properties [263]. Osteocytes have the capacity to synthesize certain matrix molecules, as abundantly as osteoblasts, visualized by in situ studies. The detection of osteocalcin, for example, is particularly robust, supporting the relevance of in vitro
models for osteoblast differentiation which show maximal levels of osteocalcin in mineralized nodules. The dentin matrix protein 1 (DMP1) was found to be expressed 20-fold higher in the differentiated osteocyte and stimulated further by mechanical loading [264, 265]. The DMP1-null mice have reduced bone mineral [266]. Interestingly, a mouse model expressing 8 kb of the DMP1 promoter-GFP transgene showed responsiveness to increased mechanical loading in osteocytes [267]. Thus, highly specialized proteins are likely to be synthesized by osteocytes to support the integrity of mineralized bone tissue. Survival of the single osteocyte, resident in lacunae and in contact with other cells, is critical for bone to function as a viable organ and structural connective tissue. Disruption of cell–cell contact leads to cell death. In vivo osteocytes are long-lived cells, but, with aging, empty lacunae from cellular apoptosis have been observed [213, 214, 217, 258]. Several mechanisms protecting the osteocyte against apoptosis are being established. The CD40 antigen is expressed in an osteocyte cell line and adding CD40 ligand to cultured cells inhibited apoptosis induced by TNFα and glucocorticoids [268]. In vivo the loss of cell viability, as occurs in areas of microcracks, is coupled to targeting and initiation of new remodeling activity [214, 269–272]. However, the specific mediators of recruitment of cells to the microcracks need to be identified. Interestingly, a higher osteocyte lacunae density was quantified in woven compared to lamellar bone, further implicating the osteocyte with bone remodeling [273]. In vitro apoptosis of osteoblasts and osteocytes is associated with the mineralized nodule [216], a finding which may be analogous to in vivo bone formation where osteoblasts at bone-forming surfaces do not all mature to an osteocyte [215]. Alternatively, in vitro osteocytes may lack requisite factors for survival, as observed in vivo [217]. Bone as a structural connective tissue can adapt its architecture to meet physical demands on the skeleton. The osteocyte syncytium is highly responsive to loading forces [271, 274]. Mechanical forces on the bone result in stress-generated electric potentials produced by strain in the organic components (piezo-electric potential) or result in electrolyte fluid flow produced by deformation of the bone (streaming potential) ([268, 275] and reviewed in reference [276]). After osmotic pressure and vascularderived pressure gradients respond to the mechanical loading, the fluid movement which is produced through bone tissue is regulated by permeability of the tissue controlled by hyaluronan [277, 278]. Mechanisms which contribute to osteocyte function as the mechanosensor of skeletal tissues are being identified [279, 280]. Mechanical strain and fluid flow regulate their communication through GAP junctions [281]. An important component in establishing a viable bone tissue is an
233
Chapter 14 The Cells of Bone
obligatory cell–cell contact mediated by GAP junctions and formed primarily by connexin 43, between osteoblast and osteocytes within a mineralized matrix [282]. Both delayed osteoblast differentiation and increased osteoblast apoptosis characterize the connexin 43-null mouse. Thus, GAP junctions and the connexin 43 hemichannels maintain cell survival [283]. Other antiapoptotic signals continued to be identified; for example, the CD40 ligand known to function as an antiapoptotic factor in dendritic cells. Fluid shear forces were reported to promote cell survival by increasing the antiapoptotic factor Bcl2 and decreasing the proapoptotic Bax protein, while disuse resulted in apoptosis [284]. Importantly, GAP junctions allow the osteocytes to be metabolically and electrically coupled to surface osteoblasts. Rapid fluxes of bone calcium across these junctions facilitate transmission of information between osteoblasts on the bone surface and osteocytes within the structure of bone itself [285]. At a molecular level, it is now realized that the GAP junctions respond to mechanical strain on the bone through PGE2 and the prostaglandin EP2 receptor [263, 281, 286] and by increasing connexin 43 expression [287]. Mechanical strain is also likely to influence cell– matrix interactions through integrin signaling. Integrins are tightly coupled to the cytoskeleton [288] and together the integrin–cytoskeleton complex facilitates the transduction of mechanical signals that may ultimately lead to modifications in gene expression. The extracellular matrix receptors, such as integrins and CD44 receptors, are thought to mediate cellular sensing of mechanical loads [289]. Mechanical strain induces changes in attachment of cellular processes to the canalicular wall [257] and focal adhesion kinase activity [290]. Therefore, several components of osteocyte cellular architecture will respond to mechanical strain. While physiologic mechanical strain can maintain bone mass, skeletal unloading has negative effects on bone, for example, skeletal unloading alleviates the anabolic action of intermittent PTH [291]. The molecular mediators of osteocyte mechanotransduction signals are being identified, increasing our understanding of osteocyte functions and responses. Modifications in signaling factors, enzymes, and transcriptional regulators have been reported [292–294]. Direct evidence that osteocytes sense mechanical loading [295] has been demonstrated by rapid changes in metabolic activity by either 3H-uridine uptake [296], increased glucose-6-phosphate dehydrogenase activity [297], activation of the kinase signal transduction pathway [290], and increased IGF-1 expression [298]. Osteocytes produce IGF-1 and release prostaglandins in response to stress [299]. Signals induced by fluid flow include prostaglandin PGE-2, cAMP, and nitric oxide (NO), all potent stimulators of bone formation and bone resorption consistent
with mechanical stimulation of bone turnover [263, 293, 300, 301]. Mechanical strain activates the endothelial nitric oxide synthase (eNOS) in osteocytes producing NO [302]. NO is a highly reactive molecule and NO produced by the eNOS isoform is essential for osteoblast function [303], while the inducible isoform (iNOS) produces cytokines that act on osteoclasts [303–306]. Recently, the neuronal NO synthase has been implicated in bone turnover [307]. Interestingly, osteocytes have the capacity to retain responses and the mechanisms contributing to their longterm potential are being identified. Skerry and colleagues considered the involvement of the NMDA and AMPA glutamate receptors, which account for the cell memory in the central nervous system in osteocyte activities [308]. Glutamate, an intercellular signaling agent, was found to be activated in response to mechanical loading, and the glutamate receptor 2 shown to be important for osteoblast survival [309]. In conclusion, the structural organization of the osteocyte in bone and its direct contact with active osteoblast or surface lining cells is consistent with the concept that bone cells respond to varying physiological signals through multiple pathways to communicate their responses.
D. Transcriptional Control of Osteoblast Differentiation The progression of osteoblast differentiation requires the sequential activation and suppression of genes that encode phenotypic proteins and regulatory factors. Signaling molecules (morphogens, growth factors) indirectly result in a cascade of gene expression through induction of transcription factors that directly engage in protein–DNA as well as protein–protein interactions [212]. Human mutations, mouse genetics, gene profiling, and differential display strategies in response to osteoinductive factors, for example, BMP2, have all contributed to an explosion in the identification of many transcriptional regulators and signaling proteins that control bone formation. Several regulatory factors have been shown to be necessary for normal skeletal development and bone formation based on genetic models [310]. Examples of recently identified factors that play key roles in regulating maturation of osteoblasts, as well as maintenance of bone architecture proven from genetic models, are presented in Table II. From these new findings, several important concepts have emerged for understanding direct effects on gene expression that integrate physiological responses. The requirements for gene expression during osteoblast differentiation are now being understood to involve (1) both negative and positive transcriptional regulatory cascades
234
JANE B. LIAN AND GARY S. STEIN
Table II.
Transcriptional Regulators of Osteoblast Differentiation
Promote osteoblast differentiation Runx2 JunB Fra1 䉭FOSB DLX3 DLX5
Osterix ATF4 C/EBP
CHOP C/EBP/LAP Inhibit osteoblast differentiation STAT1
C-fos
MSX2
TWIST
HEY1
Lian et al., 2004 [516] Hess et al., 2003 [517] Jochum et al., 2000 [518] Kveiborg et al., 2004 [519] Sabatakos et al., 2000 [520] Hassan et al., 2004 [326] Lee et al., 2003 [521] Miyama et al., 1999 [522] Tadic et al., 2002 [523] Tai et al., 2004 [524] Nakashima et al., 2002 [71] Yang et al., 2004 [525] Harrison et al. 2004 [363] Gutierrez et al., 2002 [361] Umayahara et al., 1999 [526] Pereira et al., 2004 [362] Iyer et al., 2004 [527] Kim et al., 2003 [506] Xiao et al., 2004 [507] Takayanagi et al., 2004 [508] Grigoriadis et al., 1993 [318] Sunters et al., 2004 [509] McCabe et al., 1996 [316] Ichida et al., 2004 [510] Dodig et al., 1999 [511] Yoshizawa et al., 2004 [512] Cheng et al., 2003 [513] Lee et al., 1999 [314] Bialek et al., 2004 [315] Ishii et al., 2003 [514] Zamurovic et al., 200 [515]
for supporting osteoblast phenotype development; (2) the role of co-regulatory protein interactions in response to physiologic factors for achieving gene-specific activation and suppression of transcription at appropriate times in development of a cell phenotype; and (3) combinatorial control mechanisms involving the integration of signal transduction with transcriptional control of gene expression, as well as interactions between transcription factors for cooperative effects on a gene promoter. This section will focus on selected transcription factors to illustrate these concepts. The Runx2 protein has provided a paradigm for understanding these regulatory mechanisms. 1. Regulatory Networks for Repression and Activation of Gene Transcription: AP-1 and Homeodomain Proteins
We have learned that factors with identified roles in specifying spatial differentiation and pattern formation during embryogenesis are re-expressed at selective stages of osteoblast maturation. As bone remodels in the postnatal
animal, there is a continuing requirement to re-establish and sustain a complex tissue organization. A concept is emerging that expression of many transcription factors expressed in the undifferentiated mesenchymal cell, although classified as inhibitors of osteoblast differentiation, are requisite for normal bone formation. The helix-loop-helix (HLH) family of proteins (Twist, Inhibitor of differentiation (Id), Scleraxis) are examples of negative transcriptional regulators at early stages of osteoblast development, but necessary for osteoprogenitor cell expansion [311]. They are expressed in mesoderm tissue of developing embryo, but not detected at the onset of ossification [312, 313]. The restricted expression of these factors in proliferating osteoblasts in vitro is consistent with their required downregulation in order for differentiation to proceed [314]. Another mechanism of negative regulation of osteoblastogenesis has recently been defined for twist inhibition of osteogenesis. Twist proteins transiently inhibit Runx2mediated transcription by interacting with the Runx2 DNAbinding domain. Thus, in vivo deficiency of Twist proteins results in premature osteoblast differentiation [315]. A typical molecular mechanism for control of gene expression involves a change in heterodimerization proteins to form either a transactivating or repression complex transcription of the gene at specific stages of osteoblast maturation. AP-1 factors include fos (c-fos, fra1, fra2) and jun (c-jun, jun-D, jun-B) oncogene-encoded transcription factors. These proteins regulate cell cycle and differentiation-related genes in cartilage and bone by forming homo- or heterodimeric complexes with fos, jun, and ATF member proteins at AP-1 regulatory elements. Through heterodimer formation with different partners, cellular levels of transcription can be finely modulated or selectively repressed or enhanced, as occurs on the osteocalcin gene where c-fos/c-jun down-regulates and fra2/ junD up-regulates transcription [316]. Another example revealing a novel function for an AP-1 factor is the binding of JunD to menin (product of the MEN1 tumor suppressor gene) which changes JunD from a growth suppressor to a growth-promoting factor when complexed with menin [317]. AP-1 elements are found in most genes and are responsive to growth factors, polypeptide hormones and other physiological signals. Fos- and jun-related proteins exhibit developmental stage-specific expression and activities during osteoblast differentiation in vivo and in vitro constituting a regulatory network [318–320]. Thus, in part, cellular levels of transcription factors are a determinant to which factors occupy regulatory elements. Genetic studies have established that several AP-1 family members are essential for normal bone development and osteoblast differentiation (Table I). In conclusion, the modifications observed in the representation and activities of family
Chapter 14 The Cells of Bone
members of these classes of transcription factors during osteoblast differentiation reflect linkage to transcriptional control for bone formation. The concept of a regulatory network of transcription factors for osteoblast differentiation is best illustrated by the Msx and Dlx homeodomain proteins. The importance of homeodomain proteins for normal bone formation is realized by skeletal abnormalities that result from mutations or misexpression of these factors [10, 321–324] (see also section III and Table I). The most well-studied during osteoblast differentiation are Msx2, Dlx3, and Dlx5 [56, 325, 326]. To ensure maintenance of the osteoblast phenotype, these related family members of specific classes of proteins are expressed in specific osteoblast subpopulations and temporally expressed during osteoblast differentiation with Msx2 restricted to proliferating and apoptotic cells, followed by up-regulation of Dlx3 and Dlx5 expression in postproliferative cells [216, 326–330]. The chromatin immunoprecipitation assay has allowed direct analysis of specific homeodomain (HD) proteins associated with gene promoters during osteoblast differentiation. Studies of the osteocalcin (OC) gene, which contains an HD element that supports osteoblast-specific expression of OC, have identified two molecular switches controlled by different HD proteins to regulate its transcription. Msx2 is bound to the OC HD element in the growth stage when no mRNA is detected [326, 331, 332]. When transcripts are expressed, Runx2 and Dlx3 replace Msx2. A second switch in HD protein occupancy at the OC promoter occurs at the mineralization stage with replacement of Dlx3 by Dlx5 binding to the gene [326]. In conclusion, these studies have provided evidence for coordinated control of gene expression throughout bone formation by family members of a class of transcription factors that form a regulatory network. 2. Co-regulatory Protein Interactions with Transcription Factors Support Integration of Physiologic Signals and Gene-specific Expression: Role of Runx Factors
A fundamental mechanism of transcription factor activity is modulation by co-regulatory protein interactions for integrating and transducing physiological signals. This concept is best illustrated by the Runx2 (CBFA1/AML) transcription factor introduced earlier in this chapter. It is one member of a three mammalian gene family that are designated runt homology domain (RHD)-related core-binding factors. The shared RHD binds to the DNA motif ATGGCT/G in target genes and requires a partner protein CBFβ which does not contact the DNA. Genetic studies in human and mouse have
235 shown each Runx gene and CBFβ to be essential for organogenesis and embryogenesis. In human these factors were initially designated as AML because the first gene to be identified (Runx1/AML1B) is encoded by gene loci that are rearranged in acute myelogenous leukemia (AML) [333]. Runx1/Cbfa2/AML-1B/PEBP2B is critical for definitive hematopoiesis. Runx1 is also highly expressed in perichondrium/periosteum, and downregulated in mature osteoblasts [334, 335], although a critical role in chondrogenesis has not been identified as Runx1-null mice die at 11.5 days postcoital. Runx3/Cbfa3 (AML2/PEBP2C) exhibits more restricted expression, contributing to gut and nerve development with Runx3 gene mutations associated with gastric cancers. However, Runx3 is detected in the hypertrophic cartilage zone, but its absence (Runx3−/− mouse) only slightly delays maturation and vascular invasion. At birth, a normal skeleton is found [336]. Runx2/Cbfa1/AML-3/PEBPP2A is specifically required for bone formation demonstrated by the inhibition of bone tissue formation in the Cbfa1-null mutation mouse model [63, 337]. Although the skeleton forms through the initial chondrogenic stage, mineralization is blocked. Recent studies have identified all three Runx factors in premesenchymal condensations and specific skeletal elements, although their functions at this stage prior to bone formation are unknown [130, 334, 338–341]. Additionally, all factors are expressed as isoforms which arise from alternate usage of two promoters and multiple ATGs. For Runx2, the resulting proteins differ in their N-terminal sequence and the isoforms appear to have similar functions, but at different stages of embryonic development or cellular differentiation [340, 342–344]. All Runx factors have two key properties for their functional activity: [1] they are platform proteins that interact with a spectrum of co-regulatory proteins to integrate developmental signals; and [2] Runx proteins contain a unique nuclear matrix-targeting signal (NMTS) which mediates the intranuclear trafficking of Runx2 and their co-regulatory proteins to domains associated with the nuclear scaffold. Co-regulatory proteins include CBFβ, an essential binding partner protein that interacts with the runt homology DNA-binding domain to promote Runx2 interaction with ATGGCT/G motifs in gene promoters. Absence of functional CBFβ in osteoblasts results in bone abnormalities in the mouse [135, 345, 346]. Other partner proteins contribute to gene structure by either promoting increasing transcription by acetylation (e.g., by p300) or inhibiting target gene transcription through deacetylation of histone proteins by histone deacetylase (HDAC) enzymes. Runx2 thereby has a function in modifying chromatin organization of target genes to control accessibility
236 of regulatory factors on target gene promoters. This has been clearly documented for the activation/repression of the osteocalcin gene [347–351]. Such modifications in chromatin structure involving Runx2 are critical for osteoblast differentiation [62] and normal bone development as shown by null mutation or overexpression of HDAC4, leading respectively to premature ossification or inhibition of chondrocyte maturation [352]. The C-terminus activation domain of Runx2 interacts with a variety of signaling proteins including, for example, Smads that mediate BMP/TGFB [58–60]; the Yesassociated protein (YAP) that mediates c-Src signaling [353], homeodomain (HD) proteins [326, 354]; TAZ, a transcriptional co-activator [355]; Groucho/TLE, a developmental signal which suppresses Runx2 activity [356], while a Groucho homolog Grg5 enhances Runx2 transcription and is particularly critical for growth plate development [357]. These co-regulatory factors are organized with Runx as multimeric complexes on target genes in subnuclear domains visualized as punctuate foci. Interestingly, the Runx2 nuclear matrix targeting sequence (~32 amino acids) (NMTS) is also the interacting domain for several of these co-regulatory proteins (Smads, homeodomain proteins, and YAP), emphasizing the importance of organizing gene regulatory complexes in subnuclear domains for tissue-specific control of gene expression [59, 326, 353]. The requirement for this architectural organization of Runx2 transcriptional complexes for bone formation was established in vivo by a knock-in mutation of Runx2 in the mouse [358] and also by several in vitro functional studies. Deletion of the C-terminus targeting and co-regulatory protein interaction domain results in a lethal phenotype and absence of mineralized skeleton analogous to the Runx2-null mouse. There is a requirement for Runx2 recruitment of Smad co-regulatory factors to Runx domains to execute the BMP/TGFβ transcriptional signal for target gene regulation [58, 59]. Of further significance for this critical property of Runx2 is the observation that Runx2 subnuclear targeting deficient mutant proteins, when expressed in metastatic breast cancer cells, inhibit the expression of Runx2 target genes and block the osteolytic properties of metastatic cell lines in bone [359, 360]. Thus these specialized properties of Runx2 provide a basis for controlling accessibility of promoter elements to cognate transcription factors by chromatin remodeling and supporting integration of activities in response to physiologic signals. Taken together, these findings have demonstrated a unique mechanism for tissue-specific control of gene expression through the formation of transcriptional complexes in nuclear microenvironments.
JANE B. LIAN AND GARY S. STEIN 3. Cooperative Control of Gene Transcription Through Interactions at Distinct Promoter Elements: Contribution of C/EBP and Steroid Response Elements
CAAT enhancer-binding proteins include a family of transcription factors that bind to specific motifs and function in cell phenotype differentiation. C/EBPα plays an essential role in adipogenesis, while C/EBPβ is now appreciated to function as a key component of transcriptional control for bone formation. C/EBPβ is up-regulated during osteoblast differentiation [361] and expression of a homologous C/EBP enhancer-binding protein, DDIT3, was shown to induce osteoblast differentiation [362]. Significantly, targeted expression of a truncated nonfunctional C/EBP isoform in bone results in osteopenia [363]. C/EBP and Runx2 regulatory elements are often located in gene promoters in close proximity to each that support physiologic responsiveness for bone formation. C/EBP and Runx2 proteins can form complexes by a direct protein–protein interaction that results in a 30–50% synergistic increase in osteocalcin expression [361]. This mechanism supports osteoblast-specific expression of other genes, such as sclerostin [364]. Other examples of cooperative effects between transcription factors are the requirements of Runx2 and cfos interaction for PTH responsiveness on the MMP13 promoter [365]. Large multimeric complexes are also formed at steroid hormone response elements, including VDRE and ERE motifs. The vitamin D and estrogen receptor complexes recruit chromatin remodeling factors that can bridge with transcriptional complexes at neighboring elements. In this regard, the Runx2 site adjacent to the VDRE interacts with p300 to facilitate vitamin Dmediated enhancement of osteocalcin gene expression [351]. Runx2 also mediates estrogen responses at ERE motifs [366]. In this manner, physiologic control of gene expression is coordinated among distinct regulatory elements on promoters and in a tissue-specific manner.
V. THE OSTEOCLAST: A FUNCTIONALLY UNIQUE CELL FOR PHYSIOLOGICALLY REGULATED RESORPTION OF BONE MINERAL The osteoclast is a multinucleated cell with the ability to degrade mineralized tissues dependent upon cellular attachment to the bone substratum. Osteoclasts are large cells (~100 µm) and appear in histologic sections of bone generally in resorption pits called Howships lacunae. Nuclei numbers are variable and in normal human osteoclasts in
Chapter 14 The Cells of Bone
the range of 4–10 can be seen. Nuclei numbers appear to reflect osteoclast activity and can number as many as 20–50 nuclei in Paget’s disease [367]. It appears from a limited number of studies that all osteoclast nuclei are equivalently transcriptionally active with the components of the transcriptional machinery, including RNA processing and splicing functions organized in subnuclear domains in each nucleus [368, 369]. The osteoclast is a highly polarized cell raising questions as to whether specific nuclei in osteoclasts express subsets of genes reflecting different functional activities or whether they have equivalent functions. Two components of the bone microenvironment are essential for osteoclastogenesis: (1) the mineralized bone tissue is necessary for recruitment, attachment, and activation of a functional osteoclast; and (2) osteoblast lineage cells are required for fusion of the mononuclear precursor to the multinucleated osteoclast and are responsive cells to the calcitrophic hormones which promote osteoclastic resorption. Microfractures that need repair or mechanical forces that signal adaptive changes in bone architecture result in a change in stress lines that provide a road map for homing of osteoclasts to the appropriate site [299]. The requirement for stromal osteoprogenitors or osteoblasts in mediating osteoclast differentiation is linked to the production of cytokines, CSF-1/M-CSF, RANKL, OPG, and several interleukins essential for regulating development of pre-osteoclasts and the multinucleated phenotype.
A. Cellular Features of Bone Resorption The highly specialized morphological features of the osteoclast accommodate its unique function to resorb bone and calcified cartilage. The structural properties of the osteoclast are recognized only when the cell is activated, a process that requires attachment to the mineralized surface, which induces a distinct organization of the cytoskeleton [370–372]. Characteristic features of the osteoclast that result from attachment are: (1) appearance of the sealing zone which is a podosome or focal adhesion complex, a specialized cell–extracellular matrix adhesion structure which allows the osteoclast to form a tight seal against the bone surface; (2) formation of a highly convoluted plasma membrane, the “ruffled border” on the apical bone surface (between the sealing zones); (3) a distinct organization of the cytoskeleton; (4) projections of the basal lateral membrane for secretion of minerals into the vascular space; (5) polarization of the osteoclast nuclei at the basolateral surface; and (6) numerous mitochondria to support activity of the proton pumps, as well
237 as primary lysosomal organelles for secretion of enzymes that will degrade the ECM and secondary lysosomal particles removing matrix debris. Mechanisms contributing to these functional activities in relation to osteoclast cell structure will be briefly described. Genetic and in vitro studies have defined the molecular components of the signaling pathways controlling osteoclast activity which have been recently reviewed in great detail [373, 374]. The adhesion of recruited osteoclasts to the bone surface involves classical integrin receptors αVβ3 and results in the activation of several signaling cascades and distinct morphological features of resorbing osteoclasts. Disruption of integrin mediated adhesion results in inhibition of bone resorption [375, 376]. Following αVβ3 interaction with RGD matrix proteins (vitronectin, osteopontin), small GTPases (Rho and Rac) regulate cytoskeletal re-organization which leads to cell spreading, formation of the sealing zone, and definition of the ruffled membrane [377, 378]. The αVβ3 integrin initially is randomly distributed on the plasma membrane prior to osteoclast activation, then translocates to the bone matrix attachment site, and associates with the filamentous actin in a complex forming a podosome analogous to a focal adhesion structure. The podosome structures completely surround the ruffled membrane rich in F-actin filaments that are oriented perpendicular to the bone surface forming a ring. The actin ring can be used to distinguish between resorbing osteoclasts and nonresorbing cells. Upon osteoclast adhesion to the bone surface, the proto-oncogene c-Src selectively binds the β3 integrin for transduction of the c-Src-dependent polarization signal separating the apical and basal lateral membranes [378–380]. C-Src is a nonreceptor tyrosinase kinase enzyme having an essential role in activating quiescent osteoclasts for bone resorption by formation of the ruffled membrane and polarization of the nuclei in the activated osteoclast [380]. The c-Src−/− mutant mice have many multinucleated cells present in bone with some properties of the osteoclast, but lack function because they are incapable of forming a ruffled border [381, 382]. Both c-Src dependent tyrosine phosphorylation and activation of c-Src result in formation of a complex with a proline-rich tyrosine kinase 2 (Pyk2), a member of the focal adhesion kinase (FAK) family. A major adhesion-induced kinase in osteoclasts is formed with Src, Pyk2, and c-Cbl, a requirement for osteoclast motility [383]. These components of the podosome complex form a positive and negative regulatory loop to regulate cytoskeletal rearrangement, migration, and polarization of the osteoclast [384]. c-Src binds to and phosphorylates c-Cbl which is a β3 ubiquitin ligase and in
238 turn will inhibit Src kinase activity [380]. Therefore, osteoclast migration is controlled by degradation of the podosome complex through the proteasome pathway upon polyubiquitinylation of c-Src. Formation of the complex will occur once again upon osteoclast adhesion [373]. Through these mechanisms, cyclic events of osteoclast activation and migration allow for bone formation at the resorbing site to maintain bone mass. Thus, the initial integrin-mediated adhesion attachment of the osteoclast involves a highly regulated and coordinated series of signaling events for cell attachment and structural modifications that mediate osteoclast activation and migration. The active process of bone resorption requires additional cell structural changes. Attachment of the osteoclast to bone also results in tubulin polymerization for movement of intracellular acidic vesicles along the microtubules to be targeted to the ruffled membrane. Insertion of the H+-ATPase-containing proton pump-bearing vesicles into the apical membrane leads to the infoldings that characterize the ruffled membrane, the actual resorbing component facing the resorption lacunae. The ruffled membrane is compartmentalized with respect to transcytotic vesicles forming at the center of the lacunae and endocytotic vesicles located more peripherally [385]. The tight seal of the surrounding plasma membrane to the bone surface, described above, is necessary to form a proton-impermeable acidic environment in the lacunae for degradation of mineralized bone. Acidification of the resorption lacunae is accomplished by the activities of carbonic anhydrase II and protons generated through the membrane bound H+-ATPase [386, 387]. Mice deficient in the ATP6i complex develop severe osteopetrosis due to the loss of acidification [388]. Various ion channels, chloride/carbonate, and sodium/proton antiporters, maintain cellular electroneutrality (reviewed in references [389, 390]). Osteoclasts are rich in matrix-degrading enzymes, particularly cathespin K and matrix metalloproteinases (MMPs) and a tartrate-resistant acid phosphatase (TRAP) isoenzyme which provides a histochemical marker commonly used to identify osteoclast activity. Genetic studies have identified their critical functions in osteoclastic resorption of bone. Cathepsin K-null mice not only have excessive trabeculae, but an altered ultrastructure of osteoclasts. Recently a human mutation in cathepsin K was identified in a syndrome of pychondystosis [391, 392]. TRAP-null mice revealed that cytoplasmic vesicles accumulate in TRAP−/− osteoclasts suggesting a functional role in modulating vesicular transport [393]. The MMP9null mouse exhibits a delay in osteoclast recruitment [394]. Enrichment of the matrix metalloproteinase MMP9 is a marker of chondroclasts and stimulates MMP13 in hypertrophic chondrocytes which more efficiently degrades
JANE B. LIAN AND GARY S. STEIN
unmineralized cartilage proteins. MMP9 facilitates bioavailability of the vascular endothelial growth factor, thus recruiting osteoclasts for normal development and long bone growth in conjunction with vascularization of the growth plate [395]. This complexity of osteoclast cellular features is required and coordinated for recruitment of the multinucleated cell to the bone surface for its boneresorbing function.
B. Molecular Mechanisms Mediating Osteoclast Ontogeny The multinucleated osteoclast forms only through the fusion of mononuclear precursors. Proliferation of osteoclasts has never been documented. The naturally occurring animal models of osteopetroses [396], null mutation mouse models [160], and bone marrow transplant or splenic parabiosis studies have firmly established the hematopoietic origin of the mononuclear precursors. Studies that determined expression of cell-surface antigens and phenotypic markers of pre-osteoclast lineage cells, as well as those studies which defined factors regulating commitment to a pre-osteoclast phenotype and maturation to the osteoclast [397, 398], show that the process of osteoclast differentiation involves a number of steps (Fig. 3). An osteoclast-specific cell line is formed at some point in time after formation of the colony-forming unit (CFU) for granulocytes and macrophages. Recently, however, consideration has been given to the formation of osteoclasts from pro-B cells [164]. Historically the generation of osteoclasts in vitro required co-culture of hematopoietic marrow or spleen cells with marrow stromal cells or an osteoblastic cell line. Stromal osteoblastic cells are the major source of two cytokines essential for osteoclastogenesis, the colonystimulating factor M-CSF (also referred to a CSF-1) and a membrane-bound TNF-related cytokine RANKL (receptor activator of NF-κB ligand)/TRANCE/TNFSF11 (tumor necrosis factor ligand superfamily member 11) (a type II transmembrane protein). Both factors are potent stimulators of bone resorption and participate in osteoclastogenesis at early and late stages. M-CSF interacts with the c-Fms receptor on hematopoietic lineage cells initiating a cascade of events for preosteoclast expansion and fusion. c-Fms is a tyrosine kinase receptor which recruits c-Src and phosphatidylinositol-3 kindase (PI-3K), resulting in proliferation and survival of osteoclast precursor cells [399]. The significance of M-CSF for osteoclast function was first established by neutralizing antibodies against M-CSF, which completely inhibited multinucleated cell formation in mouse bone marrow cultures and later by
239
Chapter 14 The Cells of Bone
Figure 3 Regulation of osteoclastogenesis. (A) Stages in progression of osteoclast differentiation are controlled by transcription factors (Pu.1, cFos , NFAT, NF-κB, FRA1), cytokines (interleukins, TGFβ, RankL), hormones (PTH, 1,25(OH)2 D3, E2) and extra- and intracellular signaling factors (integrins, c-Src, Traf6). (B) Markers frequently used to identify the phenotype of osteoclast lineage cells. CT-R, calcitonin receptor; TRAP, tartrate-resistant acid phosphatase; Cl C-7, specific chloride channel; CAII, carbonic anhydrase.
the finding that recombinant M-CSF injections cure the osteopetrosis in the OP/OP mouse, which is characterized by production of a nonfunctional CSF-1 protein [389, 400–402]. The M-CSF-1/c-Fms receptor interaction signals through the intracellular p38 MAPK, which leads to phosphorylation of the microphalmia-associated transcription factor (MITF) for transcription of target genes that include carbonic anhydrase II and tartrate-resistant acid phosphatase, markers of mature osteoclasts [403, 404]. Spontaneous Mitf mutations in quail and null mutation in mouse (Mitf Mi/Mi) result in an absence of osteoclasts and osteopetrosis [405, 406]. M-CSF and αvβ3 integrin interactions also occur, resulting in phosphorylation of p130(Cas) and c-cbl and consequent p130(Cas) association with Pyk2, thereby contributing to osteoclast attachment and activation [407–409]. The discoveries of RANKL which interacts with RANK on osteoclast precursors and the osteoclast inhibitory factor, osteoprotegerin (OPG), provided insight into mechanisms for the essential role of osteoblasts in mediating osteoclastogenesis. RANKL is produced by osteoblastic stromal cells and, notably, by activated T lymphocytes thereby contributing to bone resorption in inflammatory disease states [410]. RANKL was demonstrated to have competency for inducing osteoclast formation from hematopoietic cells in the absence of stromal cells, allowing
osteoclasts to be generated in vitro from precursor cells without co-culture. Mice deficient in either RANK (encoded by the gene TNFR-SFIIA) or RANKL, and human mutations in the RANK receptor exhibit the same phenotype, osteopetrosis (reviewed in reference [373]. RANKL activation of RANK signaling is transduced by the binding of TRAFs (TNF receptor associated cytoplasmic factors) to the cytoplasmic domain of RANK (TNFRSFIIA). Although multiple TRAFs are present in the cells, only the TRAF6-null mutation results in osteopetrosis [411–413]. RANKL interaction with RANK on osteoclast precursors activates gene expression through signal transduction pathways resulting mainly in NF-κB- and AP-1-mediated transcriptional events (illustrated in Fig. 4). Together then, the RANK/RANKL and cFms/ CSF-1 receptor/ligand pairs represent factors required for coupling stromal/osteoblastic cells to the formation of osteoclasts, while OPG attenuates resorption by inhibiting osteoclastogenesis. Regulation of osteoclastogenesis is achieved at multiple levels. Osteoprotegerin (OPG) encoded by TNFR-SIIB is a soluble decoy protein with strong homology to RANK receptors expressed and secreted by several tissues, including bone, cartilage, kidney, blood vessels, and osteoblast lineage cells [414, 415]. OPG functions as a soluble decoy receptor for RANKL and thereby blocks RANKL-mediated
240
JANE B. LIAN AND GARY S. STEIN
Figure 4
Signaling pathways and transcriptional targets of osteoclast differentiation. The extracellular signal is shown in bold and transcription factor targets in grey. Not all intermediates are shown. See text for details. Several extracellular signals can be transduced through common intracellular pathways. PI3k/Akt signaling leads to cell survival, ERK signaling also results in cell survival and proliferation, while the IKK/NF-κB pathway supports osteoclast resorption. ITAM, immunoreceptor tyrosine based activation motif; SHIP, Src homology 2-containing inositol phosphatase negatively regulates osteoclast formation [528].
osteoclast formation and survival of pre-existing osteoclasts [416]. Overexpression of OPG in transgenic mice resulted in severe osteopetrosis as a result of inhibition of the terminal stages of osteoclast differentiation [414, 415, 417, 418]. A spectrum of cytokines also regulates osteoclast differentiation and activity. The stromal cell-derived growth factor (SDF-1) recruits osteoclast precursors by inducing MMP and chemotaxis [419]. The prostaglandin E[2] cytokine, also produced by stromal cells, acts as a potent stimulator of bone resorption and signals through the PGE2 receptor that exists as four subtypes. Null mutation of the EP2 and Ep4 receptors confirms the importance of PGE2 for osteoclast formation [420, 421]. The cytokines II-6 and II-11, like TNFα and IL-1 that are predominantly derived from monocytes, have mitogenic effects on the mononuclear osteoclast precursor. TNFα also promotes osteoclast survival by stimulating phosphorylation of the Akt and ERK, thereby increasing the activity of these pathways [422]. In addition, IL-1 and TNF feed back on stromal cells to regulate their production of CSF-1 and IL-11 (reviewed in references [389, 423–425]). IL-6 and related cytokines that include (but are not limited to) IL-11, oncostatin M, and leukuemia-inhibitory factors, which all enhance osteoclast differentiation. The selective effects of these cytokines occur through the same signal transduction pathway [389, 397, 426, 427]. The membrane-bound cell-surface receptor consists of two components, a specific ligand-binding subunit and a
second subunit that participates in the signaling cascade and is the transducing protein gp130 [424, 428]. The signal transduction cascade includes the nonreceptor tyrosine kinases known as Janus kinases (JAKs), which in turn phosphorylate components of the receptor kinases, and a series of cytoplasmic proteins designated signal transducers and activators of transcription (STATs). Specific cytokines also inhibit osteoclastogenesis [429]. IL-18, initially described as a T-cell cytokine that promotes interferon gamma and production by lymphocytes, was shown to be an important negative regulator of osteoclast formation. It is expressed in osteoblasts but functions via M-CSF [430, 431]. IL-7, a lymphokine secreted from T cells that stimulates production of M-CSF and RANKL, was thought to activate bone resorption. However, evidence from IL-7R-null mice indicates the cytokine is a direct inhibitor of osteoclast formation. The IL-7-deficient mice exhibit an increase in osteoclast number and confirmed by in vitro and ex vivo studies [429].
C. Hormonal Regulation and Transcriptional Control of Osteoclast Differentiation and Activity The calcitrophic hormones 1,25(OH)2D3 and parathyroid hormone (PTH) influence osteoclastogenesis at multiple stages of differentiation and activation through receptors on osteoblast lineage cells. It had been very well
Chapter 14 The Cells of Bone
established that 1,25(OH)2D3 induces differentiation of mouse leukemia cells into macrophages and promotes their activation and fusion [432]. However these cells cannot form resorption pits on dentin slices but required the presences of osteoblastic cells. It is now appreciated that 1,25(OH)2D3, as does PTH and numerous stimulators of bone resorption, including glucocorticoids, T4/T3, TNFα, IL-1, IL-11, and PGE-2, regulate osteoblast expression of RANKL. To further ensure osteoclastogenesis, the calcitrophic hormones 1,25(OH)2D3 and PTH, as well as glucocorticoids, also inhibit OPG production. These hormones also modulate the synthesis and secretion of CSF-1, IL-1, IL-6, and TNFα that have mitogenic effects on osteoclast progenitors. However, direct effects of hormones on osteoclasts should not be dismissed. For example, a VDR-binding protein TAFII-17 was recently found to be expressed by Pagetic osteoclast precursors, but not normal pre-osteoclasts [433]. Although Pagetic osteoclast lineages are characterized by the measles virus nucleocapsid protein (mVNP) responsible for the formation of the large size and numbers of osteoclasts, VDR-TAF II-17 interactions may contribute to the osteoclastogenesis in Paget’s disease. PTH is an essential regulator of calcium homeostasis, which induces a cAMP signal through G protein-coupled receptors (see Chapters 16 and 17). Interestingly, the PTH-related protein (PTHrP) influences osteoclastogenesis after birth and in rapidly turned-over tissues [434]. PTH and PTHrP have homologous N-terminal sequences (amino acids 8–13), which are required for binding to the shared PTH/PTHrP receptor. As such, PTHrP also serves as an inducer of bone resorption and hypercalcemia [435–437]. The majority of evidence supports an indirect activity of PTH and PTHrP on bone resorption, mediated through stromal and osteoblast lineage cells [438]. However, there are reports of osteoclasts expressing PTH/PTHrP receptors [246, 439–444]. The mechanisms by which PTH/PTHrP affects bone resorption at high catabolic concentrations versus osteoblast activity at low anabolic concentration are being identified [445]. Resorptive activity is controlled through regulation of RANKL and OPG [253, 446, 447], while anabolic activity is mediated by AP-1 factors [448] through an IGF-1 dependent mechanism, as described earlier [449, 450]. Estrogen has a broad range of effects in regulating bone resorption. Estrogen’s protective effects on bone are exerted by several mechanisms. Estrogen inhibits PTHinduced osteoclastogenesis and attachment to bone [444]. The suppression of PTH-stimulated osteoclast activity occurs both by blocking the cAMP-dependent PKA and calcium PKC pathways and by promoting osteoclast apoptosis [444]. Equally important, several cytokines that
241 prolong osteoclast survival, IL1, IL6, II7, TNFα, M-CSF, and RANKL are negatively regulated by estrogens and thereby estrogen increases osteoclast apoptosis [451–455]. At the same time, estrogen stimulates synthesis of antiresorptive factors OPG and TGFβ [456]. Thus, estrogen deficiency will lead to a marked increase in bone turnover. Estrogen effects are indirect functioning through the osteoblast lineage cells, mainly through the ERα, although the ERβ receptor also mediates responses [457] to activate osteoclasts. A few studies have reported that osteoclasts express estrogen receptors [442, 458, 459], while others have not found this result in developing bone [460]. Nonetheless, in isolated osteoclasts, estrogen induces a rapid response on inhibitory response via pp60src signaling [461]. The role of estrogen for bone health in men is being appreciated [443, 462]. Calcitonin, a polypeptide hormone, is secreted by the thyroid C-cells, but also by neuroendocrine cells, in response to hypercalcemia. Calcitonin binding to a G protein-coupled-type calcitonin receptor, which is a characteristic marker of the mature osteoblast, rapidly terminates resorption activity [463, 464]. The inhibition of osteoclast function induces rapid disappearance of the actin rings and subsequent loss of structural morphology. Calcitonin inhibits proton extrusion via PKA and cytoskeletal proteins become reorganized for osteoclast cell retraction [372, 465–468]. A surprising finding from the calcitonin/calcitonin gene-related peptide-null mouse is that absence of calcitonin does not affect bone resorption. A compensatory mechanism is the sensitization of the osteoclast to PTH upon calcitonin withdrawal [469, 470]. Calcitonin (salmon calcitonin) has long been used clinically as an antiresorptive agent and has an added analgesic benefit, although the mechanisms contributing to pain relief are unknown. Recently identified nonhormonal regulatory factors have roles in supporting the resorptive phase of bone remodeling by inhibiting bone formation. Sclerostin, for example, is secreted by osteoclasts and negatively regulates bone formation by functioning as a BMP antagonist [471]. c-Src signaling, essential for activation of osteoclast activity, is a negative regulator of Runx2-mediated osteoblast differentiation [58, 472]. Mechanisms regulating the limit of resorption in Howship lacunae are not well understood, but a combination of pathways is likely to be operative. Calcitonin exerts a rapid response for retraction of the osteoclast in response to physiologic Ca++ concentration [473], and the simultaneous elevated phosphate concentrations induce osteoclast apoptosis, as well as up-regulate OPG, thereby limiting resorptive activity by a combination of events [474]. Amylin, a calcitonin family member, was recently reported
242 to inhibit the fusion of osteoclast precursors through an unidentified receptor [475]. Numerous cytokines (e.g., TNFα, RANKL, and IL-1) prolong the ability of osteoclasts to go through successive rounds of resorption. However, the action of TGFβ and estrogen have a negative effect on osteoclast survival time [476]. In response to hormone- and cytokine-induced signal transduction pathways, the fusion of circulating mononuclear precursors to the multinucleated osteoclasts and osteoclast activity is ultimately regulated by the transcriptional targets of signal transduction pathways that control gene expression for commitment, and differentiation of the mononuclear precursors (Fig. 4). Our understanding of the transcriptional mediators that drive osteoclast differentiation and regulate their activation derive largely from characterization of murine animal models that manifest the osteopetrotic disorders of bone resorption [396, 477, 478]. Many nonsteroidal transcriptional regulators are essential for osteoclast differentiation and activity as illustrated in Figure 3. The required extracellular signals for osteoclastogenesis (M-CSF, RANKL, IL-1, αvβ3, Ca++/ calmodulin) are transduced via intracellular proteins to transcriptional control of target genes (illustrated in Fig. 4). The first transcription factor which was implicated in osteoclastogenesis, c-fos, was revealed by a marked osteopetrosis disorder in the null mutant mouse [479]. Bone sections revealed an abundance of macrophages but an absence of osteoclasts suggesting that c-fos promotes proliferation and differentiation of a specific mononuclear precursor committed to the osteoclast pathway [480, 481]. Transgenic mice in which the myeloid and B lymphoid transcription factor PU.1 is ablated, fail to generate macrophages or osteoclasts and also develop osteopetrosis [482]. The mutant mice are successfully rescued by marrow transplantation supporting the concept of the common lineage of osteoclasts and macrophages and that the PU.1 factor is involved in establishing one of the early myeloid cells. Additionally, PU.1 directs expression of M-CSF [483]. RANKL and II-1 signaling induce the transcription factor NF-κB. Mice deficient in both subunits (p50 and p52) of NF-κB result in osteopetrosis [484]. The nuclear factor of activated T cell cytoplasmic (NFAT2/NFATc) transcription factors have been identified as targets of RANKL (NFAT1/NFATc2) in promoting osteoclast differentiation [485–487] and calcium– calmodulin signaling (NFAT2/NFATc1). Transgenic mice overexpressing NFAT1 (NFAT2c) in osteoclasts result in decreased trabecular bone mass [488]. Upon dephosphorylation, these transcription factors translocate from the cytoplasm to the nucleus. The NFATs are now considered transcriptional key regulators of gene expression responsive to signaling pathways essential for osteoclast differentiation.
JANE B. LIAN AND GARY S. STEIN
The MI gene product, Mitf (mentioned earlier), encodes a member of the basic helix-loop-helix leucine zipper protein family that is one of the transcription factor targets of MAPK/p38 signaling. Mitf and the Tfe3 member of this family have redundant roles in osteoclast formation [489, 490]. MI/MI mice develop osteopetrosis, although many TRAP- positive osteoclasts are present in trabecular bones of these mice. However, the osteoclasts fail to form ruffle borders [491], similar to the Src−/− mice. In the final analyses of the regulatory controls for osteoclastogenesis, many levels are operative from the initial ECM microenvironment to the cellular stimulates and intracellular signal transduction pathways that induce transcriptional events.
VI. PERSPECTIVES Our understanding of the biological properties of bone cells is rapidly evolving. Complex regulatory mechanisms that are operative in the establishment of skeletal stem cell populations, as well as in commitment to the osteoblast or osteoclast phenotypes, are being clarified. Although systemic and local factors that promote and regulate differentiation of lineage-specific phenotypes are being identified, a goal for the future is a better understanding of the mechanisms that facilitate interactions between subpopulations of osteoblasts and osteoclasts to support maintenance of bone mass and mineral homeostasis. Pursuing the identity of cell type-specific proteins and transcription factors that regulate their expression has been instructive. As the signaling pathways that mediate expression of genes controlling proliferation and differentiation of bone cells are further clarified, a new generation of options for diagnosis and treatment of skeletal disorders will emerge.
Acknowledgments This work was supported by NIH grants AR39588, DE12528, P01 PO1 AR48818, and P30 DK32520. The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health.
References 1. Karaplis, A. (2002). Embryonic development of bone and the molecular regulation of intramembranous and endochondral bone formation. In “Principles of Bone Biology” (J. P. Bilezikian, L. G. Raisz, and G. A. Rodan, eds.), pp. 33–58. Academic Press, San Diego, CA. 2. Olsen, B. R., Reginato, A. M., and Wang, W. (2000). Bone development. Annu. Rev. Cell Dev. Biol. 16, 191–220.
Chapter 14 The Cells of Bone 3. Zakany, J., Kmita, M., and Duboule, D. (2004). A dual role for Hox genes in limb anterior-posterior asymmetry. Science 304, 1669–1672. 4. Goodman, F. R. (2002). Limb malformations and the human HOX genes. Am. J. Med. Genet. 112, 256–265. 5. Vargesson, N., Patel, K., Lewis, J., and Tickle, C. (1998). Expression patterns of Notch1, Serrate1, Serrate2 and Delta1 in tissues of the developing chick limb. Mech. Dev. 77, 197–199. 6. Yamaguchi, A., Komori, T., and Suda, T. (2000). Regulation of osteoblast differentiation mediated by bone morphogenetic proteins, hedgehogs, and Cbfa1. Endocr. Rev. 21, 393–411. 7. Harfe, B. D., Scherz, P. J., Nissim, S., Tian, H., McMahon, A. P., and Tabin, C. J. (2004). Evidence for an expansion-based temporal Shh gradient in specifying vertebrate digit identities. Cell 118, 517–528. 8. Spinella-Jaegle, S., Rawadi, G., Kawai, S., Gallea, S., Faucheu, C., Mollat, P., Courtois, B., Bergaud, B., Ramez, V., Blanchet, A. M., Adelmant, G., Baron, R., and Roman-Roman, S. (2001). Sonic hedgehog increases the commitment of pluripotent mesenchymal cells into the osteoblastic lineage and abolishes adipocytic differentiation. J. Cell Sci. 114, 2085–2094. 9. Depew, M. J., Lufkin, T., and Rubenstein, J. L. (2002). Specification of jaw subdivisions by Dlx genes. Science 298, 381–385. 10. Ferrari, D., Sumoy, L., Gannon, J., Sun, H., Brown, A. M., Upholt, W. B., and Kosher, R. A. (1995). The expression pattern of the Distal-less homeobox-containing gene Dlx-5 in the developing chick limb bud suggests its involvement in apical ectodermal ridge activity, pattern formation, and cartilage differentiation. Mech. Dev. 52, 257–264. 11. Robledo, R. F., Rajan, L., Li, X., and Lufkin, T. (2002). The Dlx5 and Dlx6 homeobox genes are essential for craniofacial, axial, and appendicular skeletal development. Genes Dev. 16, 1089–1101. 12. Thesleff, I., Vaahtokari, A., Vainio, S., and Jowett, A. (1996). Molecular mechanisms of cell and tissue interactions during early tooth development. Anat. Rec. 245, 151–161. 13. Yoon, B. S. and Lyons, K. M. (2004). Multiple functions of BMPs in chondrogenesis. J. Cell Biochem. 93, 93–103. 14. Boulet, A. M., Moon, A. M., Arenkiel, B. R., and Capecchi, M. R. (2004). The roles of Fgf4 and Fgf8 in limb bud initiation and outgrowth. Dev. Biol. 273, 361–372. 15. Marie, P. J. (2003). Fibroblast growth factor signaling controlling osteoblast differentiation. Gene 316, 23–32. 16. ten Dijke, P., Fu, J., Schaap, P., and Roelen, B. A. (2003). Signal transduction of bone morphogenetic proteins in osteoblast differentiation. J. Bone Joint Surg. Am. 85-A Suppl 3, 34–38. 17. Westendorf, J. J., Kahler, R. A., and Schroeder, T. M. (2004). Wnt signaling in osteoblasts and bone diseases. Gene 341, 19–39. 18. Iseki, S., Wilkie, A. O., and Morriss-Kay, G. M. (1999). Fgfr1 and Fgfr2 have distinct differentiation- and proliferation-related roles in the developing mouse skull vault. Development 126, 5611–5620. 19. Delezoide, A. L., Benoist-Lasselin, C., Legeai-Mallet, L., Le Merrer, M., Munnich, A., Vekemans, M., and Bonaventure, J. (1998). Spatio-temporal expression of FGFR 1, 2 and 3 genes during human embryo-fetal ossification. Mech. Dev. 77, 19–30. 20. Rice, D. P., Rice, R., and Thesleff, I. (2003). Molecular mechanisms in calvarial bone and suture development, and their relation to craniosynostosis. Eur. J. Orthod. 25, 139–148. 21. Sun, X., Mariani, F. V., and Martin, G. R. (2002). Functions of FGF signalling from the apical ectodermal ridge in limb development. Nature 418, 501–508. 22. Zhou, Y. X., Xu, X., Chen, L., Li, C., Brodie, S. G., and Deng, C. X. (2000). A Pro250Arg substitution in mouse Fgfr1 causes increased expression of Cbfa1 and premature fusion of calvarial sutures. Hum. Mol. Genet. 9, 2001–2008. 23. Hajihosseini, M. K., Lalioti, M. D., Arthaud, S., Burgar, H. R., Brown, J. M., Twigg, S. R., Wilkie, A. O., and Heath, J. K. (2004). Skeletal development is regulated by fibroblast growth factor receptor 1 signalling dynamics. Development 131, 325–335.
243 24. Rutland, P., Pulleyn, L. J., Reardon, W., Baraitser, M., Hayward, R., Jones, B., Malcolm, S., Winter, R. M., Oldridge, M., Slaney, S. F., Pook, M. D., and Wilkie, A. O. M. (1995). Identical mutations in the FGFR2 gene cause both Pfeiffer and Crouzon syndrome phenotypes. Nat. Genet. 9, 173–176. 25. Eswarakumar, V. P., Horowitz, M. C., Locklin, R., Morriss-Kay, G. M., and Lonai, P. (2004). A gain-of-function mutation of Fgfr2c demonstrates the roles of this receptor variant in osteogenesis. Proc. Natl Acad. Sci. USA 101, 12555–12560. 26. Baroni, T., Carinci, P., Lilli, C., Bellucci, C., Aisa, M. C., Scapoli, L., Volinia, S., Carinci, F., Pezzetti, F., Calvitti, M., Farina, A., Conte, C., and Bodo, M. (2005). P253R fibroblast growth factor receptor-2 mutation induces RUNX2 transcript variants and calvarial osteoblast differentiation. J. Cell Physiol. 202, 524–535. 27. Tanimoto, Y., Yokozeki, M., Hiura, K., Matsumoto, K., Nakanishi, H., Matsumoto, T., Marie, P. J., and Moriyama, K. (2004). A soluble form of fibroblast growth factor receptor 2 (FGFR2) with S252W mutation acts as an efficient inhibitor for the enhanced osteoblastic differentiation caused by FGFR2 activation in Apert syndrome. J. Biol. Chem. 279, 45926–45934. 28. Ornitz, D. M. and Marie, P. J. (2002). FGF signaling pathways in endochondral and intramembranous bone development and human genetic disease. Genes Dev. 16, 1446–1465. 29. Amizuka, N., Davidson, D., Liu, H., Valverde-Franco, G., Chai, S., Maeda, T., Ozawa, H., Hammond, V., Ornitz, D. M., Goltzman, D., and Henderson, J. E. (2004). Signalling by fibroblast growth factor receptor 3 and parathyroid hormone-related peptide coordinate cartilage and bone development. Bone 34, 13–25. 30. Deng, C., Wynshaw-Boris, A., Zhou, F., Kuo, A., and Leder, P. (1996). Fibroblast growth factor receptor 3 is a negative regulator of bone growth. Cell 84, 911–921. 31. Valverde-Franco, G., Liu, H., Davidson, D., Chai, S., ValderramaCarvajal, H., Goltzman, D., Ornitz, D. M., and Henderson, J. E. (2004). Defective bone mineralization and osteopenia in young adult FGFR3-/mice. Hum. Mol. Genet. 13, 271–284. 32. Abzhanov, A., and Tabin, C. J. (2004). Shh and Fgf8 act synergistically to drive cartilage outgrowth during cranial development. Dev. Biol. 273, 134–148. 33. Laufer, E., Nelson, C. E., Johnson, R. L., Morgan, B. A., and Tabin, C. (1994). Sonic hedgehog and Fgf-4 act through a signaling cascade and feedback loop to integrate growth and patterning of the developing limb bud. Cell 79, 993–1003. 34. Liu, Z., Xu, J., Colvin, J. S., and Ornitz, D. M. (2002). Coordination of chondrogenesis and osteogenesis by fibroblast growth factor 18. Genes Dev. 16, 859–869. 35. Ohbayashi, N., Shibayama, M., Kurotaki, Y., Imanishi, M., Fujimori, T., Itoh, N., and Takada, S. (2002). FGF18 is required for normal cell proliferation and differentiation during osteogenesis and chondrogenesis. Genes Dev. 16, 870–879. 36. Sullivan, R., and Klagsbrun, M. (1985). Purification of cartilagederived growth factor by heparin affinity chromatography. J. Biol. Chem. 260, 2399–2403. 37. Hurley, M. M., Abreu, C., Gronowicz, G., Kawaguchi, H., and Lorenzo, J. (1994). Expression and regulation of basic fibroblast growth factor mRNA levels in mouse osteoblastic MC3T3-E1 cells. J. Biol. Chem. 269, 9392–9396. 38. Tanaka, H., Ogasa, H., Barnes, J., and Liang, C. T. (1999). Actions of bFGF on mitogenic activity and lineage expression in rat osteoprogenitor cells: effect of age. Mol. Cell Endocrinol. 150, 1–10. 39. Debiais, F., Lefevre, G., Lemonnier, J., Le Mee, S., Lasmoles, F., Mascarelli, F., and Marie, P. J. (2004). Fibroblast growth factor-2 induces osteoblast survival through a phosphatidylinositol 3-kinase-dependent, -beta-catenin-independent signaling pathway. Exp. Cell Res. 297, 235–246. 40. Xiao, L., Liu, P., Sobue, T., Lichtler, A., Coffin, J. D., and Hurley, M. M. (2003). Effect of overexpressing fibroblast growth factor 2 protein
244
41.
42.
43.
44.
45.
46.
47. 48. 49.
50.
51.
52.
53.
54.
55.
56.
57.
isoforms in osteoblastic ROS 17/2.8 cells. J. Cell Biochem. 89, 1291–1301. Montero, A., Okada, Y., Tomita, M., Ito, M., Tsurukami, H., Nakamura, T., Doetschman, T., Coffin, J. D., and Hurley, M. M. (2000). Disruption of the fibroblast growth factor-2 gene results in decreased bone mass and bone formation. J. Clin. Invest. 105, 1085–1093. Ahn, K., Mishina, Y., Hanks, M. C., Behringer, R. R., and Crenshaw, E. B., III (2001). BMPR-IA signaling is required for the formation of the apical ectodermal ridge and dorsal-ventral patterning of the limb. Development 128, 4449–4461. Canalis, E., Economides, A. N., and Gazzerro, E. (2003). Bone morphogenetic proteins, their antagonists, and the skeleton. Endocr. Rev. 24, 218–235. Derynck, R., and Zhang, Y. E. (2003). Smad-dependent and Smadindependent pathways in TGF-beta family signalling. Nature 425, 577–584. Zhao, M., Harris, S. E., Horn, D., Geng, Z., Nishimura, R., Mundy, G. R., and Chen, D. (2002). Bone morphogenetic protein receptor signaling is necessary for normal murine postnatal bone formation. J. Cell Biol. 157, 1049–1060. Yi, S. E., Daluiski, A., Pederson, R., Rosen, V., and Lyons, K. M. (2000). The type I BMP receptor BMPRIB is required for chondrogenesis in the mouse limb. Development 127, 621–630. Wrana, J. L. (2000). Crossing Smads. Sci. STKE 2000, RE1. Chubinskaya, S., and Kuettner, K. E. (2003). Regulation of osteogenic proteins by chondrocytes. Int. J. Biochem. Cell Biol. 35, 1323–1340. Settle, S. H., Jr., Rountree, R. B., Sinha, A., Thacker, A., Higgins, K., and Kingsley, D. M. (2003). Multiple joint and skeletal patterning defects caused by single and double mutations in the mouse Gdf6 and Gdf5 genes. Dev. Biol. 254, 116–130. Zhang, D., Ferguson, C. M., O’Keefe, R. J., Puzas, J. E., Rosier, R. N., and Reynolds, P. R. (2002). A role for the BMP antagonist chordin in endochondral ossification. J. Bone Miner. Res. 17, 293–300. Wang, C. K., Omi, M., Ferrari, D., Cheng, H. C., Lizarraga, G., Chin, H. J., Upholt, W. B., Dealy, C. N., and Kosher, R. A. (2004). Function of BMPs in the apical ectoderm of the developing mouse limb. Dev. Biol. 269, 109–122. Archer, C. W., Dowthwaite, G. P., and Francis-West, P. (2003). Development of synovial joints. Birth Defects Res. C. Embryo Today 69, 144–155. Grimsrud, C. D., Romano, P. R., D’Souza, M., Puzas, J. E., Reynolds, P. R., Rosier, R. N., and O’Keefe, R. J. (1999). BMP-6 is an autocrine stimulator of chondrocyte differentiation. J. Bone Miner. Res. 14, 475–482. Balint, E., Lapointe, D., Drissi, H., van der Meijden, C., Young, D. W., van Wijnen, A. J., Stein, J. L., Stein, G. S., and Lian, J. B. (2003). Phenotype discovery by gene expression profiling: mapping of biological processes linked to BMP-2-mediated osteoblast differentiation. J. Cell Biochem. 89, 401–426. Banerjee, C., Javed, A., Choi, J.-Y., Green, J., Rosen, V., van Wijnen, A. J., Stein, J. L., Lian, J. B., and Stein, G. S. (2001). Differential regulation of the two principal Runx2/Cbfa1 N-terminal isoforms in response to bone morphogenetic protein-2 during development of the osteoblast phenotype. Endocrinology 142, 4026–4039. Harris, S. E., Guo, D., Harris, M. A., Krishnaswamy, A., and Lichtler, A. (2003). Transcriptional regulation of BMP-2 activated genes in osteoblasts using gene expression microarray analysis: role of Dlx2 and Dlx5 transcription factors. Front Biosci. 8, s1249–s1265. Lee, K. S., Kim, H. J., Li, Q. L., Chi, X. Z., Ueta, C., Komori, T., Wozney, J. M., Kim, E. G., Choi, J. Y., Ryoo, H. M., and Bae, S. C. (2000). Runx2 is a common target of transforming growth factor beta1 and bone morphogenetic protein 2, and cooperation between runx2 and smad5 induces osteoblast-specific gene expression in the pluripotent mesenchymal precursor cell line C2C12. Mol. Cell Biol. 20, 8783–8792.
JANE B. LIAN AND GARY S. STEIN 58. Zaidi, S. K., Sullivan, A. J., van Wijnen, A. J., Stein, J. L., Stein, G. S., and Lian, J. B. (2002). Integration of Runx and Smad regulatory signals at transcriptionally active subnuclear sites. Proc. Natl Acad. Sci. USA 99, 8048–8053. 59. Afzal, F., Pratap, J., Ito, K., Ito, Y., Stein, J. L., van Wijnen, A. J., Stein, G. S., Lian, J. B., and Javed, A. (2004). Smad function and intranuclear targeting share a Runx2 motif required for osteogenic lineage induction and BMP2 responsive transcription. J. Cell Physiol. [Epub ahead of print]. 60. Zhang, Y. W., Yasui, N., Ito, K., Huang, G., Fujii, M., Hanai, J., Nogami, H., Ochi, T., Miyazono, K., and Ito, Y. (2000). A RUNX2/ PEBP2αA/CBFA1 mutation displaying impaired transactivation and Smad interaction in cleidocranial dysplasia. Proc. Natl Acad. Sci. USA 97, 10549–10554. 61. Yang, S., Wei, D., Wang, D., Phimphilai, M., Krebsbach, P. H., and Franceschi, R. T. (2003). In vitro and in vivo synergistic interactions between the Runx2/Cbfa1 transcription factor and bone morphogenetic protein-2 in stimulating osteoblast differentiation. J. Bone Miner. Res. 18, 705–715. 62. Young, D. W., Pratap, J., Javed, A., Weiner, B., Ohkawa, Y., van Wijnen, A., Montecino, M., Stein, G. S., Stein, J. L., Imbalzano, A. N., and Lian, J. B. (2005). SWI/SNF chromatin remodeling complex is obligatory for BMP2-induced, Runx2-dependent skeletal gene expression that controls osteoblast differentiation. J. Cell Biochem. 94, 720–730. 63. Komori, T., Yagi, H., Nomura, S., Yamaguchi, A., Sasaki, K., Deguchi, K., Shimizu, Y., Bronson, R. T., Gao, Y.-H., Inada, M., Sato, M., Okamoto, R., Kitamura, Y., Yoshiki, S., and Kishimoto, T. (1997). Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89, 755–764. 64. Otto, F., Thornell, A. P., Crompton, T., Denzel, A., Gilmour, K. C., Rosewell, I. R., Stamp, G. W. H., Beddington, R. S. P., Mundlos, S., Olsen, B. R., Selby, P. B., and Owen, M. J. (1997). Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89, 765–771. 65. Banerjee, C., McCabe, L. R., Choi, J.-Y., Hiebert, S. W., Stein, J. L., Stein, G. S., and Lian, J. B. (1997). Runt homology domain proteins in osteoblast differentiation: AML-3/CBFA1 is a major component of a bone specific complex. J. Cell Biochem. 66, 1–8. 66. Ducy, P., Zhang, R., Geoffroy, V., Ridall, A. L., and Karsenty, G. (1997). Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89, 747–754. 67. Chimal-Monroy, J., Rodriguez-Leon, J., Montero, J. A., Ganan, Y., Macias, D., Merino, R., and Hurle, J. M. (2003). Analysis of the molecular cascade responsible for mesodermal limb chondrogenesis: Sox genes and BMP signaling. Dev. Biol. 257, 292–301. 68. Mori-Akiyama, Y., Akiyama, H., Rowitch, D. H., and de Crombrugghe, B. (2003). Sox9 is required for determination of the chondrogenic cell lineage in the cranial neural crest. Proc. Natl Acad. Sci. USA 100, 9360–9365. 69. Semba, I., Nonaka, K., Takahashi, I., Takahashi, K., Dashner, R., Shum, L., Nuckolls, G. H., and Slavkin, H. C. (2000). Positionallydependent chondrogenesis induced by BMP4 is co-regulated by Sox9 and Msx2. Dev. Dyn. 217, 401–414. 70. Zhang, Z., Song, Y., Zhang, X., Tang, J., Chen, J., and Chen, Y. (2003). Msx1/Bmp4 genetic pathway regulates mammalian alveolar bone formation via induction of Dlx5 and Cbfa1. Mech. Dev. 120, 1469–1479. 71. Nakashima, K., Zhou, X., Kunkel, G., Zhang, Z., Deng, J. M., Behringer, R. R., and de Crombrugghe, B. (2002). The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell 108, 17–29. 72. Logan, C. Y. and Nusse, R. (2004). The Wnt signaling pathway in development and disease. Annu. Rev. Cell Dev. Biol. 20, 781–810. 73. Kawano, Y. and Kypta, R. (2003). Secreted antagonists of the Wnt signalling pathway. J. Cell Sci. 116, 2627–2634.
245
Chapter 14 The Cells of Bone 74. Church, V. L., and Francis-West, P. (2002). Wnt signalling during limb development. Int. J. Dev. Biol. 46, 927–936. 75. Ikeya, M., and Takada, S. (2001). Wnt-3a is required for somite specification along the anteroposterior axis of the mouse embryo and for regulation of cdx-1 expression. Mech. Dev. 103, 27–33. 76. Niemann, S., Zhao, C., Pascu, F., Stahl, U., Aulepp, U., Niswander, L., Weber, J. L., and Muller, U. (2004). Homozygous WNT3 mutation causes tetra-amelia in a large consanguineous family. Am. J. Hum. Genet. 74, 558–563. 77. Lako, M., Strachan, T., Bullen, P., Wilson, D. I., Robson, S. C., and Lindsay, S. (1998). Isolation, characterisation and embryonic expression of WNT11, a gene which maps to 11q13.5 and has possible roles in the development of skeleton, kidney and lung. Gene 219, 101–110. 78. Loganathan, P. G., Nimmagadda, S., Huang, R., Scaal, M., and Christ, B. (2005). Comparative analysis of the expression patterns of Wnts during chick limb development. Histochem. Cell Biol. 79. Hartmann, C., and Tabin, C. J. (2001). Wnt-14 plays a pivotal role in inducing synovial joint formation in the developing appendicular skeleton. Cell 104, 341–351. 80. Yamaguchi, T. P., Bradley, A., McMahon, A. P., and Jones, S. (1999). A Wnt5a pathway underlies outgrowth of multiple structures in the vertebrate embryo. Development 126, 1211–1223. 81. Yang, Y., Topol, L., Lee, H., and Wu, J. (2003). Wnt5a and Wnt5b exhibit distinct activities in coordinating chondrocyte proliferation and differentiation. Development 130, 1003–1015. 82. Topol, L., Jiang, X., Choi, H., Garrett-Beal, L., Carolan, P. J., and Yang, Y. (2003). Wnt-5a inhibits the canonical Wnt pathway by promoting GSK-3-independent {beta}-catenin degradation. J. Cell Biol. 162, 899–908. 83. Hartmann, C., and Tabin, C. J. (2000). Dual roles of wnt signaling during chondrogenesis in the chicken limb. Development 127, 3141–3159. 84. Kawakami, Y., Wada, N., Nishimatsu, S. I., Ishikawa, T., Noji, S., and Nohno, T. (1999). Involvement of Wnt-5a in chondrogenic pattern formation in the chick limb bud. Dev. Growth Differ. 41, 29–40. 85. Church, V., Nohno, T., Linker, C., Marcelle, C., and Francis-West, P. (2002). Wnt regulation of chondrocyte differentiation. J. Cell Sci. 115, 4809–4818. 86. Parr, B. A., and McMahon, A. P. (1995). Dorsalizing signal Wnt-7a required for normal polarity of D-V and A-P axes of mouse limb. Nature 374, 350–353. 87. Holmen, S. L., Zylstra, C. R., Mukherjee, A., Sigler, R. E., Faugere, M. C., Bouxsein, M. L., Deng, L., Clemens, T. L., and Williams, B. O. (2005). Essential role of beta -catenin in post natal bone acquisition. J. Biol. Chem. [Epub ahead of print]. 88. Glass, D. A., Bialek, P., Ahn, J. D., Starbuck, M., Patel, M. S., Clevers, H., Taketo, M. M., Long, F., McMahon, A. P., Lang, R. A., and Karsenty, G. (2005). Canonical wnt signaling in differentiated osteoblasts controls osteoclast differentiation. Dev. Cell 8, 751–764. 89. Day, T. F., Guo, X., Garrett-Beal, L., and Yang, Y. (2005). Wnt/betacatenin signaling in mesenchymal progenitors controls osteoblast and chondrocyte differentiation during vertebrate skeletogenesis. Dev. Cell 8, 739–750. 90. Hill, T. P., Spater, D., Taketo, M. M., Birchmeier, W., and Hartmann, C. (2005). Canonical Wnt/beta-catenin signaling prevents osteoblasts from differentiating into chondrocytes. Dev. Cell 8, 727–738. 91. Boyden, L. M., Mao, J., Belsky, J., Mitzner, L., Farhi, A., Mitnick, M. A., Wu, D., Insogna, K., and Lifton, R. P. (2002). High bone density due to a mutation in LDL-receptor-related protein 5. N. Engl. J. Med. 346, 1513–1521. 92. Little, R. D., Recker, R. R., and Johnson, M. L. (2002). High bone density due to a mutation in LDL-receptor-related protein 5. N. Engl. J. Med. 347, 943–944. 93. Akhter, M. P., Wells, D. J., Short, S. J., Cullen, D. M., Johnson, M. L., Haynatzki, G. R., Babij, P., Allen, K. M., Yaworsky, P. J., Bex, F., and
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
Recker, R. R. (2004). Bone biomechanical properties in LRP5 mutant mice. Bone 35, 162–169. Babij, P., Zhao, W., Small, C., Kharode, Y., Yaworsky, P. J., Bouxsein, M. L., Reddy, P. S., Bodine, P. V., Robinson, J. A., Bhat, B., Marzolf, J., Moran, R. A., and Bex, F. (2003). High bone mass in mice expressing a mutant LRP5 gene. J. Bone Miner. Res. 18, 960–974. Gong, Y., Slee, R. B., Fukai, N., Rawadi, G., Roman-Roman, S., Reginato, A. M., Wang, H., Cundy, T., Glorieux, F. H., Lev, D., Zacharin, M., Oexle, K., Marcelino, J., Suwairi, W., Heeger, S., Sabatakos, G., Apte, S., Adkins, W. N., Allgrove, J., Arslan-Kirchner, M., Batch, J. A., Beighton, P., Black, G. C., Boles, R. G., Boon, L. M., Borrone, C., Brunner, H. G., Carle, G. F., Dallapiccola, B., De Paepe, A., Floege, B., Halfhide, M. L., Hall, B., Hennekam, R. C., Hirose, T., Jans, A., Juppner, H., Kim, C. A., Keppler-Noreuil, K., Kohlschuetter, A., LaCombe, D., Lambert, M., Lemyre, E., Letteboer, T., Peltonen, L., Ramesar, R. S., Romanengo, M., Somer, H., Steichen-Gersdorf, E., Steinmann, B., Sullivan, B., Superti-Furga, A., Swoboda, W., van den Boogaard, M. J., van Hul, W., Vikkula, M., Votruba, M., Zabel, B., Garcia, T., Baron, R., Olsen, B. R., and Warman, M. L. (2001). LDL receptor-related protein 5 (LRP5) affects bone accrual and eye development. Cell 107, 513–523. Van Wesenbeeck, L., Cleiren, E., Gram, J., Beals, R. K., Benichou, O., Scopelliti, D., Key, L., Renton, T., Bartels, C., Gong, Y., Warman, M. L., de Vernejoul, M. C., Bollerslev, J., and van Hul, W. (2003). Six novel missense mutations in the LDL receptor-related protein 5 (LRP5) gene in different conditions with an increased bone density. Am. J. Hum. Genet. 72, 763–771. Kato, M., Patel, M. S., Levasseur, R., Lobov, I., Chang, B. H., Glass, D. A., Hartmann, C., Li, L., Hwang, T. H., Brayton, C. F., Lang, R. A., Karsenty, G., and Chan, L. (2002). Cbfa1-independent decrease in osteoblast proliferation, osteopenia, and persistent embryonic eye vascularization in mice deficient in Lrp5, a Wnt coreceptor. J. Cell Biol. 157, 303–314. Bodine, P. V., Zhao, W., Kharode, Y. P., Bex, F. J., Lambert, A. J., Goad, M. B., Gaur, T., Stein, G. S., Lian, J. B., and Komm, B. S. (2004). The Wnt antagonist secreted frizzled-related protein-1 is a negative regulator of trabecular bone formation in adult mice. Mol. Endocrinol. 18, 1222–1237. Bennett, C. N., Longo, K. A., Wright, W. S., Suva, L. J., Lane, T. F., Hankenson, K. D., and MacDougald, O. A. (2005). Regulation of osteoblastogenesis and bone mass by Wnt10b. Proc. Natl Acad. Sci. USA 102, 3324–3329. Longo, K. A., Wright, W. S., Kang, S., Gerin, I., Chiang, S. H., Lucas, P. C., Opp, M. R., and MacDougald, O. A. (2004). Wnt10b inhibits development of white and brown adipose tissues. J. Biol. Chem. 279, 35503–35509. Minina, E., Kreschel, C., Naski, M. C., Ornitz, D. M., and Vortkamp, A. (2002). Interaction of FGF, Ihh/Pthlh, and BMP signaling integrates chondrocyte proliferation and hypertrophic differentiation. Dev. Cell 3, 439–449. Schipani, E., Lanske, B., Hunzelman, J., Luz, A., Kovacs, C. S., Lee, K., Pirro, A., Kronenberg, H. M., and Juppner, H. (1997). Targeted expression of constitutively active receptors for parathyroid hormone and parathyroid hormone-related peptide delays endochondral bone formation and rescues mice that lack parathyroid hormone-related peptide. Proc. Natl Acad. Sci.USA 94, 13689–13694. Weir, E. C., Philbrick, W. M., Amling, M., Neff, L. A., Baron, R., and Broadus, A. E. (1996). Targeted overexpression of parathyroid hormonerelated peptide in chondrocytes causes chondrodysplasia and delayed endochondral bone formation. Proc. Natl Acad. Sci. USA 93, 10240–10245. Lanske, B., Karaplis, A. C., Lee, K., Luz, A., Vortkamp, A., Pirro, A., Karperien, M., Defize, L. H. K., Ho, C., Mulligan, R. C.,
246
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
JANE B. LIAN AND GARY S. STEIN Abou-Samra, A. B., Juppner, H., Segre, G. V., and Kronenberg, H. M. (1996). PTH/PTHrP receptor in early development and Indian hedgehog-regulated bone growth. Science 273, 663–666. Long, F., Chung, U. I., Ohba, S., McMahon, J., Kronenberg, H. M., and McMahon, A. P. (2004). Ihh signaling is directly required for the osteoblast lineage in the endochondral skeleton. Development 131, 1309–1318. St Jacques, B., Hammerschmidt, M., and McMahon, A. P. (1999). Indian hedgehog signaling regulates proliferation and differentiation of chondrocytes and is essential for bone formation. Genes Dev. 13, 2072–2086. Kobayashi, T., Chung, U. I., Schipani, E., Starbuck, M., Karsenty, G., Katagiri, T., Goad, D. L., Lanske, B., and Kronenberg, H. M. (2002). PTHrP and Indian hedgehog control differentiation of growth plate chondrocytes at multiple steps. Development 129, 2977–2986. Schipani, E., and Provot, S. (2003). PTHrP, PTH, and the PTH/PTHrP receptor in endochondral bone development. Birth Defects Res. C. Embryo. Today 69, 352–362. Robson, H., Siebler, T., Stevens, D. A., Shalet, S. M., and Williams, G. R. (2000). Thyroid hormone acts directly on growth plate chondrocytes to promote hypertrophic differentiation and inhibit clonal expansion and cell proliferation. Endocrinology 141, 3887–3897. O’Shea, P. J., Harvey, C. B., Suzuki, H., Kaneshige, M., Kaneshige, K., Cheng, S. Y., and Williams, G. R. (2003). A thyrotoxic skeletal phenotype of advanced bone formation in mice with resistance to thyroid hormone. Mol. Endocrinol. 17, 1410–1424. Li, T. F., Zuscik, M. J., Ionescu, A. M., Zhang, X., Rosier, R. N., Schwarz, E. M., Drissi, H., and O’Keefe, R. J. (2004). PGE2 inhibits chondrocyte differentiation through PKA and PKC signaling. Exp. Cell Res. 300, 159–169. Vermot, J., Niederreither, K., Garnier, J. M., Chambon, P., and Dolle, P. (2003). Decreased embryonic retinoic acid synthesis results in a DiGeorge syndrome phenotype in newborn mice. Proc. Natl Acad. Sci. USA 100, 1763–1768. Niederreither, K., Vermot, J., Schuhbaur, B., Chambon, P., and Dolle, P. (2002). Embryonic retinoic acid synthesis is required for forelimb growth and anteroposterior patterning in the mouse. Development 129, 3563–3574. Allen, S. P., Maden, M., and Price, J. S. (2002). A role for retinoic acid in regulating the regeneration of deer antlers. Dev. Biol. 251, 409–423. Weston, A. D., Chandraratna, R. A., Torchia, J., and Underhill, T. M. (2002). Requirement for RAR-mediated gene repression in skeletal progenitor differentiation. J. Cell Biol. 158, 39–51. Tseng, K. F., and Goldstein, S. A. (1998). Systemic over-secretion of growth hormone in transgenic mice results in a specific pattern of skeletal modeling and adaptation. J. Bone Miner. Res. 13, 706–715. Tribioli, C., and Lufkin, T. (1999). The murine Bapx1 homeobox gene plays a critical role in embryonic development of the axial skeleton and spleen. Development 126, 5699–5711. Akazawa, H., Komuro, I., Sugitani, Y., Yazaki, Y., Nagai, R., and Noda, T. (2000). Targeted disruption of the homeobox transcription factor Bapx1 results in lethal skeletal dysplasia with asplenia and gastroduodenal malformation. Genes Cells 5, 499–513. Lengner, C. J., Drissi, H., Choi, J.-Y., van Wijnen, A. J., Stein, J. L., Stein, G. S., and Lian, J. B. (2002). Activation of the bone related Runx2/Cbfa1 promoter in mesenchymal condensations and developing chondrocytes of the axial skeleton. Mech. Dev. 114, 167–170. Zeng, L., Kempf, H., Murtaugh, L. C., Sato, M. E., and Lassar, A. B. (2002). Shh establishes an Nkx3.2/Sox9 autoregulatory loop that is maintained by BMP signals to induce somitic chondrogenesis. Genes Dev. 16, 1990–2005. Murtaugh, L. C., Zeng, L., Chyung, J. H., and Lassar, A. B. (2001). The chick transcriptional repressor Nkx3.2 acts downstream of Shh to promote BMP-dependent axial chondrogenesis. Dev. Cell 1, 411–422.
122. Wilson, J., and Tucker, A. S. (2004). Fgf and Bmp signals repress the expression of Bapx1 in the mandibular mesenchyme and control the position of the developing jaw joint. Dev. Biol. 266, 138–150. 123. Lettice, L. A., Purdie, L. A., Carlson, G. J., Kilanowski, F., Dorin, J., and Hill, R. E. (1999). The mouse bagpipe gene controls development of axial skeleton, skull, and spleen. Proc. Natl Acad. Sci. USA 96, 9695–9700. 124. Akiyama, H., Chaboissier, M. C., Martin, J. F., Schedl, A., and de Crombrugghe, B. (2002). The transcription factor Sox9 has essential roles in successive steps of the chondrocyte differentiation pathway and is required for expression of Sox5 and Sox6. Genes Dev. 16, 2813–2828. 125. Lengner, C. J., Hassan, M. Q., Serra, R. W., Lepper, C., van Wijnen, A. J., Stein, J. L., Lian, J. B., and Stein, G. S. (2005). Nkx3.2 mediated repression of RUNX2 promotes chondrogenic differentiation. J. Biol. Chem. 280, 15872–15879. 126. Eames, B. F., Sharpe, P. T., and Helms, J. A. (2004). Hierarchy revealed in the specification of three skeletal fates by Sox9 and Runx2. Dev. Biol. 274, 188–200. 127. Jimenez, M. J., Balbin, M., Alvarez, J., Komori, T., Bianco, P., Holmbeck, K., Birkedal-Hansen, H., Lopez, J. M., and Lopez-Otin, C. (2001). A regulatory cascade involving retinoic acid, Cbfa1, and matrix metalloproteinases is coupled to the development of a process of perichondrial invasion and osteogenic differentiation during bone formation. J. Cell Biol. 155, 1333–1344. 128. Zelzer, E., McLean, W., Ng, Y. S., Fukai, N., Reginato, A. M., Lovejoy, S., d’Amore, P. A., and Olsen, B. R. (2002). Skeletal defects in VEGF(120/120) mice reveal multiple roles for VEGF in skeletogenesis. Development 129, 1893–1904. 129. Pratap, J., Vashi, S., Zhang, J., Languino, L., van Wijnen, A. J., Stein, J. L., Lian, J. B., and Stein, G. S. (2004). Runx2 regulates transcription of gelatinases (MMP9) in metastatic cancer cell lines and functionally related to cell migration. J. Bone Miner. Res. 19[Suppl 1], S123. (Abstract). 130. Stricker, S., Fundele, R., Vortkamp, A., and Mundlos, S. (2002). Role of runx genes in chondrocyte differentiation. Dev. Biol. 245, 95–108. 131. Iwamoto, M., Kitagaki, J., Tamamura, Y., Gentili, C., Koyama, E., Enomoto, H., Komori, T., Pacifici, M., and Enomoto-Iwamoto, M. (2003). Runx2 expression and action in chondrocytes are regulated by retinoid signaling and parathyroid hormone-related peptide (PTHrP). Osteoarthritis Cartilage 11, 6–15. 132. Takeda, S., Bonnamy, J. P., Owen, M. J., Ducy, P., and Karsenty, G. (2001). Continuous expression of Cbfa1 in nonhypertrophic chondrocytes uncovers its ability to induce hypertrophic chondrocyte differentiation and partially rescues Cbfa1-deficient mice. Genes Dev. 15, 467–481. 133. Inada, M., Yasui, T., Nomura, S., Miyake, S., Deguchi, K., Himeno, M., Sato, M., Yamagiwa, H., Kimura, T., Yasui, N., Ochi, T., Endo, N., Kitamura, Y., Kishimoto, T., and Komori, T. (1999). Maturational disturbance of chondrocytes in Cbfa1-deficient mice. Dev. Dyn. 214, 279–290. 134. Kim, I. S., Otto, F., Zabel, B., and Mundlos, S. (1999). Regulation of chondrocyte differentiation by Cbfa1. Mech. Dev. 80, 159–170. 135. Kundu, M., Javed, A., Jeon, J. P., Horner, A., Shum, L., Eckhaus, M., Muenke, M., Lian, J. B., Yang, Y., Nuckolls, G. H., Stein, G. S., and Liu, P. P. (2002). Cbfbeta interacts with Runx2 and has a critical role in bone development. Nat. Genet. 32, 639–644. 136. Miller, J., Horner, A., Stacy, T., Lowry, C., Lian, J. B., Stein, G., Nuckolls, G. H., and Speck, N. A. (2002). The core binding factor beta subunit is required for bone formation and hematopoietic maturation. Nat Genet. 32, 645–649. 137. Takamoto, M., Tsuji, K., Yamashita, T., Sasaki, H., Yano, T., Taketani, Y., Komori, T., Nifuji, A., and Noda, M. (2003). Hedgehog signaling enhances core-binding factor a1 and receptor activator of nuclear factor-kappaB ligand (RANKL) gene expression in chondrocytes. J. Endocrinol. 177, 413–421.
Chapter 14 The Cells of Bone 138. Komori, T. (2003). Requisite roles of Runx2 and Cbfb in skeletal development. J. Bone Miner. Metab. 21, 193–197. 139. Ginis, I., Luo, Y., Miura, T., Thies, S., Brandenberger, R., GerechtNir, S., Amit, M., Hoke, A., Carpenter, M. K., Itskovitz-Eldor, J., and Rao, M. S. (2004). Differences between human and mouse embryonic stem cells. Dev. Biol. 269, 360–380. 140. Tanaka, H., Murphy, C. L., Murphy, C., Kimura, M., Kawai, S., and Polak, J. M. (2004). Chondrogenic differentiation of murine embryonic stem cells: effects of culture conditions and dexamethasone. J. Cell Biochem. 93, 454–462. 141. Sottile, V., Thomson, A., and McWhir, J. (2003). In vitro osteogenic differentiation of human ES cells. Cloning Stem Cells 5, 149–155. 142. zur Nieden, N. I., Kempka, G., and Ahr, H. J. (2003). In vitro differentiation of embryonic stem cells into mineralized osteoblasts. Differentiation 71, 18–27. 143. Bourne, S., Polak, J. M., Hughes, S. P., and Buttery, L. D. (2004). Osteogenic differentiation of mouse embryonic stem cells: differential gene expression analysis by cDNA microarray and purification of osteoblasts by cadherin-11 magnetically activated cell sorting. Tissue Eng. 10, 796–806. 144. Kramer, J., Hegert, C., and Rohwedel, J. (2003). In vitro differentiation of mouse ES cells: bone and cartilage. Methods Enzymol 365, 251–268. 145. Bronson, S. K. (2003). Bone nodule formation via in vitro differentiation of murine embryonic stem cells. Methods Enzymol 365, 241–251. 146. Tsuneto, M., Yamane, T., Okuyama, H., Yamazaki, H., and Hayashi, S. (2003). In vitro differentiation of mouse ES cells into hematopoietic, endothelial, and osteoblastic cell lineages: the possibility of in vitro organogenesis. Methods Enzymol 365, 98–114. 147. Petersen, B. E., Bowen, W. C., Patrene, K. D., Mars, W. M., Sullivan, A. K., Murase, N., Boggs, S. S., Greenberger, J. S., and Goff, J. P. (1999). Bone marrow as a potential source of hepatic oval cells. Science 284, 1168–1170. 148. Makino, S., Fukuda, K., Miyoshi, S., Konishi, F., Kodama, H., Pan, J., Sano, M., Takahashi, T., Hori, S., Abe, H., Hata, J., Umezawa, A., and Ogawa, S. (1999). Cardiomyocytes can be generated from marrow stromal cells in vitro. J. Clin. Invest. 103, 697–705. 149. Jiang, Y., Jahagirdar, B. N., Reinhardt, R. L., Schwartz, R. E., Keene, C. D., Ortiz-Gonzalez, X. R., Reyes, M., Lenvik, T., Lund, T., Blackstad, M., Du, J., Aldrich, S., Lisberg, A., Low, W. C., Largaespada, D. A., and Verfaillie, C. M. (2002). Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 418, 41–49. 150. Olmsted-Davis, E. A., Gugala, Z., Camargo, F., Gannon, F. H., Jackson, K., Kienstra, K. A., Shine, H. D., Lindsey, R. W., Hirschi, K. K., Goodell, M. A., Brenner, M. K., and Davis, A. R. (2003). Primitive adult hematopoietic stem cells can function as osteoblast precursors. Proc. Natl Acad Sci. USA 100, 15877–15882. 151. Grove, J. E., Bruscia, E., and Krause, D. S. (2004). Plasticity of bone marrow-derived stem cells. Stem Cells 22, 487–500. 152. Dunbar, C. E., Tisdale, J., Yu, J. M., Soma, T., Zujewski, J., Bodine, D., Sellers, S., Cowan, K., Donahue, R., and Emmons, R. (1997). Transduction of hematopoietic stem cells in humans and in nonhuman primates. Stem Cells 15, 135–139 (Review). 153. Bruder, S. P., Ricalton, N. S., Boynton, R. E., Connolly, T. J., Jaiswal, N., Zaia, J., and Barry, F. P. (1998). Mesenchymal stem cell surface antigen SB-10 corresponds to activated leukocyte cell adhesion molecule and is involved in osteogenic differentiation. J. Bone Miner. Res. 13, 655–663. 154. Gronthos, S., Graves, S. E., Ohta, S., and Simmons, P. J. (1994). The STRO-1+ fraction of adult human bone marrow contains the osteogenic precursors. Blood 84, 4164–4173. 155. Haynesworth, S. E., Baber, M. A., and Caplan, A. I. (1992). Cell surface antigens on human marrow-derived mesenchymal cells are detected by monoclonal antibodies. Bone 13, 69–80.
247 156. Joyner, C. J., Bennett, A., and Triffitt, J. T. (1997). Identification and enrichment of human osteoprogenitor cells by using differentiation stage-specific monoclonal antibodies. Bone 21, 1–6. 157. Turksen, K., and Aubin, J. E. (1991). Positive and negative immunoselection for enrichment of two classes of osteoprogenitor cells. J. Cell Biol. 114, 373–384. 158. Otto, W. R., and Rao, J. (2004). Tomorrow’s skeleton staff: mesenchymal stem cells and the repair of bone and cartilage. Cell Prolif. 37, 97–110. 159. Chambers, I., and Smith, A. (2004). Self-renewal of teratocarcinoma and embryonic stem cells. Oncogene 23, 7150–7160. 160. Heng, B. C., Cao, T., Stanton, L. W., Robson, P., and Olsen, B. (2004). Strategies for directing the differentiation of stem cells into the osteogenic lineage in vitro. J. Bone Miner. Res. 19, 1379–1394. 161. Friedenstein, A. J., Chailakhjan, R. K., and Lalykina, K. S. (1970). The development of fibroblast colonies in monolayer cultures of guinea-pig bone marrow and spleen cells. Cell Tissue Kinet. 3, 393–403. 162. Friedenstein, A. J., Chailakhyan, R. K., and Gerasimov, U. V. (1987). Bone marrow osteogenic stem cells: in vitro cultivation and transplantation in diffusion chambers. Cell Tissue Kinet. 20, 263–272. 163. Friedenstein, A. J., Latzinik, N. V., Gorskaya Yu. F., Luria, E. A., and Moskvina, I. L. (1992). Bone marrow stromal colony formation requires stimulation by haemopoietic cells. Bone Miner. 18, 199–213. 164. Horowitz, M. C., and Lorenzo, J. A. (2004). The origins of osteoclasts. Curr. Opin. Rheumatol. 16, 464–468. 165. Malaval, L., and Aubin, J. E. (2001). Biphasic effects of leukemia inhibitory factor on osteoblastic differentiation. J. Cell Biochem. 81, 63–70. 166. Metcalf, D. (2003). The unsolved enigmas of leukemia inhibitory factor. Stem Cells 21, 5–14. 167. Owen, M. E., Cave, J., and Joyner, C. J. (1987). Clonal analysis in vitro of osteogenic differentiation of marrow CFU-F. J. Cell Sci. 87, 731–738. 168. Chaudhary, L. R., and Hruska, K. A. (2001). The cell survival signal Akt is differentially activated by PDGF-BB, EGF, and FGF-2 in osteoblastic cells. J. Cell Biochem. 81, 304–311. 169. Nakamura, T., Hanada, K., Tamura, M., Shibanushi, T., Nigi, H., Tagawa, M., Fukumoto, S., and Matsumoto, T. (1995). Stimulation of endosteal bone formation by systemic injections of recombinant basic fibroblast growth factor in rats. Endocrinology 136, 1276–1284. 170. Noda, M., and Camilliere, J. J. (1989). In vivo stimulation of bone formation by transforming growth factor-beta. Endocrinology 124, 2991–2994. 171. Hock, J. M., Canalis, E., and Centrella, M. (1990). Transforming growth factor-beta stimulates bone matrix apposition and bone cell replication in cultured fetal rat calvariae. Endocrinology 126, 421–426. 172. Hauschka, P. V., Mavrakos, A. E., Iafrati, M. D., Doleman, S. E., and Klagsbrun, M. (1986). Growth factors in bone matrix. J. Biol. Chem. 261, 12665–12674. 173. Long, M. W., Robinson, J. A., Ashcraft, E. A., and Mann, K. G. (1995). Regulation of human bone marrow-derived osteoprogenitor cells by osteogenic growth factors. J. Clin. Invest. 95, 881–887. 174. Gazzerro, E., Du, Z., Devlin, R. D., Rydziel, S., Priest, L., Economides, A. N., and Canalis, E. (2003). Noggin arrests stromal cell differentiation in vitro. Bone 32, 111–119. 175. Yoshida, Y., Tanaka, S., Umemori, H., Minowa, O., Usui, M., Ikematsu, N., Hosoda, E., Imamura, T., Kuno, J., Yamashita, T., Miyazono, K., Noda, M., Noda, T., and Yamamoto, T. (2000). Negative regulation of BMP/Smad signaling by Tob in osteoblasts. Cell 103, 1085–1097. 176. Yoshida, Y., von Bubnoff, A., Ikematsu, N., Blitz, I. L., Tsuzuku, J. K., Yoshida, E. H., Umemori, H., Miyazono, K., Yamamoto, T., and
248
177.
178.
179.
180.
181.
182.
183.
184.
185. 186.
187.
188.
189.
190.
191.
192.
JANE B. LIAN AND GARY S. STEIN Cho, K. W. (2003). Tob proteins enhance inhibitory Smad-receptor interactions to repress BMP signaling. Mech. Dev. 120, 629–637. Thies, R. S., Bauduy, M., Ashton, B. A., Kurtzberg, L., Wozney, J. M., and Rosen, V. (1992). Recombinant human bone morphogenetic protein-2 induces osteoblastic differentiation in W-20-17 stromal cells. Endocrinology 130, 1318–1324. Wang, E. A., Israel, D. I., Kelly, S., and Luxenberg, D. P. (1993). Bone morphogenetic protein-2 causes commitment and differentiation in C3H10T1/2 and 3T3 cells. Growth Factors 9, 57–71. Katagiri, T., Yamaguchi, A., Komaki, M., Abe, E., Takahashi, N., Ikeda, T., Rosen, V., Wozney, J. M., Fujisawa-Sehara, A., and Suda, T. (1994). Bone morphogenetic protein-2 converts the differentiation pathway of C2C12 myoblasts into the osteoblast lineage. J. Cell Biol. 127, 1755–1766. Einhorn, T. A., Majeska, R. J., Mohaideen, A., Kagel, E. M., Bouxsein, M. L., Turek, T. J., and Wozney, J. M. (2003). A single percutaneous injection of recombinant human bone morphogenetic protein-2 accelerates fracture repair. J. Bone Joint Surg. Am. 85-A, 1425–1435. Hock, J. M., Fonseca, J., Gunness-Hey, M., Kemp, B. E., and Martin, T. J. (1989). Comparison of the anabolic effects of synthetic parathyroid hormone-related protein (PTHrP) 1–34 and PTH 1–34 on bone in rats. Endocrinology 125, 2022–2027. Shalhoub, V., Aslam, F., Breen, E., van Wijnen, A., Bortell, R., Stein, G. S., Stein, J. L., and Lian, J. B. (1998). Multiple levels of steroid hormone-dependent control of osteocalcin during osteoblast differentiation: glucocorticoid regulation of basal and vitamin D stimulated gene expression. J. Cell Biochem. 69, 154–168. Delany, A. M., Durant, D., and Canalis, E. (2001). Glucocorticoid suppression of IGF I transcription in osteoblasts. Mol. Endocrinol. 15, 1781–1789. McCarthy, T. L., Ji, C., Chen, Y., Kim, K., and Centrella, M. (2000). Time- and dose-related interactions between glucocorticoid and cyclic adenosine 3′,5′-monophosphate on CCAAT/enhancer-binding proteindependent insulin-like growth factor I expression by osteoblasts. Endocrinology 141, 127–137. Rosen, C. J. (2003). The cellular and clinical parameters of anabolic therapy for osteoporosis. Crit. Rev. Eukaryot. Gene Expr. 13, 25–38. Qin, L., Tamasi, J., Raggatt, L., Li, X., Feyen, J. H., Lee, D. C., Dicicco-Bloom, E., and Partridge, N. C. (2005). Amphiregulin is a novel growth factor involved in normal bone development and in the cellular response to parathyroid hormone stimulation. J. Biol. Chem. 280, 3974–3981. Qin, L., Li, X., Ko, J. K., and Partridge, N. C. (2005). Parathyroid hormone uses multiple mechanisms to arrest the cell cycle progression of osteoblastic cells from G1 to S phase. J. Biol. Chem. 280, 3104–3111. Gensure, R. C., Gardella, T. J., and Juppner, H. (2005). Parathyroid hormone and parathyroid hormone-related peptide, and their receptors. Biochem. Biophys. Res. Commun. 328, 666–678. Karaplis, A. C., Luz, A., Glowacki, J., Bronson, R. T., Tybulewicz, V. L., Kronenberg, H. M., and Mulligan, R. C. (1994). Lethal skeletal dysplasia from targeted disruption of the parathyroid hormone-related peptide gene. Genes Dev. 8, 277–289. Miao, D., Li, J., Xue, Y., Su, H., Karaplis, A. C., and Goltzman, D. (2004). Parathyroid hormone-related peptide is required for increased trabecular bone volume in parathyroid hormone-null mice. Endocrinology 145, 3554–3562. Liu, F., Malaval, L., and Aubin, J. E. (2003). Global amplification polymerase chain reaction reveals novel transitional stages during osteoprogenitor differentiation. J. Cell Sci. 116, 1787–1796. Watanabe, N., Tezuka, Y., Matsuno, K., Miyatani, S., Morimura, N., Yasuda, M., Fujimaki, R., Kuroda, K., Hiraki, Y., Hozumi, N., and Tezuka, K. (2003). Suppression of differentiation and proliferation of early chondrogenic cells by Notch. J. Bone Miner. Metab. 21, 344–352.
193. Tezuka, K., Yasuda, M., Watanabe, N., Morimura, N., Kuroda, K., Miyatani, S., and Hozumi, N. (2002). Stimulation of osteoblastic cell differentiation by Notch. J. Bone Miner. Res. 17, 231–239. 194. Sowa, H., Kaji, H., Hendy, G. N., Canaff, L., Komori, T., Sugimoto, T., and Chihara, K. (2004). Menin is required for bone morphogenetic protein 2- and transforming growth factor beta-regulated osteoblastic differentiation through interaction with Smads and Runx2. J. Biol. Chem. 279, 40267–40275. 195. Chan, G. K., Miao, D., Deckelbaum, R., Bolivar, I., Karaplis, A., and Goltzman, D. (2003). Parathyroid hormone-related peptide interacts with bone morphogenetic protein 2 to increase osteoblastogenesis and decrease adipogenesis in pluripotent C3H10T 1/2 mesenchymal cells. Endocrinology 144, 5511–5520. 196. Settleman, J. (2004). Tension precedes commitment-even for a stem cell. Mol. Cell 14, 148–150. 197. McBeath, R., Pirone, D. M., Nelson, C. M., Bhadriraju, K., and Chen, C. S. (2004). Cell shape, cytoskeletal tension, and RhoA regulate stem cell lineage commitment. Dev. Cell 6, 483–495. 198. Zayzafoon, M., Gathings, W. E., and McDonald, J. M. (2004). Modeled microgravity inhibits osteogenic differentiation of human mesenchymal stem cells and increases adipogenesis. Endocrinology 145, 2421–2432. 199. Ontiveros, C., Irwin, R., Wiseman, R. W., and McCabe, L. R. (2004). Hypoxia suppresses runx2 independent of modeled microgravity. J. Cell Physiol. 200, 169–176. 200. Tang, Q. Q., Otto, T. C., and Lane, M. D. (2003). CCAAT/enhancerbinding protein beta is required for mitotic clonal expansion during adipogenesis. Proc. Natl Acad. Sci.USA 100, 850–855. 201. Mueller, E., Drori, S., Aiyer, A., Yie, J., Sarraf, P., Chen, H., Hauser, S., Rosen, E. D., Ge, K., Roeder, R. G., and Spiegelman, B. M. (2002). Genetic analysis of adipogenesis through peroxisome proliferatoractivated receptor gamma isoforms. J. Biol. Chem. 277, 41925–41930. 202. de Crombrugghe, B., Lefebvre, V., Behringer, R. R., Bi, W., Murakami, S., and Huang, W. (2000). Transcriptional mechanisms of chondrocyte differentiation. Matrix Biol. 19, 389–394. 203. Skillington, J., Choy, L., and Derynck, R. (2002). Bone morphogenetic protein and retinoic acid signaling cooperate to induce osteoblast differentiation of preadipocytes. J. Cell Biol. 159, 135–146. 204. Shi, X., Shi, W., Li, Q., Song, B., Wan, M., Bai, S., and Cao, X. (2003). A glucocorticoid-induced leucine-zipper protein, GILZ, inhibits adipogenesis of mesenchymal cells. EMBO Rep. 4, 374–380. 205. Lee, M. H., Javed, A., Kim, H. J., Shin, H. I., Gutierrez, S., Choi, J. Y., Rosen, V., Stein, J. L., van Wijnen, A. J., Stein, G. S., Lian, J. B., and Ryoo, H. M. (1999). Transient upregulation of CBFA1 in response to bone morphogenetic protein-2 and transforming growth factor β1 in C2C12 myogenic cells coincides with suppression of the myogenic phenotype but is not sufficient for osteoblast differentiation. J. Cell Biochem. 73, 114–125. 206. Gori, F., Thomas, T., Hicok, K. C., Spelsberg, T. C., and Riggs, B. L. (1999). Differentiation of human marrow stromal precursor cells: bone morphogenetic protein-2 increases OSF2/CBFA1, enhances osteoblast commitment, and inhibits late adipocyte maturation. J. Bone Miner. Res. 14, 1522–1535. 207. Jeon, M. J., Kim, J. A., Kwon, S. H., Kim, S. W., Park, K. S., Park, S. W., Kim, S. Y., and Shin, C. S. (2003). Activation of peroxisome proliferator-activated receptor-gamma inhibits the Runx2-mediated transcription of osteocalcin in osteoblasts. J. Biol. Chem. 278, 23270–23277. 208. Akune, T., Ohba, S., Kamekura, S., Yamaguchi, M., Chung, U. I., Kubota, N., Terauchi, Y., Harada, Y., Azuma, Y., Nakamura, K., Kadowaki, T., and Kawaguchi, H. (2004). PPARgamma insufficiency enhances osteogenesis through osteoblast formation from bone marrow progenitors. J. Clin. Invest. 113, 846–855. 209. Young, D. W., Zaidi, S. K., Furcinitti, P. S., Javed, A., van Wijnen, A. J., Stein, J. L., Lian, J. B., and Stein, G. S. (2004). Quantitative signature
Chapter 14 The Cells of Bone
210.
211.
212.
213. 214.
215.
216.
217.
218.
219.
220.
221.
222.
223.
224.
225.
226.
for architectural organization of regulatory factors using intranuclear informatics. J. Cell Sci. 117, 4889–4896. Kobayashi, H., Gao, Y., Ueta, C., Yamaguchi, A., and Komori, T. (2000). Multilineage differentiation of Cbfa1-deficient calvarial cells in vitro. Biochem. Biophys. Res. Commun. 273, 630–636. Pratap, J., Galindo, M., Zaidi, S. K., Vradii, D., Bhat, B. M., Robinson, J. A., Choi, J.-Y., Komori, T., Stein, J. L., Lian, J. B., Stein, G. S., and van Wijnen, A. J. (2003). Cell growth regulatory role of Runx2 during proliferative expansion of pre-osteoblasts. Cancer Res. 63, 5357–5362. Stein, G. S., Lian, J. B., Stein, J. L., van Wijnen, A. J., and Montecino, M. (1996). Transcriptional control of osteoblast growth and differentiation. Physiol. Rev. 76, 593–629. Hughes, D. E., and Boyce, B. F. (1997). Apoptosis in bone physiology and disease. Mol. Pathol. 50, 132–137. Noble, B. S., Stevens, H., Loveridge, N., and Reeve, J. (1997). Identification of apoptotic changes in osteocytes in normal and pathological human bone. Bone 20, 273–282. Parfitt, A. M., Villanueva, A. R., Foldes, J., and Rao, D. S. (1995). Relations between histologic indices of bone formation: implications for the pathogenesis of spinal osteoporosis. J. Bone Miner. Res. 10, 466–473. Lynch, M., Capparelli, C., Stein, J. L., Stein, G. S., and Lian, J. B. (1998). Apoptosis during bone-like tissue development in vitro. J. Cell Biochem. 68, 31–49. Tomkinson, A., Reeve, J., Shaw, R. W., and Noble, B. S. (1997). The death of osteocytes via apoptosis accompanies estrogen withdrawal in human bone. J. Clin. Endocrinol. Metab. 82, 3128–3135. Lynch, M. P., Stein, J. L., Stein, G. S., and Lian, J. B. (1995). The influence of type I collagen on the development and maintenance of the osteoblast phenotype in primary and passaged rat calvarial osteoblasts: modification of expression of genes supporting cell growth, adhesion, and extracellular matrix mineralization. Exp. Cell Res. 216, 35–45. Vukicevic, S., Luyten, F. P., Kleinman, H. K., and Reddi, A. H. (1990). Differentiation of canalicular cell processes in bone cells by basement membrane matrix components: regulation by discrete domains of laminin. Cell 63, 437–445. Andrianarivo, A. G., Robinson, J. A., Mann, K. G., and Tracy, R. P. (1992). Growth on type I collagen promotes expression of the osteoblastic phenotype in human osteosarcoma MG-63 cells. J. Cell Physiol. 153, 256–265. Franceschi, R. T., and Iyer, B. S. (1992). Relationship between collagen synthesis and expression of the osteoblast phenotype in MC3T3-E1 cells. J. Bone Miner. Res. 7, 235–246. Robey, P. G. (2002). Bone matrix proteoglycans and glycoproteins. In “Principles of Bone Biology” (J. P. Bilezikian, L. G. Raisz, and G. A. Rodan, eds.), pp. 225–237. Academic Press, San Diego, CA. Gronthos, S., Stewart, K., Graves, S. E., Hay, S., and Simmons, P. J. (1997). Integrin expression and function on human osteoblast-like cells. J. Bone Miner. Res. 12, 1189–1197. Xiao, G., Gopalkrishnan, R. V., Jiang, D., Reith, E., Benson, M. D., and Franceschi, R. (2002). Bone morphogenetic proteins, extracellular matrix, and mitogen-activated protein kinase signaling pathways are required for osteoblast-specific gene expression and differentiation in MC3T3-E1 cells. J. Bone Miner. Res. 17, 101–110. Cheng, S. L., Lecanda, F., Davidson, M. K., Warlow, P. M., Zhang, S. F., Zhang, L., Suzuki, S., St. John, T., and Civitelli, R. (1998). Human osteoblasts express a repertoire of cadherins, which are critical for BMP-2-induced osteogenic differentiation. J. Bone Miner. Res. 13, 633–644. Lee, Y. S., and Chuong, C. M. (1992). Adhesion molecules in skeletogenesis: I. Transient expression of neural cell adhesion molecules (NCAM) in osteoblasts during endochondral and intramembranous ossification. J. Bone Miner. Res. 7, 1435–1446.
249 227. Okazaki, M., Takeshita, S., Kawai, S., Kikuno, R., Tsujimura, A., Kudo, A., and Amann, E. (1994). Molecular cloning and characterization of OB-cadherin, a new member of cadherin family expressed in osteoblasts. J. Biol. Chem. 269, 12092–12098. 228. Arai, F., Ohneda, O., Miyamoto, T., Zhang, X. Q., and Suda, T. (2002). Mesenchymal stem cells in perichondrium express activated leukocyte cell adhesion molecule and participate in bone marrow formation. J. Exp. Med. 195, 1549–1563. 229. Jamal, H. H., and Aubin, J. E. (1996). CD44 expression in fetal rat bone: in vivo and in vitro analysis. Exp. Cell Res. 223, 467–477. 230. Hughes, D. E., Salter, D. M., and Simpson, R. (1994). CD44 expression in human bone: a novel marker of osteocytic differentiation. J. Bone Miner. Res. 9, 39–44. 231. Kato, Y., Windle, J. J., Koop, B. A., Mundy, G. R., and Bonewald, L. F. (1997). Establishment of an osteocyte-like cell line, MLO-Y4. J. Bone Miner. Res. 12, 2014–2023. 232. Pockwinse, S. M., Lawrence, J. B., Singer, R. H., Stein, J. L., Lian, J. B., and Stein, G. S. (1993). Gene expression at single cell resolution associated with development of the bone cell phenotype: ultrastructural and in situ hybridization analysis. Bone 14, 347–352. 233. Pockwinse, S. M., Stein, J. L., Lian, J. B., and Stein, G. S. (1995). Developmental stage-specific cellular responses to vitamin D and glucocorticoids during differentiation of the osteoblast phenotype: interrelationship of morphology and gene expression by in situ hybridization. Exp. Cell Res. 216, 244–260. 234. Breen, E. C., Ignotz, R. A., McCabe, L., Stein, J. L., Stein, G. S., and Lian, J. B. (1994). TGF beta alters growth and differentiation related gene expression in proliferating osteoblasts in vitro, preventing development of the mature bone phenotype. J. Cell Physiol. 160, 323–335. 235. Tang, K.-T., Capparelli, C., Stein, J. L., Stein, G. S., Lian, J. B., Huber, A. C., Braverman, L. E., and DeVito, W. J. (1996). Acidic fibroblast growth factor inhibits osteoblast differentiation in vitro: altered expression of collagenase, cell growth-related and mineralizationassociated genes. J. Cell Biochem. 61, 152–166. 236. Debiais, F., Hott, M., Graulet, A. M., and Marie, P. J. (1998). The effects of fibroblast growth factor-2 on human neonatal calvaria osteoblastic cells are differentiation stage specific. J. Bone Miner. Res. 13, 645–654. 237. Gerstenfeld, L. C., Zurakowski, D., Schaffer, J. L., Nichols, D. P., Toma, C. D., Broess, M., Bruder, S. P., and Caplan, A. I. (1996). Variable hormone responsiveness of osteoblast populations isolated at different stages of embryogenesis and its relationship to the osteogenic lineage. Endocrinology 137, 3957–3968. 238. Owen, T. A., Aronow, M. S., Barone, L. M., Bettencourt, B., Stein, G. S., and Lian, J. B. (1991). Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: dependency upon basal levels of gene expression, duration of exposure, and bone matrix competency in normal rat osteoblast cultures. Endocrinology 128, 1496–1504. 239. Bodine, P. V. N., Henderson, R. A., Green, J., Aronow, M., Owen, T., Stein, G. S., Lian, J. B., and Komm, B. S. (1998). Estrogen receptorα is developmentally regulated during osteoblast differentiation and contributes to selective responsiveness of gene expression. Endocrinology 139, 2048–2057. 240. Ishizuya, T., Yokose, S., Hori, M., Noda, T., Suda, T., Yoshiki, S., and Yamaguchi, A. (1997). Parathyroid hormone exerts disparate effects on osteoblast differentiation depending on exposure time in rat osteoblastic cells. J. Clin. Invest. 99, 2961–2970. 241. Govoni, K. E., Baylink, D. J., and Mohan, S. (2005). The multifunctional role of insulin-like growth factor binding proteins in bone. Pediatr. Nephrol. 20, 261–268. 242. Delany, A. M., Pash, J. M., and Canalis, E. (1994). Cellular and clinical perspectives on skeletal insulin-like growth factor I. J. Cell Biochem. 55, 328–333.
250 243. Zhang, M., Xuan, S., Bouxsein, M. L., von Stechow, D., Akeno, N., Faugere, M. C., Malluche, H., Zhao, G., Rosen, C. J., Efstratiadis, A., and Clemens, T. L. (2002). Osteoblast-specific knockout of the insulin-like growth factor (IGF) receptor gene reveals an essential role of IGF signaling in bone matrix mineralization. J. Biol. Chem. 277, 44005–44012. 244. Pereira, R. M., Delany, A. M., and Canalis, E. (2001). Cortisol inhibits the differentiation and apoptosis of osteoblasts in culture. Bone 28, 484–490. 245. Van Leeuwen, J. P., van Driel, M., van den Bemd, G. J., and Pols, H. A. (2001). Vitamin D control of osteoblast function and bone extracellular matrix mineralization. Crit. Rev. Eukaryot. Gene Expr. 11, 199–226. 246. Orwoll, E. S. (2001). Androgens: basic biology and clinical implication. Calcif. Tissue Int. 69, 185–188. 247. Kousteni, S., Chen, J. R., Bellido, T., Han, L., Ali, A. A., O’Brien, C. A., Plotkin, L., Fu, Q., Mancino, A. T., Wen, Y., Vertino, A. M., Powers, C. C., Stewart, S. A., Ebert, R., Parfitt, A. M., Weinstein, R. S., Jilka, R. L., and Manolagas, S. C. (2002). Reversal of bone loss in mice by nongenotropic signaling of sex steroids. Science 298, 843–846. 248. Zhou, S., Zilberman, Y., Wassermann, K., Bain, S. D., Sadovsky, Y., and Gazit, D. (2001). Estrogen modulates estrogen receptor alpha and beta expression, osteogenic activity, and apoptosis in mesenchymal stem cells (MSCs) of osteoporotic mice. J. Cell Biochem. Suppl 36, 144–155. 249. Chen, J. R., Plotkin, L. I., Aguirre, J. I., Han, L., Jilka, R. L., Kousteni, S., Bellido, T., and Manolagas, S. C. (2004). Transient versus sustained phosphorylation and nuclear accumulation of ERKs underlie anti-versus Pro-apoptotic effects of estrogens. J. Biol. Chem. 280, 4632–4638. 250. Murphy, E., and Williams, G. R. (2004). The thyroid and the skeleton. Clin. Endocrinol.(Oxf) 61, 285–298. 251. Siebler, T., Robson, H., Shalet, S. M., and Williams, G. R. (2004). Thyroid hormone and the growth plate. Horm. Res. 62 Suppl 3, 116. 252. Kindblom, J. M., Gevers, E. F., Skrtic, S. M., Lindberg, M. K., Gothe, S., Tornell, J., Vennstrom, B., and Ohlsson, C. (2005). Increased adipogenesis in bone marrow but decreased bone mineral density in mice devoid of thyroid hormone receptors. Bone 36, 607–616. 253. Huang, J. C., Sakata, T., Pfleger, L. L., Bencsik, M., Halloran, B. P., Bikle, D. D., and Nissenson, R. A. (2004). PTH differentially regulates expression of RANKL and OPG. J. Bone Miner. Res. 19, 235–244. 254. Chen, C., Koh, A. J., Datta, N. S., Zhang, J., Keller, E. T., Xiao, G., Franceschi, R. T., D’Silva, N. J., and McCauley, L. K. (2004). Impact of the mitogen-activated protein kinase pathway on parathyroid hormone-related protein actions in osteoblasts. J. Biol. Chem. 279, 29121–29129. 255. Boskey, A. L., Wright, T. M., and Blank, R. D. (1999). Collagen and bone strength. J. Bone Miner. Res. 14, 330–335. 256. Glimcher, M. J. (1989). Mechanism of calcification: role of collagen fibrils and collagen-phosphoprotein complexes in vitro and in vivo. Anat. Rec. 224, 139–153. 257. Han, Y., Cowin, S. C., Schaffler, M. B., and Weinbaum, S. (2004). Mechanotransduction and strain amplification in osteocyte cell processes. Proc. Natl Acad. Sci. USA 101, 16689–16694. 258. Jilka, R. L., Weinstein, R. S., Bellido, T., Parfitt, A. M., and Manolagas, S. C. (1998). Osteoblast programmed cell death (apoptosis): modulation by growth factors and cytokines. J. Bone Miner. Res. 13, 793–802. 259. Karsdal, M. A., Larsen, L., Engsig, M. T., Lou, H., Ferreras, M., Lochter, A., Delaisse, J. M., and Foged, N. T. (2002). Matrix metalloproteinase-dependent activation of latent transforming growth factorbeta controls the conversion of osteoblasts into osteocytes by blocking osteoblast apoptosis. J. Biol. Chem. 277, 44061–44067. 260. Jilka, R. L., Weinstein, R. S., Bellido, T., Roberson, P., Parfitt, A. M., and Manolagas, S. C. (1999). Increased bone formation by prevention
JANE B. LIAN AND GARY S. STEIN
261.
262.
263.
264.
265.
266.
267.
268.
269.
270.
271.
272. 273. 274.
275.
of osteoblast apoptosis with parathyroid hormone. J. Clin. Invest. 104, 439–446. Bellido, T., Ali, A. A., Plotkin, L. I., Fu, Q., Gubrij, I., Roberson, P. K., Weinstein, R. S., O’Brien, C. A., Manolagas, S. C., and Jilka, R. L. (2003). Proteasomal degradation of Runx2 shortens parathyroid hormone-induced anti-apoptotic signaling in osteoblasts. A putative explanation for why intermittent administration is needed for bone anabolism. J. Biol. Chem. 278, 50259–50272. Nijweide, P. J., Burger, E. H., and Klein-Nulend, J. (2002). The osteocyte. In “Principles of Bone Biology” (J. P. Bilezikian, L. G. Raisz, and G. A. Rodan, eds.), pp. 93–107. Academic Press, San Diego, CA. Cheng, B., Kato, Y., Zhao, S., Luo, J., Sprague, E., Bonewald, L. F., and Jiang, J. X. (2001). PGE(2) is essential for gap junction-mediated intercellular communication between osteocyte-like MLO-Y4 cells in response to mechanical strain. Endocrinology 142, 3464–3473. Feng, J. Q., Huang, H., Lu, Y., Ye, L., Xie, Y., Tsutsui, T. W., Kunieda, T., Castranio, T., Scott, G., Bonewald, L. B., and Mishina, Y. (2003). The Dentin matrix protein 1 (Dmp1) is specifically expressed in mineralized, but not soft, tissues during development. J. Dent. Res. 82, 776–780. Gluhak-Heinrich, J., Ye, L., Bonewald, L. F., Feng, J. Q., MacDougall, M., Harris, S. E., and Pavlin, D. (2003). Mechanical loading stimulates dentin matrix protein 1 (DMP1) expression in osteocytes in vivo. J. Bone Miner. Res. 18, 807–817. Ye, L., MacDougall, M., Zhang, S., Xie, Y., Zhang, J., Li, Z., Lu, Y., Mishina, Y., and Feng, J. Q. (2004). Deletion of dentin matrix protein-1 leads to a partial failure of maturation of predentin into dentin, hypomineralization, and expanded cavities of pulp and root canal during postnatal tooth development. J. Biol. Chem. 279, 19141–19148. Yang, W., Lu, Y., Kalajzic, I., Guo, D., Harris, M. A., GluhakHeinrich, J., Kotha, S., Bonewald, L. F., Feng, J. Q., Rowe, D. W., Turner, C. H., Robling, A. G., and Harris, S. E. (2005). Dentin matrix protein 1 gene cis-regulation: Use in osteocytes to characterize local responses to mechanical loading in vitro and In vivo. J. Biol. Chem. 280, 20680–20690. Ahuja, S. S., Zhao, S., Bellido, T., Plotkin, L. I., Jimenez, F., and Bonewald, L. F. (2003). CD40 ligand blocks apoptosis induced by tumor necrosis factor alpha, glucocorticoids, and etoposide in osteoblasts and the osteocyte-like cell line murine long bone osteocyte-Y4. Endocrinology 144, 1761–1769. Tami, A. E., Nasser, P., Verborgt, O., Schaffler, M. B., and Knothe Tate, M. L. (2002). The role of interstitial fluid flow in the remodeling response to fatigue loading. J. Bone Miner. Res. 17, 2030–2037. Noble, B. S., Peet, N., Stevens, H. Y., Brabbs, A., Mosley, J. R., Reilly, G. C., Reeve, J., Skerry, T. M., and Lanyon, L. E. (2003). Mechanical loading: biphasic osteocyte survival and targeting of osteoclasts for bone destruction in rat cortical bone. Am. J. Physiol. Cell Physiol. 284, C934-C943. Colopy, S. A., Benz-Dean, J., Barrett, J. G., Sample, S. J., Lu, Y., Danova, N. A., Kalscheur, V. L., Vanderby, R., Jr., Markel, M. D., and Muir, P. (2004). Response of the osteocyte syncytium adjacent to and distant from linear microcracks during adaptation to cyclic fatigue loading. Bone 35, 881–891. Noble, B. (2003). Bone microdamage and cell apoptosis. Eur. Cell Mater. 6, 46–55. Hernandez, C. J., Majeska, R. J., and Schaffler, M. B. (2004). Osteocyte density in woven bone. Bone 35, 1095–1099. Knothe Tate, M. L., Steck, R., Forwood, M. R., and Niederer, P. (2000). In vivo demonstration of load-induced fluid flow in the rat tibia and its potential implications for processes associated with functional adaptation. J. Exp. Biol. 203 Pt 18, 2737–2745. Weinbaum, S., Cowin, S. C., and Zeng, Y. (1994). A model for the excitation of osteocytes by mechanical loading-induced bone fluid shear stresses. J. Biomech. 27, 339–360.
Chapter 14 The Cells of Bone 276. Cullinane, D. M., and Einhorn, T. A. (2002). Biomechanics of bone. In “Principles of Bone Biology” (J. P. Bilezikian, L. G. Raisz, and G. A. Rodan, eds.), pp. 17–32. Academic Press, San Diego, CA. 277. Noonan, K. J., Stevens, J. W., Tammi, R., Tammi, M., Hernandez, J. A., and Midura, R. J. (1996). Spatial distribution of CD44 and hyaluronan in the proximal tibia of the growing rat. J. Orthop. Res. 14, 573–581. 278. Knothe Tate, M. L. (2003). “Whither flows the fluid in bone?” An osteocyte’s perspective. J. Biomech. 36, 1409–1424. 279. Robling, A. G., and Turner, C. H. (2002). Mechanotransduction in bone: genetic effects on mechanosensitivity in mice. Bone 31, 562–569. 280. Skerry, T. M., and Suva, L. J. (2003). Investigation of the regulation of bone mass by mechanical loading: from quantitative cytochemistry to gene array. Cell Biochem. Funct. 21, 223–229. 281. Cherian, P. P., Cheng, B., Gu, S., Sprague, E., Bonewald, L. F., and Jiang, J. X. (2003). Effects of mechanical strain on the function of Gap junctions in osteocytes are mediated through the prostaglandin EP2 receptor. J. Biol. Chem. 278, 43146–43156. 282. Furlan, F., Lecanda, F., Screen, J., and Civitelli, R. (2001). Proliferation, differentiation and apoptosis in connexin43-null osteoblasts. Cell Commun. Adhes. 8, 367–371. 283. Plotkin, L. I., Manolagas, S. C., and Bellido, T. (2002). Transduction of cell survival signals by connexin-43 hemichannels. J. Biol. Chem. 277, 8648–8657. 284. Bakker, A., Klein-Nulend, J., and Burger, E. (2004). Shear stress inhibits while disuse promotes osteocyte apoptosis. Biochem. Biophys. Res. Commun. 320, 1163–1168. 285. Rubin, C. T., and Lanyon, L. E. (1987). Osteoregulatory nature of mechanical stimuli: function as a determinant for adaptive remodeling in bone. J. Orthop. Res. 5, 300–310. 286. Woodiel, F. N., Fall, P. M., and Raisz, L. G. (1996). Anabolic effects of prostaglandins in cultured fetal rat calvariae: structure-activity relations and signal transduction pathway. J. Bone Miner. Res. 11, 1249–1255. 287. Lohmann, C. H., Schwartz, Z., Liu, Y., Li, Z., Simon, B. J., Sylvia, V. L., Dean, D. D., Bonewald, L. F., Donahue, H. J., and Boyan, B. D. (2003). Pulsed electromagnetic fields affect phenotype and connexin 43 protein expression in MLO-Y4 osteocyte-like cells and ROS 17/2.8 osteoblast-like cells. J. Orthop. Res. 21, 326–334. 288. Hitt, A. L., and Luna, E. J. (1994). Membrane interactions with the actin cytoskeleton. Curr. Opin. Cell Biol. 6, 120–130. 289. Wang, N., Butler, J. P., and Ingber, D. E. (1993). Mechanotransduction across the cell surface and through the cytoskeleton. Science 260, 1124–1127. 290. Boutahar, N., Guignandon, A., Vico, L., and Lafage-Proust, M. H. (2004). Mechanical strain on osteoblasts activates autophosphorylation of focal adhesion kinase and proline-rich tyrosine kinase 2 tyrosine sites involved in ERK activation. J. Biol. Chem. 279, 30588–30599. 291. Tanaka, S., Sakai, A., Tanaka, M., Otomo, H., Okimoto, N., Sakata, T., and Nakamura, T. (2004). Skeletal unloading alleviates the anabolic action of intermittent PTH[1–34] in mouse tibia in association with inhibition of PTH-induced increase in c-fos mRNA in bone marrow cells. J. Bone Miner. Res. 19, 1813–1820. 292. Wang, F. S., Wang, C. J., Chen, Y. J., Chang, P. R., Huang, Y. T., Sun, Y. C., Huang, H. C., Yang, Y. J., and Yang, K. D. (2004). Ras induction of superoxide activates ERK-dependent angiogenic transcription factor HIF-1alpha and VEGF-A expression in shock wavestimulated osteoblasts. J. Biol. Chem. 279, 10331–10337. 293. Judex, S., Zhong, N., Squire, M. E., Ye, K., Donahue, L. R., Hadjiargyrou, M., and Rubin, C. T. (2005). Mechanical modulation of molecular signals which regulate anabolic and catabolic activity in bone tissue. J. Cell. Biochem. 94, 982–994. 294. Gross, T. S., Akeno, N., Clemens, T. L., Komarova, S., Srinivasan, S., Weimer, D. A., and Mayorov, S. (2001). Selected contribution: Osteocytes upregulate HIF-1alpha in response to acute disuse and oxygen deprivation. J. Appl. Physiol. 90, 2514–2519.
251 295. Warden, S. J., and Turner, C. H. (2004). Mechanotransduction in the cortical bone is most efficient at loading frequencies of 5–10 Hz. Bone 34, 261–270. 296. el Haj, A. J., Minter, S. L., Rawlinson, S. C., Suswillo, R., and Lanyon, L. E. (1990). Cellular responses to mechanical loading in vitro. J. Bone Miner. Res. 5, 923–932. 297. Skerry, T. M., Bitensky, L., Chayen, J., and Lanyon, L. E. (1989). Early strain-related changes in enzyme activity in osteocytes following bone loading in vivo. J. Bone Miner. Res. 4, 783–788. 298. Lean, J. M., Jagger, C. J., Chambers, T. J., and Chow, J. W. (1995). Increased insulin-like growth factor I mRNA expression in rat osteocytes in response to mechanical stimulation. Am. J. Physiol. 268, E318–E327. 299. Klein-Nulend, J., van der Plas, A., Semeins, C. M., Ajubi, N. E., Frangos, J. A., Nijweide, P. J., and Burger, E. H. (1995). Sensitivity of osteocytes to biomechanical stress in vitro. FASEB J. 9, 441–445. 300. Raisz, L. G. (1999). Prostaglandins and bone: physiology and pathophysiology. Osteoarthritis Cartilage 7, 419–421. 301. Burger, E. H., and Klein-Nulen, J. (1999). Responses of bone cells to biomechanical forces in vitro. Adv. Dent. Res. 13, 93–98. 302. Zaman, G., Pitsillides, A. A., Rawlinson, S. C., Suswillo, R. F., Mosley, J. R., Cheng, M. Z., Platts, L. A., Hukkanen, M., Polak, J. M., and Lanyon, L. E. (1999). Mechanical strain stimulates nitric oxide production by rapid activation of endothelial nitric oxide synthase in osteocytes. J. Bone Miner. Res. 14, 1123–1131. 303. Nomura, S., and Takano-Yamamoto, T. (2000). Molecular events caused by mechanical stress in bone. Matrix Biol. 19, 91–96. 304. Afzal, F., Polak, J., and Buttery, L. (2004). Endothelial nitric oxide synthase in the control of osteoblastic mineralizing activity and bone integrity. J. Pathol. 202, 503–510. 305. Aguirre, J., Buttery, L., O’Shaughnessy, M., Afzal, F., Fernandez, d. M., I, Hukkanen, M., Huang, P., MacIntyre, I., and Polak, J. (2001). Endothelial nitric oxide synthase gene-deficient mice demonstrate marked retardation in postnatal bone formation, reduced bone volume, and defects in osteoblast maturation and activity. Am. J. Pathol. 158, 247–257. 306. van’t Hof, R. J., Armour, K. J., Smith, L. M., Armour, K. E., Wei, X. Q., Liew, F. Y., and Ralston, S. H. (2000). Requirement of the inducible nitric oxide synthase pathway for IL-1-induced osteoclastic bone resorption. Proc. Natl Acad. Sci. USA 97, 7993–7998. 307. van’t Hof, R. J., Macphee, J., Libouban, H., Helfrich, M. H., and Ralston, S. H. (2004). Regulation of bone mass and bone turnover by neuronal nitric oxide synthase. Endocrinology 145, 5068–5074. 308. Gu, Y., Genever, P. G., Skerry, T. M., and Publicover, S. J. (2002). The NMDA type glutamate receptors expressed by primary rat osteoblasts have the same electrophysiological characteristics as neuronal receptors. Calcif. Tissue Int. 70, 194–203. 309. Genever, P. G., and Skerry, T. M. (2001). Regulation of spontaneous glutamate release activity in osteoblastic cells and its role in differentiation and survival: evidence for intrinsic glutamatergic signaling in bone. FASEB J. 15, 1586–1588. 310. Lian, J. B., Stein, G. S., and Aubin, J. E. (2003). Bone formation: maturation and functional activities of osteoblast lineage cells. In “Primer on the Metabolic Bone Dieseases and Disorders of Mineral Metabolism” (M. J. Favus, ed.), pp. 13–28. American Society for Bone and Mineral Research, Washington, D.C. 311. Jabs, E. W. (2001). A TWIST in the fate of human osteoblasts identifies signaling molecules involved in skull development. J. Clin. Invest. 107, 1075–1077. 312. Chen, Z. F., and Behringer, R. R. (1995). Twist is required in head mesenchyme for cranial neural tube morphogenesis. Genes Dev. 9, 686–699. 313. Cserjesi, P., Brown, D., Ligon, K. L., Lyons, G. E., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., and Olson, E. N. (1995). Scleraxis: a basic
252
314.
315.
316.
317.
318.
319.
320.
321.
322.
323.
324.
325.
326.
327.
328.
JANE B. LIAN AND GARY S. STEIN helix-loop-helix protein that prefigures skeletal formation during mouse embryogenesis. Development 121, 1099–1110. Lee, M. S., Lowe, G. N., Strong, D. D., Wergedal, J. E., and Glackin, C. A. (1999). TWIST, a basic helix-loop-helix transcription factor, can regulate the human osteogenic lineage. J. Cell Biochem. 75, 566–577. Bialek, P., Kern, B., Yang, X., Schrock, M., Sosic, D., Hong, N., Wu, H., Yu, K., Ornitz, D. M., Olson, E. N., Justice, M. J., and Karsenty, G. (2004). A twist code determines the onset of osteoblast differentiation. Dev. Cell 6, 423–435. McCabe, L. R., Banerjee, C., Kundu, R., Harrison, R. J., Dobner, P. R., Stein, J. L., Lian, J. B., and Stein, G. S. (1996). Developmental expression and activities of specific fos and jun proteins are functionally related to osteoblast maturation: role of fra-2 and jun D during differentiation. Endocrinology 137, 4398–4408. Agarwal, S. K., Novotny, E. A., Crabtree, J. S., Weitzman, J. B., Yaniv, M., Burns, A. L., Chandrasekharappa, S. C., Collins, F. S., Spiegel, A. M., and Marx, S. J. (2003). Transcription factor JunD, deprived of menin, switches from growth suppressor to growth promoter. Proc. Natl Acad. Sci. USA 100, 10770–10775. Grigoriadis, A. E., Schellander, K., Wang, Z.-Q., and Wagner, E. F. (1993). Osteoblasts are target cells for transformation in c-fos transgenic mice. J. Cell Biol. 122, 685–701. Machwate, M., Jullienne, A., Moukhtar, M., and Marie, P. J. (1995). Temporal variation of c-fos proto-oncogene expression during osteoblast differentiation and osteogenesis in developing bone. J. Cell Biochem. 57, 62–70. McCabe, L. R., Kockx, M., Lian, J., Stein, J., and Stein, G. (1995). Selective expression of fos- and jun-related genes during osteoblast proliferation and differentiation. Exp. Cell Res. 218, 255–262. Lufkin, T., Mark, M., Hart, C. P., Dolle, P., LeMeur, M., and Chambon, P. (1992). Homeotic transformation of the occipital bones of the skull by ectopic expression of a homeobox gene. Nature 359, 835–841. Jabs, E. W., Muller, U., Li, X., Ma, L., Luo, W., Haworth, I. S., Klisak, I., Sparkes, R., Warman, M. L., Mulliken, J. B., Snead, M. L., and Maxson, R. (1993). A mutation in the homeodomain of the human MSX2 gene in a family affected with autosomal dominant craniosynostosis. Cell 75, 443–450. Mina, M., Gluhak, J., Upholt, W. B., Kollar, E. J., and Rogers, B. (1995). Experimental analysis of Msx-1 and Msx-2 gene expression during chick mandibular morphogenesis. Dev. Dynamics 202, 195–214. Zhao, G. Q., Zhao, S., Zhou, X., Eberspaecher, H., Solursh, M., and de Crombrugghe, B. (1994). rDlx, a novel distal-less-like homeoprotein is expressed in developing cartilages and discrete neuronal tissues. Dev. Biol. 164, 37–51. Orestes-Cardoso, S., Nefussi, J. R., Lezot, F., Oboeuf, M., Pereira, M., Mesbah, M., Robert, B., and Berdal, A. (2002). Msx1 is a regulator of bone formation during development and postnatal growth: in vivo investigations in a transgenic mouse model. Connect. Tissue Res. 43, 153–160. Hassan, M. Q., Javed, A., Morasso, M. I., Karlin, J., Montecino, M., van Wijnen, A. J., Stein, G. S., Stein, J. L., and Lian, J. B. (2004). Dlx3 transcriptional regulation of osteoblast differentiation: temporal recruitment of Msx2, Dlx3, and Dlx5 homeodomain proteins to chromatin of the osteocalcin gene. Mol. Cell Biol. 24, 9248–9261. Dodig, M., Kronenberg, M. S., Bedalov, A., Kream, B. E., Gronowicz, G., Clark, S. H., Mack, K., Liu, Y. H., Maxon, R., Pan, Z. Z., Upholt, W. B., Rowe, D. W., and Lichtler, A. C. (1996). Identification of a TAAT-containing motif required for high level expression of the COL1A1 promoter in differentiated osteoblasts of transgenic mice. J. Biol. Chem. 271, 16422–16429. Hoffmann, H. M., Catron, K. M., van Wijnen, A. J., McCabe, L. R., Lian, J. B., Stein, G. S., and Stein, J. L. (1994). Transcriptional control of the tissue-specific, developmentally regulated osteocalcin gene
329.
330.
331.
332.
333.
334.
335.
336.
337.
338.
339.
340.
341.
342.
343.
requires a binding motif for the Msx family of homeodomain proteins. Proc. Natl Acad. Sci. USA 91, 12887–12891. Ryoo, H.-M., Hoffmann, H. M., Beumer, T. L., Frenkel, B., Towler, D. A., Stein, G. S., Stein, J. L., van Wijnen, A. J., and Lian, J. B. (1997). Stage-specific expression of Dlx-5 during osteoblast differentiation: involvement in regulation of osteocalcin gene expression. Mol. Endocrinol. 11, 1681–1694. Towler, D. A., Rutledge, S. J., and Rodan, G. A. (1994). Msx-2/Hox 8.1: a transcriptional regulator of the rat osteocalcin promoter. Mol. Endocrinol. 8, 1484–1493. Hoffmann, H. M., Beumer, T. L., Rahman, S., McCabe, L. R., Banerjee, C., Aslam, F., Tiro, J. A., van Wijnen, A. J., Stein, J. L., Stein, G. S., and Lian, J. B. (1996). Bone tissue-specific transcription of the osteocalcin gene: role of an activator osteoblast-specific complex and suppressor hox proteins that bind the OC box. J. Cell Biochem. 61, 310–324. Bidder, M., Latifi, T., and Towler, D. A. (1998). Reciprocal temporospatial patterns of Msx2 and Osteocalcin gene expression during murine odontogenesis. J. Bone Miner. Res. 13, 609–619. Nimer, S. D., and Moore, M. A. (2004). Effects of the leukemiaassociated AML1-ETO protein on hematopoietic stem and progenitor cells. Oncogene 23, 4249–4254. Lian, J. B., Balint, E., Javed, A., Drissi, H., Vitti, R., Quinlan, E. J., Zhang, L., van Wijnen, A. J., Stein, J. L., Speck, N., and Stein, G. S. (2003). Runx1/AML1 hematopoietic transcription factor contributes to skeletal development in vivo. J. Cell Physiol. 196, 301–311. Levanon, D., Bettoun, D., Harris-Cerruti, C., Woolf, E., Negreanu, V., Eilam, R., Bernstein, Y., Goldenberg, D., Xiao, C., Fliegauf, M., Kremer, E., Otto, F., Brenner, O., Lev-Tov, A., and Groner, Y. (2002). The Runx3 transcription factor regulates development and survival of TrkC dorsal root ganglia neurons. EMBO J. 21, 3454–3463. Yoshida, C. A., Yamamoto, H., Fujita, T., Furuichi, T., Ito, K., Inoue, K., Yamana, K., Zanma, A., Takada, K., Ito, Y., and Komori, T. (2004). Runx2 and Runx3 are essential for chondrocyte maturation, and Runx2 regulates limb growth through induction of Indian hedgehog. Genes Dev. 18, 952–963. Mundlos, S., Otto, F., Mundlos, C., Mulliken, J. B., Aylsworth, A. S., Albright, S., Lindhout, D., Cole, W. G., Henn, W., Knoll, J. H. M., Owen, M. J., Mertelsmann, R., Zabel, B. U., and Olsen, B. R. (1997). Mutations involving the transcription factor CBFA1 cause cleidocranial dysplasia. Cell 89, 773–779. Levanon, D., Brenner, O., Negreanu, V., Bettoun, D., Woolf, E., Eilam, R., Lotem, J., Gat, U., Otto, F., Speck, N., and Groner, Y. (2001). Spatial and temporal expression pattern of Runx3 (Aml2) and Runx1 (Aml1) indicates non-redundant functions during mouse embryogenesis. Mech. Dev. 109, 413–417. Yamashiro, T., Aberg, T., Levanon, D., Groner, Y., and Thesleff, I. (2002). Expression of Runx1, -2 and -3 during tooth, palate and craniofacial bone development. Mech. Dev. 119, S107-S110. Smith, N., Dong, Y., Pratap, J., Lian, J. B., Kingsley, P., van Wijnen, A. J., Stein, J. L., Schwarz, E. M., O’Keefe, R. J., Stein, G. S., and Drissi, M. H. (2005). Overlapping expression of Runx1(Cbfa2) and Runx2(Cbfa1) transcription factors supports cooperative induction of skeletal development. J. Cell Pysiol. 203, 133–143. Yamashiro, T., Wang, X. P., Li, Z., Oya, S., Aberg, T., Fukunaga, T., Kamioka, H., Speck, N. A., Takano-Yamamoto, T., and Thesleff, I. (2004). Possible roles of Runx1 and Sox9 in incipient intramembranous ossification. J. Bone Miner. Res. 19, 1671–1677. Stewart, M., Terry, A., Hu, M., O’Hara, M., Blyth, K., Baxter, E., Cameron, E., Onions, D. E., and Neil, J. C. (1997). Proviral insertions induce the expression of bone-specific isoforms of PEBP2alphaA (CBFA1): evidence for a new myc collaborating oncogene. Proc. Natl Acad. Sci. USA. 94, 8646–8651. Ogawa, E., Maruyama, M., Kagoshima, H., Inuzuka, M., Lu, J., Satake, M., Shigesada, K., and Ito, Y. (1993). PEBP2/PEA2 represents
253
Chapter 14 The Cells of Bone
344.
345.
346.
347.
348.
349.
350.
351.
352.
353.
354.
355.
356.
357.
358.
a family of transcription factors homologous to the products of the Drosophila runt gene and the human AML1 gene. Proc. Natl Acad. Sci. USA 90, 6859–6863. Xiao, Z. S., Liu, S. G., Hinson, T. K., and Quarles, L. D. (2001). Characterization of the upstream mouse Cbfa1/Runx2 promoter. J. Cell Biochem. 82, 647–659. Yoshida, C. A., Furuichi, T., Fujita, T., Fukuyama, R., Kanatani, N., Kobayashi, S., Satake, M., Takada, K., and Komori, T. (2002). Corebinding factor beta interacts with Runx2 and is required for skeletal development. Nat. Genet. 32, 633–638. Miller, J., Horner, A., Stacy, T., Lowrey, C., Lian, J. B., Stein, G., Nuckolls, G. H., and Speck, N. A. (2002). The core-binding factor beta subunit is required for bone formation and hematopoietic maturation. Nat. Genet. 32, 645–649. Schroeder, T. M., Kahler, R. A., Li, X., and Westendorf, J. J. (2004). Histone deacetylase 3 interacts with runx2 to repress the osteocalcin promoter and regulate osteoblast differentiation. J. Biol. Chem. 279, 41998–42007. Westendorf, J. J., Zaidi, S. K., Cascino, J. E., Kahler, R., van Wijnen, A. J., Lian, J. B., Yoshida, M., Stein, G. S., and Li, X. (2002). Runx2 (Cbfa1, AML-3) interacts with histone deacetylase 6 and represses the p21(CIP1/WAF1) promoter. Mol. Cell Biol. 22, 7982–7992. Gutierrez, S., Liu, J., Javed, A., Montecino, M., Stein, G. S., Lian, J. B., and Stein, J. L. (2004). The vitamin D response element in the distal osteocalcin promoter contributes to chromatin organization of the proximal regulatory domain. J. Biol. Chem. 279, 43581–43588. Paredes, R., Arriagada, G., Cruzat, F., Olate, J., van Wijnen, A., Lian, J., Stein, G., Stein, J., and Montecino, M. (2004). The Runx2 transcription factor plays a key role in the 1α,25-dihydroxy vitamin D3-dependent upregulation of the rat osteocalcin (OC) gene expression in osteoblastic cells. J. Steroid Biochem. Mol. Biol. 89–90, 269–271. Paredes, R., Arriagada, G., Cruzat, F., Villagra, A., Olate, J., Zaidi, K., van Wijnen, A. J., Lian, J. B., Stein, G. S., Stein, J. L., and Montecino, M. (2004). The bone-specific transcription factor RUNX2 interacts with the 1α,25-dihydroxyvitamin D3 receptor to up-regulate rat osteocalcin gene expression in osteoblastic cells. Mol. Cell Biol. 24, 8847–8861. Vega, R. B., Matsuda, K., Oh, J., Barbosa, A. C., Yang, X., Meadows, E., McAnally, J., Pomajzl, C., Shelton, J. M., Richardson, J. A., Karsenty, G., and Olson, E. N. (2004). Histone deacetylase 4 controls chondrocyte hypertrophy during skeletogenesis. Cell 119, 555–566. Zaidi, S. K., Sullivan, A. J., Medina, R., Ito, Y., van Wijnen, A. J., Stein, J. L., Lian, J. B., and Stein, G. S. (2004). Tyrosine phosphorylation controls Runx2-mediated subnuclear targeting of YAP to repress transcription. EMBO J. 23, 790–799. Shirakabe, K., Terasawa, K., Miyama, K., Shibuya, H., and Nishida, E. (2001). Regulation of the activity of the transcription factor Runx2 by two homeobox proteins, Msx2 and Dlx5. Genes Cells 6, 851–856. Cui, C. B., Cooper, L. F., Yang, X., Karsenty, G., and Aukhil, I. (2003). Transcriptional coactivation of bone-specific transcription factor Cbfa1 by TAZ. Mol. Cell Biol. 23, 1004–1013. Javed, A., Guo, B., Hiebert, S., Choi, J.-Y., Green, J., Zhao, S.-C., Osborne, M. A., Stifani, S., Stein, J. L., Lian, J. B., van Wijnen, A. J., and Stein, G. S. (2000). Groucho/TLE/R-Esp proteins associate with the nuclear matrix and repress RUNX (CBFα/AML/PEBP2α) dependent activation of tissue-specific gene transcription. J. Cell Sci. 113, 2221–2231. Wang, W., Wang, Y. G., Reginato, A. M., Glotzer, D. J., Fukai, N., Plotkina, S., Karsenty, G., and Olsen, B. R. (2004). Groucho homologue Grg5 interacts with the transcription factor Runx2-Cbfa1 and modulates its activity during postnatal growth in mice. Dev. Biol. 270, 364–381. Choi, J.-Y., Pratap, J., Javed, A., Zaidi, S. K., Xing, L., Balint, E., Dalamangas, S., Boyce, B., van Wijnen, A. J., Lian, J. B., Stein, J. L., Jones, S. N., and Stein, G. S. (2001). Subnuclear targeting of
359.
360.
361.
362.
363.
364.
365.
366.
367. 368.
369.
370.
371.
372.
373. 374. 375.
376.
Runx/Cbfa/AML factors is essential for tissue-specific differentiation during embryonic development. Proc. Natl Acad. Sci. USA 98, 8650–8655. Barnes, G. L., Hebert, K. E., Kamal, M., Javed, A., Einhorn, T. A., Lian, J. B., Stein, G. S., and Gerstenfeld, L. C. (2004). Fidelity of Runx2 activity in breast cancer cells is required for the generation of metastases associated osteolytic disease. Cancer Res. 64, 4506–4513. Javed, A., Barnes, G. L., Pratap, J., Antkowiak, T., Gerstenfeld, L. C., van Wijnen, A. J., Stein, J. L., Lian, J. B., and Stein, G. S. (2005). Impaired intranuclear trafficking of Runx2 (AML3/CBFA1) transcription factors in breast cancer cells inhibits osteolysis in vivo. Proc. Natl Acad. Sci. USA 102, 1454–1459. Gutierrez, S., Javed, A., Tennant, D., van Rees, M., Montecino, M., Stein, G. S., Stein, J. L., and Lian, J. B. (2002). CCAAT/enhancerbinding proteins (C/EBP) β and δ Activate osteocalcin gene transcription and synergize with Runx2 at the C/EBP element to regulate bonespecific expression. J. Biol. Chem. 277, 1316–1323. Pereira, R. C., Delany, A. M., and Canalis, E. (2004). CCAAT/enhancer binding protein homologous protein (DDIT3) induces osteoblastic cell differentiation. Endocrinology 145, 1952–1960. Harrison, J. R., Huang, Y. F., Wilson, K. A., Kelly, P. L., Adams, D. J., Gronowicz, G. A., and Clark, S. H. (2005). Col1a1 promoter-targeted expression of p20C/EBPbeta , a truncated C/EBPbeta isoform, causes osteopenia in transgenic mice. J. Biol. Chem. 280, 8117–8124. Sevetson, B., Taylor, S., and Pan, Y. (2004). Cbfa1/RUNX2 directs specific expression of the sclerosteosis gene (SOST). J. Biol. Chem. 279, 13849–13858. Hess, J., Porte, D., Munz, C., and Angel, P. (2001). AP-1 and Cbfa/Runt physically interact and regulate PTH-dependent MMP13 expression in osteoblasts through a new OSE2/AP-1 composite element. J. Biol. Chem. 276, 20029–20038. McCarthy, T. L., Chang, W. Z., Liu, Y., and Centrella, M. (2003). Runx2 integrates estrogen activity in osteoblasts. J. Biol. Chem. 278, 43121–43129. Roodman, G. D. (1995). Osteoclast function in Paget’s disease and multiple myeloma. Bone 17, 57S-61S. Boissy, P., Saltel, F., Bouniol, C., Jurdic, P., and Machuca-Gayet, I. (2002). Transcriptional activity of nuclei in multinucleated osteoclasts and its modulation by calcitonin. Endocrinology 143, 1913–1921. Saltman, L. H., Javed, A., Ribadeneyra, J., Hussain, S., Young, D. W., Osdoby, P., Amcheslavsky, A., van Wijnen, A. J., Stein, J. L., Stein, G. S., Lian, J. B., and Bar-Shavit, Z. (2005). Organization of transcriptional regulatory machinery in osteoclast nuclei: Compartmentalization of Runx1. J. Cell Physiol. 204, 871–880. Lakkakorpi, P. T., and Vaananen, H. K. (1991). Kinetics of the osteoclast cytoskeleton during the resorption cycle in vitro. J. Bone Miner. Res. 6, 817–826. Vaananen, H. K., and Horton, M. (1995). The osteoclast clear zone is a specialized cell-extracellular matrix adhesion structure. J. Cell Sci. 108, 2729–2732. Tanaka, S., Takahashi, N., Udagawa, N., Murakami, H., Nakamura, I., Kurokawa, T., and Suda, T. (1995). Possible involvement of focal adhesion kinase, p125FAK, in osteoclastic bone resorption. J. Cell Biochem. 58, 424–435. Teitelbaum, S. L., and Ross, F. P. (2003). Genetic regulation of osteoclast development and function. Nat. Rev. Genet. 4, 638–649. Boyle, W. J., Simonet, W. S., and Lacey, D. L. (2003). Osteoclast differentiation and activation. Nature 423, 337–342. Engleman, V. W., Nickols, G. A., Ross, F. P., Horton, M. A., Griggs, D. W., Settle, S. L., Ruminski, P. G., and Teitelbaum, S. L. (1997). A peptidomimetic antagonist of the alpha(v)beta3 integrin inhibits bone resorption in vitro and prevents osteoporosis in vivo. J. Clin. Invest. 99, 2284–2292. Duong, L. T., Nakamura, I., Lakkakorpi, P. T., Lipfert, L., Bett, A. J., and Rodan, G. A. (2000). Inhibition of osteoclast function by adenovirus
254
377.
378.
379.
380.
381.
382.
383.
384.
385.
386.
387.
388.
389.
390. 391.
392.
393.
JANE B. LIAN AND GARY S. STEIN expressing antisense protein-tyrosine kinase 2. J. Biol. Chem. 276, 7484–7492. Chellaiah, M. A., Kizer, N., Biswas, R., Alvarez, U., Strauss-Schoenberger, J., Rifas, L., Rittling, S. R., Denhardt, D. T., and Hruska, K. A. (2003). Osteopontin deficiency produces osteoclast dysfunction due to reduced CD44 surface expression. Mol. Biol. Cell 14, 173–189. Faccio, R., Novack, D. V., Zallone, A., Ross, F. P., and Teitelbaum, S. L. (2003). Dynamic changes in the osteoclast cytoskeleton in response to growth factors and cell attachment are controlled by beta3 integrin. J. Cell Biol. 162, 499–509. Arias-Salgado, E. G., Lizano, S., Sarkar, S., Brugge, J. S., Ginsberg, M. H., and Shattil, S. J. (2003). Src kinase activation by direct interaction with the integrin beta cytoplasmic domain. Proc. Natl Acad. Sci. USA 100, 13298–13302. Miyazaki, T., Sanjay, A., Neff, L., Tanaka, S., Horne, W. C., and Baron, R. (2004). Src kinase activity is essential for osteoclast function. J. Biol. Chem. 279, 17660–17666. Soriano, P., Montgomery, C., Geske, R., and Bradley, A. (1991). Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64, 693–702. Boyce, B. F., Yoneda, T., Lowe, C., Soriano, P., and Mundy, G. R. (1992). Requirement of pp60c-src expression for osteoclasts to form ruffled borders and resorb bone in mice. J. Clin. Invest. 90, 1622–1627. Chiusaroli, R., Sanjay, A., Henriksen, K., Engsig, M. T., Horne, W. C., Gu, H., and Baron, R. (2003). Deletion of the gene encoding c-Cbl alters the ability of osteoclasts to migrate, delaying resorption and ossification of cartilage during the development of long bones. Dev. Biol. 261, 537–547. Lakkakorpi, P. T., Bett, A. J., Lipfert, L., Rodan, G. A., and Duong, I. T. (2003). PYK2 autophosphorylation, but not kinase activity, is necessary for adhesion-induced association with c-Src, osteoclast spreading, and bone resorption. J. Biol. Chem. 278, 11502–11512. Mulari, M. T., Zhao, H., Lakkakorpi, P. T., and Vaananen, H. K. (2003). Osteoclast ruffled border has distinct subdomains for secretion and degraded matrix uptake. Traffic 4, 113–125. Chatterjee, D., Chakraborty, M., Leit, M., Neff, L., Jamsa-Kellokumpu, S., Fuchs, R., and Baron, R. (1992). Sensitivity to vanadate and isoforms of subunits A and B distinguish the osteoclast proton pump from other vacuolar H+ ATPases. Proc. Natl Acad. Sci. USA. 89, 6257–6261. Toyosawa, S., Ogawa, Y., Chang, C. K., Hong, S. S., Yagi, T., Kuwahara, H., Wakasa, K., and Sakurai, M. (1991). Histochemistry of tartrate-resistant acid phosphatase and carbonic anhydrase isoenzyme II in osteoclast-like giant cells in bone tumours. Virchows Arch. A. Pathol. Anat. Histopathol. 418, 255–261. Li, Y. P., Chen, W., Liang, Y., Li, E., and Stashenko, P. (1999). Atp6ideficient mice exhibit severe osteopetrosis due to loss of osteoclastmediated extracellular acidification. Nat. Genet. 23, 447–451. Teitelbaum, S. L., Tondravi, M. M., and Ross, F. P. (1996). Osteoclast biology. In “Osteoporosis” (R. Marcus, D. Feldman, and J. Kelsey, eds.), pp. 61–94. Academic Press, Inc., San Diego, CA. Teitelbaum, S. L. (2000). Bone resorption by osteoclasts. Science 289, 1504–1508. Saftig, P., Hunziker, E., Wehmeyer, O., Jones, S., Boyde, A., Rommerskirch, W., Moritz, J. D., Schu, P., and von Figura, K. (1998). Impaired osteoclastic bone resorption leads to osteopetrosis in cathepsin-K-deficient mice. Proc. Natl Acad. Sci. USA 95, 13453–13458. Saftig, P., Hunziker, E., Everts, V., Jones, S., Boyde, A., Wehmeyer, O., Suter, A., and von Figura, K. (2000). Functions of cathepsin K in bone resorption. Lessons from cathepsin K deficient mice. Adv. Exp. Med. Biol. 477, 293–303. Hollberg, K., Hultenby, K., Hayman, A., Cox, T., and Andersson, G. (2002). Osteoclasts from mice deficient in tartrate-resistant acid
394.
395.
396.
397. 398. 399.
400.
401.
402.
403.
404.
405.
406.
407.
408.
409.
410.
phosphatase have altered ruffled borders and disturbed intracellular vesicular transport. Exp. Cell Res. 279, 227–238. Bergers, G., Brekken, R., McMahon, G., Vu, T. H., Itoh, T., Tamaki, K., Tanzawa, K., Thorpe, P., Itohara, S., Werb, Z., and Hanahan, D. (2000). Matrix metalloproteinase-9 triggers the angiogenic switch during carcinogenesis. Nat. Cell Biol. 2, 737–744. Engsig, M. T., Chen, Q. J., Vu, T. H., Pedersen, A. C., Therkidsen, B., Lund, L. R., Henriksen, K., Lenhard, T., Foged, N. T., Werb, Z., and Delaisse, J. M. (2000). Matrix metalloproteinase 9 and vascular endothelial growth factor are essential for osteoclast recruitment into developing long bones. J. Cell Biol. 151, 879–890. Popoff, S. N., and Marks, S. C. Jr. (1995). The heterogeneity of the osteopetroses reflects the diversity of cellular influences during skeletal development. Bone 17, 437–445. Suda, T., Nakamura, I., Jimi, E., and Takahashi, N. (1997). Regulation of osteoclast function. J. Bone Miner. Res. 12, 869–879. Roodman, G. D. (1996). Advances in bone biology: the osteoclast. Endocr. Rev. 17, 308–332. Grey, A., Chen, Y., Paliwal, I., Carlberg, K., and Insogna, K. (2000). Evidence for a functional association between phosphatidylinositol 3- kinase and c-src in the spreading response of osteoclasts to colonystimulating factor-1. Endocrinology 141, 2129–2138. Felix, R., Cecchini, M. G., and Fleisch, H. (1990). Macrophage colony-stimulating factor restores in vivo bone resorption in the op/op osteopetrotic mouse. Endocrinology 127, 2592–2594. Yoshida, H., Hayashi, S., Kunisada, T., Ogawa, M., Nishikawa, S., Okamura, H., Sudo, T., and Shultz, L. D. (1990). The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345, 442–444. Takahashi, N., Udagawa, N., Akatsu, T., Tanaka, H., Isogai, Y., and Suda, T. (1991). Deficiency of osteoclasts in osteopetrotic mice is due to a defect in the local microenvironment provided by osteoblastic cells. Endocrinology 128, 1792–1796. Luchin, A., Purdom, G., Murphy, K., Clark, M. Y., Angel, N., Cassady, A. I., Hume, D. A., and Ostrowski, M. C. (2000). The microphthalmia transcription factor regulates expression of the tartrateresistant acid phosphatase gene during terminal differentiation of osteoclasts. J. Bone Miner. Res. 15, 451–460. Mansky, K. C., Sankar, U., Han, J., and Ostrowski, M. C. (2002). Microphthalmia transcription factor is a target of the p38 MAPK pathway in response to receptor activator of NF-kappa B ligand signaling. J. Biol. Chem. 277, 11077–11083. Weilbaecher, K. N., Motyckova, G., Huber, W. E., Takemoto, C. M., Hemesath, T. J., Xu, Y., Hershey, C. L., Dowland, N. R., Wells, A. G., and Fisher, D. E. (2001). Linkage of M-CSF signaling to Mitf, TFE3, and the osteoclast defect in Mitf(mi/mi) mice. Mol. Cell 8, 749–758. Kawaguchi, N., Ono, T., Mochii, M., and Noda, M. (2001). Spontaneous mutation in Mitf gene causes osteopetrosis in silver homozygote quail. Dev. Dyn. 220, 133–140. Lakkakorpi, P. T., Nakamura, I., Nagy, R. M., Parsons, J. T., Rodan, G. A., and Duong, L. T. (1999). Stable association of PYK2 and p130(Cas) in osteoclasts and their co-localization in the sealing zone. J. Biol. Chem. 274, 4900–4907. Nakamura, I., Lipfert, L., Rodan, G. A., and Duong, L. T. (2001). Convergence of alpha(v)beta(3) integrin- and macrophage colony stimulating factor-mediated signals on phospholipase Cgamma in prefusion osteoclasts. J. Cell Biol. 152, 361–374. Nakamura, I., Rodan, G. A., and Duong, l. T. (2003). Distinct roles of p130Cas and c-Cbl in adhesion-induced or macrophage colonystimulating factor-mediated signaling pathways in prefusion osteoclasts. Endocrinology 144, 4739–4741. Horwood, N. J., Kartsogiannis, V., Quinn, J. M., Romas, E., Martin, T. J., and Gillespie, M. T. (1999). Activated T lymphocytes support osteoclast formation in vitro. Biochem. Biophys. Res. Commun. 265, 144–150.
Chapter 14 The Cells of Bone 411. Lomaga, M. A., Yeh, W. C., Sarosi, I., Duncan, G. S., Furlonger, C., Ho, A., Morony, S., Capparelli, C., Van, G., Kaufman, S., van der, H. A., Itie, A., Wakeham, A., Khoo, W., Sasaki, T., Cao, Z., Penninger, J. M., Paige, C. J., Lacey, D. L., Dunstan, C. R., Boyle, W. J., Goeddel, D. V., and Mak, T. W. (1999). TRAF6 deficiency results in osteopetrosis and defective interleukin-1, CD40, and LPS signaling. Genes Dev. 13, 1015–1024. 412. Armstrong, A. P., Tometsko, M. E., Glaccum, M., Sutherland, C. L., Cosman, D., and Dougall, W. C. (2002). A RANK/TRAF6-dependent signal transduction pathway is essential for osteoclast cytoskeletal organization and resorptive function. J. Biol. Chem. 277, 44347–44356. 413. Naito, A., Azuma, S., Tanaka, S., Miyazaki, T., Takaki, S., Takatsu, K., Nakao, K., Nakamura, K., Katsuki, M., Yamamoto, T., and Inoue, J. (1999). Severe osteopetrosis, defective interleukin-1 signalling and lymph node organogenesis in TRAF6-deficient mice. Genes Cells 4, 353–362. 414. Simonet, W. S., Lacey, D. L., Dunstan, C. R., Kelley, M., Chang, M. S., Luthy, R., Nguyen, H. Q., Wooden, S., Bennett, L., Boone, T., Shimamoto, G., DeRose, M., Elliott, R., Colombero, A., Tan, H. L., Trail, G., Sullivan, J., Davy, E., Bucay, N., Renshaw-Gegg, L., Hughes, T. M., Hill, D., Pattison, W., Campbell, P., and Boyle, W. J. (1997). Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89, 309–319. 415. Wong, B. R., Josien, R., Lee, S. Y., Sauter, B., Li, H. L., Steinman, R. M., and Choi, Y. (1997). TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor. J. Exp. Med. 186, 2075–2080. 416. Yasuda, H., Shima, N., Nakagawa, N., Mochizuki, S. I., Yano, K., Fujise, N., Sato, Y., Goto, M., Yamaguchi, K., Kuriyama, M., Kanno, T., Murakami, A., Tsuda, E., Morinaga, T., and Higashio, K. (1998). Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): a mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 139, 1329–1337. 417. Yasuda, H., Shima, N., Nakagawa, N., Yamaguchi, K., Kinosaki, M., Mochizuki, S., Tomoyasu, A., Yano, K., Goto, M., Murakami, A., Tsuda, E., Morinaga, T., Higashio, K., Udagawa, N., Takahashi, N., and Suda, T. (1998). Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc. Natl Acad. Sci. USA 95, 3597–3602. 418. Lacey, D. L., Timms, E., Tan, H. L., Kelley, M. J., Dunstan, C. R., Burgess, T., Elliott, R., Colombero, A., Elliott, G., Scully, S., Hsu, H., Sullivan, J., Hawkins, N., Davy, E., Capparelli, C., Eli, A., Qian, Y. X., Kaufman, S., Sarosi, I., Shalhoub, V., Senaldi, G., Guo, J., Delaney, J., and Boyle, W. J. (1998). Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93, 165–176. 419. Yu, X., Huang, Y., Collin-Osdoby, P., and Osdoby, P. (2003). Stromal cell-derived factor-1 (SDF-1) recruits osteoclast precursors by inducing chemotaxis, matrix metalloproteinase-9 (MMP-9) activity, and collagen transmigration. J. Bone Miner. Res. 18, 1404–1418. 420. Miyaura, C., Inada, M., Suzawa, T., Sugimoto, Y., Ushikubi, F., Ichikawa, A., Narumiya, S., and Suda, T. (2000). Impaired bone resorption to prostaglandin E2 in prostaglandin E receptor EP4-knockout mice. J. Biol. Chem. 275, 19819–19823. 421. Li, X., Okada, Y., Pilbeam, C. C., Lorenzo, J. A., Kennedy, C. R., Breyer, R. M., and Raisz, L. G. (2000). Knockout of the murine prostaglandin EP2 receptor impairs osteoclastogenesis in vitro. Endocrinology 141, 2054–2061. 422. Lee, S. E., Chung, W. J., Kwak, H. B., Chung, C. H., Kwack, K. B., Lee, Z. H., and Kim, H. H. (2001). Tumor necrosis factor-alpha supports the survival of osteoclasts through the activation of Akt and ERK. J. Biol. Chem. 276, 49343–49349. 423. Girasole, G., Passeri, G., Jilka, R. L., and Manolagas, S. C. (1994). Interleukin-11: a new cytokine critical for osteoclast development. J. Clin. Invest. 93, 1516–1524.
255 424. Romas, E., Udagawa, N., Zhou, H., Tamura, T., Saito, M., Taga, T., Hilton, D. J., Suda, T., Ng, K. W., and Martin, T. J. (1996). The role of gp130-mediated signals in osteoclast development: regulation of interleukin 11 production by osteoblasts and distribution of its receptor in bone marrow cultures. J. Exp. Med. 183, 2581–2591. 425. Elias, J. A., Tang, W., and Horowitz, M. C. (1995). Cytokine and hormonal stimulation of human osteosarcoma interleukin-11 production. Endocrinology 136, 489–498. 426. Suda, T., Udagawa, N., Nakamura, I., Miyaura, C., and Takahashi, N. (1995). Modulation of osteoclast differentiation by local factors. Bone 17, 87S-91S. 427. Devlin, R. D., Reddy, S. V., Savino, R., Ciliberto, G., and Roodman, G. D. (1998). IL-6 mediates the effects of IL-1 or TNF, but not PTHrP or 1, 25(OH)2D3, on osteoclast-like cell formation in normal human bone marrow cultures. J. Bone Miner. Res. 13, 393–399. 428. Kishimoto, T., Akira, S., Narazaki, M., and Taga, T. (1995). Interleukin-6 family of cytokines and gp130. Blood 86, 1243–1254. 429. Lee, S. K., Kalinowski, J. F., Jastrzebski, S. L., Puddington, L., and Lorenzo, J. A. (2003). Interleukin-7 is a direct inhibitor of in vitro osteoclastogenesis. Endocrinology 144, 3524–3531. 430. Udagawa, N., Horwood, N. J., Elliott, J., Mackay, A., Owens, J., Okamura, H., Kurimoto, M., Chambers, T. J., Martin, T. J., and Gillespie, M. T. (1997). Interleukin-18 (interferon-gamma-inducing factor) is produced by osteoblasts and acts via granulocyte/macrophage colony-stimulating factor and not via interferon-gamma to inhibit osteoclast formation. J. Exp. Med. 185, 1005–1012. 431. Horwood, N. J., Udagawa, N., Elliott, J., Grail, D., Okamura, H., Kurimoto, M., Dunn, A. R., Martin, T., and Gillespie, M. T. (1998). Interleukin 18 inhibits osteoclast formation via T cell production of granulocyte macrophage colony-stimulating factor. J. Clin. Invest. 101, 595–603. 432. Suda, T., Ueno, Y., Fujii, K., and Shinki, T. (2003). Vitamin D and bone. J. Cell Biochem. 88, 259–266. 433. Kurihara, N., Reddy, S. V., Araki, N., Ishizuka, S., Ozono, K., Cornish, J., Cundy, T., Singer, F. R., and Roodman, G. D. (2004). Role of TAFII-17, a VDR binding protein, in the increased osteoclast formation in Paget’s Disease. J. Bone Miner. Res. 19, 1154–1164. 434. Faucheux, C., Horton, M. A., and Price, J. S. (2002). Nuclear localization of type I parathyroid hormone/parathyroid hormone-related protein receptors in deer antler osteoclasts: evidence for parathyroid hormone-related protein and receptor activator of NF-kappaBdependent effects on osteoclast formation in regenerating mammalian bone. J. Bone Miner. Res. 17, 455–464. 435. Nakchbandi, I. A., Weir, E. E., Insogna, K. L., Philbrick, W. M., and Broadus, A. E. (2000). Parathyroid hormone-related protein induces spontaneous osteoclast formation via a paracrine cascade. Proc. Natl Acad. Sci. USA 97, 7296–7300. 436. Vilardaga, J. P., Lin, I., and Nissenson, R. A. (2001). Analysis of parathyroid hormone (PTH)/secretin receptor chimeras differentiates the role of functional domains in the pth/ pth-related peptide (PTHrP) receptor on hormone binding and receptor activation. Mol. Endocrinol. 15, 1186–1199. 437. Huang, Z., Chen, Y., Pratt, S., Chen, T. H., Bambino, T., Nissenson, R. A., and Shoback, D. M. (1996). The N-terminal region of the third intracellular loop of the parathyroid hormone (PTH)/PTH-related peptide receptor is critical for coupling to cAMP and inositol phosphate/Ca2+ signal transduction pathways. J. Biol. Chem. 271, 33382–33389. 438. Fuller, K., Owens, J. M., and Chambers, T. J. (1998). Induction of osteoclast formation by parathyroid hormone depends on an action on stromal cells. J. Endocrinol. 158, 341–350. 439. Gay, C. V., Zheng, B., and Gilman, V. R. (2003). Co-detection of PTH/PTHrP receptor and tartrate resistant acid phosphatase in osteoclasts. J. Cell Biochem. 89, 902–908.
256 440. Tong, H., Lin, H., Wang, H., Sakai, D., and Minkin, C. (1995). Osteoclasts respond to parathyroid hormone and express mRNA for its receptor. J. Bone Miner. Res. 10[Suppl. 1], S322. (Abstract). 441. Teti, A., Rizzoli, R., and Zambonin Zallone, A. (1991). Parathyroid hormone binding to cultured avian osteoclasts. Biochem. Biophys. Res. Commun. 174, 1217–1222. 442. Oreffo, R. O., Kusec, V., Virdi, A. S., Flanagan, A. M., Grano, M., Zambonin-Zallone, A., and Triffitt, J. T. (1999). Expression of estrogen receptor-alpha in cells of the osteoclastic lineage. Histochem. Cell Biol. 111, 125–133. 443. Orwoll, E. S. (2003). Men, bone and estrogen: unresolved issues. Osteoporos. Int. 14, 93–98. 444. Liu, B. Y., Wu, P. W., Bringhurst, F. R., and Wang, J. T. (2002). Estrogen inhibition of PTH-stimulated osteoclast formation and attachment in vitro: involvement of both PKA and PKC. Endocrinology 143, 627–635. 445. Swarthout, J. T., D’Alonzo, R. C., Selvamurugan, N., and Partridge, N. C. (2002). Parathyroid hormone-dependent signaling pathways regulating genes in bone cells. Gene 282, 1–17. 446. Fu, Q., Jilka, R. L., Manolagas, S. C., and O’Brien, C. A. (2002). Parathyroid hormone stimulates receptor activator of NFkappa B ligand and inhibits osteoprotegerin expression via protein kinase A activation of cAMP-response element-binding protein. J. Biol. Chem. 277, 48868–48875. 447. Locklin, R. M., Khosla, S., Turner, R. T., and Riggs, B. L. (2003). Mediators of the biphasic responses of bone to intermittent and continuously administered parathyroid hormone. J. Cell Biochem. 89, 180–190. 448. Demiralp, B., Chen, H. L., Koh, A. J., Keller, E. T., and McCauley, L. K. (2002). Anabolic actions of parathyroid hormone during bone growth are dependent on c-fos. Endocrinology 143, 4038–4047. 449. Bikle, D. D., Sakata, T., Leary, C., Elalieh, H., Ginzinger, D., Rosen, C. J., Beamer, W., Majumdar, S., and Halloran, B. P. (2002). Insulin-like growth factor I is required for the anabolic actions of parathyroid hormone on mouse bone. J. Bone Miner. Res. 17, 1570–1578. 450. Yamaguchi, M., Ogata, N., Shinoda, Y., Akune, T., Kamekura, S., Terauchi, Y., Kadowaki, T., Hoshi, K., Chung, U. I., Nakamura, K., and Kawaguchi, H. (2005). Insulin receptor substrate-1 is required for bone anabolic function of parathyroid hormone in mice. Endocrinology 146, 2620–2628. 451. Manolagas, S. C., Kousteni, S., and Jilka, R. L. (2002). Sex steroids and bone. Recent Prog. Horm. Res. 57, 385–409. 452. Srivastava, S., Toraldo, G., Weitzmann, M. N., Cenci, S., Ross, F. P., and Pacifici, R. (2001). Estrogen decreases osteoclast formation by down-regulating receptor activator of NF-kappa B ligand (RANKL)induced JNK activation. J. Biol. Chem. 276, 8836–8840. 453. Srivastava, S., Weitzmann, M. N., Kimble, R. B., Rizzo, M., Zahner, M., Milbrandt, J., Ross, F. P., and Pacifici, R. (1998). Estrogen blocks M-CSF gene expression and osteoclast formation by regulating phosphorylation of Egr-1 and its interaction with Sp-1. J. Clin.Invest. 102, 1850–1859. 454. Sunyer, T., Lewis, J., Collin-Osdoby, P., and Osdoby, P. (1999). Estrogen’s bone-protective effects may involve differential IL-1 receptor regulation in human osteoclast-like cells. J. Clin. Invest. 103, 1409–1418. 455. Shevde, N. K., Bendixen, A. C., Dienger, K. M., and Pike, J. W. (2000). Estrogens suppress RANK ligand-induced osteoclast differentiation via a stromal cell independent mechanism involving c-Jun repression. Proc. Natl Acad. Sci. USA 97, 7829–7834. 456. Bord, S., Ireland, D. C., Beavan, S. R., and Compston, J. E. (2003). The effects of estrogen on osteoprotegerin, RANKL, and estrogen receptor expression in human osteoblasts. Bone 32, 136–141. 457. Windahl, S. H., Hollberg, K., Vidal, O., Gustafsson, J. A., Ohlsson, C., and Andersson, G. (2001). Female estrogen receptor beta-/- mice are
JANE B. LIAN AND GARY S. STEIN
458.
459.
460.
461.
462. 463.
464.
465.
466.
467.
468.
469.
470.
471.
472.
473.
474.
partially protected against age-related trabecular bone loss. J. Bone Miner. Res. 16, 1388–1398. Robinson, J. A., Riggs, B. L., Spelsberg, T. C., and Oursler, M. J. (1996). Osteoclasts and transforming growth factor-beta: estrogenmediated isoform-specific regulation of production. Endocrinology 137, 615–621. van der Eerden, B. C., Lowik, C. W., Wit, J. M., and Karperien, M. (2004). Expression of estrogen receptors and enzymes involved in sex steroid metabolism in the rat tibia during sexual maturation. J. Endocrinol. 180, 457–467. Egerbacher, M., Helmreich, M., Rossmanith, W., and Haeusler, G. (2002). Estrogen receptor-alpha and estrogen receptor-beta are present in the human growth plate in childhood and adolescence, in identical distribution. Horm. Res. 58, 99–103. Pascoe, D., and Oursler, M. J. (2001). The Src signaling pathway regulates osteoclast lysosomal enzyme secretion and is rapidly modulated by estrogen. J. Bone Miner. Res. 16, 1028–1036. Khosla, S., Melton, L. J., III, and Riggs, B. L. (2001). Estrogens and bone health in men. Calcif. Tissue Int. 69, 189–192. Lee, S. K., Goldring, S. R., and Lorenzo, J. A. (1995). Expression of the calcitonin receptor in bone marrow cell cultures and in bone: a specific marker of the differentiated osteoclast that is regulated by calcitonin. Endocrinology 136, 4572–4581. Hattersley, G., and Chambers, T. J. (1989). Calcitonin receptors as markers for osteoclastic differentiation: correlation between generation of bone-resorptive cells and cells that express calcitonin receptors in mouse bone marrow cultures. Endocrinology 125, 1606–1612. Kajiya, H., Okamoto, F., Fukushima, H., and Okabe, K. (2003). Calcitonin inhibits proton extrusion in resorbing rat osteoclasts via protein kinase A. Pflugers Arch. 445, 651–658. Komarova, S. V., Shum, J. B., Paige, L. A., Sims, S. M., and Dixon, S. J. (2003). Regulation of osteoclasts by calcitonin and amphiphilic calcitonin conjugates: role of cytosolic calcium. Calcif. Tissue Int. 73, 265–273. Nakamura, I., Takahashi, N., Sasaki, T., Tanaka, S., Udagawa, N., Murakami, H., Kimura, K., Kabuyama, Y., Kurokawa, T., and Suda, T. (1995). Wortmannin, a specific inhibitor of phosphatidylinositol-3 kinase, blocks osteoclastic bone resorption. FEBS Lett. 361, 79–84. Lakkakorpi, P. T., and Vaananen, H. K. (1990). Calcitonin, prostaglandin E2, and dibutyryl cyclic adenosine 3′, 5′-monophosphate disperse the specific microfilament structure in resorbing osteoclasts. J. Histochem. Cytochem. 38, 1487–1493. Hoff, A. O., Catala-Lehnen, P., Thomas, P. M., Priemel, M., Rueger, J. M., Nasonkin, I., Bradley, A., Hughes, M. R., Ordonez, N., Cote, G. J., Amling, M., and Gagel, R. F. (2002). Increased bone mass is an unexpected phenotype associated with deletion of the calcitonin gene. J. Clin. Invest. 110, 1849–1857. Zaidi, M., Moonga, B. S., and Abe, E. (2002). Calcitonin and bone formation: a knockout full of surprises. J. Clin. Invest. 110, 1769–1771. Kusu, N., Laurikkala, J., Imanishi, M., Usui, H., Konishi, M., Miyake, A., Thesleff, I., and Itoh, N. (2003). Sclerostin is a novel secreted osteoclast-derived bone morphogenetic protein antagonist with unique ligand specificity. J. Biol. Chem. 278, 24113–24117. Marzia, M., Sims, N. A., Voit, S., Migliaccio, S., Taranta, A., Bernardini, S., Faraggiana, T., Yoneda, T., Mundy, G. R., Boyce, B. F., Baron, R., and Teti, A. (2000). Decreased c-Src expression enhances osteoblast differentiation and bone formation. J. Cell Biol. 151, 311–320. Zaidi, M., Moonga, B. S., and Huang, C. L. (2004). Calcium sensing and cell signaling processes in the local regulation of osteoclastic bone resorption. Biol. Rev. Camb. Philos. Soc. 79, 79–100. Kanatani, M., Sugimoto, T., Kano, J., Kanzawa, M., and Chihara, K. (2003). Effect of high phosphate concentration on osteoclast differentiation as well as bone-resorbing activity. J. Cell Physiol. 196, 180–189.
Chapter 14 The Cells of Bone 475. Dacquin, R., Davey, R. A., Laplace, C., Levasseur, R., Morris, H. A., Goldring, S. R., Gebre-Medhin, S., Galson, D. L., Zajac, J. D., and Karsenty, G. (2004). Amylin inhibits bone resorption while the calcitonin receptor controls bone formation in vivo. J. Cell Biol. 164, 509–514. 476. Kobayashi, Y., Hashimoto, F., Miyamoto, H., Kanaoka, K., MiyazakiKawashita, Y., Nakashima, T., Shibata, M., Kobayashi, K., Kato, Y., and Sakai, H. (2000). Force-induced osteoclast apoptosis in vivo is accompanied by elevation in transforming growth factor beta and osteoprotegerin expression. J. Bone Miner. Res. 15, 1924–1934. 477. McLean, W., and Olsen, B. R. (2001). Mouse models of abnormal skeletal development and homeostasis. Trends Genet. 17, S38-S43. 478. Marks, S. C., Jr., and Odgren, P. R. (2002). Structure and development of the skeleton. In “Principles of Bone Biology” (J. P. Bilezikian, L. G. Raisz, and G. A. Rodan, eds.), pp. 3–15. Academic Press, San Diego, CA. 479. Johnson, R. S., Spiegelman, B. M., and Papaioannou, V. (1992). Pleiotropic effects of a null mutation in the c-fos proto- oncogene. Cell 71, 577–586. 480. Grigoriadis, A. E., Wang, Z.-Q., Cecchini, M. G., Hofstetter, W., Felix, R., Fleisch, H. A., and Wagner, E. F. (1994). c-Fos: a key regulator of osteoclast-macrophage lineage determination and bone remodeling. Science 266, 443–448. 481. Wang, Z.-Q., Ovitt, C., Grigoriadis, A. E., Mohle-Steinlein, U., Ruther, U., and Wagner, E. F. (1992). Bone and haematopoietic defects in mice lacking c-fos. Nature 360, 741–745. 482. Tondravi, M. M., McKercher, S. R., Anderson, K., Erdmann, J. M., Quiroz, M., Maki, R., and Teitelbaum, S. L. (1997). Osteopetrosis in mice lacking haematopoietic transcription factor PU.1. Nature 386, 81–84. 483. Zhang, D. E., Hetherington, C. J., Chen, H. M., and Tenen, D. G. (1994). The macrophage transcription factor PU.1 directs tissue-specific expression of the macrophage colony-stimulating factor receptor. Mol. Cell Biol. 14, 373–381. 484. Xing, L., Franzoso, G., Siebenlist, U., and Boyce, B. (1997). Targeted disruption of the p50 and p52 subunits of NF-kB transcription factor leads to failure of osteoclast formation and osteopetrosis. J. Bone Miner. Res. 12, S136. (Abstract). 485. Ikeda, F., Nishimura, R., Matsubara, T., Tanaka, S., Inoue, J., Reddy, S. V., Hata, K., Yamashita, K., Hiraga, T., Watanabe, T., Kukita, T., Yoshioka, K., Rao, A., and Yoneda, T. (2004). Critical roles of c-Jun signaling in regulation of NFAT family and RANKL-regulated osteoclast differentiation. J. Clin. Invest. 114, 475–484. 486. Komarova, S. V., Pereverzev, A., Shum, J. W., Sims, S. M., and Dixon, S. J. (2005). Convergent signaling by acidosis and receptor activator of NF-kappaB ligand (RANKL) on the calcium/calcineurin/ NFAT pathway in osteoclasts. Proc. Natl Acad. Sci. USA 102, 2643–2648. 487. Ortega-Perez, I., Cano, E., Were, F., Villar, M., Vazquez, J., and Redondo, J. M. (2005). c-Jun NH2-terminal kinase (JNK) positively regulates NFATc2 transactivation through phosphorylation within the N-terminal regulatory domain. J. Biol. Chem. 280, 20867–20878. 488. Matsuo, K., Galson, D. L., Zhao, C., Peng, L., Laplace, C., Wang, K. Z., Bachler, M. A., Amano, H., Aburatani, H., Ishikawa, H., and Wagner, E. F. (2004). Nuclear factor of activated T-cells (NFAT) rescues osteoclastogenesis in precursors lacking c-Fos. J. Biol. Chem. 279, 26475–26480. 489. Steingrimsson, E., Tessarollo, L., Pathak, B., Hou, L., Arnheiter, H., Copeland, N. G., and Jenkins, N. A. (2002). Mitf and Tfe3, two members of the Mitf-Tfe family of bHLH-Zip transcription factors, have important but functionally redundant roles in osteoclast development. Proc. Natl Acad. Sci. USA 99, 4477–4482. 490. Hershey, C. L., and Fisher, D. E. (2004). Mitf and Tfe3: members of a b-HLH-ZIP transcription factor family essential for osteoclast development and function. Bone 34, 689–696.
257 491. Hodgkinson, C. A., Moore, K. J., Nakayama, A., Steingrimsson, E., Copeland, N. G., Jenkins, N. A., and Arnheiter, H. (1993). Mutations at the mouse microphthalmia locus are associated with defects in a gene encoding a novel basic-helix-loop-helix-zipper protein. Cell 74, 395–404. 492. Lu, M. F., Cheng, H. T., Lacy, A. R., Kern, M. J., Argao, E. A., Potter, S. S., Olson, E. N., and Martin, J. F. (1999). Paired-related homeobox genes cooperate in handplate and hindlimb zeugopod morphogenesis. Dev. Biol. 205, 145–157. 493. Herbrand, H., Pabst, O., Hill, R., and Arnold, H. H. (2002). Transcription factors Nkx3.1 and Nkx3.2 (Bapx1) play an overlapping role in sclerotomal development of the mouse. Mech. Dev. 117, 217–224. 494. Takeuchi, J. K., Koshiba-Takeuchi, K., Suzuki, T., Kamimura, M., Ogura, K., and Ogura, T. (2003). Tbx5 and Tbx4 trigger limb initiation through activation of the Wnt/Fgf signaling cascade. Development 130, 2729–2739. 495. Smits, P., Dy, P., Mitra, S., and Lefebvre, V. (2004). Sox5 and Sox6 are needed to develop and maintain source, columnar, and hypertrophic chondrocytes in the cartilage growth plate. J. Cell Biol. 164, 747–758. 496. Smits, P., and Lefebvre, V. (2003). Sox5 and Sox6 are required for notochord extracellular matrix sheath formation, notochord cell survival and development of the nucleus pulposus of intervertebral discs. Development 130, 1135–1148. 497. Pfander, D., Cramer, T., Schipani, E., and Johnson, R. S. (2003). HIF1alpha controls extracellular matrix synthesis by epiphyseal chondrocytes. J. Cell Sci. 116, 1819–1826. 498. Bi, W., Deng, J. M., Zhang, Z., Behringer, R. R., and de Crombrugghe, B. (1999). Sox9 is required for cartilage formation. Nat. Genet. 22, 85–89. 499. Rice, R., Rice, D. P., Olsen, B. R., and Thesleff, I. (2003). Progression of calvarial bone development requires Foxc1 regulation of Msx2 and Alx4. Dev. Biol. 262, 75–87. 500. Bendall, A. J., and Abate-Shen, C. (2000). Roles for Msx and Dlx homeoproteins in vertebrate development. Gene 247, 17–31. 501. Miao, D., Liu, H., Plut, P., Niu, M., Huo, R., Goltzman, D., and Henderson, J. E. (2004). Impaired endochondral bone development and osteopenia in Gli2-deficient mice. Exp. Cell Res. 294, 210–222. 502. Buttitta, L., Mo, R., Hui, C. C., and Fan, C. M. (2003). Interplays of Gli2 and Gli3 and their requirement in mediating Shh-dependent sclerotome induction. Development 130, 6233–6243. 503. Vortkamp, A., Lee, K., Lanske, B., Segre, G. V., Kronenberg, H. M., and Tabin, C. J. (1996). Regulation of rate of cartilage differentiation by Indian hedgehog and PTH-related protein. Science 273, 613–622. 504. Jemtland, R., Divieti, P., Lee, K., and Segre, G. V. (2003). Hedgehog promotes primary osteoblast differentiation and increases PTHrP mRNA expression and iPTHrP secretion. Bone 32, 611–620. 505. Hsu, S. C., Galceran, J., and Grosschedl, R. (1998). Modulation of transcriptional regulation by LEF-1 in response to Wnt-1 signaling and association with beta-catenin. Mol. Cell Biol. 18, 4807–4818. 506. Kim, S., Koga, T., Isobe, M., Kern, B. E., Yokochi, T., Chin, Y. E., Karsenty, G., Taniguchi, T., and Takayanagi, H. (2003). Stat1 functions as a cytoplasmic attenuator of Runx2 in the transcriptional program of osteoblast differentiation. Genes Dev. 17, 1979–1991. 507. Xiao, L., Naganawa, T., Obugunde, E., Gronowicz, G., Ornitz, D. M., Coffin, J. D., and Hurley, M. M. (2004). Stat1 controls postnatal bone formation by regulating fibroblast growth factor signaling in osteoblasts. J. Biol. Chem. 279, 27743–27752. 508. Takayanagi, H., Kim, S., Koga, T., and Taniguchi, T. (2005). Stat1mediated cytoplasmic attenuation in osteoimmunology. J. Cell Biochem. 94, 232–240. 509. Sunters, A., Thomas, D. P., Yeudall, W. A., and Grigoriadis, A. E. (2004). Accelerated cell cycle progression in osteoblasts overexpressing the c-fos proto-oncogene: induction of cyclin A and enhanced CDK2 activity. J. Biol. Chem. 279, 9882–9891.
258 510. Ichida, F., Nishimura, R., Hata, K., Matsubara, T., Ikeda, F., Hisada, K., Yatani, H., Cao, X., Komori, T., Yamaguchi, A., and Yoneda, T. (2004). Reciprocal roles of MSX2 in regulation of osteoblast and adipocyte differentiation. J. Biol. Chem. 279, 34015–34022. 511. Dodig, M., Tadic, T., Kronenberg, M. S., Dacic, S., Liu, Y. H., Maxson, R., Rowe, D. W., and Lichtler, A. C. (1999). Ectopic Msx2 overexpression inhibits and Msx2 antisense stimulates calvarial osteoblast differentiation. Dev. Biol. 209, 298–307. 512. Yoshizawa, T., Takizawa, F., Iizawa, F., Ishibashi, O., Kawashima, H., Matsuda, A., Endo, N., and Kawashima, H. (2004). Homeobox protein MSX2 acts as a molecular defense mechanism for preventing ossification in ligament fibroblasts. Mol. Cell Biol. 24, 3460–3472. 513. Cheng, S. L., Shao, J. S., Charlton-Kachigian, N., Loewy, A. P., and Towler, D. A. (2003). Msx2 promotes osteogenesis and suppresses adipogenic differentiation of multipotent mesenchymal progenitors. J. Biol. Chem. 278, 45969–45977. 514. Ishii, M., Merrill, A. E., Chan, Y. S., Gitelman, I., Rice, D. P., Sucov, H. M., and Maxson, R. E., Jr. (2003). Msx2 and Twist cooperatively control the development of the neural crest-derived skeletogenic mesenchyme of the murine skull vault. Development 130, 6131–6142. 515. Zamurovic, N., Cappellen, D., Rohner, D., and Susa, M. (2004). Coordinated activation of Notch, Wnt and TGF-beta signaling pathways in BMP-2 induced osteogenesis: Notch target gene Hey1 inhibits mineralization and Runx2 transcriptional activity. J. Biol. Chem. 279, 37704–37715. 516. Lian, J. B., Javed, A., Zaidi, S. K., Lengner, C., Montecino, M., van Wijnen, A. J., Stein, J. L., and Stein, G. S. (2004). Regulatory controls for osteoblast growth and differentiation: role of Runx/Cbfa/AML factors. Crit. Rev. Eukaryot. Gene Expr. 14, 1–41. 517. Hess, J., Hartenstein, B., Teurich, S., Schmidt, D., Schorpp-Kistner, M., and Angel, P. (2003). Defective endochondral ossification in mice with strongly compromised expression of JunB. J. Cell Sci. 116, 4587–4596. 518. Jochum, W., David, J. P., Elliott, C., Wutz, A., Plenk, H., Jr., Matsuo, K., and Wagner, E. F. (2000). Increased bone formation and osteosclerosis in mice overexpressing the transcription factor Fra-1. Nat. Med. 6, 980–984. 519. Kveiborg, M., Sabatakos, G., Chiusaroli, R., Wu, M., Philbrick, W. M., Horne, W. C., and Baron, R. (2004). DeltaFosB induces osteosclerosis and decreases adipogenesis by two independent cell-autonomous mechanisms. Mol. Cell Biol. 24, 2820–2830.
JANE B. LIAN AND GARY S. STEIN 520. Sabatakos, G., Sims, N. A., Chen, J., Aoki, K., Kelz, M. B., Amling, M., Bouali, Y., Mukhopadhyay, K., Ford, K., Nestler, E. J., and Baron, R. (2000). Overexpression of DeltaFosB transcription factor(s) increases bone formation and inhibits adipogenesis. Nat. Med. 6, 985–990. 521. Lee, M. H., Kim, Y. J., Kim, H. J., Park, H. D., Kang, A. R., Kyung, H. M., Sung, J. H., Wozney, J. M., Kim, H. J., and Ryoo, H. M. (2003). BMP-2-induced Runx2 expression is mediated by Dlx5, and TGF-beta 1 opposes the BMP-2-induced osteoblast differentiation by suppression of Dlx5 expression. J. Biol. Chem. 278, 34387–34394. 522. Miyama, K., Yamada, G., Yamamoto, T. S., Takagi, C., Miyado, K., Sakai, M., Ueno, N., and Shibuya, H. (1999). A BMP-inducible gene, dlx5, regulates osteoblast differentiation and mesoderm induction. Dev. Biol. 208, 123–133. 523. Tadic, T., Dodig, M., Erceg, I., Marijanovic, I., Mina, M., Kalajzic, Z., Velonis, D., Kronenberg, M. S., Kosher, R. A., Ferrari, D., and Lichtler, A. C. (2002). Overexpression of Dlx5 in chicken calvarial cells accelerates osteoblastic differentiation. J. Bone Miner. Res. 17, 1008–1014. 524. Tai, G., Polak, J. M., Bishop, A. E., Christodoulou, I., and Buttery, L. D. (2004). Differentiation of osteoblasts from murine embryonic stem cells by overexpression of the transcriptional factor osterix. Tissue Eng. 10, 1456–1466. 525. Yang, X., Matsuda, K., Bialek, P., Jacquot, S., Masuoka, H. C., Schinke, T., Li, L., Brancorsini, S., Sassone-Corsi, P., Townes, T. M., Hanauer, A., and Karsenty, G. (2004). ATF4 is a substrate of RSK2 and an essential regulator of osteoblast biology; implication for CoffinLowry Syndrome. Cell 117, 387–398. 526. Umayahara, Y., Billiard, J., Ji, C., Centrella, M., McCarthy, T. L., and Rotwein, P. (1999). CCAAT/enhancer-binding protein delta is a critical regulator of insulin-like growth factor-I gene transcription in osteoblasts. J. Biol. Chem. 274, 10609–10617. 527. Iyer, V. V., Kadakia, T. B., McCabe, L. R., and Schwartz, R. C. (2004). CCAAT/enhancer-binding protein-beta has a role in osteoblast proliferation and differentiation. Exp. Cell Res. 295, 128–137. 528. Takeshita, S., Namba, N., Zhao, J. J., Jiang, Y., Genant, H. K., Silva, M. J., Brodt, M. D., Helgason, C. D., Kalesnikoff, J., Rauh, M. J., Humphries, R. K., Krystal, G., Teitelbaum, S. L., and Ross, F. P. (2002). SHIP-deficient mice are severely osteoporotic due to increased numbers of hyper-resorptive osteoclasts. Nat. Med. 8, 943–949.
Chapter 15
Signaling in Bone T. John Martin Natalie A. Sims
St. Vincent’s Institute of Medical Research, 9 Princes Street, Fitzroy 3065, Australia University of Melbourne, Department of Melbourne, St. Vincent’s Health, 41 Victoria Pde, Fitzroy 3065, Australia
VIII. RANK Signaling IX. Coupling of Bone Formation to Resorption – Release of Growth Factors from Bone Matrix X. Coupling of Bone Formation to Resorption – Autocrine/Paracrine Regulation by Differentiating Osteoblasts XI. Coupling of Bone Formation to Resorption – Are Osteoclasts a Source of Coupling Activity? References
I. II. III. IV. V.
Abstract Introduction The Control of Osteoclasts Signaling in the Control of Osteoclast Activity Signals from the Osteoblast Lineage that Control Osteoclast Formation VI. Hormone and Cytokine Influences on the Contact-dependent Regulation of Osteoclasts VII. Discovery of the Physiological Signaling Mechanisms in Osteoclast Control
I. ABSTRACT
themselves, growing in the resorption space, can communicate through cell contact and paracrine signaling mechanisms to differentiate. Finally, it is proposed that transiently activated osteoclasts can contribute to the coupling of bone formation to resorption by producing activity that influences preosteoblast participation in bone formation.
The activities of osteoblasts and osteoclasts to form and resorb bone are controlled by circulating hormones and locally generated cytokines and growth factors. The latter are especially important, comprising complex signaling systems in bone that determine how bone is formed and resorbed in the remodeling process. Cells of the osteoblast lineage regulate osteoclast formation from hemopoietic precursors through contact-dependent mechanisms that are controlled by hormones and by the production of locally generated inhibitors. The key effectors are products of the tumor necrosis factor ligand and receptor family. Local signaling that results in bone formation during remodeling takes place in several ways. Growth factors released from resorbed bone matrix can contribute to preosteoblast differentiation and bone formation. The preosteoblasts Dynamics of Bone and Cartilage Metabolism
II. INTRODUCTION Bone formation and resorption proceed throughout life. The processes are more rapid during skeletal growth, at which stage the term modeling is used. Modeling takes place from the beginning of skeletogenesis during fetal life until the end of the second decade when the longitudinal growth of the skeleton is completed. In the modeling process bone is formed at a location different from the sites 259
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
260 of resorption, leading to a change in the shape or macroarchitecture of the bone. It is responsible for determining the size and shape of bone, such as the simultaneous widening of long bones and development of the medullary cavity by bone formation at the periosteal surface and resorption at the endosteal surface, respectively. The remodeling process, which continues throughout adult life, is an integral part of the calcium homeostatic system and provides a mechanism for self-repair and adaptation to physical stress. In the adult human skeleton, approximately 5–10% of the existing bone is replaced every year. A characteristic feature of bone remodeling is that the process does not occur uniformly throughout the skeleton. Remodeling of bone occurs in focal or discrete packets known as basic multicellular units (BMUs) of bone turnover. The cellular sequence is always initiated by osteoclastic bone resorption to be followed by osteoblastic new bone formation. This sequence of events is initiated asynchronously throughout the skeleton, at sites that are geographically and chronologically separated from each other. Both bone formation and resorption occur at the same place so that there is no change in the shape of the bone. After a certain amount of bone is removed as a result of osteoclastic resorption and the osteoclasts have moved away from the site, a reversal phase takes place in which a cement line is laid down. Osteoblasts then synthesize matrix, which becomes mineralized. The processes of bone resorption and formation are kept in balance by systemic hormones and cytokines, with the maintenance of a normal, healthy skeletal mass depending on information transfer taking place among osteoblasts, osteoclasts, immune cells, and constituents of the bone matrix. Remodeling thus maintains the mechanical integrity of the skeleton by replacing old bone with new bone [1–5]. The maintenance of adequate trabecular and cortical bone requires that bone formation and resorption should be balanced, such that a high or low level of resorption is usually associated with a similar change in the level of bone formation. The theory that resorption is followed by an equal amount of formation has come to be known as “coupling”. However, during life, the effects of growth and aging, including changes in mechanical stress, mean that this theory of equal bone replacement rarely holds true. During growth there is a positive balance, with the amount of bone replaced at individual BMUs exceeding that lost [3], and with aging there is a negative balance at individual BMUs [6], with gradual attrition of bone. In common metabolic states, such as post-menopausal osteoporosis, while coupling exists and both bone formation and resorption are occurring at a higher level than normal, the amount of bone formed is not equal to that resorbed and bone density is reduced.
T. JOHN MARTIN AND NATALIE A. SIMS
Until the early 1980s it was understood that bone metabolism was regulated by circulating hormones. Parathyroid hormone (PTH) and 1,25(OH)2 vitamin D3, promoted bone resorption, sex steroids had some poorly defined beneficial effects on the skeleton, and it was thought that there must be unknown factors promoting bone formation. The discoveries of subsequent years revealed that, although circulating hormones are important controlling factors, the key influences are locally generated cytokines which influence bone cell function and communication in complex ways, and often are themselves regulated in turn by the hormones. Discoveries of the many intercellular communication and signaling pathways in bone, together with the recent contributions from mouse and human genetics, have contributed enormously to the understanding of bone physiology and pathology, and have provided many new approaches to the development of drugs to prevent and treat bone diseases. Indeed, many cytokines that had originally been discovered by virtue of their actions on the immune or hematopoietic systems have been revealed as fundamental local factors in the control of bone cell function. This is exemplified most directly and simply by the remarkable number of skeletal phenotypes that exist in genetically altered mice, which either under- or overexpress these cytokines or their receptors.
III. THE CONTROL OF OSTEOCLASTS Osteoclasts are the only cells that can resorb bone. Their formation and activity are controlled by signals emanating predominantly from cells of the osteoblast lineage, but also from T and B cells. Further to this, in disease states other cell types have been shown to support osteoclast formation, such as breast cancer cells and synovial fibroblasts in rheumatoid arthritis. The proposal that osteoblasts mediate hormonal stimulation of osteoclast formation and activity [7] led eventually to the identification of the molecules responsible for physiological control of osteoclast formation and activity by osteoblasts. In addition to these communication pathways directing the osteoclast lineage, it has long been proposed that a local “coupling factor” plays a key role in bone remodeling by linking bone resorption to subsequent formation. Although such a coupling factor has never been identified, there have been suggestions that such activity is released from bone matrix during resorption [8, 9]. Advances of the last few years allow consideration of other contributing factors, including the possibility that signaling from the activated osteoclast could play a part in the process, helping to stimulate osteoblast lineage cells to replace the resorbed bone in each BMU. In this way,
261
Chapter 15 Signaling in Bone
intercellular communication appears to exist in both directions between osteoblasts and osteoclasts.
IV. SIGNALING IN THE CONTROL OF OSTEOCLAST ACTIVITY The first indications of the importance of intercellular signaling in bone came in the early 1980s, in that, when osteoclasts were isolated from newborn rat or mouse bone, they required the presence of contaminating osteoblastic cells in order to be fully active and resorb bone [10, 11]. The observations that isolated osteoblasts of various origins responded to bone-resorbing hormones and possessed receptors for these factors, in addition to the lack of evidence demonstrating receptors or direct responses to these hormones in osteoclasts, led to the concept that bone-resorbing factors must act first on osteoblasts, most likely bone-lining cells. This was proposed to release factors that influence the bone-resorbing activity of osteoclasts [7, 12]. Furthermore, since osteoclasts are derived from hemopoietic progenitors and not from a local bone cell, Chambers came to the same conclusion by arguing that, since the osteoclast derives from a “wandering” cell, it made sense to have its activity programmed by an authentic bone cell, i.e. the osteoblast [13]. Work over the next few years established the concept of an osteoblast-derived stimulator of osteoclast resorption [10, 14, 15] and by the mid-1980s the field was at the stage of accepting that intercellular signaling may be an important mechanism in bone cell biology. Among the questions raised were whether a single stromal/osteoblastic cell factor existed that is responsible for osteoclast activation, and if so, is it cell-associated or secreted? Would the formation of osteoclasts be at least as important as regulation of their activity? The next real advances came with the development of methods to study osteoclast formation in vitro.
V. SIGNALS FROM THE OSTEOBLAST LINEAGE THAT CONTROL OSTEOCLAST FORMATION Major advances required the development of murine bone marrow cultures, with reproducible assays of osteoclast formation [16–19]. Takahashi observed that more than 90% of the TRAP-positive mononucleated cell clusters and multinucleated cells formed in mouse marrow cultures, in response to bone-resorbing stimuli, were located near colonies of alkaline phosphatase-positive mononucleated
cells (possibly osteoblasts) [18]. This suggested that osteoblastic cells are involved in osteoclast formation, in addition to the evidence produced in the few earlier years of their influence on osteoclast activity. Proof of this came from the experimental system established by Takahashi [16] in which osteoclasts were formed from osteoblast-rich cultures from newborn mouse calvariae grown in co-culture with mouse spleen cells treated with 1,25(OH)2D3. Importantly, when the osteoblastic cells were separated from the spleen cells by a 0.45−µm membrane filter, but shared the same media, no osteoclast formation occurred. Similar results were obtained with bone marrow-derived stromal cell lines including MC3T3-G2/PA6, ST2, and KS-4 [20, 21]. Thus, it was clear that direct cell-to-cell contact of osteoclast precursors with osteoblasts, or their precursors, is necessary for osteoclast differentiation.
VI. HORMONE AND CYTOKINE INFLUENCES ON THE CONTACTDEPENDENT REGULATION OF OSTEOCLASTS With increasing acceptance of a role for cells of the osteoblast lineage in controlling osteoclast formation and activity by a contact-dependent mechanism, it was important to understand how this process was regulated by hormones and cytokines that stimulate osteoclast formation. Prostaglandin (PG)-induced osteoclast formation in mouse bone marrow cultures was mediated by cAMP [22]. Likewise, PTH and PTHrP, acting through their common receptor, promoted osteoclast formation in marrow cultures by a cAMP-dependent mechanism [23, 24], and the effect of interleukin-1 (IL-1) resulted from the generation of PGE2 as an intermediate effector [25]. There remains no convincing evidence for the existence of functional PTH/ PTHrP receptors in osteoclasts, capable of supporting a direct stimulatory effect of PTH on the osteoclast. A second signaling mechanism for regulation of osteoclastogenesis by osteoblasts was provided by the steroid hormone, 1,25(OH)2D3 which had very similar effects on osteoclast formation in marrow cultures and in co-cultures of osteoblasts with hemopoietic cells [17]. 1,25(OH)2D3 signals by combining with its receptor and translocating to the nucleus to influence transcriptional events. Finally, a membrane-bound receptor complex involving a 130 kDa glycoprotein (gp130) [26] provides for osteoclast formation under the influence of the group of cytokines that use this signaling mechanism. In co-cultures of mouse stromal cells with hemopoietic cells, simultaneous
262
T. JOHN MARTIN AND NATALIE A. SIMS
treatment with IL-6 and its soluble receptor (sIL-6R) induced osteoclast formation, but when added separately they were ineffective [26]. The other cytokines in this group, IL-11, leukemia inhibitory factor (LIF), and oncostatin M (OSM), all of which use gp130 as a common transducer, also stimulated osteoclast formation [26]. When cells from IL-6R-overexpressing transgenic mice were used in crossover co-cultures with hemopoietic cells from wildtype mice, the expression of IL-6R by osteoblastic cells was shown to be indispensable for the induction of osteoclasts [27]. This clearly demonstrated that stimulation of osteoclast formation by IL-6 required action of the cytokine on the osteoblast, despite the fact that osteoclasts possessed its receptor. A central role of the gp130-coupled cytokines in osteoclast development was further suggested by the observations that PTH, 1, 25(OH)2D3 , PGE2 , IL-1, and TNFα all promoted IL-11 production by osteoblastic cells [28]. Furthermore, addition of neutralizing anti-gp130 to co-cultures fully blocked stimulation of osteoclast formation in response to IL-1, and partly blocked the responses to PTH, 1,25(OH)2D3 , and PGE2. Thus the concept of stromal/osteoblastic regulation of osteoclastogenesis through local signaling mechanisms was firmly established, along with its regulation by a number of circulating and local factors. Despite the fact that they fell into three main classes with respect to their initial signaling mechanisms (Fig. 1), it seemed that a common pathway for these agents was the membrane stromal factor called variously “stromal osteoclast-forming activity” (SOFA) [29] or “osteoclast differentiation factor” (ODF) [19]. It was assumed that these agents must converge in their actions at some stage before finally generating the crucial membrane factor [30, 31].
VII. DISCOVERY OF THE PHYSIOLOGICAL SIGNALING MECHANISMS IN OSTEOCLAST CONTROL Of the many multifunctional cytokines that had some role in osteoclast formation, none provided an explanation for the molecular regulation of osteoclast formation and activity that was evident from the foregoing studies. In a form of murine osteopetrosis resulting from a mutation in the coding region of the M-CSF gene in the op/op mouse [32, 33], M-CSF was found to play a role in both proliferation and differentiation of osteoclast progenitors [34]. On the other hand, M-CSF inhibited the bone-resorbing activity of isolated osteoclasts [35], and osteoclasts were found to be rich in M-CSF receptors [36]. Bone resorption in organ culture was reduced by M-CSF, GM-CSF, and IL-3 [37], and all three cytokines inhibited osteoclast generation in mouse bone marrow cultures. The conclusion from these and other observations was that none of these hemopoietic growth factors fulfilled criteria expected of one that is specific for osteoclast formation, and certainly not the predicted ODF/SOFA. The hypothesis was compelling, but, by the mid-1990s there had been no significant advances towards identifying these control mechanisms. Resolution of the question came in 1997 with the discovery by two groups independently that osteoclast formation is controlled physiologically by regulated interactions among members of the TNF ligand and receptor families. The discovery of osteoprotegerin (OPG), a soluble member of the TNF receptor superfamily, revealed it as a powerful inhibitor of osteoclast formation [38, 39].
Figure 1 Three distinct signaling pathways in cells of the osteoblast lineage result in osteoclast formation, through a mechanism depending on contact between the osteoblastic cells and the hemopoietic precursors of osteoclasts. The membrane protein responsible for promoting osteoclast differentiation was discovered to be RANKL, binding to its receptor, RANK. The process is inhibited by the decoy receptor, OPG (not shown).
263
Chapter 15 Signaling in Bone
This provided the means of identifying and cloning the elusive ODF/SOFA, which proved to be a TNF ligand family member that came to be called receptor activator of nuclear factor κB ligand (RANKL), as the common factor mediating osteoclast formation in response to all known stimuli [40, 41]. As a membrane protein, RANKL fulfilled the predictions of earlier work, that osteoclast differentiation required contact-dependent activation of hemopoietic precursors. The communication with the hemopoietic lineage results from RANKL binding to its receptor on the osteoclast lineage, RANK, thereby initiating signaling essential for osteoclast differentiation. Figure 1 illustrates the regulated production of RANKL that is essential for osteoclast formation and maintenance of activity. The boneresorbing cytokines and hormones, with disparate signaling mechanisms, converge in promoting RANKL production. The decoy receptor, OPG, has an essential physiological role as a paracrine regulator of osteoclast formation, produced by the osteoblasts and binding RANKL to limit its activation of osteoclast formation through its receptor, RANK. All of these discoveries have been validated by studies in genetically altered mice, establishing clearly the essential physiological role of these TNF ligand and receptor family members in controlling osteoclast formation and activity. First, transgenic overexpression of OPG in mice results in generation of mice with osteopetrosis because of failure to form osteoclasts [38]. Genetic ablation of OPG, on the other hand, leads to severe high turnover osteoporosis [42]. Removal of the stimulatory pathway by genetic ablation of RANKL also results in osteopetrosis because RANKL is necessary for normal osteoclast formation [43]. Finally, genetic ablation of the hemopoietic cell receptor, RANK, also leads to osteopetrosis [44]. Because this signaling pathway is functional also in immune cells, RANK-null mice have severe abnormalities in that system, reflecting an intriguing link with the immune system, which was evident very early in the discovery process. The role of OPG as a paracrine effector has been illustrated by in vitro experiments that lead to the conclusion that constitutive production of this decoy receptor is necessary to limit the osteoclast formation resulting from RANKL stimulation [45]. Two other groups were successful in identifying and cloning RANKL, but these groups were interested in its role in the immune response [46, 47]. Wong et al. [46] identified and characterized a TNF-related activationinduced cytokine (TRANCE) during a search for apoptosisregulatory genes in murine T-cell hybridomas, finding it to be predominantly expressed on T cells and in lymphoid organs and controlled by the T-cell receptor through a calcineurin-regulated pathway. The putative receptor for
TRANCE was detected on mature dendritic cells [46, 48]. In studying the processing and presentation of antigens by dendritic cells to T cells, Anderson et al. [47] characterized receptor activator of NF-κB (RANK), a new member of the TNF receptor family derived from dendritic cells, and its ligand RANKL, which they recognized to be identical to TRANCE [46]. Furthermore, production of soluble RANKL by activated T cells directly stimulated osteoclast formation in vitro and in vivo [49], with important pathophysiological implications. It is apparent that lymphocytes exert important regulatory controls on osteoclast function. Physiological control of bone resorption is thus dependent on the essential regulatory function of RANKL, not only in promoting osteoclast formation, but also their survival and activity [50], as predicted from the earlier demonstration of osteoclast activation through contact with osteoblastic cells [10, 51]. By treating with RANKL and M-CSF it became possible to prepare osteoclasts in relatively large numbers without the participation of stromal/osteoblastic precursors, including the preparation of human osteoclasts from peripheral blood [52, 53]. As predicted, the important osteotropic hormones and cytokines, including 1,25(OH)2D3, IL-11, PTH, and PGE2, stimulated osteoblast RANKL production, triggering osteoclast development, and the same treatments reduced production of OPG. The physiological function of OPG as a paracrine regulator, suggested by the phenotypes of OPG-null and OPG-transgenic mice, was demonstrated using in vitro organ and cell culture methods also [45]. The functions of RANKL extend to pathological states of increased bone resorption, where increased local RANKL production contributes to bone destruction in disease states such as breast cancer metastases to bone [54] in rheumatoid arthritis [49], multiple myeloma [55], and osteoporosis [56].
VIII. RANK SIGNALING The activation of RANK by its ligand initiates signaling events in cells of the monocyte–macrophage series that result in the expression of genes required for osteoclast differentiation. The key initial step is binding of the intracellular domain of RANK to TNFR-associated cytoplasmic factors (TRAFs). The C-terminal region within residues 544 to 616 is required for binding to TRAF 1, 2, 3, 5, and 6 [57, 58], and there is an additional separate binding site for TRAF 6 at amino acids 340–421 [58]. RANK signaling was found to activate both NF-κB and JNK activity, and deletion of the 72 C-terminal residues of RANK resulted
264 in a major reduction in NF-κB and JNK activity [57, 58]. Further ablation of the TRAF 6-binding region ablated all residual NF-κB and JNK activity [58]. Mutation of the TRAF 6-binding site in RANK completely inhibited NF-κB activation, while JNK activation was inhibited to a lesser extent. It was concluded that binding of RANK to TRAF 6 is pivotal for NF-κB activation, and the TRAF 2-binding site was subsequently shown to be required for JNK activation [57]. Earlier work had shown that inactivating mutations of both the p50 and p52 components of NF-κB in mice resulted in osteopetrosis because of failed osteoclast development [59]. In keeping with the importance of NF-κB signaling in osteoclast formation, mice rendered null for the TRAF 6 gene have osteopetrosis [60, 61], consistent with a critical role for RANK/TRAF 6 in the activation of osteoclastic bone resorption. A novel aspect of signaling in osteoclast development is the ability of RANKL to limit its own osteoclastogenic effect by promoting interferon-β (IFN-β) production by monocytic osteoclast precursors [62]. This inhibits osteoclast formation by preventing RANKL-induced expression of c-fos. The latter was known as an essential transcription factor in osteoclast differentiation, since c-fos-null mice were osteopetrotic because of failed osteoclast formation [63]. Participation by IFN-β in this internal control process provides another mechanism of local signaling that could regulate bone remodeling. The extent of osteoclastic resorption within individual BMUs needs to be limited. If it were to continue unchecked, resorption would be excessive. RANKL-induced IFN-β provides an appealing mechanism, contained within the osteoclast itself, by which osteoclast formation could be controlled at critical sites in normal bone remodeling. Further elucidation of the RANK signaling pathway came with the identification that RANKL treatment considerably enhanced the expression and nuclear location of nuclear activator of activated T cells (NFAT)-2 (NFATc1), a transcription factor regulated by calcineurin and calcium [64]. It was shown by several methods that NFATc1 activation is necessary and sufficient for osteoclast differentiation. NFATc1 up-regulation did not occur in either c-fos or TRAF 6-null mice, suggesting that both pathways are necessary for up-regulation of NFATc1 by RANKL. A further, novel aspect of RANKL signaling emerged in this work with the demonstration that RANKL-induced oscillations in cytoplasmic calcium in osteoclast precursors were necessary for its effect. Both calcium chelators and calcineurin inhibitors suppressed NFATc1 expression and nuclear translocation. These discoveries provided the first real link between the observations of failed osteoclast development in mice deficient in RANK signaling and c-fos-null mice.
T. JOHN MARTIN AND NATALIE A. SIMS
IX. COUPLING OF BONE FORMATION TO RESORPTION – RELEASE OF GROWTH FACTORS FROM BONE MATRIX While much is known about the communication processes among cells of bone that determine osteoclast formation and activity, less is understood of the signaling processes that govern osteoblast differentiation and bone formation. In particular it is unknown how the extent of bone formation in response to resorption is controlled, and how the two processes are so closely linked. Interventions that increase bone resorption are usually accompanied by increased bone formation. A recent example of this comes from observations in mice overexpressing TGFβ in osteoblasts, that had increased bone resorption and formation, with a net bone loss [65]. By crossing these mice with other genetically manipulated mice these authors showed that the increased formation was dependent upon osteoclastic resorption and was not a direct effect of TGF. The very nature of the remodeling process, occurring as it does in different parts of the skeleton at different times, highlights the importance of locally generated and regulated factors in the process. When the cycle is initiated, say by PTH or by mechanical strain which would generate cytokines or prostanoids [66], the thin layer of nonmineralized matrix under these cells is initially digested by collagenase to expose the mineralized matrix which osteoclasts can resorb [67]. PTH stimulates collagenase production and secretion in osteoblastic cells [68]; in mice whose collagenase is unable to digest type I collagen, the bone resorbing action of PTH is severely attenuated [69]. According to this model of initiation of resorption, osteoclasts or their late precursors near to final maturation and activation must be available close to the sites where they are required. Although this is likely, there is no direct in vivo evidence for it. Toward the end of resorption, mononuclear cells are seen at the bottom of resorption pits, where they remove demineralized collagen. The mononuclear cells ascribed this function have sometimes been fibroblasts, but more commonly macrophages. Recent work, applying light and electron microscopy to a number of in vitro models, describes how bone-lining cells, through the mediation of matrix metalloproteinases, carry out the task of cleaning resorption pits after osteoclasts have completed their task [70]. In order to do this the cells must use differentiated properties that they acquire and retain as part of the osteoblast lineage. These include especially the capacity to respond appropriately to the cytokines and growth factors that most likely govern these steps. A favored hypothesis for some years is that this “coupling” is regulated by bone formation factors released
Chapter 15 Signaling in Bone
from the bone matrix during bone resorption [9, 10]. Indeed, a large number of substances which are mitogenic to osteoblasts or which stimulate bone formation in vivo can be extracted from bone matrix [71]. These include insulin-like growth factor (IGF) I and II [72], acidic and basic fibroblast growth factor (FGF) [73], transforming growth factor ß (TGFß) 1 and 2 [74], and TGFß heterodimers [75], bone morphogenetic proteins (BMPs) 2, 3, 4, 6, and 7 [76–78], platelet-derived growth factor (PDGF) [79], and probably others. Several questions should be considered regarding the role of these substances in the coupling of bone formation to bone resorption: (1) which cells produce them and under what circumstances?; (2) do they stimulate bone formation in vivo?; (3) can they be released from the matrix in active form and in controlled amounts during bone resorption?; (4) is there evidence for an increase in the abundance of these substances at sites of bone remodeling?; and (5) are there regulated mechanisms by which they are activated? IGF-I, IGF-II, bFGF, TGFβ, and PDGF are produced by rat osteoblastic cells. IGF-I and IGF-II production is enhanced by stimulators of bone formation, such as PGE2 and PTH [80, 81]. Elevated levels of IGF-I mRNA were found in bone from estrogen-deficient rats, where bone turnover is increased [82, 83]. During bone growth in rats, there is a close association between osteogenesis and IGF-I expression [83]. However, following marrow ablation which causes a substantial increase in bone formation, the rise in IGF-I mRNA was seen after the appearance of differentiated osteoblasts, suggesting that it did not initiate bone formation in that system [84]. In human bone, the major form of IGF is IGF-II, which is also produced by human bone cells in culture [9, 10]. Bone is one of the most abundant sources of TGFβ [85], and this growth factor is produced by all osteoblastic cells examined. The BMPs are members of the TGFβ superfamily. BMP-2 and BMP-4 are produced in adult bovine pre-odontoblasts [86] and human fetal teeth [87]. BMP-7 (OP-1) was localized in human embryos in hypertrophic chondrocytes, osteoblasts, and periosteum as well as other tissues [77], while BMP-3 was found in human embryonic lung and kidney in addition to perichondrium, periosteum, and osteoblasts [78]. Both bFGF [88] and PDGF [79] were shown to be produced by bone cells or bone explants in culture. These factors could thus be involved in bone remodeling but the time and site for their synthesis and secretion in vivo has not yet been determined. Prostaglandin E, primarily E2, is another bone cell-produced cytokine, which is up-regulated by mechanical strain in vitro [89] and stimulates both bone resorption and formation [90]. Many growth factors stimulate bone formation in vivo. IGF-I, injected into humans or rats, increases both bone
265 resorption and bone formation [91], yet reports on its relative effects on trabecular and cortical bone are inconclusive [92]. When injected together with the IGF-binding protein IGFBP-3 into rats, it was reported to increase bone volume [93]. BMPs injected into bone stimulate bone formation locally and produce a positive bone balance. TGFβ, from the same family of proteins, has a similar effect. When injected next to the periosteum or endosteum, local bone formation is substantially augmented in rats and other species [94, 95]. At the same time, endocortical bone resorption is increased; thus, like IGF-I, TGFβ seems to stimulate both resorption and formation, however, the local balance is clearly positive. bFGF, injected both locally and systemically, was also reported to increase bone formation [96]. PGE1 and PGE2 have long been known to be potent stimulators of bone formation when given either locally or systemically, both to humans or experimental animals [97, 98]. These substances could thus contribute to the bone formation observed in remodeling, if secreted or released in active form at the appropriate site and time. It was proposed that TGFß, which is produced as an inactive precursor in bone and bone cells is present in the matrix and can be activated by acidification or proteolytic cleavage by resorbing osteoclasts [99]. Consistent with this, TGFß activity was recovered from conditioned media of in vitro resorbing osteoclasts [100]. It remains to be shown if the other growth factors also survive the proteolytic cleavage of the acidic hydrolases present in resorption lacunae. Other questions raised by this model of coupling, via growth factor release from the matrix, relate to the time course and the distance between the resorption and formation processes and whether activation can be controlled with sufficient precision in this way. Osteoclastic bone resorption proceeds for about 2 weeks before formation follows and continues for 3–4 months. The osteoblast precursors, which should respond to the “coupling factors” are many microns away from where active osteoclast resorption is in progress. The osteoblastic lineage cells produce TGFβ in latent form and the IGFs as complexes bound to a family of specific, high-affinity binding proteins (IGFBPs), which regulate their bioavailability [101]. TGFβ may be released from latent complexes at appropriate sites in bone by plasmin generated locally by plasminogen activators, in a manner controlled temporally and spatially by hormones and cytokines [102]. A similar local control could free IGF-1 from association with its inhibitory binding protein [103]. Although there is no obvious skeletal phenotype in mice with inactivated genes for plasminogen activators, in vivo investigation of such possibilities would require treatment of such animals with anabolic agents such as PTH.
266
X. COUPLING OF BONE FORMATION TO RESORPTION – AUTOCRINE/PARACRINE REGULATION BY DIFFERENTIATING OSTEOBLASTS The above theory of coupling requires dissolution of growth factors from matrix and their activation by acidification and/or proteases. An extension of this that is not necessarily exclusive arises from the work of Boyde and colleagues [104], in which they showed in vitro that rat calvarial cells grown on bone slices with mechanically excavated crevices and grooves made bone in those defects, filling them exactly to a flat surface. The findings suggested that it is the topography of the bone that determines the timing, siting, and extent of new bone formation, and that in vivo this would take place in the resorbed spaces vacated by osteoclasts. Both the proposed growth factor involvement and the work of Gray et al. [104] imply that, once the formation process is established, the participating cells themselves are able to sense spatial limits, and most likely do so by chemical communication that takes place between the developing osteoblasts. The likely mediators of these signaling processes are the same growth factors and cytokines that are proposed to be of matrix origin.
XI. COUPLING OF BONE FORMATION TO RESORPTION – ARE OSTEOCLASTS A SOURCE OF COUPLING ACTIVITY? Observations in genetically manipulated mice suggest that the osteoclast itself could also be the source of an activity that contributes to the fine control of the coupling process. Generation of coupling activity was suggested by increased bone formation in OPG−/− mice [105], which are severely osteoporotic because of excessive osteoclast formation. In bone sections from mice in this high bone-turnover state sites of active bone resorption very commonly had active osteoblasts located nearby, suggesting that the coupling activity in this high-turnover state was more likely derived from osteoclasts themselves. Some indication of an osteoclast role comes also from human genetics. In individuals with the osteopetrotic syndrome, ADOII, due to inactivating mutations in the chloride-7 channel (ClC-7), bone resorption is deficient because of the failure of the osteoclast acidification process. Bone formation in these patients is nevertheless normal, rather than diminished as might be expected because of the greatly impaired resorption [106]. Furthermore, in mice deficient in either c-src [107], ClC-7 [108], or tyrosine phosphatase epsilon [109], bone resorption is inhibited
T. JOHN MARTIN AND NATALIE A. SIMS
without inhibition of formation. In these three knockout mouse lines osteoclast resorption is greatly reduced by the mutation, although osteoclast numbers are not reduced. Indeed osteoclast numbers are actually increased because of reduced osteoclast apoptosis. A possibility is that these osteoclasts, although unable to resorb bone, are nevertheless capable of generating a factor (or factors) required for bone formation. On the other hand, mice lacking c-fos, which are unable to generate osteoclasts, have reduced bone formation as well as resorption [63]. The cytokines that signal through gp130 play an important role in intercellular communication processes in bone, with evidence indicating that they can be involved in regulation of both bone resorption and formation [28, 110]. We addressed this by studying mice in which each of the two gp130-dependent signaling pathways was specifically attenuated, and found that inactivation of the SHP2/ras/ MAPK signaling pathway (gp130Y757F/Y757F mice) yielded mice with greater osteoclast numbers and bone resorption, as well as greater bone formation than wild-type mice. This increased bone remodeling resulted in less bone because the increase in resorption was relatively greater than that in formation. In other words the coupling process was imprecise in a way that resembles the result of estrogen withdrawal, as in ovariectomy. gp130Y757F/Y757F mice crossed with IL-6-null mice had similarly high osteoclast numbers and increased bone resorption, however these mice showed no corresponding increase in bone formation and thus had extremely low bone mass. Thus resorption alone is insufficient to promote the coupled bone formation, but the active osteoclasts are the likely source. Furthermore this indicated that stimulation of bone formation coupled to the high level of bone resorption in gp130Y757F/Y757F mice is an IL-6-dependent process, though it does not necessarily show that it is mediated by IL-6 itself [111]. Evidence of quite a different nature points further to a role for the activated osteoclast in the coupling process. Mice rendered null for calcitonin (CT−/− mice) have increased bone formation [112], as do those deficient in the CT receptor (CTR+/− mice) [113]. These unexpected findings can be explained by the hypothesized role of active osteoclasts. The best documented action of calcitonin is its acute inhibition of osteoclast function following injection [114]. If transient osteoclast activation is necessary for bone formation, then removal of calcitonin would result in osteoclasts remaining active in a way that allows them to continue contributing to bone formation, thus giving rise to the phenotype expressed in CT−/− and in CTR+/− mice. The real physiological function of CT has never been clearly understood, but these new observations, together with what is proposed, could reveal it as a short-term regulator of osteoclast activity which
267
Chapter 15 Signaling in Bone
Figure 2 Three main pathways contributing to the process of coupling of bone formation to resorption. 1. Osteoclasts resorb matrix, releasing stored growth factors that are able to promote osteoblast precursor division or maturation [9]. 2. Preosteoblasts in the resorption space divide, communicate with each other through gap junctions and paracrine signals to program differentiation and sense spatial requirements [104]. 3. Osteoclasts, activated by RANKL generated in osteoblasts, release coupling activity that is required for the osteoblast response [123].
through its action on the osteoclast also regulates bone formation. Finally, the finding that effective inhibition of resorption markedly attenuates, or even abolishes, the anabolic effect of PTH, also points to a role for the osteoclast. The anabolic effect of PTH is lost in c-fos−/− mice that are osteopetrotic because of failure of osteoclast formation [115]. Treatment of patients with osteoporosis concomitantly with PTH and a bisphosphonate resulted in significant early blunting of the anabolic response to PTH [116, 117], with the obvious implication that combining the anabolic PTH with antiresorptive bisphosphonate would be contraindicated [118, 119]. If some aspect of resorption is required for the anabolic effect of PTH, how is that connection made? Treatment of rats with a single subcutaneous injection of PTH results in a transient increase in mRNA for RANKL and a decrease in that for OPG, with maximum effect at 1 hour, and returning to control within 2 hours, leading us to suggest that a subtle or transient increase in osteoclast formation or activation might be needed to prepare the bone surface for new matrix deposition [120, 121]. Finally and most importantly, Holtrop et al. [122] showed that intravenous injection of PTH in young rats resulted in transient activation of osteoclasts in vivo, evident within 30 minutes, and followed only some hours later at high PTH doses by increased osteoclast number. Thus it is clear that rapid activation of osteoclasts in response to a single injection of PTH can occur, and it is likely that this is the consequence of a transient increase in RANKL production by cells at sites available for activation of BMUs. Our interpretation of the foregoing and other data is that what is needed for full expression of the anabolic action of PTH, in addition to its direct effects on committed
preosteoblasts, is a transient effect on the osteoclast, achieved by promoting activation, rather than formation, of osteoclasts [123]. The precise way in which the osteoclast is involved in the anabolic process needs to be clearly understood because of the implications for sequential or combined use of therapeutic resorption inhibitors and anabolic agents, and for the development of new anabolic agents. Figure 2 illustrates this schematically, with the hypothesis that there can be at least three main pathways involved in the coupling of bone formation to resorption. The first is the release of any of several growth factors from bone matrix as a result of resorption. The second is the communication among osteoblast precursors, through gap junctions and growth factors, that equips them to sense the space that needs to be filled after resorption. The third is the postulated production of coupling activity from activated osteoclasts, a regulated process that could depend on the controlled generation of RANKL and transient osteoclast activation.
Acknowledgement Work from the authors’ laboratories supported by the National Health & Medical Research Council of Australia.
References 1. Frost, H. M. (1964). Dynamics of bone remodeling. In “Bone Biodynamics” (H. M. Frost, ed.), pp. 315–333. Little, Brown & Co., Boston, MA. 2. Harris, W. H., and Heaney, R. P. (1969). Skeletal renewal and metabolic bone disease. N. Engl. J. Med. 280, 193–202.
268 3. Parfitt, A. M. (1982). The coupling of bone formation to bone resorption: a critical analysis of the concept and of its relevance to the pathogenesis of osteoporosis. Metab. Bone Dis. Relat. Res. 4, 1–6. 4. Eriksen, E. F. (1986). Normal and pathological remodeling of human trabecular bone: three dimensional reconstruction of the remodeling sequence in normals and in metabolic bone disease. Endocr. Rev. 7, 379–408. 5. Martin, T. J., and Rodan, G. A. (2001). Coupling of bone resorption and formation during bone remodeling. In “Osteoporosis” (J., Kelsey, ed.), 2nd ed., pp. 361–372. Academic Press, San Diego, CA. 6. Lips, P., Courpron, P., and Meunier, P. J. (1978). Mean wall thickness of trabecular bone packets in the human iliac crest: changes with age. Calcif. Tissue Res. 26, 13–17. 7. Rodan, G. A., and Martin, T. J. (1981). Role of osteoblasts in hormonal control of bone resorption – a hypothesis. Calcif. Tissue Int. 33, 349–351. 8. Howard, G. A., Bottemiller, B. L., Turner, R. T., Rader, J. I., and Baylink, D. J. (1981). Parathyroid hormone stimulates bone formation and resorption in organ culture: evidence for a coupling mechanism. Proc. Natl Acad. Sci. USA 78, 3204–3208. 9. Mohan, S., and Baylink, D. J. (1991). The role of IGF-II in the coupling of bone formation to resorption. In “Modern Concepts of Insulin-like Growth Factors” (E. M. Spencer, ed.), pp. 19–174. Elsevier, New York. 10. Chambers, T. J., and Magnus, C. J. (1982). Calcitonin alters behaviour of isolated osteoclasts. J. Pathol. 136, 27–39. 11. Chambers, T. J. (1985). The pathobiology of the osteoclast. J. Clin. Pathol. 38, 241–252. 12. Martin, T. J., Partridge, N. C., Greaves, M., Atkins, D., and Ibbotson, K. J. (1979). Prostaglandin effects on bone and role in cancer hypercalcemia. In “Molecular Endocrinology” (M. Szelke, ed.), pp. 251–264. Elsevier, Amsterdam. 13. Chambers, T. J. (1980). The cellular basis of bone resorption. Clin. Orthop. 283–293. 14. McSheehy, P. M., and Chambers, T. J. (1986). Osteoblast-like cells in the presence of parathyroid hormone release soluble factor that stimulates osteoclastic bone resorption. Endocrinology 119, 1654–1659. 15. McSheehy, P. M., and Chambers, T. J. (1986). Osteoblastic cells mediate osteoclastic responsiveness to parathyroid hormone. Endocrinology 118, 824–828. 16. Takahashi, N., Akatsu, T., Udagawa, N., Sasaki, T., Yamaguchi, A., Moseley, J. M., Martin, T. J., and Suda, T. (1988). Osteoblastic cells are involved in osteoclast formation. Endocrinology 123, 2600–2602. 17. Takahashi, N., Akatsu, T., Sasaki, T., Nicholson, G. C., Moseley, J. M., Martin, T. J., and Suda, T. (1988). Induction of calcitonin receptors by 1 alpha, 25-dihydroxyvitamin D3 in osteoclast-like multinucleated cells formed from mouse bone marrow cells. Endocrinology 123, 1504–1510. 18. Takahashi, N., Yamana, H., Yoshiki, S., Roodman, G. D., Mundy, G. R., Jones, S. J., Boyde, A., and Suda, T. (1988). Osteoclast-like cell formation and its regulation by osteotropic hormones in mouse bone marrow cultures. Endocrinology 122, 1373–1382. 19. Suda, T., Takahashi, N., and Martin, T. J. (1992). Modulation of osteoclast differentiation. Endocr. Rev. 13, 66–80. 20. Udagawa, N., Takahashi, N., Akatsu, T., Sasaki, T., Yamaguchi, A., Kodama, H., Martin, T. J., and Suda, T. (1989). The bone marrow-derived stromal cell lines MC3T3-G2/PA6 and ST2 support osteoclast-like cell differentiation in cocultures with mouse spleen cells. Endocrinology 125, 1805–1813. 21. Yamashita, T., Asano, K., Takahashi, N., Akatsu, T., Udagawa, N., Sasaki, T., Martin, T. J., and Suda, T. (1990). Cloning of an osteoblastic cell line involved in the formation of osteoclast-like cells. J. Cell. Physiol. 145, 587–595. 22. Akatsu, T., Takahashi, N., Debari, K., Morita, I., Murota, S., Nagata, N., Takatani, O., and Suda, T. (1989). Prostaglandins promote osteoclastlike
T. JOHN MARTIN AND NATALIE A. SIMS
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
cell formation by a mechanism involving cyclic adenosine 3′, 5′monophosphate in mouse bone marrow cell cultures. J. Bone. Miner. Res. 4, 29–35. Akatsu, T., Takahashi, N., Udagawa, N., Sato, K., Nagata, N., Moseley, J. M., Martin, T. J., and Suda, T. (1989). Parathyroid hormone (PTH)-related protein is a potent stimulator of osteoclast-like multinucleated cell formation to the same extent as PTH in mouse marrow cultures. Endocrinology 125, 20–27. Evely, R. S., Bonomo, A., Schneider, H. G., Moseley, J. M., Gallagher, J., and Martin, T. J. (1991). Structural requirements for the action of parathyroid hormone-related protein (PTHrP) on bone resorption by isolated osteoclasts. J. Bone Miner. Res. 6, 85–93. Akatsu, T., Takahashi, N., Udagawa, N., Imamura, K., Yamaguchi, A., Sato, K., Nagata, N., and Suda, T. (1991). Role of prostaglandins in interleukin-1-induced bone resorption in mice in vitro. J. Bone Miner. Res. 6, 183–189. Tamura, T., Udagawa, N., Takahashi, N., Miyaura, C., Tanaka, S., Yamada, Y., Koishihara, Y., Ohsugi, Y., Kumaki, K., Taga, T., Kishimoto, T., and Suda, T. (1993). Soluble interleukin-6 receptor triggers osteoclast formation by interleukin 6. Proc. Natl Acad. Sci. USA 90, 11924–11928. Udagawa, N., Takahashi, N., Katagiri, T., Tamura, T., Wada, S., Findlay, D. M., Martin, T. J., Hirota, H., Taga, T., Kishimoto, T., and Suda, T. (1995). Interleukin (IL)-6 induction of osteoclast differentiation depends on IL-6 receptors expressed on osteoblastic cells but not on osteoclast progenitors. J. Exp. Med. 182, 1461–1468. Romas, E., Udagawa, N., Zhou, H., Tamura, T., Saito, M., Taga, T., Hilton, D. J., Suda, T., Ng, K. W., and Martin, T. J. (1996). The role of gp130-mediated signals in osteoclast development: regulation of interleukin 11 production by osteoblasts and distribution of its receptor in bone marrow cultures. J. Exp. Med. 183, 2581–2591. Chambers, T. J., Owens, J. M., Hattersley, G., Jat, P. S., and Noble, M. D. (1993). Generation of osteoclast-inductive and osteoclastogenic cell lines from the H-2KbtsA58 transgenic mouse. Proc. Natl Acad. Sci. USA 90, 5578–5582. Martin, T. J., and Ng, K. W. (1994). Mechanisms by which cells of the osteoblast lineage control osteoclast formation and activity. J. Cell Biochem. 56, 357–366. Suda, T., Udagawa, N., Nakamura, I., Miyaura, C., and Takahashi, N. (1995). Modulation of osteoclast differentiation by local factors. Bone 17, 87S-91S. Felix, R., Cecchini, M. G., and Fleisch, H. (1990). Macrophage colony stimulating factor restores in vivo bone resorption in the op/op osteopetrotic mouse. Endocrinology 127, 2592–2594. Yoshida, H., Hayashi, S., Kunisada, T., Ogawa, M., Nishikawa, S., Okamura, H., Sudo, T., and Shultz, L. D. (1990). The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345, 442–444. Takahashi, N., Udagawa, N., Akatsu, T., Tanaka, H., Shionome, M., and Suda, T. (1991). Role of colony-stimulating factors in osteoclast development. J. Bone Miner. Res. 6, 977–985. Hattersley, G., Dorey, E., Horton, M. A., and Chambers, T. J. (1988). Human macrophage colony-stimulating factor inhibits bone resorption by osteoclasts disaggregated from rat bone. J. Cell Physiol. 137, 199–203. Hofstetter, W., Wetterwald, A., Cecchini, M. C., Felix, R., Fleisch, H., and Mueller, C. (1992). Detection of transcripts for the receptor for macrophage colony-stimulating factor, c-fms, in murine osteoclasts. Proc. Natl Acad. Sci. USA 89, 9637–9641. Lorenzo, J. A., Sousa, S. L., Fonseca, J. M., Hock, J. M., and Medlock, E. S. (1987). Colony-stimulating factors regulate the development of multinucleated osteoclasts from recently replicated cells in vitro. J. Clin. Invest. 80, 160–164. Simonet, W. S., Lacey, D. L., Dunstan, C. R., Kelley, M., Chang, M. S., Luthy, R., Nguyen, H. Q., Wooden, S., Bennett, L., Boone, T., Shimamoto, G., DeRose, M., Elliott, R., Colombero, A., Tan, H. L.,
269
Chapter 15 Signaling in Bone
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
Trail, G., Sullivan, J., Davy, E., Bucay, N., Renshaw-Gegg, L., Hughes, T. M., Hill, D., Pattison, W., Campbell, P., Sander, S., Van, G., Tarpley, J., Derby, P., Lee, R., and Boyle, W. J. (1997). Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89, 309–319. Tsuda, E., Goto, M., Mochizuki, S., Yano, K., Kobayashi, F., Morinaga, T., and Higashio, K. (1997). Isolation of a novel cytokine from human fibroblasts that specifically inhibits osteoclastogenesis. Biochem. Biophys. Res. Commun. 234, 137–142. Lacey, D. L., Timms, E., Tan, H. L., Kelley, M. J., Dunstan, C. R., Burgess, T., Elliott, R., Colombero, A., Elliott, G., Scully, S., Hsu, H., Sullivan, J., Hawkins, N., Davy, E., Capparelli, C., Eli, A., Qian, Y. X., Kaufman, S., Sarosi, I., Shalhoub, V., Senaldi, G., Guo, J., Delaney, J., and Boyle, W. J. (1998). Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93, 165–176. Yasuda, H., Shima, N., Nakagawa, N., Yamaguchi, K., Kinosaki, M., Mochizuki, S., Tomoyasu, A., Yano, K., Goto, M., Murakami, A., Tsuda, E., Morinaga, T., Higashio, K., Udagawa, N., Takahashi, N., and Suda, T. (1998). Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc. Natl Acad. Sci. USA 95, 3597–3602. Bucay, N., Sarosi, I., Dunstan, C. R., Morony, S., Tarpley, J., Capparelli, C., Scully, S., Tan, H. L., Xu, W., Lacey, D. L., Boyle, W. J., and Simonet, W. S. (1998). Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification. Genes Dev. 12, 1260–1268. Kong, Y. Y., Feige, U., Sarosi, I., Bolon, B., Tafuri, A., Morony, S., Capparelli, C., Li, J., Elliott, R., McCabe, S., Wong, T., Campagnuolo, G., Moran, E., Bogoch, E. R., Van, G., Nguyen, L. T., Ohashi, P. S., Lacey, D. L., Fish, E., Boyle, W. J., and Penninger, J. M. (1999). Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 402, 304–309. Dougall, W. C., Glaccum, M., Charrier, K., Rohrbach, K., Brasel, K., De Smedt, T., Daro, E., Smith, J., Tometsko, M. E., Maliszewski, C. R., Armstrong, A., Shen, V., Bain, S., Cosman, D., Anderson, D., Morrissey, P. J., Peschon, J. J., and Schuh, J. (1999). RANK is essential for osteoclast and lymph node development. Genes Dev 13, 2412–2424. Udagawa, N., Takahashi, N., Yasuda, H., Mizuno, A., Itoh, K., Ueno, Y., Shinki, T., Gillespie, M. T., Martin, T. J., Higashio, K., and Suda, T. (2000). Osteoprotegerin produced by osteoblasts is an important regulator in osteoclast development and function. Endocrinology 141, 3478–3484. Wong, B. R., Rho, J., Arron, J., Robinson, E., Orlinick, J., Chao, M., Kalachikov, S., Cayani, E., Bartlett, F. S., 3rd, Frankel, W. N., Lee, S. Y., and Choi, Y. (1997). TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J. Biol. Chem. 272, 25190–25194. Anderson, D. M., Maraskovsky, E., Billingsley, W. L., Dougall, W. C., Tometsko, M. E., Roux, E. R., Teepe, M. C., DuBose, R. F., Cosman, D., and Galibert, L. (1997). A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 390, 175–179. Wong, B. R., Josien, R., Lee, S. Y., Sauter, B., Li, H. L., Steinman, R. M., and Choi, Y. (1997). TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor. J. Exp. Med. 186, 2075–2080. Horwood, N. J., Elliott, J., Martin, T. J., and Gillespie, M. T. (1998). Osteotropic agents regulate the expression of osteoclast differentiation factor and osteoprotegerin in osteoblastic stromal cells. Endocrinology 139, 4743–4746. Jimi, E., Akiyama, S., Tsurukai, T., Okahashi, N., Kobayashi, K., Udagawa, N., Nishihara, T., Takahashi, N., and Suda, T. (1999). Osteoclast differentiation factor acts as a multifunctional regulator in
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
murine osteoclast differentiation and function. J. Immunol. 163, 434–442. Jimi, E., Nakamura, I., Amano, H., Taguchi, Y., Tsurukai, T., Tamura, M., Takahashi, N., and Suda, T. (1996). Osteoclast function is activated by osteoblastic cells through a mechanism involving cell-to-cell contact. Endocrinology 137, 2187–2190. Quinn, J. M., Elliott, J., Gillespie, M. T., and Martin, T. J. (1998). A combination of osteoclast differentiation factor and macrophagecolony stimulating factor is sufficient for both human and mouse osteoclast formation in vitro. Endocrinology 139, 4424–4427. Matsuzaki, K., Udagawa, N., Takahashi, N., Yamaguchi, K., Yasuda, H., Shima, N., Morinaga, T., Toyama, Y., Yabe, Y., Higashio, K., and Suda, T. (1998). Osteoclast differentiation factor (ODF) induces osteoclast-like cell formation in human peripheral blood mononuclear cell cultures. Biochem. Biophys. Res. Commun. 246, 199–204. Thomas, R. J., Guise, T. A., Yin, J. J., Elliott, J., Horwood, N. J., Martin, T. J., and Gillespie, M. T. (1999). Breast cancer cells interact with osteoblasts to support osteoclast formation. Endocrinology 140, 4451–4458. Lai, F. P., Cole-Sinclair, M., Cheng, W. J., Quinn, J. M., Gillespie, M. T., Sentry, J. W., and Schneider, H. G. (2004). Myeloma cells can directly contribute to the pool of RANKL in bone bypassing the classic stromal and osteoblast pathway of osteoclast stimulation. Br. J. Haematol. 126, 192–201. Eghbali-Fatourechi, G., Khosla, S., Sanyal, A., Boyle, W. J., Lacey, D. L., and Riggs, B. L. (2003). Role of RANK ligand in mediating increased bone resorption in early postmenopausal women. J. Clin. Invest. 111, 1221–1230. Darnay, B. G., Ni, J., Moore, P. A., and Aggarwal, B. B. (1999). Activation of NF-kappaB by RANK requires tumor necrosis factor receptor-associated factor (TRAF) 6 and NF-kappaB-inducing kinase. Identification of a novel TRAF6 interaction motif. J. Biol. Chem. 274, 7724–7731. Galibert, L., Tometsko, M. E., Anderson, D. M., Cosman, D., and Dougall, W. C. (1998). The involvement of multiple tumor necrosis factor receptor (TNFR)-associated factors in the signaling mechanisms of receptor activator of NF-kappaB, a member of the TNFR superfamily. J. Biol. Chem. 273, 34120–34127. Franzoso, G., Carlson, L., Xing, L., Poljak, L., Shores, E. W., Brown, K. D., Leonardi, A., Tran, T., Boyce, B. F., and Siebenlist, U. (1997). Requirement for NF-kappaB in osteoclast and B-cell development. Genes Dev. 11, 3482–3496. Naito, A., Azuma, S., Tanaka, S., Miyazaki, T., Takaki, S., Takatsu, K., Nakao, K., Nakamura, K., Katsuki, M., Yamamoto, T., and Inoue, J. (1999). Severe osteopetrosis, defective interleukin-1 signalling and lymph node organogenesis in TRAF6-deficient mice. Genes Cells 4, 353–362. Lomaga, M. A., Yeh, W. C., Sarosi, I., Duncan, G. S., Furlonger, C., Ho, A., Morony, S., Capparelli, C., Van, G., Kaufman, S., van der Heiden, A., Itie, A., Wakeham, A., Khoo, W., Sasaki, T., Cao, Z., Penninger, J. M., Paige, C. J., Lacey, D. L., Dunstan, C. R., Boyle, W. J., Goeddel, D. V., and Mak, T. W. (1999). TRAF6 deficiency results in osteopetrosis and defective interleukin-1, CD40, and LPS signaling. Genes Dev 13, 1015–1024. Takayanagi, H., Kim, S., Matsuo, K., Suzuki, H., Suzuki, T., Sato, K., Yokochi, T., Oda, H., Nakamura, K., Ida, N., Wagner, E. F., and Taniguchi, T. (2002). RANKL maintains bone homeostasis through c-Fos-dependent induction of interferon-beta. Nature 416, 744–749. Grigoriadis, A. E., Wang, Z. Q., Cecchini, M. G., Hofstetter, W., Felix, R., Fleisch, H. A., and Wagner, E. F. (1994). c-Fos: a key regulator of osteoclast-macrophage lineage determination and bone remodeling. Science 266, 443–448. Takayanagi, H., Kim, S., Koga, T., Nishina, H., Isshiki, M., Yoshida, H., Saiura, A., Isobe, M., Yokochi, T., Inoue, J., Wagner, E. F., Mak, T. W., Kodama, T., and Taniguchi, T. (2002). Induction and activation of the
270
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
transcription factor NFATc1 (NFAT2) integrate RANKL signaling in terminal differentiation of osteoclasts. Dev. Cell 3, 889–901. Erlebacher, A., Filvaroff, E. H., Ye, J. Q., and Derynck, R. (1998). Osteoblastic responses to TGF-beta during bone remodeling. Mol. Biol. Cell 9, 1903–1918. Rodan, G. A. (1991). Mechanical loading, estrogen deficiency, and the coupling of bone formation to bone resorption. J. Bone Miner. Res. 6, 527–530. Chambers, T. J., and Fuller, K. (1985). Bone cells predispose bone surfaces to resorption by exposure of mineral to osteoclastic contact. J. Cell Sci. 76, 155–165. Partridge, N. C., Jeffrey, J. J., Ehlich, L. S., Teitelbaum, S. L., Fliszar, C., Welgus, H. G., and Kahn, A. J. (1987). Hormonal regulation of the production of collagenase and a collagenase inhibitor activity by rat osteogenic sarcoma cells. Endocrinology 120, 1956–1962. Chiusaroli, R., Maier, A., Knight, M. C., Byrne, M., Calvi, L. M., Baron, R., Krane, S. M., and Schipani, E. (2003). Collagenase cleavage of type I collagen is essential for both basal and parathyroid hormone (PTH)/PTH-related peptide receptor-induced osteoclast activation and has differential effects on discrete bone compartments. Endocrinology 144, 4106–4116. Everts, V., Delaisse, J. M., Korper, W., Jansen, D. C., TigchelaarGutter, W., Saftig, P., and Beertsen, W. (2002). The bone lining cell: its role in cleaning Howship’s lacunae and initiating bone formation. J. Bone Miner. Res. 17, 77–90. Hauschka, P. V., Mavrakos, A. E., Iafrati, M. D., Doleman, S. E., and Klagsbrun, M. (1986). Growth factors in bone matrix. Isolation of multiple types by affinity chromatography on heparin-Sepharose. J. Biol. Chem. 261, 12665–12674. Schmid, C., and Ernst, M. (1992). Insulin-like growth factors. In “Cytokines and Bone Metabolism”, pp. 229–265. CRC Press, Boca Raton, FL. Rodan, S. B., and Rodan, G. A. (1992). Fibroblast growth factor and platelet-derived growth factor. In “Cytokines and Bone Metabolism”, pp. 115–145. CRC Press, Boca Raton, FL. Centrella, M., McCarthy, T. L., and Canalis, E. (1991). Transforming growth factor-beta and remodeling of bone. J. Bone Joint Surg. Am. 73, 1418–1428. Ogawa, Y., Schmidt, D. K., Dasch, J. R., Chang, R. J., and Glaser, C. B. (1992). Purification and characterization of transforming growth factor-beta 2.3 and -beta 1.2 heterodimers from bovine bone. J. Biol. Chem. 267, 2325–2328. Harris, S. E., Sabatini, M., Harris, M. A., Feng, J. Q., Wozney, J., and Mundy, G. R. (1994). Expression of bone morphogenetic protein messenger RNA in prolonged cultures of fetal rat calvarial cells. J. Bone Miner. Res. 9, 389–394. Vukicevic, S., Latin, V., Chen, P., Batorsky, R., Reddi, A. H., and Sampath, T. K. (1994). Localization of osteogenic protein-1 (bone morphogenetic protein-7) during human embryonic development: high affinity binding to basement membranes. Biochem. Biophys. Res. Commun. 198, 693–700. Vukicevic, S., Paralkar, V. M., Cunningham, N. S., Gutkind, J. S., and Reddi, A. H. (1990). Autoradiographic localization of osteogenin binding sites in cartilage and bone during rat embryonic development. Dev. Biol. 140, 209–214. Rydziel, S., Ladd, C., McCarthy, T. L., Centrella, M., and Canalis, E. (1992). Determination and expression of platelet-derived growth factor-AA in bone cell cultures. Endocrinology 130, 1916–1922. McCarthy, T. L., Centrella, M., Raisz, L. G., and Canalis, E. (1991). Prostaglandin E2 stimulates insulin-like growth factor I synthesis in osteoblast-enriched cultures from fetal rat bone. Endocrinology 128, 2895–2900. McCarthy, T. L., Centrella, M., and Canalis, E. (1989). Parathyroid hormone enhances the transcript and polypeptide levels of insulin-like
T. JOHN MARTIN AND NATALIE A. SIMS
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
growth factor I in osteoblast-enriched cultures from fetal rat bone. Endocrinology 124, 1247–1253. Turner, R. T., Backup, P., Sherman, P. J., Hill, E., Evans, G. L., and Spelsberg, T. C. (1992). Mechanism of action of estrogen on intramembranous bone formation: regulation of osteoblast differentiation and activity. Endocrinology 131, 883–889. Shinar, D. M., Endo, N., Halperin, D., Rodan, G. A., and Weinreb, M. (1993). Differential expression of insulin-like growth factor-I (IGF-I) and IGF-II messenger ribonucleic acid in growing rat bone. Endocrinology 132, 1158–1167. Suva, L. J., Seedor, J. G., Endo, N., Quartuccio, H. A., Thompson, D. D., Bab, I., and Rodan, G. A. (1993). Pattern of gene expression following rat tibial marrow ablation. J. Bone Miner. Res. 8, 379–388. Centrella, M., Horowitz, M. C., Wozney, J. M., and McCarthy, T. L. (1994). Transforming growth factor-beta gene family members and bone. Endocr. Rev. 15, 27–39. Nakashima, M., Nagasawa, H., Yamada, Y., and Reddi, A. H. (1994). Regulatory role of transforming growth factor-beta, bone morphogenetic protein-2, and protein-4 on gene expression of extracellular matrix proteins and differentiation of dental pulp cells. Dev. Biol. 162, 18–28. Heikinheimo, K. (1994). Stage-specific expression of decapentaplegicVg-related genes 2, 4, and 6 (bone morphogenetic proteins 2, 4, and 6) during human tooth morphogenesis. J. Dent. Res. 73, 590–597. Globus, R. K., Plouet, J., and Gospodarowicz, D. (1989). Cultured bovine bone cells synthesize basic fibroblast growth factor and store it in their extracellular matrix. Endocrinology 124, 1539–1547. Somjen, D., Binderman, I., Berger, E., and Harell, A. (1980). Bone remodeling induced by physical stress is prostaglandin E2 mediated. Biochim. Biophys. Acta. 627, 91–100. Raisz, L. G., and Martin, T. J. (1984). Prostaglandins in bone and mineral metabolism. In “Bone and Mineral Research” (W. A. Pecked, ed.), pp. 286–310. Elsevier, Amsterdam. Spencer, E. M., Liu, C. C., Si, E. C., and Howard, G. A. (1991). In vivo actions of insulin-like growth factor-I (IGF-I) on bone formation and resorption in rats. Bone 12, 21–26. Tobias, J. H., Chow, J. W., and Chambers, T. J. (1992). Opposite effects of insulin-like growth factor-I on the formation of trabecular and cortical bone in adult female rats. Endocrinology 131, 2387–2392. Bagi, C. M., Brommage, R., Deleon, L., Adams, S., Rosen, D., and Sommer, A. (1994). Benefit of systemically administered rhIGF-I and rhIGF-I/IGFBP-3 on cancellous bone in ovariectomized rats. J. Bone Miner. Res. 9, 1301–1312. Critchlow, M. A., Bland, Y. S., and Ashhurst, D. E. (1994). The effects of age on the response of rabbit periosteal osteoprogenitor cells to exogenous transforming growth factor-beta 2. J. Cell Sci. 107 (Pt 2), 499–516. Joyce, M. E., Roberts, A. B., Sporn, M. B., and Bolander, M. E. (1990). Transforming growth factor-beta and the initiation of chondrogenesis and osteogenesis in the rat femur. J. Cell Biol. 110, 2195–2207. Kawaguchi, H., Kurokawa, T., Hanada, K., Hiyama, Y., Tamura, M., Ogata, E., and Matsumoto, T. (1994). Stimulation of fracture repair by recombinant human basic fibroblast growth factor in normal and streptozotocin-diabetic rats. Endocrinology 135, 774–781. Marks, S. C., Jr., and Miller, S. (1988). Local infusion of prostaglandin E1 stimulates mandibular bone formation in vivo. J. Oral Pathol. 17, 500–505. Lund, J. E. (1982). The toxicology of PGE1 and PGE2. In “Prostaglandins in Clinical Medicine”, pp. 93–109. Yearbook Medical Publishers, New York. Bonewald, L. F., Wakefield, L., Oreffo, R. O., Escobedo, A., Twardzik, D. R., and Mundy, G. R. (1991). Latent forms of transforming
271
Chapter 15 Signaling in Bone
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110. 111.
growth factor-beta (TGF beta) derived from bone cultures: identification of a naturally occurring 100-kDa complex with similarity to recombinant latent TGF beta. Mol. Endocrinol. 5, 741–751. Oreffo, R. O., Mundy, G. R., Seyedin, S. M., and Bonewald, L. F. (1989). Activation of the bone-derived latent TGF beta complex by isolated osteoclasts. Biochem. Biophys. Res. Commun. 158, 817–823. Underwood, L. E., Thissen, J. P., Lemozy, S., Ketelslegers, J. M., and Clemmons, D. R. (1994). Hormonal and nutritional regulation of IGF-I and its binding proteins. Horm. Res. 42, 145–151. Martin, T. J., Allan, E. H., and Fukumoto, S. (1993). The plasminogen activator and inhibitor system in bone remodeling. Growth Regul. 3, 209–214. Campbell, P. G., Novak, J. F., Yanosick, T. B., and McMaster, J. H. (1992). Involvement of the plasmin system in dissociation of the insulin-like growth factor-binding protein complex. Endocrinology 130, 1401–1412. Gray, C., Boyde, A., and Jones, S. J. (1996). Topographically induced bone formation in vitro: implications for bone implants and bone grafts. Bone 18, 115–123. Nakamura, M., Udagawa, N., Matsuura, S., Mogi, M., Nakamura, H., Horiuchi, H., Saito, N., Hiraoka, B. Y., Kobayashi, Y., Takaoka, K., Ozawa, H., Miyazawa, H., and Takahashi, N. (2003). Osteoprotegerin regulates bone formation through a coupling mechanism with bone resorption. Endocrinology 144, 5441–5449. Cleiren, E., Benichou, O., Van Hul, E., Gram, J., Bollerslev, J., Singer, F. R., Beaverson, K., Aledo, A., Whyte, M. P., Yoneyama, T., deVernejoul, M. C., and Van Hul, W. (2001). Albers-Schonberg disease (autosomal dominant osteopetrosis, type II) results from mutations in the ClCN7 chloride channel gene. Hum. Mol. Genet. 10, 2861–2867. Marzia, M., Sims, N. A., Voit, S., Migliaccio, S., Taranta, A., Bernardini, S., Faraggiana, T., Yoneda, T., Mundy, G. R., Boyce, B. F., Baron, R., and Teti, A. (2000). Decreased c-Src expression enhances osteoblast differentiation and bone formation. J. Cell Biol. 151, 311–320. Karsdal, M. A., Henriksen, K., Sorensen, M. G., Gram, J., Schaller, S., Dziegiel, M. H., Heegard, A. M., Christophersen, P., Martin, T. J., Christiansen, C., and Bollerslev, J. (2005). Acidification of the osteoclast resorption compartment provides insight into the coupling of bone formation to bone resorption. Am. J. Pathol. 166, 467–476. Chiusaroli, R., Knobler, H., Luxenburg, C., Sanjay, A., Granot-Attas, S., Tiran, Z., Miyazaki, T., Harmelin, A., Baron, R., and Elson, A. (2004). Tyrosine phosphatase epsilon is a positive regulator of osteoclast function in vitro and in vitro. Mol. Biol. Cell 15, 234–244. Heymann, D., and Rousselle, A. V. (2000). gp130 Cytokine family and bone cells. Cytokine 12, 1455–1468. Sims, N. A., Jenkins, B. J., Quinn, J. M., Nakamura, A., Glatt, M., Gillespie, M. T., Ernst, M., and Martin, T. J. (2004). Glycoprotein 130
112.
113.
114. 115.
116.
117.
118. 119.
120.
121.
122.
123.
regulates bone turnover and bone size by distinct downstream signaling pathways. J. Clin. Invest. 113, 379–389. Hoff, A. O., Catala-Lehnen, P., Thomas, P. M., Priemel, M., Rueger, J. M., Nasonkin, I., Bradley, A., Hughes, M. R., Ordonez, N., Cote, G. J., Amling, M., and Gagel, R. F. (2002). Increased bone mass is an unexpected phenotype associated with deletion of the calcitonin gene. J. Clin. Invest. 110, 1849–1857. Dacquin, R., Davey, R. A., Laplace, C., Levasseur, R., Morris, H. A., Goldring, S. R., Gebre-Medhin, S., Galson, D. L., Zajac, J. D., and Karsenty, G. (2004). Amylin inhibits bone resorption while the calcitonin receptor controls bone formation in vitro. J. Cell Biol. 164, 509–514. Sexton, P. M., Findlay, D. M., and Martin, T. J. (1999). Calcitonin. Curr. Med. Chem. 6, 1067–1093. Demiralp, B., Chen, H. L., Koh, A. J., Keller, E. T., and McCauley, L. K. (2002). Anabolic actions of parathyroid hormone during bone growth are dependent on c-fos. Endocrinology 143, 4038–4047. Black, D. M., Greenspan, S. L., Ensrud, K. E., Palermo, L., McGowan, J. A., Lang, T. F., Garnero, P., Bouxsein, M. L., Bilezikian, J. P., and Rosen, C. J. (2003). The effects of parathyroid hormone and alendronate alone or in combination in postmenopausal osteoporosis. N. Engl. J. Med. 349, 1207–1215. Finkelstein, J. S., Hayes, A., Hunzelman, J. L., Wyland, J. J., Lee, H., and Neer, R. M. (2003). The effects of parathyroid hormone, alendronate, or both in men with osteoporosis. N. Engl. J. Med. 349, 1216–1226. Khosla, S. (2003). Parathyroid hormone plus alendronate—a combination that does not add up. N. Engl. J. Med. 349, 1277–1279. Martin, T. J. (2004). Does bone resorption inhibition affect the anabolic response to parathyroid hormone? Trends Endocrinol. Metab. 15, 49–50. Onyia, J. E., Bidwell, J., Herring, J., Hulman, J., and Hock, J. M. (1995). In vitro, human parathyroid hormone fragment (hPTH 1–34) transiently stimulates immediate early response gene expression, but not proliferation, in trabecular bone cells of young rats. Bone 17, 479–484. Ma, Y. L., Cain, R. L., Halladay, D. L., Yang, X., Zeng, Q., Miles, R. R., Chandrasekhar, S., Martin, T. J., and Onyia, J. E. (2001). Catabolic effects of continuous human PTH (1–38) in vitro is associated with sustained stimulation of RANKL and inhibition of osteoprotegerin and gene-associated bone formation. Endocrinology 142, 4047– 4054. Holtrop, M. E., King, G. J., Cox, K. A., and Reit, B. (1979). Time-related changes in the ultrastructure of osteoclasts after injection of parathyroid hormone in young rats. Calcif. Tissue Int. 27, 129–135. Martin, T. J., and Sims, N. A. (2005). On the coupling of bone formation to resorption. Trends Molecular Med. 11, 76–81.
This Page Intentionally Left Blank
Chapter 16
Parathyroid Hormone: Structure, Function and Dynamic Actions Lorraine A. Fitzpatrick John P. Bilezikian
I. II. III. IV. V. VI. VII.
Global Development, Amgen, Thousand Oaks, CA Departments of Medicine and Pharmacology, College of Physicians and Surgeons, Columbia University, New York, NY
VIII. Structure of the PTH/PTHrP (PTH1R) Receptor IX. Activation of the Cyclic Adenosine Monophosphate Second-Messenger System by Parathyroid Hormone X. Identification of a Second PTH Receptor XI. Physiological Actions of PTH XII. Cell-to-Cell Communication: Osteoblasts and Osteoclasts XIII. Preferential Actions of PTH at Selected Skeletal Sites References
Introduction Structure of the PTH Gene Chromosome Location Control of Gene Expression Biosynthesis of Parathyroid Hormone Metabolism of Parathyroid Hormone Receptor Interactions of Parathyroid Hormone and Parathyroid Hormone-Related Protein
I. INTRODUCTION
1,25-dihydroxy vitamin D and enhances the intestinal absorption of calcium. These actions of PTH are mediated through a G protein-coupled receptor system in the cells of target tissues. A rise in extracellular calcium inhibits further secretion of parathyroid hormone.
Parathyroid hormone (PTH) is an 84 amino acid hormone that regulates calcium homeostasis via its actions on target tissues. PTH maintains serum calcium concentrations within a narrow physiological range by direct actions on bone and kidney tissue and an indirect action, via 1,25-dihydroxy vitamin D, on the intestinal tract. PTH release and gene expression, in turn, are regulated by serum calcium concentrations. Hypocalcemia stimulates the release of PTH from the parathyroid gland. By reabsorption of calcium in the distal convoluted tubule of the kidney or by osteoclast-mediated bone resorption, serum calcium concentrations increase. PTH also stimulates renal 1-α-hydroxylase activity in the kidney which increases Dynamics of Bone and Cartilage Metabolism
II. STRUCTURE OF THE PTH GENE The gene that encodes PTH is representative of typical eukaryotic genes with consensus sequences for initiation of RNA synthesis, RNA splicing and polyadenylation. Restriction enzyme analysis of the human PTH gene has indicated polymorphism in the cleavage products in different individuals that are useful for genetic analysis [1]. 273
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
274
Lorraine A. Fitzpatrick and John P. Bilezikian
Complementary DNAs encoding for human, rat, chicken, bovine and dog PTH have been cloned. These genes have several structural features in common: all contain two introns or intervening sequences and three exons (Fig. 1) [2]. The introns vary in size among species. The first intron is large; the second intron is smaller (about 100 base pairs) in human, bovine, and rat genes. The first intron encodes the 5′ noncoding sequences; the second exon encodes most of the “prepro” sequence of prepro parathyroid hormone. The mature PTH sequence and the 3′ noncoding region are contained within the third exon. RNA is transcribed from the sequences in both the introns and exons. Subsequently, RNA sequences derived from the introns are spliced. The mature PTH mRNA is then translated into prepro PTH. Considerable homology among the mammalian PTH genes is seen between human and bovine proteins (85%) and between human and rat proteins (75%). Identity is less in the 3′ noncoding region. In contrast, human and bovine PTH genes contain two functional TATA transcription
Figure 1
start sites where the rat and chicken genes have only one. Both initiation sites of transcription are utilized in the bovine and human genes. Tissue specificity of PTH gene expression is determined by coding regions upstream of the structural gene. Analysis of these regulatory sequences has been difficult due to the lack of a well-defined cell line for gene expression. A cAMP-response element has been identified and other sequences that bind 1,25(OH)2D3 receptors have been recognized [3]. PTH gene expression is regulated transcriptionally by the same mechanism, in part, by calcium and phosphate [4, 5].
III. CHROMOSOME LOCATION The human PTH gene has been localized to the short arm of chromosome 11 and is close to the calcitonin gene [6, 7]. Additional localization studies have indicated that the human PTH gene is located at band 11p15 [8].
Primary structure of human preparathyroid hormone. The arrows indicate sites of specific cleavages which occur in the sequence of biosynthesis and peripheral metabolism. The biologically active sequence is noted. Reprinted from Choren and Rosenblatt [2a], with permission.
Chapter 16 Parathyroid Hormone: Structure, Function and Dynamic Actions
Restriction fragment length polymorphisms indicate that the human PTH gene is linked to genes encoding calcitonin, catalase, insulin, H-ras, and β-globin [1]. The development of mice homozygous for a null PTH allele assisted in assessing the role of PTH in the modulation of skeletal development in the fetus. Mice carrying the disrupted PTH gene (PTH−/− mice) were derived by homologous recombination in embryonic stem cells [9]. Parathyroid glands were greatly enlarged and PTH expression was absent in these mice, compared to wild-type littermates. In the PTH−/− mice, the skull was abnormal and mineralization of skull bones was enhanced. The vertebral column had smaller vertebral bodies and mineralized metacarpal and metatarsal bones were shorter. Lengths of the tibae were normal and the hypertrophic cartilage zone was enlarged with a higher ratio of the hypertrophic zone to proliferating zone. However, proliferation of chondrocytes was not altered, and there was enhancement of the deposition of type X collagen in the hypertrophic zone. Cortical thickness of long bones was increased, but the trabecular components were diminished. Reduced osteoblast numbers were present in the primary spongiosa, with increased apoptosis. PTH appeared to be essential for normal cartilage remodeling, and the reduced osteoblasts resulted in reduced trabecular bone volume. When PTH−/− mice were placed on a low-calcium diet, renal 1-α-hydyroxylase expression increased in spite of the lack of PTH. There was a rise in 1,25(OH)2D3 levels, marked osteoclastogenesis, and profound bone resorption. These latter studies indicate that in the complete absence of PTH, calcium stores can be mobilized through the actions of 1,25(OH)2D3 to guard against extreme hypocalcemia [10]. Clinical studies have attempted to link gene polymorphisms in the PTH gene to specific human disease states. No linkages were found in bone phenotypes in Chinese subjects [11], or in sporadic idiopathic hypoparathyroidism [12]. PTH gene polymorphism accounted for about 7–9% of the total variances of bone dimensional variables in a group of 91 healthy Caucasian women. Bone diameter, cortical thickness, and cross-sectional area at standard sites of the metacarpals, radius, and femur were measured with radiogrammetry. Higher metacarpal diameter and cross-sectional cortical area and a slower decrease in radial cortical area with age were associated with the absence of the BstBI restriction site of the PTH gene [13]. In secondary hyperparathyroidism associated with renal hemodialysis, PTH genotypes were determined by PCR and by restriction fragment length polymorphisms (RFLPs) of BstBI and DraII. There were no differences in these genotypes between hemodialysis patients and healthy controls. There was a significant difference in the serum intact PTH levels between Bb/bb and BB genotypes
275
(P < 0.02) [14]. In another study, PTH genotypes were determined by PCR and BstBI or DRAII RFLPs were determined in patients with primary hyperparathyroidism. The BstBI polymorphism of the PTH gene was closely linked to bone mineral density, but not the development of primary hyperparathyroidism. However, serum calcium was higher in those with the BB genotype, compared to the BB/bb genotypes, suggesting that the severity of primary hyperparathyroidism may be linked to this genotype [15]. A novel mutation in the signal peptide of the prepro PTH gene was associated with autosomal-recessive familialisolated hypoparathyroidism in a child of consanguineous parents who presented with neonatal hypocalcemia seizures. Replacement of thymine with a cytosine was found in the first nucleotide of position 23 in the 25-amino acid signal peptide, resulting in the replacement of the normal Ser (TCG) with a Pro (CCG). The parents were heterozygous for the allele [16].
IV. CONTROL OF GENE EXPRESSION A. The Role of Extracellular Calcium Minute-to-minute regulation of serum calcium levels is dependent on the regulation of secretion of PTH by calcium. Long-term homeostasis is dependent on several additional regulatory mechanisms. Expression of the PTH gene occurs almost exclusively in the parathyroid chief cells. Early studies indicated that increased levels of extracellular calcium resulted in a reversible fall in PTH mRNA in cultured parathyroid cells [17]. Suppression was most pronounced after 72 hours of incubation. The rate of PTH gene transcription was also reduced in the presence of high concentrations of calcium. In vivo studies in intact rats revealed rapid changes in PTH mRNA in response to changes in serum calcium concentrations [18]. The largest effects were noted in response to lowered levels of calcium (2.6–2.1 mmol/liter). Increased levels of serum calcium, whether induced by infusion of calcium or by transplantation of Walker carcinosarcoma 256 cells, had little effect on PTH mRNA levels. One major difference in the in vivo studies in comparison to the in vitro work is in the time course of response to calcium. In vivo, high calcium levels lead to a rapid fall in PTH mRNA within a few hours; in vitro, a longer time frame is required to detect changes in PTH mRNA in the models tested. Another major difference is that in vivo, hypercalcemia does not alter PTH mRNA levels, suggesting that at the molecular level the parathyroid gland is responsive to the signal of hypocalcemia.
276 The parathyroid cell senses changes in extracellular calcium through a calcium-sensing receptor (CaR). This receptor is a membrane-bound, seven transmembrane protein that has three major domains: an extracellular 612-amino acid ligand-binding segment, a hydrophobic 250-amino acid membrane-spanning section, and a cytosolic COOH-terminal tail of ~250 amino acids [19]. Activation of the CaR by extracellular calcium ions engages the mitogen-activating protein kinase C cascade through both G protein-linked phospholipase C and a member of the Src tyrosine kinase family [20, 21]. The pathway eventually leads to activation of phospholipase A2 and the generation of arachidonic acid [20]. Thus, PTH secretion is inhibited via the CaR when it is activated by elevated concentrations of ionized calcium. Prolonged hypercalcemia eventually leads to reduced parathyroid cell proliferation, an effect that is also likely to be mediated by the CaR [22]. Conversely, low serum levels of ionized calcium reduce CaR activity and PTH secretion increases. Many studies to evaluate the genetic mechanism of the action of calcium and the search for DNA sequences responsible for the transcriptional effects of calcium, have been hampered by the lack of a well-differentiated parathyroid cell line that produces PTH in response to changes in extracellular calcium. Several short sequences located several thousand base pairs upstream from the start site of PTH gene transcription may be involved in the regulation of PTH by calcium. A negative calcium regulatory element (nCaRE) has been identified in the PTH gene. It is also present in the atrial natriuretic peptide gene [23]. Further studies by the same investigators revealed the identity of a redox factor protein [1] that activates several transcription factors and binds nCaRE. The level of ref1 mRNA and protein were elevated by an increase in extracellular calcium concentration. Unfortunately, these studies had to be performed in a nonparathyroid cell line so the role of the protein in the regulation of PTH expression remains to be determined. The in vivo effects of calcium also may be due to posttranscriptional mechanisms. Preliminary evidence indicates that regions in the 3′ UTR of PTH mRNA binds specifically to cytosolic proteins and may be involved in the post-transcriptional increase in PTH gene expression induced by decreased calcium concentration [24]. Bovine parathyroid cells were incubated in the presence of low extracellular calcium (0.4 mM) and resulted in a post-transcriptional increase in the membrane-bound polysomal content of PTH mRNA [25]. Using a cell-free assay system, Vadher et al. [26] presented preliminary evidence that the 3′-untranslated region is important in the translational regulation of PTH synthesis by cytosolic
Lorraine A. Fitzpatrick and John P. Bilezikian
regulatory proteins. These in vitro studies add support for a post-translational role of calcium on PTH gene expression. Taken together, these studies suggest that posttranslational regulation of PTH mRNA may involve calcium-sensitive proteins binding to the 3′ UTR region. Additional work indicates that 26 nucleotide is the minimal sequence for protein binding in the PTH mRNA 3′ UTR. By using chimeric growth hormone PTH 63 nucleotide transcripts, the investigators were able to determine that the protein- binding region of the PTH mRNA 3′ UTR is necessary and sufficient to confer responsiveness to calcium and phosphate. This element determined PTH mRNA stability and levels [27].
B. The role of 1,25 dihydroxyvitamin D In vivo and in vitro studies support the significant role of 1,25 dihydroxyvitamin D in the regulation of PTH gene expression via a well-defined feedback loop between PTH and 1,25(OH)2D3. The renal synthesis of 1-α hydroxylase is increased by PTH; in turn, 1,25(OH)2D3 increases serum calcium concentrations by enhancing intestinal absorption of calcium. 1,25(OH)2D3 reduces transcription of the PTH gene. In bovine parathyroid cells in primary culture, 1,25(OH)2D3 exposure results in reduced levels of PTH mRNA and PTH gene transcription rates [28, 29]. In vitro studies in rats confirm the regulatory role of 1,25(OH)2D3 in PTH gene expression. Rats that are injected with 1,25(OH)2D3 at levels that do not alter serum calcium levels show decreased transcription of the PTH gene and reduced levels of PTH mRNA [30]. Transfection assays and DNA-binding studies have attempted to identify specifically the sequences responsible for modulation of the PTH gene transcription by 1,25(OH)2D3. Utilizing a rat pituitary cell line (GH4), Okazaki et al. linked 684 base pairs of the 5′-flanking region of the human PTH gene to a reporter gene. Transfection of the rat pituitary cell line resulted in responsiveness of gene expression to 1,25(OH)2D3 [31]. DNA binding studies identified DNA sequences upstream of the PTH gene that bind to 1,25(OH)2D3 receptors in vitro. A specific 26 base pair sequence located 125 base pairs upstream from the transcription start site on the PTH gene binds to 1,25(OH)2D3 receptors. Confirmation of this activity was performed by transfection of a pituitary cell line with the sequence linked to a reporter gene, with the result that 1,25(OH)2D3 suppressed the reporter gene [3]. In contrast, the human PTH DNA sequence does not mediate repression of transcription in the osteoblast cell line, ROS 17/2.8, in spite of the fact that vitamin D response elements (VDRE) are active in these cells. One difference is that
Chapter 16 Parathyroid Hormone: Structure, Function and Dynamic Actions
human PTH DNA sequences contain a single copy of a hexameric motif homologous to those repeated in the up-regulatory vitamin D response elements. When nuclear extracts from bovine parathyroid cells are incubated with the vitamin D receptor, the receptor binds to the downregulatory human PTH DNA sequence independent of the presence of the retinoid X receptor (RXR). In nuclear extracts from GH4CI pituitary cells, two vitamin D-containing complexes are present, one of which contains RXR. In nuclear extracts derived from the ROS 17/2.8 osteoblasts, a single VDR-dependent complex containing RXR is present. The negative 1,25(OH)2D3 response element contains one copy of a motif that is duplicated in the mouse osteopontin gene, a gene that is up-regulated by 1,25(OH)2D3. When the osteocalcin vitamin D response element is utilized as a probe, only VDR–RXR-containing complexes are generated from the nuclear extracts of all three cell types. These experiments indicate that transcriptional repression in response to 1,25(OH)2D3 differs from the up-regulatory response elements in sequence composition and in the ability to bind VDR independent of RXR [32]. Calreticulin is a calcium-binding protein present in the lumen of the endoplasmic reticulum of the cell. Using a model of chronic hypocalcemia, with increased PTH gene expression, calreticulin protein was increased in the nuclear fraction of parathyroids. In vitro, calreticulin blocked DNA binding of the VDR-RXRβ heterodimers to the cPTH VDRE. The studies suggest that calreticulin may prevent the transcriptional effect of 1,25 (OH)2D3 on the PTH promoter [33].
C. Phosphate as a Modulator of PTH Gene Expression PTH has a potent effect on the tubular reabsorption of phosphate by the kidney. This observation has been utilized as a clinical test to evaluate target organ responsiveness to parathyroid hormone. In severe renal failure, hyperphosphatemia results in secondary hyperparathyroidism. Clinically, the effects of the associated hypocalcemia and decrease in serum levels of 1,25(OH)2D3 have complicated our understanding of the role of phosphate on PTH secretion. Slatopolsky and Bricker [34] showed that in experimental renal failure, restriction of dietary phosphate prevented secondary hyperparathyroidism. Clinical studies confirmed the utility of restricting oral phosphate intake. These initial in vitro and in vivo studies suggested that phosphate directly alters that production of 1,25(OH)2D3. Recently, a series of elegant, carefully designed studies have shown that the effect of phosphate on PTH is independent
277
of serum calcium and 1,25(OH)2D3 levels. Uremic dogs placed on diets deficient in both calcium and phosphate have lower levels of serum phosphate and calcium but no change in the concentration of 1,25(OH)2D3 [35]. In hypophosphatemic rats, phosphate regulates the PTH gene independent of the effects of phosphate on calcium and 1,25(OH)2D3 [36]. These studies indicate that phosphate regulates PTH in a hypophosphatemic, normocalcemic model in the presence of normal levels of 1,25(OH)2D3. Nuclear transcript run-on assays show that the decrease in PTH mRNA due to low phosphate was post-transcriptional in contrast to the transcriptional effects of 1,25(OH)2D3 on the PTH gene. Recently, protein–mRNA binding studies in hypophosphatemic rats reveal decreased binding of PTH cytosolic proteins. Confirmation of the functional role was provided with an in vitro degradation assay. The functional life of intact transcripts was reduced from 40 minutes in parathyroid proteins obtained from control rats to 5 minutes in the proteins from hypophosphatemic animals. In contrast, parathyroid proteins from hypocalcemic rats had increased binding and intact transcripts were present for a longer period of time (180 minutes). The study also indicates that degradation of PTH mRNA is dependent on sequences in the 3′ UTR region [37]. In human parathyroid tissue, it appears that phosphate directly regulates PTH secretion and PTH mRNA levels [38]. An additional mechanism by which 1,25(OH)2D3 may regulate PTH gene expression is through alterations in the 1,25(OH)2D3 receptor. Modulation of the concentration of the 1,25(OH)2D3 receptor may alter the effect of 1,25(OH)2D3 at its target sites. The administration of 1,25(OH)2D3 to rats results in an increase in 1,25(OH)2D3 receptors in parathyroid tissue with a concomitant decrease in PTH mRNA [39]. In another study, weanling rats fed a calcium-deficient diet became markedly hypocalcemic with high serum levels of 1,25(OH)2D3. There was no change in 1,25(OH)2D3 receptor mRNA levels in spite of the high levels of 1,25(OH)2D3; furthermore, PTH mRNA levels were increased. The low serum calcium levels may prevent the increase in parathyroid 1,25(OH)2D3 receptor levels and suppression of PTH mRNA levels. These findings have been confirmed in an avian model [40] and in vitamin D-deficient rats [41]. The use of calcitriol analogs that are biologically active but result in less hypercalcemia also have helped to separate the effects of calcium from the actions of vitamin D on PTH gene expression. Detailed dose–response curves suggest that 1,25(OH)2D3 is the most effective of the vitamin D forms to inhibit PTH levels in vivo. The ability of 1,25(OH)2D3 to decrease PTH gene expression is a significant therapeutic management tool for patients at
278 risk of developing secondary hyperparathyroidism. The downside to the use of this agent is the possibility of raising the serum calcium level. The negative regulation of the PTH gene in response to 1,25(OH)2D3 is mediated by a 684 base pair region upstream of the transcription start site [42]. Another transcriptional regulator, Nuclear factor Y (NF-Y), is an abundant transcription factor and regulates the expression of about 30% of all eukaryotic genes. NF-Y binds to the extended VDRE region of the human PTH gene [43], resulting in enhanced transcriptional activity. Further enhancement is due to additional binding sites for NF-Y, establishing this factor as an important regulator of the expression of the PTH gene and a necessary ingredient for full transcriptional activity of the promoter. Disruption of the synergism between the NF-Y molecules bound to the PTH gene promoter region and modulation of the transcriptional activity of NF-Y by VDR and 1,25(OH)2D3 results in repression of the gene.
D. Other Regulators of PTH Gene Expression Other circulating factors besides calcium, 1,25(OH)2D3 and phosphate may modulate PTH gene transcription. A cAMP response element on the PTH gene suggests that hormones that stimulate adenylyl cyclase activity may increase PTH gene transcription. Another potential regulator is glucocorticoids that increase mRNA levels in dispersed, hyperplastic human parathyroid cells [44] and attenuate the decrease in PTH mRNA in response to 1,25(OH)2D3 in dispersed bovine parathyroid cells [45]. Other intracellular signaling mechanisms are capable of PTH gene regulation. Protein kinases A and C regulate PTH mRNA in vitro in dispersed bovine parathyroid cells [46, 47]. Inhibition of levels of protein kinase C by the addition of staurosporine or a phorbol ester results in decreased levels of PTH mRNA, stimulators of protein kinase A increased PTH mRNA levels. The physiological relevance of the role of intracellular second messengers in the minute-to-minute control of calcium homeostasis remains to be delineated. Estrogen is thought to play a role in the regulation of PTH gene expression due to the proposed roles of PTH in the pathophysiology of postmenopausal osteoporosis. In primary cultures of bovine parathyroid cells, estrogen increases the secretion of PTH. Rat parathyroid glands contain estrogen receptors for estrogen and in ovariectomized rats, estradiol administration increases PTH mRNA [48]. Progesterone has also been demonstrated to enhance secretion of PTH and synthesis of PTH mRNA in cell culture. The addition of 19-nor progestin R5020 to ovariectomized rats results in a two-fold increase in PTH
Lorraine A. Fitzpatrick and John P. Bilezikian
mRNA [49]. In the rat, PTH mRNA levels are increased in the proestrus and estrus phase, suggesting alteration of PTH gene expression in response to the rat estrus cycle.
V. BIOSYNTHESIS OF PARATHYROID HORMONE Mammalian PTH is an 84-amino acid polypeptide. The primary amino acid sequence is highly conserved among mammalian species, with strongest homology in the N-terminus. The greatest variation is in the mid-molecule between residues 39 and 52, and the C-terminus contains several stretches of homology [50]. Parathyroid hormone is synthesized in the parathyroid cell as a 115-amino acid single-chain polypeptide termed “preproparathyroid hormone”. The “pre” sequence is the “leader” or “signal” sequence and is identical to signal sequences associated with other polypeptide hormones. The “preproPTH” sequences are known for human, rat, bovine, pig and chicken. All have in common the 25-residue “pre” and 6-residue “pro” sequences. The “pre” sequence contains a hydrophobic stretch of amino acids preceded by a positively charged residue. The signal sequence is cleaved from the amino terminus during the process of synthesis. The signal sequence binds to a signal recognition particle (SRP), an intracellular RNA–protein complex that recognizes the signal sequence. The SRP binds to a receptor on the rough endoplasmic reticulum and directs preproPTH to a protein-lined channel, thereby “docking” the protein. A signal peptidase on the inner surface of the endoplasmic reticulum cleaves the signal sequence. The resultant peptide, “proparathyroid hormone” is left in the cisternae of the endoplasmic reticulum. The “pro” region is a highly positively charged hexapeptide and its cleavage is necessary to initiate biological activity of the hormone [51]. Packaged into vesicles, ProPTH progresses to the Golgi apparatus where the amino-terminal “pro” sequences are removed by the proprotein convertase furin [52]. PTH is concentrated into dense core secretory vesicles that fuse with the plasma membrane to release the contents in response to a fall in extracellular calcium concentration.
VI. METABOLISM OF PARATHYROID HORMONE Proteolysis of newly synthesized PTH is influenced by extracellular calcium [53, 54]. Most of the hormone secreted under conditions of hypercalcemia is already fragmented whereas, in the case of hypocalcemia, the
Chapter 16 Parathyroid Hormone: Structure, Function and Dynamic Actions
secreted molecule tends to be intact and active. Variable extracellular calcium levels, which are similar to mechanisms in proteolytic cleavage of the stored hormone, provides a dynamic regulatory mechanism. Intact PTH is rapidly cleared from the circulation with a half-life of approximately 2 minutes [55]. The liver plays the dominant role in PTH metabolism, removing 60–70% of the hormone. The kidney removes 20–30% by glomerular filtration. Uptake by other cells such as bone cells accounts for a minor component. Clearance by the liver is under the direction of the high-capacity, nonsaturable Kupffer cells where PTH undergoes extensive proteolysis. Some of the C-terminal fragments are released into the bloodstream to be removed ultimately by the kidneys [56]. PTH is not only rapidly cleared from the circulation, but it is also rapidly cleaved by endoproteases resulting in a series of carboxy-terminal fragments that are in the circulation. Carboxy-terminal fragments make up 50–90% of the total circulating PTH immunoactivity, in spite of the fact that, quantitatively, only 10–20% of secreted intact PTH is converted to circulating carboxy-terminal fragments. This is because clearance of these fragments is limited by glomerular filtration mechanisms in the kidney. In renal insufficiency, large amounts of carboxy-terminal fragments are found in the circulation.
VII. RECEPTOR INTERACTIONS OF PARATHYROID HORMONE AND PARATHYROID HORMONE-RELATED PROTEIN The discovery that PTH and PTHrP share a common receptor has generated additional information about the structure–function relationship of these two peptides. The gene for PTHrP is located on chromosome 12 but it is thought that PTH and PTHrP share a common ancestral gene [57]. There is strong homology in the first half of the active amino-terminal core of PTH and PTHrP in that eight of the first 13 amino acid residues are identical. Beyond these first 13 residues, the remainder of the first 34 amino acids differ in sequence, but topographically may fold in ways that are very similar to each other. In the nonhomologous 14–34 portions, substantial threedimensional structure is shared between the two peptides, helping to account for their ability to bind to a common receptor [58–60]. PTHrP is synthesized in adult and fetal tissue and mediates several paracrine/autocrine functions. The physicochemical properties of PTH receptors have been partially characterized with affinity cross-linking PTH ligands and solubilization of the receptor [61].
279
Biological properties of PTH and PTHrP have further defined the important residues in the ligand and allowed creation of analogs that can modify the activity of the receptor. PTH(1-34) and PTHrP(1-36) bind to the same receptor with equal affinity [62]. The human PTH receptor binds both PTH and PTHrP but binds to PTH preferentially [63]. Diversity exists in the receptor mRNAs of different lengths that encode the PTH receptor in several tissues and organs, suggesting that certain receptors may specifically bind specific configurations of the PTH molecule.
VIII. STRUCTURE OF THE PTH/PTHrP (PTH1R) RECEPTOR cDNAs encoding the PTH/PTHrP (also designated PTH1R) receptors predict that they would have topographical features in common with other G protein-coupled receptors. The PTH/PTHrP receptors have an aminoterminal extracellular sequence, three extracellular loops, three intracellular loops, and a carboxy-terminal cytoplasmic region and are approximately 590 amino acids in length. The hydrophilic regions are linked by seven relatively hydrophobic membrane-spanning helices (Fig. 2). PTH1Rs from different species are highly homologous; rat and mouse PTH1Rs are 99% identical at the amino acid level and human and rat PTH1Rs are 91% identical. The greatest amino acid differences are limited to discrete areas in the first extracellular loop and to the carboxyterminal sequences. Renal and bone receptors are identical within each species suggesting that the same receptor transcript is present in these two traditional target tissues [63]. The PTH1R is activated equivalently by intact and N-terminal PTH and PTHrP peptides and capable of coupling to several different G-proteins. This ligand–receptor interaction results in activation of multiple signaling pathways concurrently including cAMP, PLC, cytoplasmic calcium, and PKC. The intracellular tail of the receptor couples to a pertussis toxin-sensitive Gi protein that inhibits adenylyl cyclase [64]. The membrane-spanning helices of PTH1Rs are strongly conserved across species and serve a vital role in the signal transduction across the membrane. Studies that utilized mutant or chimeric receptors have generated information about the importance of several domains in the PTH/ PTHrP receptors responsible for ligand-binding, G-protein binding and receptor activation. The PTH1R is capable of stimulating multiple second-messenger pathways. Stably transfected LLC-PK1 porcine cells expressing rat or opposum PTH1Rs exhibited high-affinity binding of PTH, dose-dependent activation of cyclic AMP, and release
280 Primary sequence and typological structure of the human PTH2R. The seven putative transmembrane-spanning regions are indicated based on alignment with other members of the secretin/PTH receptor subfamily. Conserved residues with the opossum kidney PTH1R are indicated by the black circles. Reprinted from Turner et al. [54], with permission.
Figure 2
Chapter 16 Parathyroid Hormone: Structure, Function and Dynamic Actions
of intracellular calcium were evident. The EC50 of the cyclic AMP response was 20–50-fold lower than the intracellular calcium response. PTH also altered cell proliferation that was mimicked by the addition of forskolin, phorbol esters, calcium ionophores, and cAMP analogs. The effect on phosphate transport was only seen when a phorbol ester was added [65]. These data suggest that activation of multiple intracellular pathways is possible with a single ligand. Scanning mutagenetic techniques of the intracellular regions of the receptor have provided evidence for the roles of various intracellular regions critical for receptor activation, receptor–ligand binding and activation of various second messengers. For example, activation of phospholipase C requires one of the intracellular loops. Mutation of residues in the N-terminal region of the third intracellular loop of the opossum PTH1R and expression in COS-7 cells results in normal binding of ligand but impaired adenylyl cyclase and phospholipase C activation. These data suggest that the N-terminal region of the third intracellular loop plays a critical role in the coupling of Gs- and Gq-mediated second-messenger systems [66]. Further studies of the interactions between PTH and PTH1R have taken advantage of mutagenetic approaches. The interaction of PTH(1-34) with PTH1R involves highaffinity binding of the C terminus of the ligand with portions of the receptor’s extracellular N-terminal domain and the extracellular loops in the transmembrane domain. The N-terminus of the ligand interacts with the transmembrane domains to catalyze the G-protein activation(s) required for signal transduction [50]. The critical role in receptor activation of the juxtamembrane (“J”) domain is highlighted by the fact that several mutations in this area result in constitutive (i.e., ligand-independent) receptor activation (Jansen’s metaphyseal chondrodysplasia) [67]. One of the most interesting and provocative studies regarding structure–function relationships between receptors and ligands is evaluation of the cavity formed by the receptor into which the ligands fit. Specific amino acid residues are responsible for optimal ligand binding and signal transduction. The polypeptides secretin and parathyroid hormone were used as models of two ligands with no sequence homology and no cross reactivity with the other’s receptor. Mutation of a single amino acid in the second transmembrane domain of the secretin receptor to the corresponding amino acid in the PTH receptor resulted in a receptor that could bind to and transmit the intracellular signals affected by PTH. The reciprocal mutation in the PTH receptor conferred responsiveness to secretin in a similar manner. Thus, transmembrane residues on receptors recognize a specific ligand and restrict the access and activation of inappropriate ligands [68].
281
The PTH1R undergoes internalization after phosphorylation. β-arrestins may terminate receptor–G protein coupling and promote receptor endocytosis, although there may be differences in the recycling of the receptor depending on the cell type [50]. PTH1R internalization can be dissociated from receptor activation and, in some cells, N-truncated PTH peptide such as PTH (7-34) and PTH (7–84) may promote PTH1R endocytosis via a β-arrestin independent mechanism [69].
IX. ACTIVATION OF THE CYCLIC ADENOSINE MONOPHOSPHATE SECOND-MESSENGER SYSTEM BY PARATHYROID HORMONE Parathyroid hormone was one of the first hormones to be shown to utilize the cAMP second-messenger system. In the kidney, cAMP is involved in PTH-mediated effects to reduce calcium excretion, enhance tubular reabsorption of phosphate and to stimulate 1,25(OH)2D3 formation. Cyclic AMP is also believed to be the mediator of many of the actions of PTH on bone [70, 71]. Clinical evidence indicates the important role of cAMP in the mediation of the effects of PTH. Patients suffering from a hereditary syndrome of PTH resistance, pseudohypoparathyroidism type 1, do not respond to exogenously administered PTH with an increase in urinary cAMP [72]. The full-length 84-amino acid polypeptide is not required for activation of adenylyl cyclase. PTH(1-34) is as potent as the intact molecule. Carboxy-terminal fragments are inactive and progressive loss in adenylyl cyclase-stimulating activity occurs with the stepwise deletion of amino acids from the amino-terminal end of PTH(1-34) [73]. PTH(1-25) is also inactive, suggesting that the 25–34 positions are important for the binding of PTH to its receptor. Stepwise removal of the amino acids from the aminoterminal end of the molecule rapidly reduces the adenylyl cyclase-stimulating properties such that PTH(2–34) is a very weak stimulator. Further truncation leads to analogs that are competitive inhibitory, such as PTH(7-34). Activation of cAMP stimulates protein kinase A [74]. The osteoblast PTH receptor is also coupled to Gαq, leading to the activation of phospholipase C (PLC) and hydrolysis of phosphatidylinositol bisphosphonate (PIP2) to inositol triphosphate (IP3) and diacylglycerol (DAG). IP3 binds to specific receptors and mediates the release of calcium from intracellular stores [75, 76]. DAG is a critical cofactor for the activation of PKC isoenzymes [77]. Recent studies have indicated that PTH stimulates hydrolysis of phosphatidylcholine through activation of phospholipase D
282
Lorraine A. Fitzpatrick and John P. Bilezikian
in osteoblast-like cells [78]. Phospholipase D-mediated hydrolysis of phosphatidylcholine produces sustained generation of DAG and more prolonged activation of PKC. Thus, PTH stimulation of phospholipase D is an additional pathway that activates PKC isoenzymes in the parathyroid cell. PKC isoenzymes can regulate PTH-stimulated interleukin-6 (IL-6) promoter activity in osteoblastic cells, suggesting that PKC has an important role in bone remodeling due to the IL-6-mediated bone resorption that can occur [75, 76, 79].
X. IDENTIFICATION OF A SECOND PTH RECEPTOR The classical PTH1R recognizes the amino-terminal portion of PTH and the homologous N-terminus region of PTHrP. Activation of PTH1R is indistinguishable if the ligand is PTH(1-34), PTHrP(1-36) or intact PTH(1-84). The second receptor, known as PTH2R, differs substantially from PTH1R in that PTH2R responds only to PTH as a ligand. Homologs of this receptor have been identified in rat and in zebrafish. It is expressed widely in many tissues, particularly the central nervous system. The true ligand for the PTH2R is probably not PTH but rather, in the central nervous system, a PTH2R-selective activator termed tuberoinfundibular peptide (TIP). This 39-amino acid peptide shows limited sequence homology to PTH and is a selective agonist of PTH2Rs. Selectivity of the receptors for PTH and PTH/PTHrP has been explored by constructing chimeras or PTH1R and PTH2R in which the extracellular amino-terminal domains were exchanged. In cells expressing PTH2R with the PTH1R amino terminus, the cAMP response to PTH or PTHrP is identical. Mutations to the PTH1R sequence were made in each of the seven transmembrane spanning domains. Mutations in the transmembrane domains 3 and 7 resulted in receptors unable to respond to PTHrP [80]. In the ligand itself, position 23 (Trp in PTH and Phe in PTHrP) determines binding selectivity and position 5 (Ile in PTH and His in PTHrP) is responsible for signaling selectivity. To determine the site in PTH2R that discriminates between the two residues at site 5, PTH2R and PTH1R chimeras were constructed, expressed in COS-7 cells and tested for their response to cAMP. Two single residues in the membranespanning/loop region of the receptor determine signaling selectively, Ile244 at the extracellular end of the transmembrane helix 3 and Try318 at the carboxy-terminal portion of extracellular loop 2 [81]. It appears that the C-terminal portion of TIP39 can bind well to PTH1R. A third PTH receptor, PTH3R, has been identified in zebrafish [82]. This receptor is more closely related,
by amino acid sequence, to mammalian PTH1R than to PTH2R. It is uncertain whether this receptor shows preferential affinity for PTH or to PTHrP. Moreover, since a mammalian variant has not yet been identified, its role in human subjects is uncertain. Large portions of the PTH C-terminus are preserved across species, suggesting the possibility of additional independent biological function(s) for this region of the PTH molecule. This putative, distinct PTH receptor with specificity for the carboxyl-terminal region of PTH was suggested by the response of osteoblasts to the C-terminal portion of PTH(53-84) [83, 84]. Several investigators have identified a C-terminal specific radiolabeled ligand that binds to parathyroid cells, osteosarcoma cells, and to clonal osteocytic cells derived from the PTH1R-null mice [85, 86]. Further work has indicated that the important binding determinants are in the region hPTH(24-27) [50]. In another report, the removal of the C-terminal Gln84 residue inhibited the binding and activity of skeletal C-terminal PTHRs [83]. Recently, a unique clonal osteocytic cell line isolated from the null PTH1R mouse was found to express abundant CPTHRs. Various recombinant and synthetic CPTH peptides were used to map ligand determinants of the CPTHR binding and bioavailability [87]. Eight specific, highly conserved amino acid residues within the PTH sequence were identified to play a key role in the affinity of PTH for CPTHR. These amino acids are displayed over three critical domains, and when all are present, result in maximal binding affinity.
XI. PHYSIOLOGICAL ACTIONS OF PTH A. Catabolic Actions of PTH at the Cellular Level The complexity of the skeleton with its many components has made evaluation of the cellular effects of PTH a challenge to elucidate. Both in vivo and in vitro approaches have been utilized. With in vitro studies, the usual physiological milieu is lacking and potentially important intercommunication among various cells as well as cytokines/ growth factor interactions in response to PTH may be missed. With in vivo experiments, such interactions among cell types can be appreciated but it is difficult to assign a particular role, even when using a specific cell type. Thus, both kinds of studies are needed for optimal insight into the mechanism of PTH action in bone. Since PTH can be catabolic to bone, it is reasonable to focus on the osteoclast. However, the osteoclast is relatively inaccessible for direct studies. Our understanding of the catabolic actions of PTH is further diminished by the fact that osteoclasts do not respond directly to PTH. They do
Chapter 16 Parathyroid Hormone: Structure, Function and Dynamic Actions
not contain receptors for PTH, and hence, all interactions appear to be the result of direct effects on the osteoblast [90]. When osteoblasts are stimulated by PTH, it is believed that the subsequent generation of factor or factors result in stimulation of osteoclast function and numbers (for review see references [91, 93]). The initial rapid response to PTH involves the cells lining the endosteal surface of bone [93]. This rapid response (less than 1 hour) that results in the release of calcium into the circulation, is an important regulatory mechanism for the quick correction of hypocalcemia. This initial phase of PTH activity is associated with increased metabolic activity of the osteoclast and does not require new protein synthesis. A second phase of calcium mobilization that is dependent on protein synthesis occurs approximately 24 hours later. This second phase results in an increase in the number and activity of osteoclasts. Current concepts attribute such indirect actions on the osteoclast to direct effects on the osteoblast: incubation of osteoclasts with osteoblast-like cells results in osteoclast activation. Osteoblasts and adjacent bone marrow stromal cells contain PTH1 receptors. The increased number and activity of osteoclasts [94] may also be mediated through cells of the osteoblast lineage such as lining cells [90]. Several clonal, conditionally immortalized PTH-responsive, bone marrow stromal cell lines derived from mouse marrow, were shown to support formation of tartrate-resistant acid phosphatase-positive multinucleated cells in response to PTH. These multinucleated cells contained calcitonin receptors and formed resorption lacunae on dentine slices, consistent with the osteoclast phenotype [95]. The precise mechanism by which PTH stimulates the osteoclast to activate bone resorption remains unknown. Many different enzymes or other factors are released by PTH. They include collagenase, lysosomal hydroxylase, acid phosphatases, carbonic anhydrase, H+, K+-ATPases, Na+–Ca++ exchange systems, cathepsin B, and cysteine proteases. An acidic environment, essential for the protonation of the alkaline salt hydroxyapatite, is formed by the osteoclast when it seals by podosomes the area between the osteoclast membrane and the mineralized surface. Several enzymes are present at this site such as H+, K+-ATPase. Omeprozole blocks this proton pump resulting in inhibition of PTH-mediated bone resorption [96]. Carbonic anhydrase, also activated by PTH, generates hydrogen ion for H+, K+-ATPase activity. PTH-mediated bone resorption is attenuated by inhibition of carbonic anhydrase [97]. Amiloride and its analog, 3′4′-dichlorobenzamil (DCB) affects calcium transport systems such as ATPdependent calcium pumps. These compounds inhibit PTHinduced bone resorption in neonatal mouse calvaria [98]. The mechanism of bone resorption remains controversial,
283
however, as other studies suggest that PTH is not required for the buffering of protons in neonatal mouse calvaria [99]. The hypothetical construct suggests that PTH induces catabolic activation in the osteoblast that in turn influences the osteoclast. PTH induces the production of plasmin metalloproteinases such as tissue-type plasminogen activator (tPA) that may initiate bone resorption via breakdown of the extracellular matrix. In neonatal rat osteoblast cell cultures, bovine PTH(1-34) increases tPA by approximately 6–8-fold. After induction of osteoblast differentiation with 1,25-dihydroxyvitamin D3, PTH had no effect on tPA [100]. Other catabolic actions of PTH in the osteoblast include stimulation of hyaluronase synthesis [101] and induction of collagenase-3 gene transcription [102].
B. Anabolic Actions of PTH at the Cellular Level The multiple anabolic actions of PTH on the skeleton are due directly to its effects on osteoblasts that contain receptors for PTH. PTH induces the retraction of preosteoblasts from the bone surface through a calcium-dependent, proteolytic modification of the cytoskeleton. Regulation of cell attachment via regulation of E-cadherins [103] might be a mechanism by which PTH leads to an increase in osteoblast number. PTH may also be mitogenic for bone cells in vivo [104]. The role of PTH in the differentiation of osteoblasts was evaluated in a bone morphogeneic protein (BMP)dependent system using a mesenchymal progenitor cell (C3H10T1/2). Constitutive expression of the PTH/PTHrP receptor resulted in stimulation of osteogenic development. PTH(1-34) stimulated the early and suppressed the late stages of osteogenic development. The effect appeared to be mediated by a cAMP signaling cascade. Other PTH peptides such as PTH(28-48) or PTH(53-84) did not result in significant responses. Parathyroid hormone enhances collagenase synthesis, inhibits type I collagen synthesis, and reduces alkaline phosphatase activity in the osteoblast [105–107]. Further studies have better defined the roles of carboxyl-terminal fragments of PTH versus amino-terminal fragments in these effects. In the osteoblast-like cell line UMR-106, PTH(1-84) and PTH(1-34) inhibited cell proliferation and stimulated alkaline phosphatase activity; C-terminal fragments had no effect. Expression of type I procollagen was stimulated by PTH(35-84), PTH(53-84), and PTH(69-84), inhibited by PTH(1-34), and unaffected by PTH(1-84). These results suggest that the carboxy-terminus region of PTH may contain information in its sequence to influence bone formation [108].
284 The effects of PTH on 24-hydroxylase in osteoblasts are opposite to those in the kidney. In combination with 1,25-dihydroxyvitamin D3, PTH enhances mRNA levels of the 24-hydroxylase cytochrome P450 components. This synergism is in marked contrast to regulation of 24-hydroxylase in the kidney where PTH and 1,25dihydroxyvitamin D have antagonistic effects [109]. Parathyroid hormone alters the nuclear matrix in osteoblasts. The rat type I collagen alpha 1(I) polypeptide chain (COL1A1) promoter confirmation is linked to cell structure via the nuclear matrix. Parathyroid hormone increased the binding of a soluble nuclear protein (NMP4) and decreased COL1A1 mRNA in osteosarcoma cells [110]. The same laboratory demonstrated that PTH can alter osteoblast gene expression via changes in the organization of the nucleus structural proteins [111]. The mechanism by which PTH exerts an anabolic effect is unknown. Nuclear matrix proteins such as NuMA, topoisomerase II-α and -β mediate nuclear organization. NuMA may maintain the structural integrity of the interphase nucleus and form discrete transcriptional domains. Further work on the changes in nuclear matrix in response to PTH has been described. PTH treatment down-regulated the number of topoisomerase II-α-immunopositive cells, and correlated with a decrease in S-phase cells in bone tissues and cultured osteoblast-like cells. The authors suggest that anabolic doses of PTH attenuate the proliferative capacity of osteogenic cells by targeting topoisomerase II-α expression [112]. Other investigators have suggested that an amphiregulin may mediate in part an anabolic effect on osteoblasts. Using microassays to target gene expression profile changes in PTH-treated UMR-105-01 osteoblast osteosarcoma cell line, amphiregulin (AR), a member of the epidermal growth factor family, was identified. AR is bifunctional because it inhibits the growth of several human tumor cells, and stimulates the proliferation of other normal cells such as fibroblasts. PTH highly up-regulates expression of amphiregulin, and is a potent growth factor for preosteoblasts. However, AR is bifunctional in bone, as it inhibits further differentiation of these cells. AR-null mice have less tibial trabecular bone than wild-type mice, suggesting a role of AR in PTH-mediated bone formation and resorption [113].
XII. CELL-TO-CELL COMMUNICATION: OSTEOBLASTS AND OSTEOCLASTS Cell-to-cell communication is enhanced by parathyroid hormone. PTH enhances the formation of gap junctions
Lorraine A. Fitzpatrick and John P. Bilezikian
in calvarial osteoblasts [114]. The enhancement of gap junctions, primarily composed of connexin 43 (Cx43), is due to an increased rate of Cx43 gene transcription by PTH. These results, shown for UMR 106 cells, vary when other osteoblast-like cells are studied. The differences may be due to the stage of differentiation of the particular cell model studied. PTH changes the level of Cx43 gene expression in proliferating and maturing osteoblasts but has little effect on nondividing, differentiated osteoblasts [115, 116]. The factors responsible for communication between cells of the osteoblast lineage and those that are osteoclast in nature are the subject of much interest and speculation. Utilizing a human osteosarcoma cell line, SaOS-2, Elias et al. [117] demonstrated that PTH stimulates production of IL-11 protein and mRNA. They proposed this cytokine as the link between the osteoblast and osteoclast. Other investigators have evaluated the effect of IGF-1 in mediating cell signaling between these two bone cells. In newborn rat calvaria, PTH stimulated the production of IGF-1 and IGFBP-3 at the message and peptide levels [118]. Other studies have implicated PTH-induced production of prostaglandins by the osteoblast to induce bone resorption by the osteoclast. In primary cultures of human osteoblastlike cells, PTH has been shown to induce cyclooxygenase-2 gene expression resulting in prostaglandin E2 production [119]. At the molecular level, the influence of PTH on c-fos gene expression in UMR-106 cells was explored [120]. Both PTH(1-34) and PTHrP(1-34) induced c-fos gene expression. Antisense oligonucleotides inhibited PTH-mediated cell proliferation in the osteoblast-like cells and also inhibited PTH-enhanced osteoclast-like cell formation. Those results suggest PTH might regulate bone formation and bone resorption by effects on c-fos gene expression.
A. Nonclassical Actions of PTH PTH promotes the intestinal absorption of calcium though the regulation of renal 1-α-hydroxylation of vitamin D to 1,25(OH)2D3. In pathological or pharmacologic studies, PTH may regulate calcium metabolism via direct stimulation of intestinal calcium absorption [121]. Other nonclassical actions of PTH include the stimulation of gluconeogenesis and alanine uptake in animals [122], a natriuretic and calciuric effect in thyroparathyroidectomized dogs [123], induction of chronotropic effects in rat cardiomyocytes (375), stimulation of intracellular calcium response in pancreatic islets, cardiomyocytes,
285
Chapter 16 Parathyroid Hormone: Structure, Function and Dynamic Actions
hepatocytes, and proximal renal tubular cells. Most of these actions are consistent with activation of PTH1R.
XIII. PREFERENTIAL ACTIONS OF PTH AT SELECTED SKELETAL SITES A. Animal Models Many studies in animals and in humans have attempted to understand the basis of a selective effect of PTH on cortical bone for catabolic events and on cancellous bone for anabolic events. The effect of PTH on indices of bone formation such as bone mineral density are significantly increased after only 10 days and biomechanical properties improved after 15 days in a rat model [124]. In another study, Wistar rats were administered parathyroid hormone after ovariectomy. The entire skeleton showed an increase in bone mineral content during the period of administration of PTH and no effect of withdrawal was noted. The metaphysis was highly sensitive to PTH and the lumbar vertebrae and diaphysis had moderate changes in bone mass in response to PTH, while the skull and caudal vertebral bodies were not responsive to PTH. After withdrawal of PTH, BMD decreased markedly at the sites that were initially responsive to PTH, readministration of PTH resulted in accretion of BMD at PTHsensitive sites [125]. Weekly administration of PTH to ovariectomized rats effectively stimulated bone formation in both the cancellous and cortical compartments suggesting that administration may not have to be as frequent as originally planned in the initial studies in humans [126]. An ovariectomized rat model was utilized to test the effect of analogs of human PTH(1-34) which differ from the native sequence in that their receptor-activating properties could promote bone formation. Single substitutions for serine in the 3-position result in peptides that are partial agonists in the kidney. Cancellous bone volume was significantly lower in the vehicle-treated group of animals and none of the compounds altered cancellous bone volume. All three peptides produced marked dose-related increases in bone formation rates but the two analogs were less potent than hPTH(1-34). All three peptides produced doserelated increases in osteocalcin. Bone resorption markers had variable effects: pyridinoline cross-link excretion was altered by treatment with hPTH(1-34), but the analog [His]hPTH(1-34) caused a dose-dependent decrease in this parameter of bone resorption [127]. Evidence exists for the role of PTH in the responsiveness of the skeleton to mechanical stress [128]. PTH increased the mechanical strength of the femur diaphysis in
aged rats. In normal rats, mechanical stimulation of the caudal vertebra induced an osteogenic response. In the absence of PTH, this response was attenuated and restored by a single injection of PTH before loading. C-fos expression in osteocytes was detectable only in rats receiving both PTH and mechanical stimulation [129]. One of the most informative models regarding the effect of PTH on the skeleton is a mouse model in which the genes encoding the PTHR1 peptide have been ablated by homologous recombination. These mice have skeletal dysplasia due to accelerated endochondral bone formation and die at birth or in utero. Targeted expression of constitutively active PTH1 receptors resulted in delayed mineralization, decelerated conversion of proliferative chondrocytes into hypertrophic cells, and delay of vascular invasion [130]. Targeted disruption of either PTHr1 [131] or PTHrP [132] in mice and defective PTHrP/PTHR1 signaling in man [133, 134] leads to a lethal form of skeletal dysplasia manifest by accelerated differentiation and decreased proliferation of growth plate chondrocytes. Transgenic mice models are used increasingly to predict skeletal response to PTH. The efficacy of intermittent PTH treatment on bone varies among the tested strains of mice with differences in peak bone mass and structure. To assess the responses of various skeletal sites with high or low cancellous bone mass, mature C57BL/6 mice were ovariectomized or sham operated and treatment with PTH(1-34) or vehicle [135]. After 4–11 weeks of ovariectomy, there was a 44% loss of cancellous bone in the proximal tibia and a 25% loss of cancellous bone in the vertebra, with impaired trabeculae architecture and high bone turnover. In the intact animals, PTH increased cancellous bone volume to a greater extent in the vertebral body compared to the proximal tibia. In the ovariectomized mice, PTH increased cancellous bone volume to a greater extent in the vertebral body, a site with moderate cancellous bone loss, compared to the proximal tibia. Treatment added a small amount of cortical bone to the tibia, but did not significantly increase cortical width of the vertebral body. These comparisons suggest that intermittent PTH is more effective at sites where there is an adequate number of trabecular present at the beginning of the treatment. The authors also proposed that there was an interaction of PTH anabolic action and mechanical loading, since the tibia had a greater increase in cortical bone.
B. Human Studies Several informative studies that address the effect of PTH on bone markers in human subjects were performed
286 in patients with primary or secondary hyperparathyroidism. For a complete description of studies in patients with primary hyperparathyroidism, refer to Chapter 46. In a group of patients with secondary hyperparathyroidism and hypovitaminosis D, treatment with either cholecalciferol or calcitriol resulted in a lowering of N-telopeptide excretion [136]. In a small group of patients with prostate cancer, the excessive bone formation that can occur in metastatic disease results in hypocalcemia and subsequent stimulation of PTH in response to the calcium demand. In these patients, serum bone alkaline phosphatase and urinary levels of bone collagen are markedly elevated. Infusion of the bisphosphonate, pamidronate, significantly decreased serum levels of calcium, phosphorus, bone alkaline phosphatase and urinary markers of bone resorption [137]. In patients with primary hyperparathyroidism, or in normal volunteers receiving PTH infusions (1-38), serum ICTP (C-terminal of type I collagen) was elevated and serum PICP (the carboxy-terminal propeptide of type I collagen) was decreased [138]. Osteoporotic and normal postmenopausal women were studied before and after calcium deprivation to assess responsiveness to endogenous PTH. Baseline values of bone turnover, which included osteocalcin and urinary deoxypyridinoline and pyridinoline cross-link excretion, were higher in a group of postmenopausal osteoporotic women in spite of decreased serum PTH levels as compared to normal postmenopausal women. Calcium deprivation resulted in similar changes in serum levels of calcium, PTH, 1,25(OH)2D3, and pyridinoline and deoxypyridinoline cross-link excretion. Serum osteocalcin increased and serum procollagen carboxy-terminal propeptide decreased in normal women who were calcium-deprived but these bone turnover markers were not altered in the postmenopausal osteoporotic group. These data suggest that there is no difference in the skeletal responsiveness to PTH in patients with osteoporosis [139]. To determine the role of PTH in age-related and nocturnal increases in bone resorption, calcium infusion was utilized to suppress endogenous PTH secretion in young and elderly normal women. Serum PTH and urinary cross-linked N-telopeptide of type I collagen (NTx) were circadian in pattern. Peak levels of PTH occurred in the mid-afternoon and at night, and a rise in urinary NTx was found at night. At baseline, levels of urinary NTx were higher in the elderly compared to young women. Calcium infusion reduced the PTH peaks but did not alter the nocturnal increase in urinary NTx excretion. The authors suggest that PTH is responsible for the age-related increase in bone resorption but does not mediate the circadian pattern of bone resorption [140]. In contrast, administration of recombinant PTH(1-84) to healthy postmenopausal
Lorraine A. Fitzpatrick and John P. Bilezikian
women resulted in no change in urinary deoxypryridinoline, although urinary cAMP excretion was appropriately elevated in response to PTH administration [141]. Racial differences exist in the incidence of hip and vertebral fractures: African-American women have a lower incidence of fracture at these sites compared to Caucasian women. One possible mechanism is a difference in the sensitivity of the skeleton to PTH. Cosman and colleagues infused PTH(1-34) in healthy premenopausal black and white women and measured indices of bone turnover. Baseline levels of 1,25(OH)2D3 were significantly lower in black women. There were trends toward higher PTH and lower urinary calcium and pyridinoline levels in this group. No differences in serum calcium or endogenous levels of PTH were noted among the groups during PTH infusion. Black women had lower levels of urinary calcium during the PTH infusion and no differences were detected in bone formation markers among black and white women. Marked differences were noted in markers of bone resorption. Cross-linked N-telopeptides, cross-linked C-telopeptides and free pyridinoline cross-links were much more elevated in white subjects as compared to blacks. These findings suggest that black women have decreased sensitivity and may be an explanation for the relative preservation of skeletal tissue [142]. In postmenopausal women receiving estrogen replacement therapy, daily administration of PTH(1-34) (400 units/ day) over a 3-year period resulted in a continuous increase in bone mineral density compared to no increase in the estrogen-treatment-only control group. The increase in vertebral BMD was 13% while the increase at the hip was less at 2.7%; no loss of bone was noted at any skeletal site tested. Serum osteocalcin measurements were increased 55% during the first 6 months of treatment while a marker of bone resorption, excretion of crosslinked N-telopeptide, increased only by 20%. These changes were interpreted to be due to an uncoupling of the bone formation–resorption cycle and the differences fell toward baseline during the study progression. These data suggest that PTH has potent anabolic effects on the human skeleton [143]. The effects of PTH as an anabolic agent on bone markers is given complete coverage in Chapter 46. In a randomized, double-blind, placebo-controlled trial of idiopathic osteoporosis in men, the effect of administration of parathyroid hormone was evaluated on markers at bone turnover. Bone formation markers that included osteocalcin, carboxyterminal propeptide of type I collagen, and bone-specific alkaline phosphatase, increased in the PTH-treated group. The carboxy-terminal propeptide of type I collagen peaked at 67% above baseline at 6 months and bone-specific alkaline phosphatase peaked at 169% above baseline at 9 months. Osteocalcin was highest (230%) at 1 year.
287
Chapter 16 Parathyroid Hormone: Structure, Function and Dynamic Actions
PTH administration also dramatically altered bone resorption markers. Pyridinoline was 131% above baseline at 9 months and urinary N-telopeptide was 375% above baseline at 1 year. The relationship between pyridinoline and markers of bone formation were highly correlated in the PTH-treated group but not in the control group. Elevations in bone formation were consistent with a 13.5% increase in bone mass at the lumbar spine in the PTH-treated group at 18 months. Pyridinoline levels at baseline and osteocalcin at 3 months were the best predictors of response to treatment. Parathyroid hormone is the potent stimulator of bone turnover even in men with idiopathic low turnover osteoporosis [144]. Several randomized, placebo-controlled, multicenter studies have evaluated the role of PTH(1-34), teriparatide, as a therapeutic intervention for postmenopausal osteoporosis. In 1637 postmenopausal women with prior vertebral fractures, daily subcutaneous injection of placebo, 20 or 40 µg of teriparatide were administered [145]. The relative risks of fracture compared to the placebo group were 0.35 (95% CI 0.22–0.55) and 0.31 (95% CI 0.19–0.50) in the 20 and 40 µg groups, respectively. New nonvertebral fractures occurred in 6% of women in the placebo group and 3% of each treatment group (RR 0.47 and 0.46, respectively). The 20-µg and 40-µg doses of teriparatide increased BMD by 9 and 13% in the lumbar spine and by 3 and 6% in the femoral neck compared to placebo (20-µg and 40-µg, respectively). In this same trial, the relationship between prior fractures and the risk of new fractures was evaluated [146]. Among the placebo patients with mild, moderate, or severe prevalent vertebral fractures, 10%, 13%, and 28%, respectively, developed vertebral fractures and 4%, 8%, and 23% developed moderate or severe vertebral fractures (both P < 0.001). In the teriparatide-treated group, there were no significant increases in vertebral fracture risk in these subgroups. Similar results were noted regarding nonvertebral fractures. To understand the mechanism by which PTH affects cortical or trabecular bone, a study that evaluated the combination of teriparatide with alendronate or either alone provided additional insights [147]. Using DXA as a measurement, the findings with teriparatide were similar to other trials with increases in BMD at the lumbar spine and total hip, but decreases at the distal one-third radius. Quantitative CT was used to measure volumetric bone mineral density of trabecular bone at the spine and hip and volumetric bone mineral density and geometric variables in cortical bone at the hip. The integral volumetric density (cortical plus trabecular bone) at the spine was similar to the pattern seen in the areal density of the spine. The volumetric density of the trabecular bone at the spine increased markedly, however the increase in the teriparatide
group was approximately twice as great as that found in the alendronate plus teriparatide group. The volumetric density of the trabecular bone at the hip increased with teriparatide administration. In contrast, the volumetric density of cortical bone at the total hip decreased significantly in the parathyroid hormone group.
References 1. Miric, A., and Levine, M. A. (1992). Analysis of the preproPTH gene by denaturing gradient gel electrophoresis in familial isolated hypoparathyroidism. J. Clin. Endocrinol. Metab. 74, 509–516. 2. Kronenberg, H. M., Igarashi, T., Freeman, M. W., Okazaki, T., Brand, S. J., Wiren, K. M., and Potts, J. T., Jr. (1986). Structure and expression of the human parathyroid hormone gene. Recent. Prog. Horm. Res. 42, 641–663. 3. Demay, M. B., Kiernan, M. S., DeLuca, H. F., and Kronenberg, H. M. (1992). Sequences in the human parathyroid hormone gene that bind the 1,25-dihydroxyvitamin D3 receptor and mediate transcriptional repression in response to 1,25-dihydroxyvitamin D3. Proc. Natl Acad. Sci. USA 89, 8097–8101. 4. Kilav, R., Silver, J., and Naveh-Many, T. (1995). Parathyroid hormone gene expression in hypophosphatemic rats. J. Clin. Invest. 96, 327–333. 5. Moallem, E., Silver, J., Kilav, R., and Naveh-Many, T. (1998). RNA protein binding and post-transcriptional regulation of PTH gene expression by calcium and phosphate. J. Biol. Chem. 273, 5253–5259. 6. Antonarakis, S. E., Phillips, J. A., Mallonee, R. L., Kazazian, H. H. J., Fearon, E. R., Waber, P. G., Kronenberg, H. M., Ullrich, A., and Meyers, D. A. (1983). Beta-globin locus is linked to the parathyroid hormone (PTH) locus and lies between the insulin and PTH loci in man. Proc. Natl Acad. Sci. USA 80, 6615–6619. 7. Mayer, H., Breyel, E., Bostock, C., and Schmidtke, J. (1983). Assignment of the human parathyroid hormone gene to chromosome 11. Hum. Genet. 64, 283–285. 8. Zabel, B. U., Kronenberg, H. M., Bell, G. I., and Shows, T. B. (1985). Chromosome mapping of genes on the short arm of human chromosome 11: parathyroid hormone gene is at 11p15 together with the genes for insulin, c-Harvey-ras 1, and b-hemoglobin. Cytogenet. Cell. Genet. 39, 200–205. 9. Miao, D., He, B., Karaplis, A. C., and Goltzman, D. (2002). Parathryoid hormone is essential for normal fetal bone formation. J. Clinical Invest. 109, 1173–1182. 10. Miao, D., He, B., Lanske B., et al. (2004). Skeletal abnormalities in PTH-null mice are influenced by dietary calcium. Endocrinology 145, 2046–2053. 11. Zhou, X-G., Liu, Y-Z., Li, and et al. (2003). Parathyroid hormone gene with bone phenotypes in Chinese. Biochem. Biophys. Res. Comm. 307, 666–671. 12. Goswami, R., Mohappatra, T., Gupta, N., and et al. (2004). Parathyroid hormone gene polymorphism and sporadic idiopathic hypoparathyroidism. J. Clin. Endocrinol. Metab. 89, 4840–4845. 13. Gong, G., Johnson, M. L., Barger-Lux, M. J., and Heaney, R. P. (1999). Association of bone dimensions with a parathyroid hormone gene polymorphism in women. Osteoporosis Int. 9, 307–311. 14. Gohda, T., Shou, I., Fukui, M., and et al. (2002). Parathyroid hormone gene polymorphism and secondary hyperparathyroidism in hemodialysis patients. J. Kidney Dis. 39, 1255–12560. 15. Kanzawa, M., Sugimoto, T., Kobayashi, T., Kobayashi, A., and Chihara, K. (1999). Parathyroid hormone gene polymorphisms in primary hyperparathyroidism. Clin. Endocrinol. 50, 583–588.
288 16. Sunthornthepvarakul, T., Churesigaew, S., and Ngowngarmratana, S. (1999). A novel mutation of the signal peptide of the preproparathyroid hormone gene associated with autosomal recessive familial isolated hypoparathyroidism. J. Clin. Endocrinol. Metab. 84, 3792–3796. 17. Russell, J., Lettieri, D., and Sherwood, L. M. (1983). Direct regulation of calcium of cytoplasmic messenger ribonucleic acid coding for preproparathyroid hormone in isolated bovine parathyroid cells. J. Clin. Invest. 72, 1851–1855. 18. Naveh-Many, T., Friedlander, M. M., Mayer, H., and Silver, J. (1989). Calcium regulates parathyroid hormone messenger ribonucleic acid (mRNA), but not calcitonin mRNA in vivo in the rat. Dominant role of 1,25-dihydroxyvitamin D. Endocrinology 125, 275–280. 19. Brown, E. M. (2005). Editorial: mutant extracellular calcium-sensing receptors and severity of disease. J. Clin. Endocrinol. Metab. 90, 1246–1248. 20. Kifor, O., Diaz, R., Butters, R., and Brown, E. M. (1997). The Ca2+sensing receptor (CaR) activates phospholipases C, A2, and D in bovine parathyroid and CaR-transfected, human embryonic kidney (HEK293) cells. J. Bone Miner. Res. 12, 715–725. 21. Kifor, O., MacLeod, R. J., Diaz, R., Bai, M., Yamaguchi, T., Yao, T., Kifor, I., and Brown, E. M. (2001). Regulation of MAP kinase by calcium-sensing receptor in bovine parathyroid and CaR-transfected HEK293 cells. Am. J. Physiol. Renal. Physiol. 280, F291–F302. 22. Parfitt, A. M. (2001). Parathyroid growth. In “The Parathyroids Basic and Clinical Concepts” (J. P. Bilezikian, R. Marcus, and M. A. Levine, eds), 2nd ed., pp. 293–329. Academic Press, San Diego, CA. 23. Okazaki, T., Zajac, J. D., Igarashi, T., Ogata, E., and Kronenberg, H. M. (1991). Negative regulatory elements in the human parathyroid hormone gene. J. Biol. Chem. 266, 21903–21910. 24. Moallem, E., Kilav, R., Silver, J., and Naveh-Many, T. (1998). RNAprotein binding and post-transcriptional regulation of parathyroid hormone gene expression by calcium and phosphatase. J. Biol. Chem. 273, 5253–5259. 25. Hawa, N. S., O’Riordan, J. L., and Farrow, S. M. (1993). Posttranscriptional regulation of bovine parathyroid hormone synthesis. J. Mol. Endocrinol. 10, 43–49. 26. Vadher, S., Hawa, N. S., O’Riordan, J. L., and Farrow, S. M. (1996). Translational regulation of parathyroid hormone gene expression and RNA: protein interactions. J. Bone Miner. Res. 11, 746–753. 27. Kilav, R., Bell, O., Le, S-Y., Silver, J., and Naveh-Many, T. (2004). The parathyroid hormone mRNA3′-untranslated region AU-rich element is an unstructured functional element. J. Biol. Chem. 279, 2109–2116. 28. Silver, J., Russell, J., and Sherwood, L. M. (1985). Regulation by vitamin D metabolites of messenger ribonucleic acid for preproparathyroid hormone in isolated bovine parathryoid cells. Proc. Natl. Acad. Sci. USA 82, 4270–4273. 29. Russell, J., Lettieri, D., and Sherwood, L. M. (1986). Suppression by 1,25(OH)2D3 of transcription of the pre-proparathyroid hormone gene. Endocrinology 119, 2864–2866. 30. Silver, J., Naveh-Many, T., Mayer, H., Schmelzer, H. J., and Popovtzer, M. M. (1986). Regulation by vitamin D metabolites of parathyroid hormone gene transcription in vivo in the rat. J. Clin. Invest. 78, 1296–1301. 31. Okazaki, T., Igarashi, T., and Kronenberg, H. M. (1988). 5′-flanking region of the parathyroid hormone gene mediates negative regulation by 1,25-(OH)2 vitamin D3. J. Biol. Chem. 263, 2203–2208. 32. Mackey, S. L., Heymont, J. L., Kronenberg, H. M., Demay, M. B. (1996). Vitamin D receptor binding to the negative human parathyroid hormone vitamin D response element does not require the retinoid x receptor. Mol. Endocrinol. 10, 298–305. 33. Sela-Brown, A., Russell, J., Koszewski, N. J., and et al. (1998). Calreticulin inhibits vitamin D’s action on the PTH gene in vitro and may prevent vitamin D’s effect in vivo in hypocalcemic rats. Mol. Endocrinol. 12, 1193–1200.
Lorraine A. Fitzpatrick and John P. Bilezikian 34. Slatopolsky, E., and Bricker, N. S. (1973). The role of phosphorus restriction in the prevention of secondary hyperparathyroidism in chronic renal disease. Kidney Int. 4, 141–145. 35. Lopez-Hilker, S., Dusso, A. S., Rapp, N. S., Martin, K. J., and Slatopolsky, E. (1990). Phosphorus restriction reverses hyperparathyroidism in uremia independent of changes in calcium and calcitriol. Am. J. Physiol. 259, F432–F437. 36. Kilav, R., Silver, J., Biber, J., Murer, H., and Naveh-Many, T. (1995). Coordinate regulation of the rat renal parathyroid hormone receptor mRNA and the Na-Pi cotransporter mRNA and protein. Am. J. Physiol. 268, F1017–F1022. 37. Moallem, E., Kilav, R., Silver, J., and Naveh-Many, T. (1998). RNAProtein binding and post-transcriptional regulation of parathyroid hormone gene expression by calcium and phosphate. J. Biol. Chem. 273, 5253–5259. 38. Almaden, Y., Hernandez, A., Torregrosa, V., Campistol, J., Torres, A., and Rodriguez, M. S. (1995). High phosphorous directly stimulates PTH secretion by human parathyroid tissue. J. Am. Soc. Nephrol. 6, 957. 39. Naveh-Many, T., Marx, R., Keshet, E., Pike, J. W., and Silver, J. (1990). Regulation of 1,25-dihydroxyvitamin D3 receptor gene expression by 1,25-dihydroxyvitamin D3 in the parathyroid in vivo. J. Clin. Invest. 86, 1968–1975. 40. Russell, J., Bar, A., Sherwood, L. M., and Hurwitz, S. (1993). Interaction between calcium and 1,25-dihydroxyvitamin D3 in the regulation of preproparathyroid hormone and vitamin D receptor messenger ribonucleic acid in avian parathyroids. Endocrinology 132, 2639–2644. 41. Brown, A. J., Zhong, M., Finch, J., Ritter, C., and Slatopolsky, E. (1995). The roles of calcium and 1,25-dihydroxyvitamin D3 in the regulation of vitamin D receptor expression by rat parathyroid glands. Endocrinology 136, 1419–1425. 42. Okazaki, T., Igarashi, T., and Kronenberg, H. M. (1988). 5′-Flanking region of the parathyroid hormone gene mediates negative regulation by 1,25-(OH)2 vitamin D3. J. Biol. Chem. 263, 2203–2208. 43. Jääskeläainen, T., Huhtakangas, J., and Mäenpää, P. H. (2005). Negative regulation of human parathyroid hormone gene promoter by vitamin D3 through nuclear factor Y. Biochem. Biophys. Res. Commun. 328, 831–837. 44. Peraldi, M. N., Rondeau, E., Jousset, V., et al. (1990). Dexamethasone increases preproparathyroid hormone messenger RNA in human hyperplastic parathyroid cell in vitro. Eur. J. Clin. Invest. 20, 392–397. 45. Karmali, R., Farrow, S., Hewison, M., Barker, S., and O’Riordan, J. L. H. (1989). Effects of 1,25-dihydroxyvitamin D3 and cortisol on bovine and human parathyroid cells. J. Endocrinol. 123, 137–142. 46. Moallem, E., Silver, J., and Naveh-Many, T. (1995). Regulation of parathyroid hormone messenger RNA levels by protein kinase A and C in bovine parathyroid cells. J. Bone Miner. Res. 10, 447–452. 47. Clarke, B. L., Hassager, C., and Fitzpatrick, L. A. (1993). Regulation of parathyroid hormone release by protein kinase-C is dependent on extracellular calcium in bovine parathyroid cells. Endocrinology 132, 1168–1175. 48. Naveh-Many, T., Almogi, G., Livni, N., and Silver, J. (1992). Estrogen receptors and biologic response in rat parathyroid tissue and C cells. J. Clin. Invest. 90, 2434–2438. 49. Epstein, E., Silver, J., Almogi, G., Livni, N., and Naveh-Many, T. (1996). Parathryoid hormone mRNA levels are increased by progestins and vary during the rat estrous cycle. Am. J. Physiol. 33, E158–E163. 50. Murray, T. M., Rao, L. G., Divieti, P., and Bringhurst, F. R. (2005). Parathyroid hormone secretion and action: evidence for discrete receptors for the carboxyl-terminal region and related biological actions of carboxyl-terminal ligands. Endocrine Rev. 26, 78–113. 51. Kronenberg, H. M., Bringhurst, F. R., Segre, G. V., and Potts, J. T., Jr. (1994). Parathyroid hormone biosynthesis and metabolism. In “The Parathyroids Basic and Clinical Concepts” (J. P. Bilezikian, R. Marcus, and M. A. Levine, eds.), pp. 125–137. Raven Press, Ltd., New York.
Chapter 16 Parathyroid Hormone: Structure, Function and Dynamic Actions 52. Lazure, C., Gauthier, D., Jean, F., Boudreault, A., Seidah, N. G., Bennett, H. P., and Hendy, G. N. (1998). In vitro cleavage of internally quenched fluorogenic human proparathyroid hormone and proparathyroid-related peptide substrates by furin. Generation of a potent inhibitor. J. Biol. Chem. 273, 8572–8580. 53. Habener, J. F., Kemper, B., and Potts, J. T., Jr. (1975). Calcium-dependent intracellular degradation of parathyroid hormone: a possible mechanism for the regulation of hormone stores. Endocrinology 97, 431–441. 54. Chu, L. L. H., MacGregor, R. R., Anast, C. S., et al. (1973). Studies on the biosynthesis of rat parathyroid hormone and proparathyroid hormone: adaptation of the parathyroid gland to dietary restriction of calcium. Endocrinology 93, 915–924. 55. Bringhurst, F. R., Stern, A. M., Yotts, M., Mizrahi, N., Segre, G. V., and Potts, J. T., Jr. (1988). Peripheral metabolism of PTH: fate of biologically active amino terminus in vivo. Am. J. Physiol. 255, E886–E893. 56. Bringhurst, F. R., Segre, G. V., Lampman, G. W., and Potts, J. T., Jr. (1982). Metabolism of parathyroid hormone by Kupffer cells: analysis by reverse-phase high-performance liquid chromatography. Biochemistry 21, 4252–4258. 57. Yasuda, T., Banville, D., Hendy, G. N., and Goltzman, D. (1989). Characterization of the human parathyroid hormone-like peptide gene: functional and evolutionary aspects. J. Biol. Chem. 264, 7720–7725. 58. Caulfield, M. P., McKee, R. L., Goldman, M. E., et al. (1990). The bovine renal parathyroid hormone (PTH) receptor has equal affinity for two different amino acid sequences: the receptor binding domains of PTH and PTH-related protein are located within the 14–34 region. Endocrinology 127, 83–87. 59. Strewler, G. J., Stern, P. H., Jacobs, J. W., et al. (1987). Parathyroid hormone-like protein from human renal carcinoma cells. Structural and functional homology with parathyroid hormone. J. Clin. Invest. 80, 1803–1807. 60. Suva, L. J., Winslow, G. A., Wettenhall, R. E., et al. (1987). A parathyroid hormone-related protein implicated in malignant hypercalcemia, cloning and expression. Science 237, 893–896. 61. Karpf, D. B., Arnaud, C. D., Bambino, T., et al. (1988). Structure properties of the renal parathyroid hormone receptor: hydrodynamic analysis and protease sensitivity. Endocrinology 123, 2611–2619. 62. Jüppner, H., Abou-Samra, A. B., Freeman, M., et al. (1991). A G Proteinlinked receptor for parathyroid hormone and parathyroid hormonerelated peptide. Science 254, 1024–1026. 63. Schipani, E., Karga, H., Karaplis, A. C., et al. (1993). Identical cDNAs encode a human renal and bone parathyroid hormone/parathyroid hormone-related peptide receptor. Endocrinology 132, 2157–2165. 64. Iida-Klein, A., Guo, J., Xie, L. Y., and et al. (1995). Trunication of the carboxyl-terminal region of the rat parathyroid hormone (PTH)/ PTH-related peptide receptor enhances PTH stimulation of the adenylyl cyclase but not phospholipase C. J. Biol. Chem. 270, 8458–8465. 65. Bringhurst, F. R., Juppner, H., Guo, J., Urena, P., Potts, J. T., Jr., Kronenberg, H. M., Abou-Samra, A. B., and Segre, G. V. (1993). Cloned, stably expressed parathyroid hormone (PTH)/PTH-related peptide receptors activate multiple messenger signals and biological responses in LLC-PK1 kidney cells. Endocrinology 132, 2090–2098. 66. Huang, Z., Chen, Y., Pratt, S., Chen, T. H., Bambino, T., Nissenson, R. A., and Shoback, D. M. (1996). The N-terminal region of the third intracellular loop of the parathyroid hormone (PTH)/PTH-related peptide receptor is critical for coupling to cAMP and inositol phosphate/Ca2+ signal transduction pathways. J. Biol. Chem. 271, 33382–33389. 67. Schipani, E., Langman, C. B., Parfitt, A. M., and et al. (1996). Constitutively activated receptors for parathyroid hormone and parathyroid hormone-related peptide in Jansen’s metaphyseal chondrodysplasia. New Engl. J. Med. 335, 708–714. 68. Turner, P. R., Bambino, T., and Nissenson, R. A. (1996). A putative selectivity filter in the G-protein-coupled receptors for parathyroid hormone and secretion. J. Biol. Chem. 271, 9205–9208.
289
69. Sneddon, W. B., Syme, C. A., Bisello, and et al. (2003). Activationindependent parathyroid hormone receptor internalization is regulated by NHERF1 (EBP50). J. Biol. Chem. 278, 43787–43796. 70. Majesca, R. J., Rodan, S. B., and Rodan, G. A. (1980). Parathyroid hormone-responsive clonal cell lines from rat osteosarcoma. Endocrinology 107, 1494–1503. 71. Chase, L. R., Fedak, S. A., and Aurbach, G. D. (1969). Activation of skeletal adenyl cyclase by parathyroid hormone in vitro. Endocrinology 84, 761–768. 72. Chase, L. R., Melson, G. L., and Aurbach, G. D. (1969). Pseudohypoparathyroidism: defective excretion of 3′, 5′-AMP in response to parathyroid hormone. J. Clin. Invest. 48, 1832–1844. 73. Tregear, G. W., Van Rietschoten, J., Greene, E., et al. (1973). Bovine parathyroid hormone: Minimum chain length of synthetic peptide required for biological activity. Endocrinology 93, 1349–1353. 74. Partridge, N. C., Alcorn, D., Michelangeli, V. P., Kemp, B. E., Ryan G. B., and Martin, T. J. (1981). Functional properties of hormonally responsive cultured normal and malignant rat osteoblastic cells. Endocrinology 108, 213–219. 75. Rasmussen, H. (1986). The calcium messenger system (first of two parts). N. Engl. J. Med. 314, 1094–1101. 76. Rasmussen, H. (1986). The calcium messenger system (second of two parts). N. Engl. J. Med. 314, 1164–1170. 77. Ohno, S., Akita, Y., Hata, O., Osada, S., Kubo, K., Konno, Y., and et al. (1991). Structural and functional diversities of a family of signal transducing protein kinases, protein kinase C family, two distinct classes of PKC, conventional cPKC and novel nPKC. Advan. Enzyme Reg. 31, 287–303. 78. Singh, A. T. K., Bhattacharyya, R. S., Radeff, J. M., and Stern, P. H. (2003). Regulation of parathyroid hormone-stimulated phospholipase D in UMR-106 cells by calcium, MAP kinase, and small G Proteins. J. Bone Miner. Res. 18, 1453–1460. 79. Radeff, J. M., Singh, A. T. K., Stern, P. H. (2004). Role of protein kinase A, phospholipase C and phospholipase D in parathyroid hormone receptor regulation of protein kinase Cα and interleukin-6 in UMR-106 osteoblastic cells. Cell Signalling 16, 105–114. 80. Turner, P. R., Mefford, S., Bambino, T., and Nissenson, R. A. (1998). Transmembrane residues together with the amino terminus limit the response of the parathyroid hormone (PTH) 2 receptor to PTH-related peptide. J. Biol. Chem. 273, 3830–3837. 81. Bergwitz, C., Jusseaume, S. A., Luck, M. D., Juppner, H., and Gardella, T. J. (1997) Residues in the membrane-spanning and extracellular loop regions of the parathyroid hormone (PTH) –2 receptor determine signaling selectivity for PTH and PTH-related peptide. J. Biol. Chem. 272, 28861–28868. 82. Gensure, R. C., Ponugoti, B., Gunes, Y., and et al. (2004). Identification and characterization of two parathyroid hormone-like molecules in zebrafish. Endocrinology 145, 1634–1639. 83. Takasu, H., Baba, H., Inomata, N., and et al. (1996). The 69–84 amino acid region of the parathyroid hormone molecule is essential for the interaction of the hormone with the binding sites with carboxyl-terminal specificity. Endocrinology 137, 5537–5543. 84. Murray, T. M., Rao, L. G., and Muzaffar, S. A. (1991). Dexamethasonetreated ROS17/2.8 rat osteosarcoma cells are responsive to human carboxylterminal parathyroid hormone peptide hPTH (53–84): stimulation of alkaline phosphatase. Calcif. Tiss. Int. 49, 120–123. 85. Inomata, N., Akiyama, M., Kubota, N., and Jüppner, H. (1995). Characterization of a novel parathyroid hormone (PTH) receptor with specificity for the carboxyl-terminal region of PTH-(1-84). Endocrinology 136, 4732–4740. 86. Divieti, P., Inomata, N., Chapin, K., Singh, R., Jüppner, H., and Bringhurst, F. R. (2001). Receptors for the carboxyl-terminal region of PTH (1-84) are highly expressed in osteocytic cells. Endocrinology 142, 916–925.
290 87. Divieti, P., Geller, A. I., Sulima, G., Jüppner, H., and Bringhurst, F. R. (2005). Receptors specific for the carboxyl-terminal region of parathyroid hormone on bone-derived cells: determinants of ligand binding and bioactivity. Endocrinology 146, 1863–1870. 88. Eichinger, A., Fiaschi-Taesch, N., Massfelder, T., Fritsch, S., Barthelmebs, M., and Helwig, J. J. (2002). Transcript expression of the turberoinfundibular peptide (TIP)39/PTH2 receptor system and non-PTH1 receptor-mediated tonic effects of TIP39 and other PTH2 receptor ligands in renal vessels. Endocrinology 143, 30336–3043. 89. John, M. R., Arai, M., Rubin, D. A., Jonsson, K. B., and Jüppner, H. (2002). Identification and characterization of the murine and human gene encoding the tuberoinfundibular peptide of 39 residues. Endocrinology 143, 1047–1057. 90. McSheehy, P. M. J., and Chambers, T. J. (1986) Osteoblastic cells mediate osteoclastic responsiveness to parathyroid hormone. Endocrinology 118, 824–828. 91. Martin, T. J. and Udagawa, N. (1998). Hormonal regulation of osteoclast function. Trends Endocrinol. Metab. 9, 6–12. 92. Roodman, G. D. (1998). Osteoclast differentiation and activity. Biochem. Soc. Trans. 26, 7–13. 93. Talmage, R.V., et al. (1978). The demand for bone calcium in maintenance of plasma calcium concentration. In “Mechanisms of Localized Bone Loss” (J. E. Horton, T. M. Tarplay, and W. F. Davis, eds.), pp. 73–91. Information Retrieval, Washington, DC. 94. Rodan, G. A., and Martin, T. J. (1981). Role of osteoblasts in hormonal control of bone resorption – a hypothesis. Calcif. Tiss. Int. 33, 349. 95. Liu, B. Y., Guo, J., Lanske, B., Divieti, P., Kronenberg, H. M., and Bringhurst, F. R. (1998). Conditionally immortalized murine bone marrow stromal cells mediate parathyroid hormone-dependent osteoclastogenesis in vitro. Endocrinology 139, 1952–1964. 96. Tuukkanen, J., and Vaananen, H. K. (1986). Omeprazole, a specific inhibitor of H+-K+-ATPase, inhibits bone resorption in vitro. Calcif. Tiss. Int. 38, 123–125. 97. Hall, G. E., and Kenny, A. D. (1986). Bone resorption induced by parathyroid hormone and dibutyryl cyclic AMP: role of carbonic anhydrase. J. Pharmacol. Exp. Ther. 223, 778–782. 98. Krieger, N. S., and Kim, S. G. (1988). Dichlorobenzamil inhibits stimulated bone resorption in vitro. Endocrinology 122, 415–420. 99. Bushinsky, D. A. (1987). Effects of parathyroid hormone on net proton flux from neonatal mouse calvariae. Am. J. Physiol. 252(4, part 2), F585–F589. 100. Catherwood, B. D., Titus, L., Evans, C. O., Rubin, J., Boden, S. D., and Nanes, M. S. (1994). Increased expression of tissue plasminogen activator messenger ribonucleic acid is an immediate response to parathyroid hormone in neonatal rat osteoblasts. Endocrinology 134, 1429–1436. 101. Midura, R. J., Evanko, S. P., and Hascall, V. C. (1994). Parathyroid hormone stimulates hyaluronan synthesis in an osteoblast-like cell line. J. Biol. Chem. 269, 13200–13206. 102. Selvamurugan, N., Chou, W. Y., Pearman, A. T., Pulumati, M. R., and Partridge, N. C. (1998). Parathyroid hormone regulates the rat collagenase-3 promoter in osteoblastic cells through the cooperative interaction of the activator protein-1 site and the runt domain binding sequence. J. Biol. Chem. 273, 10647–10657. 103. Babich, M., and Foti, L. R. (1994). E-cadherins identified in osteoblastic cells: effects of parathyroid hormone and extracellular calcium on localization. Life Sci. 54, 201–208. 104. Schlüter, K. D., Hellstern, H., Wingender, E., and Mayer, H. (1989). The central part of parathyroid hormone stimulates thymidine incorporation of chondrocytes. J. Biol. Chem. 264, 11087–11092. 105. Hall, A. K., and Dickson, I. R. (1985). The effects of parathyroid hormone on osteoblast-like cells from embryonic chick calcaria. Acta Endocrinol. 108, 217–233.
Lorraine A. Fitzpatrick and John P. Bilezikian 106. Heath, J. K., Atkinson, S. J., Meikle, M. C., and Reynolds, J. J. (1984). Mouse osteoblasts synthesize collagenase in response to bone resorbing agents. Biochim. Biophys. Acta. 802, 151–154. 107. Simon, L. S., Slovik, S. M., Neer, R. M., and Krane, S. M. (1988). Changes in serum levels of type I and III procollagen extension peptides during infusion of human parathyroid hormone fragment (1-34). J. Bone Miner. Res. 3, 241–246. 108. Nasu, M., Sugimoto, T., Kaji, H., et al. (1998). Carboxyl-terminal parathyroid hormone fragments stimulate type-1 procollagen and insulin-like growth factor-binding protein-5 mRNA expression in osteoblastic UMR-106 cells. Endocr. J. 45, 229–234. 109. Armbrecht, J. H., and Hodam, T. L. (1994). Parathyroid hormone and 1,25-dihydroxyvitamin D synergistically induce the 1,25-dihydroxyvitamin D-24-hydroxylase in rat UMR-106 osteoblast-like cells. Biochem. Biophys. Res. Commun. 205, 674–679. 110. Alvarez, M., Thunyakitpisal, P., Morrison, P., Onyia, J., Hock, J., and Bidwell, J. P. (1998). PTH-responsive osteoblast nuclear matrix architectural transcription factor binds to the rat type I collagen promoter. J. Cell Biochem. 69, 336–352. 111. Torrungruang, K., Feister, H., Swartz, D., et al. (1998). Parathyroid hormone regulates the expression of the nuclear mitotic apparatus protein in the osteoblast-like cells. Bone 22, 317–324. 112. Feister, H. A., Onyia, J. E., Miles, R. R., and et al. (2000). The expression of the nuclear matrix proteins NuMA, topoisomerase II-α and -β in bone and osseous cell culture: regulation by parathyroid hormone. Bone 26, 227–234. 113. Qin, L., Tamasis, J., Raggatt, L., and et al. (2005). Amphiregulin is a novel growth factor involved in normal bone development and in the cellular response to parathyroid hormone stimulation. J. Biol. Chem. 280, 3974–3981. 114. Massas, R., and Benayahu, D. (1998). Parathyroid hormone effect on cell-to-cell communication in stromal and osteoblastic cells. J. Cell. Biochem. 69, 81–86. 115. Schiller, P. C., Roos, B. A., and Howard, G. A. (1997). Parathyroid hormone up-regulation of connexin 43 gene expression in osteoblasts depends on cell phenotype. J. Bone Miner. Res. 12, 2005–2013. 116. Civitelli, R., Ziambaras, K., Warlow, P. M., Lecanda, R., Nelson, T., Harley, J., Atal, N., Beyer, E. C., and Steinberg, T. H. (1998). Regulation of connexin43 expression and function by prostaglandin E2 (PGE2) and parathyroid hormone (PTH) in osteoblastic cells. J. Cell Biochem. 68, 8–21. 117. Elias, J. A., Tang, W., and Horowitz, M. C. (1995). Cytokine and hormonal stimulation of human osteosarcoma interleukin-11 production. Endocrinology 136, 489–498. 118. Schmid, C., Schlapfer, I., Peter, M., Boni-Schnetzler, M., and Schwander, J. (1994). Growth hormone and parathyroid hormone stimulate IGFBP-3 in rat osteoblasts. Am. J. Physiol. 267, E226–E233. 119. Conover, C. A. (1995). Insulin-like growth factor binding protein proteolysis in bone cell models. Prog. Growth Factor Res. 6, 301–309. 120. Kano, J., Sugimoto, T., Kanatani, M., Kuroki, Y., Tsukamoto, T., Fukase, M., and Chinara, K. (1994). Second messenger signaling of c-fos gene induction by parathyroid hormeon (PTH) and PTH-related peptide in osteoblastic osteosarcoma cells: its role in osteoblast proliferation and osteoclast-like cell formation. J. Cell. Physiol. 161, 358–366. 121. Nemere, I., and Larsson, D. (2002). Does PTH have a direct effect on intestine? J. Cell. Biochem. 86, 29–34. 122. Hruska, K. A., Blondin, J., Bass, R., et al. (1979). Effect of intact parathyroid hormone on hepatic glucose release in the dog. J. Clin. Invest. 64, 1016–1023. 123. Puschett, J. B., McGowan, J., Fragola, J., Chen, T. C., et al. (1984). Difference between 1-84 parathyroid hormone and the 1-34 fragment on renal tubular calcium transport in the dog. Miner. Electrolyte Metab. 10, 271–274.
Chapter 16 Parathyroid Hormone: Structure, Function and Dynamic Actions 124. Toromanoff, A., Ammann, P., and Riond, J. L. (1998). Early effects of short-term parathyroid hormone administration on bone mass, mineral content, and strength in female rats. Bone 22, 217–223. 125. Kishi, T., Hagino, H., Kishimoto, H., et al. (1998). Bone responses at various skeletal sites to human parathyroid hormone in ovariectomized rats: effects of long-term administration, withdrawal, and readministration. Bone 22, 515–522. 126. Okimoto, N., Tsurukami, H., Okazaki, Y., et al. (1998). Effects of a weekly injection of human parathyroid hormone (1-34) and withdrawal on bone mass, strength, and turnover in mature ovariectomized rats. Bone 22, 523–531. 127. Lane, N. E., Kimmel, D. B., Nilsson, M. H., Cohen, F. E., Newton, S., Nissenson, R. A., and Strewler, G. J. (1996). Bone-selective analogs of human PTH(1-34) increase bone formation in an ovariectomized rat model. J. Bone Miner. Res. 11, 614–625. 128. Ejersted, C., Oxlund, H., Eriksen, E. F., et al. (1998). Withdrawal of parathyroid hormone treatment causes rapid resorption of newly formed vertebral cancellous and endocortical bone in old rats. Bone 23, 43–52. 129. Chow, J. W. M., Fox, S., Jagger, C. J., et al. (1998). Role for parathyroid hormone in mechanical responsiveness of rat bone. Am. J. Physiol. Endocrin. Metab. 37, E146–E154. 130. Schipani, E., Lanske, B., Hunzelman, J., et al. (1997). Targeted expression of constitutively active receptors for parathyroid hormone and parathyroid hormone-related peptide delays endochondral bone formation and rescues mice that lack parathyroid hormone-related peptide. Proc. Natl. Acad. Sci. USA 94, 13689–13694. 131. Karaplis, A. C., Luz, A., Glowacki, J., and et al. (1994). Lethal skeletal dysplasia from targeted disruption of the parathyroid hormone-related peptide gene. Genes Development 8, 277–289. 132. Lanske, B., Karaplis, A. C., Lee, K., and et al. (1996). PTH/PTHrP receptor in early development and indian hedgehog-regulated bone growth. Science 273, 663–666. 133. Jobert, A-S., Zhang, P., Couvineau, A., and et al. (1998). Absence of functional receptors for parathyroid hormone and parathyroid hormone-related peptide in blomstrand chondrodysplasia. J. Clin. Invest. 102, 34–40. 134. Karaplis, A. C., He, B., Nguyen, M. T. A., and et al. (1998). Inactivating mutation in the human parathyroid hormone receptor type 1 gene in blomstrand chondrodysplasia. Endocrinology 139, 5255–5258. 135. Zhou, H., Iida-Klein, A., Lu, S. S., and et al. (2003). Anabolic action of parathyroid hormone on cortical and cancellous bone differs between axial and appendicular skeletal sites in mice. Bone 32, 513–520. 136. Theiler, R., Bischoff, H., Tyndall, A., and Stahelin, H. B. (1998). Elevated PTH levels in hypovitaminosis D are more rapidly suppressed
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
291
by the administration of 1,25dihydroxy-vitamin D3 than by vitamin D3. Int. J. Vitam. Nutr. Res. 68, 36–41. Berruti, A., Sperone, P., Fasolis, G., Torta, M., Fontana, D., Dogliotti, L., and Angeli, A. (1997). Pamidronate administration improves the secondary hyperparathyroidism due to “Bone Hunger Syndrome” in a patient with osteoblastic metastases from prostate cancer. Prostate 33, 252–255. Brahm, H., Ljunggren, O., Larsson, K., Lindh, E., and Ljunghall, S. (1994) Effects of infusion of parathyroid hormone and primary hyperparathyroidism on formation and breakdown of type I collagen. Calcif. Tissue Int. 55, 412–416. Ebeling, P. R., Jones, J. D., Burritt, M. F., Duerson, C. R., Lane, A. W., Hassager, C., Kumar, R., and Riggs, B. L. (1992). Skeletal responsiveness to endogenous parathyroid hormone in postmenopausal osteoporosis. J. Clin. Endocrinol. Metab. 75, 1033–1038. Ledger, G. A., Burritt, M. F., Kao, P. C., O’Fallon, W. M., Riggs, B. L. and Khosla, S. (1995). Role of parathyroid hormone in mediating nocturnal and age-related increases in bone resorption. J. Clin. Endocrinol. Metab. 80, 3304–3310. Schwietert, H. R., Groen, E. W., Sollie, F. A., and Jonkman, J. H. (1997). Single-dose subcutaneous administration of recombinant human parathyroid hormone [rhPTH(1-84)] in healthy postmenopausal volunteers. Clin. Pharmacol. Ther. 61, 360–376. Cosman, F., Morgan, D. C., Nieves, J. W., Shen, V., Luckey, M. M., Dempster, D. W., Lindsay, R., and Parisien, M. (1997). Resistance to bone resorbing effects of PTH in black women. J. Bone Miner. Res. 12, 958–966. Lindsay, R., Nieves, J., Formica, C., et al. (1997). Randomised controlled study of effect of parathyroid hormone on vertebral-bone mass and fracture incidence among postmenopausal women on oestrogen with osteoporosis. Lancet 350, 550–555. Kurland, E. S., Cosman, F., McMahon, D. J., Shen, V., Lindsay, R., Rosen, C. J., and Bilezikian, J. P. (2000). Changes in bone markers predict bone accural in osteoporotic men treated with parathyroid hormone. J. Bone Miner. Res. 85, 3069–3076. Neer, R. M., Arnaud, C. D., Zanchetta, J. R., and et al. (2001). Effect of parathyroid hormone (1–34) on fractures and bone mineral density in postmenopausal women with osteoporosis. New Engl. J. Med. 344, 1434–1441. Gallagher, J. C., Genant, H. K., Crans, G. G., Vargas, S. J., and Krege, J. H. (2005). Teriparatide reduces the fracture risk associated with increasing number and severity of osteoporotic fractures. J. Clin. Endocrinol. Metab. 90, 1583–1587. Black, D. M., Greenspan, S. L., Ensrud, K. E., Palmero, L., and et al. (2003). The effects of parathyroid hormone and alendronate alone or in combination in postmenopausal osteoporosis. New Engl. J. Med. 349, 1207–1215.
This Page Intentionally Left Blank
Chapter 17
Interaction of Parathyroid Hormone-related Peptide with the Skeleton David Goltzman
Calcium Research Laboratory, Department of Medicine, Royal Victoria Hospital, McGill University, Montreal, H3A 1A1, Canada
I. Abstract II. Introduction III. Molecular Biology and Mechanism of Action
IV. The Skeletal Actions of PTHrP V. Summary References
I. ABSTRACT
lineage, resulting in enhanced bone formation and also, indirectly, in augmented osteoclastic bone resorption; however, unlike PTH which serves postnatally to defend against a decrease in extracellular calcium, endogenous PTHrP appears to play an important anabolic role in the postnatal skeleton. Overall, these findings point to a critical role for PTHrP in normal skeletal development and in its maintenance. Further elucidation of the molecular role and interactions of PTHrP may therefore reveal new facets of the pathogenesis of a variety of metabolic bone diseases and potentially point to new directions for therapeutic interventions.
Parathyroid hormone-related peptide (PTHrP) was discovered as a mediator of hypercalcemia associated with malignancy but is now known to be expressed by a large number of normal fetal and adult studies. The aminoterminal region of PTHrP reveals limited but significant homology with parathyroid hormone (PTH), resulting in the interaction of the first 34–36 residues of either peptide with a single seven-transmembrane spanning G proteinlinked receptor termed the PTH/PTHrP receptor or the PTH-1 receptor. PTHrP plays a critical role in chondrocyte biology and in endochondral bone formation. Deficient PTHrP action in utero results in a distorted epiphyseal growth plate and marked skeletal deformity with ensuing neonatal death due to respiratory compromise. The complex processes resulting in normal endochondral bone development involve additional factors such as the hedgehogsignaling pathway with which PTHrP interacts. PTHrP, like PTH, binds to receptors on cells of the osteoblast Dynamics of Bone and Cartilage Metabolism
II. INTRODUCTION The association of malignancies and hypercalcemia was first reported in 1936 [1] and the production and secretion by tumors of a circulating factor causing hypercalcemia was postulated by Fuller Albright in 1941 [2]. 293
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
294
DAVID GOLTZMAN
The hypercalcemia of malignancy syndrome is characterized by hypercalcemia and hypophosphatemia, biochemical features that are usually indicative of oversecretion of parathyroid hormone (PTH), the major regulator of calcium homeostasis. Initial studies that searched for immunoreactive PTH within the tumors, however, were equivocal or negative. However, PTH-like activity in tumor extracts was evident in assays that measured renal glucose-6-phosphate dehydrogenase (G6PD) activity by a cytochemical technique [3] and adenylate cyclase activity in osteoblast cell lines and in renal membranes. The latter assay proved to be critical for monitoring the purification and the final isolation of the PTH-like peptide from human lung [4], breast [5], and kidney [6] cancer cell lines. The finding of structural similarity of the amino-terminal region of this peptide (PTHrP, for PTH-related peptide) with the corresponding domain of PTH, and the resultant capacity of both molecules to interact with the same G protein-coupled cell-surface receptor (PTH/PTHrP or PTH-1 receptor) via this region [7, 8], appear to account for the ability of circulating, tumor-derived PTHrP to inappropriately activate receptors in classic PTH target organs such as bone and kidney and to elicit PTH-like bioactivity. By the same mechanism, PTHrP may also enhance osteoclastic activity when released locally by tumor cells that have already metastasized to bone [9, 10]. Further studies, however, have demonstrated the importance of PTHrP, not only as an osteolytic factor in malignancy but also as a paracrine/ autocrine, and probably an intracrine regulator, of many
Figure 1
physiologic processes, most notably of bone and cartilage metabolism. The pivotal role of the PTHrP signaling pathway in endochondral bone development and in adult skeletal homeostasis will be the focus of this review.
III. MOLECULAR BIOLOGY AND MECHANISM OF ACTION A. Organization of the PTHrP Gene The human gene for PTHrP has been mapped to the short arm of chromosome 12 [11], whereas the PTH gene is assigned to human chromosome 11. An evolutionary relationship between the PTHrP and PTH genes has been suggested, because human chromosomes 11 and 12 are thought to have arisen by a tetraploidization event of a common ancestral chromosome. Indeed, the PTHrP and PTH genes and their respective gene clusters have been maintained as syntetic groups in human, rat, and mouse genomes [12]. Considerably more complex than the PTH gene, however, the human PTHrP gene comprises at least seven exons, spans more than 15 kb of genomic DNA (Fig. 1), and utilizes at least three distinct promoters, two containing a classical TATA box and one including a GC-rich region [13–16]. Exon 1, subdivided into Ia, Ib, and Ic, and exon II encode different 5′ untranslated regions. Exon III encodes the prepro coding region and exon IV encodes most of the
Organization of the human, rat, mouse and chicken PTHrP genes. Bold numbers depict numbering of exons. Nonbold numbers depict amino acid positions in the precursor (denoted by – symbols) or in the mature peptide. Arrows indicate promoters. Open boxes indicate 5′ untranslated regions, dark gray boxes denote coding exons and light gray boxes denote exons encoding 3′ untranslated sequences. Exons joined by solid Vs denote obligatory splicing whereas exons joined by broken Vs denote alternative splicing.
Chapter 17 Interaction of Parathyroid Hormone-related Peptide with the Skeleton
mature peptide sequence. At the end of exon IV, the splice site interrupts codon exon IV, encodes a stop codon and a 3′ noncoding sequence. Exon VI encodes 36 additional amino acids, a stop codon, and a second 3′ noncoding region. Exon VII encodes two extra amino acids, a stop codon, and a third noncoding region. Hence, by the use of alternative splicing of exons, a multitude of human PTHrP RNA transcripts can occur, giving rise to three different isoforms of 139, 141, and 173 amino acids in length which are identical as far as amino acid 139. In contrast with this complexity, the rat [17], mouse [18], and chicken [19] PTHrP genes are considerably simpler in organization and consist of four or five exons and only a single promoter, corresponding to the downstream promoter in the human gene. Overall, this simpler organization predicts that, in most species, a single promoter will be found and the 141-amino acids isoform of the peptide will predominate. In the mouse, the mature peptide is 139 amino acids rather than 141 amino acids because of a sixbase pair deletion (corresponding to amino acids 130 and 131 of human and rat protein). Moreover, the chicken gene can encode two isoforms, one of 139 amino acids arising by read-through of exon III into exon IV, and one of 141 amino acids, which is the predominant form, arising from splicing to exon V. Human, rat, mouse, dog [20], and chicken peptides are virtually identical in the amino terminal and midregion of the molecule but diverge beyond residue 112. This striking conservation throughout this large evolutionary range suggested early on that PTHrP is of essential biological importance.
295
increased PTHrP expression in the uterus, as does intrauterine occupancy in preterm myometrium [28], however the precise mechanism of these effects is unknown. Both stimulation and inhibition generally occur predominantly at the transcriptional level. 2. Post-translational Processing
With the availability of the PTHrP cDNA-based amino acid sequence, it became evident that the mature form of the protein is preceded by a 36-residue prepro sequence. This sequence has all the classic features of a signal peptide that would target the nascent protein to the rough endoplasmic reticulum. While the precise site of cleavage by the signal peptidase has yet to be determined, it is likely that it occurs at position −9, thereby generating a nascent propeptide extending from position –8. The evidence that the pro sequence is subsequently cleaved to generate the mature form of the protein at position Ala +1 is based on (1) direct sequencing of the secreted PTHrP form by three independent laboratories, (2) the presence of an endoproteolytic site for prohormone convertases of the furin and PC1/3 family proximal to it [29], and (3) the homology of this region with that of PTH. Closer examination of the mature PTHrP sequence reveals additional clusters of basic amino acids that could potentially serve as endoproteolytic processing sites (Fig. 2). Hence, cleavage appears to occur carboxy-terminal to arginine at position 37, most likely yielding the secretory form, PTHrP (1–36). A midregion fragment that begins at amino acid 38 and terminates in the 80–100 region has also been identified, both within cells and outside cells following
B. Regulation of PTHrP Production In contrast to PTH, which is synthesized almost exclusively in the parathyroid glands, PTHrP is expressed in many fetal and adult cells and tissues. 1. Regulation of Gene Expression
Several factors have been shown to regulate PTHrP gene expression in a variety of normal tissues or cells, at least in part, at the transcriptional level. In most cells, including for example cultured keratinocytes, breast epithelial cells, and smooth muscle cells, growth factors, such as EGF [21], IGF1 [22], and TGFb, stimulate PTHrP gene expression [23], whereas 1,25 (OH)2 vitamin D3 and low calcemic analogs of 1,25(OH)2 vitamin D3 inhibit this production [21, 24]. Other steroids including glucocorticoids, androgens [25] and estrogens [26] have also been shown to inhibit PTHrP expression in tissues expressing their cognate receptors. PTHP mRNA in lactating mammary tissue is stimulated by prolactin [27], and estrogen administration causes
Figure 2 PTHrP as a polyhormone. Distinct biological activities have been associated with the different regions of the peptide which are depicted. Numbers within the molecule indicate the sites where proteolytic processing is believed to occur. Numbers at the carboxyl terminal region (139, 141, 173) denote alternative carboxyl terminal amino acids.
296
DAVID GOLTZMAN
secretion [30]. An additional carboxy-terminal PTHrP fragment may exist that encompasses the highly conserved pentapeptide 107–111. It would seem, therefore, that PTHrP could function as a polyhormone precursor of a number of biologically active fragments with distinct biological actions. 3. Regulation of PTHrP Secretion
Both constitutive and regulated secretion of PTHrP have been described. Elevated calcium concentrations have been reported to stimulate PTHrP secretion via the calciumsensing receptor (CaSR) in prostate, breast, and Leydig cell cancer cell lines. Thus, calcium-activated CaSR may transactivate the EGF receptor by activation of a matrix metalloproteinase, which leads to cleavage of proheparin-binding EGF to heparin-binding EGF. This then leads to phosphorylation of EGF receptor kinase (ERK) 1/2 and resultant PTHrP secretion [31]. In contrast, in lactating mammary epithelium, low calcium levels have been reported to stimulate PTHrP secretion via the CaSR [32].
C. Receptors and Signal Transduction Pathways In view of the extensive post-translational processing that PTHrP undergoes, it would not be surprising if multiple receptors for the various forms of the protein exist. These can be divided primarily into four categories that recognize determinants in (1) the animo-terminal region of the protein, (2) the midregion form, (3) the nuclear localization sequence, and (4) the carboxyl-terminal form. The amino-terminal domain of PTHrP, because of limited but important homology with eight of the first 13 residues in PTH, binds to the classical PTH-1 receptor (PTHR-1) and activates both the adenylyl cyclase/protein kinase A pathway as well as the calcium/inositol phosphate/ protein kinase C pathway (Fig. 3). More recently, evidence for PTHR-1 stimulation of mitogen-activated protein kinase (MAPK) has also been reported [33, 34] and the sodium/hydrogen exchanger regulatory factor (NHERF) has been reported to interact with the carboxyl-terminal intracellular tail of the PTHR-1 to increase the formation of inositol triphosphate while reducing cyclic AMP formation [35]. The interaction between PTHrP and PTHR-1 is sufficient to confer functions of PTHrP which mimic those of PTH in bone and kidney. In fact, most of the well-characterized actions of PTHrP appear to be mediated by this receptor for the NH2-terminal domain of PTHrP. The expression of PTHR-1 has been observed in a wide variety of tissues that might well be PTHrP targets, although expression of the receptor seems highest in bone, cartilage, and kidney.
Figure 3
Interaction of PTHrP with the PTHR-1. The amino terminal domain of PTHrP (and PTH) binds to the seven-transmembrane spanning receptor, causing the guanyl nucleotide binding proteins Gsα and Gqα to activate adenylyl cyclase and phospholipase C respectively. Therefore cAMP or inositol triphosphate (IP3) and diacylglycerol (DAG) are produced. Cyclic AMP and DAG will stimulate protein kinase A (PKA) and protein kinase C (PKC) respectively and IP3 will increase levels of intracellular calcium (CA++). The βγ subunits of Gs and Gq, as well as PKC may also participate in increasing levels of mitogen-activated protein kinase (MAPK). The sodium/hydrogen exchanger regulatory factor (NHERF) appears to associate with the PTHR-1 to increase IP3 and may also reduce adenylyl cyclase activity by interacting with the Gi/o protein.
The existence of a “receptor” that recognizes the N-terminal domain of PTHrP but is distinct from the classic type 1 receptor has also been suggested [36, 37]. This high-capacity, low-affinity receptor was identified in keratinocytes and pancreatic β-cells and signals only through the protein kinase C pathway. Cloning and molecular characterization of this receptor, however, have not been accomplished. Even less is known about a putative receptor that binds the midregion form of PTHrP. The existence of such a receptor, likely linked to the protein kinase C pathway, is circumstantial and is based primarily on the finding that residues 67–86 of PTHrP are critical for the control of placental calcium transport [38]. While the receptor has not yet been identified, support for its existence has come from work with the PTHrP gene knockout mouse in which placental calcium transport is severely impaired and can only be restored by administration of PTHrP(1-86) and PTHrP(67-86) peptides. PTHrP(1-34) and PTH(1-84) have no effect [39]. Amino acids 87–107 of the mature form of PTHrP encode a nuclear localization signal (NLS) consisting of two basic clusters separated by a spacer region, analogous to the prototypical nucleoplasmin NLS [40, 41]. Studies have shown that elimination of the NLS from prepro
Chapter 17 Interaction of Parathyroid Hormone-related Peptide with the Skeleton
PTHrP effectively abolished intranuclear localization of the recombinant protein. Moreover, the NLS alone was capable of translocating a heterologous cytoplasmic protein, β-galactosidase, to the nucleolus when expressed as a fusion protein in COS-7 cells. Selective protein import into the nucleus proceeds through the nuclear pore complex and requires the interaction between the NLS and an importin–beta1 receptor complex [42]. Receptors for the carboxyl-terminal form of PTHrP have been inferred from studies reporting that picomolar concentrations of this peptide can profoundly inhibit resorption in an isolated osteoclast bone resorption assay [43, 44]. In addition, the C-terminal peptide PTHrP(107139) has been shown to elicit a rise in cytosolic calcium in cultured hypocampal neurons [45]. It is possible, therefore, that novel receptors for carboxyl-terminal fragments might be expected in osteoclasts and neurons. The cloning and characterization of these receptors will be necessary in order to establish the physiological relevance associated with their activation.
D. Endocrine, Autocrine/Paracrine, and Intracrine Actions Although PTHrP was initially discovered through its “endocrine” effects (i.e., it is released from tumors into the circulation and it acts at distant sites such as the skeleton and kidneys), this mechanism of action is generally considered to be the exception rather than the rule. In fact, under normal circumstances, the only other endocrine effect ascribed to PTHrP is its influence on placental calcium transport, presumably following release from the fetal parathyroids. This scenario notwithstanding, PTHrP, unlike PTH, does not circulate in appreciable amounts in normal subjects [46]; rather, it is thought to exert its biological effects locally in a paracrine/autocrine fashion and in this manner may influence growth, differentiation, and differentiated function in a variety of cell types. For example, studies in cultured keratinocytes [40, 47] demonstrated a role for PTHrP in inhibiting growth and supporting differentiation, and overexpression of PTHrP in breast myoepithelial cells decreased ductal proliferation and branching morphogenesis [48, 49]. In some tissues such as mammary tissue, PTHrP is an epithelial-derived factor that acts in a paracrine manner on mesenchymal cells expressing the PTH-1 receptor to influence mammary gland development [49]. Similarly, PTHrP derived from the endothelium can act on vascular smooth muscle cells to regulate vascular tone and blood pressure. Alternatively, PTHrP can be secreted by vascular smooth muscle cells, and in turn, feed back in
297
an autocrine mode to alter the contractility of these cells. Likewise, keratinocytes, pancreatic β cells, and uterine smooth muscle cells can both secrete and respond to PTHrP. Furthermore, PTHrP has also been reported to alter differentiated functions, such as smooth muscle relaxation and neurotransmission. In addition to the autocrine/paracrine effects, experimental evidence has been presented suggesting that PTHrP may also have an intracrine site of action. This conclusion was first based on the following observations: (1) endogenous PTHrP localizes to the nucleus of murine bone cells in vitro as well as in situ [46], (2) immunogold labeling for the endogenous peptide in situ was observed over the dense fibrillar component of nucleoli, a subnucleolar structure thought to represent the major sight of transcription of gene coding for rRNA, and (3) the nuclear actions of PTHrP have been subsequently confirmed in a human keratinocyte cell line (HaCaT) [50] and in cultured vascular smooth muscle cells [51]. In view of the fact that PTHrP contains a leader sequence, which directs the nascent peptide to the secretory apparatus of the cell in which it is biosynthesized, it is uncertain how such a peptide might be trafficked to the cytoplasm and have access to the nucleus. Three possibilities have been described. One is that PTHrP undergoes reverse transport from the endoplasmic reticulum to the cytoplasm, with at least some of PTHrP becoming subject to ubiquitination and proteosomal degradation [52]. A second is that, after secretion, PTHrP is internalized by the PTHR-1 [53] which itself has been reported to contain a nuclear localization signal, or by a distinct membrane receptor which has not yet been characterized [54]. In either case, how PTHrP would escape from endosomes subsequent to receptor-mediated endocytosis has not yet been determined. A third and perhaps the most likely possibility, is that, as has been demonstrated, some of the PTHrP is translated from an alternative non-AUG start site downstream from the initiator methionine, which excludes the leader sequence but allows the retention of the peptide, including the nuclear localization signal within the cell cytoplasm [55]. Studies in vascular smooth muscle cells [56] and keratinocytes [57] have shown that PTHrP expression occurs in a cell cycle-dependent manner with higher expression occurring in the G2 and M phases of the cycle. The cell cycle-dependent localization of PTHrP is regulated by the activity of the cyclin-dependent kinases (cdk) cdc2 and cdk2, which phosphorylate PTHrP at threonine85 within a consensus cdc2/cdk2 site. Phosphorylation increases as cells progress from G1 to G2 and M of the cell cycle and leads to decreased nuclear entry, perhaps by enhancing binding to a cytoplasmic retention factor.
298
DAVID GOLTZMAN
PTHrP appears to bind with high affinity to importin b1 and the GDP-bound protein, Ran. After translocation across the nuclear envelope, GTP-bound Ran may release PTHrP into the nucleus where it may act in the nucleolus to bind RNA [58], which may possibly regulate mRNA processing or mRNA transport. Functional effects of intranuclear PTHrP may then be to delay apoptosis, as shown in chondrocytic cells in vitro and to increase cell proliferation in vascular smooth muscle in vitro [51]. Further studies are, however, required to demonstrate the effects of nuclear localization of PTHrP in vivo. It would appear, therefore, that PTHrP influences a spectrum of cellular function by exerting its effects at, near, or in its site of synthesis.
IV. THE SKELETAL ACTIONS OF PTHrP A. PTHrP Signaling and Chondrocyte Biology 1. Chondrocyte Proliferation and Differentiation
Endochondral ossification is a complex, multistep process involving the formation of cartilaginous skeleton from aggregated mesenchymal cells and its subsequent replacement by bone. The cellular and molecular events that regulate the highly ordered progression of chondrocytes within the growth plate through the stages of proliferation and differentiation must be under precise spatial and temporal control. Ultimately, these events determine the extent and rate of skeletal growth. In the cartilaginous growth plate, PTHrP is expressed by perichondral, proliferating, and hypertrophic chondrocytes, and the PTHR-1 is expressed in proliferating and hypertrophic chondrocytes. The generation of mice homozygous for a disrupted PTHrP gene provided the first direct evidence of a physiological role for this protein in chondrocyte biology [59, 60]. PTHrP-null mice die in the immediate postnatal period, likely from respiratory failure in association with widespread abnormalities of endochondral bone development. Characterized by diminished proliferation, accelerated differentiation, and premature apoptotic death of chondrocytes, this form of ostoechondrodysplasia results in the untimely and rapid maturation of the skeleton [59, 60]. The critical role of PTHrP as an inhibitor of the chondrocyte differentiation program [61] has been further substantiated by the targeted overexpression of PTHrP in chondrocytes by means of the mouse collagen type II promoter [62]. This targeting induces a novel form of chondrodysplasia that is the antithesis of the PTHrP-null
phenotype and is characterized by a delay in endochondral ossification so profound that mice are born with cartilaginous endochondral skeletons. Therefore, PTHrP enhances chondrocyte proliferation and negatively regulates the switch from a proliferative immature chondrocyte to a post proliferative mature, hypertrophic chondrocyte. PTHrP blocks chondrocyte differentiation via the cAMP pathway and this process may involve phosphorylation and activation of the transcription factor SOX9, a master regulator of early chondrogenesis. 2. PTHrP and Chondrocyte Apoptosis
Hypertrophic chondrocytes in the growth plate are thought to undergo apoptosis immediately prior to ossification [63, 64] and, therefore, represent the terminal stage of differentiation in the chondrogenic lineage. PTHrP expression might be expected to delay, or even prevent, the progression to terminal differentiation and eventual programmed cell death. Studies with the chondrocytic cell line CFK2 overexpressing PTHrP demonstrated enhanced cell survival under conditions that prompted is apoptotic death [41]. In addition, quantitative analysis of the growth plate of PTHrP-null mice revealed significantly more apoptotic chondrocytes near the chondro-osseous junction compared to wild-type littermates [63]. Thus, PTHrP influences not only chondrocyte proliferation and differentiation, but also programmed cell death. In several cell types, apoptosis is regulated by the ratio of expression of the cell death inhibitor, Bcl-2, and the cell death inducer, Bax. Bcl-2 is expressed in growth plate chondrocytes in a pattern similar to that of PTHrP [65], with the highest levels detected in late proliferative and prehypertrophic chondrocyte. Both in vitro and in transgenic mice, PTHrP overexpression causes a marked increase in Bcl-2 with no detectable change in Bax levels. A shift of the Bcl-2 downstream of the Bcl-2/Bax ratio in favor of Bcl-2 delays terminal differentiation, prolongs chondrocyte survival, and leads to the accumulation of cells in their prehypertrophic stage. These observations place Bcl-2 downstream of PTHrP in the pathway that controls chondrocyte maturation and endochondral skeletal development. 3. INTERACTION OF PTHrP WITH OTHER CARTILAGE REGULATORY FACTORS
Indian hedgehog (Ihh), a member of a family of proteins important for embryonic patterning, is highly expressed in the prehypertrophic zone between proliferating and hypertrophic chondrocytes and in the upper hypertrophic chondrocytes [66, 67]. Ihh seems, in growth plate cartilage, to be necessary and sufficient for PTHrP gene expression (Fig. 4).
Chapter 17 Interaction of Parathyroid Hormone-related Peptide with the Skeleton
299
and is essential for vascular invasion at the chondro-osseous junction. Runx2 can activate the Ihh/PTHrP pathway but at present it is unclear whether PTHrP alters Runx2 expression in chondrocytes. 4. Defining the Role of the PTH-1 Receptor
Figure 4 Effect of PTHrP and other factors on chondrocyte proliferation and differentiation in the growth plate. Stimulatory effects of the factors are shown by the solid arrows (t) and inhibitory effect by the solid symbol, . The effects of Indian hedgehog (Ihh), bone morphogenetic protein (BMP), the transcription factor Runx2, and of fibroblast growth factor (FGF) on each other are depicted by hatched arrows ( ) when stimulatory and by the hatched symbol, , when inhibitory.
Ihh stimulation of PTHrP therefore increases chondrocyte proliferation and decreases chondrocyte hypertrophy. This therefore reduces Ihh production, resulting in a negative feedback loop. Ihh, independently from PTHrP, also stimulates chondrocyte proliferation but may also enhance chondrocyte differentiation by blocking the effects of PTHrP [68]. The fibroblast growth factor (FGF) receptor, FGFR3, a tyrosine kinase receptor, is expressed in proliferating chondrocytes and can inhibit Ihh expression, thereby indirectly reducing PTHrP expression. Activation of FGFR3 inhibits chondrocyte proliferation independent of its effect on Ihh but may also increase chondrocyte differentiation, apoptosis, and vascular invasion at the chondro-osseous junction. FGF18 appears to be a major ligand for FGFR3 [69]. Studies with double mutants lacking both FGFR3 and PTHrP indicate that the stimulatory effect of PTHrP on chondrocyte proliferation supercedes the inhibitory effect of FGFR3. In contrast to FGF which inhibits Ihh, bone morphogenetic proteins (BMPs) form a positive autoregulatory feedback loop with Ihh, although a role for BMPs in stimulating proliferation and inhibiting differentiation independent of the Ihh/PTHrP axis has been suggested [66]. The transcription factor Runx2 (also known as Cbfa1) is expressed in early hypertrophic chondrocytes and enhances chondrocyte maturation in co-operation with Runx3 and/or core-binding factor b (CBFb), a co-transcription factor that forms heterodimers with Runx proteins. Runx2 also activates vascular endothelial growth factor (VEGF)
The PTH-1 receptor appears to mediate most PTHrP actions in the cartilaginous growth plate. Thus, the generation of PTH-1 receptor knockout mice provided the first evidence that this receptor mediates the cartilaginous effects of PTHrP [67]. The small number of homozygous mice that do survive to the peripartum period exhibit a phenotype similar to that observed in the PTHrP-negative mutants, although they are proportionally smaller. Skeletal alterations are characterized by accelerated differentiation of chondrocytes leading to a severe and lethal form of osteochondrodysplasia. In contrast, targeted overexpression of constitutively active PTH-1 receptor to the growth plate delays endochondral bone formation [70]. Of note, however, is the observation that most PTHR1-negative animals exhibit a more severe phenotype that the PTHrP-null mice characterized by early embryonic lethality (embryonic day E14.5). It was originally speculated that PTHrP synthesized by maternal decidual cells could complement the absence of fetal PTHrP but not that of its receptor. Alternatively, the more severe phenotype may implicate the presence of another member of the PTH/PTHrP ligand family that interacts with the common receptor. The existence of such a protein, however, remains speculative at present. Perhaps the most intriguing explanation for the early demise of these animals implicates a possible dual mechanism of action of PTHrP on cellular function, i.e., the amino-terminal end of the protein acting on the cell-surface receptor and the more carboxyl region acting within the nucleus. In the absence of the amino-terminal receptor, the unopposed nuclear effects of PTHrP would lead to severe imbalance of cellular proliferation and differentiation and hence to early lethality of the receptor-negative mutants. In contrast, ablation of the ligand would eliminate both membrane receptor- and nuclear-mediated activities, leading to a more “coordinated” impairment and a less severe phenotype. At present, however, it is reasonable to conclude that the small size and early death of the PTH-1 receptor-null mice remains unexplained. The critical role of the PTHR-1 in endochondral bone formation is emphasized, however, by the discovery that two human chondro-osseous dysplasias, Blomstrand’s lethal chondrodysplasia (BLC) [71, 72] and Jansen’s metaphyseal chondrodysplasia (JMC) are caused by receptor mutations. BLC appears to be autosomal recessive, is caused by inactivating mutations of PTHR-1, and is characterized
300
DAVID GOLTZMAN
by prenatal lethality, premature and abnormal bone mineralization and ossification, and shortened limbs. The proliferative zone in the mutant growth plate is virtually absent and endochondral bone formation is markedly advanced simulating the phenotype of the PTHR-1 knockout mice. JMC is a rare autosomal dominant disorder caused by an activating mutation of the PTHR-1 leading to ligand-independent (constitutive) activation of the receptor. It is characterized by short-limbed dwarfism secondary to severe abnormalities of the growth plate and by hypercalcemia [73–75]. Although laboratory findings of JMC are reminiscent of primary hyperparathyroidism, parathyroid glands appear normal and circulating PTH levels are normal or undetectable. A heterozygous missense mutation of PTHR-1, which causes constitutive activation of the receptor, has also been identified in enchondromatosis, characterized by multiple benign cartilage tumors [76].
B. PTHrP and Osteoblast Biology 1. PTHrP Haploinsufficiency Leads to Osteopenia
PTHrP and PTH-1 receptors are expressed in cells of the osteogenic lineage [77–83]. Although PTH is essential for normal trabecular bone formation in the fetus [84] and therefore plays an anabolic role in utero, accumulating evidence indicates that PTH subserves a role postnatally primarily in defending against a decline in blood calcium, whereas PTHrP assumes a role in promoting bone formation in place of PTH as the animal matures (Fig. 5). Thus, heterozygous PTHrP-null mice, while phenotypically normal at birth, by 3 months of age exhibit a form of osteopenia characterized by a marked decrease in trabecular thickness and connectivity [85]. Moreover, their bone marrow contains an abnormally high number of adipocytes. Since the same pluripotent stromal cells in the bone marrow compartment can give rise to adipocytes and osteoprogenitor cells [86], the increased number of adipocytes and osteopenia in these mice could be attributed to altered stem cell differentiation as a consequence of PTHrP haploinsufficiency. Furthermore increased apoptotic osteoblastic cells are also observed. Taken together, these findings, and a variety of in vitro studies, suggest that PTHrP is important for the orderly commitment of pluripotential bone marrow stromal cells toward the osteogenic lineage and for their subsequent maturation and survival. In these animals, a physiological concentration of PTH is unable to compensate for PTHrP haploinsufficiency. One possible explanation is that a defect in skeletal progenitor cells was caused by PTHrP deficiency in utero and was
Figure 5 Effects of PTHrP and of PTH on the cartilaginous growth plate and on trabecular bone in the fetus and postnatally. In the fetus and postnatally, PTHrP acts as a paracrine factor to increase proliferation and reduce differentiation in the growth plate. Postnatally, PTHrP as a paracrine factor is anabolic for trabecular bone. PTH, as a hormone, is anabolic for trabecular bone in the fetus. As its main role post natally, PTH functions to preserve calcium homeostasis by producing net bone resorption. Circles depict chondrocytes and black areas denote trabecular bone.
only manifest in the postnatal state. Alternatively, domains of PTHrP not shared with the PTH molecule may subserve unique functions that are important for normal adult skeletal development. Further evidence for the anabolic boneforming effect of PTHrP came from crosses of mice deficient in PTH due to targeted deletion of the PTH gene, with PTHrP heterozygous mice lacking one allele encoding PTHrP [87]. PTH knockout mice have increased trabecular bone volume with diminished bone turnover consistent with the primary role of PTH in resorbing bone in order to maintain calcium homeostasis. Nevertheless, introducing PTHrP haploinsufficiency in these mice reduced trabecular bone of the PTH knockout mice to below wild-type levels by decreasing osteoprogenitor cell recruitment and enhancing osteoblast apoptosis and thereby diminishing bone formation. Consequently PTHrP plays an important bone-forming role in the post-natal state. More recent studies employing a cell-specific knockout model, in which osteoblastic PTHrP was specifically deleted, have demonstrated that osteoblastic PTHrP per se plays a critical role in enhancing osteoblastic bone formation. 2. PTHrP and the Bone Remodeling Process
In common with PTH, PTHrP in the postnatal skeleton appears to regulate both bone formation and bone resorption and both actions occur after binding to cells of the osteoblast phenotype [88]. Anabolic effects appear to predominate when exposure of skeletal cells to PTHrP is
301
Chapter 17 Interaction of Parathyroid Hormone-related Peptide with the Skeleton
intermittent, whereas continuous exposure appears to elicit a skeletal catabolic effect [89]. The underlying mechanism for this pattern is unclear at present. The anabolic effects of the NH2-terminal region of both PTHrP and PTH occur via their action on the PTH-1 receptor, however, at least some of the effects of PTHrP and PTH on bone formation may be partly mediated through cytokines or growth factors such as IGF-1 [90]. The effects of PTHrP (and PTH) on osteoclastic bone resorption also appear to be mediated by interaction with PTH-1 receptors. Multinucleated osteoclasts are derived by differentiation of mononuclear precursors presumably of the monocyte/macrophage lineage. The interaction of PTHrP with its receptor on the osteoblast appears to increase soluble or cell-bound cytokines, most important of which is the receptor activator of nuclear factor kappa beta ligand (RANKL) (Fig. 6). Simultaneously, PTHrP (and PTH) can reduce concentrations of the soluble RANKL antagonist, osteoprotegerin (OPG). The dual effect thereby facilitates the capacity of RANKL to bind to its cognate receptor on osteoclastic precursors and multinucleated mature osteoclasts to enhance bone resorption. Interestingly, reduction in osteoblastic PTHrP in conditional knockout experiments not only reduced bone formation but even more dramatically reduced bone resorption, thus illustrating a role for osteoblastic PTHrP as a “coupling” factor. Several important steroid modulators may influence the action of PTHrP and of PTH on bone anabolism and catabolism. Thus, glucocorticoids may increase PTH-1 mRNA receptor expression [90], PTH binding, and PTH-stimulated camp production, effects which could lead to a net increase in PTHrP-mediated bone resorption. Estrogen, on the
other hand, may inhibit PTHrP-and PTH-stimulated adenylate cyclase stimulation in osteoblasts [91, 92], and estrogen deficiency in postmenopausal women leads to increased skeletal catabolic effects of continuous PTH administration [93]. It is important to note that the capacity either of circulating PTH or of circulating or locally released PTHrP to exert anabolic or catabolic effects will, most likely, be determined not only by ambient levels of these two peptides but also by the state of occupancy of the receptors by either of these ligands, which may in turn alter receptor internalization and thus the efficacy of the peptides. 3. Potential Therapeutic Role in Osteoporosis
The anabolic effects of intermittent PTH administration on bone and its therapeutic potential in osteoporosis have been extensively explored [94]. With the recognition that PTHrP is the endogenous ligand for the PTH/PTHrP receptor in osteoblasts, its use as an anabolic agent has also been investigated. Indeed the anabolic action of intermittently administered PTHrP(1-36) has now been documented in women with post menopausal osteoporosis [95]. Further studies on the mechanism of PTHrP and PTH as skeletal anabolic agents are clearly needed in order to produce additional agents, which may have improved efficacy profiles and routes of administration.
V. SUMMARY The discovery of PTHrP has led to improved understanding of the pathogenesis of hypercalcemia of malignancy
Figure 6 Effect of PTHrP on osteoclastic bone resorption. PTHrP interacts with osteoblastic stromal cells to increase expression of receptor activator of nuclear factor kappa b ligand (RANKL) and to decrease expression of the soluble decoy receptor osteoprotegerin (OPG). RANKL can activate its receptor RANK which can then increase the production of active osteoclasts from hematopoietic stem cells, resulting in increased bone resorption.
302
DAVID GOLTZMAN
and may also be relevant for understanding malignancyinduced bone resorption even in the absence of hypercalcemia. Unexpectedly, and of equal importance, this humoral factor has been found to play a role in a number of normal developmental processes and in the differentiated function of a variety of normal tissues. Most notable is its crucial role in the development of the cartilaginous growth plate as demonstrated by gene disruption technology. Analyses of knockout mice have also emphasized the function of PTHrP in normal osteoblastic bone formation. Powerful new in vitro and in vivo molecular technologies are also providing insights into detailed intracellular signaling mechanisms of PTHrP and its interaction with other regulatory factors. Continued advances in this field should provide important new understanding of the pathophysiology of osteochondrodysplasias, osteopetrosis, hyperparathyroidism, and osteoporosis and should facilitate the development of new therapeutic tools for these disorders.
Acknowledgments This work on PTHrP was supported by the Canadian Institutes of Health Research and by the National Cancer Institute of Canada.
References 1. Gutman, A. B., Tyson, T. L., and Gutman E. B. (1936). Inorganic phosphorus and phosphatase activity in hyperparathyroidism, Paget’s disease, multiple myeloma and neoplastic disease of the bones. Arch. Intern. Med. 57, 379–413. 2. Albright, F. (1941). Case records of the Massachusetts General Hospital: Case 27461. N. Engl. J. Med. 225, 789–791. 3. Goltzman, D., Stewart, A. F., and Broadus, A. E. (1981). Malignancyassociated hypercalcemia: Evaluation with a sytochemical bioassay for parathyroid hormone. J. Clin. Endocrinol. Metab. 53, 899–904. 4. Moseley, J. M., Kubota, M., Diefenbach-Jagger, H., Wettenhall, R. E. H., Kemp, B. E., Suva, L. J., Rodda, C. P., Ebeling, P. R., Hudson, P. J., Zajac, J. D., and Martin, T. J. (1987). Parathyroid hormone-related protein purified from a human lung cancer cell line. Proc. Natl Acad. Sci. USA. 84, 5048–5052. 5. Stewart, A. F., Wu, T., Goumas, D., Burtis, W. J., and Broadus, A. E. (1987). N-terminal amino acid sequence of two novel tumor-derived adenylate cyclase-stimulating proteins: Identification of parathyroid hormone-like and parathyroid hormone-unlike domains. Biochem. Biophys. Res. Commun. 146, 672–678. 6. Strewler, G. J., Ster, P. H., Jacobs, J. W., Eveloff, L., Klein, R. F., Leung, S. C., Rosenblatt, M., and Nissenson, R. A. (1987). Parathyroid hormone-like protein from human renal carcinoma cells. Structural and functional homology with parathyroid hormone. J. Clin. Invest. 80, 1803–1807. 7. Juppner, H., Abou-Samra, A-B., Freeman, M., Kong, X.-F., Schipani, E., Richards, J., Kolakowski, L. F., Jr., Hock, J., Potts, J. T., Jr., Kronenberg, H. M., and Segre, G. V. (1991). A G protein-linked receptor for parathyroid hormone and parathyroid hormone-related peptide. Science 254, 1024–1026.
8. Abou-Samra, A.-B., Juppner, H., Force, T., Freeman, M. W., Kong, J. T., Jr., Kronenberg, H. M., and Segre, G. V. (1992). Expression cloning of a common receptor for parathyroid hormone and parathyroid hormone-related peptide from rat osteoblast-like cells: A single receptor stimulates intracellular accumulation of both camp and inositol trisphosphates and increases intracellular free calcium. Proc. Natl, Acad. Sci. USA. 89, 2732–2736. 9. Guise, T. A., Yin, J. J., Taylor, S. D., Kumagai, Y., Dallas, M., Boyce, B. F., Yoneda, T., and Mundy, G. R. (1996). Evidence for a causal role of parathyroid hormone-related peptide protein in the pathogenesis of human breast cancer-mediated osteolysis. J. Clin. Invest. 98, 1544–1549. 10. Rabbani, S. A., Gladu, J., Harakidas, P., Jamison, B., and Goltzman, D. (1999). Over-production of parathyroid hormone-related peptide results in increased osteolytic skeletal metastasis by prostate cancer cells in vivo. Int. J. Cancer. 80, 257–264. 11. Mangin, M., Webb, A. C., Dreyer, B, E., Posillico, J, T., Ikeda, K., Weir, E. C., Stewart, A. F., Bander, N. H., Milstone, L., Barton, D. E., Francke, U., and Broadus, A. E. (1988). Identification of a cDNA encoding a parathyroid hormone-like peptide from a human tumor associated with hypercalcemia of malignancy. Proc. Natl Acad. Sci. USA. 85, 597. 12. Hendy, G., Sakaguchi, A. Y., Yasuda, T., Weber, D. K., Wang, L.-M., Yoshida, M. C., Banville, D., and Goltzman, D. (1988). Gene for parathyroid hormone-like peptide is on rat chromosome 2. Biochem. Biophys. Res. Commun. 157, 558–562. 13. Mangin, M., Ikeda, K., Dreyer, B. E., and Broadus, A. E. (1989). Isolation and characterization of the human parathyroid hormone-like peptide gene. Proc. Natl Acad. Sci. USA. 86, 2408–2412. 14. Yasuda, T., Banville, D., Hendy, G. N., and Goltzman, D. (1989). Characterization of the human parathyroid hormone-like peptide gene. J. Biol. Chem. 264, 7720–7725. 15. Mangin, M., Ikeda, K., Dreyer, B. E., and Broadus. A. E. (1990). Identification of an up-stream promoter of the human parathyroid hormone-related peptide gene. Mol. Endocrinology 4, 851–858. 16. Vasavada, R. C., Wysolmerski, J. J., Broadus, A. E., and Philbrick, W. M. (1993). Identification and characterization of a GC-rich promoter of the human parathyroid hormone-related peptide gene. Mol. Endocrinology 7, 273–283. 17. Karaplis, A. C., Yasuda, T., Hendy, G. N., Goltzman, D., and Banville, D. (1990). Gene encoding parathyroid hormone-like peptide: Nucleotide sequences of the rat gene and comparison with the human homologue. Mol. Endocrinol. 4, 441–446. 18. Mangin, M., Ikeda, K., and Broadus, A. E. (1990). Structure of the mouse gene encoding parathyroid hormone-related peptide. Gene 95, 195–202. 19. Thiede, M., and Rutledge, S. J. (1990). Nucleotide sequences of a parathyroid hormone-related peptide expressed by the 10 day chicken embryo. Nucleic Acids Res. 18, 3062. 20. Rosol, T. J., Steinmeyer, C. L., McCauley, L. K., Grone, A., DeWille, J. W., and Capen, C. C. (1995). Sequence of the cDNAs encoding canine parathyroid hormone-related protein and parathyroid hormone. Gene 160, 241–243. 21. Kremer, R., Karaplis, A. C., Henderson, J., Gulliver, W., Banville, D., Hendy, H. N., and Goltzman, D. (1991). Regulation of parathyroid-like peptide in cultured normal human keratinocytes: Effect of growth factors and 1,25 dihydroxyvitamin D3 on gene expression and secretion. J. Clin. Invest. 87, 884–893. 22. Sebag, M., Henderson, J., Goltzman, D., and Kremer, R. (1994). Regulation of parathyroid hormone-related peptide production in normal human mammary epithelial cells in vitro. Am. J. Physiol. 267(3 Pt 1), C723–730. 23. Akilu, F., Park, M., Goltzman, D., and Rabbani, S. A. (1997). Induction of parathyroid hormone-related peptide by the Ras oncogene: Role of Ras farnesylation inhibitors as potential therapeutic agents for hypercalcemia of malignancy. Cancer Res. 57, 4515–4522.
Chapter 17 Interaction of Parathyroid Hormone-related Peptide with the Skeleton 24. Haq, M., Kremer, R., Goltzman, D., and Rabbani, S. A. (1993). A vitamin D analogue (EB1089) inhibits parathyroid hormone related protein production and prevents the development of malignancy-associated hypercalcemia in vivo. I. Clin. Invest. 911, 2416–2422. 25. Pizzi, H., Gladu, J., Carpio, L., Miao, D., Goltzman, D., and Rabbani S. A. (2003). Androgen regulation of parathyroid hormone-related peptide production in human prostate cancer cells. Endocrinology 144, 858–867. 26. Rabbani, S. A., Khalili, P., Arakelian, A., Pizzi, H., Chen, G., and Goltzman, D. (2005). Regulation of parathyroid hormone related peptide by estradiol (E2): effect on tumor growth and metastasis in vitro and in vivo. Endocrinology 146, 2885–2894. 27. Thiede, M. A. (1989). The mRNA encoding a parathyroid hormonelike peptide is produced in mammary tissue in response to elevations in serum prolactin. Mol. Endocrinology 3, 1443–1447. 28. Thiede, M. A., Daifotis, A. G., Weir, E. C., Brines, M. L., Burtis, W. J., Ikeda, K., Dreyer, B. E., Garfield, R. E., and Broadus, A. E. (1990). Intrauterine occupancy controls expression of the parathyroid hormone-related peptide gene in preterm rat myometrium. Proc. Natl Acad. Sci. USA. 87, 6969–6973. 29. Liu, B., Goltzman, D., and Rabbani, S. A. (1995). Processing of parathyroid hormone-related peptide (pro-PTHrP) by the prohormone convertase furin: Effect on biological activity. Am. J. Physiol. 268, E832-E838. 30. Soifer, N. E., Dee, K. E., Insogna, K. L., Burtis, W. J., Matovcik, L. M., Wu, T. L., Milstone, L. M., Broadus, A. E., Philbrick, W. M., and Stewart, A. F. (1992). Secretion of a novel mid-region fragment of parathyroid hormone-related protein by three different cell lines in culture. J. Biol. Chem. 267, 18236–18243. 31. Yano, S., Macleod, R. J., Chattopadhyay, N., Tfelt-Hensen, J., Kifor, O., Butters, R. R., and Brown, E. M. (2004). Calcium-sensing receptor activation stimulates parathyroid hormone-related protein secretion in prostate cancer cells: role of epidermal growth factor receptor transactivation. Bone 35(3), 664–672. 32. VanHouten, J., Dann, P., McGeoch, G., Brown, E. M., Krapcho, K., Neville, M., and Wysolmerski, J. J. (2004). The calcium-sensing receptor regulates mammary gland parathyroid hormone-related protein production and calcium transport. J. Clin. Invest 113, 598–608. 33. Chan, G. K., Deckelbaum, R. A., Bolivar, I., Goltzman, D., and Karaplis, A. C. (2001). PTHrP inhibits adipocyte differentiation by down-regulating PPAR gamma activity via a MAPK-dependent pathway. Endocrinology 142, 4900–4909. 34. Miao, D., Tong, X. K., Chan, G. K., Panda, D., McPherson, P. S., and Goltzman, D. (2001). Parathyroid hormone-related peptide stimulates osteogenic cell proliferation through protein kinase C activation of the Ras/mitogen-activated protein kinase signaling pathway. J. Biol Chem. 276, 32204–32213. 35. Mahon, M. J., Donowitz, M., Yun, C. C., and Segre, G. V. (2002). Na+ H+-exchange regulatory factor 2 directs parathyroid hormone receptor signalling. Nature 417, 858–861. 36. Orloff, J. J., Ganz, M. B., Ribaudo, A. E., Buris, W. J., Reiss, M., Milstone, L. M., and Stewart, A. F. (1992). Analysis of PTHrP binding and signal transduction mechanisms in benign and malignant squamous cells. Am. J. Physiol. 262, E599–E607. 37. Orloff, J. J., Kats, Y., Urena, P., Schipani, E., Vasavada, R. C., Philbrick, W. M., Behal, A., Abou-Samra, A.-B., Segre, G. V., and Juppner, H. (1995). Further evidence for a novel receptor for aminoterminal parathyroid hormone-related protein on keratinocytes and squamous carcinoma cell lines. Endocrinology 136, 3016–3023. 38. Care, A. E., Abbas, S. K., Pickard, D. W., Barri, M., Drinkhill, M., Findlay, J. B. C., White, I. R., and Caple, I. W. (1990). Stimulation of ovine placental transport of calcium and magnesium by mid-molecule fragments of human PTHrP. Exp. Physiol. 75, 605–608. 39. Kovacs, C. S., Lanske, B., Hunzelman, J. L., Guo, J., Karaplis, A. C., and Kronenberg, H. M. (1996). Parathyroid hormone-related peptide (PTHrP) regulated fetal-placental calcium transport through a receptor
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
303
distinct from the PTH/PTHrP receptor. Proc. Natl Acad. Sci. USA. 93, 15233–15238. Kaiser, S. M., Laneuville, P., Bernier, S. M., Rhim, J. S., Kremer, R., and Goltzman, D. (1992). Enhanced growth of a human keratinocytes cell line induced by antisense RNA for parathyroid hormone-related peptide. J. Biol. Chem. 267, 13623–13628. Henderson, J. E., Amizurka, N., Warshawsky, H., Biasotto, D., Lanske, B. M. K., Goltzman, D., and Karaplis, A. C. (1995). Nucleolar localization of parathyroid hormone-related peptide enhances survival of chondrocytes under conditions that promote apoptotic cell death. Mol. Cell. Biol. 15, 4064–4075. Jans, D. A., Thomas, R. J., and Gillespie, M. T. (2003). Parathyroid hormone-related protein (PTHrP): a nucleocytoplasmic shuttling protein with distinct paracrine and intracrine roles. Vitam. Horm. 66, 345–384. Fenton, A. J., Kemp, B. E., Kent, G. N., Moseley, J. M., Zheng, M. H., Rowe, D. J., Britto, J. M., Martin, T. J., and Nicholson, G. C. (1991). A carboxyl-terminal peptide from the parathyroid hormone-related protein inhibits bone resorption by osteoclasts. Endocrinology 129, 1762–1768. Fenton, A. J., Kemp, B. E., Hammonds, J., Mitchelhill, K., Moseley, J. M., Martin, T. J., and Nicholson, G. C. (1991). A potent inhibitor of osteoclastic bone resorption within a highly conserved pentapeptide region of parathyroid hormone-related protein, PTHrP [107–111]. Endocrinology 129, 3424–3426. Fukayama, S., Tashjian, Jr, A., David, J. N., and Chisholm, J. C. (1995). Signaling by N-and C-terminal sequences of parathyroid hormone-related protein in hippocampal neurons. Proc. Natl Acad. Sci. USA 92, 10182–10186. Henderson, J., Shustik, C., Kremer, R., Rabbani, S. A., Hendy, G. N., and Goltzman, D. (1990). Circulating concentrations of parathyroid hormone-like peptide in malignancy and in hyperparathyroidism. J. Bone Miner. Res. 5, 105–113. Kaiser, S. M., Sebag, M., Rhim, J. S., Kremer, R., and Goltzman, D. (1994). Antisense-mediated inhibition of parathyroid hormone-related peptide production in a keratinocytes cell line impedes differentiation. Mol. Endocrinol. 8, 139–147. Wysolmerski, J. J., McCaughern-Carucci, J. F., Daifotis, A. G., Broadus, A. E., and Philbrick, W. M. (1995). Overexpression of parathyroid hormone-related protein or parathyroid hormone in transgenic mice impairs branching morphogenesis during mammary gland development. Development 12, 3539–3547. Wysolmerski, J. J., Philbrick, W. M., Dunbar, M. E., Lanske, B., Kronenberg, H., Karaplis, A., and Broadus, A. E. (1998). Rescue of the parathyroid hormone-related protein knockout mouse demonstrates that parathyroid hormone-related protein is essential for mammary gland development. Development 125, 1285–1294. Lam, M. H., Olsen, S. L., Rankin, W. A., Ho, P. W., Martin, T. J., Gillespie, M. T., and Moseley, J. M. (1997). PTHrP and cell division: Expression and localization of PTHrP in a keratinocytes cell line (HaCaT) during the cell cycle. J. Cell. Physiol. 173, 433–446. Massfelder, T., Dann, P., Wu, T. L., Vasavada, R., Helwig, J. J., and Steward, A. F. (1997). Opposing mitogenic and anti-mitogenic action of parathyroid hormone-related protein in vascular smooth muscle cells: a critical role for nuclear targeting. Proc. Natl Acad. Sci. USA. 94, 13630–13635. Meerovitch, K., Wing, S., and Goltzman, D. (1998). Proparathyroid hormone-related protein is associated with the chaperone protein BiP and undergoes proteasome-mediated degradation. J. Biol. Chem. 273, 21025–21030. Aarts, M. M., Rix, A., Guo, J., Bringhurst, R., and Henderson, J. E. (1999). The nucleolar targeting signal (NTS) of parathyroid hormone related protein mediates endocytosis and nucleolar translocation. J. Bone Miner. Res. 14, 1493–1503. Watson, P. H., Fraher, L. J., Hendy, G. N., Chung, U. I., Kisiel, M., Natale, B. V., and Hodsman, A. B. (2000). Nuclear localization of the
304
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
type 1 PTH/PTHrP receptor in rat tissues. J. Bone Miner Res. 15, 1033–1044. Nguyen, M., He, B., and Karaplis, A. (2001). Nuclear forms of parathyroid hormone-related peptide are translated from non-AUG start sites downstream from the initiator methionine. Endocrinology 142, 694–703. Okano, K., Pirola, C. J., Wang, H. M., Forrester, J. S., Fagin, J. A., and Clemens, T. L. (1995). Involvement of cell cycle and mitogen-activated pathways in induction of parathyroid hormone-related protein gene expression in rat aortic smooth muscle cells. Endocrinology 136, 1782–1789. Lam, M. H., Olsen, S. L., Rankin, W. A., Ho, P. W., Martin, T. J., Gillespie, M. T., and Moseley, J. M. (1997). PTHrP and cell division: expression and localization of PTHrP in a keratinocyte cell line (HaCaT) during the cell cycle. J. Cell. Physiol. 173, 433–446. Aarts, M. M., Levy, D., He, B., Stregger, S., Chem, T., Richards, S., and Henderson, J. E. (1999). Parathyroid hormone-related protein interacts with RNA. J. Biol. Chem. 274, 4832–4838. Karaplis, A. C., Luz, A., Glowacki, J., Bronson, R. T., Tybulewicz, V. L. J., Kronenberg, H. M., and Mulligan, R. C. (1994). Lethal skeletal dysplasia from targeted disruption of the parathyroid hormone-related peptide (PTHrP) gene. Genes Dev. 8, 277–289. Amizuka, N., Henderson, J. E., Warshawsky, H., Goltzman, D., and Karaplis, A. C. (1994). Parathyroid hormone-related peptide (PTHrP)-depleted mice show abnormal epiphyseal cartilage development and altered endochondral bone formation. J. Cell Biol. 126, 1611–1623. Lee, K., Lanske, B., Karaplis, A. C., Deeds, J. D., Kohno, H., Nissenson, R. A., Kronenberg, H. M., and Segre, G. V. (196). Parathyroid hormone-related peptide delays terminal differentiation of chondrocytes during endochondral bone development. Endocrinology 137, 5055–5067. Weir, E. C., Philbrick, W. M., Amling, M., Neff, L. A., Baron, R., and Broadus, A. E. (1996). Targeted overexpression of parathyroid hormone-related peptide in chondrocytes causes chondrodysplasia and delayed endochondral bone formation. Proc. Natl Acad. Sci. USA. 93, 10240–10245. Amizuka, N., Henderson, J. E., Hoshi, K., Warshawsky, H., Ozawa, H., Goltzman, D., and Karaplis, A. C. (1996). Programmed cell death of chondrocytes and aberrant chondrogenesis in mice homozygous for parathyroid hormone-related peptide gene deletion. Endocrinology 137, 5055–5067. Farnum, C. E., and Wilsman, N. J. (1989). Condensation of hypertrophic chondrocytes at the chondro-osseous junction of growth plate cartilage in Yucatan swine: relationship to long bone growth. Am. Anat. 186, 346–358. Amling, M., Neff, L., Tanaka, S., Inoue, D., Kuida, K., Weir, E. C., Philbrick, W. M., Broadus, A. E., and Baron, R. (1997). Bcl-2 downstream of parathyroid hormone-related peptide in a signal pathway that regulates chondrocytes maturation during skeletal development. J. Cell. Biol. 136, 205–213. Provot, S., and Schipani, E. (2005). Molecular mechanisms of endochondral bone development. Biochem. Biophys. Res. Commun. 328, 658–665. Lanske, B., Karaplis, A. C., Lee, K., Luz, A., Vortkamp, A., Pirro, A., Karperien, M., Defize, L. H., Ho, C., Mulligan, R. C., Abou-Samra, A. -B., Juppner, H., Segre, G. V., and Kronenberg, H. M. (1996). PTH/PTHrP receptor in early development and Indian Hedgehog-regulated bone growth. Science 273, 663–666. Decklebaum, R. A., Chan, G., Miao, D., Goltzman, D., and Karaplis, A. C. (2002). Ihh enhances differentiation of CfK-2 chondrocytic cells and antagonizes PTHrP-mediated activation of PKA. J. Cell Sci. 115 (Pt 14), 3015–3025. Davidson, D., Blanc, A., Filion, D., Wang, H., Plut, P., Pfeffer, G., Buschmann, M. D., and Henderson, J. E. (2005). FGF18 signals
DAVID GOLTZMAN
70.
71.
72.
73. 74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
through FGFR3 to promote chondrogenesis. J. Biol. Chem. 27, 20509–20515. Schipani, E., Lanske, B., Hunzelman, J., Luz, A., Kovacs, C. S., Le, K., Pirro, A., Kronenberg, H. M., and Juppner, H. (1997). Targeted expression of constitutively active receptors for parathyroid hormone and parathyroid hormone-related peptide delays endochondral bone formation and rescues mice that lack parathyroid hormone-related peptide. Proc. Natl Acad. Sci. USA. 94, 13689–13694. Zhang, P., Jobert, A. S., Couvineau, A., and Sillve, C. (1998). A homozygous inactivating mutation in the parathyroid hormone/ parathyroid hormone-related peptide receptor causing Blomstrand chondrodysplasia. J. Clin. Endocrinol. 83, 3365–3368. Karaplis, A. C., He, B., Nguyen, M. T., Young, I. D., Semeraro, D., Ozawa, H., and Amizuka, N. (1998). Inactivating mutation in the human parathyroid hormone receptor type 1 gene in Blomstrand chondrodysplasia. Endocrinology 139, 5255–5258. Kruse, K. (1993). Calcium metabolism in the Jansen type of metaphyseal dysplasia. Eur. J. Pediatr. 1522, 912–915. Schipani, E., Kruse, K., and Juppner, H. (1995). A constitutively active mutant PTH-PTHrP receptor in Jansen-type metaphyseal chondrodysplasia. Science 268, 98–100. Schipani, E., Langman, C. B., Parfitt, A. M., Jensen, G. S., Kikuch, S., Kooh, S. W., Cole, W. G., and Juppner, H. (1996). Constitutive activated receptors for parathyroid hormone-related peptide in Jansen metaphyseal chondrodysplasia. N. Engl. J. Med. 335, 708–714. Hopyan, S., Gokgoz, N., Poon, R., Gensure, R. C., Yu, C., Cole, W. G., Bell, R. S., Juppner, H., Andrulis, I. L., Wunder, J. S., and Alman, B. A. (2002). A mutant PTH/PTHrP type 1 receptor in enchondromatosis. Nat. Genet. 30, 306–310. Lee, K., Deeds, J. D., Bond, A. T., Juppner, H., About-Samra, A.-B., and Segre, G. V. (1993). In situ localization of PTH/PTHrP receptor mRNA in the bone of fetal and young rats. Bone 14, 341–345. Lee, K., Deeds, J. D., Chiba, S., Un-No, M., Bond, A. T., and Segre, G. V. (1994). Parathyroid hormone induces sequential c-fos expression in bone cells in vivo: In situ localization of its receptor and c-fos messenger ribonucleric acids. Endocrinology 134, 441–450. Lee, K., Deeds, J. D., and Segre, G. V. (1995). Expression of parathyroid hormone related-peptide and its receptor messenger ribonucleic acids during fetal development of rats. Endocrinology 136, 453–463. Fermor, B., and Skerry, T. M. (1995). PTH/PTHrP receptor expression on osteoblasts and osteocytes but not resorbing bone surfaces in growing rats. J. Bone Miner. Res. 10, 1935–1943. Celic, S., Chilco, P. J., Zajac, J. D., Martin, T. J., and Findley, D. M. (1996). A type 1 collagen substrate increases PTH/PTHrP receptor mRNA expression and suppresses PTHrP mRNA expression in UMR106-06 osteoblast-like cells. J. Endocrinol. 150, 299–308. Suda, N., Gillespie, M. T., Traianedes, K., Zhou, H., Ho, P. W. M., Hards, D. K., Allan, E. H., Martin, T. J., and Moseley, J. M. (1996). Expression of parathyroid hormone-related protein in cells of osteoblast lineage. J. Cell. Physiol. 166, 94–104. McCauley, L. K., Koh, A. J., Beecher, C. A., Cui, Y., Rosol, T. l. J., and Franceschi, R. T. (1996). PTh/PthrP receptor is temporally regulated during osteoblast differentiation and is associated with collagen synthesis. J. Cell. Biochem. 61, 638–647. Miao, D., He, B., Karaplis, A. C., Goltzman, D. (2002). Parathyroid hormone is essential for normal fetal bone formation. J. Clin. Invest. 109, 1173–1182. Amizuka, N., Karaplis, A. C., Henderson, J. E., Warshawsky, H., Lipman, M. L., Matsuki, Y., Eijiri, S., Tanaka, M., Izumi, N., Ozawa, H., and Goltzman, D. (1996). Haploinsufficiency of parathyroid hormonerelated peptide (PTHrP) results in abnormal postnatal bone development. Dev. Biol. 175, 166–176. Owen, M., and Friedenstein, A. J. (1988). Stromal stem cells: marrowdeprived osteogenic precursors. Ciba Foundation Symposium 136, 42–60.
Chapter 17 Interaction of Parathyroid Hormone-related Peptide with the Skeleton 87. Miao, D., Li, J., Xue, Y., Su, H., Karaplis, A. C., and Goltzman, D. (2004). Parathyroid hormone-related peptide is required for increased trabecular bone volume in parathyroid hormone-null mice. Endocrinology 145, 3554–3562. 88. Rouleau, M. F., Mitchell, J., and Goltzman, D. (1990). Characteristics of the major parathyroid hormone target cell in the endosteal metaphysis of rat long bones. J. Bone Miner. Res. 5, 1043–1053. 89. Canalis, E., McCarthy, T. L., and Centrella, M. (1990). Differential effects of continuous and transient treatment with parathyroid hormone related peptide (PTHrP) on bone collagen synthesis. Endocrinology 126, 1806–1812. 90. Urena, P., Iida-Klein, A., Kong, X.-F., Juppner, H., Kronenberg, H. M., About-Samra, A.-B, and Segre, G. V. (1994). Regulation of parathyroid (PTH)/PTH-related peptide receptor messenger ribonucleic acid by glucocorticoids and PTH in ROS 17/2.8 and OK Cells. Endocrinology 134, 451–456. 91. Ernst, M., Heath, J. K., and Rodan, G. A. (1989). Estradiol effects on proliferation, messenger ribonucleric acid for collagen and insulin-like
92.
93.
94. 95.
305
growth factor-I, and parathyroid hormone-stimulated adenylate cyclase activity in osteoblastic cells from calvariae and long bones. Endocrinology 125, 825–833. Fukayama, S., and Tashjian, A. H. (1989). Direct modulation by estradiol of the response of human bone cells (SaOS-2) to human parathyroid hormone (PTH) and PTH-related protein. Endocrinology 124, 397–401. Joborn, C., Ljunghall, S., Larsson, K., Lindh, E., Naessen, T., Wide, L., Akerstrom, G., and Rastad, J. (1991). Skeletal responsiveness to parathyroid hormone in healthy females: Relationship to menopause and estrogen replacement. Clin. Endocrinol. 34, 335–339. Reeve, J. (1996). PTH: A future role in the management of osteoporosis? J. Bone Miner. Res. 11, 440–445. Horowitz, M. J., Tedesco, M. B., Gundberg, C., Garcia-Ocana, A, Stewart, A. F. (2003). Short-term, high-dose parathyroid hormonerelated protein as a skeletal anabolic agent for the treatment of postmenopausal osteoporosis. J. Clin. Endocrinol. Metab. 88, 569–575.
This Page Intentionally Left Blank
Chapter 18
The Vitamin D Hormone and its Nuclear Receptor: Mechanisms Involved in Bone Biology Geert Carmeliet, Annemieke Verstuyf, Christa Maes, Guy Eelen, and Roger Bouillon
V. Pathology and Therapy Related to Vitamin D Availability, Metabolism, and Function VI. Conclusions References
I. Introduction II. Metabolism of Vitamin D III. Nuclear Vitamin D Receptor IV. Vitamin D and Bone Cells
I. INTRODUCTION
retinoid X receptor and associating specifically with vitamin D responsive elements (VDREs) in target genes. Upon binding of 1,25(OH)2D3, VDR releases corepressors and attracts general transcription factors and coactivators, some of them acetylating histones to create a permissive chromatin surrounding. At the cellular level, 1,25(OH)2D3 treatment inhibits osteoblast proliferation but induces osteoblast differentiation when given at the differentiation stage and protects osteoblastic cells from undergoing apoptosis. In addition, 1,25(OH)2D3 promotes osteoclast differentiation and alters chondrocyte development.
The active metabolite of vitamin D, 1α,25 dihydroxyvitamin D3 [1,25(OH)2D3], has multiple functions in humans and animals but its major role is related to bone metabolism and mineral homeostasis. Vitamin D is metabolized to its active form by two sequential hydroxylations on carbon-25 and carbon-1. The genomic action of 1,25(OH)2D3 is mediated by the nuclear vitamin D receptor (VDR) that binds its ligand with high affinity. VDR is a phosphoprotein that regulates gene expression by heterodimerizing with Dynamics of Bone and Cartilage Metabolism
Laboratorium for Experimental Medicine and Endocrinology, K. U. Leuven, Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium
307
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
308
GEERT CARMELIET ET AL .
Vitamin D deficiency or mutations in the 25-hydroxyvitamin D-1α-hydroxylase gene or VDR gene lead to rickets and osteomalacia, a phenotype that is also observed in the corresponding knock-out mouse models. Although bone mineralization can be maintained in the absence of VDR or its ligand when sufficient mineral is supplied, the precise role of the vitamin D system in fine tuning of bone mineralization and homeostasis remains to be determined.
II. METABOLISM OF VITAMIN D A. Source and Synthesis of Vitamin D The secosteroid vitamin D3 can be produced endogenously in the skin by the action of UV light (290–300 nm), converting 7-dehydrocholesterol into previtamin D3 that is in thermoequilibrium with vitamin D3 (Fig. 1). A second source of vitamin D is the dietary intake, whereby 50% of vitamin D is absorbed by the enterocytes and transported via the chylomicrons to the general circulation. A first step in the activation pathway of vitamin D3 is the hydroxylation at carbon 25 catalyzed by a cytochrome P450 (CYP) enzyme and takes place primarily in the liver to produce 25-hydroxyvitamin D3 [25(OH)D3], the major circulating form of vitamin D3. Several CYPs, localized either in the inner mitochondrial membrane or in
Figure 1
the microsomes, can catalyze this reaction in vitro but CYP27A1 and CYP2R1 are the two viable candidates. Mutations in the human and mouse genes encoding the mitochondrial CYP27A1 protein impair bile acid synthesis, but have no consequences for vitamin D metabolism [1, 2]. On the other hand, molecular analysis of a patient with selective 25-hydroxyvitamin D deficiency and symptoms of vitamin D deficiency, including skeletal abnormalities, identified a mutation in the CYP2R1 gene (chromosome 11p15.2), encoding a microsomal enzyme [3]. Nevertheless, this patient still had measurable levels of 1,25(OH)2D3 and responded to vitamin D treatment. 25(OH)D3 is biologically inactive and requires further hydroxylation in the kidney to the active hormone 1,25(OH)2D3 by another mitochondrial cytochrome P450 enzyme, 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). The production of 1,25(OH)2D3 is regulated primarily at this final step by several factors including parathyroid hormone (PTH) (positive feedback), calcium, phosphate, and 1,25(OH)2D3 (negative feedback). The proximal renal tubule is the principal site of 1α-hydroxylase activity but expression has also been detected in several tissues including bone and cartilage. The CYP27B1 gene has been cloned in several species [4–8] and the human gene is mapped on chromosome 12q13.3. Mutations in the human gene are responsible for the rare autosomal recessive disease pseudovitamin D-deficiency rickets (PDDR) and will be discussed in Section V.A.2.
Sources, bioactivation of vitamin D3 and metabolism, actions of 1,25(OH)2D3. The active 1,25(OH)2D3 hormone is synthesized from the vitamin D precursor by two successive enzymatic hydroxylations on carbon-25 and carbon-1 in liver and kidney respectively. 1,25(OH)2D3 acts principally on intestine, kidney, and bone in order to keep serum calcium constant. The major pathway in catabolism is via 24-hydroxylase that has a broad tissue distribution.
309
Chapter 18 The Vitamin D Hormone and its Nuclear Receptor
B. Transport of Vitamin D In blood, vitamin D and its active metabolites are transported by binding to a specific plasma protein vitamin Dbinding protein (DBP). The vitamin D-binding site of DBP is a cleft located at the surface of the molecule and partly in contact with the surrounding solvent as evidenced from the crystal structure of the human DBP [9]. Mice deficient in DBP had low levels of total vitamin D metabolites, but were otherwise normal, suggesting that intracellular concentrations of 1,25(OH)2D3 are still adequate [10]. However, when dietary vitamin D was reduced a clear functional 1,25(OH)2D3 deficiency state with osteomalacia developed rapidly in DBP-null mice, indicating that DBP provided a degree of protection against dietary-induced vitamin D deficiency. Unexpectedly, DBP did not create a buffer for 1,25(OH)2D3 as DBP-null mice were resistant to vitamin D toxicity, possibly reflecting a more rapid entry into metabolic breakdown pathways and a more rapid urinary excretion of 25(OH)D3. Indeed, after renal glomerular filtration of 25(OH)D3 in complex with DBP it is reabsorbed in the proximal tubular cells by the endocytic receptor megalin, rather than entering the cell by simple diffusion. The importance of this pathway was underlined by the phenotype of mice with a total or kidney-specific megalin gene defect [11, 12]. The latter mice show hypocalcemia and osteomalacia as a consequence of hypovitaminosis D, indicating that renal reuptake of the 25(OH)D3–DBP complex is essential to preserve vitamin D stores.
C. 24-Hydroxylase: Catabolism or Specific Function? An alternative hydroxylation of 25(OH)D3 occurs on carbon 24 by the multifunctional enzyme 24-hydroxylase (CYP24), mapped on human chromosome 20q13 [13]. This enzyme not only initiates the catabolic cascade of 25(OH)D3 and 1α,25(OH)2D3 [14, 15] by 24-hydroxylation but catalyzes also the dehydrogenation of the 24-OH group and performs 23-hydroxylation, resulting in 24-oxo1,23,25-(OH)3D3 [16]. This C24 oxidation pathway leads finally to calcitroic acid, which is the major end product of 1,25(OH)2D3 and this catabolic breakdown is regulated by 1,25(OH)2D3 itself. This indicates that CYP24 may play an important role in regulating the functions of 1,25(OH)2D3 in the target cells. In vivo evidence for the role of CYP24 in the catabolism of 1,25(OH)2D3 was provided by the generation of CYP24-null mice [17]. These mice were unable to clear 1,25(OH)2D3 from the bloodstream in response to 1,25(OH)2D3 treatment. Whether 24,25(OH)2D3 has a
physiological role in bone metabolism has previously been a matter of debate [18, 19]. CYP24-null mice, born from CYP24-null mothers, have impaired intramembranous bone formation that was however normalized after crossing them with vitamin D receptor (VDR)-null mice. This finding indicates that this phenotype was caused by increased 1,25(OH)2D3 levels acting via the VDR, and not by the absence of 24,25(OH)2D3.
III. NUCLEAR VITAMIN D RECEPTOR A. General Characteristics of the Protein and Gene The actions of 1,25(OH)2D3 are mediated by the nuclear vitamin D receptor (VDR), a member of the superfamily of steroid/thyroid hormone receptors [20]. The VDR has been cloned from several species, among which are humans, mouse, rat, chick, quail, Xenopus, zebrafish, rainbow trout, and Japanese flounder [21–28]. The recently cloned VDR from lamprey, an ancient vertebrate without calcified skeleton, indicates that the VDR predates the origin of a calcified skeleton [29]. VDR is a 55-kDa protein made up of 427 amino acids and is present at a low concentration (10–100 fmol/mg protein), consistent with the fact that it is a potent transcription regulatory molecule. The human VDR is encoded by a single gene mapped to chromosome 12q13. The gene is comprised of at least 14 exons; the noncoding 5′ end of the gene includes exons 1a, 1b, 1c, 1d, 1e, and 1f, whereas eight additional exons (exons 2–9) encode the structural portion of the VDR gene (Fig. 2) [30, 31]. For human VDR, a multitude of different transcripts were found that mostly vary in their 5′ untranslated region but encode the same 427-amino acid protein. Transcripts originating from the most distal promoter (near the 1f exon) were only found in calcitropic vitamin D target tissues (kidney, intestine, and parathyroid) [31] and illustrate that different receptor isoforms may contribute to tissuespecific effects.
B. Structure–Function Analysis A deeper insight into the structural and functional properties of the VDR protein and a better understanding of ligand-binding has been provided by site-directed mutation studies and by three-dimensional modeling of the protein by X-ray crystallography. From N- to C-terminus the VDR contains several functionally distinct domains which are listed below (Fig. 2).
310
GEERT CARMELIET ET AL .
Figure 2
Gene and protein structure of the human vitamin D receptor (hVDR). The hVDR gene is composed of at least 14 exons and is mapped to chromosome 12q13. Indicated on the protein structure are the domains for DNA binding (DBD) and ligand binding (LBD) as well as the A/B domain and the hinge region. Two major phosphorylation sites are indicated by P.
1. The A/B-Domain
Unlike other nuclear receptors, the VDR contains a truncated A/B-domain which lacks the usual transactivation function (AF1). Therefore, little or no function was ascribed to this region. Evidence for a substantial role of the A/B-domain in transcriptional activity of the receptor came from recent studies on N-terminal variants of the VDR. Two of these variants with 50 or 23 amino acid extensions at their N-terminus display reduced transcriptional activity [32]. On the other hand, a polymorphic variant, called F/M4, lacking three amino acids at the N-terminal start, displays an increased transcriptional activity presumably due to a better interaction with the transcription factor TFIIB [33]. Multiple studies have linked the F/M4 polymorphism to increased bone mineral density [e.g. 34, 35]. 2. The C-Domain or DNA-Binding Domain
The DNA-binding domain (DBD) consists of two zincfinger modules within a 66 amino acid core sequence that is highly conserved among nuclear receptors [36]. The N-terminal module mediates specific DNA binding in the major groove of the DNA-binding site, whereas the C-terminal module is involved in binding a dimerization partner [37, 38]. In addition to this core sequence, the DBD contains a variable C-terminal extension (CTE) which is also involved in receptor dimerization and in directing the ligand-bound VDR, together with its dimerization partner to correctly spaced VDREs (see below) [39]. To function as a transcription factor, VDR needs to reside in the nucleus. In a screen for possible nuclear localization signals (NLS), two stretches of basic amino acids at both ends of the VDR region 79–105, which is partially located within the DBD, were shown to be crucial for nuclear accumulation [40]. A sequence of five basic amino acids positioned between the two zinc-finger modules is likely to function as an additional NLS [41].
3. THE D-DOMAIN OR HINGE-REGION
This region with poor sequence conservation confers flexibility to the receptor. An NLS of 20 amino acids in this region was found to facilitate nuclear transfer of the receptor [42]. 4. THE E-DOMAIN OR LIGAND-BINDING DOMAIN
The multifunctional ligand-binding domain (LBD) serves three major functions; binding of 1,25(OH)2D3, hetero- or homodimerization, and recruitment of regulatory factors needed for transcriptional activation. a. Ligand Binding-Function The VDR binds its ligand with extremely high affinity, in the range of 10−10 M. The ligand-binding pocket (LBP) comprises many segments spanning the entire C-terminal domain of 300 amino acids [43]. Great progress in understanding ligand binding and in identifying the crucial amino acids came from crystallization studies of the VDR in complex with 1,25(OH)2D3 [44]. Much like in other nuclear receptors the hydrophobic LBP of the VDR consists of 13 α-helices and a three-stranded β-sheet. In the LBP the A-ring of 1,25(OH)2D3 adopts a chair B conformation [45] and the connection between the A- and the C-ring lies in a hydrophobic channel formed by amino acids Ser275, Trp286, and Leu233. Contacts between specific amino acids lining the LBP and the hydroxyl groups of 1,25(OH)2D3 were found crucial for ligand binding. The 1α-OH group forms a hydrogen bond with Ser237 and with Arg274 whereas the 3-OH moiety is hydrogen-bonded with Ser278 and Tyr143. The 25-OH group connects through hydrogen bonds with His305 and His397. Studies with VDR constructs bearing mutations in the above-mentioned amino acids stress their importance in ligand binding [46]. b. Dimerization VDR forms homo- or heterodimers that can interact with specific DNA segments (VDREs, see below). The receptor
Chapter 18 The Vitamin D Hormone and its Nuclear Receptor
for 9-cis-retinoic acid, RXR, is the preferred dimerization partner for VDR [47–49] but also for other nuclear receptors and might therefore be a key element in cross-talk between nuclear receptor-mediated signaling [50, 51]. Part of the dimerization function of the receptor maps to two regions (amino acids 239–269 and 317–401) in the ligand-binding domain that are highly conserved across the steroid hormone receptor family [52]. In addition, regions in the DNA-binding domain are also involved in dimerization (see Section III.B.2). c. Transcriptional Activation Upon binding of 1,25(OH)2D3, VDR recruits RXR and this heterodimer can bind to specific DNA sequences in the promoter region of target genes called vitamin D responsive elements (VDREs) (Fig. 3). The consensus high-affinity VDRE consists of two hexameric half sites (PuG(G/T)TCA) arranged as a direct repeat spaced by three base pairs (DR3-type VDRE). Other natural VDREs exist, among which is an inverted palindrome of two hexameric binding sites known as IP9-type VDRE [53]. RXR occupies the 5′ half-site and VDR the 3′ half-site in cases of positive gene regulation. Reversal of the VDR-RXR polarity on the VDRE has been suggested to suppress gene expression (reviewed in reference [54]). Binding of the VDR–RXR dimer on the VDRE induces DNA-bending which facilitates transcription complex assembly [55].
311
To contact the basal transcription complex, VDR–RXR releases corepressors and recruits coactivator molecules (Fig. 3). Upon ligand binding in the VDR-LBP, helix 12 closes off the LBP, much like a mousetrap, and exposes its activation function 2 (AF2) to which coactivators can bind through a conserved LXXLL motif in their amino acid sequence. The liganded VDR induces RXR to close off its LBP with helix 12 without binding of the cognate ligand 9-cis-retinoic acid. This “phantom ligand effect” allows RXR to recruit coactivators as well [56]. One type of attracted coactivators recruits histone acetyl-transferase (HAT) activity; they acetylate histone tails and create a permissive chromatin surrounding for gene transcription. This class contains CBP/p300 and the p160 family of proteins like SRC-1, GRIP1/TIF2, and ACTR (reviewed in reference [57]). Another type of coactivator is the multimeric vitamin D receptor interacting proteins (DRIP) complex, of which the 205-kDa subunit (DRIP205) interacts directly with the VDR–RXR heterodimer. The complex is not associated with HAT activity but instead recruits RNA polymerase II, the key enzyme needed for gene transcription [58]. Two other types of coactivators are NCoA-62/ski-interacting protein (SKIP) and WINAC. The former is thought to link transcriptional activation by nuclear receptors with mRNA splicing [59], whereas the latter interacts with VDR through the Williams’ syndrome
Figure 3 Simplified model of gene transcription by 1,25(OH)2D3. Ligand-bound VDR heterodimerizes with RXR and binds to VDRE sequence upstream of the target gene. Upon binding of 1,25(OH)2D3, VDR releases corepressors (not shown in figure) and attracts coactivator molecules (represented here by CBP/p300, p160, DRIP, and NCoA-62/SKIP) and general transcription factors (GTF, RNA polymerase II). Coactivators with histone acetyl transferase (HAT)-activity acetylate histones to create a permissive chromatin surrounding.
312
GEERT CARMELIET ET AL .
transcription factor (WSTF) and displays ATP-dependent chromatin-remodeling activity [60]. VDR also directly contacts the N-terminal region of the general transcription factor IIB (TFIIB), a component of the basal transcription apparatus. The VDR–TFIIB interaction is not dependent upon 1,25(OH)2D3 and is disrupted by the ligand. A model is proposed in which unliganded VDR recruits TFIIB to the promoter region of vitamin D-responsive target genes where this transcription factor can be incorporated into the transcription apparatus after ligand-induced disruption from the VDR [61–63]. Recently TAFII-17, a subunit of the general transcription factor TFIID, was found to interact with VDR at low doses of 1,25(OH)2D3. Increased levels of TAFII-17 were linked to the bone-remodeling disorder Paget’s disease [64].
C. Modulation of Vitamin D Receptor 1. Phosphorylation, “Cross Talk”, and Receptor Activity
VDR is a phosphoprotein and the majority of the phosphorylation of VDR is on serine residues. Nevertheless the exact functional consequences of this phosphorylation have been difficult to determine. VDR becomes hyperphosphorylated upon 1,25(OH)2D3 treatment [65] which is partly catalyzed by casein kinase II that phosphorylates VDR at Ser-208 in the LBD and hereby amplifies VDR transcriptional activity [66]. Accordingly, a recent study indicated that 1,25(OH)2D3-dependent transcription was enhanced when phosphatase inhibitors were used. This enhancement may be due, at least in part, to an increased interaction between the VDR and the coactivator protein DRIP205 [67]. In addition, two 1,25(OH)2D3-independent phosphorylations of VDR antagonize this positive phosphorylation. Protein kinase C-mediated phosphorylation of Ser-51 decreased the DNA binding of the receptor [68]. Analogously, protein kinase A phosphorylates serines 182–185 in the LBD of VDR that seems to interfere with transactivation [69]. Taken together, VDR activity is both positively and negatively regulated depending on the type of phosphorylation and this might represent a mechanism whereby other cell-signaling systems interfere – cross talk – with the genomic (or nongenomic) actions of 1,25(OH)2D3 [50]. 2. Regulation of Vitamin D Receptor Abundance
Several studies have demonstrated that there is a strong correlation between the VDR level and the biological response in target cells [70]. A wide variety of factors, including PTH [71], pharmacological activators of the
protein kinase A pathway [72], transforming growth factor β (TGFβ) [73], estrogens [74], thyroid hormone [75], glucocorticoids [76], and retinoic acid [77] can alter VDR mRNA levels in a tissue-specific pattern (heterologous regulation). Interestingly, both cell cycle [78] and the differentiation state of cells in culture [79] influence the extent of VDR mRNA expression. Autoregulation of VDR expression has been demonstrated both in vitro and in vivo, again in a very tissuespecific fashion and could serve as a means of signal amplification. It is well established that 1,25(OH)2D3 increases the VDR content by stabilizing the receptor [80, 81]. The role of 1,25(OH)2D3 in the transcriptional regulation of the VDR gene is less evident and 1,25(OH)2D3-inducibility of the hVDR gene has not been found with the part of the promoter characterized [30, 82, 83].
IV. VITAMIN D AND BONE CELLS A. Genomic Versus Nongenomic Action of Vitamin D As a transcription factor, 1,25(OH)2D3 affects the expression of several genes in osteoblasts, osteoclasts, and chondrocytes, thereby regulating cellular growth and differentiation of these cells. Although over 50 genes have been reported to be regulated by 1,25(OH)2D3, only a small number have been reported to contain VDREs. Nongenomic responses to 1,25(OH)2D3 have also been described. They develop at the plasma membrane and comprise rapid (seconds to minutes) changes in ion channel activities, activation of second-messenger pathways, and elevation of cytosolic calcium concentrations [84, 85]. Recently, Zanello and Norman [86] described that the presence of a functional VDR was essential for the modulation of Cl− and Ca2+ channel activities by 1,25(OH)2D3 as well as for 1,25(OH)2D3-dependent rapid exocytose of osteoblast-secretory granules. Moreover, the classical VDR seems to be the 1,25(OH)2D3-binding protein associated with plasma membrane caveolae [87]. Analogously, inactivation of the VDR was found to abrogate rapid 1,25(OH)2D3mediated changes in intracellular Ca2+ in calvarial osteoblasts [88]. A new receptor conformational ensemble model proposes that the VDR has two ligand-binding pockets, which, when occupied by the appropriately shaped ligand will result in the onset of either rapid or genomic responses [89, 90]. Another study however reported that the rapid effect of 1,25(OH)2D3 on intracellular calcium and protein kinase C was independent of VDR [91], suggesting the existence of membrane-associated receptor.
313
Chapter 18 The Vitamin D Hormone and its Nuclear Receptor
B. Effects of 1,25(OH)2D3 on Osteoblast Growth, Differentiation, and Apoptosis The general concept is that 1,25(OH)2D3 has a biphasic effect on osteoblasts: it abrogates or stimulates the normal developmental pathway or gene-expression profiles depending on whether it is given during the proliferation or differentiation stage. 1,25(OH)2D3 given during the proliferative period of rat calvaria osteoblast cultures inhibited proliferation and down-regulated collagen synthesis and alkaline phosphatase activity. No increase in osteocalcin expression was observed [92] and nodule formation was blocked [93] suggesting a stop in osteoblast differentiation. In contrast, 1,25(OH)2D3 treatment of mature osteoblasts resulted in up-regulation of some osteoblast-associated genes, such as osteopontin and osteocalcin and in stimulation of calcium accumulation [94]. At this stage of differentiation 1,25(OH)2D3 may promote further maturation of the osteoblast. However, differences in species, cell types, and cell culture conditions, as well as interactions with growth factor and cytokine signaling may result in diverse and sometimes contradictory biological actions of 1,25(OH)2D3. Despite this controversy, the general agreement is in favor of antiproliferative and prodifferentiative properties of 1,25(OH)2D3. In addition, 1,25(OH)2D3 protects osteoblastic cells from undergoing apoptosis, as shown for both human and rat osteoblastic cells treated with several apoptosis-inducing agents [95–97].
C. 1,25(OH)2D3-Regulated Gene Expression Related to Osteoblast Characteristics 1. Cell Cycle Genes
The antiproliferative effect of 1,25(OH)2D3 on several cell types including osteoblasts is due to inhibition of the transition from the G1 to the S-phase of the cell cycle [98, 99]. Cyclin-dependent kinase inhibitors like p21CIP1/WAF1 and p27KIP1 are believed to be the main mediators of this effect [100], although conflicting results on this matter exist [101–104]. A study on 1,25(OH)2D3-induced growth inhibition in murine osteoblasts revealed an early and persistent VDR-dependent down-regulation of cyclin D1 and several genes with key functions in the process of DNA replication. Similar findings in other cell types stress the general nature and the importance of this repressed gene expression in the antiproliferative effect of 1,25(OH)2D3 [105]. 2. TRANSCRIPTION FACTORS IN OSTEOBLAST DIFFERENTIATION
During osteoblast differentiation, 1,25(OH)2D3 regulates the expression of numerous genes belonging to different
families, some of them containing an identified functional VDRE. The expression of the transcription factor Runx2 that is essential for osteoblast differentiation [106], is down-regulated by 1,25(OH)2D3 in rodent osteoblastic cells via the VDR/RXR heterodimer recognizing a functional VDRE [107]. On the other hand, in primary human osteoblasts the effect of 1,25(OH)2D3 on Runx2 mRNA expression was found to be dependent on the duration of the treatment, being inhibitory after 1 h and stimulatory after 48 h [108]. 3. Extracellular Matrix Proteins
The expression of several noncollagenous extracellular matrix proteins is regulated by 1,25(OH)2D3 with osteocalcin being the best characterized. Transcription of the osteocalcin gene is principally regulated by Runx2 through multiple Runx2-binding sites in the promoter. 1,25(OH)2D3 stimulates the expression of osteocalcin strongly in human and rat osteoblastic cells. The VDRE in the rat osteocalcin gene is flanked by Runx2-binding sites and all these regulatory domains contribute to the extensive chromatin reorganization of the involved promoter regions at the time of gene activation [109–111]. In contrast, the expression of the mouse osteocalcin gene is inhibited by 1,25(OH)2D3 [112, 113], which may be explained by differences in the organization of Runx2 motifs in the promoter regions [114]. This illustrates that 1,25(OH)2D3 can play different roles in the regulation of the same gene in various species. Two other genes coding for extracellular matrix proteins have been shown to contain a functional VDRE involved in enhanced transcription by 1,25(OH)2D3: the mouse osteopontin [115] and the rat bone sialoprotein gene [116]. At the same time, genes encoding the proteinases MMP-13 and MT1-MMP [117, 118], functioning in extracellular matrix remodeling, are positively regulated by 1,25(OH)2D3 treatment. 4. Growth Factors
Not only local factors involved in bone remodeling but also their receptors and/or downstream effectors are modulated by 1,25(OH)2D3, but again different responses have been observed depending on the system used and on the stage of cell differentiation. As an example the effect on insulin-like growth factor I (IGF-I) production is either stimulatory or inhibitory, whereas on IGF receptors and IGF-binding proteins it is mainly stimulatory. TGFβ and TGFβ receptors are up-regulated by 1,25(OH)2D3, while Smad proteins that are downstream effectors of TGFβ signaling can interact (as a coactivator) with VDR (for review see reference [119]). A consistent stimulatory effect of 1,25(OH)2D3 is seen on vascular endothelial growth factor (VEGF) [120].
314
GEERT CARMELIET ET AL .
D. Effects of 1,25(OH)2D3 on Osteoclastogenesis and Osteoclast Activity 1. Indirect Role of 1,25(OH)2D3 on Osteoclastogenesis
1,25(OH)2D3 is a potent stimulator of osteoclastic bone resorption by stimulating osteoclast formation and activity. It has been shown that 1,25(OH)2D3 acts as a fusigen for committed osteoclast precursors [121]. Unlike osteoblasts, osteoclasts do not express a VDR, but osteoclast precursors seem to contain the receptor. Therefore, it is unlikely that 1,25(OH)2D3 acts on mature osteoclasts directly but rather exerts its effects on osteoclastogenesis indirectly by modulating osteoblasts. Recently, it has been shown that 1,25(OH)2D3 acts directly on osteoblasts/ stromal cells to increase the production of membranebound receptor activator of nuclear factor (NF)-kB ligand (RANKL) [122], that plays an essential role in osteoclast differentiation and activation [123]. RANKL acts then via binding to its signal-transducing receptor RANK, present on (pre-)osteoclasts. This induction of RANKL is mediated by VDR/RXR binding to functional VDRE in the promoter region [124, 125]. In addition, up-regulation of macrophage-colony stimulating factor (M-CSF) [126] and suppression of the RANKL-decoy receptor OPG [122, 127, 128] in osteoblastic cells are involved in 1,25(OH)2D3enhanced osteoclast formation. This indirect osteoblastmediated mechanism may explain previous observation of coculture experiments with VDR-deficient mice: no osteoclasts were formed in response to 1,25(OH)2D3 when VDR-null osteoblastic cells were cocultured with wild-type (WT) spleen cells, while normal osteoclasts were present when WT osteoblastic cells and VDR-null spleen cells were cocultured [129]. Interestingly, the action of 1,25(OH)2D3 is not essential for osteoclast formation in vivo since VDR-deficient mice have normal number of osteoclasts [130]. In addition, the expression or signaling of several other cytokines and growth factors, which stimulate (interleukin-1 (IL-1), IL-6, IL-11) or inhibit (IL-4, TGFβ) bone resorption, are regulated by 1,25(OH)2D3 (reviewed in reference [131]). 2. 1,25(OH)2D3-Regulated Genes Involved in Osteoclast Activity
Several markers of the osteoclast phenotype are induced by 1,25(OH)2D3 in cultures of osteoclast precursors. One of them is the vitronectin receptor αvβ3 integrin that enables osteoclasts to adhere to bone-matrix proteins containing the RGD sequence, such as osteopontin, fibronectin, vitronectin, an event critical in the process of
bone resorption [132, 133]. Treatment of avian osteoclast precursors with 1,25(OH)2D3 increases αvβ3 expression [134]. The promoter sequence comprises three hexameric direct repeat half-sites separated by three and nine nucleotides to which, respectively, VDR/RXR and RAR/ RXR heterodimers bind [135]. Another 1,25(OH)2D3-regulated protein is the nonreceptor tyrosine kinase pp60c-src [136], that probably plays an important role in bone resorption since c-src-deficient mice develop osteopetrosis [137]. Chapel et al. [138] showed that 1,25(OH)2D3 accelerates pp60c-src kinasespecific activity without affecting its expression. 1,25(OH)2D3 also increases mRNA levels of carbonic anhydrase-II (CA-II) that produces hydrogen protons necessary to acidify the resorption lacuna. A VDRE consensus sequence has been identified in the promoter of the chicken CA-II gene [139], but the VDR/RXR complex binds to it with an inverse polarity compared to the classical VDRE. In contrast, the 1,25(OH)2D3-induced CA-II expression in murine marrow cultures is suggested not to be the result of increased transcription but rather a part of the differentiation process [140].
E. Effect on Chondrocytes Chondrocytes express the VDR and 1,25(OH)2D3 has a biphasic effect on chondrocyte proliferation, being stimulatory at physiological concentrations (10−12 M) and inhibitory at high concentrations (10−8 M) [141, 142]. Characteristics of chondrocyte differentiation, such as alkaline phosphatase expression, collagen production, and matrix calcification, are affected by 1,25(OH)2D3 treatment in vitro [142, 143]. The expression of and signaling by several factors modulating chondrocyte development, such as IGF-I and IGF-I receptor, is influenced by 1,25(OH)2D3 [142]. 1,25(OH)2D3 has also been reported to increase apoptosis in mouse hypertrophic chondrocytes [144].
V. PATHOLOGY AND THERAPY RELATED TO VITAMIN D AVAILABILITY, METABOLISM, AND FUNCTION A. Pathology Vitamin D is either primarily or secondarily involved in many metabolic bone diseases. Abnormalities in vitamin D availability, metabolism, or action are responsible for calciopenic rickets/osteomalacia. During development, defective chondrocyte mineralization occurs at the growth plates and contributes to disorganization of chondrocyte
315
Chapter 18 The Vitamin D Hormone and its Nuclear Receptor
arrangement. This leads to epiphyseal widening, a typical hallmark of rickets, retardation of longitudinal bone growth and skeletal deformities. In osteomalacia, the characteristic pathology changes involve impaired mineralization of bone matrix or osteoid during bone remodeling, manifested by a decrease in the mineralizing surface and mineralization rate. This results in the accumulation of large surfaces of unmineralized osteoid. Focus will be on the types of rickets or osteomalacia caused by genetic defect while diseases that are acquired or secondary to renal or gastrointestinal dysfunction will only be briefly mentioned. 1. Vitamin D Deficiency
Classic nutritional rickets, caused by the simultaneous deprivation of sunlight exposure and dietary vitamin D [145] (Table I) has become uncommon because this vitamin is frequently added to dairy products. In the absence of such health policies, mild or severe cases of rickets are still endemic, especially in countries where calcium intake is low (e.g. northern China, many African countries) and/or exposure to sunshine is avoided for cultural reasons. Vitamin D deficiency is not a primary cause of osteoporosis but it can be secondarily involved. Indeed a low vitamin D status is quite common in the elderly and by inducing secondary hyperparathyroidism it certainly accelerates the negative balance of bone turnover [146]. Moreover, vitamin D metabolism or action is frequently disturbed in secondary osteoporosis. For instance, this fat-soluble vitamin is subject to malabsorption, and levels may be low in chronic liver disease and primary biliary cirrhosis. Chronic renal failure results in renal rickets and secondary hyperparathyroidism when compromised renal mass reduces 1α-hydroxylase activity. Some anticonvulsant drugs increase the turnover of vitamin D into inactive compounds, resulting in a decrease in serum levels of 1,25(OH)2D3.
Table I. 1.
Substrate availability
2.
Metabolism Genetic
Acquired 3.
Vitamin D action Genetic
Tumor-induced osteomalacia is a syndrome of renal phosphate wasting and low or inappropriate normal levels of 1,25(OH)2D3 [147]. 2. Hereditary Defects
Genetic defects causing calciopenic rickets include disorders with abnormal vitamin D metabolism or resistance to the effects of the active form of vitamin D. Genetic defects are found at several steps of this pathway but rickets are part of the clinical picture in only two disorders. A specific hereditary defect in the gene coding for the 25-hydroxyvitamin D-1α hydroxylase (1α-hydroxylase) or CYP27B1 impairs the final, critical step in the biosynthesis of 1,25(OH)2D3 resulting in vitamin D-dependent rickets type I (VDDR-I), also called pseudo-vitamin D-deficiency (PDDR) [148, 149]. This disease is characterized by failure to thrive, muscle weakness, skeletal deformities, hypocalcemia, secondary hyperparathyroidism and low serum 1,25(OH)2D3 concentrations caused by impaired activity of the 1α-hydroxylase [148]. Several mutations in the VDR are responsible for vitamin D-dependent rickets type II (VDDR-II), also called hereditary hypocalcemic vitamin D-resistant rickets (HVDRR) [150, 151]. The greatest number of these mutations can be localized to the DNA-binding zinc finger region and a smaller group to the ligand-binding domain, some of them affecting the heterodimerization of VDR with RXR. HVDRR results in a phenotype characterized by early-onset rickets, hypocalcemia, secondary hyperparathyroidism, and markedly increased 1,25(OH)2D3 levels. The alopecia observed in some kindreds with mutant VDRs has not been observed in vitamin D deficiency, suggesting that VDR not only mediates the bone mineral homeostatic actions of vitamin D but also plays a role in the differentiation of hair follicles. These patients do not respond to physiological
Pathology: Vitamin D-Related Calcium/Bone Diseases
Mechanism ↓ Intake ↓ Sun exposure
Pathology Rickets/Osteomalacia
DBP CYP24
KO mice KO mice
CYP27B1
KO mice Human-chr12q13
Chronic renal failure Oncogenic osteomalacia VDR
KO mice Human-chr12q13
More susceptible to osteomalacia Defect in intramembranous ossification Rickets HVDRR I Renal osteodystrophy Osteomalacia and hypophosphatemia Rickets HVDRR II
316
GEERT CARMELIET ET AL .
doses of 1,25(OH)2D3, but the bone pathology is cured by frequent intravenous infusions of calcium [152] and is prevented by high oral doses of calcium [153]. Both of these two diseases display an autosomalrecessive trait, but clinical features and response to administrated 1,25(OH)2D3 are distinct. The phenotypes of mice with targeted ablation of 1α-hydroxylase or VDR exhibit the clinical abnormalities observed in the VDDR-I and -II patients respectively and offered opportunity to investigate in vivo functions of 1,25(OH)2D3 and the molecular basis of its actions. a. VDR-Null Mice: Model of Vitamin D-dependent Rickets Type II a.1. Mineral homeostasis and rickets of VDR-null mice VDR-null or VDR-mutant mice have been created by four different groups [131, 154–156]. They are born phenotypically normal, but show growth retardation starting after weaning. At that age, hypocalcemia, hyperparathyroidism, and hypophosphatemia develop and 1,25(OH)2D3 serum levels become 10- to 100-fold elevated, coincident with extremely low 24,25(OH)2D3 serum levels. Inspection and X-ray analysis of long bones showed a reduced length and the typical features of rickets including widening of epiphyseal growth plates, thinning of the cortex, cupping, and widening of the metaphysis. The time of onset of symptoms at 3 weeks of age suggests that the vitamin D endocrine system is principally required for maintaining bone mineral homeostasis when the organism is deprived of a consistent and plentiful supply of calcium, such as occurs after weaning in mammals. Amling and coworkers [157] studied histomorphometric and biomechanical parameters in Boston VDR-null mice, fed regular chow. They found by the age of 70 days an expanded length of
Figure 4
the growth plate and disorganization of the chondrocyte columns, a marked increase in bone volume due to increased osteoid, an increased number of osteoblasts, but normal number of osteoclasts, impaired mineral apposition rate, and reduced stiffness. However, normalization of mineral ion homeostasis (calcemia and phosphatemia) by feeding these mice a diet enriched with calcium, lactose, and phosphorus (the socalled “rescue diet”) prevented hyperparathyoidism and rickets [157, 158]. These findings are consistent with the therapeutic effect of high doses of calcium in VDRR type II (see Section V.A.2). In addition, the ratio of Ca/P in the diet is a major determinant for rescue of bone mineralization and turnover in VDR-null mice, with the optimal ratio being a factor of 2 [159, 160]. The underlying mechanism is suggested to be a Ca/P ratio-dependent intestinal transport of calcium and phosphorus. These data suggest that, in an ideal metabolic environment, the VDR is not essential for the development or maintenance of normal bone, although 1,25(OH)2D3 has been shown to have significant effects on the expression of genes by osteoblasts and osteoclasts. Thus the principal action of the VDR is its role in intestinal calcium absorption and/or renal calcium reabsorption. Intestinal active calcium absorption, assessed in two VDR-null strains, was reduced to one third of normal mice [155]. At the molecular level only the expression of the epithelial calcium channels TRPV6 and TRPV5 was severely impaired in the absence of a functional VDR, with less effect on calbindin-D9K and the plasma membrane calcium ATPase (PMCAIb), [155] (Fig. 4). The urinary calcium excretion in VDR-null mice under a normal diet is inappropriately high given the hypocalcemia
Gene targets of 1,25(OH)2D3 action in the enterocyte. In the absence of a functional VDR, active intestinal calcium absorption is impaired reflected at the molecular level by severely decreased gene expression of TRPV6 but only modest decreases in calbindin-D9K and PMCA1b expression. The thickness of the arrow reflects VDR-related changes.
Chapter 18 The Vitamin D Hormone and its Nuclear Receptor
in these mice [156, 161]. Normalization of serum calcium levels by the rescue diet resulted in a manifest increase of calcium excretion suggesting that renal tubular calcium reabsorption is impaired in VDR-null mice. The parameter that correlated best herewith was the expression of calbindin-D9K which was found to be decreased in all strains of VDR-null mice, both on the normal and the rescue diet [131, 155, 156, 161, 162]. a.2. Mechanisms underlying the disorganization of growth plate Rickets or the expansion and disorganization of the growth plate are a characteristic feature of VDR inactivation (Fig. 5). Analysis of growth plate morphology demonstrated normal resting and proliferating chondrocyte layers, but expansion of the hypertrophic chondrocyte layer. This latter finding could potentially be a consequence of an osteoclast defect, since 1,25(OH)2D3 has been shown to be a potent inducer of RANKL produced by osteoblasts (see Section IV.D). However, histomorphometric analyses have demonstrated normal numbers of osteoclasts in the metaphyses of several strains of VDR-null mice, indicating that other osteoclast-activating factors such as PTH and interleukin-1α are capable of inducing the formation of mature resorbing osteoclasts in vivo in the absence of VDR and that the VDR is not essential for osteoclastmediated bone resorption. This also suggests that the expansion of the hypertrophic chondrocyte layer is not secondary to an osteoclast defect. Osteoclasts are brought to the hypertrophic chondrocytes concomitant with vascular invasion. This process is initiated by secretion of VEGF, a signaling molecule produced by hypertrophic chondrocytes. No impairment of VEGF RNA production by these
Figure 5
317
cells was observed in VDR-null mice, indicating that hypertrophic chondrocytes were capable of signaling vascular invasion [163]. Alternatively, the expansion of the hypertrophic chondrocyte layer may be due to altered chondrocyte development in VDR-null mice, but present data are not conclusive. A recent study shows that in VDR-null mice, this feature is not secondary to increased chondrocyte proliferation or impaired differentiation but rather a consequence of decreased apoptosis [163]. The authors also postulate that hypophosphatemia, rather than hypocalcemia, is the major determinant of decreased chondrocyte apoptosis. However, excess PTH secretion in these mice might mimic the PTHrP paracrine effects on chondrocyte survival [164]. In contrast, no manifest alteration in the number of apoptotic cells or in the expression of apoptosis-related genes was detected in another VDR-null strain [165]. In addition, a functionally redundant VDR may be present. To investigate this possibility, mice deficient in both VDR and RXRγ were generated since RXRγ-null mice exhibit no discernible abnormalities [165]. The phenotype observed in double-null mutant mice was basically similar to those in VDR-null mice, except that growth plate development was more severely impaired and was not prevented by dietary supplementation. A selective impairment of hypertrophic chondrocyte development was observed despite normalized mineral homeostasis produced by the high-mineral diet. These findings indicate that the combined actions of VDR- and RXRγmediated signals are essential for normal development of growth plate cartilage and might be explained by a functional,
Molecular mechanisms of bone pathology in VDR-null mice. Inactivation of VDR leads to increased bone volume, mainly due to increased osteoid, which normalizes when a rescue diet with an adequate Ca/P ratio is given. The disorganization of the growth plate in VDR-null mice is not due to defective osteoclast formation or chondrocyte proliferation as both processes are normal in these mice. VEGF gene expression by hypertrophic chondrocytes is normal, suggesting normal vascular invasion of the growth plate. Rather, apoptosis of hypertrophic chondrocytes is impaired, possibly as a consequence of hypophosphatemia, although the effect of increased PTH levels can not be excluded.
318 but redundant, VDR-like receptor in growth plate chondrocytes. b. 1a-Hydroxylase-Null Mice: Model of Vitamin DDependent Type Rickets I Mice with inactivation of 1a-hydroxylase(1α(OH)ase) were generated and demonstrate failure to thrive [166, 167]. Regarding calcium homeostasis, 1α(OH)ase-null models demonstrate hypocalcemia, severe secondary hyperparathyroidism, hypophosphatemia, elevated serum alkaline phosphatase, and increased phosphaturia, analogous to the phenotype of the VDR-null mice. However, 1α(OH)ase-null mice genuinely lack the capacity to synthesize 1,25(OH)2D3: circulating levels of 1,25(OH)2D3 were undetectable, whereas circulating levels of 25(OH)D3 doubled. This emphasizes the non-redundant role played by the renal 1α(OH)ase in producing the hormonally active metabolite of vitamin D. Furthermore, this confirms that the phenotype of the 1α(OH)ase-null mice matches the clinical manifestations of VDDR-I [168]. Typical features of advanced rickets are observed histologically in bone of 1α(OH)ase-null mice: disorganization and widening of the columnar alignment of hypertrophic chondrocytes, resulting in increased width of the growth plate, impaired calcification of the hypertrophic cartilage and accumulation of osteoid in trabecular and cortical bone. Feeding the 1α(OH)ase-null mice a rescue diet normalizes serum calcium levels and trabecular bone volume, but not the width of the growth plate [169]. A phenotypic rescue, even with normalization of the growth plate, was also seen after treating the 1α(OH)ase-null mice with 1,25(OH)2D3, the treatment of choice for VDRR-I therapy.
B. 1,25(OH)2D3 and Analogs as Therapeutics in Bone Pathology The role of 1,25(OH)2D3 in the treatment of bone disorders characterized by defective mineralization, such as rickets and renal osteodystrophy, is well established. Moreover prevention or correction of subclinical 1,25(OH)2D3 deficiency can substantially reduce the incidence of hip and other fractures in senile osteoporosis, at least when given in combination with oral calcium supplementation [170]. The major goal is the reduction of PTH secretion and the subsequent decreased bone resorption. However, 1,25(OH)2D3 is also suggested as therapeutic for osteoporosis, a disorder characterized by decreased bone formation or at least a discrepancy between bone formation and resorption [171]. Indeed, a large New Zealand study demonstrated a major reduction in the number of new vertebral fractures during 1,25(OH)2D3
GEERT CARMELIET ET AL .
therapy [172]. However, the optimal therapeutic range for 1,25(OH)2D3 appears to be quite narrow and may vary among patients. An alternative therapy can be established with the vitamin D3 analog, 1α(OH)D3 (alfacalcidiol). 1α(OH)D3 is less effective on intestinal calcium absorption compared with its effect on bone because it must first be hydroxylated at the 25-position to become fully active [173]. Nevertheless, therapeutic dose monitoring is also needed to avoid side effects such as hypercalcemia and hypercalciuria. Efforts to develop new synthetic analogs for bone diseases have been modest. Presently some analogs are being evaluated in preclinical or human clinical phase II trials for the prevention or cure of osteoporosis. The vitamin D analog, 2β-(3-hydroxypropoxy)-1,25(OH)2D3 (ED-71), is characterized by high calcemic activity and strong binding to DBP (two-fold that of 1,25(OH)2D3) [174]. In normal or ovariectomized rats, ED-71 stimulates bone mass and bone formation rates [175] and is a more potent inhibitor of bone resorption than 1α(OH)D3 [176]. ED-71 is in final stages of clinical osteoporosis trials. Several other compounds such as 1α-(OH)D2 and 19-nor-1,25(OH)2D3 analogs as well as 2-methylene-19nor-20-epi-1α,25(OH)2D3 (2MD); 1αfluoro-25(OH)16,23diene-26,27bishomo-20-epi vitamin D3 (Ro 26-9228), and GS 1790 are now being evaluated for the same purpose in preclinical or early clinical settings. In vitro studies demonstrated that the analog 2MD was 100 times more potent in osteoclast-inducing activity compared with 1,25(OH)2D3, probably related to its induction of increased VDR interaction with RXR and co-activators (SRC-1 and DRIP 205) [177]. 2MD also induced numerous mineralized bone nodules (10−12 M) in primary human osteoblast precursors and fetal mouse calvarial cells, whereas 1α,25(OH)2D3 exhibited little effect even at 10−8 M [178]. 2MD increased total body bone mineral density in ovariectomized rats in a 23-week period, whereas 1,25(OH)2D3 only prevented the bone loss [178]. Three weeks of oral treatment of ovariectomized rats with Ro 269228 had a bone-protecting effect without inducing hypercalcemia. Analysis of biochemical and bone histomorphometrical indices suggested that Ro 26-9228 inhibited bone resorption and increased the number of differentiated osteoblasts. Gene expression was differentially regulated in duodenum and bone in analog-treated rats [179]. This tissue-selective action of the analog has been explained by significant interactions of Ro 26-9228-VDR with RXR and cofactors such as GRIP and DRIP in osteoblasts but much less in intestinal CaCo-2 cells [180]. Bone disease in patients with chronic renal failure is due to a complex set of mechanisms such as impaired 1,25(OH)2D3 synthesis, vitamin D resistance, secondary hyperparathyroidism, and abnormal mineral handling
319
Chapter 18 The Vitamin D Hormone and its Nuclear Receptor
(hyperphosphatemia, aluminum or fluoride excess, acidosis). Whereas 1α-(OH)D3 and 1,25(OH)2D3 are widely used for the prevention and cure of renal osteodystrophy, several analogs have been evaluated for this indication with the aim of better PTH suppression with less risk for inducing hypercalcemia or hyperphosphatemia [181]. In the United States, 19-nor-1,25(OH)2D2 (paricalcitol, Zemplar®) and 1α(OH)D2 (doxercalciferol, Hecterol®) are available for the treatment of secondary hyperparathyroidism in renal-failure patients. 22-oxacalcitriol (OCT, maxacalcitol, Oxarol®) and 1,25-dihydroxy-26,26,26, 27,27,27-hexafluorovitamin D3 (falecalcitriol, Hornel®, and Fulstan®) have been approved in Japan. 22-Oxacalcitriol has similar effects as 1,25(OH)2D3 on cell proliferation and inhibition of PTH secretion in vitro and in vivo [182] in experimental animals and patients with chronic renal failure [183]. Similarly, the vitamin D2 analogs 19-nor1,25(OH)2D2 [184] and 1α(OH)D2 [185] and the analog falecalcitriol [186] have been evaluated in vivo for suppressing PTH levels, together with the effects on calcium and phosphate levels [187]. A double-blind, randomized, multicenter study, comparing the efficacy of paricalcitol and calcitriol in renal disease patients undergoing hemodialysis demonstrated that paricalcitol reduced PTH concentration more rapidly, with fewer sustained episodes of either hypercalcemia or increased Ca × P product than 1,25(OH)2D3 therapy [188]. An uncontrolled, retrospective assessment of a clinical data base of about 67000 patients undergoing hemodialysis and receiving either paricalcitol or calcitriol, demonstrated that paricalcitol was associated with a lower mortality rate (especially by cardiovascular events) over the 36-month follow-up period (18%) than calcitriol (22.3%). However, such an uncontrolled study has serious limitations and properly designed, randomized, controlled trials are still needed to confirm these data [189].
VI. CONCLUSIONS Insight into the molecular mechanisms of 1,25(OH)2D3 action, via binding to VDR and interacting with VDRE promoting altered transcriptional activity, has enlarged considerably during the last years. Many elements of this complex machinery are not yet fully identified; the mechanism of negatively regulated genes by 1,25(OH)2D3 especially remains an open question. Certainly the participation of co-activators or co-repressors has been established and their functional significance elucidated. In addition, the knowledge of the three-dimensional structure of DBP and of the VDR provided valuable information for the understanding of ligand–protein and protein–protein interactions.
The development of several KO mice (DBP, megalin, VDR, 1α-hydroxylase, and 24-hydroxylase) strengthens the importance of 1,25(OH)2D3–VDR system in mineral and bone homeostasis. The exact function of 1,25(OH)2D3 in bone metabolism awaits new physiological in vivo models with tissue-specific inactivation of the gene of interest.
References 1. Calli, J. J., and Russell, D. W. (1991). Characterization of human sterol 27-hydroxylase. A mitochondrial cytochrome P-450 that catalyzes multiple oxidation reaction in bile acid biosynthesis. J. Biol. Chem. 266, 7774–7778. 2. Rosen, H., Reshef, A., Maeda, N., Lippoldt, A., Shpizen, S., Triger, L., Eggertsen, G., Bjorkhem, I., and Leitersdorf, E. (1998). Markedly reduced bile acid synthesis but maintained levels of cholesterol and vitamin D metabolites in mice with disrupted sterol 27-hydroxylase gene. J. Biol. Chem. 273, 14805–14812. 3. Cheng, J. B., Levine, M. A., Bell, N. H., Mangelsdorf, D. J., and Russell, D. W. (2004). Genetic evidence that the human CYP2R1 enzyme is a key vitamin D 25-hydroxylase. Proc. Natl Acad. Sci. USA 101, 7711–7715. 4. St-Arnaud, R., Messerlian, S., Moir, J. M., Omdahl, J. L., and Glorieux, F. H. (1997). The 25-hydroxyvitamin D 1-alpha-hydroxylase gene maps to the pseudovitamin D-deficiency rickets (PDDR) disease locus. J. Bone Miner. Res. 12, 1552–1559. 5. Takeyama, K., Kitanaka, S., Sato, T., Kobori, M., Yanagisawa, J., and Kato, S. (1997). 25-hydroxyvitamin D3 1α-hydroxylase and vitamin D synthesis. Science 277, 1827–1830. 6. Monkawa, T., Yoshida, T., Wakino, S., Shinki, T., Anazawa, H., DeLuca, H. F., Suda, T., Hayashi, M., and Saruta, T. (1997). Molecular cloning of cDNA and genomic DNA for human 25-hydroxyvitamin D3 1α-hydroxylase. Biochem. Biophys. Res. Comm. 239, 527–533. 7. Shinki, T., Shimada, H., Wakino, S., Anazawa, H., Hayashi, M., Saruta, T., DeLuca, H. F., and Suda, T. (1997). Cloning and expression of rat 25-hydroxyvitamin D3-1α-hydroxylase cDNA. Proc. Natl Acad. Sci. USA 94, 1290–1295. 8. Fu, G. K., Lin, D., Zhang, M. Y., Bikle, D. D., Shackleton, C. H. L., Miller, W. L., and Portale, A. A. (1997). Cloning of human 25-hydroxyvitamin D-1α-hydoxylase and mutations causing vitamin Ddependent rickets type 1. Molec. Endocrinol. 11, 1961–1970. 9. Verboven, C., Rabijns, A., De Maeyer, M., Van Baelen, H., Bouillon, R., and De Ranter, D. (2002). A structural basis for the unique binding features of the human vitamin D-binding protein. Nat. Struct. Biol. 9, 131–136. 10. Safadi, F. F., Thornton, P., Magiera, H., Hollis, B. W., Gentile, M., Haddad, J. G., Liebhaber, S. A., and Cooke, N. E. (1999). Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J. Clin. Invest. 103, 239–251. 11. Nykjaer, A., Dragun, D., Walther, D., Vorum, H., Jacobsen, C., Herz, J., Melsen, F., Christensen, E. I., and Willnow, T. E. (1999). An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) Vitamin D3. Cell 96, 507–515. 12. Leheste, J. R., Melsen, F., Wellner, M., Jansen, P., Schlichting, U., Renner-Müller, I., Andreassen, T. T., Wolf, E., Bachmann, S., Nykjaer, A., and Willnow, T. E. (2003). Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J. 17, 247–249. 13. Ohyama, Y., Noshiro, M., Eggertsen, G., Gotoh, O., Kato, Y., Bjorkhem, I., and Okuda, K. (1993). Structural characterization of the
320
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
gene encoding rat 25-hydroxyvitamin D3 24-hydoxylase. Biochemistry 32, 76–82. Lohnes, D., and Jones, G. (1987). Side chain metabolism of vitamin D3 in osteosarcoma cell line UMR-106. Characterization of products. J. Biol. Chem. 262, 14394–14401. Reddy, G. S., and Tserng, K. Y. (1989). Calcitroic acid, end product of renal metabolism of 1,25-dihydroxyvitamin D3 through C24 oxidation pathway. Biochemistry 28, 1763–1769. Akiyoshi-Shibata, M., Sakaki, T., Ohyama, Y., Noshiro, M., Okuda, K., and Yabusaki, Y. (1994). Further oxidation of hydroxycalcidiol by calcidiol 24-hydroxylase. A study with the mature enzyme expressed in Escherichia Coli. Eur. J. Biochem. 224, 335–343. St.-Arnaud, R., Arabian, A., Travers, R., Barletta, F., Raval-Pandya, M., Chapin, K., Depovere, J., Mathieu, C., Christakos, S., Demay, M. B., and Glorieux, F. H. (2000). Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase-ablated mice is due to elevated 1,25dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141, 2658–2666. Yamaura, M., Nakamura, T., Nagai, Y., Yoshishara, A., and Suzuki, K. (1993). Reduced mechanical competence of bone by ovariectomy and its prevention with 24R,25-dihydroxyvitamin D3 administration in beagles. Calcif. Tissue Int. 25, 49–56. Schwartz, Z., Bonewald, L. F., Caulfield, K., Brooks, B., and Boyan, B. D. (1993). Direct effects of transforming growth factor-beta on chondrocytes are modulated by vitamin D metabolites in a cell maturation-specific manner. Endocrinology 132, 1544–1552. Mangelsdorf, D. J., Thummel, C., Beato, M., Herrlich, P., Schütz, G., Umesono, K., Blumberg, B., Kastner, P., Mark, M., Chambon, P., and Evans, R. M. (1995). The nuclear receptor superfamily: the second decade. Cell 83, 835–839. Li, Y. C., Bergwitz, C., Jüppner, H., and Demay, M. B. (1997). Cloning and characterization of the vitamin D receptor from Xenopus laevis. Endocrinology 138, 2347–2353. McDonnell, D. P., Mangelsdorf, D. J., Pike, J. W., Haussler, M. R., and O’Malley, B. W. (1987). Molecular cloning of complementary DNA encoding the avian receptor for vitamin D. Science 235, 1214–1217. Elaroussi, M. A., Prahl, J. M., and DeLuca, H. F. (1994). The avian vitamin D receptors: primary structures and their origins. Proc. Natl Acad. Sci. USA 91, 11596–11600. Kamei, Y., Kawada, T., Fukuwatari, T., Ono, T., Kato, S., and Sugimoto, E. (1995). Cloning and sequencing of the gene encoding the mouse vitamin D receptor. Gene 152, 281–282. Burmester, J., Wiese, R., Maeda, N., and DeLuca, H. F. (1988). Structure and regulation of the rat 1,25-dihydroxyvitamin D3 receptor. Proc. Natl Acad. Sci. USA 85, 9499–9502. Baker, A. R., McDonnell, D. P., Hughes, M., Crisp, T. M., Mangelsdorf, D. J., Haussler, M. R., Pike, J. W., Shine, J., and O’Malley, B. W. (1988). Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc. Natl Acad. Sci. USA 85, 3294–3298. Suzuki, T., Suzuki, N., Srivastava, A. S., and Kurokawa, T. (2000). Identification of cDNAs encoding two subtypes of vitamin D receptor in flounder, Paralichthys olivaceus. Biochem. Biophys. Res. Commun. 270, 40–45. Sugiura, S. H., and Ferraris, R. P. (2004). Dietary phosphorus-responsive genes in the intestine, pyloric ceca, and kidney of rainbow trout. Am. J. Physiol. 287, R541–550. Whitfield, G. K., Dang, H. T., Schluter, S. F., Bernstein, R. M., Bunag, T., Manzon, L. A., Hsieh, G., Dominguez, C. E., Youson, J. H., Haussler, M. R., and Marchalonis, J. J. (2003). Cloning of a functional vitamin D receptor from the lamprey (Petromyzon marinus), an ancient vertebrate lacking a calcified skeleton and teeth. Endocrinology 144, 2704–2716. Miyamoto, K., Kesterson, R. A., Yamamoto, H., Taketani, Y., Nishiwaki, E., Tatsumi, S., Inoue, Y., Morita, K., Takeda, E., and
GEERT CARMELIET ET AL .
31.
32.
33.
34.
35.
36. 37. 38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
Pike, J. W. (1997). Structural organization of the human vitamin D receptor chromosomal gene and its promoter. Mol. Endocrinol. 11, 1165–1179. Crofts, L. A., Hancock, M. S., Morrison, N. A., and Eisman, J. A. (1998). Multiple promoters direct the tissue-specific expression of novel N-terminal variant human vitamin D receptor gene transcripts. Proc. Natl Acad. Sci. USA. 95, 10529–10534. Sunn, K. L., Cock, T. A., Crofts, L. A., Eisman, J. A., and Gardiner, E. M. (2001). Novel N-terminal variant of human VDR. Mol. Endocrinol. 15, 1599–1609. Jurutka, P. W., Remus, L. S., Whitfield, G. K., Thompson, P. D., Hsieh, J. C., Zitzer, H., Tavakkoli, P., Galligan, M. A., Dang, H. T., Haussler, C. A., and Haussler, M. R. (2000). The polymorphic N terminus in human vitamin D receptor isoforms influences transcriptional activity by modulating interaction with transcription factor IIB. Mol. Endocrinol. 14, 401–420. Ames, S. K., Ellis, K. J., Gunn, S. K., Copeland, K. C., and Abrams, S. A. (1999). Vitamin D receptor gene Fok1 polymorphism predicts calcium absorption and bone mineral density in children. J. Bone Miner. Res. 14, 740–746. Harris, S. S., Eccleshall, T. R., Gross, C., Dawson-Hughes, B., and Feldman, D. (1997). The vitamin D receptor start codon polymorphism (FokI) and bone mineral density in premenopausal American black and white women. J. Bone Miner. Res. 12, 1043–1048. Khorasanizadeh, S., and Rastinejad, F. (2001). Nuclear-receptor interactions on DNA-response elements. Trends Biochem. Sci. 26, 384–390. Freedman, L. P. (1992). Anatomy of the steroid receptor zinc finger region. Endocr. Rev. 13, 129–145. Rastinejad, F., Perlmann, T., Evans, R. M., and Sigler, P. B. (1995). Structural determinants of nuclear receptor assembly on DNA direct repeats. Nature 375, 203–211. Shaffer, P. L., and Gewirth, D. T. (2002). Structural basis of VDR-DNA interactions on direct repeat response elements. EMBO J. 21, 2242–2252. Luo, Z., Rouvinen, J., and Maenpaa, P. H. (1994). A peptide C-terminal to the second Zn finger of human vitamin D receptor is able to specify nuclear localization. Eur. J. Biochem. 223, 381–387. Hsieh, J. C., Shimizu, Y., Minoshima, S., Shimizu, N., Haussler, C. A., Jurutka, P. W., and Haussler, M. R. (1998). Novel nuclear localization signal between the two DNA-binding zinc fingers in the human vitamin D receptor. J. Cell. Biochem. 70, 94–109. Michigami, T., Suga, A., Yamazaki, M., Shimizu, C., Cai, G., Okada, S., and Ozono, K. (1999). Identification of amino acid sequence in the hinge region of human vitamin D receptor that transfers a cytosolic protein to the nucleus. J. Biol. Chem. 274, 33531–33538. Allegretto, E. A., Pike, J. W., and Haussler, M. R. (1987). C-terminal proteolysis of the avian 1,25-dihydroxyvitamin D3 receptor. Biochem. Biophys. Res. Commun. 147, 479–485. Rochel, N., Wurtz, J. M., Mitschler, A., Klaholz, B., and Moras, D. (2000). Crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol. Cell. 5, 173–179. Bouillon, R., Okamura, W. H., and Norman, A. W. (1995). Structurefunction relationships in the vitamin D endocrine system. Endocr. Rev. 16, 200–257. Choi, M., Yamamoto, K., Masuno, H., Nakashima, K., Taga, T., and Yamada, S. (2001). Ligand recognition by the vitamin D receptor. Bioorg. Med. Chem. 9, 1721–1730. Sone, T., Ozono, K., and Pike, J. W. (1991). A 55-kilodalton accessory factor facilitates vitamin D receptor DNA binding. Mol. Endocrinol. 5, 1578–1586. Kliewer, S. A., Umesono, K., Mangelsdorf, D. J., and Evans, R. M. (1992). Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone and vitamin D3 signaling. Nature 355, 446–449. Carlberg, C., Bendik, I., Wyss, A., Meier, E., Sturzenbecker, L. J., Grippo, J. F., and Hunziker, W. (1993). Two nuclear signalling pathways for vitamin D. Nature 361, 657–660.
Chapter 18 The Vitamin D Hormone and its Nuclear Receptor 50. Christakos, S., Raval-Pandya, M., Werny, J., and Yang, W. (1996). Genomic mechanisms involved in the pleiotropic actions of 1,25dihydroxyvitamin D3. Biochem. J. 316, 361–371. 51. Carlberg, C. (1995). Mechanisms of nuclear signalling by vitamin D3. Interplay with retinoid and thyroid hormone signalling. Eur. J. Biochem. 231, 517–527. 52. Jin, C. H., Kerner, S. A., Hong, M. H., and Pike, J. W. (1996). Transcriptional activation and dimerization functions in the human vitamin D receptor. Mol. Endocrinol. 10, 945–957. 53. Schrader, M., Nayeri, S., Kahlen, J. P., Muller, K. M., and Carlberg, C. (1995). Natural vitamin D3 response elements formed by inverted palindromes: polarity-directed ligand sensitivity of vitamin D3 receptor-retinoid X receptor heterodimer-mediated transactivation. Mol. Cell. Biol. 15, 1154–1161. 54. Haussler, M. R., Whitfield, G. K., Haussler, C. A., Hsieh, J. C., Thompson, P. D., Selznick, S. H., Dominguez, C. E., and Jurutka, P. W. (1998). The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J. Bone Miner. Res. 13, 325–349. 55. Kimmel-Jehan, C., Darwish, H. M., Strugnell, S. A., Jehan, F., Wiefling, B., and DeLuca, H. F. (1999). DNA bending is induced by binding of vitamin D receptor-retinoid X receptor heterodimers to vitamin D response elements. J. Cell. Biochem. 74, 220–228. 56. Bettoun, D. J., Burris, T. P., Houck, K. A., Buck, D. W. 2nd, Stayrook, K. R., Khalifa, B., Lu, J., Chin, W. W., and Nagpal, S. (2003). Retinoid X receptor is a nonsilent major contributor to vitamin D receptor-mediated transcriptional activation. Mol. Endocrinol. 17, 2320–2328. 57. Rachez, C., and Freedman, L. P. (2000). Mechanisms of gene regulation by vitamin D3 receptor: a network of coactivator interactions. Gene. 246, 9–21. 58. Rachez, C., Gamble, M., Chang, C. P., Atkins, G. B., Lazar, M. A., and Freedman, L. P. (2000). The DRIP complex and SRC-1/p160 coactivators share similar nuclear receptor binding determinants but constitute functionally distinct complexes. Mol. Cell. Biol. 20, 2718–2726. 59. Auboeuf, D., Honig, A., Berget, S. M., and O’Malley, B. W. (2002). Coordinate regulation of transcription and splicing by steroid receptor coregulators. Science 298, 416–419. 60. Kitagawa, H., Fujiki, R., Yoshimura, K., Mezaki, Y., Uematsu, Y., Matsui, D., Ogawa, S., Unno, K., Okubo, M., Tokita, A., Nakagawa, T., Ito, T., Ishimi, Y., Nagasawa, H., Matsumoto, T., Yanagisawa, J., and Kato, S. (2003). The chromatin-remodeling complex WINAC targets a nuclear receptor to promoters and is impaired in Williams syndrome. Cell 113, 905–917. 61. MacDonald, P. N., Sherman, D. R., Dowd, D. R., Jefcoat, S. C. Jr., and DeLisle, R. K. (1995). The vitamin D receptor interacts with general transcription factor IIB. J. Biol. Chem. 270, 4748–4752. 62. Blanco, J. C. G., Wang, I. M., Tsai, S. Y., Tsai, M. J., O’Malley, B. W., Jurutka, P. W., Haussler, M. R., and Ozato, K. (1995). Transcription factor TFIIB and the vitamin D receptor cooperatively activate liganddependent transcription. Proc. Natl Acad. Sci. USA 92, 1535–1539. 63. Masuyama, H., Jefcoat, S. C. Jr., and MacDonald, P. N. (1997). The N-terminal domain of trancription Factor IIB is required for direct interaction with the vitamin D receptor and participates in vitamin Dmediated transcription. Mol. Endocrinol. 11, 218–228. 64. Kurihara, N., Reddy, S. V., Araki, N., Ishizuka, S., Ozono, K., Cornish, J., Cundy, T., Singer, F. R., and Roodman, G. D. (2004). Role of TAFII-17, a VDR binding protein, in the increased osteoclast formation in Paget’s Disease. J. Bone Miner. Res. 19, 1154–1164. 65. Brown, T. A., and DeLuca, H. F. (1990). Phosphorylation of the 1,25-dihydroxyvitamin D3 receptor. A primary event in 1,25dihydroxyvitamin D3 action. J. Biol. Chem. 265, 10025–10029. 66. Jurutka, P. W., Hsieh, J-C., Nakajima, S., Haussler, C. A., Whitfield, G. K., and Haussler, M. R. (1996). Human vitamin D receptor phosphorylation by casein kinase II at Ser-208 potentiates transcriptional activation. Proc. Natl Acad. Sci. USA 93, 3519–3524.
321
67. Bareletta, F., Freedman, L. P., and Christakos, S. (2002). Enhancement of VDR-mediated transcription by phosphorylation: correlation with increased interaction between the VDR and DRIP205, a subunit of the VDR-interacting protein coactivator complex. Mol. Endocrinol. 16, 301–314. 68. Hsieh, J-C., Jurutka, P. W., Nakajima, S., Galligan, M. A., Haussler, C. A., Shimizu, Y., Shimizu, N., Whitfield, G. K., and Haussler, M. R. (1993). Phosphorylation of the human vitamin D receptor by protein kinase C. Biochemical and functional evaluation of the serine 51 recognition site. J. Biol. Chem. 268, 15118–15126. 69. Nakajima, S., Yamagata, M., Sakai, N., and Ozono, K. (2000). Effect of cyclic adenosine 3′,5′-monophosphate and protein kinase A on ligand-dependent transactivation via the vitamin D receptor. Mol. Cell. Endocrinol. 159, 45–51. 70. Chen, T. L., and Feldman, D. (1981). Regulation of 1,25dihydroxyvitamin D3 receptors in cultured mouse bone cells. Correlation of receptor concentration with the rate of cell division. J. Biol. Chem. 256, 5561–5566. 71. Pols, H. A. P., van Leeuwen, J. P. T. M., Schilte, J. P., Visser, T. J., and Birkenhager, J. C. (1988). Heterologous up-regulation of the 1,25dihydroxyvitamin D3 receptor by parathyroid hormone (PTH) and PTH-like peptide in osteoblast-like cells. Biochem. Biophys. Res. Commun. 156, 588–594. 72. van Leeuwen, J. P. T. M., Birkenhager, J. C., Wijngaarden, T. V. V., vanden Bemd, G. J. C. M., and Pols, H. A. P. (1992). Regulation of 1,25dihydroxyvitamin D3 receptor gene expression by parathyroid hormone and cAMP-agonists. Biochem. Biophys. Res. Commun. 185, 881–886. 73. Staal, A., Van Wijnen, A. J., Desai, R. K., Pols, H. A., Birkenhager, J. C., DeLuca, H. F., Denhardt, D. T., Stein, J. L., van Leeuwen, J. P., Stein, G. S., and Lian, J. B. (1996). Antagonistic effects of transforming growth factor-beta on vitamin D3 enhancement of osteocalcin and osteopontin transcription: reduced interactions of vitamin D receptor/ retinoid X receptor complexes with vitamin D response elements. Endocrinology 137, 2001–2011. 74. Walters, M. R. (1981). An estrogen-stimulated 1,25-dihydroxyvitamin D3 receptor in rat uterus. Biochem. Biophys. Res. Commun. 103, 721–726. 75. Mahonen, A., Pirskanen, A., and Maenpaa, P. H. (1991). Homologous and heterlologous regulation of 1,25-dihydroxyvitmain D-3 receptor mRNA levels in human osteosarcoma cells. Biochim. Biophys. Acta 188, 111–118. 76. Chen, T. L., Cone, C. M., Morey-Holton, E., and Feldman, D. (1983). 1,25-Dihydroxyvitmain D3 receptors in cultured rat osteoblasts-like cells. Glucocorticoid treatment increases receptor content. J. Biol. Chem. 258, 4350–4355. 77. Petkovitch, P. M., Heersche, J. N. M., Tinker, D. O., and Jones, G. (1984). Retinoic acid stimulates 1,25-dihydroxyvitamin D3 binding in rat osteosarcoma cells. J. Biol. Chem. 259, 8274–8290. 78. Krishnan, A. V., and Feldman, D. (1991). Stimulation of 1,25dihydroxyvitamin D3 receptor gene expression in cultured cells by serum and growth factors. J. Bone Miner. Res. 6, 1099–1107. 79. Zhao, X., and Feldman, D. (1993). Regulation of vitamin D receptor abundance and responsiveness during differentiation of HT-29 human colon cancer cells. Endocrinology 132, 1808–1814. 80. Wiese, R. J., Uhland-Smith, A., Ross, T. K., Prahl, J. M., and DeLuca, H. F. (1992). Up-regulation of the vitamin D receptor in response to 1,25-dihydroxyvitamin D3 results from ligand-induced stabilization. J. Biol. Chem. 267, 20082–20086. 81. Arbour, N. C., Prahl, J. M., and DeLuca, H. F. (1993). Stabilization of the vitamin D receptor in rat osteosarcoma cells through the action of 1,25(OH)2D3. Mol. Endocrinol. 7, 1307–1312. 82. Jehan, F., and DeLuca, H. F. (1997). Cloning and characterization of the mouse vitamin D receptor promoter. Proc. Natl Acad. Sci. USA 94, 10138–10143. 83. Byrne, I. M., Flanagan, L., Tenniswood, M. P., and Welsh, J. (2000). Identification of a hormone-responsive promoter immediately upstream
322
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
of exon 1c in the human vitamin D receptor gene. Endocrinology 141, 2829–2836. Caffrey, J. M., and Farach-Carson, M. C. (1989). Vitamin D3 metabolites modulate dihydropyridine-sensitive calcium currents in clonal rat osteosarcoma cells. J. Biol. Chem. 264, 20265–20274. Zanello, L. P., and Norman, A. W. (1997). Stimulation by 1α,25(OH)2-vitamin D3 of whole cell chloride currents in osteoblastic ROS 17/2.8 cells. A structure-function study. J. Biol. Chem. 272, 22617–22622. Zanello, L. P., and Norman, A. W. (2004). Rapid modulation of osteoblast ion channel responses by 1α,25(OH)2-vitamin D3 require the presence of a functional vitamin D nuclear receptor. Proc. Natl Acad. Sci. USA 101, 1589–1594. Huhtakangas, J. A., Olivera, C. J., Bishop, J. E., Zanello, L. P., and Norman, A. W. (2004). The vitamin D receptor is present in caveolaeenriched plasma membranes and binds 1α,25(OH)2-vitamin D3 in vivo and in vitro. Mol. Endocrinol. 18, 2660–2671. Erben, R. G., Soegiarto, D. W., Weber, K., Zeitz, U., Lieberherr, M., Gniadecki, R., Moller, G., Adamski, J., and Balling, R. (2002). Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and non genomic functions of vitamin D. Mol. Endocrinol. 16, 1524–1537. Mizwicki, M. T., Keidel, D., Bula, C. M., Bishop, J. E., Zanello, L. P., Wurtz, J. M., Moras, D., Norman, A. W. (2004). Identification of an alternative ligand-binding pocket in the nuclear vitamin D receptor and its functional importance in 1α,25(OH)2-vitamin D3 signalling. Proc. Natl Acad. Sci. USA 101, 12876–12881. Norman, A. W., Mizwicki, M. T., and Norman, D. P. G. (2004). Steroid-hormone rapid actions, membrane receptors and a conformational ensemble model. Nat. Rev. Drug Discov. 3, 27–41. Wali, R. K., Kong, J., Sitrin, M. D., Bissonnette, M., and Li, Y. C. (2003). Vitamin D receptor is not required for the rapid actions of 1,25dihydroxyvitamin D3 to increase intracellular calcium and activate protein kinase C in mouse osteoblasts. J. Cell. Biochem. 88, 794–801. Ishida, H., Bellows, C. G., Aubin, J. E., and Heersche, J. N.M. (1993). Characterizaion of the 1,25(OH)2D3-induced inhibition of bone nodule formation in long-term cultures of fetal rat calvaria cells. Endocrinology 132, 61–66. Owen, T. A., Aronow, M. S., Barone, L. M., Bettencourt, B., Stein, G. S., and Lian, J. B. (1991). Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: Dependency upon basal levels of gene expression, duration of exposure and bone matrix competency in normal rat osteoblast cultures. Endocrinology 128, 1496–1504 Matsumoto, T., Igarashi, C., Taksuchi, Y., Harada, S., Kikuchi, T., Yamamoto, H., and Ogata, E. (1991). Stimulation by 1,25-dihydroxyvitamin D3 of in vitro mineralization induced by osteoblast-like MC3T3-E1 cells. Bone 12, 27–32. Hansen, C. M., Hansen, D., Holm, P. K., and Binderup, L. (2001). Vitamin D compounds exert anti-apoptotic effects in human osteosarcoma cells in vitro. J. Steroid Biochem. Mol. Biol. 77, 1–11. Duque, G., El Abdaimi, K., Henderson, J. E., Lomri, A., and Kremer, R. (2004). Vitamin D inhibits Fas ligand-induced apoptosis in human osteoblasts by regulating components of both the mitochondrial and Fas-related pathways. Bone 35, 57–64. Morales, O., Samuelsson, M. K., Lindgren, U., and Haldosen, L. A. (2004). Effects of 1α,25-dihydroxyvitamin D3 and growth hormone on apoptosis and proliferation in UMR 106 osteoblast-like cells. Endocrinology 145, 87–94. Wang, Q. M., Jones, J. B., and Studzinski, G. P. (1996). Cyclindependent kinase inhibitor p27 as a mediator of the G1-S phase block induced by 1,25-dihydroxyvitamin D3 in HL60 cells. Cancer Res. 56, 264–267. Verlinden, L., Verstuyf, A., Convents, R., Marcelis, S., Van Camp, M., and Bouillon, R. (1998). Action of 1,25(OH)2D3 on the cell cycle
GEERT CARMELIET ET AL .
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
genes, cyclin D1, p21 and p27 in MCF-7 cells. Mol. Cell. Endocrinol. 142, 57–65. Liu, M., Lee, M. H., Cohen, M., Bommakanti, M., and Freedman, L. P. (1996). Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev. 10, 142–153. Akutsu, N., Lin, R., Bastien, Y., Bestawros, A., Enepekides, D. J., Black, M. J., and White, J. H. (2001). Regulation of gene expression by 1α,25-dihydroxyvitamin D3 and its analog EB1089 under growthinhibitory conditions in squamous carcinoma cells. Mol. Endocrinol. 15, 1127–1139. Hager, G., Formanek, M., Gedlicka, C., Thurnher, D., Knerer, B., and Kornfehl, J. (2001). 1,25(OH)2 vitamin D3 induces elevated expression of the cell cycle-regulating genes P21 and P27 in squamous carcinoma cell lines of the head and neck. Acta Otolaryngol. 121, 103–109. Hershberger, P. A., Modzelewski, R. A., Shurin, Z. R., Rueger, R. M., Trump, D. L., and Johnson, C. S. (1999). 1,25-Dihydroxycholecalciferol (1,25-D3) inhibits the growth of squamous cell carcinoma and down-modulates p21Waf1/Cip1 in vitro and in vivo. Cancer Res. 59, 2644–2649. Wade, W. N., Willingham, M. C., Koumenis, C., and Cramer, S. D. (2002). p27Kip1 is essential for the antiproliferative action of 1,25dihydroxyvitamin D3 in primary, but not immortalized, mouse embryonic fibroblasts. J. Biol. Chem. 277, 37301–37306. Eelen, G., Verlinden, L., Van Camp, M., Van Hummelen, P., Marchal, K., De Moor, B., Mathieu, C., Carmeliet, G., Bouillon, R., and Verstuyf, A. (2004). The effects of 1alpha,25-dihydroxyvitamin D3 on the expression of DNA replication genes. J. Bone Miner. Res. 19, 133–146. Ducy, P., Zhang, R., Geoffroy, V., Ridall, A. L., and Karsenty, G. (1997). Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89, 747–754. Drissi, H., Pouliot, A., Koolloos, C., Stein, J. L., Lian, J. B., Stein, G. S., and van Wijnen, A. J. (2002). 1,25-(OH)2-vitamin D3 suppresses the bone-related Runx2/Cbfa1 gene promoter. Exp. Cell Res. 274, 323–333. Viereck, V., Siggelkow, H., Tauber, S., Raddatz, D., Schutze, N., and Hufner, M. (2002). Differential regulation of Cbfa1/Runx2 and osteocalcin gene expression by vitamin-D3, dexamethasone, and local growth factors in primary human osteoblasts. J. Cell. Biochem. 86, 348–356. Javed, A., Gutierrez, S., Montecino, M., van Wijnen, A. J., Stein, J. L., Stein, G. S., and Lian, J. B. (1999). Multiple Cbfa/AML sites in the rat osteocalcin promoter are required for basal and vitamin D-responsive transcription and contribute to chromatin organization. Mol. Cell. Biol. 19, 7491–7500. Sierra, J., Villagra, A., Paredes, R., Cruzat, F., Gutierrez, S., Javed, A., Arriagada, G., Olate, J., Imschenetzky, M., Van Wijnen, A. J., Lian, J. B., Stein, G. S., Stein, J. L., and Montecino, M. (2003). Regulation of the bone-specific osteocalcin gene by p300 requires Runx2/Cbfa1 and the vitamin D3 receptor but not p300 intrinsic histone acetyltransferase activity. Mol. Cell. Biol. 23, 3339–3351. Gutierrez, S., Liu, J., Javed, A., Montecino, M., Stein, G. S., Lian, J. B., and Stein, J. L. (2004). The vitamin D response element in the distal osteocalcin promoter contributes to chromatin organization of the proximal regulatory domain. J. Biol. Chem. 279, 43581–43588. Lian, J. B., Shalhoub, V., Aslam, F., Frenkel, B., Green, J., Hamrah, M., Stein, G. S., and Stein, J. L. (1997). Species-specific glucocorticoid and 1,25-dihydroxyvitamin D responsiveness in mouse MC3T3-E1 osteoblasts: dexamethasone inhibits osteoblast differentiation and vitamin D down-regulates osteocalcin gene expression. Endocrinology 138, 2117–2127. Zhang, R., Ducy, P., and Karsenty, G. (1997). 1,25-dihydroxyvitamin D3 inhibits osteocalcin expression in mouse through an indirect mechanism. J. Biol. Chem. 272, 110–116. Ducy, P., and Karsenty, G. (1995). Two distinct osteoblast-specific cis-acting elements control expression of a mouse osteocalcin gene. Mol. Cell. Biol. 15, 1858–1869.
Chapter 18 The Vitamin D Hormone and its Nuclear Receptor 115. Noda, M., Vogel, R. L., Craig, A. M., Prahl, J., Deluca, H. F., and Denhardt, D. T. (1990). Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25dihydroxyvitamin D3 enhancement of mouse secreted phophoprotein 1 (Spp-1 or osteopontin) gene expression. Proc. Natl Acad. Sci. USA 87, 9995–9999. 116. Kim, R. H., Li, J. J., Ogata, Y., Yamauchi, M., Freedman, L. P., and Sodek, J. (1996). Identification of a vitamin D3-response element that overlaps a unique inverted TATA box in the rat bone sialoprotein gene. Biochem. J. 318, 219–226. 117. Uchida, M., Shima, M., Chikazu, D., Fujieda, A., Obara, K., Suzuki, H., Nagai, Y., Yamato, H., and Kawaguchi, H. (2001). Transcriptional induction of matrix metalloproteinase-13 (collagenase-3) by 1α,25dihydroxyvitamin D3 in mouse osteoblastic MC3T3-E1 cells. J. Bone Miner. Res. 16, 221–230. 118. Luo, X. H., and Liao, E. Y. (2003). 1α,25-dihydroxyvitamin D3 regulates the expression of membrane-type matrix metalloproteinase-1 in normal human osteoblast-like cells. Endocr. Res. 29, 353–362. 119. Van Driel, M., Pols, H. A. P., and van Leeuwen, J. P. T. M. (2004). Osteoblast differentiation and control by vitamin D and vitamin D metabolites. Curr. Pharm. Des. 10, 2535–2555. 120. Wang, D. S., Miura, M., Demura, H., and Sato, K. (1997). Anabolic effects of 1,25-dihydroxyvitamin D3 on osteoblasts are enhanced by vascular endothelial growth factor produced by osteoblasts and by growth factors produced by endothelial cells. Endocrinology 138, 2953–2962. 121. Takahashi, N., Udagawa, N., Tanaka, S., Murakami, H., Owan, I., Tamura, T., and Suda, T. (1993). Postmitotic osteoclast precursors are mononuclear cells which express macrophage-associated phenotypes. Dev. Biol. 163, 212–221. 122. Nakashima, T., Kobayashi, Y., Yamasaki, S., Kawakami, A., Eguchi, K., Sasaki, H., and Sakai, H. (2000). Protein expression and functional difference of membrane-bound and soluble receptor activator of NF-kappaB ligand: modulation of the expression by osteotropic factors and cytokines. Biochem. Biophys. Res. Commun. 275, 768–775. 123. Lacey, D. L., Timms, E., Tan, H. L., Kelley, M. J., Dunstan, C. R., Burgess, T., Elliott, R., Colombero, A., Elliott, G., Scully, S., Hsu, H., Sullivan, J., Hawkins, N., Davy, E., Capparelli, C., Eli, A., Qian, Y. X., Kaufman, S., Sarosi, I., Shalhoub, V., Senaldi, G., Guo, J., Delaney, J., and Boyle, W. J. (1998). Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93, 165–176. 124. Kitazawa, R., and Kitazawa, S. (2002). Vitamin D3 augments osteoclastogenesis via vitamin D-responsive element of mouse RANKL gene promoter. Biochem. Biophys. Res. Commun. 290, 650–655. 125. Kitazawa, S., Kajimoto, K., Kondo, T., and Kitazawa, R. (2003). Vitamin D3 supports osteoclastogenesis via functional vitamin D response element of human RANKL gene promoter. J. Cell. Biochem. 89, 771–777. 126. Rubin, J., Fan, X., Thornton, D., Bryant, R., and Biskobing, D. (1996). Regulation of murine osteoblast macrophage colony-stimulating factor production by 1,25(OH)2D3. Calcif. Tissue Int. 59, 291–296. 127. Horwood, N. J., Elliott, J., Martin, T. J., and Gillespie, M. T. (1998). Osteotropic agents regulate the expression of osteoclast differentiation factor and osteoprotegerin in osteoblastic stromal cells. Endocrinology 139, 4743–4746. 128. Kondo, T., Kitazawa, R., Maeda, S., and Kitazawa, S. (2004). 1α,25dihydroxyvitamin D3 rapidly regulates the mouse osteoprotegerin gene through dual pathways. J. Bone Miner. Res. 19, 1411–1419. 129. Takeda, S., Yoshizawa, T., Nagai, Y., Yamato, H., Fukumoto, S., Sekine, K., Kato, S., Matsumoto, T., and Fujita, J. (1997). Stimulation of osteoclast formation by 1,25-dihydroxyvitamin D requires its binding to vitamin D receptor (VDR) in osteoblastic cells: studies using VDR knockout mice. Endocrinology 140, 1005–1008. 130. Yoshizawa, T., Handa, Y., Uematsu, Y., Takesa, S., Sekine, K., Yoshihara, Y., Kawakami, T., Arioka, K., Sato, H., Uchiyama, Y.,
131.
132. 133. 134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146. 147.
148.
149.
323 Masushige, S., Fukamizu, A., Matsumoto, T., and Kato, S. (1997). Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat. Genet. 16, 391–396. Gurlek, A., Pittelkow, M. R., and Kumar, R. (2002). Modulation of growth factor/cytokine synthesis and signaling by 1α,25dihydroxyvitamin D3: implications in cell growth and differentiation. Endocr. Rev. 23, 763–786. Roodman, G. D. (1996). Advances in bone biology: the osteoclast. Endocr. Rev. 17, 308–332. Rodan, S. B., and Rodan, G. A. (1997). Integrin function in osteoclasts. J. Endocrinol. 154, S47–S56. Mimura, H., Cao, X., Ross, F. P., Chiba, M., and Teitelbaum, S. L. (1994). 1,25-dihydroxyvitamin D3 transcriptionally activates the β3-integrin subunit gene in avian osteoclast precursors. Endocrinology 134, 1061–1066. Cao, X., Ross, F. P., Zhang, L., MacDonald, P. N., Chappel, J., and Teitelbaum, S. L. (1993). Cloning of the promoter for the avian integrin β3 subunit gene and its regulation by 1,25-dihydroxyvitamin D3. J. Biol. Chem. 268, 27371–27380. Yoneda, T., Niewolna, M., Lowe, C., Izbicka, E., and Mundy, G. R. (1993). Hormonal regulation of pp60c-src expression during osteoclast formation in vitro. Mol. Endocrinol. 7, 1313–1318. Soriano, P., Montgomery, C., Geske, R., and Bradley, A. (1991). Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64, 693–702. Chappel, J., Ross, F. P., Abu-Amer, Y., Shaw, A., and Teitelbaum, S. L. (1997). 1,25-dihydroxyvitamin D3 regulates pp60c-src activity and expression of a pp60c-src activating phosphatase. J. Cell. Biochem. 67, 432–438. Quelo, I., Kahlen, J. P., Rascle, A., Jurdic, P., and Carlberg, C. (1994). Identification and characterization of a vitamin D3 response element of chicken carbonic anhydrase-II. DNA Cell Biol. 13, 1181–1187. Biskobing, D. M., Fan, D., Fan, X., and Rubin, J. (1997). Induction of carbonic anhydrase II expression in osteoclast progenitors requires physical contact with stromal cells. Endocrinology 138, 4852–4857. Klaus, G., Weber, L., Rodriguez, J., Fernandez, P., Klein, T., Grulich-Henn, J., Huegel, U., Ritz, E., and Mehls, O. (1998). Interaction of IGF-I and 1,25(OH)2D3 on receptor expression and growth stimulation in rat growth plate chondrocytes. Kidney Int. 53, 1152–1161. Krohn, K., Haffner, D., Hugel, U., Himmele, R., Klaus, G., Mehls, O., and Schaefer, F. (2003). 1,25(OH)2D3 and dihydrotestoterone interact to regulate proliferation and differentiation of epiphyseal chondrocytes. Calcif. Tissue Int. 73, 400–410. Schwartz, Z., Schlader, D. L., Ramirez, V., Kennedy, M. B., and Boyan, B. D. (1989). Effects of vitamin D metabolites on collagen production and cell proliferation of growth zone and resting zone cartilage cells in vitro. J. Bone Miner. Res. 4, 199–207. Donohue, M. M., and Demay, M. B. (2002). Rickets in VDR null mice is secondary to decreased apoptosis of hypertrophic chondrocytes. Endocrinology 143, 3691–3694. Schmidt-Gayk, H., Bouillon, R., and Roth, H. J. (1997). The determination of vitamin D metabolites (calcidiol and calcitriol) and their clinical significance. Scand. J. Clin. Lab. Invest. 57, 35–45. Bouillon, R., Carmeliet, G., and Boonen, S. (1997). Ageing and calcium metabolism. Baill. Clin. Endocrinol. Metab. 11, 341–365. Brame, L. A., White, K. E., and Econs, M. J. (2004). Renal phosphate wasting disorders: clinical features and pathogenesis. Semin. Nephrol. 24, 39–47. Fu, G. K., Lin, D., Zhang, M. Y., Bikle, D. D., Shackleton, C. H. L., Miller, W. L., and Portale, A. A. (1997). Cloning of human 25-hydroxyvitamin D-1α-hydroxylase and mutations causing vitamin D-dependent rickets type 1. Molec. Endocrinol. 11, 1961–1970. Kitanaka, S., Takeyama, K., Murayama, A., Sato, T., Okumura, K., Nogami, M., Hasegawa, Y., Niimi, H., Yanagisawa, J., Tanaka, T., and Kato, S. (1998). Inactivating mutations in the 25-hydroxyvitamin D3
324
150.
151.
152.
153. 154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
GEERT CARMELIET ET AL . 1α-hydroxylase gene in patients with pseudovitamin D-deficiency rickets. N. Engl. J. Med. 338, 653–661. Fraser, D., Kooh, S. W., Kind, H. P., Holick, M. F., Tanaka, Y., and DeLuca, H. F. (1973). Pathogenesis of hereditary vitamin D-dependent rickets. An inborn error of vitamin D metabolism involving defective conversion of 25-hydroxyvitamin D to 1α,25-hydroxyvitamin D. N. Engl. J. Med. 289, 817–822. Malloy, P. J., Pike, J. W., and Feldman, D. (1999). The vitamin D receptor and the syndrome of hereditary 1,25-dihydroxyvitamin Dresistant rickets. Endocr. Rev. 20, 156–188. Balsan, S., Garabedian, M., Larchet, M., Gorski, A. M., Cournot, G., Tau, C., Bourdeau, A., Silve, C., and Ricour, C. (1986). Long-term nocturnal calcium infusions can cure rickets and promote normal mineralization in hereditary resistance to A,25-dihydroxyvitamin D. J. Clin. Invest. 77, 1661–1667. Hochberg, Z., Tiosano, D., and Even, L. (1992). Calcium therapy for calcitriol-resistant rickets. J. Pediatr. 121, 803–808. Li, Y. C., Pirro, A. E., Amling, M., Delling, G., Baron, R., Bronson, R., and Demay, M. B. (1997). Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc. Natl Acad. Sci. USA 94, 9831–9835. Van Cromphaut, S. J., Dewerchin, M., Hoenderop, J. G. J., Stockmans, I., Van Herck, E., Kato, S., Bindels, R. J. M., Collen, D., Carmeliet, P., Bouillon, R., and Carmeliet, G. (2001). Duodenal calcium absorption in vitamin D receptor-knockout mice: Functional and molecular aspects. Proc. Natl Acad. Sci. USA 98, 13324–13329. Erben, R. G., Soegiarto, D. W., Weber, K., Zeitz, U., Lieberherr, M., Gniadecki, R., Moller, G., Adamski, J., and Balling, R. (2002). Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol. Endocrinol. 16, 1524–1537. Amling, M., Priemel, M. H. T., Chapin, K., Rueger, J. M., Baron, R., and Demay, M. B. (1999). Rescue of the skeletal phenotype of vitamin D receptor ablated mice in the setting of normal mineral ion homeostasis: formal histomorphometric and biomechanical analyses. Endocrinology 140, 4982–4987. Li, Y. C., Amling, M., Pirro, A. E., Priemel, M., Meuse, J., Baron, R., Delling, G., and Demay, M. B. (1998). Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice. Endocrinology 139, 4391–4396. Masuyama, R., Nakaya, Y., Tanaka, S., Tsurukami, H., Nakamura, T., Watanabe, S., Yoshizawa, T., Kato, S., and Suzuki, K. (2001). Dietary phosphorus restriction reverses the impaired bone mineralization in vitamin D receptor knockout mice. Endocrinology 142, 494–497. Masuyama, R., Nakaya, Y., Katsumata, S., Kajita, Y., Uehara, M., Tanaka, S., Sakai, A., Kato, S., Nakamura, T., and Suzuki, K. (2003). Dietary calcium and phosphorus ratio regulates bone mineralization and turnover in vitamin D receptor knockout mice by affecting intestinal calcium and phosphorus absorption. J. Bone Miner. Res. 18, 1217–1226. Li, Y. C., Bolt, M. J. G., Cao, L. P., and Sitrin, M. D. (2001). Effects of vitamin D receptor inactivation on the expression of calbindins and calcium metabolism. Am. J. Physiol. Endocrinol. Metab. 281, E558–E564. Li, Y. C., Pirro, A. E., and Demay, M. B. (1998). Analysis of vitamin D-dependent calcium-binding protein messenger ribonucleic acid expression in mice lacking the vitamin D receptor. Endocrinology 139, 847–851. Donohue, M. M., and Demay, M. B. (2002). Rickets in VDR null mice is secondary to decreased apoptosis of hypertrophic chondrocytes. Endocrinology 143, 3691–3694. Amling, M., Neff, L., Tanaka, S., Inoue, D., Kuida, K., Weir, E., Philbrick, W. M., Broadus, A. E., and Baron, R. (1997). Bcl-2 lies downstream of parathyroid hormone-related peptide in a signaling
165.
166.
167.
168. 169.
170.
171. 172.
173.
174.
175.
176.
177.
178.
179.
180.
pathway that regulates chondrocyte maturation during skeletal development. J. Cell Biol. 136, 205–213. Yagishita, N., Yamamoto, Y., Yoshizawa, T., Sekine, K., Uematsu, Y., Murayama, H., Nagai, Y., Krezel, W., Chambon, P., Matsumoto, T., and Kato, S. (2001). Aberrant growth plate development in VDR/RXRγ double null mutant mice. Endocrinology 142, 5332–5341. Dardenne, O., Prud’homme, J., Arabian, A., Glorieux, F. H., and St-Arnaud, R. (2001). Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142, 3135–3141. Panda, D. K., Miao, D., Tremblay, M. L., Sirois, J., Farookhi, R., Hendy, G. N., and Goltzman, D. (2001). Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: Evidence for skeletal, reproductive, and immune dysfunction. Proc. Natl. Acad. Sci. USA 98, 7498–7503. Bouillon, R. (1998). The many faces of rickets. N. Engl. J. Med. 338, 653–661. Panda, D. K., Miao, D., Bolivar, I., Li, J., Huo, R., Hendy, G. N., and Goltzman D. (2004). Inactivation of the 25-hydroxyvitamin D 1α-hydroxylase and vitamin D receptor demonstrates independent and interdependent effects of calcium and vitamin D on skeletal and mineral homeostasis. J. Biol. Chem. 279, 16754–16766. Chapuy, M., Arlot, M., Duboeuf, F., Brun, J., Crouzet, B., Arnaud, S., Delmas, P. D., and Meunier, P. J. (1992). Vitamin D3 and calcium to prevent hip fractures in elderly women. N. Eng. J. Med. 327, 1637–1642. Civitelli, R. (1995). The role of vitamin D metabolism in the treatment of osteoporosis. Calcif. Tissue Int. 57, 409–414. Tilyard, M. W., Spears, G. F., Thomson, J., and Dovey, S. (1992). Treatment of postmenopausal osteoporosis with calcitriol or calcium. N. Engl. J. Med. 326, 357–362. Akesson, K., Lau, K-H. W., and Baylink, D. J. (1997). Rationale for active vitamin D analog therapy in senile osteoporosis. Calcif. Tissue Int. 60, 100–105. Okano, T., Tsugawa, N., Masuda, S., Takeuchi, A., Kabayashi, T., Takita, Y., and Nishii, Y. (1989). Regulatory activities of 2β(3-hydroxypropoxy)-1α,25-dihydroxy- vitamin D3, a novel synthetic vitamin D3 derivative, on calcium metabolism. Biochem. Biophys. Res. Commun. 163, 1444–1449. Tsurukami, H., Nakamura, T., Suzuki, K., Sato, K., Higuchi, Y., and Nishii, Y. (1994). A novel synthetic vitamin D analogue, 2β(3-hydroxypropoxy)1α,25-dihydroxyvitamin D3 (ED-71), increases bone mass by stimulating the bone formation in normal and ovariectomized rats. Calcif. Tissue Int. 54, 142–149. Uchiyama, Y., Higuchi, Y., Takeda, S., Masaki, T., Shira-Ishi, A., Sato, K., Kubodera, N., Ikeda, K., and Ogata, E. (2002). ED-71, a vitamin D analog, is a more potent inhibitor of bone resorption than alfacalcidol in an estrogen-deficient rat model of osteoporosis. Bone 30, 582–588. Yamamoto, H., Shevde, N. K., Warrier, A., Plum, L. A., DeLuca, H. F., and Pike, J. W. (2003). 2-Methylene-19-nor-(20S)1,25-dihydroxyvitamin D3 potently stimulates gene-specific DNA binding of the vitamin D receptor in osteoblasts. J. Biol. Chem. 278, 31756–31765. Shevde, N. K., Plum, L. A., Clagett-Dame, M., Yamamoto, H., Pike, J. W., and DeLuca, H. F. (2002). A potent analog of 1α,25dihydroxyvitamin D3 selectively induces bone formation. Proc. Natl. Acad. Sci. USA 99, 13487–13491. Peleg, S., Uskokovic, M., Ahene, A., Vickery, G., and Avnur, Z. (2002). Cellular and molecular events associated with the bone-protecting activity of the noncalcemic vitamin D analog Ro-26–9228 in osteopenic rats. Endocrinology 143, 1625–1636. Peleg, S., Ismail, A., Uskokovic, M. R., and Avnur, Z. (2003). Evidence for tissue- and cell-type selective activation of the vitamin D receptor by Ro-26–9228, a noncalcemic analog of vitamin D3. J. Cell. Biochem. 88, 267–273.
Chapter 18 The Vitamin D Hormone and its Nuclear Receptor 181. Brown, A. J., Dusso, A. S., and Slatopolsky, E. (2002). Vitamin D analogues for secondary hyperparathyroidism. Nephrol. Dial. Transplant. 17 Suppl 10, 10–19. 182. Brown, A. J., Ritter, C. R., Finch, J. L., Morrissey, J., Martin, K., Murayama, E., Nishii, Y., and Slatopolsky, E. (1989). The noncalcemic analogue of vitamin D, 22-oxacalcitriol, suppresses parathyroid hormone synthesis and secretion. J. Clin. Invest. 84, 728–732. 183. Kurokawa, K., Akizawa, T., Suzuki, M., Akita, T., Ogata, E., and Slatopolsky, E. (1996). Effect of 22-oxacalcitriol on hyperparathyroidism of dialysis patients: results of a preliminary study. Nephrol. Dial. Transplant. 11, 121–124. 184. Lindberg, J., Martin, K. J., Gonzalez, E. A., Acchiardo, S. R., Valdin, J. R., and Soltanek, C. (2001). A long-term, multicenter study of the efficacy and safety of paricalcitol in end-stage renal disease. Clin. Nephrol. 56, 315–323. 185. Maung, H. M., Elangovan, L., Frazao, J. M., Bower, J. D., Kelley, B. J., Acchiardo, S. R., Rodriguez, H. J., Norris, K. C., Sigala, J. F., Rutkowski, M., Robertson, J. A., Goodman, W. G., Levin, B. S., Chesney, R. W., Mazess, R. B., Kyllo, D. M., Douglass, L. L.,
186.
187.
188.
189.
325 Bishop, C. W., and Coburn, J. W. (2001). Efficacy and side effects of intermittent intravenous and oral doxercalciferol (1α-hydroxyvitamin D2) in dialysis patients with secondary hyperparathyroidism: a sequential comparison. Am. J. Kidney Dis. 37, 532–543. Nishizawa, Y., Morii, H., Ogura, Y., and DeLuca, H. F. (1991). Clinical trial of 26,26,26,27,27,27-hexafluoro-1,25-dihydroxyvitamin D3 in uremic patients on hemodialysis: preliminary report. Contrib. Nephrol. 90, 196–203. Akiba, T., Marumo, F., Owada, A., Kurihara, S., Inoue, A., Chida, Y., Ando, R., Shinoda, T., Ishida, Y., and Ohashi, Y. (1998). Controlled trial of falecalcitriol versus alfacalcidol in suppression of parathyroid hormone in hemodialysis patients with secondary hyperparathyroidism. Am. J. Kidney Dis. 32, 238–246. Sprague, S. M., Llach, F., Amdahl, M., Taccetta, C., and Batlle, D. (2003). Paricalcitol versus calcitriol in the treatment of secondary hyperparathyroidism. Kidney Int. 63, 1483–1490. Teng, M., Wolf, M., Lowrie, E., Ofsthun, N., Lazarus, J. M., and Thadhani, R. (2003). Survival of patients undergoing hemodialysis with paricalcitol or calcitriol therapy. N. Engl. J. Med. 349, 446–456.
This Page Intentionally Left Blank
Chapter 19
Sex Steroid Effects on Bone Metabolism David G. Monroe, Ph.D. Thomas C. Spelsberg, Ph.D. and S. Khosla, M.D.
Biochemistry and Molecular Biology, Mayo Clinic College of Medicine, Rochester, MN Professor of Biochemistry, Mayo Clinic College of Medicine, Rochester, MN
I. Abstract II. Introduction III. Molecular Structures, Synthesis, Mechanism of Action of Major Sex Steroids, and Transcriptional Coregulator Function IV. Effects of Sex Steroids on Bone Cells and Bone Turnover
V. Effects of Estrogens and Androgens on Bone Metabolism in Men Versus Women VI. Effects of Sex Steriods on Extraskeletal Calcium Homeostasis VII. Summary References
I. ABSTRACT
significant effects on the male skeleton, estrogen may also play a key role in males, and conversely, androgens have important skeletal effects in women. These dual roles for estrogen and androgens have long been known in reproductive tissues, and now also appear to be true for the skeleton. Sex steroids also have important, indirect effects on bone metabolism via their effects on intestinal calcium absorption and renal calcium handling. While much has been learned in recent years on sex steroid effects on bone, this is a rapidly evolving area and future studies are likely to reveal new insights, both into the mechanisms of sex steroid action on bone, as well as the relative contributions of estrogens versus androgens towards bone metabolism in both sexes.
Sex steroids have major effects on bone and calcium metabolism. Thus, hypogonadism in either sex is associated with rapid bone loss, which is preventable with gonadal steroid replacement. In this chapter, we review the molecular structures, biosynthesis, and mechanism of action of the major clinically relevant sex steroids. The identification of estrogen and androgen receptors in bone cells was followed by an explosion of studies on the effects of sex steroids, and in particular, estrogen, on osteoblasts and osteoclasts. This has led to the identification of candidate autocrine and paracrine mediators of estrogen and androgen action in bone, although important gaps remain in our understanding of the direct effects of sex steroids on bone cell function. Recent genetic, epidemiologic, and direct interventional evidence has also challenged traditional notions of the relative importance of estrogen on bone metabolism in men. While androgens clearly have Dynamics of Bone and Cartilage Metabolism
II. INTRODUCTION Estrogens and androgens play major roles in skeletal homeostasis. Estrogen deficiency following the menopause 327
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
328
DAVID G. MONROE ET AL.
is recognized as the major cause of postmenopausal osteoporosis. The effects of estrogen on bone and overall calcium homeostasis have been the subject of intensive investigation. Androgens also have important effects on bone, although the relative contributions of estrogens and androgens towards bone metabolism in both sexes are still under investigation. In this chapter, we review the molecular structures, synthesis, and role of the major sex steroids in bone and calcium metabolism.
III. MOLECULAR STRUCTURES, SYNTHESIS, MECHANISM OF ACTION OF MAJOR SEX STEROIDS, AND TRANSCRIPTIONAL COREGULATOR FUNCTION A. Sex Steroid Structure and Synthesis Figure 1 provides an overview of the major steps in the biosynthetic pathway for the sex steroids [1, 2]. In the ovary, estrone and estradiol are formed from androstenedione and
testosterone, respectively. This reaction is mediated by the enzyme, aromatase, which is a cytochrome P450 enzyme (P450 aromatase) present in the ovary [3]. This enzyme is also present in numerous other locations, including the testis, adipose tissue, and bone cells (osteoblasts and osteoclasts) [4, 5]. During ovulation and pregnancy, progesterone is secreted by the ovaries, although its role in bone metabolism is not as well defined as that of estrogens or androgens. Individual cell types in these tissues are often responsible for the synthesis of each of the species of steroids. The steroidogenic pathway is essentially similar in the testes, with the exception that testosterone is the major secretory product, although there is some conversion of both androstenedione and testosterone to estrone and 17β-estradiol, respectively (Fig. 1). In androgen target tissues, which include bone [6], testosterone is further converted to dihydrotestosterone (DHT). Unlike testosterone, DHT cannot be aromatized to estrogens. The adrenal glands are a significant source of weak androgens, principally dehydroepiandrosterone (DHEA) and androstenedione (Fig. 1). DHEA is further converted by the enzyme steroid sulfotransferase to dehydroepiandrosterone
Figure 1 Biosynthetic pathways involved in the formation of estrogens, androgens, and progestins. (Adapted from Monroe et al. [10].)
329
Chapter 19 Sex Steroid Effects on Bone Metabolism
sulfate (DHEAS). DHEA and DHEAS represent the major circulating adrenal androgens. In peripheral tissues, these can be further converted to the more potent sex steroids, testosterone and 17β-estradiol, although there is some evidence that adrenal androgens may have independent and direct effects on some tissues, including bone [7]. In the circulation, the major sex steroids, testosterone and estradiol, are bound to serum-binding proteins, and only 1–3% of the total circulating sex steroids are free in solution. Approximately 35–55% of the sex steroids are bound to albumin, which constitutes a high-capacity, lowaffinity reservoir of the steroids. Both the free fraction and the albumin-bound fraction have access to target tissues [8] and thus represent the “bioavailable” sex steroid. Since sex hormone-binding globulin (SHBG) has a high affinity for testosterone and estradiol, the fraction of these sex steroids bound to SHBG is likely unavailable to target tissues. In contrast to testosterone and estradiol, adrenal androgens do not bind SHBG to a significant degree, and more than 90% of adrenal androgens circulate loosely bound to albumin [9].
B. Mechanisms of Action of Steroids Considerable work has defined the mechanisms of action of sex steroids in target tissues (for reviews, see references [1, 10–12]). Sex steroids rapidly diffuse into cells through the plasma membrane where they bind their cognate receptors, which are proteins of high molecular weight present in low concentrations (approximately 5000–40 000 molecules per cell), primarily in the nucleus. The steroid hormones bind their specific receptor and modify the tertiary structure of the receptor producing
Figure 2
a novel conformation. The result of this conformational change is dissociation from specific chaperone proteins [13, 14], homo- or heterodimerization [15–17], several post-translational modifications including phosphorylation [18–20], and recruitment of specific transcriptional coregulator protein complexes [21]. In the classical pathway, the liganded ER then binds estrogen response elements (EREs) and this leads to the activation of specific genes (Fig. 2A). However, the ER is also known to function through pathways that do not involve direct binding to DNA, such as through protein–protein interactions (Fig. 2B) [22]. In addition, recent evidence has implicated estrogen signaling through a membrane receptor, with subsequent activation of specific cellular kinase pathways [23] (Fig. 2C); the latter appear to be directly involved in estrogen regulation of cellular apoptosis [24, 25]. Two highly related isoforms of the estrogen receptor exist (termed ERα and ERβ) that are encoded by unique genes. Both estrogen receptor isoforms bind 17β-estradiol with similar affinities and utilize similar DNA response elements, whereas androgens bind to the androgen receptor and use their own unique DNA response elements. Interestingly, while almost all other ligand-dependent hormone receptors have two isoforms of the receptor (i.e. ER, GR, PR), only one known androgen receptor isoform exists. The ligand-dependent steroid hormone receptors, including the estrogen and androgen receptors, are classified as type I nuclear hormone receptors. Type I receptors are soluble and recognize their respective DNA-binding elements only when activated by ligand. Type II receptors include receptors that bind nonsteroidal compounds such as thyroid hormone (thyroid receptor; TR), retinoic acid (RAR, RXR), and vitamin D (VDR). These receptors are
Classical and nonclassical pathways of estrogen action. See text for details. (Adapted from Manolagas et al. [188].)
330
DAVID G. MONROE ET AL.
bound to DNA in the absence of ligand where they function to repress gene transcription. Activation of transcription is elicited through ligand binding. In this chapter, we will concentrate on the type I receptors (specifically estrogen and androgen receptors). All type I receptors share a similar domain structure [26]. While the DNA-binding domains of ERα and ERβ are similar, their transactivation domains, tissue distribution, and molecular sizes differ [21]. The N-terminal region of the receptor shows less conservation of sequence among the steroid hormone receptors and is involved in transcriptional activation functions. The ERβ species is highly expressed in prostate, ovary, brain, and bladder, but is found in limiting quantities in the uterus, kidney, pituitary, and epididymis (these tissues contain high levels of the ERα species [27]). Recent studies also indicate that bone cells express both ERα and ERβ [28, 29]. Several studies are now indicating that coexpression of ERα and ERβ, leading to heterodimerization, has unique functions apart from either ER isoform alone. Finally, a number of studies using mice with deletion of either ERα or ERβ indicate that ERα is likely the major ER species mediating estrogen effects on bone [30], with ERβ either substituting for ERα [31] or perhaps enhancing ERα action under certain conditions [30].
Figure 3
C. Transcriptional Coregulator Function Nuclear hormone receptor coactivators are involved in enhancing the ligand-dependent transcriptional signal of nuclear hormone receptors (see Fig. 3). SRC1, the founding member of the SRC family, was originally identified in a yeast two-hybrid screen of a human B-lymphocyte cDNA library using the ligand-binding domain of progesterone receptor (PR) as bait [32]. Further analysis demonstrated that SRC1 exhibits coactivation properties, not only with PR but also with numerous other nuclear hormone receptors including ER [21, 33]. Two additional members of the SRC family, SRC2 [34] and SRC3 [35], were also identified which interact with and coactivate a wide variety of nuclear hormone receptors including ER. Gene deletion experiments in mice demonstrate that, while homozygous deletion of SRC1 and SRC2 results in partial hormone resistance [36], the phenotype of SRC1 and SRC2 knockout animals otherwise appear largely normal. However, recent studies using SRC1 knockout mice have demonstrated that loss of SRC1 leads to impaired estrogen action, principally in trabecular bone [37]. Interestingly, the SRC3 mutant demonstrates dwarfism and a more severe reproductive phenotype [38]. Numerous other transcriptional
General pathway for E2 and SERM action. The ligand enters the cell through diffusion where it encounters inactive ER monomers. Homo- or heterodimerization can occur with ERα and ERβ. Depending on the type of estrogen response element present at any given promoter, the ER dimer binds either directly or can bind indirectly through protein tethering (AP1 or fos/jun complex, in this case) and causing activation of gene transcription.
331
Chapter 19 Sex Steroid Effects on Bone Metabolism
coregulators involved in either activation or repression of nuclear receptor function have been described elsewhere [21, 39]. The SRC family of proteins mediate the interactions with nuclear receptors through a centrally located receptorinteracting domain (RID), which contains three LXXLL motifs necessary for interaction with nuclear receptors [40, 41]. Binding of ligand to ER induces specific conformational changes, providing a unique surface for association of SRCs through the RID. Although all the SRCs are capable of binding ER, it is unclear whether the different SRCs preferentially interact with a particular ER dimer (e.g. ERα, ERβ, or ERα/β). A recent study suggests that SRC1 overexpression preferentially enhances ERβ transcriptional activation of a transiently transfected ERE construct, whereas SRC2 overexpression preferentially enhances ERα transcriptional activation in human bone cells [42]. However, whether these preferences are due to preferential binding or another phenomenon (e.g. posttranslational modifications) is unknown.
IV. EFFECTS OF SEX STEROIDS ON BONE CELLS AND BONE TURNOVER A. Estrogen Estrogen deficiency is recognized as the most relevant factor in the pathogenesis of postmenopausal bone loss [43, 44]. Recent studies have begun to elucidate the molecular mediators of estrogen deficiency on bone resorption. The identification of both ERα and ERβ on human osteoblasts [45] (Fig. 4) and osteoclasts [46–49] has demonstrated that estrogen likely has direct effects on bone, as opposed to more indirect effects of systemic
Figure 4 Demonstration of the presence of ERs (left) and ARs (right) in human bone cells. (Adapted from Colvard et al. [115].)
estrogen deficiency. The modulation of ER-dependent processes within both osteoblasts and osteoclasts has been utilized clinically through estrogen replacement therapy, which has been shown in numerous studies to be effective in preventing and treating osteoporosis [50]. A summary of the effects of estrogen on osteoblasts and osteoclasts is described in Figure 5. The OPG/RANKL/RANK system has become recognized as one of the major regulatory mechanisms influencing osteoclastogenesis [51–54] (see Fig. 5). Osteoblasts express surface RANKL that binds to surface RANK receptors on osteoclast precursor cells, leading to differentiation and activation of these precursors into mature osteoclasts. A soluble decoy receptor, osteoprotegerin (OPG), is secreted from the preosteoblast/stromal cell that inhibits osteoclastogenesis by competing with RANKL for RANK. Estrogen can exert its antiresorptive effect by up-regulating OPG [55–58], thus inhibiting bone resorption. Estrogen can also decrease the expression of RANKL on the surface of osteoblasts leading to a similar inhibition of bone resorption [59]. Estrogens also have direct effects on the responsiveness and formation of osteoclasts [60, 61], thereby decreasing bone resorption. In addition to the effects of estrogen on the OPG and RANKL expression, estrogen also regulates the production of additional cytokines in osteoblasts or bone marrow mononuclear cells, thus modulating osteoclastic activity in a paracrine fashion [62]. There is now an increasing body of evidence that bone-resorbing cytokines, such as interleukin (IL)-1, IL-6, tumor necrosis factor α (TNF-α),
Figure 5 osteocytes.
Summary of estrogen effects on osteoblasts, osteoclasts, and
332 macrophage-colony stimulating factor (M-CSF), and prostaglandins may be potential candidates for mediating the bone loss following estrogen deficiency. Both IL-1 and M-CSF production are increased in estrogen-deficient model systems [63, 64], which can be inhibited using specific antagonists [65–67]. Additionally, the boneresorptive effects of TNF-α are well documented [68] and can be reversed using a soluble type I TNF receptor [69]. Numerous other studies indicate that IL-6 plays a key role in mediating bone loss following estrogen deficiency [70–74]. However, it is likely that, in vivo, multiple cytokines act cooperatively in inducing bone resorption following estrogen deficiency, and that a single cytokine may only partially account for the effects of estrogen deficiency on the skeleton. In contrast to the inhibitory effects of estrogen on bone resorption, the effects of estrogen on bone formation and on osteoblast proliferation and differentiation are less clear. Since estrogen deficiency is associated with both increased bone resorption and impaired compensatory bone formation [75], it is likely that estrogen has significant effects on bone formation. In support of this, estrogen has been shown to increase production of IGF-I [76] and transforming growth factor-β [77] by osteoblastic cells in vitro. Also, estrogen has been reported to stimulate fibroblast [78] and osteoblast [76] type I collagen synthesis, which represents >90% of the biosynthetic capacity of these cells. However, estrogen treatment has been reported both to stimulate [79] and to inhibit [80] bone formation in experimental animals, so this issue is far from resolved. In vitro, estrogen effects on osteoblast proliferation and differentiation markers have been variable, depending on the model system used [76, 81, 82]. Since these initial studies were conducted before ERβ was discovered, many of these apparent discrepancies may be due to the ER isoform expressed and the receptor concentration, which can greatly influence both the magnitude and direction of the estrogen-dependent response. A number of more recent studies, using osteoblastic cell systems expressing ERα, have found a consistent inhibitory effect of estrogen on proliferation [83–85], whereas no effect of estrogen was seen using cells expressing ERβ [83]. Additionally, Waters et al. [86] found that osteoblastic matrix formation, mineralization, and the expression and activity of bone marker genes are not only dependent on the ER isoform, but also on the stage of osteoblastic differentiation (e.g. matrix synthesis, early/late mineralization stages). Interestingly, Cao et al. [87] found that in the MG-63 osteosarcoma line that ERβ mediates the synthesis of bone matrix proteins. Collectively, these data suggest that estrogen directly regulates osteoblast proliferation and differentiation, although the net effect of estrogen on
DAVID G. MONROE ET AL.
osteoblasts likely depends on factors such as species differences, cell system heterogeneity, differentiation stage, ER isoform expression, and receptor concentration. The advent of microarray technology has advanced the understanding of osteoblastic gene expression. A number of studies using an osteoblastic cell model expressing either ERα or ERβ has demonstrated that estrogen differentially regulates a unique subset of genes through either ER isoform [83, 88, 89] (see Fig. 6). These unique estrogendependent regulatory patterns are most likely due to differences in recruitment of transcriptional coregulator molecules. Indeed, as noted earlier, recent studies have indicated that ERα and ERβ have preferential affinities for the coregulators SRC1 and SRC2 in transient transfection assays in the hFOB osteoblast model [42]. Recent evidence also suggests that, in addition to the osteoblast, the osteocyte may also be a target for estrogen action. Estrogen withdrawal associated with GnRH therapy has, for example, been shown to induce apoptosis of osteocytes in iliac bone [90]. Furthermore, estrogen treatment inhibits osteocyte apoptosis induced by pro-apoptotic stimuli [23]. Since osteocytes may be involved in mechanosensing and transducing loading responses [91], these effects of estrogen deficiency could impair the skeletal response to loading. In addition to effects on bone resorption and possibly on bone formation in the adult skeleton, estrogen clearly also has major effects on skeletal development. Thus, the increase in estrogen levels at the onset of puberty in girls is associated with the pubertal increase in growth velocity as well as the ultimate closure of the epiphyseal growth plate and cessation of linear growth [92]. The pubertal growth spurt in girls is also associated with an increase in bone mass through a combination of increases in bone length, bone diameter, cortical bone width, and cancellous bone mass.
Figure 6 Microarrray analysis of U2OS cells stably expressing either ERα or ERβ (U2OS-ERα, U2OS-ERβ) treated with 10 nM E2 for 24 h. The data demonstrate that ~80% of genes regulated by either ER isoform are regulated uniquely (e.g. not regulated by the other ER isoform). Data taken from [83].
333
Chapter 19 Sex Steroid Effects on Bone Metabolism
The mechanism of action of estrogen on growth plate cartilage is at present poorly understood. It appears, however, that estrogen stimulates maturation of cartilage without increasing the growth rate [93, 94]. In addition, the effects of estrogen on cartilage growth and maturation are likely mediated by interactions with growth hormone and insulin-like growth factor-I. Thus, although estrogen stimulates growth hormone secretion, it also inhibits IGF-I synthesis in the growth plate [95]. Therefore, the longitudinal growth rate may be influenced by the relative actions of estrogen and growth hormone on the growth plate. Finally, studies on selective estrogen receptor modulators (SERMs), such as raloxifene, are providing new insights into possible alternate pathways for estrogen action in bone. These compounds have tissue-specific actions and can serve as either estrogen antagonists or agonists. For example, they can affect the skeleton without stimulating breast or endometrial tissue [96]. Although they bind the ER, they induce conformational changes in the ER that are different from estradiol that affects transcriptional coregulator interactions [21, 97–99]. The SERM–ER complex may modulate the expression of genes via alternate, nonclassical pathways, rather than activating estrogen-responsive genes via the classical steroid pathway. For example, estrogen regulates the expression of the IL-6 gene, not by direct interaction of the estrogen/ER complex with DNA, but by binding and inhibiting the activity of transcription factors, such as NF-κβ and C/EBPβ, which are required for IL-6 gene transcription [70, 100, 101] (Fig. 2). It is possible that SERMs may also activate this pathway in bone cells. Similarly, raloxifene has been shown to activate the gene for TGF-β3 in human osteosarcoma MG63 cells transfected with an ERα lacking the DBD. This effect may be mediated by a raloxifene response element in DNA that is distinct from the estrogen response element [102, 103]. Both ERα and ERβ interact with AP-1 sites (fos/jun) [104]; however, the ER isoforms have different affinities and transcriptional activation potentials for various SERMs [105, 106]. When estradiol is present, ERα activates transcription from the AP-1 site, whereas ERβ inhibits transcription from this site. In contrast, in the presence of the SERM raloxifene, ERβ stimulates transcription from the AP-1 site, whereas ERα inhibits it [104]. Further elucidation of the mechanism of action of these SERMs is likely to provide new insights into estrogen action in bone and in other tissues.
B. Progesterone The effects of progesterone on bone metabolism are much less clear than those of estrogen. However, progesterone
receptors have been identified in primary human osteoblastic cells, human osteosarcoma cells, and immortalized fetal osteoblast cells [107, 108], although estrogen is generally required to induce progesterone receptors in these cells, as in other cell systems [109, 110]. In addition, progesterone has been shown to increase the proliferation and differentiation of human osteoblastic cells [111]. Although progesterone can stimulate transcriptional activation of bone markers such as type I collagen, osteocalcin [112] and other bone-related cytokines such as IGF-II [111], transforming growth factor TGFβ1–3 [113], progesterone does not appear to have consistent effects on postmenopausal bone loss in humans [112, 114]. Furthermore, while estrogen treatment of primary human osteoblasts stimulates OPG mRNA production [55–58], progesterone had no significant effect on OPG expression [57]. Thus, the effects of progesterone on the human skeleton likely differ significantly from those of estrogen. Indeed, since progesterone is a known antagonist of estrogen in certain cell processes, and an agonist in others, it is possible that the major role of progesterone is to modulate estrogen action in the processes of bone metabolism.
C. Androgens Human bone cells have been demonstrated to have specific androgen-binding sites [115], and the androgen receptor concentrations found have been similar to estrogen receptor concentrations in these cells (Fig. 4). Moreover, the number of specific androgen-binding sites in osteoblasts (1000–3000 sites/cell) is similar to the number of binding sites present in other androgen-responsive tissues, such as the prostate [116]. In addition to the androgen receptor, bone cells have also been shown to have 5α-reductase activity [117], which is responsible for the conversion of testosterone to DHT, the major biologically active metabolite of testosterone in most tissues. Testosterone may also be aromatized to estrogens in various tissues by the microsomal enzyme, aromatase, and there is also evidence that bone cells possess an aromatase activity [118]. As in the case of estrogen, there is evidence that androgens inhibit bone resorption. Thus, orchiectomy, like ovariectomy, is associated with increased bone resorption and rapid bone loss [119]. In addition, Tenover [120] has shown that 3 months of testosterone treatment of men with borderline serum testosterone levels resulted in a 28% reduction in urinary hydroxyproline excretion, an index of bone resorption. In another study of eugonadal men with osteoporosis, who were treated with 6 months of intramuscular testosterone, testosterone therapy was associated
334 with significant reductions in bone resorption markers [121]. Finally, Riggs et al. [122] performed bone biopsies in 29 women with postmenopausal osteoporosis before and after either estrogen or oxandrolone (a synthetic androgen) treatment and found similar decreases in resorption surfaces, and comparable changes in bone resorption rate assessed by calcium kinetics. Androgen receptors have also been identified on osteoclasts, and androgens, like estrogen, have been shown to directly decrease bone resorption in vitro by avian and human osteoclast-like cells [123]. In addition to these direct effects on osteoclasts, androgens have also been shown to regulate the production of a number of bone-resorbing factors by osteoblasts or marrow stromal cells. Thus, both DHT and testosterone have been shown to reduce PGE2 production in calvarial organ cultures exposed to stimulation with parathyroid hormone or IL-1 [124]. Similarly, androgens have been shown to inhibit IL-6 production by stromal and osteoblastic cells as well as the stimulation of osteoclastogenesis by marrow osteoclast precursors [125, 126]. While androgens have generally been considered anabolic for bone, the evidence for this in adults is relatively sparse. Thus, in the studies noted above, Tenover [120] found that testosterone treatment had no effect on the bone formation marker, serum osteocalcin. In contrast, Raisz et al. [127] found that postmenopausal women treated with estrogen plus 2.5 mg of methyltestosterone had a 24% higher serum osteocalcin level after 3 months of treatment, as compared to a 40% lower osteocalcin level in the women treated with estrogen alone. In addition, Baran et al. [128] performed bone biopsies in a hypogonadal man before and after 6 months of testosterone treatment and noted increases in relative osteoid volume, total osteoid surface, linear extent of bone formation, and bone mineralization. Similarly, Cantatore et al. [129] administered synthetic androgens in postmenopausal women and found increases in makers of bone formation, at least over the short term (3–6 months), and Morley et al. [130] also noted that 3 months of testosterone therapy in elderly hypogonadal men resulted in a significant increase in serum osteocalcin levels. Thus, at least over the short term, androgens may have an anabolic effect on bone, although further studies are needed to address this issue. As in the case of estrogen, in vitro studies of androgen effects on osteoblast proliferation and differentiation have yielded inconsistent results. Androgens have been shown to have mitogenic effects on normal and transformed osteoblast-like cells in most [81, 131], but not all systems [132, 133]. With respect to osteoblast differentiation, androgens have been shown to increase alkaline phosphatase activity in primary cultures of osteoblast-like cells and the
DAVID G. MONROE ET AL.
percentage of alkaline phosphatase-positive cells, suggesting a shift towards a more differentiated phenotype [131, 134, 135]. On the other hand, studies using a fetal human osteoblast cell line stably transfected with the androgen receptor have found a decrease in alkaline phosphatase activity in these cells following DHT exposure [133]. Androgen effects on type I collagen synthesis have been variable, with some studies showing a stimulatory effect [81, 132, 134], and others finding no effect [124], or even a decrease [133] in collagen synthesis. Finally, studies using primary osteoblast cultures have suggested that some of the effects of androgens in these cells may be mediated, at least in part, by increased TGF-β production, or by alterations in the insulin-like growth factor/insulin-like growth factor-binding protein system [136, 137]. In addition to testosterone and DHT, there is some evidence that adrenal androgens may have significant effects on bone metabolism. Thus, Durbridge et al. [138] found that, in rats, adrenalectomy alone resulted in loss of metaphyseal trabecular bone of an extent similar to that produced by oophorectomy. These investigators interpreted these data as being consistent with an important role for adrenal androgens in the maintenance of skeletal mass in rats. In addition, Turner et al. [139] have shown that, in rats, treatment with DHEA reduced the loss of cancellous bone following oophorectomy, indicating that adrenal androgens may prevent the bone loss induced by estrogen deficiency. DHEA, which is essentially a prohormone, can be converted locally in various tissues to androstenedione and then into potent androgens and/or estrogens, and it has generally been assumed that the effects of adrenal androgens on target tissues are due to the effects of these androgens or estrogens [140]. However, the recent description of specific binding sites for DHEA [141] raises the possibility that DHEA or similar compounds may have direct effects on bone. In support of this, Bodine et al. [7] have shown that DHEA caused a rapid reduction (30 minutes post treatment) in steady-state levels of c-fos mRNA in normal human osteoblastic cells, whereas testosterone or DHT had no effects on c-fos expression. However, all three agents resulted in significant increases in TGF-β activity in these cells. These studies demonstrate that bone cells have the ability to respond to DHEA, and that the effects of DHEA may or may not be identical to the effects of testosterone, depending on the specific metabolism and perhaps independent action of DHEA in bone cells. Perhaps the major role of androgens in bone metabolism occurs at puberty. As in females, adolescence in males is associated with rapid longitudinal growth, and marked increases in axial and appendicular bone mass [142]. The increase in bone mass is associated with increases in markers of bone formation and is closely linked to pubertal
Chapter 19 Sex Steroid Effects on Bone Metabolism
stage and to testosterone levels [92, 143], suggesting that testicular androgen production plays a major role in mediating the pubertal increase in bone mass. In addition, however, the increase in adrenal androgens that occurs before the onset of puberty also likely affects the acquisition of bone mass. Thus, longitudinal growth rate increases during adrenarche [144] and bone mass has been shown to increase before the onset of sexual development, likely related to the rise in adrenal androgens [145]. Androgens are also clearly responsible for the sexual dimorphism of the skeleton that is evident during adolescence: the male skeleton is larger in most dimensions compared to the female skeleton, the diameter and cortical thickness of long bones is greater in men than in women, and vertebral size is larger in men [146]. Despite these unequivocal effects of androgens on skeletal growth, however, it remains unresolved which of these effects are due to direct androgen effects on bone and which are due to indirect effects. For example, like estrogen, testosterone can influence growth hormone secretion [147] or the local production of IGF-I or its binding proteins [148], and thus affect bone mass indirectly. It is likely, therefore, that androgen effects on bone during adolescence are due both to direct and indirect effects.
V. EFFECTS OF ESTROGENS AND ANDROGENS ON BONE METABOLISM IN MEN VERSUS WOMEN Based on the sexual dimorphism of the skeleton, it had generally been believed that estrogens were the major determinant of bone metabolism in women and androgens primarily affected bone metabolism in men. Recent evidence, however, indicates considerable overlap of the effects of sex steroids on bone in men and women. Data from several “experiments of nature” are consistent, for example, with the concept that estrogen plays a major role in bone metabolism in men. Smith et al. [149] described a male with homozygous mutations in the ERα gene who, even in the presence of normal testosterone levels, had unfused epiphyses and marked osteopenia, along with elevated indices of bone turnover. Subsequently, Morishima et al. [150] and Carani et al. [151] reported clinical findings in two males with homozygous mutations in the aromatase gene. In both instances, bone mineral density was significantly reduced and bone turnover markers were markedly elevated despite normal testosterone levels. Treatment with testosterone did not significantly affect bone metabolism in one patient, whereas treatment with estrogen markedly increased bone mineral density in both
335 patients [151, 152]. Moreover, a number of observational studies [153–157] have demonstrated that serum estrogen levels (particularly bioavailable estrogen levels) are better predictors of bone density in men at all measured sites except at some cortical bone sites in the appendicular skeleton. In addition, rates of bone loss in elderly men are also associated with serum estrogen levels [158–160]. Direct evidence for the key role played by estrogen in bone metabolism in men was provided by Falahati-Nini et al. [161] who suppressed endogenous testosterone and estrogen production in a group (n = 59) of older men (mean age, 68 years) using a combination of a GnRH agonist and an aromatase inhibitor. Physiologic estradiol and testosterone levels were maintained by placing the men on the respective patches. After baseline measurements of bone turnover markers, the men were then randomized to one of four groups in order to rigorously delineate the relative contributions of estrogen and testosterone towards regulating bone resorption and formation. Group A (−T, −E) had both patches withdrawn; Group B (−T, +E) had the testosterone patch withdrawn but continued the estrogen patch; Group C (+T, −E) had the estrogen patch withdrawn, but continued the testosterone patch; and Group D (+T, +E) continued both patches. All subjects were continued on GnRH and the aromatase inhibitor. Figure 7A shows the changes in the bone resorption markers, urine deoxypyridinoline (Dpd) and N-telopeptide of type I collagen (NTx), after the subjects had been on the variable treatments for a period of 3 weeks. As is evident, significant increases in both urinary Dpd and NTx excretion, Group A (−T, −E), were prevented completely by continuing T and E replacement [Group D (+T, +E)]. E alone (Group B) was almost completely able to prevent the increase in bone resorption, whereas T alone (Group C) was much less effective. Using a two-factor ANOVA model, the effects of E on urinary Dpd and NTx excretion were highly significant (P = 0.005 and 0.0002, respectively). E accounted for 70% or more of the total effect of sex steroids on bone resorption in these older men, while T could account for no more than 30% of the effect. Using a somewhat different design, Leder et al. [162] have confirmed an independent effect of T on bone resorption, although the data in the aggregate clearly favor a more prominent effect of E on the control of bone resorption in men. Figure 7B shows the corresponding changes in the bone formation markers, serum osteocalcin, and PINP. The reductions in both osteocalcin and PINP levels with the induction of sex steroid deficiency (Group A) were prevented with continued E and T replacement (Group D). Interestingly, serum osteocalcin, which is a marker of function of the mature osteoblast and osteocyte, was maintained
336
DAVID G. MONROE ET AL.
Figure 7 Percent changes in (A) bone resorption markers (urinary deoxypyridinoline [Dpd] and N-telopeptide of type I collagen [NTx]) and (B) bone formation markers (serum osteocalcin and N-terminal extension peptide of type I collagen [PINP]) in a group of elderly men (mean age 68 years) made acutely hypogonadal and treated with an aromatase inhibitor (Group A), treated with E alone (Group B), T alone (Group C), or both (Group D). See text for details. Asterisks indicate significance for change from baseline: *, P < 0.05; **, P < 0.01; ***, P < 0.001. The estrogen and testosterone effects were analyzed using two-factor ANOVA models: Dpd: E effect, P = 0.005; T effect, P = 0.232; NTx: E effect, P = 0.0002; T effect, P = 0.085; Osteocalcin: E effect, P = 0.002; T effect, P = 0.013; PINP: E effect, P = 0.0001; T effect, P = 0.452. (Adapted from Falahati-Nini et al. [161].)
by either E or T (ANOVA P values of 0.002 and 0.013, respectively). By contrast, serum PINP, which represents type I collagen synthesis throughout the various stages of osteoblast differentiation, was maintained by E (ANOVA P value 0.0001), but not T. Collectively, these findings provided conclusive proof of an important (and indeed, dominant) role for E in bone metabolism in the mature skeleton of adult men. Similar findings were subsequently reported by Taxel et al. [163] in a study of 15 elderly men treated with an aromatase inhibitor for 9 weeks, where suppression of E production resulted in significant increases in bone resorption markers and a suppression of bone formation markers. Similarly, several lines of evidence indicate that androgens may have important effects on bone in women. In premenopausal women, serum androgen levels are significantly related to bone mineral density [155, 164], suggesting that androgens may contribute to the development of peak bone mass in women. Moreover, hyperandrogenic women with hirsutism have increased bone mass relative to control women [165]. Finally, testosterone levels correlate
with bone mass and rates of fall in bone mass in perimenopausal women [164].
VI. EFFECTS OF SEX STEROIDS ON EXTRASKELETAL CALCIUM HOMEOSTASIS In addition to direct effects on bone cells and on bone remodeling, it is becoming clear that sex steroids may affect bone turnover indirectly through an action on extraskeletal calcium homeostasis, including effects on intestinal calcium absorption, renal calcium handling, and perhaps direct effects on PTH secretion. Thus, Gallagher et al. [166] found that estrogen treatment increased both serum total 1,25(OH)2D levels and calcium absorption in postmenopausal osteoporotic women, although this effect may have been due to the increase in serum PTH in response to the estrogen-induced decrease in bone resorption. However, in perimenopausal women before and
337
Chapter 19 Sex Steroid Effects on Bone Metabolism
6 months after oophorectomy, Gennari et al. [167] found that the increase in calcium absorption in response to treatment with 1,25(OH)2D was blunted in the presence of estrogen deficiency, suggesting a direct effect of estrogen on intestinal calcium absorption. Of note, estrogen receptors have been found in an untransformed cell line from rat small intestinal crypts (IEC-6 cells) as well as epithelial cells from rat intestine using both classical receptor-binding techniques and RT-PCR [168]. Thus, estrogen may affect bone turnover both directly by acting on bone cells and indirectly by altering intestinal calcium absorption and hence serum PTH levels. Estrogen receptors have also been demonstrated in renal tubules [169], and estrogen has been reported to increase adenylate cyclase activity of renal membranes from hens [170] and from cultured opossum kidney cells [171]. Using estimates based on regression analysis, Nordin et al. [172] found that early postmenopausal women had a “renal calcium leak” that they attributed to estrogen deficiency. McKane et al. [173] assessed renal calcium transport by direct measurements both at baseline and during administration of a saturating dose of PTH in early postmenopausal women before and after 6 months of estrogen treatment. They demonstrated a PTH-independent decrease in tubular calcium reabsorption in the estrogendeficient women compared to the estrogen-replete women, consistent with a direct effect of estrogen on renal calcium conservation. In addition to the indirect effects on serum PTH levels, there is evidence that estrogen may directly regulate PTH secretion. Some [174], but not other [175], investigators have demonstrated the presence of estrogen receptors in parathyroid glands. Cosman et al. [176] reported that chronic estrogen replacement therapy in elderly women decreased maximal PTH secretory rate, as assessed by measurement of serum PTH during EDTA-induced hypocalcemia, although Vincent et al. [177] failed to demonstrate an effect of acute estrogen treatment or withdrawal on PTH secretion. Studies in experimental animals and parathyroid cells in vitro suggest that estrogen increases PTH secretion [178]. Thus, estrogen administration to oophorectomized rats increased PTH mRNA in the parathyroid glands four-fold [174], and in vitro estrogen treatment of surgically removed human hyperplastic parathyroid glands increased PTH secretion in a dose- and time-dependent manner [177]. Some of the contrasting effect of estrogen on PTH secretion may be related to acute versus chronic effects, although more work is needed to address this issue. There are several lines of evidence suggesting that androgens may also have similar, indirect effects on bone metabolism. Thus, there are data indicating that androgens
may significantly enhance intestinal calcium absorption. Hope et al. [179] reported that duodenal active calcium transport decreased in male rats following orchiectomy and testosterone treatment reversed this. In early studies, Lafferty et al. [180] found that calcium absorption increased following short-term (2–3 months) testosterone treatment in three subjects studied, although these changes did not persist with long-term therapy. Need et al. [181] studied 27 women with postmenopausal osteoporosis and found a significant increase in the hourly fractional rate of radiocalcium absorption from (mean ± SEM) 0.79 ± 0.06 to 0.93 ± 0.05 following 3 months of therapy with the anabolic steroid, nandrolone decanoate. Some evidence also suggests that androgens may affect renal calcium transport. The androgen receptor gene is expressed in kidney epithelial cells [182] and androgens have been shown to modulate calcium fluxes in mouse kidney cortex preparations [183]. Reifenstein and Albright [184] initially noted that testosterone propionate decreased the urinary excretion of calcium. Several studies using synthetic androgens have also suggested an effect of androgens on renal calcium handling. Thus, Chesnut et al. [185] noted a 32% decrease in urinary calcium excretion in 23 women with postmenopausal osteoporosis following 8–32 months of therapy with the anabolic steroid, stanozolol. Similarly, Need et al. [181] indirectly estimated renal tubular calcium reabsorption and found that this increased significantly in 27 women with postmenopausal osteoporosis treated for 3 months with nandrolone decanoate. Finally, in recent studies, Mason et al. [186] showed that DHT administration increased the tubular reabsorption of calcium in oophorectomized rats. Thus, both estrogen and androgens may affect bone metabolism indirectly via alterations in overall calcium homeostasis and subsequent changes in serum PTH levels. Indeed, loss of these indirect effects of sex steroids in elderly women and in some men may account, in large part, for the age-related increase in serum PTH levels in both sexes [187]. This, in turn, results in an age-related increase in bone turnover and, in the setting of a concomitant defect in osteoblast function, in bone loss.
VII. SUMMARY Sex steroids clearly have major effects on all aspects of bone metabolism. These include effects on longitudinal growth at the time of puberty, the acquisition of peak bone mass, and the regulation of bone turnover in adults. While there are several candidate factors for mediating the autocrine and paracrine effects of estrogen and, to a lessor extent, androgens, on osteoclasts and osteoblasts, the precise
338
DAVID G. MONROE ET AL.
mechanism of direct sex steroid effects on the skeleton remains to be clearly defined. Estrogen likely has a significant effect on bone not just in females, but also in males. Conversely, androgens also have important skeletal effects in both sexes. In addition, sex steroids also affect bone turnover indirectly, via extraskeletal effects on calcium homeostasis. Collectively, these direct and indirect actions of sex steroids on bone and calcium metabolism are critical for normal bone mass acquisition and maintenance.
Acknowledgment Supported by Research Grants AG-004875, AR-027065 and AR-30582 from the National Institutes of Health, U.S. Public Health Service.
References 1. Monroe, D. G., and Spelsberg, T. (2004). General mechanisms of steroid receptors and receptor coregulator action. In: “Advances in Molecular and Cell Biology” (V. Miller, M. Hay, eds.), 34th ed., Vol. 34, pp. 39–48. Elsevier B.V., Amsterdam. 2. Hsueh, A. J. W., and Billig, H. (1995). Ovarian hormone synthesis and mechanism of action. In: “Endocrinology” (L. J. DeGroot, M. Besser, H. G. Burger, J. L. Jameon, D. L. Loriaux, J. C. Marshall, W. D. Odell, J. T. Potts, A. H. Rubenstein, eds.), pp. 2019–2030. W.B. Saunders, Philadelphia, PA. 3. Simpson, E. R., Mahendroo, M. S., Means, G. D., Kilgore, M. W., Hinshelwood, M. M., Graham-Lorence, S., Amarneh, B., Ito, Y., Fisher, C. R., Michael, M. D., Mendelson, C. R., and Bulun, S. E. (1994). Aromatase cytochrome P450, the enzyme responsible for estrogen biosynthesis. Endocr. Rev. 15(3), 342–355. 4. Simpson, E. R. (2003). Sources of estrogen and their importance. J. Steroid Biochem. Mol. Biol. 86(3–5), 225–230. 5. Nawata, H., Tanaka, S., Takayanagi, R., Sakai, Y., Yanase, T., Ikuyama, S., and Haji, M. (1995). Aromatase in bone cell: association with osteoporosis in postmenopausal women. J. Steroid Biochem. Mol. Biol. 53(1–6), 165–174. 6. Dinneen, S., Alzaid, A., Miles, J., and Rizza, R. (1993). Metabolic effects of the nocturnal rise in cortisol on carbohydrate metabolism in normal humans. J. Clin. Invest. 92(5), 2283–2290. 7. Bodine, P. V., Riggs, B. L., and Spelsberg, T. C. (1995). Regulation of c-fos expression and TGF-beta production by gonadal and adrenal androgens in normal human osteoblastic cells. J. Steroid Biochem. Mol. Biol. 52(2), 149–158. 8. Manni, A., Pardridge, W. M., Cefalu, W., Nisula, B. C., Bardin, C. W., Santner, S. J., and Santen, R. J. (1985). Bioavailability of albumin-bound testosterone. J. Clin. Endocrinol. Metab. 61(4), 705–710. 9. Cekan, S. Z., Xing, S., and Ritzen, M. (1984). On the binding of steroid sulfates to albumin. Experientia 40(9), 949–951. 10. Monroe, D. G., and Spelsberg, T. C. (2004). Gonadal steroids and receptors. In: “Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism” (M. J. Favus, ed), 5th ed., pp. 32–38. Lippincott, Williams and Wilkins, Philadelphia, PA. 11. Monroe, D. G., and Spelsberg, T. C. (2003). A case for estrogen receptors on cell membranes and nongenomic actions of estrogen. Calcif. Tissue Int. 72(3), 183–184. 12. Monroe, D. G., Secreto, F. J., and Spelsberg, T. (2003). Overview of estrogen action in osteoblasts: Role of the ligand, receptor, and the coregulators. J. Musculoskeletal Neuronal. Int. 3(4), 357–362.
13. Pratt, W. B., and Toft, D. O. (2003). Regulation of signaling protein function and trafficking by the hsp90/hsp70-based chaperone machinery. Exp. Biol. Med. (Maywood) 228(2), 111–133. 14. Smith, D. F., and Toft, D. O. (1993). Steroid receptors and their associated proteins. Mol. Endocrinol. 7(1), 4–11. 15. Cowley, S. M., Hoare, S., Mosselman, S., and Parker, M. G. (1997). Estrogen receptors alpha and beta form heterodimers on DNA. J. Biol. Chem. 272(32), 19858–19862. 16. Jisa, E., and Jungbauer, A. (2003). Kinetic analysis of estrogen receptor homo- and heterodimerization in vitro. J. Steroid Biochem. Mol. Biol. 84(2–3), 141–148. 17. Tsai, S. Y., Carlstedt-Duke, J., Weigel, N. L., Dahlman, K., Gustafsson, J. A., Tsai, M. J., and O’Malley, B. W. (1988). Molecular interactions of steroid hormone receptor with its enhancer element: evidence for receptor dimer formation. Cell 55(2), 361–369. 18. Driggers, P. H., and Segars, J. H. (2002). Estrogen action and cytoplasmic signaling pathways. Part II: the role of growth factors and phosphorylation in estrogen signaling. Trends Endocrinol. Metab. 13(10), 422–427. 19. Lannigan, D. A. (2003). Estrogen receptor phosphorylation. Steroids 68(1), 1–9. 20. Takimoto, G. S., Tasset, D. M., Eppert, A. C., and Horwitz, K. B. (1992). Hormone-induced progesterone receptor phosphorylation consists of sequential DNA-independent and DNA-dependent stages: analysis with zinc finger mutants and the progesterone antagonist ZK98299. Proc. Natl Acad. Sci. USA 89(7), 3050–3054. 21. Smith, C. L., and O’Malley, B. W. (2004). Coregulator function: a key to understanding tissue specificity of selective receptor modulators. Endocr. Rev. 25(1), 45–71. 22. McDonnell, D. P., and Norris, J. D. (1997). Analysis of the molecular pharmacology of estrogen receptor agonists and antagonists provides insights into the mechanism of action of estrogen in bone. Osteoporos. Int. 7 Suppl 1, S29–34. 23. Kousteni, S., Bellido, T., Plotkin, L. I., O’Brien, C. A., Bodenner, D. L., Han, L., Han, K., DiGregorio, G. B., Katzenellenbogen, J. A., Katzenellenbogen, B. S., Roberson, P. K., Weinstein, R. S., Jilka, R. L., and Manolagas, S. C. (2001). Nongenotropic, sex-nonspecific signaling through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell 104(5), 719–730. 24. Kousteni, S., Chen, J. R., Bellido, T., Han, L., Ali, A. A., O’Brien, C. A., Plotkin, L., Fu, Q., Mancino, A. T., Wen, Y., Vertino, A. M., Powers, C. C., Stewart, S. A., Ebert, R., Parfitt, A. M., Weinstein, R. S., Jilka, R. L., and Manolagas, S. C. (2002). Reversal of bone loss in mice by nongenotropic signaling of sex steroids. Science 298(5594), 843–846. 25. Kousteni, S., Han, L., Chen, J. R., Almeida, M., Plotkin, L. I., Bellido, T., and Manolagas, S. C. (2003). Kinase-mediated regulation of common transcription factors accounts for the bone-protective effects of sex steroids. J. Clin. Invest. 111(11), 1651–1664. 26. Parl, F. F. (2000). Estrogen Receptor and Breast Cancer. IOS Press, Amsterdam. 27. Kuiper, G. G., Enmark, E., Pelto-Huikko, M., Nilsson, S., and Gustafsson, J. A. (1996). Cloning of a novel receptor expressed in rat prostate and ovary. Proc. Natl Acad. Sci. USA 93(12), 5925–5930. 28. Arts, J., Kuiper, G. G., Janssen, J. M., Gustafsson, J. A., Lowik, C. W., Pols, H. A., van Leeuwen, J. P. (1997). Differential expression of estrogen receptors alpha and beta mRNA during differentiation of human osteoblast SV-HFO cells. Endocrinology 138(11), 5067–5070. 29. Onoe, Y., Miyaura, C., Ohta, H., Nozawa, S., and Suda, T. (1997). Expression of estrogen receptor beta in rat bone. Endocrinology 138(10), 4509–4512. 30. Sims, N. A., Clement-Lacroix, P., Minet, D., Fraslon-Vanhulle, C., Gaillard-Kelly, M., Resche-Rigon, M., and Baron, R. (2003). A functional androgen receptor is not sufficient to allow estradiol to protect bone after gonadectomy in estradiol receptor-deficient mice. J. Clin. Invest. 111(9), 1319–1327. 31. Sims, N. A., Dupont, S., Krust, A., Clement-Lacroix, P., Minet, D., Resche-Rigon, M., Gaillard-Kelly, M., and Baron, R. (2002). Deletion of
Chapter 19 Sex Steroid Effects on Bone Metabolism
32.
33.
34.
35.
36.
37.
38.
39.
40.
41. 42.
43.
44.
45.
46.
47.
48.
49.
estrogen receptors reveals a regulatory role for estrogen receptors-beta in bone remodeling in females but not in males. Bone 30(1), 18–25. Onate, S. A., Tsai, S. Y., Tsai, M. J., and O’Malley, B. W. (1995). Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science 270(5240), 1354-1357. McKenna, N. J., Lanz, R. B., and O’Malley, B. W. (1999). Nuclear receptor coregulators: cellular and molecular biology. Endocr. Rev. 20(3), 321–344. Voegel, J. J., Heine, M. J., Zechel, C., Chambon, P., and Gronemeyer, H. (1996). TIF2, a 160 kDa transcriptional mediator for the ligand-dependent activation function AF-2 of nuclear receptors. Embo. J. 15(14), 3667–3675. Li, H., Gomes, P. J., and Chen, J. D. (1997). RAC3, a steroid/nuclear receptor-associated coactivator that is related to SRC-1 and TIF2. Proc. Natl Acad. Sci. USA 94(16), 8479–8484. Xu, J., Qiu, Y., DeMayo, F. J., Tsai, S. Y., Tsai, M. J., and O’Malley, B. W. (1998). Partial hormone resistance in mice with disruption of the steroid receptor coactivator-1 (SRC-1) gene. Science 279(5358), 1922–1925. Modder, U. I., Sanyal, A., Kearns, A. E., Sibonga, J. D., Nishihara, E., Xu, J., O’Malley, B. W., Ritman, E. L., Riggs, B. L., Spelsberg, T. C., and Khosla, S. (2004). Effects of loss of steroid receptor coactivator-1 on the skeletal response to estrogen in mice. Endocrinology 145(2), 913–921. Xu, J., Liao, L., Ning, G., Yoshida-Komiya, H., Deng, C., and O’Malley, B. W. (2000). The steroid receptor coactivator SRC-3 (p/CIP/RAC3/AIB1/ACTR/TRAM-1) is required for normal growth, puberty, female reproductive function, and mammary gland development. Proc. Natl Acad. Sci. USA 97(12), 6379–6384. Martini, P. G., and Katzenellenbogen, B. S. (2003). Modulation of estrogen receptor activity by selective coregulators. J. Steroid Biochem. Mol. Biol. 85(2–5), 117–122. Heery, D. M., Kalkhoven, E., Hoare, S., and Parker, M. G. (1997). A signature motif in transcriptional co-activators mediates binding to nuclear receptors. Nature 387(6634), 733–736. Leo, C., and Chen, J. D. (2000). The SRC family of nuclear receptor coactivators. Gene 245(1), 1–11. Monroe, D. G., Johnsen, S. A., Subramaniam, M., Getz, B. J., Khosla, S., Riggs, B. L., and Spelsberg, T. C. (2003). Mutual antagonism of estrogen receptors alpha and beta and their preferred interactions with steroid receptor coactivators in human osteoblastic cell lines. J. Endocrinol. 176(3), 349–357. Riggs, B. L., Khosla, S., and Melton, L. J., III (2002). Sex steroids and the construction and conservation of the adult skeleton. Endocr. Rev. 23(3), 279–302. Riggs, B. L., Wahner, H. W., Melton, L. J., III, Richelson, L. S., Judd, H. L., and Offord, K. P. (1986). Rates of bone loss in the appendicular and axial skeletons of women. Evidence of substantial vertebral bone loss before menopause. J. Clin. Invest. 77(5), 1487–1491. Eriksen, E. F., Colvard, D. S., Berg, N. J., Graham, M. L., Mann, K. G., Spelsberg, T. C., and Riggs, B. L. (1988). Evidence of estrogen receptors in normal human osteoblast-like cells. Science 241(4861), 84–86. Gyda, M., Corisdeo, S., Zaidi, M., and Troen, B. R. (2001). Macrophage colony-stimulating factor suppresses osteoblast formation. Biochem. Biophys. Res. Commun. 285(2), 328–334. Hoyland, J. A., Mee, A. P., Baird, P., Braidman, I. P., Mawer, E. B., and Freemont, A. J. (1997). Demonstration of estrogen receptor mRNA in bone using in situ reverse-transcriptase polymerase chain reaction. Bone 20(2), 87–92. Vidal, O., Kindblom, L. G., and Ohlsson, C. (1999). Expression and localization of estrogen receptor-beta in murine and human bone. J. Bone Miner. Res. 14(6), 923–929. Oursler, M. J., Pederson, L., Fitzpatrick, L., Riggs, B. L., and Spelsberg, T. (1994). Human giant cell tumors of the bone (osteoclastomas) are estrogen target cells. Proc. Natl Acad. Sci. USA 91(12), 5227–5231.
339 50. Riggs, B. L., and Melton, L. J., III (1992). The prevention and treatment of osteoporosis [published erratum appears in N. Engl. J. Med. 1993 Jan 7; 328(1):65]. N. Engl. J. Med. 327, 620–627. 51. Khosla, S. (2001). Minireview: the OPG/RANKL/RANK system. Endocrinology 142(12), 5050–5055. 52. Hofbauer, L. C., Khosla, S., Dunstan, C. R., Lacey, D. L., Boyle, W. J., and Riggs, B. L. (2000). The roles of osteoprotegerin and osteoprotegerin ligand in the paracrine regulation of bone resorption. J. Bone Miner. Res. 15(1), 2–12. 53. Troen, B. R. (2003). Molecular mechanisms underlying osteoclast formation and activation. Exp. Gerontol. 38(6), 605–614. 54. Bell, N. H. (2001). Advances in the treatment of osteoporosis. Curr. Drug Targets Immune Endocr. Metabol. Disord. 1(1), 93–102. 55. Bord, S., Ireland, D. C., Beavan, S. R., and Compston, J. E. (2003). The effects of estrogen on osteoprotegerin, RANKL, and estrogen receptor expression in human osteoblasts. Bone 32(2), 136–141. 56. Hofbauer, L. C., Khosla, S., Dunstan, C. R., Lacey, D. L., Spelsberg, T. C., and Riggs, B. L. (1999). Estrogen stimulates gene expression and protein production of osteoprotegerin in human osteoblastic cells. Endocrinology 140(9), 4367–4370. 57. Liao, E. Y., Luo, X. H., and Su, X. (2002). Comparison of the effects of 17β-E2 and progesterone on the expression of osteoprotegerin in normal human osteoblast-like cells. J. Endocrinol. Invest. 25(9), 785–790. 58. Saika, M., Inoue, D., Kido, S., and Matsumoto, T. (2001). 17β-estradiol stimulates expression of osteoprotegerin by a mouse stromal cell line, ST-2, via estrogen receptor-alpha. Endocrinology 142(6), 2205–2212. 59. Eghbali-Fatourechi, G., Khosla, S., Sanyal, A., Boyle, W. J., Lacey, D. L., and Riggs, B. L. (2003). Role of RANK ligand in mediating increased bone resorption in early postmenopausal women. J. Clin. Invest. 111(8), 1221–1230. 60. Srivastava, S., Toraldo, G., Weitzmann, M. N., Cenci, S., Ross, F. P., and Pacifici, R. (2001). Estrogen decreases osteoclast formation by down-regulating receptor activator of NF-κB ligand (RANKL)induced JNK activation. J. Biol. Chem. 276(12), 8836–8840. 61. Kanatani, M., Sugimoto, T., Takahashi, Y., Kaji, H., Kitazawa, R., and Chihara, K. (1998). Estrogen via the estrogen receptor blocks cAMPmediated parathyroid hormone (PTH)-stimulated osteoclast formation. J. Bone Miner. Res. 13(5), 854–862. 62. Manolagas, S. C., and Jilka, R. L. (1995). Bone marrow, cytokines, and bone remodeling. Emerging insights into the pathophysiology of osteoporosis. N. Engl. J. Med. 332(5), 305–311. 63. Pacifici, R., Brown, C., Puscheck, E., Friedrich, E., Slatopolsky, E., Maggio, D., McCracken, R., and Avioli, L. V. (1991). Effect of surgical menopause and estrogen replacement on cytokine release from human blood mononuclear cells. Proc. Natl Acad. Sci. USA 88(12), 5134–5138. 64. Tanaka, S., Takahashi, N., Udagawa, N., Tamura, T., Akatsu, T., Stanley, E. R., Kurokawa, T., and Suda, T. (1993). Macrophage colonystimulating factor is indispensable for both proliferation and differentiation of osteoclast progenitors. J. Clin. Invest. 91(1), 257–263. 65. Kimble, R. B., Vannice, J. L., Bloedow, D. C., Thompson, R. C., Hopfer, W., Kung, V. T., Brownfield, C., and Pacifici, R. (1994). Interleukin-1 receptor antagonist decreases bone loss and bone resorption in ovariectomized rats. J. Clin. Invest. 93(5), 1959–1967. 66. Ammann, P., Rizzoli, R., Bonjour, J. P., Bourrin, S., Meyer, J. M., Vassalli, P., and Garcia, I. (1997). Transgenic mice expressing soluble tumor necrosis factor-receptor are protected against bone loss caused by estrogen deficiency. J. Clin. Invest. 99(7), 1699–1703. 67. Kimble, R. B., Srivastava, S., Ross, F. P., Matayoshi, A., and Pacifici, R. (1996). Estrogen deficiency increases the ability of stromal cells to support murine osteoclastogenesis via an interleukin-1and tumor necrosis factor-mediated stimulation of macrophage colony-stimulating factor production. J. Biol. Chem. 271(46), 28890–28897. 68. Nanes, M. S. (2003). Tumor necrosis factor-alpha: molecular and cellular mechanisms in skeletal pathology. Gene 321, 1–15.
340 69. Kitazawa, R., Kimble, R. B., Vannice, J. L., Kung, V. T., and Pacifici, R. (1994). Interleukin-1 receptor antagonist and tumor necrosis factor binding protein decrease osteoclast formation and bone resorption in ovariectomized mice. J. Clin. Invest. 94(6), 2397–2406. 70. Manolagas, S. C., Kousteni, S., and Jilka, R. L. (2002). Sex steroids and bone. Recent Prog. Horm. Res. 57, 385–409. 71. Girasole, G., Jilka, R. L., Passeri, G., Boswell, S., Boder, G., Williams, D. C., and Manolagas, S. C. (1992). 17β-estradiol inhibits interleukin-6 production by bone marrow-derived stromal cells and osteoblasts in vitro: a potential mechanism for the antiosteoporotic effect of estrogens. J. Clin. Invest. 89(3), 883–891. 72. Jilka, R. L., Hangoc, G., Girasole, G., Passeri, G., Williams, D. C., Abrams, J. S., Boyce, B., Broxmeyer, H., and Manolagas, S. C. (1992). Increased osteoclast development after estrogen loss: mediation by interleukin-6. Science 257(5066), 88–91. 73. Kassem, M., Harris, S. A., Spelsberg, T. C., and Riggs, B. L. (1996). Estrogen inhibits interleukin-6 production and gene expression in a human osteoblastic cell line with high levels of estrogen receptors. J. Bone Miner. Res. 11(2), 193–199. 74. Poli, V., Balena, R., Fattori, E., Markatos, A., Yamamoto, M., Tanaka, H., Ciliberto, G., Rodan, G. A., and Costantini, F. (1994). Interleukin-6 deficient mice are protected from bone loss caused by estrogen depletion. Embo. J. 13(5), 1189–1196. 75. Ivey, J. L., and Baylink, D. J. (1981). Postmenopausal osteoporosis: proposed roles of defective coupling and estrogen deficiency. Metab. Bone Dis. Relat. Res. 3(1), 3–7. 76. Ernst, M., Heath, J. K., and Rodan, G. A. (1989). Estradiol effects on proliferation, messenger ribonucleic acid for collagen and insulin-like growth factor-I, and parathyroid hormone-stimulated adenylate cyclase activity in osteoblastic cells from calvariae and long bones. Endocrinology 125(2), 825–833. 77. Oursler, M. J., Cortese, C., Keeting, P., Anderson, M. A., Bonde, S. K., Riggs, B. L., and Spelsberg, T. C. (1991). Modulation of transforming growth factor-beta production in normal human osteoblast-like cells by 17β-estradiol and parathyroid hormone. Endocrinology 129(6), 3313–3320. 78. Brincat, M., Moniz, C. F., Kabalan, S., Versi, E., O’Dowd, T., Magos, A. L., Montgomery, J., and Studd, J. W. (1987). Decline in skin collagen content and metacarpal index after the menopause and its prevention with sex hormone replacement. Br. J. Obstet. Gynaecol. 94(2), 126–129. 79. Kawaguchi, H., Pilbeam, C. C., Vargas, S. J., Morse, E. E., Lorenzo, J. A., and Raisz, L. G. (1995). Ovariectomy enhances and estrogen replacement inhibits the activity of bone marrow factors that stimulate prostaglandin production in cultured mouse calvariae. J. Clin. Invest. 96(1), 539–548. 80. Westerlind, K. C., Wakley, G. K., Evans, G. L., and Turner, R. T. (1993). Estrogen does not increase bone formation in growing rats. Endocrinology 133(6), 2924–2934. 81. Gray, C., Colston, K. W., Mackay, A. G., Taylor, M. L., and Arnett, T. R. (1992). Interaction of androgen and 1,25-dihydroxyvitamin D3: effects on normal rat bone cells. J. Bone Miner. Res. 7(1), 41–46. 82. Gray, T. K., Mohan, S., Linkhart, T. A., and Baylink, D. J. (1989). Estradiol stimulates in vitro the secretion of insulin-like growth factors by the clonal osteoblastic cell line, UMR106. Biochem. Biophys. Res. Commun. 158(2), 407–412. 83. Monroe, D. G., Getz, B. J., Johnsen, S. A., Riggs, B. L., Khosla, S., and Spelsberg, T. C. (2003). Estrogen receptor isoform-specific regulation of endogenous gene expression in human osteoblastic cell lines expressing either ERα or ERβ. J. Cell. Biochem. 90(2), 315–326. 84. Kassem, M., Okazaki, R., De Leon, D., Harris, S. A., Robinson, J. A., Spelsberg, T. C., Conover, C. A., and Riggs, B. L. (1996). Potential mechanism of estrogen-mediated decrease in bone formation: estrogen increases production of inhibitory insulin-like growth factor-binding protein-4. Proc. Assoc. Am. Physicians 108(2), 155–164.
DAVID G. MONROE ET AL. 85. Robinson, J. A., Harris, S. A., Riggs, B. L., and Spelsberg, T. C. (1997). Estrogen regulation of human osteoblastic cell proliferation and differentiation. Endocrinology 138(7), 2919–2927. 86. Waters, K. M., Rickard, D. J., Riggs, B. L., Khosla, S., Katzenellenbogen, J. A., Katzenellenbogen, B. S., Moore, J., and Spelsberg, T. C. (2001). Estrogen regulation of human osteoblast function is determined by the stage of differentiation and the estrogen receptor isoform. J. Cell. Biochem. 83(3), 448–462. 87. Cao, L., Bu, R., Oakley, J. I., Kalla, S. E., and Blair, H. C. (2003). Estrogen receptor-β modulates synthesis of bone matrix proteins in human osteoblast-like MG63 cells. J. Cell. Biochem. 89(1), 152–164. 88. Kian Tee, M., Rogatsky, I., Tzagarakis-Foster, C., Cvoro, A., An, J., Christy, R. J., Yamamoto, K. R., and Leitman, D. C. (2004). Estradiol and selective estrogen receptor modulators differentially regulate target genes with estrogen receptors alpha and beta. Mol. Biol. Cell 15(3), 1262–1272. 89. Stossi, F., Barnett, D. H., Frasor, J., Komm, B., Lyttle, C. R., and Katzenellenbogen, B. S. (2004). Transcriptional profiling of estrogen-regulated gene expression via estrogen receptor (alpha) or estrogen receptor (beta) in human osteosarcoma cells: Distinct and common target genes for these receptors. Endocrinology 145(7), 3473–3486. 90. Tomkinson, A., Reeve, J., Shaw, R. W., and Noble, B. S. (1997). The death of osteocytes via apoptosis accompanies estrogen withdrawal in human bone. J. Clin. Endocrinol. Metab. 82(9), 3128–3135. 91. Pitsillides, A. A., Rawlinson, S. C., Suswillo, R. F., Bourrin, S., Zaman, G., and Lanyon, L. E. (1995). Mechanical strain-induced NO production by bone cells: a possible role in adaptive bone (re)modeling? Faseb. J. 9(15), 1614–1622. 92. Bonjour, J. P., Theintz, G., Buchs, B., Slosman, D., and Rizzoli, R. (1991). Critical years and stages of puberty for spinal and femoral bone mass accumulation during adolescence. J. Clin. Endocrinol. Metab. 73(3), 555–563. 93. Gustafsson, P. O., Kasstrom, H., Lindberg, L., and Olsson, S. E. (1975). Growth and mitotic rate of the proximal tibial epiphyseal plate in hypophysectomized rats given estradiol and human growth hormone. Acta Radiol. Suppl. 344, 69–74. 94. Strickland, A. L., and Sprinz, H. (1973). Studies of the influence of estradiol and growth hormone on the hypophysectomized immature rat epiphyseal cartilage growth plate. Am. J. Obstet. Gynecol. 115(4), 471–477. 95. Ho, K. K., and Weissberger, A. J. (1992). Impact of short-term estrogen administration on growth hormone secretion and action: distinct route-dependent effects on connective and bone tissue metabolism. J. Bone Miner. Res. 7(7), 821–827. 96. Delmas, P. D., Bjarnason, N. H., Mitlak, B. H., Ravoux, A. C., Shah, A. S., Huster, W. J., Draper, M., and Christiansen, C. (1997). Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. N. Engl. J. Med. 337(23), 1641–1647. 97. Brzozowski, A. M., Pike, A. C., Dauter, Z., Hubbard, R. E., Bonn, T., Engstrom, O., Ohman, L., Greene, G. L., Gustafsson, J. A., and Carlquist, M. (1997). Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389(6652), 753–758. 98. Tamrazi, A., Carlson, K. E., and Katzenellenbogen, J. A. (2003). Molecular sensors of estrogen receptor conformations and dynamics. Mol. Endocrinol. 17(12), 2593–2602. 99. Heldring, N., Nilsson, M., Buehrer, B., Treuter, E., and Gustafsson, J. A. (2004). Identification of tamoxifen-induced coregulator interaction surfaces within the ligand-binding domain of estrogen receptors. Mol. Cell. Biol. 24(8), 3445–3459. 100. Ray, A., Prefontaine, K. E., and Ray, P. (1994). Down-regulation of the interleukin-6 promoter by estrogen is mediated by NF-κB and C/EBP. J. Biol. Chem. 269, 12940–12946.
Chapter 19 Sex Steroid Effects on Bone Metabolism 101. Stein, B., and Yang, M. X. (1995). Repression of the interleukin-6 promoter by estrogen receptor is mediated by NF-kappa B and C/EBP beta. Mol. Cell. Biol. 15(9), 4971–4979. 102. Yang, N. N., Venugopalan, M., Hardikar, S., and Glasebrook, A. (1996). Identification of an estrogen response element activated by metabolites of 17β-estradiol and raloxifene. Science 273(5279), 1222–1225. 103. Yang, N. N., Bryant, H. U., Hardikar, S., Sato, M., Galvin, R. J., Glasebrook, A. L., and Termine, J. D. (1996). Estrogen and raloxifene stimulate transforming growth factor-β3 gene expression in rat bone: a potential mechanism for estrogen- or raloxifene-mediated bone maintenance. Endocrinology 137(5), 2075–2084. 104. Paech, K., Webb, P., Kuiper, G. G., Nilsson, S., Gustafsson, J., Kushner, P. J., and Scanlan, T. S. (1997). Differential ligand activation of estrogen receptors ERα and ERβ at AP1 sites. Science 277(5331), 1508–1510. 105. Kuiper, G. G., and Gustafsson, J. A. (1997). The novel estrogen receptor-beta subtype: potential role in the cell- and promoter-specific actions of estrogens and anti-estrogens. FEBS Lett. 410(1), 87–90. 106. Kuiper, G. G., Carlsson, B., Grandien, K., Enmark, E., Haggblad, J., Nilsson, S., and Gustafsson, J. A. (1997). Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 138(3), 863–870. 107. MacNamara, P., O’Shaughnessy, C., Manduca, P., and Loughrey, H. C. (1995). Progesterone receptors are expressed in human osteoblast-like cell lines and in primary human osteoblast cultures. Calcif. Tissue Int. 57(6), 436–441. 108. Wei, L. L., Leach, M. W., Miner, R. S., and Demers, L. M. (1993). Evidence for progesterone receptors in human osteoblast-like cells. Biochem. Biophys. Res. Commun. 195(2), 525–532. 109. Flototto, T., Niederacher, D., Hohmann, D., Heimerzheim, T., Dall, P., Djahansouzi, S., Bender, H. G., and Hanstein, B. (2004). Molecular mechanism of estrogen receptor (ER)alpha-specific, estradioldependent expression of the progesterone receptor (PR) B-isoform. J. Steroid Biochem. Mol. Biol. 88(2), 131–142. 110. Rickard, D. J., Waters, K. M., Ruesink, T. J., Khosla, S., Katzenellenbogen, J. A., Katzenellenbogen, B. S., Riggs, B. L., and Spelsberg, T. C. (2002). Estrogen receptor isoform-specific induction of progesterone receptors in human osteoblasts. J. Bone Miner. Res. 17(4), 580–592. 111. Tremollieres, F. A., Strong, D. D., Baylink, D. J., and Mohan, S. (1992). Progesterone and promegestone stimulate human bone cell proliferation and insulin-like growth factor-2 production. Acta Endocrinol. (Copenh.) 126(4), 329–337. 112. Azizi, G., Hansen, A., and Prestwood, K. M. (2003). Effect of micronized progesterone on bone turnover in postmenopausal women on estrogen replacement therapy. Endocr. Res. 29(2), 133–140. 113. Luo, X. H., Liao, E. Y., and Su, X. (2002). Progesterone upregulates TGF-β isoforms (β1, β2, and β3) expression in normal human osteoblast-like cells. Calcif. Tissue Int. 71(4), 329–334. 114. Prior, J. C., Vigna, Y. M., Wark, J. D., Eyre, D. R., Lentle, B. C., Li, D. K., Ebeling, P. R., and Atley, L. (1997). Premenopausal ovariectomy-related bone loss: a randomized, double-blind, one-year trial of conjugated estrogen or medroxyprogesterone acetate. J. Bone Miner. Res. 12(11), 1851–1863. 115. Colvard, D. S., Eriksen, E. F., Keeting, P. E., Wilson, E. M., Lubahn, D. B., French, F. S., Riggs, B. L., and Spelsberg, T. C. (1989). Identification of androgen receptors in normal human osteoblast-like cells. Proc. Natl Acad. Sci. USA 86(3), 854–857. 116. Sanchez-Visconti, G., Herrero, L., Rabadan, M., Pereira, I., and Ruiz-Torres, A. (1995). Ageing and prostate: age-related changes in androgen receptors of epithelial cells from benign hypertrophic glands compared with cancer. Mech. Ageing Dev. 82(1), 19–29. 117. Schweikert, H. U., Rulf, W., Niederle, N., Schafer, H. E., Keck, E., and Kruck, F. (1980). Testosterone metabolism in human bone. Acta Endocrinol. (Copenh.) 95(2), 258–264.
341 118. Nakano, Y., Morimoto, I., Ishida, O., Fujihira, T., Mizokami, A., Tanimoto, A., Yanagihara, N., Izumi, F., and Eto, S. (1994). The receptor, metabolism and effects of androgen in osteoblastic MC3T3-E1 cells. Bone Miner. 26(3), 245–259. 119. Stepan, J. J., Lachman, M., Zverina, J., Pacovsky, V., and Baylink, D. J. (1989). Castrated men exhibit bone loss: effect of calcitonin treatment on biochemical indices of bone remodeling. J. Clin. Endocrinol. Metab. 69(3), 523–527. 120. Tenover, J. S. (1992). Effects of testosterone supplementation in the aging male. J. Clin. Endocrinol. Metab. 75(4), 1092–1098. 121. Anderson, F. H., Francis, R. M., Peaston, R. T., and Wastell, H. J. (1997). Androgen supplementation in eugonadal men with osteoporosis: effects of six months’ treatment on markers of bone formation and resorption. J. Bone Miner. Res. 12(3), 472–478. 122. Riggs, B. L., Jowsey, J., Goldsmith, R. S., Kelly, P. J., Hoffman, D. L., and Arnaud, C. D. (1972). Short- and long-term effects of estrogen and synthetic anabolic hormone in postmenopausal osteoporosis. J. Clin. Invest. 51(7), 1659–1663. 123. Pederson, L., Kremer, M., Judd, J., Pascoe, D., Spelsberg, T. C., Riggs, B. L., and Oursler, M. J. (1999). Androgens regulate bone resorption activity of isolated osteoclasts in vitro. Proc. Natl Acad. Sci. USA 96(2), 505–510. 124. Pilbeam, C. C., and Raisz, L. G. (1990). Effects of androgens on parathyroid hormone and interleukin-1-stimulated prostaglandin production in cultured neonatal mouse calvariae. J. Bone Miner. Res. 5(11), 1183–1188. 125. Bellido, T., Jilka, R. L., Boyce, B. F., Girasole, G., Broxmeyer, H., Dalrymple, S. A., Murray, R., and Manolagas, S. C. (1995). Regulation of interleukin-6, osteoclastogenesis, and bone mass by androgens. The role of the androgen receptor. J. Clin. Invest. 95(6), 2886–2895. 126. Hofbauer, L. C., Ten, R. M., and Khosla, S. (1999). The anti-androgen hydroxyflutamide and androgens inhibit interleukin-6 production by an androgen-responsive human osteoblastic cell line. J. Bone Miner. Res. 14(8), 1330–1337. 127. Raisz, L. G., Wiita, B., Artis, A., Bowen, A., Schwartz, S., Trahiotis, M., Shoukri, K., and Smith, J. (1996). Comparison of the effects of estrogen alone and estrogen plus androgen on biochemical markers of bone formation and resorption in postmenopausal women. J. Clin. Endocrinol. Metab. 81(1), 37–43. 128. Baran, D. T., Bergfeld, M. A., Teitelbaum, S. L., and Avioli, L. V. (1978). Effect of testosterone therapy on bone formation in an osteoporotic hypogonadal male. Calcif. Tissue Res. 26(2), 103–106. 129. Cantatore, F. P., Carrozzo, M., Magli, D., D’Amore, M., and Pipitone, V. (1988). Serum osteocalcin levels in normal humans of different sex and age. An epidemiological study. Panminerva Med. 30(1), 23–25. 130. Morley, J. E., Perry, H. M., III, Kaiser, F. E., Kraenzle, D., Jensen, J., Houston, K., Mattammal, M., and Perry, H. M., Jr. (1993). Effects of testosterone replacement therapy in old hypogonadal males: a preliminary study. J. Am. Geriatr. Soc. 41(2), 149–152. 131. Kasperk, C. H., Wergedal, J. E., Farley, J. R., Linkhart, T. A., Turner, R. T., and Baylink, D. J. (1989). Androgens directly stimulate proliferation of bone cells in vitro. Endocrinology 124(3), 1576–1578. 132. Benz, D. J., Haussler, M. R., Thomas, M. A., Speelman, B., and Komm, B. S. (1991). High-affinity androgen binding and androgenic regulation of alpha 1(I)-procollagen and transforming growth factor-beta steady state messenger ribonucleic acid levels in human osteoblastlike osteosarcoma cells. Endocrinology 128(6), 2723–2730. 133. Hofbauer, L. C., Hicok, K. C., and Khosla, S. (1998). Effects of gonadal and adrenal androgens in a novel androgen-responsive human osteoblastic cell line. J. Cell. Biochem. 71(1), 96–108. 134. Kasperk, C. H., Faehling, K., Borcsok, I., and Ziegler, R. (1996). Effects of androgens on subpopulations of the human osteosarcoma cell line SaOS2. Calcif. Tissue Int. 58(5), 376–382.
342 135. Kasperk, C. H., Wakley, G. K., Hierl, T., and Ziegler, R. (1997). Gonadal and adrenal androgens are potent regulators of human bone cell metabolism in vitro. J. Bone Miner. Res. 12(3), 464–471. 136. Kasperk, C., Fitzsimmons, R., Strong, D., Mohan, S., Jennings, J., Wergedal, J., and Baylink, D. (1990). Studies of the mechanism by which androgens enhance mitogenesis and differentiation in bone cells. J. Clin. Endocrinol. Metab. 71(5), 1322–1329. 137. Gori, F., Hofbauer, L., Conover, C. A., and Khosla, S. (1999). Effects of androgens on the insulin-like growth factor system in an androgen-responsive human osteoblastic cell line. Endocrinology 140, 5579–5586. 138. Durbridge, T. C., Morris, H. A., Parsons, A. M., Parkinson, I. H., Moore, R. J., Porter, S., Need, A. G., Nordin, B. E., and Vernon-Roberts, B. (1990). Progressive cancellous bone loss in rats after adrenalectomy and oophorectomy. Calcif. Tissue Int. 47(6), 383–387. 139. Turner, R. T., Lifrak, E. T., Beckner, M., Wakley, G. K., Hannon, K. S., and Parker, L. N. (1990). Dehydroepiandrosterone reduces cancellous bone osteopenia in ovariectomized rats. Am. J. Physiol. 258(4 Pt 1), E673–677. 140. Labrie, F., Belanger, A., Simard, J., Van, L.-T., and Labrie, C. (1995). DHEA and peripheral androgen and estrogen formation: intracinology. Ann. N. Y. Acad. Sci. 774, 16–28. 141. Meikle, A. W., Dorchuck, R. W., Araneo, B. A., Stringham, J. D., Evans, T. G., Spruance, S. L., and Daynes, R. A. (1992). The presence of a dehydroepiandrosterone-specific receptor binding complex in murine T cells. J. Steroid Biochem. Mol. Biol. 42(3–4), 293–304. 142. Lu, P. W., Briody, J. N., Ogle, G. D., Morley, K., Humphries, I. R., Allen, J., Howman-Giles, R., Sillence, D., and Cowell, C. T. (1994). Bone mineral density of total body, spine, and femoral neck in children and young adults: a cross-sectional and longitudinal study. J. Bone Miner. Res. 9(9), 1451–1458. 143. Riis, B. J., Krabbe, S., Christiansen, C., Catherwood, B. D., and Deftos, L. J. (1985). Bone turnover in male puberty: a longitudinal study. Calcif. Tissue Int. 37(3), 213–217. 144. Parker, L. N. (1991). Adrenarche. Endocrinol. Metab. Clin. North Am. 20(1), 71–83. 145. Slemenda, C. W., Reister, T. K., Hui, S. L., Miller, J. Z., Christian, J. C., and Johnston, C. C., Jr. (1994). Influences on skeletal mineralization in children and adolescents: evidence for varying effects of sexual maturation and physical activity. J. Pediatr. 125(2), 201–207. 146. Fehily, A. M., Coles, R. J., Evans, W. D., and Elwood, P. C. (1992). Factors affecting bone density in young adults. Am. J. Clin. Nutr. 56(3), 579–586. 147. Mauras, N., Haymond, M. W., Darmaun, D., Vieira, N. E., Abrams, S. A., and Yergey, A. L. (1994). Calcium and protein kinetics in prepubertal boys. Positive effects of testosterone. J. Clin. Invest. 93(3), 1014–1019. 148. Hobbs, C. J., Plymate, S. R., Rosen, C. J., and Adler, R. A. (1993). Testosterone administration increases insulin-like growth factor-I levels in normal men. J. Clin. Endocrinol. Metab. 77(3), 776–779. 149. Smith, E. P., Boyd, J., Frank, G. R., Takahashi, H., Cohen, R. M., Specker, B., Williams, T. C., Lubahn, D. B., and Korach, K. S. (1994). Estrogen resistance caused by a mutation in the estrogen-receptor gene in a man. N. Engl. J. Med. 331(16), 1056–1061. 150. Morishima, A., Grumbach, M. M., Simpson, E. R., Fisher, C., and Qin, K. (1995). Aromatase deficiency in male and female siblings caused by a novel mutation and the physiological role of estrogens. J. Clin. Endocrinol. Metab. 80(12), 3689–3698. 151. Carani, C., Qin, K., Simoni, M., Faustini-Fustini, M., Serpente, S., Boyd, J., Korach, K. S., and Simpson, E. R. (1997). Effect of testosterone and estradiol in a man with aromatase deficiency. N. Engl. J. Med. 337(2), 91–95. 152. Morishima, A., Grumbach, M. M., and Bilezikian, J. P. (1997). Estrogen markedly increases bone mass in an estrogen deficient young man with aromatase deficiency. J. Bone Miner. Res. 12(Suppl 1), S126.
DAVID G. MONROE ET AL. 153. Center, J. R., Nguyen, T. V., White, C. P., and Eisman, J. A. (1997). Male osteoporosis predictors: Sex hormones and calcitropic hormones. J. Bone Miner. Res. 12(Suppl 1), S368. 154. Greendale, G. A., Edelstein, S., and Barrett-Connor, E. (1997). Endogenous sex steroids and bone mineral density in older women and men: the Rancho Bernardo Study. J. Bone Miner. Res. 12(11), 1833–1843. 155. Khosla, S., Melton, L. J., 3rd, Atkinson, E. J., O’Fallon, W. M., Klee, G. G., and Riggs, B. L. (1998). Relationship of serum sex steroid levels and bone turnover markers with bone mineral density in men and women: a key role for bioavailable estrogen. J. Clin. Endocrinol. Metab. 83(7), 2266–2274. 156. Slemenda, C. W., Longcope, C., Zhou, L., Hui, S. L., Peacock, M., and Johnston, C. C. (1997). Sex steroids and bone mass in older men. Positive associations with serum estrogens and negative associations with androgens. J. Clin. Invest. 100(7), 1755–1759. 157. Amin, S., Zhang, Y., Sawin, C. T., Evans, S. R., Hannan, M. T., Kiel, D. P., Wilson, P. W., and Felson, D. T. (2000). Association of hypogonadism and estradiol levels with bone mineral density in elderly men from the Framingham study. Ann. Intern. Med. 133(12), 951–963. 158. Khosla, S., Melton, L. J., III, Atkinson, E. J., and O’Fallon, W. M. (2001). Relationship of serum sex steroid levels to longitudinal changes in bone density in young versus elderly men. J. Clin. Endocrinol. Metab. 86(8), 3555–3561. 159. Gennari, L., Merlotti, D., Martini, G., Gonnelli, S., Franci, B., Campagna, S., Lucani, B., Dal Canto, N., Valenti, R., Gennari, C., and Nuti, R. (2003). Longitudinal association between sex hormone levels, bone loss, and bone turnover in elderly men. J. Clin. Endocrinol. Metab. 88(11), 5327–5333. 160. Van Pottelbergh, I., Goemaere, S., and Kaufman, J. M. (2003). Bioavailable estradiol and an aromatase gene polymorphism are determinants of bone mineral density changes in men over 70 years of age. J. Clin. Endocrinol. Metab. 88(7), 3075–3081. 161. Falahati-Nini, A., Riggs, B. L., Atkinson, E. J., O’Fallon, W. M., Eastell, R., and Khosla, S. (2000). Relative contributions of testosterone and estrogen in regulating bone resorption and formation in normal elderly men. J. Clin. Invest. 106(12), 1553–1560. 162. Leder, B. Z., LeBlanc, K. M., Schoenfeld, D. A., Eastell, R., and Finkelstein, J. S. (2003). Differential effects of androgens and estrogens on bone turnover in normal men. J. Clin. Endocrinol. Metab. 88(1), 204–210. 163. Taxel, P., Kennedy, D. G., Fall, P. M., Willard, A. K., Clive, J. M., and Raisz, L. G. (2001). The effect of aromatase inhibition on sex steroids, gonadotropins, and markers of bone turnover in older men. J. Clin. Endocrinol. Metab. 86(6), 2869–2874. 164. Slemenda, C., Hui, S. L., Longcope, C., and Johnston, C. C. (1987). Sex steroids and bone mass. A study of changes about the time of menopause. J. Clin. Invest. 80(5), 1261–1269. 165. Buchanan, J. R., Hospodar, P., Myers, C., Leuenberger, P., and Demers, L. M. (1988). Effect of excess endogenous androgens on bone density in young women. J. Clin. Endocrinol. Metab. 67(5), 937–943. 166. Gallagher, J. C., Riggs, B. L., and DeLuca, H. F. (1980). Effect of estrogen on calcium absorption and serum vitamin D metabolites in postmenopausal osteoporosis. J. Clin. Endocrinol. Metab. 51(6), 1359–1364. 167. Gennari, C., Agnusdei, D., Nardi, P., and Civitelli, R. (1990). Estrogen preserves a normal intestinal responsiveness to 1,25dihydroxyvitamin D3 in oophorectomized women. J. Clin. Endocrinol. Metab. 71(5), 1288–1293. 168. Thomas, M. L., Xu, X., Norfleet, A. M., and Watson, C. S. (1993). The presence of functional estrogen receptors in intestinal epithelial cells. Endocrinology 132(1), 426–430.
Chapter 19 Sex Steroid Effects on Bone Metabolism 169. Hagenfeldt, Y., and Eriksson, H. A. (1988). The estrogen receptor in the rat kidney. Ontogeny, properties and effects of gonadectomy on its concentration. J. Steroid Biochem. 31(1), 49–56. 170. Forte, L. R., Langeluttig, S. G., Biellier, H. V., Poelling, R. E., Magliola, L., and Thomas, M. L. (1983). Upregulation of kidney adenylate cyclase in the egg-laying hen: role of estrogen. Am. J. Physiol. 245(3), E273–280. 171. Stock, J. L., Coderre, J. A., Burke, E. M., Danner, D. B., Chipman, S. D., and Shapiro, J. R. (1992). Identification of estrogen receptor mRNA and the estrogen modulation of parathyroid hormone-stimulated cyclic AMP accumulation in opossum kidney cells. J. Cell. Physiol. 150(3), 517–525. 172. Nordin, B. E., Need, A. G., Morris, H. A., Horowitz, M., and Robertson, W. G. (1991). Evidence for a renal calcium leak in postmenopausal women. J. Clin. Endocrinol. Metab. 72(2), 401–407. 173. McKane, W. R., Khosla, S., Burritt, M. F., Kao, P. C., Wilson, D. M., Ory, S. J., and Riggs, B. L. (1995). Mechanism of renal calcium conservation with estrogen replacement therapy in women in early postmenopause – a clinical research center study. J. Clin. Endocrinol. Metab. 80(12), 3458–3464. 174. Naveh-Many, T., Almogi, G., Livni, N., and Silver, J. (1992). Estrogen receptors and biologic response in rat parathyroid tissue and C cells. J. Clin. Invest. 90(6), 2434–2438. 175. Prince, R. L., MacLaughlin, D. T., Gaz, R. D., and Neer, R. M. (1991). Lack of evidence for estrogen receptors in human and bovine parathyroid tissue. J. Clin. Endocrinol. Metab. 72(6), 1226–1228. 176. Cosman, F., Nieves, J., Horton, J., Shen, V., and Lindsay, R. (1994). Effects of estrogen on response to edetic acid infusion in postmenopausal osteoporotic women. J. Clin. Endocrinol. Metab. 78(4), 939–943. 177. Vincent, A., Riggs, B. L., Atkinson, E. J., Oberg, A. L., and Khosla, S. (2003). Effect of estrogen replacement therapy on parathyroid hormone secretion in elderly postmenopausal women. Menopause 10, 165–171.
343 178. Duarte, B., Hargis, G. K., and Kukreja, S. C. (1988). Effects of estradiol and progesterone on parathyroid hormone secretion from human parathyroid tissue. J. Clin. Endocrinol. Metab. 66(3), 584–587. 179. Hope, W. G., Ibarra, M. J., and Thomas, M. L. (1992). Testosterone alters duodenal calcium transport and longitudinal bone growth rate in parallel in the male rat. Proc. Soc. Exp. Biol. Med. 200(4), 536–541. 180. Lafferty, F. W., Spencer, G. E., Jr., and Pearson, O. H. (1964). Effects of androgens, estrogens and high calcium intakes on bone formation and resorption in osteoporosis. Am. J. Med. 36, 514–528. 181. Need, A. G., Horowitz, M., Morris, H. A., Walker, C. J., and Nordin, B. E. (1987). Effects of nandrolone therapy on forearm bone mineral content in osteoporosis. Clin. Orthop. 225, 273–278. 182. Stefani, S., Aguiari, G. L., Bozza, A., Maestri, I., Magri, E., Cavazzini, P., Piva, R., and del Senno, L. (1994). Androgen responsiveness and androgen receptor gene expression in human kidney cells in continuous culture. Biochem. Mol. Biol. Int. 32(4), 597–604. 183. Goldstone, A. D., Koenig, H., and Lu, C. Y. (1983). Androgenic stimulation of endocytosis, amino acid and hexose transport in mouse kidney cortex involves increased calcium fluxes. Biochim. Biophys. Acta. 762(2), 366–371. 184. Reifenstein, E. C., and Albright, F. (1946). The metabolic effects of steroid hormones in osteoporosis. J. Clin. Invest. 26, 24–56. 185. Chesnut, C. H., III, Ivey, J. L., Gruber, H. E., Matthews, M., Nelp, W. B., Sisom, K., and Baylink, D. J. (1983). Stanozolol in postmenopausal osteoporosis: therapeutic efficacy and possible mechanisms of action. Metabolism 32(6), 571–580. 186. Mason, R. A., and Morris, H. A. (1997). Effects of dihydrotestosterone on bone biochemical markers in sham and oophorectomized rats. J. Bone Miner. Res. 12(9), 1431–1437. 187. Riggs, B. L., Khosla, S., and Melton, L. J., III (1998). A unitary model for involutional osteoporosis: estrogen deficiency causes both type I and type II osteoporosis in postmenopausal women and contributes to bone loss in aging men. J. Bone Miner. Res. 13(5), 763–773. 188. Manolagas, S. C., and Kousteni, S. (2001). Perspective: nonreproductive sites of action of reproductive hormones. Endocrinology 142(6), 2200–2204.
This Page Intentionally Left Blank
Chapter 20
Physiology of Calcium and Phosphate Homeostases René Rizzoli, M.D. and Jean-Philippe Bonjour, M.D.
VII. Calcium and Bone Growth VIII. Body Distribution of Phosphorus IX. Determinants of Extracellular Phosphate Concentration X. Homeostatic Responses to Changes in Phosphate Supply or Demand XI. Conclusions References
I. Abstract II. Introduction III. Body Distribution of Calcium IV. Determinants of Extracellular Calcium Concentration V. Relative Importance of the Various Calcium Fluxes in Controlling Extracellular Calcium Homeostasis VI. Homeostatic Responses to Hypocalcemia
I. ABSTRACT
tubular calcium and phosphate transports, or by releasing calcium from intracellular stores, calcium itself plays the role of an effector on homeostatic mechanisms.
Calcium and phosphate homeostases are controlled by bidirectional calcium and phosphate fluxes, occurring at the levels of intestine, bone, and kidney. The latter organ plays a central role in regulating the extracellular concentration of either ion. Sensitive and efficient regulatory mechanisms, involving extracellular calcium sensing, are triggered by changes in calcium demand or supply. Similarly, the renal handling of phosphate can adjust its capacity to meet the need for phosphate of the organism. Not only calciotropic peptides or steroid hormones are capable of modifying the different calcium and phosphate fluxes to various extents, but also a variety of local factors are implicated in the regulation of calcium and phosphate homeostasis, in order to protect the organism against a deficiency or an overload. Finally, by directly influencing renal Dynamics of Bone and Cartilage Metabolism
Division of Bone Diseases, WHO Collaborating Center for Osteoporosis Prevention, Department of Rehabilitation and Geriatrics, University Hospitals, CH - 1211 Geneva 14 (Switzerland)
II. INTRODUCTION Calcium and phosphate play prominent roles in the regulation of cell function. In addition, both are tightly connected to the process of bone mineralization, in which they participate in the formation of hydroxyapatite crystal deposited in specific regions with the collagen fibril networks. The hydroxyapatite crystal ensures the structural rigidity of the skeleton in its function of standing body support, and in the protection and housing of bone marrow. The regulation of calcium and phosphate homeostases is aimed at maintaining extracellular calcium and phosphate 345
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
346 concentration and balance as constant as possible, to protect the organism against deficiency or overload of these ions. Extracellular calcium concentrations are maintained remarkably stable, because of the high sensitivity of a variety of cell systems or organs, including the central nervous system, muscle, and exo-/endocrine glands, to small variations of extracellular calcium concentrations. In contrast, to fulfill the requirements of adequate mineral supply for osteoid tissue mineralization, the level of extracellular inorganic phosphate is adjusted to meet the demand of the organism. The homeostasis of both ions and their balance are controlled by a series of hormones and factors tightly interrelated in complex regulatory systems. The production of some of these agents is regulated by the concentration of the solute they are controlling, through negative feedback mechanisms, justifying thereby the term of calciotropic hormones as far as calcium homeostasis control is concerned.
III. BODY DISTRIBUTION OF CALCIUM Calcium is the fifth inorganic element in abundance in the body. For a 70-kg subject, the calcium mass represents about 1300 grams, 99% of which is in bone and teeth, mostly as hydroxyapatite [1]. A small portion of bone calcium (approximately 1%) is rapidly exchangeable with extracellular fluid, contributing to the regulation of the homeostasis of extracellular calcium concentration, by serving as buffer and storage [2]. During puberty, there is nearly a doubling of body mineral stores through an increase in the size of the skeleton, with minor changes in volumetric bone density, i.e. the amount of bone in bone [3, 4]. By the end of the second decade, most of the body mineral capital is accumulated, though a very few percents of bone consolidation have been suggested to occur during the third decade, particularly in males [5]. Approximately 1% of total body calcium is intracellular. At the intracellular level, calcium homeostasis is controlled by an influx following an electrical and chemical gradient, through selective calcium channels [6–8]. As compared with the 1 mmol/L extracellular calcium concentration, a 10 000-fold lower concentration in the cytosol is maintained by the constant extrusion of calcium, through a calcium– magnesium ATPase and sodium–calcium exchange mechanisms [9, 10]. Intracellular calcium homeostasis is also maintained by a dynamic equilibrium with intracellular mitochondrial and microsomal stores [11, 12]. The concentration of cytosolic-free calcium is around 0.1 µmol/L. This concentration is critical in controlling cell membrane permeability, a large variety of enzymatic reactions, endocrine and exocrine hormone secretions,
RENÉ RIZZOLI AND JEAN-PHILIPPE BONJOUR
and in regulating cardiac, skeletal and smooth muscles contraction (Table I). Calcium is also implicated in the control of cell replication and apoptosis [13, 14]. Regulation of intracellular enzymatic functions is achieved through the interaction with various calcium-binding proteins such as, for instance, calmodulin or troponin, or the actin–myosin system, for muscle contraction [2, 8]. About 0.1% of total body calcium is in the extracellular compartment. Extracellular calcium concentration plays a major role in the integrity and stability of cell membrane, in intercellular adhesion, in blood clotting, and it influences neuromuscular excitability. Plasma calcium is tightly regulated in a narrow range, particularly the ionized form, which amounts to approximately 50% of total plasma calcium, and which represents the physiologically active form, whereas 40% is bound to proteins, mostly albumin, and 10% is under the form of ultrafiltrable ion complexes. The binding equilibrium of calcium with albumin is determined by the pH. Indeed, acute acidosis is associated with a decreased binding and, thereby, a higher proportion of the ionized form. In opposite, acute alkalosis decreases ionized calcium by increasing calcium bound to serum albumin.
IV. DETERMINANTS OF EXTRACELLULAR CALCIUM CONCENTRATION Extracellular calcium concentration is maintained in a dynamic equilibrium through fluxes occurring at the level of the intestine, bone, and kidney (Fig. 1). At steady state, when body ions balance is zero, as in nongrowing individuals, the amounts of solutes entering the extracellular space Table I.
Physiological Roles of Calcium and Phosphate Calcium
Structural Hydroxyapatite (99% body constituent calcium) Exchangeable pool (mineral storage) Function
Intracellular signal transduction Cell adhesion Cell proliferation and differentiation Membrane permeability (neuromuscular excitability, muscle contraction, neurotransmission) Cytoskeleton (cell motility) Exo-/endosecretion Coagulation
Phosphate Hydroxyapatite (85% body phosphorus) Nucleic acids Carbohydrates Lipids Energy storage and delivery Intracellular signal transduction Enzyme activity Acid–base homeostasis
347
Chapter 20 Physiology of Calcium and Phosphate Homeostases
Figure 1
Main fluxes (mg/day) controlling calcium homeostasis.
are matched by the amounts leaving it. By controlling the calcium output, the kidney plays a central role in the maintenance of calcium homeostasis.
A. Intestinal Flux Net intestinal absorption of calcium represents the difference between the amounts of solutes absorbed and secreted into the gut lumen. In humans, under normal conditions, intestinal absorption of calcium constitutes approximately 20% of ingested calcium. Net intestinal absorption of calcium depends on dietary intakes, on the capacity of the intestinal wall to transport calcium, on the bioavailability of calcium present in the intestinal lumen, and on the secretory flux. Under normal conditions, the secretory flux does not appear to vary markedly. However, it could be increased in pathological conditions such as coeliac disease. The intestinal calcium absorptive capacity is mainly controlled by calcitriol, which stimulates the transport through both genomic and nongenomic mechanisms [15–19]. The interaction of calcitriol with its specific nuclear receptor triggers the synthesis of a variety of proteins, including its receptor itself, calbindins, and the calcium–magnesium ATPase pump, which is located in the enterocyte basolateral membrane. Duodenum possesses the highest concentration of calcitriol receptors, and is the site of the intestine most sensitive to the vitamin metabolite in terms of calcium absorptive capacity. However, the short length of this segment and the rapid transit time suggest that it may not play a quantitative major role in the overall net intestinal calcium absorption. Thus, jejunum and ileum could quantitatively absorb more, despite a less efficient
calcium transport capacity. Parathyroid hormone does not exert any direct effect on the intestinal cells [20, 21]. The importance of bioavailability of calcium at absorptive sites is illustrated by the impairment of calcium absorption induced by the formation of complexes with anions, such as phosphate, sulfate, phytate, or oxalate [22, 23]. For instance, the colonic mucosa is equipped with a powerful vitamin D-sensitive mechanism of calcium transport. However, the absorption is quantitatively little, since calcium is in the large intestine lumen under a form not available to the site of absorption [24]. At steady state, a 24-hour urinary excretion of calcium is mainly the reflection of daily net intestinal calcium absorption. The intestinal absorptive capacity can be evaluated by measuring calcium isotope absorption. A deconvolution analysis of calcium isotopes after oral and intravenous simultaneous administration of 2 different tracers, allows one to determine intestinal calcium transport. A simple evaluation is obtained by measuring the increase in urinary calcium after an oral calcium load [25, 26].
B. Bone Fluxes On average, about 1% of total bone calcium exchanges every month, through a mechanism involving bidirectional fluxes. The main regulators of these fluxes are parathyroid hormone and calcitriol, and possibly calcitonin as an inhibitor of osteoclastic bone resorption under certain conditions [18, 19, 27, 28]. A large variety of substances, either circulating or produced locally, or present in the bone matrix, such as prostaglandins, thyroid hormones, glucocorticoids, sex hormones, growth factors, or interleukins,
348
RENÉ RIZZOLI AND JEAN-PHILIPPE BONJOUR
components of the RANK-ligand/osteoprotegerin system, produced by the immune or hematopoietic systems, or bone cells, are capable of influencing bone remodeling, and thereby the bidirectional calcium fluxes [29–32]. In fasting urine, calcium excretion related to creatinine is a direct reflection of net bone [33–35]. Indeed, after an overnight fast, provided there is no calcium supplement taken in the previous evening and the patient does not have a postmicturitional urine residue, calcium appearing in the urine originates mostly from bone. The adjustment by creatinine corrects for urine concentration. Multiplying this ratio by serum creatinine provides urinary calcium excretion per glomerular filtration rate unit.
C. Soft Tissues A minute amount of body calcium is intracellular. Thus, any calcium shift from or into the intracellular compartment does not significantly influence extracellular calcium homeostasis, except maybe in response to parathyroid hormone. The transient decrease in plasma calcium observed in the minutes following parathyroid hormone injection has been attributed to a parathyroid hormonemediated acute transfer of calcium into cells [36, 37].
D. Renal Fluxes Approximately 75% of plasma calcium is ultrafiltrable. After filtration, more than 95% of calcium is reabsorbed. Thus, the amount of calcium appearing in the urine
Figure 2
represents the difference between the amounts filtered and reabsorbed. At steady state, this amount is mainly the reflection of net fluxes into the extracellular fluid of calcium originating from intestine and bone. The proximal tubule reabsorbs 60–70% of calcium, this reabsorption is tightly connected to that of sodium [38] (Fig. 2). Then, 20–30% of filtered calcium is reabsorbed along the loop ascending limb, and 10% at the level of the distal tubule. These two latter reabsorptive sites are influenced by parathyroid hormone. The renal tubular capacity to reabsorb calcium is the main determinant of extracellular calcium concentration. Any change of this capacity is able to induce variations of plasma calcium from 1.5–3.8 mmol/L [33, 34]. This concept has been established by the study of the relationship between urinary calcium excretion and plasma calcium in patients suffering from a lack or an excess of parathyroid hormone. Another example of the central control of extracellular calcium by the renal tubule is given by the syndrome of familial hypocalciuric hypercalcemia, the features of which include an increase in renal tubular reabsorption of calcium independent of parathyroid hormone [39–41]. Various situations or pharmacological agents can modulate the renal tubular reabsorption of calcium. Alkalosis stimulates renal tubular reabsorption of calcium, whereas acidosis decreases it. Thiazides and lithium salts increase the reabsorption of calcium, through mechanisms which are independent of parathyroid hormone [39, 40, 42]. Phosphate deficiency, pharmacological doses of calcitonin and loop diuretics are associated with an increase in calcium clearance [38]. Even large variations of the glomerular filtration rate do not cause major changes in calcemia, since the renal tubule can easily maintain the
Central role of the kidney in controlling calcium homeostasis.
349
Chapter 20 Physiology of Calcium and Phosphate Homeostases
calcium excretion by modulating its reabsorptive capacity. However, the capacity of the kidney to excrete calcium can be overwhelmed under certain circumstances, particularly in case of large calcium loads resulting from extensive bone destruction and/or very high net calcium intestinal absorption, for instance as a consequence of vitamin D excess, which can lead to an increase in extracellular calcium levels.
V. RELATIVE IMPORTANCE OF THE VARIOUS CALCIUM FLUXES IN CONTROLLING EXTRACELLULAR CALCIUM HOMEOSTASIS An important role in the regulation of calcium fluxes and balance between the various body compartments is played by the above-mentioned systemic hormones. Other hormones such as insulin, growth hormone, insulin-like growth factors, parathyroid hormone-related protein [28, 43–45], glucocorticoids, and sex hormones, as well as locally produced and acting interleukins [46], transforming growth factors and colony-stimulating factors, which are capable of directly influencing calcium metabolism, could also modify the target cell sensitivity to parathyroid hormone and/or calcitriol [29–31, 47–49]. However, their secretion does not appear to be directly controlled by variations in extracellular calcium and/or calcium demand. Any disturbance of the above-described fluxes can result in an alteration of extracellular calcium homeostasis. For instance, an excess of vitamin D is associated with an increase in intestinal absorption of calcium and of bone resorption, and leads to hypercalcemia when the renal excretion capacity is overwhelmed [50, 51]. On the other hand, when renal tubular reabsorption of calcium is stimulated, such as by parathyroid hormone or by parathyroid hormone-related protein, plasma calcium levels can rise, despite very minute changes in calcium influx into the extracellular fluid compartment [33, 34]. Then, despite an increase in renal tubular reabsorption of calcium, urinary excretion is elevated as a consequence of a higher filtered load. The relative and quantitative contribution of calcium mobilization from bone, and of renal tubular reabsorption of calcium, to hypercalcemia induced by parathyroid hormone-related protein, can be estimated in studying the model of thyroparathyroidectomized rats chronically infused with parathyroid hormone-related protein [52]. Thyroparathyroidectomy prevents the contraregulation to variations in extracellular calcium concentrations by endogenous hormones. The elevation of plasma calcium is determined by both increased bone resorption and
enhanced renal tubular reabsorption of calcium. However, the complete inhibition of bone resorption by a bisphosphonate, at a dose which fully normalized fasting urinary calcium excretion, taken as a reflection of net bone resorption, is associated with an approximately 30% decrease, but not a correction of plasma calcium. Thus, the residual hypercalcemia can be attributed to a renal tubular reabsorption effect, which accounted for more than two-thirds of the elevated plasma calcium (Fig. 3A). Indeed, it is well established that bisphosphonates are devoid of any direct effect on the renal handling of calcium [53]. To influence the renal tubular reabsorption, the administration of an agent (the free radical scavenger WR-2721), known to impair the tubular reabsorption of calcium through a parathyroid hormone-independent mechanism [54, 55], is able to acutely increase calcium excretion and to further reduce plasma calcium (Fig. 3B). The predominance of stimulated renal tubular reabsorption of calcium or of increased bone resorption in determining an altered extracellular calcium homeostasis can be demonstrated in a variety of clinical disorders associated with hypercalcemia [50] (Table II, Fig. 4).
VI. HOMEOSTATIC RESPONSES TO HYPOCALCEMIA To maintain the extracellular calcium concentration as constant as possible, the response to acute variations implies changes in the various fluxes, without a necessarily major alteration in total body calcium stores. In contrast, when the organism is chronically submitted to calcium deficiency, its capacity to absorb calcium from the gut or to retain it in the kidney cannot match the need; then, to maintain extracellular calcium concentration, the skeletal mineral is mobilized, leading to a progressive decrease in bone mineral mass [1]. This mechanism might be implicated in the pathogenesis of senile osteoporosis. To recognize changes in extracellular calcium concentration, parathyroid cells have a sensitive calcium sensor, capable of transmitting the information to the parathyroid hormonesynthesizing and -releasing machinery.
A. Extracellular Calcium-sensing Receptor A central role in the regulation of calcium homeostasis is played by parathyroid hormone produced by the parathyroid glands, which recognize alterations in plasma calcium [2, 28, 56]. Any change in plasma calcium is detected by a cell membrane-associated calcium sensor/receptor, which can also be activated by other divalent cations [2, 41, 57].
350
RENÉ RIZZOLI AND JEAN-PHILIPPE BONJOUR
Figure 3 Relative contribution of net bone resorption and renal tubular reabsorption to parathyroid hormone-related protein-induced hypercalcemia in thyroparathyroidectomized rats. (A) Pair-fed animals were chronically infused by subcutaneous osmotic minipumps with synthetic parathyroid hormonerelated protein (1–34) during 7 days. The bisphosphonate pamidronate was given subcutaneously at a dose which normalized bone resorption, as reflected by the correction of fasting urinary calcium excretion (adapted from Rizzoli et al., J Bone Miner Res 4: 759–765, 1989). (B) The free radical scavenger WR-2721 was acutely administered to parathyroid hormone-related proteininfused and pamidronate-treated rats. The arrows connect the points before and after WR-2721 acute administration. WR-2721 acutely increased the clearance of calcium.
This 1078-amino acid polypeptide belongs to the receptor family of 7 transmembrane domains cell membraneassociated and guanine nucleotide-binding protein-coupled. This structure is present in parathyroid cells, in C cells of the thyroid, in renal epithelial cells, and in the central nervous system. A raise in extracellular calcium concentration stimulates protein kinase C and leads to an increase in
cytosolic free calcium, thereby inhibiting parathyroid hormone synthesis and release. Conversely, a decrease in plasma calcium triggers the exocytosis of PTH within seconds, whereas it takes hours to increase PTH synthesis, and days to stimulate parathyroid cell proliferation [2, 28, 56]. Various mutations have been reported, which account for hyper- or hyposecretion of parathyroid hormone
Chapter 20 Physiology of Calcium and Phosphate Homeostases Table II.
Hypercalcemic Disorders Increased bone resorption
Endocrine disorders Primary hyperparathyroidism Hyperthyroidism Malignancy Granulomatous disorders (with calcitriol production) Disuse Drug-induced Vitamin D poisoning Milk-Alkali syndrome Thiazide diuretics Lithium salts Benign familial hypocalciuric hypercalcemia
Increased renal tubular reabsorption of calcium
+ ++ + or ++ ++
+ − + or − −
++
−
++ − − − −
− + + + +
Figure 4
351
in relation to variations in extracellular calcium concentration [58]. In familial hypocalciuric benign hypercalcemia, circulating levels of parathyroid hormone are insufficiently suppressed for the degree of calcemia, and the renal tubular reabsorption of calcium is increased through a parathyroid hormone-independent mechanism [39, 40, 59]. This disorder appears to be due to mutations associated with hypofunction of the cell membrane calcium-sensing mechanism [58, 60]. Thus, higher plasma calcium levels are necessary to inhibit PTH secretion. A similar syndrome can be reproduced in transgenic mice by the incorporation of transgenes displaying the same mutations [61]. Interestingly, the same biochemical pattern can be encountered in patients treated with lithium salts [62, 63]. Conversely, in certain cases of familial
Relationship between bone resorption, as evaluated by the bone resorption index (BRI) and tubular reabsorption of calcium index (TRCaI) in rehydrated patients with malignant (A) or non malignant (B) hypercalcemia. In parentheses the mean plasma calcium concentration is given. This figure is taken from Buchs et al. (Bone 12: 47–56, 1991), with the permission of the publisher.
352
RENÉ RIZZOLI AND JEAN-PHILIPPE BONJOUR
hypoparathyroidism, a mutation-induced hyperactive calcium-sensing receptor has been found [59]. Under these conditions, parathyroid hormone production is not stimulated despite reduced calcium levels. Calcium-sensing receptor activity can be modulated by calcimimetics, which enhance its sensitivity to extracellular calcium [58]. Under these conditions, PTH secretion is inhibited. Conversely, calcilytics are negative modulators of the calcium-sensing receptor, leading thereby to a stimulation of PTH secretion. There are other structures recognizing changes in extracellular calcium concentration [64–67]. In osteoblasts, they may be distinct from those present in parathyroid cells [68–70].
B. Effectors The perturbation of plasma calcium is corrected by the stimulation of the tubular reabsorption of calcium, and the mobilization of calcium from bone mineral (Fig. 5). Parathyroid hormone itself, as well as a decrease in calcium and/or phosphate concentrations, directly stimulates the synthesis of calcitriol at the renal level. The latter hormone contributes to an increase in plasma calcium through a mobilization of calcium from bone and a stimulation of intestinal calcium absorption. Thus, calcitriol plays a central role in the intestinal adaptation to low calcium intake. This adaptative mechanism is blunted in the elderly, as a consequence of a decreased synthesis of, and a lower response to, calcitriol [71, 72]. It appears therefore that the integrated control of extracellular calcium homeostasis is governed by the requirement of maintaining extracellular calcium concentration in a very narrow range. This system
Figure 5
is quite efficient in its capacity to respond to a calcium need. Conversely, prevention of overwhelming by calcium is ensured by the reversal of these systemic mechanisms. Although calcitonin is able to inhibit bone resorption and to increase the renal clearance of calcium, when given in pharmacological doses, it is improbable that this hormone significantly participates in the physiological defense against hypercalcemia. During body growth, or in the third trimester of pregnancy, when calcification of the fetus bone takes place, the need for calcium results in an increase in intestinal calcium absorption. Elevated levels of calcitriol are responsible for stimulating intestinal calcium absorption [16, 18, 19]. On the other hand, with, for instance, estrogen deprivation at menopause, bone bidirectional calcium fluxes are increased, with resorption overcoming accretion, resulting in a net negative skeletal balance. Under these conditions, a reduced intestinal calcium absorption and/or a higher urinary calcium excretion can be viewed as a homeostatic mechanism attempting to prevent the extracellular calcium compartment from being overloaded. A similar reaction can occur with prolonged immobilization and the consequent net calcium loss.
VII. CALCIUM AND BONE GROWTH Several observational studies have suggested that increasing the calcium intake would promote a greater bone mass gain, and thereby a higher peak bone mass [5]. Furthermore, several prospective randomized, doubleblind, placebo-controlled intervention trials indicate that calcium supplementation can increase bone mass gain,
Homeostatic response to a decrease in extracellular (EC) calcium concentration.
353
Chapter 20 Physiology of Calcium and Phosphate Homeostases
although the magnitude of the calcium effects appears to vary according to the skeletal sites examined, the stage of pubertal maturation at the time of the intervention, and the spontaneous calcium intake [73–78]. Furthermore, these effects could be modulated by an interaction with vitamin D receptor genotype [79]. The positive effects of calcium supplementation have essentially been ascribed to a reduction in bone remodeling. Indeed, in one study the plasma level of osteocalcin, a biochemical marker of bone remodeling in adults, was significantly reduced in the calciumsupplemented children [73]. This interpretation is also in keeping with the currently favored mechanism proposed to account for the inhibitory effect of calcium on age-related bone loss. In the elderly, the calcium effect on bone remodeling is usually ascribed to an inhibition of parathyroid hormone secretion, as the plasma level tends to increase with aging [80–85]. In a double-blind, placebo-controlled study on the effects of calcium supplementation in prepubertal girls, changes in scanned bone area and in standing height suggest that calcium supplementation could affect bone modeling in addition to bone remodeling [76, 77]. Indeed, milk calcium-enriched foods enhanced the gain of both mean scanned bone area of several skeletal sites, and statural height in the group of spontaneously low-calcium consumers, to the level achieved by the spontaneously highcalcium consumers. Morphometric analysis of the changes observed in the lumbar spine and in femoral diaphysis suggests that calcium could enhance both the longitudinal and the cross-sectional growth of the bones [76, 77]. Effects of calcium on the secretion or action of growth factors, including the IGFs-IGF-binding proteins system, could be implicated [86, 87]. Bone mineral density was measured 7.5 years after the end of calcium supplementation. In these young adult girls, it appeared that menarche occurred earlier in the calcium-supplemented group, and that persistent effects of calcium were mostly detectable in those subjects with an earlier puberty [77]. Finally, because of a higher response to calcium supplements preferentially observed in a certain vitamin D receptor genotype [79], it remains to be determined whether the interaction between vitamin D receptor genotype and the influence of calcium supplementation on bone mass accrual affects modeling and/or remodeling.
VIII. BODY DISTRIBUTION OF PHOSPHORUS Phosphorus is the sixth element in abundance in the body, mostly under the form of phosphate. Its mass represents about 1% of body weight, i.e. 700 g for an individual of
70 kg [1]. More than 80% of phosphate is in bone, the remaining in soft tissues, in the cellular and extracellular compartments. Inside the cells, phosphate can be either under an inorganic form, or an organic one, as a constituent of carbohydrates, lipids, or nucleic acids (Table I). It plays a crucial role in the energy storage and delivery, in the regulation of a large variety of different enzymes activity, and in intracellular signal transduction mechanisms. It also contributes to ensure intracellular and plasma membrane stability, and is implicated in the modulation of hemoglobin affinity for oxygen. Phosphate enters the cells through an active transport system, the energy for which is provided by the extra/intracellular sodium gradient [88, 89]. Extracellular phosphate represents 0.1% of body phosphate. In plasma, one-third of phosphate is inorganic, of which 90% is ultrafiltrable. Two-thirds are parts of circulating phospholipids. At physiological pH, the divalent form of inorganic phosphate predominates. This anion contributes to the maintenance of extracellular acid–base homeostasis. Inorganic phosphate concentration varies in relation to age and body growth rate. In adults, its concentration ranges between 0.8 and 1.4 mmol/L, whereas it is 1.4–2.7 and 1.3–2.0 in the neonate and the adolescent, respectively. There is a nycthemeral rhythm in inorganic phosphate concentration, with highest values encountered in the late afternoon.
IX. DETERMINANTS OF EXTRACELLULAR PHOSPHATE CONCENTRATION The levels of extracellular phosphate are determined by the balance between dynamic fluxes from and into extracellular compartment, which occur in intestine, bone, soft tissue, and kidney (Fig. 6).
A. Intestinal Fluxes A balanced normal diet provides sufficient amounts of phosphate in most circumstances, so that phosphate deficiency from dietary origin usually does not occur. In the intestinal lumen, phosphate can be complexed with various cations, including aluminum or calcium, which thereby prevents its absorption. This property is applied in the prevention of phosphate overload in the frame of advanced renal failure and phosphate retention. The jejunum exhibits the highest absorptive capacity. Two mechanisms are involved. The first one, which predominates in the proximal small intestine, is an active and saturable transport system, activated by calcitriol [17, 21]. The second one is a passive
354
RENÉ RIZZOLI AND JEAN-PHILIPPE BONJOUR
Figure 6
Main fluxes (mg/day) controlling phosphate homeostasis.
and nonsaturable transport, which plays quantitatively the most important role. Indeed, an increase in phosphate intake is accompanied by a commensurate increment of net intestinal phosphate absorption. The saturable component becomes negligible at high phosphate intake. Under normal conditions, 60–70% of dietary phosphate is absorbed. At steady state, daily urinary phosphate excretion is a direct reflection of phosphate absorbed in the intestine, and thus of the phosphate intake.
B. Bone Fluxes The processes of bone formation and resorption implicate bidirectional fluxes of phosphate from and into bone mineral. Osteoclastic bone resorption releases phosphate into the extracellular fluid. The deposition of phosphate into newly formed osteoid tissue is dependent on its extracellular concentration. In addition, the plasma membrane of osteoblastic cells, as well as extracellular matrix vesicles found in epiphyseal cartilage and in woven bone, are endowed with a saturable, carrier-mediated phosphate transport system, modulated by various hormones and growth factors [90–92]. These extracellular matrix vesicles could play a role in the initiation of the calcification process. The plasma membrane-associated receptor for gibbon ape leukemia virus [93], which functions as a sodium-dependent phosphate transporter, distinct from the renal type I and II sodium-dependent phosphate transporters, has been found in osteoblastic cells and is regulated by IGF-I [94]. Thus, this type III phosphate transporter could participate in the control of the flux of phosphate deposition into bone.
C. Renal Fluxes Approximately 70% of the filtered inorganic phosphate is reabsorbed in the proximal tubule through saturable sodium co-transport mechanisms [95], whereas about 20% of the filtered load is excreted in the urine (Fig. 7). An important step in the transfer of phosphate from the lumen to peritubular capillaries takes place in the epithelial cell brush-border membrane. At least two different sodiumdependent phosphate transporters have been identified in the renal tubule, which share little homology in their amino acid sequence [88, 89]. The type I transporter appears to be constitutive, nonselective, and unregulated. The type II transporter is controlled by dietary phosphate intake or by parathyroid hormone/parathyroid hormone-related protein. Short-term exposure to a low-phosphate diet stimulates the insertion of type II phosphate transporters into the apical membrane, whereas parathyroid hormone decreases it. Prolonged low dietary phosphate intakes increase the expression of type II transporter. Its presence in the brush border membrane is markedly depressed in X-linked hypophosphatemic rachitic mice [88, 89]. However, this defect has been mapped to another gene than that coding for the type II phosphate transporter. The gene mutated in the case of X-linked hypophosphatemia codes for an endopeptidase, for which the substrate is, however, still not clearly elucidated [96–99]. The maximal tubular reabsorption capacity (TmPi/ GFR) is the main determinant of plasma phosphate concentration [95]. Since the tubular reabsorption is a saturable process, fractional reabsorption for a given TmPi/GFR varies according to the filtered load. Therefore, fractional
355
Chapter 20 Physiology of Calcium and Phosphate Homeostases
Figure 7
Central role of the kidney in controlling phosphate homeostasis.
excretion or reabsorption cannot be taken as a reliable reflection of the tubular reabsorptive capacity. This capacity is controlled by the phosphate supply and the need for phosphate of the organism [100]. Not only is the low dietary phosphate-mediated stimulation of phosphate transport independent of parathyroid hormone, but it prevents the phosphaturic response to the hormone [101]. During growth, a state of need for phosphate, or during phosphate deficiency, TmPi/GFR is increased [95]. In proximal tubule kidney cells, phosphate transport is activated by IGF-I, whereas it is decreased by parathyroid hormone, or its tumoral analog parathyroid hormonerelated protein [102–104]. Other factors could influence the sodium-dependent phosphate transport [47, 48]; however, their contribution to alterations observed in physiological or pathophysiological situations remains to be established. IGF-I, which plays a prominent role in the longitudinal growth of bone, also increases the renal synthesis of calcitriol [105], which in turn stimulates the intestinal absorption of phosphate [17, 21] (Table III). Table III.
Thus, by acting indirectly on the intestine through calcitriol, and directly on the kidney, IGF-I contributes to make positive the phosphate balance, favoring thereby the mineralization of newly formed bone (Fig. 8). Parathyroid hormone, or parathyroid hormone-related protein, as well as a state of phosphate overload, or elevated extracellular calcium concentration reduce TmPi/GFR. Calcitriol appears to be devoid of any direct and significant effect on the renal tubular reabsorption of phosphate under physiological conditions. Factor(s) other than PTH or PTHrP are known to increase phosphate excretion, through influencing renal tubular reabsorption of phosphate. Although the molecules implicated are not fully elucidated, the name “phosphatonin” has been suggested [106, 107]. Evidence for the existence of such factor(s) has been obtained from pathological conditions such as tumor-associated osteomalacia, X-linked hypophosphatemia, or autosomal dominant hypophosphatemic rickets. Genomic and proteomic procedures allowed one to identify fibroblast
Factors Affecting Phosphate Renal Tubular Reabsorption
Factors that decrease Pi reabsorption Phosphate loading Parathyroid hormone/PTHrP/cAMP Volume expansion Hypercalcemia Carbonic anhydrase inhibitors Fibroblast growth factor 23 Frizzled receptor protein 4
Factors that increase Pi reabsorption Phosphate depletion Parathyroidectomy Volume contraction Hypocalcemia Growth hormone Insulin-like growth factor-I
Figure 8
Effects of IGF-I on bone and on mineral homeostasis.
356
RENÉ RIZZOLI AND JEAN-PHILIPPE BONJOUR
growth factor 23 (FGF23), frizzled receptor protein 4 (FRP4), and matrix extracellular phosphoglycoprotein (MEPE) has exhibited phosphaturic activity [107–110] (Fig. 9).
D. Soft Tissues In constrast to calcium, phosphate fluxes from and into the intracellular compartment can alter extracellular phosphate concentration. During fasting, phosphate mobilization from soft tissues, such as the liver, can contribute to an increase in plasma phosphate. Inversely, during the postabsorptive phase, extracellular phosphate is transferred into soft tissues, as a result of incorporation into carbohydrates or lipids under the influence of insulin [111, 112]. Hyperphosphatemia, in the frame of a tumor lysis syndrome following the initiation of cytotoxic therapy of hematologic malignancies, or hypophosphatemia occurring during the treatment of diabetic ketoacidosis with insulin, are examples of the role played by soft tissues in phosphate homeostasis under pathological conditions [113, 114].
X. HOMEOSTATIC RESPONSES TO CHANGES IN PHOSPHATE SUPPLY OR DEMAND Phosphate deficiency stimulates calcitriol production, which increases the transfer of phosphate from the intestine and mobilization from bone mineral (Fig. 10). The renal synthesis and release of calcitriol are related to the phosphate supply and the demand of the organism. On the other hand, low extracellular phosphate concentration is associated with higher calcium concentration through a physicochemical process. In turn, the inhibition of parathyroid hormone release decreases urinary phosphate excretion.
Figure 9
Effects of “phosphatonins” on Pi homeostasis.
Figure 10 Renal Pi handling and serum FGF23 levels in relation to dietary Pi intakes (from Ferrari et al., J Clin Endocrinol Metab 2005, with permission of the publisher).
The most powerful mechanism for phosphate retention is certainly the system, the molecular nature of which is still not elucidated, which allows the kidney to adapt very tightly its capacity of reabsorbing phosphate to dietary supplies and to the need of the organism in phosphate, in a parathyroid hormone-independent manner [88, 89, 95, 96, 100, 101, 115]. FGF23 could contribute to this dietary phosphate-induced modulation of renal tubular transport. Indeed, increasing dietary phosphate intakes was associated with higher FGF23 levels in young healthy males undergoing acute changes in phosphate supply [116] (Fig. 11). The adaptation to low phosphate intakes in terms of calcitriol production and changes in renal tubular phosphate reabsorption is missing in X-linked hypophosphatemic rickets [115]. This indicates that the homeostatic response to meet the phosphate supply and need is altered in this disorder, leading to a renal phosphate leak and inadequate calcitriol levels for the degree of hypophosphatemia, despite the need for phosphate for bone growth and mineralization. Various mutations in the gene to which X-linked hypophosphatemic rickets has been mapped, have been reported, both in humans and in mice [97–99].
357
Chapter 20 Physiology of Calcium and Phosphate Homeostases
Sensitive and efficient regulatory mechanisms involving local calcium sensing are triggered by changes in calcium demand or supply. Similarly, the renal handling of phosphate can adjust its capacity to meet the need of the organism. The regulation of calcium and phosphate homeostases aims at fighting against a deficiency or an overload. In response to a variation of any regulatory flux, a series of homeostatic responses and adjustments is triggered, leading to a new steady state, in which the initially altered variable has been corrected.
Acknowledgments Figure 11
Homeostatic response to a decrease in extracellular (EC) phosphate concentration.
The role played by this PHEX (phosphate-regulating gene with homology to endopeptidase on the X chromosome) gene product, which is a zinc-binding endopeptidase, in the regulation of phosphate homeostasis, is still not elucidated. As mentioned above, this phosphate homeostatic mechanism could play a role in the increase of renal tubular reabsorption of phosphate observed during growth, possibly in association with IGF-I, but also in disorders characterized by an increased demand, as for instance in extensive osteoblastic metastases [117]. An inhibition of the same system could also be implicated to avoid phosphate overload in the frame of a decreased nephron mass, as for instance in progressive renal failure [118, 119]. There is also evidence that elevated extracellular phosphate concentration could also be able to directly increase parathyroid hormone production [120]. This mechanism is apparently less efficacious than the one triggered by hypocalcemia. Because of the phosphaturic effects of parathyroid hormone, such a regulatory system certainly represents an attempt to maintain the homeostasis of phosphate economy.
XI. CONCLUSIONS The physiology of calcium and phosphate homeostases is regulated by coordinated bidirectional calcium and phosphate fluxes, occurring at the levels of intestine, bone, and kidney. These fluxes are influenced by calciotropic peptides or steroid hormones, and by a variety of locally produced factors. By directly modifying renal tubular calcium and phosphate transport, or by releasing calcium from intracellular stores, calcium itself functions as a regulator. In the control of extracellular concentration of either ion, the kidney tubule reabsorptive capacity plays a central role.
We thank Mrs M. Perez for her expert secretarial assistance.
References 1. Broadus, A.E. (2003). Mineral balance and homeostasis. In “Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism” (M. J. Favus, ed.), pp. 105–111. American Society for Bone and Mineral Research, Washington, D.C. 2. Brown, E. M., and MacLeod, R. J. (2001). Extracellular calcium sensing and extracellular calcium signaling. Physiol. Rev. 81, 239–297. 3. Bonjour, J. P., Theintz, G., Buchs, B., Slosman, D., and Rizzoli, R. (1991). Critical years and stages of puberty for spinal and femoral bone mass accumulation during adolescence. J. Clin. Endocrinol. Metab. 73, 555–563. 4. Theintz, G., Buchs, B., Rizzoli, R., Slosman, D., Clavien, H., Sizonenko, P. C., and Bonjour, J. P. (1992). Longitudinal monitoring of bone mass accumulation in healthy adolescents: evidence for a marked reduction after 16 years of age at the levels of lumbar spine and femoral neck in female subjects. J. Clin. Endocrinol. Metab. 75, 1060–1065. 5. Bonjour, J. P., and Rizzoli, R. (2001). Bone acquisition in adolescence. In “Osteoporosis” (R. Marcus, D. Feldman, and J. Kelsey, eds.), pp. 621–638. Academic Press, San Diego, CA. 6. Blair, H. C., Robinson, L. J., and Zaidi, M. (2005). Osteoclast signalling pathways. Biochem. Biophys. Res. Commun. 328, 728–738. 7. Parekh, A. B., and Putney, J. W., Jr. (2005). Store-operated calcium channels. Physiol. Rev. 85, 757–810. 8. Dvorak, M. M., and Riccardi, D. (2004). Ca2+ as an extracellular signal in bone. Cell Calcium 35, 249–255. 9. Szabadkai, G., and Rizzuto, R. (2004). Participation of endoplasmic reticulum and mitochondrial calcium handling in apoptosis: more than just neighborhood? FEBS Lett. 567, 111–115. 10. Reeves, J. P., Condrescu, M., Chernaya, G., and Gardner, J. P. (1994). Na+/Ca2+ antiport in the mammalian heart. J. Exp. Biol. 196, 375–388. 11. Carafoli, E. (2002). Calcium signaling: a tale for all seasons. Proc. Natl Acad. Sci. USA 99, 1115–1122. 12. Carafoli, E. (2004). Calcium-mediated cellular signals: a story of failures. Trends Biochem. Sci. 29, 371–379. 13. Rodland, K. D. (2004). The role of the calcium-sensing receptor in cancer. Cell Calcium 35, 291–295. 14. McConkey, D. J., and Orrenius, S. (1996). The role of calcium in the regulation of apoptosis. J. Leukoc. Biol. 59, 775–783. 15. Rizzoli, R., Fleisch, H., and Bonjour, J. P. (1977). Effect of thyroparathyroidectomy of calcium metabolism in rats: role of 1,25dihydroxyvitamin D3. Am. J. Physiol. 233, E160–164.
358 16. Lopez-Hilker, S., Dusso, A. S., Rapp, N. S., Martin, K. J., and Slatopolsky, E. (1990). Phosphorus restriction reverses hyperparathyroidism in uremia independent of changes in calcium and calcitriol. Am. J. Physiol. 259, F432–437. 17. Hoenderop, J. G., Nilius, B., and Bindels, R. J. (2005). Calcium absorption across epithelia. Physiol. Rev. 85, 373–422. 18. Holick, M. F. (2003). Vitamin D: photobiology, metabolism, mechanism of action, and clinical applications. In “Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism” (M. J. Favus, ed.), pp. 129–137. American Society for Bone and Mineral Research, Washington, D.C. 19. Feldman, D., Malloy, P. J., and Gross, C. (2001). Biology, action, and clinical implications. In “Osteoporosis” (R. Marcus, D. Feldman, and J. Kelsey, eds.), pp. 257–303. Academic Press, San Diego, CA. 20. Bonjour, J. P., Fleisch, H., and Trechsel, U. (1977). Calcium absorption in diphosphonate-treated rats: effect of parathyroid function, dietary calcium and phosphorus. J. Physiol. 264, 125–139. 21. Rizzoli, R., Fleisch, H., and Bonjour, J. P. (1977). Role of 1,25dihydroxyvitamin D3 on intestinal phosphate absorption in rats with a normal vitamin D supply. J. Clin. Invest. 60, 639–647. 22. Heaney, R. P., Weaver, C. M., and Recker, R. R. (1988). Calcium absorbability from spinach. Am. J. Clin. Nutr. 47, 707–709. 23. Heaney, R. P., and Weaver, C. M. (1990). Calcium absorption from kale. Am. J. Clin. Nutr. 51, 656–657. 24. Ammann, P., Rizzoli, R., and Fleisch, H. (1986). Calcium absorption in rat large intestine in vivo: availability of dietary calcium. Am. J. Physiol. 251, G14–18. 25. Pak, C. Y., Kaplan, R., Bone, H., Townsend, J., and Waters, O. (1975). A simple test for the diagnosis of absorptive, resorptive and renal hypercalciurias. N. Engl. J. Med. 292, 497–500. 26. Lemann, J., and Favus, M. J. (2003). Intestinal absorption of calcium, magnesium and phosphate. In “Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism” (M. J. Favus, ed.), pp. 93–97. American Society for Bone and Mineral Research, Washington, D.C. 27. Hefti, E., Trechsel, U., Fleisch, H., and Bonjour, J. P. (1983). Nature of calcemic effect of 1,25-dihydroxyvitamin D3 in experimental hypoparathyroidism. Am. J. Physiol. 244, E313–316. 28. Jüppner, H., and Kronenberg, H. M. (2003). Parathyroid hormone. In “Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism” (M. J. Favus, ed.), pp. 117–124. American Society for Bone and Mineral Research, Washington, D.C. 29. Mundy, G. R., Chen, D., and Oyajobi, B. O. (2003). Bone remodeling. In “Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism” (M. J. Favus, ed.), pp. 46–58, American Society for Bone and Mineral Research, Washington, D.C. 30. Manolagas, S. C. (2000). Birth and death of bone cells: basic regulatory mechanisms and implications for the pathogenesis and treatment of osteoporosis. Endocr. Rev. 21, 115–137. 31. Martin, T. J., and Rodan, G. A. (2001). Coupling of bone resorption and formation during bone remodeling. In “Osteoporosis” (R. Marcus, D. Feldman, and J. Kelsey, eds.), pp. 361–371. Academic Press, San Diego, CA. 32. Hofbauer, L. C., and Schoppet, M. (2004). Clinical implications of the osteoprotegerin/RANKL/RANK system for bone and vascular diseases. JAMA 292, 490–495. 33. Peacock, M., Robertson, W. G., and Nordin, B. E. (1969). Relation between serum and urinary calcium with particular reference to parathyroid activity. Lancet 1, 384–386. 34. Nordin, B. E., and Peacock, M. (1969). Role of kidney in regulation of plasma-calcium. Lancet 2, 1280–1283. 35. Bonjour, J. P., Philippe, J., Guelpa, G., Bisetti, A., Rizzoli, R., Jung, A., Rosini, S., and Kanis, J. A. (1988). Bone and renal components in hypercalcemia of malignancy and responses to a single infusion of clodronate. Bone 9, 123–130.
RENÉ RIZZOLI AND JEAN-PHILIPPE BONJOUR 36. Parsons, J. A., and Robinson, C. J. (1971). Calcium shift into bone causing transient hypocalcaemia after injection of parathyroid hormone. Nature 230, 581–582. 37. Parsons, J. A., Neer, R. M., and Potts, J. T., Jr. (1971). Initial fall of plasma calcium after intravenous injection of parathyroid hormone. Endocrinology 89, 735–740. 38. Friedman, P. A. (2000). Mechanisms of renal calcium transport. Exp. Nephrol. 8, 343–350. 39. Marx, S. J., Spiegel, A. M., Levine, M. A., Rizzoli, R. E., Lasker, R. D., Santora, A. C., Downs, R. W., Jr., and Aurbach, G. D. (1982). Familial hypocalciuric hypercalcemia: the relation to primary parathyroid hyperplasia. N. Engl. J. Med. 307, 416–426. 40. Attie, M. F., Gill, J. R., Jr., Stock, J. L., Spiegel, A. M., Downs, R. W., Jr., Levine, M. A., and Marx, S. J. (1983). Urinary calcium excretion in familial hypocalciuric hypercalcemia. Persistence of relative hypocalciuria after induction of hypoparathyroidism. J. Clin. Invest. 72, 667–676. 41. Thakker, R. V. (2004). Diseases associated with the extracellular calcium-sensing receptor. Cell Calcium 35, 275–282. 42. Rizzoli, R., Hugi, K., Fleisch, H., and Bonjour, J. P. (1981). Effect of hydrochlorothiazide on 1,25-dihydroxyvitamin D3-induced changes in calcium metabolism in experimental hypoparathyroidism in rats. Clin. Sci. (Lond.) 60, 101–107. 43. Rizzoli, R., Pizurki, L., and Bonjour, J. P. (1989). Tumoral parathyroid hormone-related protein: a new hormone affecting calcium and phosphate transport. NIPS 4, 195–198. 44. Strewler, G. J., and Nissenson, R. A. (2003). Parathyroid hormonerelated protein. In “Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism” (M. J. Favus, ed.), pp. 125–129. American Society for Bone and Mineral Research, Washington, D.C. 45. Bisello, A., Horwitz, M. J., and Stewart, A. F. (2004). Parathyroid hormone-related protein: an essential physiological regulator of adult bone mass. Endocrinology 145, 3551–3553. 46. Caverzasio, J., Rizzoli, R., Vallotton, M. B., Dayer, J. M., and Bonjour, J. P. (1993). Stimulation by interleukin-1 of renal calcium reabsorption in thyroparathyroidectomized rats. J. Bone Miner. Res. 8, 1219–1225. 47. Pizurki, L., Rizzoli, R., Caverzasio, J., and Bonjour, J. P. (1990). Effect of transforming growth factor-alpha and parathyroid hormone-related protein on phosphate transport in renal cells. Am. J. Physiol. 259, F929–935. 48. Law, F., Rizzoli, R., and Bonjour, J. P. (1993). Transforming growth factor-beta inhibits phosphate transport in renal epithelial cells. Am. J. Physiol. 264, F623–628. 49. Law, F., Bonjour, J. P., and Rizzoli, R. (1994). Transforming growth factor-beta: a down-regulator of the parathyroid hormone-related protein receptor in renal epithelial cells. Endocrinology 134, 2037–2043. 50. Buchs, B., Rizzoli, R., and Bonjour, J. P. (1991). Evaluation of bone resorption and renal tubular reabsorption of calcium and phosphate in malignant and nonmalignant hypercalcemia. Bone 12, 47–56. 51. Rizzoli, R., Stoermann, C., Ammann, P., and Bonjour, J. P. (1994). Hypercalcemia and hyperosteolysis in vitamin D intoxication: effects of clodronate therapy. Bone 15, 193–198. 52. Rizzoli, R., Caverzasio, J., Chapuy, M. C., Martin, T. J., and Bonjour, J. P. (1989). Role of bone and kidney in parathyroid hormonerelated peptide-induced hypercalcemia in rats. J. Bone Miner. Res. 4, 759–765. 53. Rizzoli, R., Caverzasio, J., Bauss, F., and Bonjour, J. P. (1992). Inhibition of bone resorption by the bisphosphonate BM 21.0955 is not associated with an alteration of the renal handling of calcium in rats infused with parathyroid hormone-related protein. Bone 13, 321–325. 54. Hirschel-Scholz, S., Caverzasio, J., and Bonjour, J. P. (1985). Inhibition of parathyroid hormone secretion and parathyroid hormoneindependent diminution of tubular calcium reabsorption by WR-2721, a unique hypocalcemic agent. J. Clin. Invest. 76, 1851–1856.
Chapter 20 Physiology of Calcium and Phosphate Homeostases 55. Hirschel-Scholz, S., Caverzasio, J., Rizzoli, R., and Bonjour, J. P. (1986). Normalization of hypercalcemia associated with a decrease in renal calcium reabsorption in Leydig cell tumor-bearing rats treated with WR-2721. J. Clin. Invest. 78, 319–322. 56. Chen, R. A., and Goodman, W. G. (2004). Role of the calcium-sensing receptor in parathyroid gland physiology. Am. J. Physiol. Renal. Physiol. 286, F1005–1011. 57. Brown, E. M., Gamba, G., Riccardi, D., Lombardi, M., Butters, R., Kifor, O., Sun, A., Hediger, M. A., Lytton, J., and Hebert, S. C. (1993). Cloning and characterization of an extracellular Ca(2+)-sensing receptor from bovine parathyroid. Nature 366, 575–580. 58. Hu, J., and Spiegel, A. M. (2003). Naturally occurring mutations of the extracellular Ca2+-sensing receptor: implications for its structure and function. Trends Endocrinol. Metab. 14, 282–288. 59. Brown, E. M., Pollak, M. R., and Hebert, S. C. (1998). The extracellular calcium-sensing receptor: its role in health and disease. Annu. Rev. Med. 49, 15–29. 60. Pollak, M. R., Brown, E. M., Chou, Y. H. W., Hebert, S. C., Marx, S. J., Steinmann, B., Levi, T., Seidman, C. E., and Seidman, J. G. (1993). Mutations in the human Ca2+ sensing receptor gene cause familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Cell 75, 1297–1303. 61. Ho, C., Conner, D. A., Pollak, M. R., Ladd, D. J., Kifor, O., Warren, H. B., Brown, E. M., Seidman, J. G., and Seidman, C. E. (1995). A mouse model of human familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Nat. Genet. 11, 389–394. 62. Mallette, L. E., and Eichhorn, E. (1986). Effects of lithium carbonate on human calcium metabolism. Arch. Intern. Med. 146, 770–776. 63. Bendz, H., Sjodin, I., Toss, G., and Berglund, K. (1996). Hyperparathyroidism and long-term lithium therapy – a cross-sectional study and the effect of lithium withdrawal. J. Intern. Med. 240, 357–365. 64. House, M. G., Kohlmeier, L., Chattopadhyay, N., Kifor, O., Yamaguchi, T., Leboff, M. S., Glowacki, J., and Brown, E. M. (1997). Expression of an extracellular calcium-sensing receptor in human and mouse bone marrow cells. J. Bone Miner. Res. 12, 1959–1970. 65. Sands, J. M., Naruse, M., Baum, M., Jo, I., Hebert, S. C., Brown, E. M., and Harris, H. W. (1997). Apical extracellular calcium/polyvalent cation-sensing receptor regulates vasopressin-elicited water permeability in rat kidney inner medullary collecting duct. J. Clin. Invest. 99, 1399–1405. 66. Chattopadhyay, N., Cheng, I., Rogers, K., Riccardi, D., Hall, A., Diaz, R., Hebert, S. C., Soybel, D. I., and Brown, E. M. (1998). Identification and localization of extracellular Ca(2+)-sensing receptor in rat intestine. Am. J. Physiol. 274, G122–130. 67. Buchs, N., Manen, D., Bonjour, J. P., and Rizzoli, R. (2000). Calcium stimulates parathyroid hormone-related protein production in Leydig tumor cells through a putative cation-sensing mechanism. Eur. J. Endocrinol. 142, 500–505. 68. Quarles, L. D., Hartle, J. E., 2nd, Siddhanti, S. R., Guo, R., and Hinson, T. K. (1997). A distinct cation-sensing mechanism in MC3T3-E1 osteoblasts functionally related to the calcium receptor. J. Bone Miner. Res. 12, 393–402. 69. Pi, M., and Quarles, L. D. (2004). A novel cation-sensing mechanism in osteoblasts is a molecular target for strontium. J. Bone Miner. Res. 19, 862–869. 70. Yamaguchi, T., Kifor, O., Chattopadhyay, N., and Brown, E. M. (1998). Expression of extracellular calcium (Ca2 + o)-sensing receptor in the clonal osteoblast-like cell lines, UMR-106 and SAOS-2. Biochem. Biophys. Res. Commun. 243, 753–757. 71. Gallagher, J. C., Riggs, B. L., Eisman, J., Hamstra, A., Arnaud, S. B., and DeLuca, H. F. (1979). Intestinal calcium absorption and serum vitamin D metabolites in normal subjects and osteoporotic patients: effect of age and dietary calcium. J. Clin. Invest. 64, 729–736. 72. Ebeling, P. R., Sandgren, M. E., DiMagno, E. P., Lane, A. W., DeLuca, H. F., and Riggs, B. L. (1992). Evidence of an age-related
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
359 decrease in intestinal responsiveness to vitamin D: relationship between serum 1,25-dihydroxyvitamin D3 and intestinal vitamin D receptor concentrations in normal women. J. Clin. Endocrinol. Metab. 75, 176–182. Johnston, C. C., Jr., Miller, J. Z., Slemenda, C. W., Reister, T. K., Hui, S., Christian, J. C., and Peacock, M. (1992). Calcium supplementation and increases in bone mineral density in children. N. Engl. J. Med. 327, 82–87. Lloyd, T., Andon, M. B., Rollings, N., Martel, J. K., Landis, J. R., Demers, L. M., Eggli, D. F., Kieselhorst, K., and Kulin, H. E. (1993). Calcium supplementation and bone mineral density in adolescent girls. JAMA 270, 841–844. Lee, W. T., Leung, S. S., Xu, Y. C., Wang, S. H., Zeng, W. P., Lau, J., and Fairweather-Tait, S. J. (1995). Effects of double-blind controlled calcium supplementation on calcium absorption in Chinese children measured with stable isotopes (42Ca and 44Ca). Br. J. Nutr. 73, 311–321. Bonjour, J. P., Carrie, A. L., Ferrari, S., Clavien, H., Slosman, D., Theintz, G., and Rizzoli, R. (1997). Calcium-enriched foods and bone mass growth in prepubertal girls: a randomized, double-blind, placebocontrolled trial. J. Clin. Invest. 99, 1287–1294. Chevalley, T., Rizzoli, R., Hans, D., Ferrari, S., and Bonjour, J. P. (2005). Interaction between calcium intake and menarcheal age on bone mass gain: an eight-year follow-up study from prepuberty to postmenarche. J. Clin. Endocrinol. Metab. 90, 44–51. Chevalley, T., Bonjour, J. P., Ferrari, S., Hans, D., and Rizzoli, R. (2005). Skeletal site selectivity in the effects of calcium supplementation on areal bone mineral density gain: a randomized, double-blind, placebo-controlled trial in prepubertal boys. J. Clin. Endocrinol. Metab. 90, 3342–3349. Ferrari, S. L., Rizzoli, R., Slosman, D. O., and Bonjour, J. P. (1998). Do dietary calcium and age explain the controversy surrounding the relationship between bone mineral density and vitamin D receptor gene polymorphisms? J. Bone Miner. Res. 13, 363–370. Boonen, S., Rizzoli, R., Meunier, P. J., Stone, M., Nuki, G., Syversen, U., Lehtonen-Veromaa, M., Lips, P., Johnell, O., and Reginster, J. Y. (2004). The need for clinical guidance in the use of calcium and vitamin D in the management of osteoporosis: a consensus report. Osteoporos Int. 15, 511–519. Shea, B., Wells, G., Cranney, A., Zytaruk, N., Robinson, V., Griffith, L., Hamel, C., Ortiz, Z., Peterson, J., Adachi, J., Tugwell, P., and Guyatt, G. (2003). Calcium supplementation on bone loss in postmenopausal women. Cochrane Database Syst. Rev. CD004526. Shea, B., Wells, G., Cranney, A., Zytaruk, N., Robinson, V., Griffith, L., Hamel, C., Ortiz, Z., Peterson, J., Adachi, J., Tugwell, P., and Guyatt, G. (2004). Calcium supplementation on bone loss in postmenopausal women. Cochrane Database Syst. Rev. CD004526. Chapuy, M. C., Arlot, M. E., Duboeuf, F., Brun, J., Crouzet, B., Arnaud, S., Delmas, P. D., and Meunier, P. J. (1992). Vitamin D3 and calcium to prevent hip fractures in the elderly women. N. Engl. J. Med. 327, 1637–1642. Chevalley, T., Rizzoli, R., Nydegger, V., Slosman, D., Rapin, C. H., Michel, J. P., Vasey, H., and Bonjour, J. P. (1994). Effects of calcium supplements on femoral bone mineral density and vertebral fracture rate in vitamin-D-replete elderly patients. Osteoporos. Int. 4, 245–252. Dawson-Hughes, B., Harris, S. S., Krall, E. A., and Dallal, G. E. (1997). Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or older. N. Engl. J. Med. 337, 670–676. Honda, Y., Fitzsimmons, R. J., Baylink, D. J., and Mohan, S. (1995). Effects of extracellular calcium on insulin-like growth factor II in human bone cells. J. Bone Miner. Res. 10, 1660–1665. Yoo, A., Tanimoto, H., Akesson, K., Baylink, D. J., and Lau, K. H. (1998). Effects of calcium depletion and repletion on serum insulin-like growth factor I and binding protein levels in weanling rats. Bone 22, 225–232.
360 88. Murer, H., Hernando, N., Forster, I., and Biber, J. (2003). Regulation of Na/Pi transporter in the proximal tubule. Annu. Rev. Physiol. 65, 531–542. 89. Bacic, D., Wagner, C. A., Hernando, N., Kaissling, B., Biber, J., and Murer, H. (2004). Novel aspects in regulated expression of the renal type IIa Na/Pi-cotransporter. Kidney Int. Suppl. S5–S12. 90. Pizurki, L., Rizzoli, R., Caverzasio, J., and Bonjour, J. P. (1991). Stimulation by parathyroid hormone-related protein and transforming growth factor-alpha of phosphate transport in osteoblast-like cells. J. Bone Miner. Res. 6, 1235–1241. 91. Montessuit, C., Bonjour, J. P., and Caverzasio, J. (1995). Expression and regulation of Na-dependent P(i) transport in matrix vesicles produced by osteoblast-like cells. J. Bone Miner. Res. 10, 625–631. 92. Caverzasio, J., and Bonjour, J. P. (1996). Characteristics and regulation of Pi transport in osteogenic cells for bone metabolism. Kidney Int. 49, 975–980. 93. Kavanaugh, M. P., Miller, D. G., Zhang, W., Law, W., Kozak, S. L., Kabat, D., and Miller, A. D. (1994). Cell-surface receptors for gibbon ape leukemia virus and amphotropic murine retrovirus are inducible sodium-dependent phosphate symporters. Proc. Natl Acad. Sci. USA 91, 7071–7075. 94. Palmer, G., Bonjour, J. P., and Caverzasio, J. (1997). Expression of a newly identified phosphate transporter/retrovirus receptor in human SaOS-2 osteoblast-like cells and its regulation by insulin-like growth factor I. Endocrinology 138, 5202–5209. 95. Bonjour, J. P., and Caverzasio, J. (1984). Phosphate transport in the kidney. Rev. Physiol. Biochem. Pharmacol. 100, 161–214. 96. Custer, M., Spindler, B., Verrey, F., Murer, H., and Biber, J. (1997). Identification of a new gene product (diphor-1) regulated by dietary phosphate. Am. J. Physiol. 273, F801–806. 97. The HYP Consortium (1995). A gene (PEX) with homologies to endopeptidases is mutated in patients with X-linked hypophosphatemic rickets. The HYP Consortium. Nat. Genet. 11, 130–136. 98. Du, L., Desbarats, M., Viel, J., Glorieux, F. H., Cawthorn, C., and Ecarot, B. (1996). cDNA cloning of the murine Pex gene implicated in X-linked hypophosphatemia and evidence for expression in bone. Genomics 36, 22–28. 99. Econs, M. J., and Francis, F. (1997). Positional cloning of the PEX gene: new insights into the pathophysiology of X-linked hypophosphatemic rickets. Am. J. Physiol. 273, F489–498. 100. Trohler, U., Bonjour, J. P., and Fleisch, H. (1976). Inorganic phosphate homeostasis. Renal adaptation to the dietary intake in intact and thyroparathyroidectomized rats. J. Clin. Invest. 57, 264–273. 101. Trohler, U., Bonjour, J. P., and Fleisch, H. (1976). Renal tubular adaptation to dietary phosphorus. Nature 261, 145–146. 102. Caverzasio, J., Rizzoli, R., and Bonjour, J. P. (1986). Sodium-dependent phosphate transport inhibited by parathyroid hormone and cyclic AMP stimulation in an opossum kidney cell line. J. Biol. Chem. 261, 3233–3237. 103. Caverzasio, J., and Bonjour, J. P. (1989). Insulin-like growth factor I stimulates Na-dependent Pi transport in cultured kidney cells. Am. J. Physiol. 257, F712–717. 104. Pizurki, L., Rizzoli, R., Moseley, J., Martin, T. J., Caverzasio, J., and Bonjour, J. P. (1988). Effect of synthetic tumoral PTH-related peptide
RENÉ RIZZOLI AND JEAN-PHILIPPE BONJOUR
105.
106.
107. 108.
109.
110.
111. 112.
113. 114.
115.
116.
117.
118.
119.
120.
on cAMP production and Na-dependent Pi transport. Am. J. Physiol. 255, F957–961. Caverzasio, J., Montessuit, C., and Bonjour, J. P. (1990). Stimulatory effect of insulin-like growth factor-1 on renal Pi transport and plasma 1,25-dihydroxyvitamin D3. Endocrinology 127, 453–459. Dunstan, C. R., Zhou, H., and Seibel, M. J. (2004). Fibroblast growth factor 23: a phosphatonin regulating phosphate homeostasis? Endocrinology 145, 3084–3086. Schiavi, S. C., and Kumar, R. (2004). The phosphatonin pathway: new insights in phosphate homeostasis. Kidney Int. 65, 1–14. Riminucci, M., Collins, M. T., Fedarko, N. S., Cherman, N., Corsi, A., White, K. E., Waguespack, S., Gupta, A., Hannon, T., Econs, M. J., Bianco, P., and Gehron Robey, P. (2003). FGF-23 in fibrous dysplasia of bone and its relationship to renal phosphate wasting. J. Clin. Invest. 112, 683–692. Berndt, T., Craig, T. A., Bowe, A. E., Vassiliadis, J., Reczek, D., Finnegan, R., Jan De Beur, S. M., Schiavi, S. C., and Kumar, R. (2003). Secreted frizzled-related protein 4 is a potent tumor-derived phosphaturic agent. J. Clin. Invest. 112, 785–794. Quarles, L. D. (2003). FGF23, PHEX, and MEPE regulation of phosphate homeostasis and skeletal mineralization. Am. J. Physiol. Endocrinol. Metab. 285, E1–9. Marinella, M. A. (2005). Refeeding syndrome and hypophosphatemia. J. Intensive Care Med. 20, 155–159. Takeda, E., Yamamoto, H., Nashiki, K., Sato, T., Arai, H., and Taketani, Y. (2004). Inorganic phosphate homeostasis and the role of dietary phosphorus. J. Cell Mol. Med. 8, 191–200. Bohannon, N. J. (1989). Large phosphate shifts with treatment for hyperglycemia. Arch. Intern. Med. 149, 1423–1425. Jones, D. P., Mahmoud, H., and Chesney, R. W. (1995). Tumor lysis syndrome: pathogenesis and management. Pediatr. Nephrol. 9, 206–212. Hruska, K. A., Rifas, L., Cheng, S. L., Gupta, A., Halstead, L., and Avioli, L. (1995). X-linked hypophosphatemic rickets and the murine Hyp homologue. Am. J. Physiol. 268, F357–362. Ferrari, S. L., Bonjour, J. P., and Rizzoli, R. (2005). Fibroblast growth factor-23 relationship to dietary phosphate and renal phosphate handling in healthy young men. J. Clin. Endocrinol. Metab. 90, 1519–1524. Buchs, N., Bonjour, J. P., and Rizzoli, R. (1998). Renal tubular reabsorption of phosphate is positively related to the extent of bone metastatic load in patients with prostate cancer. J. Clin. Endocrinol. Metab. 83, 1535–1541. Caverzasio, J., Gloor, H. J., Fleisch, H., and Bonjour, J. P. (1982). Parathyroid hormone-independent adaptation of the renal handling of phosphate in response to renal mass reduction. Kidney Int. 21, 471–476. Bonjour, J. P., Caverzasio, J., and Rizzoli, R. (1992). Phosphate homeostasis, 1,25-dihydroxyvitamin D3, and hyperparathyroidism in early chronic renal failure. Trends Endocrinol. Metab. 3, 391–395. Slatopolsky, E., Finch, J., Denda, M., Ritter, C., Zhong, M., Dusso, A., MacDonald, P. N., and Brown, A. J. (1996). Phosphorus restriction prevents parathyroid gland growth. High phosphorus directly stimulates PTH secretion in vitro. J. Clin. Invest. 97, 2534–2540.
Chapter 21
The Central Control of Bone Remodeling Paul A. Baldock
Senior Research Officer, Bone Research Program, Garvan Institute of Medical Research, 384 Victoria Street, Sydney NSW 2010, Australia
Susan J. Allison
Postgraduate Scholar, Bone Research Program, Garvan Institute of Medical Research, 384 Victoria Street, Sydney NSW 2010, Australia
Herbert Herzog
Adjunct Professor, Faculty of Medicine, The University of New South Wales, Principal Research Fellow, Head Obesity and Energy Homeostasis Research Group, Director Neurobiology Program, Garvan Institute of Medical Research, 384 Victoria Street, Sydney NSW 2010, Australia
Edith M. Gardiner
Associate Professor, School of Medicine, The University of Queensland, Head, Skeletal Biology Unit, Centre for Diabetes & Endocrine Research, Ground Floor, C Wing, Bldg 1, Princess Alexandra Hospital, Ipswich Road, Brisbane QLD 4102, Australia
V. Interaction between Leptin and Y2-Regulated Bone Antiosteogenic Pathways VI. Concluding Remarks References
I. Introduction II. Actions of Leptin III. Sympathetic Nervous System IV. Neuropeptide Y and the Y Receptors
I. INTRODUCTION
formation, and that both activities require an intact sympathetic nervous system [1, 3, 4]. Leptin pro-resorptive activity appears to act through two distinct mechanisms, one via sympathetic signals that increase RANKL expression in immature osteoblasts and another signaling through the hypothalamic receptor for cocaine amphetamine regulated transcript, CART, which inhibits osteoblastic RANKL expression via an as-yet-undefined mechanism [3]. Importantly, mice lacking the sympathetic system were resistant to ovariectomy-associated bone loss [3]. This finding, together with the potent bone anabolic responses observed with genetic disruption of the leptin and the
Control of bone physiology through the hypothalamopituitary-corticotropic axis has long been recognized and studied, but more recent discoveries have revealed direct central neural mechanisms in addition to the known neuroendocrine pathways. The finding in 2000 that leptin acts through the hypothalamus to inhibit bone formation was rapidly followed by evidence that the hypothalamic neuropeptide Y2 receptor also mediates a distinct but perhaps interacting antiosteogenic pathway [1, 2]. It is now known that leptin regulates bone resorption as well as Dynamics of Bone and Cartilage Metabolism
361
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
362
PAUL A. BALDOCK ET AL .
Y2-mediated antiosteogenic pathways, emphasizes the potential value of bone-specific therapeutics that target these central control mechanisms for prevention and repair of hypogonadal bone loss. This chapter outlines the current state of understanding and highlights important unresolved issues and future research opportunities in this exciting and rapidly moving area of bone research.
II. ACTIONS OF LEPTIN A. Leptin Role in Energy Homeostasis in Mouse Leptin, the product of the obese gene, is a small polypeptide hormone secreted into the peripheral circulation by adipocytes. Serum levels of leptin, proportional to total body adiposity, are sensed by the leptin receptor in the hypothalamus as part of the energy homeostatic mechanism. Mice with mutations in the obese gene (ob/ob) or its receptor (db/db) display marked obesity accompanied by hyperglycemia, hyperinsulinemia, hypercorticism, and hypogonadism [5–7]. Confirming the potent central action of the hormone, intracerebroventricular (icv) administration of leptin caused a dose-dependent reduction in body weight and food intake in both ob/ob and control mice [6, 8, 9]. Body temperature, locomotor activity, glucose and insulin levels, and body fat mass were also partially corrected after leptin treatment of ob/ob mice [6, 8–10]. Leptin effects in humans and rodents are mediated through receptors in the periphery and the central nervous system. The predominant “short” Ob-Ra form of the leptin receptor is widely expressed in peripheral tissues including kidney, lung, adrenal gland, and choroid plexus [11]. The “long” Ob-Rb receptor variant, generated by alternative splicing [12, 13], is the only isoform that has been shown to mediate signal transduction in vivo [14]. It is most highly expressed in the hypothalamus, particularly in the ventromedial hypothalamic nucleus (VMH) and arcuate nucleus (ARC), which are thought to be the most critical targets for leptin actions [11]. It is the Ob-Rb isoform that is deleted in the db/db obese mouse model [13]. Neuronspecific, but not hepatocyte-specific, disruption of the leptin receptor was associated with obesity, indicating that the nervous system is a primary target for the weight-reducing and neuroendocrine effects of leptin [15].
this was found not to be the case, as cancellous bone volume was markedly increased in both models [1]. This increased bone mass was associated with a marked increase in bone formation, which surpassed the elevation of bone resorption that was also observed. A key role for hypothalamic leptin activity in regulating bone mass was indicated by icv infusion of leptin, which decreased cancellous bone volume in both ob/ob and wild-type mice (Fig. 1) [1]. Consistent with a central mechanism of action, leptin was not detectable in the serum of the icvadministered ob/ob mice. The lack of skeletal effect by circulating leptin was further demonstrated by central icv infusion of leptin to one mouse of a pair of parabiosed ob/ob mice. Although effective cross-circulation was established, correction of the ob/ob skeletal phenotype by a decrease of bone mass occurred only in the recipient, and not the contralateral mouse [4]. In addition, in a separate study using similar duration of leptin treatment, peripheral administration of a 25-fold greater dose of leptin in wild-type mice caused no alteration to cancellous bone volume or bone formation rate [16]. These findings provided strong evidence that, similar to the central control of obesity, the regulation of osteoblast activity on cancellous bone surfaces is mediated by leptin inhibitory actions within the hypothalamus. Subsequent studies implicating
Figure 1
B. Leptin Deficiency and Bone in Mice The hypogonadism and hypercorticism of ob/ob and db/db mice would be expected to result in reduced cancellous bone mass due to elevated bone loss but, unexpectedly,
Central leptin administration reduces cancellous bone mass. Histological sections from vertebrae of 4-month-old ob/ob (A) and wildtype (B) mice following central infusion into the third ventricle with leptin or PBS for 28 days. In both leptin-deficient and wild-type mice, central leptin administration reduced trabecular bone mass and volume. Underlined numbers indicate a statistically significant difference between experimental and control groups of mice (P < 0.05). (Adapted from Ducy et al. Cell 100: 197–207, 2000.)
363
Chapter 21 The Central Control of Bone Remodeling
the sympathetic nervous system as the downstream mediator of leptin antiosteogenic action are covered later in this chapter. The unfolding story of leptin antiosteogenic control provides exciting new insights into neural mechanisms in skeletal physiology, but there are reports in the literature that suggest additional levels of complexity. For example, leptin-resistant Zucker ( fa/fa) rats, similar to the db/db mouse, carry an inactivating Gln269Pro mutation in the extracellular domain of the Ob-Rb leptin receptor gene [17], but unlike either db/db or ob/ob mice, these rats exhibit reduced femoral bone mineral density (BMD) and calcium content [18], suggesting the possibility of rodent species or strain differences in leptin control of bone. Also, leptin action in cortical bone may be distinct from its antiosteogenic effects in the cancellous compartment, as femoral length and total body BMC are decreased in ob/ob mice and peripheral leptin treatment of immature (4-weekold) ob/ob mice partially restored both parameters [19]. As leptin can stimulate growth plate chondrocyte proliferation and differentiation in vitro [20], it is possible that the ob/ob-associated reductions in bone length and BMC may relate to growth plate rather than osteoblastic effects. A subsequent study also reported lower femur length and bone mineral content, as well as decreased mid-shaft cortical area and thickness in leptin-deficient mice; cortical thickness of lumbar vertebrae was also reduced but, paradoxically, vertebral length and lumbar BMC and BMD were increased [21]. Whether these differences between effects of leptin deficiency on cortical and trabecular compartments or between appendicular and axial skeleton may relate to effects on muscle mass, marrow adipogenesis [21] or growth plate remains to be resolved.
C. Leptin Excess and Bone in Mice The marked increase in bone formation that occurs with ablation of the leptin response in the ob/ob and db/db mice provides important information about the existence of this key inhibitory pathway and potential skeletal consequences of reduced leptin concentration, but does not necessarily predict how bone will respond to elevated serum leptin. In this respect, an initial insight was provided by the lack of change in cancellous bone volume in high-fat-fed wildtype mice despite a 50% increase in body weight and, presumably, a parallel increase in serum leptin [1]. A later study directly testing the effect of raised serum leptin concentration revealed that the magnitude of bone loss achieved with leptin excess was far less than the increase in bone resulting from leptin deficiency [22]. In lypodystrophic A-ZIP/F1 mice, a four-fold reduction in leptin was associated with a doubling of cancellous bone volume associated with an estimated 50% increase in osteoblast activity [1]. By contrast, a transgene-induced increase in serum leptin reduced cancellous bone volume and osteoblast activity by only around 20%, whether the leptin increase was four-fold or 180-fold [22] (Fig. 2). This observation suggested an upper limit to the antiosteogenic potential of leptin, perhaps due to central resistance to the hormone, which is characterized by altered transport of leptin across the blood–brain barrier [23, 24] and reduced signaling from hypothalamic leptin receptors [25]. There are reports that peripheral leptin administration is associated with reduced ovariectomy-induced bone loss in rats [26] and increased bone strength in mice [27]. The clear contrast between these findings and the bone anabolic effects due to leptin reduction in the ob/ob mouse could
Figure 2 Peripheral leptin elevation reduces cancellous bone mass. Vertebrae of 6-month-old models in which transgenic leptin overproduction in liver led to modest (A) and extreme (B) elevation of serum leptin (ng/ml). In addition to severe lipodystrophy (not shown here), cancellous bone volume (BV/TV) was significantly decreased in both models, associated with a reduction in osteoblast activity (BFR/BS) (*P < 0.05). (Adapted from Elefteriou et al. Proc. Nat. Acad. Sci. USA 101: 3258–3263, 2004.)
364
PAUL A. BALDOCK ET AL .
be explained if, at elevated levels, positive peripheral leptin effects on bone mass may override those of the central regulatory mechanism. This possibility is discussed further below.
D. Leptin Action in Human Bone 1. Leptin Deficiency and Bone in Humans
The association between leptin deficiency and elevated bone formation has been convincingly demonstrated by genetic studies in mice, but the limited genetic evidence in humans is mixed. There is evidence supporting an osteogenic effect of leptin, as one of four individuals in a family carrying a missense mutation in the leptin gene exhibited very low bone mass despite morbid obesity, but a high degree of consanguinity and multiple endocrine defects cloud interpretation of this observation [28]. In a different study that also supports an osteogenic leptin effect, administration of leptin for one year to a young girl with a leptin mutation was associated with loss of body fat and an increase in whole-body BMD [29]. These findings are inconsistent with the indirect evidence of leptin antiosteogenic function provided by an anecdotal report of advanced bone age in a single leptin-deficient child and more documented evidence of a similar effect in prepubertal patients affected by congenital generalized lipodystrophy [22]. Also supporting leptin antiosteogenic action, a Gln223Arg polymorphism in the human leptin receptor gene, while not associated with altered body weight or serum leptin, was associated with higher lumbar BMD in young men, although the effect was dependent upon genotype at the estrogen receptor α locus and linkage disequilibrium has not been excluded [30]. The encoded amino acid change, which is in the extracellular domain of the leptin receptor, has previously been associated with reduced leptin-binding affinity in carriers of the allele [31], which would be consistent with the observed increase in peak BMD in the study by Koh and colleagues. It is therefore not possible to deduce the physiological role(s) of leptin in human bone biology based on human genetic evidence. Indirect clinical evidence suggests that if there is an association between leptin deficiency and high bone mass in humans, it may not be an overriding factor in some circumstances. BMD was low after self-imposed calorie restriction in anorexia nervosa, which can lead to extremely low levels of the nutritionally dependent hormones leptin and IGF-1 and amenorrhea, suggesting that, unlike the ob/ob mice, leptin deficiency did not compensate for the skeletal consequences of low IGF-1 levels and hypogonadism in these patients [32, 33]. Reinforcing the potential importance of IGF-1 in this context, leptin treatment in
patients with hypothalamic amenorrhea due to strenuous exercise or low body weight resulted in increased levels of both IGF-1 and bone formation markers [34]. Such studies involving reduced nutritional intake, excessive exercise and disrupted menses are not, however, amenable to simple interpretation in terms of direct leptin effects on bone vs. indirect effects acting through altered reproductive and other neuroendocrine function. 2. Normal to Excess Leptin and Bone in Humans
Initial correlations between bone mass and body weight or fat mass [35] were attributed either to variations in mechanical load or estrogen production related to fat tissue mass, but after the discovery of leptin, this additional factor was also investigated. In cross-sectional studies, serum leptin levels and BMD are directly correlated in pre- and postmenopausal females by some investigators [36] but not others [37]. Furthermore, leptin and whole body bone area, as well as change in bone area over time, were correlated in one study of pubertal girls [38], a positive association between leptin level and bone mineral content was observed in a group of healthy nonobese women [39], and a weak correlation was observed in postmenopausal osteoporotic women but not the controls [40]. This uncertain relationship between leptin and bone may be genderspecific, as studies in males have reported either no association [36] or a negative association [41] between leptin and BMD. The possibility that some of these associations (or lack thereof) may relate to measurement artifacts associated with the particular technologies for measuring bone density employed in each study is a subject of ongoing discussion [35]. In this context, it is interesting that at least one study that did detect a positive association between leptin and BMD found no association between leptin and biochemical markers of bone resorption or formation, and concluded that leptin is not likely to play a significant direct role in regulating bone cell activity [42]. This interpretation is challenged, however, by direct examinations of bone cell responses to leptin in vitro. 3. Studies of Bone Cell Responses to Leptin IN VITRO
Evidence for presence of the signal-transducing Ob-Rb leptin receptor or leptin-binding sites has been detected on ossifying fetal cartilage [43], immortalized marrow stromal cells [44], chondrocytes [19, 20, 27, 45], primary osteoblasts [19, 27, 46–48], and some but not all osteosarcoma cell lines [47]. Adding further to the possibility of peripheral actions of leptin is evidence of leptin production by primary human and rat osteoblast cultures [49, 50].
Chapter 21 The Central Control of Bone Remodeling
A number of in vitro functional studies support direct leptin effects on chondrocytes and osteoblasts. Leptin has been observed to stimulate proliferation and differentiation of cultured growth plate chondrocytes [20] and to induce osteoblast differentiation and suppress adipogenesis in a human stromal cell line [44]. The hormone also has direct stimulatory effects on cell proliferation, differentiation, and function in human and rat osteoblastic cultures [27, 51]. In addition, leptin can inhibit osteoclastogenesis in vitro [26, 52]. One notable exception to these studies detected no osteoblastic expression of leptin or the leptin receptor long form, and no effect of leptin treatment on formation of mineralized nodules in primary mouse osteoblastic cultures [1]. The majority of the in vitro evidence, however, supports peripheral actions of leptin in bone biology. It therefore remains possible that, particularly at high leptin concentrations, competing interactions of opposing peripheral osteogenic and central antiosteogenic actions of leptin may in part explain the inconsistencies in the literature.
III. SYMPATHETIC NERVOUS SYSTEM A. Evidence for Sympathetic Control of Bone Remodeling by Leptin The sympathetic nervous system (SNS) is known to mediate some of the effects of leptin on body composition, with decreased sympathetic tone noted in leptin-deficient ob/ob mice [53]. Based on those observations, Takeda and colleagues investigated whether the central antiosteogenic actions of leptin might also be mediated via the sympathetic nervous system and found that dopamine β-hydroxylase (DBH)-deficient mice, which are unable to produce epinephrine and norepinephrine, the catecholamine ligands for adrenergic receptors, also have high bone mass [4]. Importantly, icv infusion of leptin failed to reduce the bone mass of DBH mice, indicating a requirement for a functional autonomic nervous system for leptin antiosteogenic actions, and identifying the sympathetic system as a neuronal mediator of leptin effects on bone [4]. This conclusion is supported by studies using β-adrenergic receptor agonists and antagonists. Administration of the β-agonist isoproterenol restored sympathetic activity in ob/ob mice and reduced bone mass in both wild-type and ob/ob mice without affecting body weight (Fig. 3) [4]. Conversely, administration of the nonselective β-adrenergic receptor antagonist propranolol increased bone mass in wild-type mice and prevented bone loss following ovariectomy, demonstrating that modulation of SNS activity affects bone remodeling (Fig. 4) [4]. Mice lacking
365 functional β2-adrenergic receptors (Adrb2-/-) have normal fat mass and fertility, accompanied by an even more marked high bone mass phenotype than ob/ob mice or wild-type mice receiving β-receptor antagonists [3]. As would be expected, icv leptin infusion into Adrb2-/- mice was unable to reduce bone mass, further demonstrating that intact and functional sympathetic nervous system signaling is required for the antiosteogenic actions of leptin [3]. Furthermore, transplantation of nonadherent bone marrow cells from wild-type mice normalized bone formation in Adrb2-/- mice, whereas the reciprocal transplant led to increased bone formation, consistent with sympathetic signaling at least in part via cells of the osteoblast lineage to control bone mass (Fig. 5) [3]. Involvement of β-adrenergic signaling in bone formation has been addressed with mixed results in rodent surgical models. In one study, treatment with propanolol increased bone strength in intact rats and enhanced bone formation in the repair of surgically introduced bone defects in rats [54]. Not all reports have been supportive of an autonomic antiosteogenic circuit, however, as administration of the specific β2-AR agonist clenbuterol prevented a reduction in mineralization of bone caused by sectioning of the sciatic nerve in rats [55]. However, in this study, the anabolic effect of clenbuterol in skeletal muscle was proportional to the inhibition of bone loss, implying an indirect effect through maintenance of loading by skeletal muscle. Consistent with the in vivo evidence for sympathetic control of bone formation, β-adrenergic receptors (β-AR) have been identified on osteoblasts and in osteoblast-like cell lines. β1- and β2-adrenergic receptors are expressed at different levels in various human osteoblast-like cell lines [56] and β2-adrenergic receptors have been identified on rat osteoblast-like cells, and in mouse primary osteoblast cultures [4, 57], suggesting that sympathetic activity might control bone formation through direct modulation of osteoblast function. Supporting this, administration of the specific β2-AR agonist, formoterol, induced cAMP levels and expression of the immediate early gene c-fos in the SaOS-2 human osteosarcoma cell line [56]. Likewise, c-fos expression was inhibited by administration of a specific β2-AR antagonist, demonstrating that β2-adrenergic receptors in osteoblasts are indeed functionally coupled to intracellular signaling pathways. Furthermore, in vitro studies have confirmed the ability of adrenergic signaling to alter bone remodeling. Administration of norepinephrine stimulated bone resorption in neonatal mouse calvariae in organ culture [57] and treatment of the MC3T3-E1 preosteoblastic mouse cell line with epinephrine stimulated expression of osteoclast differentiation factor [58], suggesting that β-adrenergic stimulation of resorption may occur via an indirect osteoblast-mediated pathway.
366
PAUL A. BALDOCK ET AL .
Pharmacological β-adrenergic agonism reduces cancellous bone mass. Proximal tibia and vertebrae from 1-month-old ob/ob (A) and wild-type (C) mice following 6 week treatment with the β-adrenergic agonist isoproterenol (3 mg/kg/day for ob/ob and 30 mg/kg/day for wild-type). Cancellous bone volume was significantly reduced in both groups (B, D) following β-agonism associated with a reduction in osteoblast activity (BFR/BS) and number (N.Ob./B.Pm.) (*P < 0.01). (Adapted from Takeda et al. Cell 111: 305–317, 2002.)
Figure 3
Figure 4 Pharmacological β-adrenergic antagonism increases cancellous bone mass. Tibia and vertebrae from 1-month-old wild-type mice following 5-week treatment with the β-adrenergic antagonist propranolol (orally, 0.4 mg/day). Cancellous bone volume (A, BV/TV) was significantly increased following β-antagonism associated with (B) an increase in osteoblast activity (BFR/BS) and number (N.Ob./B.Pm.) (*P < 0.01). (Adapted from Takeda et al. Cell 111: 305–317, 2002.)
367
Chapter 21 The Central Control of Bone Remodeling
Figure 5 Reciprocal bone marrow transplants confirm sympathetic control of bone mass. Vertebrae from 6-month-old recipient mice, collected 4 months after lethal irradiation followed by transplantation with nonadherent marrow cells from donor mice. Transplantation of bone marrow cells from wild-type mice into Adrb2-/mice, which lack β2 adrenergic receptor, reduced bone volume (BV/TV) and osteoblast activity (BFR) and surface (Ob.S/BS) to normal levels and increased urinary elimination of deoxypryidinoline (Dpd/creat.), an indicator of osteoclast function. Conversely, transplantation of Adrb2-/- marrow cells increased bone formation and decreased bone resorption in wild-type recipients (*P < 0.05). (Adapted from Elefteriou et al. Nature 434: 514–520, 2005.)
In addition to increased bone formation, Adrb2-/- mice also exhibited significantly decreased bone resorption and, importantly, this conferred a resistance to gonadectomyinduced bone loss. Furthermore, as mentioned in the previous section, icv leptin treatment failed to normalize the baseline Adrb2-/- phenotype [3]. Together with the in vitro evidence from the previous section, these observations indicate that leptin also regulates bone resorption via sympathetic signaling. Transplantation of wild-type bone marrow cells restored normal levels of bone resorption in Adrb2-/- mice, and the reciprocal transplant reduced resorption parameters, consistent with sympathetic signaling via cells of the osteoblast lineage to control bone resorption [3]. In addition, the β-agonist isoproterenol did not increase osteoclast formation in co-cultures of wild-type bone marrow macrophages with Adrb2-/- osteoblasts, although an increase did occur in co-cultures with wild-type osteoblasts. Also, isoproterenol increased expression of the osteoclast differentiation factor RANKL in cultured wild-type but not Adrb2-/- osteoblasts, via a mechanism involving the osteoblast-specific transcription factor ATF4. Further evidence indicated that sympathetic and PTH signaling target different subsets of osteoblasts, with sympathetic control acting through undifferentiated osteoblasts expressing β2-AR and ATF-4, and PTH acting
by an ATF-4-independent mechanism later in osteoblastic differentiation [3].
B. Clinical Relevance of b-adrenergic Control of Bone Beta-blocker usage to treat cardiovascular diseases is prevalent, but effects on skeletal mass and strength had not been widely investigated prior to the evidence from mouse studies cited above. The effects of β-adrenergic antagonists on bone turnover, bone mineral density (BMD), and fracture risk have now, however, been assessed in human population-based studies. β-blocker use was associated with a reduction in fracture risk and increased BMD at the hip and forearm in women over 50 years of age [59], and reduced fracture risk in women and men between 30 and 79 years of age [60], consistent with the rodent model data. However, a third observational study contradicted these findings, with β-blocker usage associated with a three-fold increase in fracture risk and reduced serum osteocalcin, an osteoblastic marker of bone formation, in perimenopausal women [61]. The difference in this latter finding may relate to variations in study design. Thus, there is some epidemiological evidence to support the hypothesis
368 that β-adrenergic signaling is involved in bone metabolism; however, placebo-controlled randomized clinical trials are necessary to more effectively assess potential therapeutic benefit from bone-specific β-adrenergic blockade in hypogonadal and age-related osteoporosis and other conditions that could benefit from stimulation of bone formation.
C. Role of CART and Melanocortin Receptors in Bone Remodeling The finding that Adrb2-/- mice were resistant to gonadectomy-induced bone loss due to lack of a bone resorption response was somewhat surprising, as this was a fundamental physiological difference from the elevated bone resorption observed in the hypogonadic ob/ob mouse model, for which gonadectomized Adrb2-/- mice had until then been considered a phenocopy [3]. This discrepancy between the two models appears to relate to another neuropeptide, CART (cocaine amphetamine-regulated transcript), a known hypothalamic target of positive leptin regulation that is low in ob/ob mice [1] but normal in Adrb2-/- [3]. Cart-/- mice displayed a low bone mass phenotype associated with increased bone resorption (Fig. 6), and also exhibited an enhanced resorptive response to icv leptin infusion, demonstrating both that
Figure 6 Cart gene deletion associated with decreased cancellous bone mass. Vertebrae from 6-month-old Cart-/- mice, displaying reduced cancellous bone volume (BV/TV) associated with increased osteoclast perimeter (Ob.Nb/BPm and Oc.S/BS) and bone resorption (Dpd/creat.), with no change in bone formation (BFR) (*P < 0.05). (Adapted from Elefteriou et al. Nature 434: 514–520, 2005.)
PAUL A. BALDOCK ET AL .
leptin-mediated sympathetic regulation of bone mass functions in the absence of CART, and that CART is an inhibitor of bone resorption [3]. CART transcripts were not detected in bone cells and in vitro analyses of Cart-/bone cells yielded no evidence of cell-autonomous CART effects, but RANKL expression was up-regulated in Cart-/- bones, implicating this osteoblastic regulator of osteoclastogenesis as a mechanistic element of the Cart-/phenotype. Earlier work from this group using chemical ablation of different regions of the hypothalamus indicated that the ventromedial hypothalamic nucleus (VMH) mediates the leptin antiosteogenic actions, with the arcuate nucleus (ARC) not involved in that regulatory circuit [4]. These more recent data implicating CART in central control of bone resorption that is independent of the sympathetic nervous system are consistent with those findings, as CART neurons are enriched in the ARC, as are neurons that produce POMC (pro-opiomelanocortin), the precursor of α-melanocyte-stimulating hormone (α-MSH) [62, 63]. After release from ARC neurons, α-MSH binds to the melanocortin-4 and -5 receptors (MC4R, MC5-R) on target neurons, thereby activating them and leading to anorexic effects on energy homeostasis [64, 65]. The Ay yellow agouti protein is a high-affinity competitive antagonist for MC4R [66], and Ay/a mice are sensitive to leptin antiosteogenic function, indicating that MC4R is not directly involved [4]. Mc4r-/- mice exhibit increased hypothalamic Cart expression and late-onset high bone mass that was associated with markedly reduced osteoclast number but normal bone formation parameters [3]. Thus, the Cart and Mc4r expression changes in Mc4r-/mice are mutually consistent and compatible with the reduction in bone resorption. In this context, it is interesting to note that markers of bone resorption are decreased in MC4R-deficient patients, lending explanation to the increased bone density previously observed in these patients [67]. Together, these observations support a model in which neural control of bone resorption may be regulated by two distinct neural mediators. Tonic inhibition by leptinsensitive VMH neurons of sympathetic signaling to immature osteoblasts via β2-adrenergic receptors induces osteoclastic differentiation, whereas CART and MC4R on arcuate neurons inhibit resorptive activity. Both these pathways appear to act on the osteoblastic cell lineage by modifying expression of the osteoclast differentiation factor, RANKL. However, the precise mechanism(s) by which CART and MC4R act is as yet unknown. Another unknown in this scenario is how the CART and MC4Rmediated circuits may relate to a tonic antiosteogenic pathway that involves neuropeptide Y responsive neurons
369
Chapter 21 The Central Control of Bone Remodeling
in the ARC. What is known about the skeletal roles of these Y receptor neurons is detailed in the following section.
IV. NEUROPEPTIDE Y AND THE Y RECEPTORS A. NPY System Ligands The neuropeptide Y (NPY) system involves the actions of three ligands, neuropeptide Y, peptide YY, and pancreatic polypeptide. These ligands are produced in specific distributions. NPY is produced by neurons of the central and peripheral nervous systems, present in many sympathetic nerve fibers, especially around blood vessels, and also in some parasympathetic nerves. NPY is a wellcharacterized vasoconstrictor in these neurons, in addition to enhancing the action of other pressor agents [68–70]. Furthermore, NPY-ergic neurons are abundant in the brain, with high levels in the hypothalamic arcuate nucleus (ARC) and ventromedial hypothalamus (VMH) [71]. Central NPY action is associated with the regulation of food intake, cardiac and respiratory activity, and the release of pituitary hormones [71, 72]. Peptide YY (PYY) is produced primarily from the endocrine L cells of the gastrointestinal tract with some expression in the pancreatic islets, and its function is related to satiety control and gastrointestinal regulation [73]. PYY expression has also been reported in brainstem neurons, although the functional significance of this localization is unknown [74]. Pancreatic polypeptide (PP) is produced by endocrine islet cells of the pancreas and regulates endocrine functions of the pancreas as well as satiety centrally [75]; to date, production of this neuropeptide has not been detected in peripheral or central neural tissue [74].
cAMP synthesis. Y1 receptors can also couple to phospholipase C to provoke release of Ca2+ from intracellular stores [79, 80]. Because of the multiplicity of Y receptors and the range of their physiological targets, receptorspecific antagonists have been developed with the goal of dissociating the various effects of NPY, with reasonable success for Y1 and Y5, but limited success for Y2 and none for Y4 [81]. To date, problems with solubility, toxicity, and inability to cross the blood–brain barrier have limited the potential utility of even the most promising nonpeptidic ligands for in vivo research and possible clinical application. Fortunately, the recent development of Y receptor knockout mouse models has enabled significant research progress, as will be evident below. Y1 and Y2 receptors are located at the post- and presynaptic terminals of the neuroeffector junction, respectively. They are both abundant in central tissue, with hypothalamic expression of Y1 predominantly in the paraventricular nucleus (PVN) and Y2 in the arcuate nucleus (ARC). Importantly, Y2 and leptin receptors are co-expressed in the ARC, which lies outside the blood–brain barrier and is therefore exposed to the peripheral circulation [69]. The Y2 receptor is predominantly presynaptic and inhibitory, and in some instances thought to act as an autoreceptor [82]. Activation of Y2 receptors modulates neurotransmitter synthesis and release, as well as production of neuropeptides. In the context of this chapter, it is particularly important that Y2 activation inhibits NPY production by Y2-expressing (and leptin receptor-expressing) NPY-ergic neurons [82]. Y4 receptor is widely distributed in central and peripheral tissues, with abundant expression in specific brain stem nuclei like the area postrema, also located outside the blood–brain barrier [69].
C. NPY in Bone Tissue B. The Y Receptors The neuropeptide Y system mediates its effects through activation of at least five different receptors: Y1, Y2, Y4, and Y5 receptors, which are present in humans and rodents, and y6, which is a pseudogene in humans but generates a functional receptor in the mouse [76, 77]. While each individual Y receptor binds NPY and PYY with equal affinities, the outright binding affinity characteristic of each receptor varies, with stronger binding for Y2 and Y1 compared to the other receptors. In contrast, the Y4 receptor has a low affinity for NPY and PYY, but the highest affinity of all the Y receptors for PP [78]. The Y receptors are seven-transmembrane receptors coupled to inhibitory G proteins, thereby mediating inhibition of
There is evidence that NPY-positive autonomic nerves are present in healthy bone tissue, particularly in association with blood vessel walls, suggestive of a role in modulation of vascular tone rather than direct effects on bone cells [83–86]. However, NPY immunoreactive nerve fibers were also detected in the synovium and bone marrow of ankle joints in arthritic rats [87]. In addition, sympathetic denervation has been observed to reduce NPY immunoreactive nerves in the periosteum. A functional study indicated the presence of NPY-responsive receptors on osteoblastic cells, with NPY treatment inhibiting the cAMP response in osteoblastic cell lines that had also been treated with PTH or norepinephrine, although NPY treatment alone did not alter basal cAMP levels [88, 89]. These findings are consistent with the known inhibitory
370
PAUL A. BALDOCK ET AL .
action of NPY on the cAMP response in other systems. There have been a few contradictory reports about NPY receptor transcripts in bone, with positive results reported for human periosteum-derived osteoblastic cells and human osteosarcoma cell lines [90], and transcripts encoding an isoform of the Y1 receptor detected in bone marrow cells and hematopoietic cell lines [91]. Others, however, have failed to detect Y receptor transcripts in primary osteoblastic cultures using RT-PCR detection [2]. It is essential that this question be resolved, given the accumulating evidence that Y receptors are involved in the regulation of bone formation, as described below.
D. Y2 Receptor Deletion Effects on Bone Because of the presence of both Y2 and leptin receptors on neurons of the arcuate nucleus [92, 93] and the emerging evidence that central leptin signaling inhibits bone
formation, a potential role for Y2 in controlling bone mass was investigated. Cancellous bone volume in Y2 receptor germline knockout mice (Y2-/-) of both genders was doubled, with increases in trabecular number and thickness [2]. Central Y2 receptors were critical for this phenotype, as a doubling of trabecular bone volume was observed only 5 weeks after conditional deletion of hypothalamic Y2 receptors from adult mice (Fig. 7). This increase in bone volume was primarily the result of greater osteoblastic activity, with the only detectable change in bone resorption a modest elevation in osteoclast number. Changes in circulating levels of insulin-like growth factor-1 (IGF-1), free T4, and testosterone were unchanged, ruling out these hypothalamo-pituitary effectors as potential mediators of the Y2-dependent bone regulatory signal [2]. Corticosterone levels were modestly but not significantly reduced [2], plasma calcium and leptin levels were unchanged, and there was no difference in Y2-/- body weight despite evidence that Y2 receptors on leptin-responsive arcuate neurons do
Figure 7 Hypothalamic Y2 receptor deletion increased cancellous bone formation. Histological sections from distal femurs of 4-month-old mice after conditional deletion of hypothalamic Y2 receptor at 3 months of age (A). Y2 floxed mice (Y2lox/lox) were injected with CRE (CRE-Y2lox/lox) or GFP (GFP-Y2lox/lox) expressing adeno-associated virus in the hypothalamus at 11 weeks of age. Five weeks later, trabecular bone volume was significantly increased (B), associated with greater bone formation rate (C) (**P < 0.001, *P < 0.05). (Adapted from Baldock et al. J Clin Invest 109: 915–921, 2002.)
371
Chapter 21 The Central Control of Bone Remodeling
modify whole-body energy homeostasis [94]. Much of the evidence therefore strongly suggested that the hypothalamic Y2-dependent antiosteogenic circuit may act through a neural mechanism.
E. Pancreatic Polypeptide and Y4 Receptor Two additional changes in Y2-/- mice that were of interest were an elevation of plasma PP concentration and increased hypothalamic NPY expression [95]. Elevation of PP in a transgenic mouse model had no effect on cortical or cancellous bone structure, effectively ruling out the increase in PP as a sole source of the Y2-/- -associated increase in bone formation [95]. This did not, however, exclude the possibility that Y4 activation in response to significantly elevated hypothalamic NPY levels in the Y2 KO mice might have stimulated osteoblast activity. As with Y2 deletion, germline Y4 receptor knockout led to increased plasma PP concentration, accompanied by a lean phenotype with decreased food intake [96], but, as for the PP transgenic mice, there was no discernible bone phenotype in Y4-/- mice [2]. Importantly, Y4 deletion did not alter hypothalamic expression of neurotransmitters involved in energy homeostasis such as NPY [95], thus providing no insights into the skeletal importance of the elevated hypothalamic NPY seen in the Y2-/- model. To address this, the bones of Y2Y4 double-mutant mice were examined, in which hypothalamic NPY expression was markedly elevated [95].
F. Y2Y4 Receptor Double Deletion Remarkably, coordinate germline deletion of both Y2 and Y4 receptors resulted in a synergistic anabolic effect, with a threefold increase in trabecular bone volume, compared with the twofold increase in Y2-/- mice (Fig. 8) [95]. This anabolic response was associated with increased bone formation rate and mineral apposition rate, as in the Y2-/- model. Importantly, this synergism between Y2 and Y4 receptors in the control of bone was only evident in male mice, although both male and female Y2-/-Y4-/mice were very lean, with body weight and white adipose tissue mass reduced compared to either Y2-/- or Y4-/- mice. Notably, plasma leptin was significantly reduced by about 60% in male mice, but not in female mice [95], which suggested that the added increase in cancellous bone volume in the Y2Y4 KO male mice might be an additive effect of both Y2 and leptin-mediated pathways. Consistent with this interpretation were the reduced, mid-diaphyseal cortical bone area and thickness in the Y2Y4 double
Figure 8
Deletion of Y2 and Y4 receptors associated with synergistic increase in cancellous bone volume. Histological sections from distal femoral metaphysis of male 4-month-old wild-type (A), Y2-/- (B), Y4-/- (C), and Y2-/-Y4-/- (D) mice. Greater trabecular bone volume was evident in Y2-/-Y4-/- compared to Y2-/-, with Y4-/- not different from wild-type (E). (*P < 0.05 vs. wild-type, # P < 0.05 vs. Y2-/-). (Adapted from Sainsbury et al. Mol Cell Biol 23: 5225–5233, 2003.)
mutants and the more marked increase in NPY transcript levels in the ARC of Y2-/-Y4-/- mice [95]. These data were consistent with a Y4-mediated pathway that is separate from but may interact with the proposed Y2 neural pathway and may act via a humoral, perhaps leptin-mediated, mechanism. This possibility is examined in greater detail in the next section.
V. INTERACTION BETWEEN LEPTIN AND Y2-REGULATED BONE ANTIOSTEOGENIC PATHWAYS The actions of leptin and the Y receptor ligand NPY are intimately linked in the hypothalamus. Deficiency of leptin or its receptor in the ob/ob and db/db mouse mutant strains, respectively, results in strongly elevated hypothalamic NPY expression. This is due to lack of leptin-induced inhibition of NPY expression and secretion, and has been shown to contribute to the massive obesity, hypercorticism, and reproductive defects characteristic of ob/ob mice [97]. NPY ablation reverses these defects, supporting a role for
372 NPY as a downstream mediator of leptin in energy homeostatic control [98]. As mentioned previously, Y2 and leptin receptors are co-expressed on NPY-ergic neurons in the arcuate nucleus, and Y2 receptor is thought to act as an autoreceptor, modulating the expression and secretion of NPY [82]. As deficiency in either the leptin or Y2 receptor results in increased NPY expression, the two receptors appear to share a common signaling pathway to regulate energy homeostasis, and could therefore also share a common centrally mediated pathway to regulate bone formation. The similarity between the high bone mass and elevated bone formation phenotypes of ob/ob and Y2-/- models, and the elevation of hypothalamic NPY expression in both models [2, 97], suggests that the high bone mass phenotype may be due to the increased NPY expression. However, continuous icv administration of NPY into wild-type mice for 28 days actually decreased bone volume, suggesting that leptin and NPY might in fact use different pathways to control bone mass and body weight [1]. One issue that was not addressed in that study, however, is that icv NPY infusion would cause an increase in total fat mass, even in pair-fed animals. This increased adiposity would consequently lead to increased leptin levels, which could have caused the antiosteogenic effects that were observed. The chemical lesion studies that support the model in which neural regulation of bone resorption may be by two distinct mediators, as mentioned earlier in this chapter, also demonstrated that distinct neuron populations are responsible for the hypothalamic control of energy homeostasis and bone mass by leptin [4]. Administration of monosodium glutamate (MSG) to ablate NPY-synthesizing neurons and arcuate nucleus (ARC) structures did not affect bone mass or leptin antiosteogenic function. Intracerebroventricular infusion of leptin into MSG-ablated mice did, however, fail to reduce body weight, indicating that, although NPY-synthesizing neurons are dispensable for leptin action on bone, they are required for leptin action on body weight. In contrast, gold thioglucose-sensitive neurons in the ventromedial hypothalamic nuclei (VMH) were required for leptin antiosteogenic function, as destruction of these neurons led to increased bone formation and was not reversed by icv infusion of leptin [4]. Therefore, it appears that the control of antiosteogenic and anorexigenic networks by leptin differs, as the ARC neuronal population responsible for regulating anorexic function is not necessary for leptin effects on bone, which instead appear to act via the neuronal populations present in the VMH [4]. Evidence for interaction between these distinct leptinand Y2-sensitive antiosteogenic pathways was sought in Y2-/-ob-/- double knockout mice [99]. Interestingly, bone
PAUL A. BALDOCK ET AL .
formation of the ob/ob model was not further enhanced by Y2 receptor deletion, although cancellous bone volume in both the Y2-/-ob-/- and ob/ob models was lower than that of Y2-/- mice due to increased bone resorption associated with the hypogonadal status of the leptin-deficient models (Fig. 9). Importantly, the levels of osteoblast activity were comparably elevated in all three models, indicating that Y2 receptor deletion did not convey additional osteoblastic stimulation in the absence of leptin, and thus suggesting that these models may share a common pathway to regulate osteoblast activity. In the presence of elevated hypothalamic NPY, however, a different result was obtained. Recombinant adeno-associated viral vector expressing NPY under the control of a neuron-specific promoter (AAV-NPY) was bilaterally injected into the hypothalamus of wild-type and Y2-/- mice, dramatically increasing hypothalamic NPY [99]. The resultant stimulation of the orexigenic pathways resulted in pronounced feeding, which doubled body weight and markedly raised plasma leptin levels in both genotypes in 3 weeks. Consistent with previous observations, NPY overexpression and leptin overproduction was associated with significantly reduced osteoblast activity in both Y2-/- and wild-type groups (Fig. 10). Importantly, however, the Y2-/- osteoblast activity remained twice that of wild-type mice, despite the elevated circulating leptin level and the overall reduction in bone formation in both mouse genotypes, indicating that action of the Y2-mediated pathway was independent of the leptin/NPY antiosteogenic pathway. The fact that the
Figure 9
Presence of leptin mutation associated with lower cancellous bone formation and higher resorption compared to Y2-deficient mice. Histomorphometric assessment of male 4-month-old wild-type, Y2-/-, ob/ob, and Y2-/-/ob mice. Cancellous bone volume (BV/TV) in ob/ob and Y2-/-/ob mice was greater than wild-type but less than Y2-/- (A). This difference was associated with greater osteoclast surface in the hypogonadal leptin-deficient ob/ob and Y2-/-/ob mice (C), rather than a difference in mineral apposition rate, which was comparably elevated relative to wild-type in all three mutant models (B). aP < 0.05 vs. wild-type, bP < 0.05 vs. ob/ob, cP < 0.05 vs. Y2/ob. (Adapted from Baldock et al. J Bone Miner Res 2005, 20, 1851–1857.)
373
Chapter 21 The Central Control of Bone Remodeling
Figure 10
Y2-associated anabolic activity maintained despite reduced bone formation accompanying hypothalamic NPY overproduction. Injection of either control or NPY-expressing adeno-associated virus (empty or NPY) into the hypothalamus resulted in a rapid increase in body weight (A) and serum leptin (B) in both wild-type (wt) and Y2-/- mice. NPY overproduction in wild-type mice led to a marked reduction in osteoblast activity, evident as decreases in osteoid width and volume (C, D), with no change in osteoblast number (E). Osteoblast activity in Y2-/- mice was also reduced in association with NPY overproduction, but osteoid parameters remained significantly elevated compared to the wild-type NPY-overproducing mice, as in empty controls. aP < 0.05 vs. wt-empty, bP < 0.05 vs. Y2-/--empty, cP < 0.05 wt-NPY vs. Y2-/−-NPY. (Adapted from Baldock et al. J. Bone. Miner. Res. 2005, in press.)
Y2-associated osteoblastic advantage was not evident in the Y2-/-/ob bones suggests the possibility that a permissive level of leptin may be necessary for detection of the Y2 pathway activity, or else that the hypothalamic Y2- and leptin-mediated antiosteogenic pathways may share a common feedback mechanism.
VI. CONCLUDING REMARKS The identification of nerve fibers within bone first led to the proposal that the regulation of bone remodeling may be influenced by neuronal mechanisms. The presence of neurotransmitters and neuropeptides and their receptors was subsequently revealed by immunohistochemical techniques, indicating the the neuro-osteogenic control of bone was likely to occur by direct mechanisms [100]. Indeed, many of the neurotransmitters and other neuropeptides present in bone, including vasoactive intestinal peptide (VIP), calcitonin gene-related protein (CGRP), glutamate, and substance P (SP) have been demonstrated to have direct effects on osteoblast and osteoclast cells in culture [101]. However, the first clues about the central control mechanisms came with the discoveries that leptin acts through a hypothalamic relay to inhibit bone formation via the sympathetic nervous system, and hypothalamic Y receptors mediate a distinct but possibly interacting antiosteogenic pathway. Experimental evidence since then has implicated other key hypothalamic neuropeptides in the regulation of bone growth, formation and resorption, which include NPY, CART, and α-MSH. There is no doubt that this list will expand rapidly as investigators strive to gain greater understanding of this recently discovered
aspect of bone physiology, but other necessary breakthroughs must come with better knowledge of the neural circuits between brain and bone, the bone cell responses to local neurotransmitter and neuropeptide release, and the extent to which interactions occur between neural, endocrine, paracrine, and autocrine regulatory mechanisms to integrate the skeletal response to physiological demands. In practical terms, greater understanding of the actions and controls on these potent central antiosteogenic and proresorptive pathways will be invaluable in the development of anabolic therapies to be used in the treatment of postmenopausal osteoporosis, other osteopenic, and orthopaedic conditions.
References 1. Ducy, P., Amling, M., Takeda, S., Priemel, M., Schilling, A. F., Beil, F. T., Shen, J., Vinson, C., Rueger, J. M., and Karsenty, G. (2000). Leptin inhibits bone formation through a hypothalamic relay: a central control of bone mass. Cell 100, 197–207. 2. Baldock, P. A., Sainsbury, A., Couzens, M., Enriquez, R. F., Thomas, G. P., Gardiner, E. M., and Herzog, H. (2002). Hypothalamic Y2 receptors regulate bone formation. J. Clin. Invest. 109, 915–921. 3. Elefteriou, F., Ahn, J. D., Takeda, S., Starbuck, M., Yang, X., Liu, X., Kondo, H., Richards, W. G., Bannon, T. W., Noda, M., Clement, K., Vaisse, C., and Karsenty, G. (2005). Leptin regulation of bone resorption by the sympathetic nervous system and CART. Nature 434, 514–520. 4. Takeda, S., Elefteriou, F., Levasseur, R., Liu, X., Zhao, L., Parker, K. L., Armstrong, D., Ducy, P., and Karsenty, G. (2002). Leptin regulates bone formation via the sympathetic nervous system. Cell 111, 305–317. 5. Tartaglia, L. A., Dembski, M., Weng, X., Deng, N., Culpepper, J., Devos, R., Richards, G. J., Campfield, L. A., Clark, F. T., Deeds, J., Muir, C., Sanker, S., Moriarty, A., Moore, K. J., Smutko, J. S., Mays, G. G., Wool, E. A., Monroe, C. A., and Tepper, R. I. (1995).
374
6.
7. 8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20. 21.
22.
23.
Identification and expression cloning of a leptin receptor, OB-R. Cell 83, 1263–1271. Halaas, J. L., Gajiwala, K. S., Maffei, M., Cohen, S. L., Chait, B. T., Rabinowitz, D., Lallone, R. L., Burley, S. K., and Friedman, J. M. (1995). Weight-reducing effects of the plasma protein encoded by the obese gene. Science 269, 543–546. Friedman, J. M., and Halaas, J. L. (1998). Leptin and the regulation of body weight in mammals. Nature 395, 763–770. Campfield, L. A., Smith, F. J., Guisez, Y., Devos, R., and Burn, P. (1995). Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 269, 546–549. Pelleymounter, M. A., Cullen, M. J., Baker, M. B., Hecht, R., Winters, D., Boone, T., and Collins, F. (1995). Effects of the obese gene product on body weight regulation in ob/ob mice. Science 269, 540–543. Chehab, F. F., Lim, M. E., and Lu, R. (1996). Correction of the sterility defect in homozygous obese female mice by treatment with the human recombinant leptin. Nat. Genet. 12, 318–320. Fei, H., Okano, H. J., Li, C., Lee, G. H., Zhao, C., Darnell, R., and Friedman, J. M. (1997). Anatomic localization of alternatively spliced leptin receptors (Ob-R) in mouse brain and other tissues. Proc. Natl Acad. Sci. USA 94, 7001–7005. Lee, G., Proenca, R., Montez, J. M., Carroll, K. M., Darvishzadeh, J. G., Lee, J. I., and Friedman, J. M. (1996). Abnormal splicing of the leptin receptor in diabetic mice. Nature 379, 632–635. Chen, H., Charlat, O., Tartaglia, L. A., Woolf, E. A., Weng, X., Ellis, S. J., Lakey, N. D., Culpepper, J., Moore, K. J., Breitbart, R. E., Duyk, G. M., Tepper, R. I., and Morgenstern, J. P. (1996). Evidence that the diabetes gene encodes the leptin receptor: identification of a mutation in the leptin receptor gene in db/db mice. Cell 84, 491–495. Vaisse, C., Halaas, J. L., Horvath, C. M., Darnell, J., Stoffel, M., and Friedman, J. M. (1996). Leptin activation of Stat3 in the hypothalamus of wild-type and ob/ob mice but not db/db mice. Nat. Genet. 14, 95–97. Cohen, P., Zhao, C., Cai, X., Montez, J. M., Rohani, S. C., Feinstein, P., Mombaerts, P., and Friedman, J. M. (2001). Selective deletion of leptin receptor in neurons leads to obesity. J. Clin. Invest. 108, 1113–1121. Kveiborg, M., Chiusaroli, R., Sims, N. A., Wu, M., Sabatakos, G., Horne, W. C., and Baron, R. (2002). The increased bone mass in deltaFosB transgenic mice is independent of circulating leptin levels. Endocrinology 143, 4304–4309. Takaya, K., Ogawa, Y., Isse, N., Okazaki, T., Satoh, N., Masuzaki, H., Mori, K., Tamura, N., Hosoda, K., and Nakao, K. (1996). Molecular cloning of rat leptin receptor isoform complementary DNAs– identification of a missense mutation in Zucker fatty (fa/fa) rats. Biochem. Biophys. Res. Commun. 225, 75–83. Mathey, J., Horcajada-Molteni, M. N., Chanteranne, B., Picherit, C., Puel, C., Lebecque, P., Cubizoles, C., Davicco, M. J., Coxam, V., and Barlet, J. P. (2002). Bone mass in obese diabetic Zucker rats: influence of treadmill running. Calcif. Tissue Int. 70, 305–311. Steppan, C. M., Crawford, D. T., Chidsey-Frink, K. L., Ke, H., and Swick, A. G. (2000). Leptin is a potent stimulator of bone growth in ob/ob mice. Regul. Pept. 92, 73–78. Nakajima, R., Inada, H., Koike, T., and Yamano, T. (2003). Effects of leptin to cultured growth plate chondrocytes. Horm. Res. 60, 91–98. Hamrick, M. W., Pennington, C., Newton, D., Xie, D., and Isales, C. (2004). Leptin deficiency produces contrasting phenotypes in bones of the limb and spine. Bone 34, 376–383. Elefteriou, F., Takeda, S., Ebihara, K., Magre, J., Patano, N., Kim, C. A., Ogawa, Y., Liu, X., Ware, S. M., Craigen, W. J., Robert, J. J., Vinson, C., Nakao, K., Capeau, J., and Karsenty, G. (2004). Serum leptin level is a regulator of bone mass. Proc. Natl Acad. Sci. USA 101, 3258–3263. Oh-I, S., Shimizu, H., Sato, T., Uehara, Y., Okada, S., and Mori, M. (2005). Molecular mechanisms associated with leptin
PAUL A. BALDOCK ET AL .
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35. 36.
37.
38.
39.
resistance: n-3 polyunsaturated fatty acids induce alterations in the tight junction of the brain. Cell Metab. 1, 331–341. Caro, J. F., Kolaczynski, J. W., Nyce, M. R., Ohannesian, J. P., Opentanova, I., Goldman, W. H., Lynn, R. B., Zhang, P. L., Sinha, M. K., and Considine, R. V. (1996). Decreased cerebrospinal-fluid/serum leptin ratio in obesity: a possible mechanism for leptin resistance. Lancet 348, 159–161. Munzberg, H., Flier, J. S., and Bjorbaek, C. (2004). Region-specific leptin resistance within the hypothalamus of diet-induced obese mice. Endocrinology 145, 4880–4889. Burguera, B., Hofbauer, L. C., Thomas, T., Gori, F., Evans, G. L., Khosla, S., Riggs, B. L., and Turner, R. T. (2001). Leptin reduces ovariectomy-induced bone loss in rats. Endocrinology 142, 3546–3553. Cornish, J., Callon, K. E., Bava, U., Lin, C., Naot, D., Hill, B. L., Grey, A. B., Broom, N., Myers, D. E., Nicholson, G. C., and Reid, I. R. (2002). Leptin directly regulates bone cell function in vitro and reduces bone fragility in vivo. J. Endocrin. 175, 405–415. Ozata, M., Ozdemir, I. C., and Licinio, J. (1999). Human leptin deficiency caused by a missense mutation: multiple endocrine defects, decreased sympathetic tone, and immune system dysfunction indicate new targets for leptin action, greater central than peripheral resistance to the effects of leptin, and spontaneous correction of leptin-mediated defects. J. Clin. Endocrinol. Metab. 84, 3686–3695. Farooqi, I. S., Jebb, S. A., Langmack, G., Lawrence, E., Cheetham, C. H., Prentice, A. M., Hughes, I. A., McCamish, M. A., and O’Rahilly, S. (1999). Effects of recombinant leptin therapy in a child with congenital leptin deficiency. New Engl. J. Med. 341, 879–884. Koh, J. M., Kim, D. J., Hong, J. S., Park, J. Y., Lee, K. U., Kim, S. Y., and Kim, G. S. (2002). Estrogen receptor alpha gene polymorphisms Pvu II and Xba I influence association between leptin receptor gene polymorphism (Gln223Arg) and bone mineral density in young men., Eur. J. Endocrinol. 147, 777–783. Quinton, N. D., Lee, A. J., Ross, R. J., Eastell, R., and Blakemore, A. I. (2001). A single nucleotide polymorphism (SNP) in the leptin receptor is associated with BMI, fat mass and leptin levels in postmenopausal Caucasian women. Hum. Genet. 108, 233–236. Miller, K. K., Grinspoon, S., Gleysteen, S., Grieco, K. A., Ciampa, J., Breu, J., Herzog, D. B., and Klibanski, A. (2004). Preservation of neuroendocrine control of reproductive function despite severe undernutrition. J. Clin. Endocrinol. Metab. 89, 4434–4438. Soyka, L. A., Grinspoon, S., Levitsky, L. L., Herzog, D. B., and Klibanski, A. (1999). The effects of anorexia nervosa on bone metabolism in female adolescents. J. Clin. Endocrinol. Metab. 84, 4489–4496. Welt, C. K., Chan, J. L., Bullen, J., Murphy, R., Smith, P., DePaoli, A. M., Karalis, A., and Mantzoros, C. S. (2004). Recombinant human leptin in women with hypothalamic amenorrhea. New Engl. J. Med. 351, 987–997. Reid, I. R. (2002). Relationships among body mass, its components, and bone. Bone 31, 547–555. Thomas, T., Burguera, B., Melton, L. J., Atkinson, E. J., O’Fallon, W. M., Riggs, B. L., and Khosla, S. (2001). Role of serum leptin, insulin, and estrogen levels as potential mediators of the relationship between fat mass and bone mineral density in men versus women. Bone 29, 114–120. Rauch, F., Blum, W. F., Klein, K., Allolio, B., and Schonau, E. (1998). Does leptin have an effect on bone in adult women? Calcif. Tissue Int. 63, 453–455. Matkovic, V., Ilich, J. Z., Skugor, M., Badenhop, N. E., Goel, P., Clairmont, A., Klisovic, D., Nahhas, R. W., and Landoll, J. D. (1997). Leptin is inversely related to age at menarche in human females. J. Clin. Endocrinol. Metab. 82, 3239–3245. Pasco, J. A., Henry, M. J., Kotowicz, M. A., Collier, G. R., Ball, M. J., Ugoni, A. M., and Nicholson, G. C. (2001). Serum leptin levels are associated with bone mass in nonobese women. J. Clin. Endocrinol. Metab. 86, 1884–1887.
Chapter 21 The Central Control of Bone Remodeling 40. Odabasi, E., Ozata, M., Turan, M., Bingol, N., Yonem, A., Cakir, B., Kutlu, M., and Ozdemir, I. C. (2000). Plasma leptin concentrations in postmenopausal women with osteoporosis. Eur. J. Endocrinol. 142, 170–173. 41. Sato, M., Takeda, N., Sarui, H., Takami, R., Takami, K., Hayashi, M., Sasaki, A., Kawachi, S., Yoshino, K., and Yasuda, K. (2001). Association between serum leptin concentrations and bone mineral density, and biochemical markers of bone turnover in adult men. J. Clin. Endocrinol. Metab. 86, 5273–5276. 42. Goulding, A., and Taylor, R. W. (1998). Plasma leptin values in relation to bone mass and density and to dynamic biochemical markers of bone resorption and formation in postmenopausal women. Calcif. Tissue Int. 63, 456–458. 43. Hoggard, N., Hunter, L., Duncan, J. S., Williams, L. M., Trayhurn, P., and Mercer, J. G. (1997). Leptin and leptin receptor mRNA and protein expression in the murine fetus and placenta. Proc. Natl Acad. Sci. USA 94, 11073–11078. 44. Thomas, T., Gori, F., Khosla, S., Jensen, M. D., Burguera, B., and Riggs, B. L. (1999). Leptin acts on human marrow stromal cells to enhance differentiation to osteoblasts and to inhibit differentiation to adipocytes. Endocrinology 140, 1360–1368. 45. Maor, G., Rochwerger, M., Segev, Y., and Phillip, M. (2002). Leptin acts as a growth factor on the chondrocytes of skeletal growth centers. J. Bone Miner. Res. 17, 1034–1043. 46. Enjuanes, A., Supervia, A., Nogues, X., and Diez-Perez, A. (2002). Leptin receptor (OB-R) gene expression in human primary osteoblasts: confirmation. J. Bone Miner. Res. 17, 1135. 47. Reseland, J. E., Syversen, U., Bakke, I., Qvigstad, G., Eide, L. G., Hjertner, O., Gordeladze, J. O., and Drevon, C. A. (2001). Leptin is expressed in and secreted from primary cultures of human osteoblasts and promotes bone mineralization. J. Bone Miner. Res. 16, 1426–1433. 48. Lee, Y. J., Park, J. H., Ju, S. K., You, K. H., Ko, J. S. and Kim, H. M. (2002). Leptin receptor isoform expression in rat osteoblasts and their functional analysis. FEBS Lett. 528, 43–47. 49. Morroni, M., De Matteis, R., Palumbo, C., Ferretti, M., Villa, I., Rubinacci, A., Cinti, S., and Marotti, G. (2004). In vivo leptin expression in cartilage and bone cells of growing rats and adult humans. J. Anat. 205, 291–296. 50. Reseland, J. E., Syversen, U., Bakke, I., Qvigstad, G., Eide, L. G., Hjertner, O., Gordeladze, J. O., and Drevon, C. A. (2002). Leptin is expressed in and secreted from primary cultures of human osteoblasts and promotes bone mineralization. J. Bone Miner. Res. 16, 1426–1433. 51. Gordeladze, J. O., Drevon, C. A., Syversen, U., and Reseland, J. E. (2002). Leptin stimulates human osteoblastic cell proliferation, de novo collagen synthesis, and mineralization: Impact on differentiation markers, apoptosis, and osteoclastic signaling. J. Cell Biochem. 85, 825–836. 52. Holloway, W. R., Collier, F. M., Aitken, C. J., Myers, D. E., Hodge, J. M., Malakellis, M., Gough, T. J., Collier, G. R., and Nicholson, G. C. (2002). Leptin inhibits osteoclast generation. J. Bone Miner. Res. 17, 200–209. 53. Satoh, N., Ogawa, Y., Katsuura, G., Numata, Y., Tsuji, T., Hayase, M., Ebihara, K., Masuzaki, H., Hosoda, K., Yoshimasa, Y., and Nakao, K. (1999). Sympathetic activation of leptin via the ventromedial hypothalamus: leptin-induced increase in catecholamine secretion. Diabetes 48, 1787–1793. 54. Minkowitz, B., Boskey, A. L., Lane, J. M., Pearlman, H. S., and Vigorita, V. J. (1991). Effects of propranolol on bone metabolism in the rat. J. Orthop. Res. 9, 869–875. 55. Zeman, R. J., Hirschman, A., Hirschman, M. L., Guo, G., and Etlinger, J. D. (1991). Clenbuterol, a beta 2-receptor agonist, reduces net bone loss in denervated hindlimbs. Am. J. Physiol. 261, E285–289. 56. Kellenberger, S., Muller, K., Richener, H., and Bilbe, G. (1998). Formoterol and isoproterenol induce c-fos gene expression in osteoblastlike cells by activating beta2-adrenergic receptors. Bone 22, 471–478.
375 57. Moore, R. E., Smith, C. K., Bailey, C. S., Voelkel, E. F., and Tashjian, A. H. (1993). Characterization of beta-adrenergic receptors on rat and human osteoblast-like cells and demonstration that betareceptor agonists can stimulate bone resorption in organ culture. Bone Miner. 23, 301–315. 58. Takeuchi, T., Tsuboi, T., Arai, M., and Togari, A. (2000). Adrenergic stimulation of osteoclastogenesis mediated by expression of osteoclast differentiation factor in MC3T3-E1 osteoblast-like cells. Biochem. Pharmacol. 61, 5. 59. Pasco, J. A., Henry, M. J., Sanders, K. M., Kotowicz, M. A., Seeman, E., and Nicholson, G. C. (2004). Beta-adrenergic blockers reduce the risk of fracture partly by increasing bone mineral density: Geelong Osteoporosis Study. J. Bone Miner. Res. 19, 19–24. 60. Schlienger, R. G., Kraenzlin, M. E., Jick, S. S., and Meier, C. R. (2004). Use of beta-blockers and risk of fractures. JAMA 292, 1326–1332. 61. Rejnmark, L., Vestergaard, P., Kassem, M., Christoffersen, B. R., Kolthoff, N., Brixen, K., and Mosekilde, L. (2004). Fracture risk in perimenopausal women treated with beta-blockers. Calcif. Tissue Int. 75, 365–372. 62. Bertagna, X. (1994). Proopiomelanocortin-derived peptides. Endocrinol. Metab. Clin. North Am. 23, 467–485. 63. Solomon, S. (1999). POMC-derived peptides and their biological action. Ann. N Y Acad. Sci. 885, 22–40. 64. Seeley, R. J., Drazen, D. L., and Clegg, D. J. (2004). The critical role of the melanocortin system in the control of energy balance. Annu. Rev. Nutr. 24, 133–149. 65. Mountjoy, K. G., Wu, C. S., Cornish, J., and Callon, K. E. (2003). alpha-MSH and desacetyl-alpha-MSH signaling through melanocortin receptors. Ann. N Y Acad. Sci. 994, 58–64. 66. Shimizu, H., Shargill, N. S., Bray, G. A., Yen, T. T., and Gesellchen, P. D. (1989). Effects of MSH on food intake, body weight and coat color of the yellow obese mouse. Life Sci. 45, 543–552. 67. Farooqi, I. S., Yeo, G. S., Keogh, J. M., Aminian, S., Jebb, S. A., Butler, G., Cheetham, T., and O’Rahilly, S. (2000). Dominant and recessive inheritance of morbid obesity associated with melanocortin 4 receptor deficiency. J. Clin. Invest. 106, 271–279. 68. Pernow, J., Ohlen, A., Hokfelt, T., Nilsson, O., and Lundberg, J. M. (1987). Neuropeptide Y: presence in perivascular noradrenergic neurons and vasoconstrictor effects on skeletal muscle blood vessels in experimental animals and man. Regul. Pept. 19, 313–324. 69. Parker, R. M., and Herzog, H. (1999). Regional distribution of Y-receptor subtype mRNAs in rat brain. Eur. J. Neurosci. 11, 1431–1448. 70. Morris, J. L. (1994). Selective constriction of small cutaneous arteries by NPY matches distribution of NPY in sympathetic axons. Regul. Pept. 49, 225–236. 71. Hokfelt, T., Broberger, C., Zhang, X., Diez, M., Kopp, J., Xu, Z., Landry, M., Bao, L., Schalling, M., Koistinaho, J., DeArmond, S. J., Prusiner, S., Gong, J., and Walsh, J. H. (1998). Neuropeptide Y: some viewpoints on a multifaceted peptide in the normal and diseased nervous system. Brain Res. Brain Res. Rev. 26, 154–166. 72. Wettstein, J. G., Earley, B., and Junien, J. L. (1995). Central nervous system pharmacology of neuropeptide Y. Pharmacol. Ther. 65, 397–414. 73. Lundberg, J. M., Terenius, L., Hokfelt, T., and Tatemoto, K. (1984). Comparative immunohistochemical and biochemical analysis of pancreatic polypeptide-like peptides with special reference to presence of neuropeptide Y in central and peripheral neurons. J. Neurosci. 4, 2376–2386. 74. Ekblad, E., and Sundler, F. (2002). Distribution of pancreatic polypeptide and peptide YY. Peptides 23, 251–261. 75. Ueno, N., Inui, A., Iwamoto, M., Kaga, T., Asakawa, A., Okita, M., Fujimiya, M., Nakajima, Y., Ohmoto, Y., Ohnaka, M., Nakaya, Y., Miyazaki, J. I., and Kasuga, M. (1999). Decreased food intake and body weight in pancreatic polypeptide-overexpressing mice. Gastroenterology 117, 1427–1432.
376 76. Starback, P., Wraith, A., Eriksson, H., and Larhammar, D. (2000). Neuropeptide Y receptor gene y6: multiple deaths or resurrections? Biochem. Biophys. Res. Commun. 277, 264–269. 77. Lin, S., Boey, D., Couzens, M., Lee, N., Sainsbury, A., and Herzog, H. (2005). Compensatory changes in [125I]-PYY binding in Y receptor knockout mice suggest the potential existence of further Y receptor(s). Neuropeptides 39, 21–28. 78. Bard, J. A., Walker, M. W., Branchek, T. A., and Weinshank, R. L. (1995). Cloning and functional expression of a human Y4 subtype receptor for pancreatic polypeptide, neuropeptide Y, and peptide YY. J. Biol. Chem. 270, 26762–26765. 79. Selbie, L. A., Darby, K., Schmitz-Peiffer, C., Browne, C. L., Herzog, H., Shine, J., and Biden, T. J. (1995). Synergistic interaction of Y1-neuropeptide Y and alpha 1b-adrenergic receptors in the regulation of phospholipase C, protein kinase C, and arachidonic acid production. J. Biol. Chem. 270, 11789–11796. 80. Herzog, H., Hort, Y. J., Ball, H. J., Hayes, G., Shine, J., and Selbie, L. A. (1992). Cloned human neuropeptide Y receptor couples to two different second messenger systems. Proc. Natl Acad. Sci. USA 89, 5794–5798. 81. Balasubramaniam, A. (2003). Neuropeptide Y (NPY) family of hormones: progress in the development of receptor selective agonists and antagonists. Curr. Pharm. Des. 9, 1165–1175. 82. King, P. J., Williams, G., Doods, H., and Widdowson, P. S. (2000). Effect of a selective neuropeptide Y Y(2) receptor antagonist, BIIE0246 on neuropeptide Y release. Eur. J. Pharmacol. 396, R1–R3. 83. Bjurholm, A., Kreicbergs, A., Terenius, L., Goldstein, M., and Schultzberg, M. (1988). Neuropeptide Y-, tyrosine hydroxylase- and vasoactive intestinal polypeptide-immunoreactive nerves in bone and surrounding tissues. J. Auton. Nerv Syst. 25, 119–125. 84. Ahmed, M., Bjurholm, A., Kreicbergs, A., and Schultzberg, M. (1993). Neuropeptide Y, tyrosine hydroxylase and vasoactive intestinal polypeptide-immunoreactive nerve fibers in the vertebral bodies, discs, dura mater, and spinal ligaments of the rat lumbar spine. Spine 18, 268–273. 85. Lindblad, B. E., Nielsen, L. B., Jespersen, S. M., Bjurholm, A., Bunger, C., and Hansen, E. S. (1994). Vasoconstrictive action of neuropeptide Y in bone. The porcine tibia perfused in vivo. Acta. Orthop. Scand. 65, 629–634. 86. Sisask, G., Bjurholm, A., Ahmed, M., and Kreicbergs, A. (1996). The development of autonomic innervation in bone and joints of the rat. J. Auton. Nerv. Syst. 59, 27–33. 87. Ahmed, M., Bjurholm, A., Theodorsson, E., Schultzberg, M., and Kreicbergs, A. (1995). Neuropeptide Y- and vasoactive intestinal polypeptide-like immunoreactivity in adjuvant arthritis: effects of capsaicin treatment. Neuropeptides 29, 33–43. 88. Bjurholm, A., Kreicbergs, A., Schultzberg, M., and Lerner, U. H. (1992). Neuroendocrine regulation of cyclic AMP formation in osteoblastic cell lines (UMR-106–01, ROS 17/2.8, MC3T3-E1, and Saos-2) and primary bone cells. J. Bone Miner. Res. 7, 1011–1019.
PAUL A. BALDOCK ET AL . 89. Bjurholm, A. (1991). Neuroendocrine peptides in bone. Int. Orthop. 15, 325–329. 90. Togari, A., Arai, M., Mizutani, S., Mizutani, S., Koshihara, Y., and Nagatsu, T. (1997). Expression of mRNAs for neuropeptide receptors and beta-adrenergic receptors in human osteoblasts and human osteogenic sarcoma cells. Neurosci. Lett. 233, 125–128. 91. Nakamura, M., Sakanaka, C., Aoki, Y., Ogasawara, H., Tsuji, T., Kodama, H., Matsumoto, T., Shimizu, T., and Noma, M. (1995). Identification of two isoforms of mouse neuropeptide Y-Y1 receptor generated by alternative splicing. Isolation, genomic structure, and functional expression of the receptors. J. Biol. Chem. 270, 30102–30110. 92. Baskin, D. G., Breininger, J. F., and Schwartz, M. W. (1999). Leptin receptor mRNA identifies a subpopulation of neuropeptide Y neurons activated by fasting in rat hypothalamus. Diabetes 48, 828–833. 93. Broberger, C., Landry, M., Wong, H., Walsh, J. N., and Hokfelt, T. (1997). Subtypes Y1 and Y2 of the neuropeptide Y receptor are respectively expressed in pro-opiomelanocortin- and neuropeptide-Ycontaining neurons of the rat hypothalamic arcuate nucleus. Neuroendocrinology 66, 393–408. 94. Sainsbury, A., Schwarzer, C., Couzens, M., Fetissov, S., Furtinger, S., Jenkins, A., Cox, H. M., Sperk, G., Hokfelt, T., and Herzog, H. (2002). Important role of hypothalamic Y2 receptors in body weight regulation revealed in conditional knockout mice. Proc. Natl Acad. Sci. USA 99, 8938–8943. 95. Sainsbury, A., Baldock, P. A., Schwarzer, C., Ueno, N., Enriquez, R. F., Couzens, M., Inui, A., Herzog, H., and Gardiner, E. M. (2003). Synergistic effects of Y2 and Y4 receptors on adiposity and bone mass revealed in double knockout mice. Mol. Cell Biol. 23, 5225–5233. 96. Sainsbury, A., Schwarzer, C., Couzens, M., Jenkins, A., Oakes, S. R., Ormandy, C. J., and Herzog, H. (2002). Y4 receptor knockout rescues fertility in ob/ob mice. Genes Dev. 16, 1077–1088. 97. Stephens, T. W., Basinski, M., Bristow, P. K., Bue-Valleskey, J. M., Burgett, S. G., Craft, L., Hale, J., Hoffmann, J., Hsiung, H. M., Kriauciunas, A., MacKellar, W., Rosteck, P. R. Jr., Schoner, B., Smith, D., Tinsley, F. C., Zhang, X., and Heiman, M. (1995). The role of neuropeptide Y in the antiobesity action of the obese gene product. Nature 377, 530–532. 98. Erickson, J. C., Hollopeter, G., and Palmiter, R. D. (1996). Attenuation of the obesity syndrome of ob/ob mice by the loss of neuropeptide Y. Science 274, 1704–1707. 99. Baldock, P. A., Sainsbury, A., Allison, S., Lin, D., Couzens, M., Boey, D., Enriquez, R. F., During, M., Herzog, H., and Gardiner, E. M. (2005). Hypothalamic control of bone formation: Distinct actions of leptin and Y2 receptor pathways. J. Bone Miner. Res. 20, 1851–1857. 100. Konttinen, Y., Imai, S., and Suda, A. (1996). Neuropeptides and the puzzle of bone remodeling. State of the art. Acta Orthop. Scand. 67, 632–639. 101. Lerner, U.H., and Lundberg, P. (2002). Kinins and Neuro-osteogenic factors. In Principles of Bone Biology, G. A. Rodan (ed.). Academic Press, San Diego, 1, 773–799.
Chapter 22
New Concepts in Bone Remodeling David W. Dempster and Hua Zhou
I. II. III. IV.
V. Possible Mechanisms whereby a Reduction in Activation Frequency may Protect against Fracture References
Introduction An Overview of the Remodeling Cycle Functions of Bone Remodeling The Role of Apoptosis in Regulating Bone Balance
II. AN OVERVIEW OF THE REMODELING CYCLE
I. INTRODUCTION One of the many remarkable features of the mammalian skeleton is its ability to constantly renew itself throughout life. This is achieved by means of the concerted efforts of a diverse group of cells, collectively referred to as the bone remodeling unit (BRU), which includes osteoclasts, osteoblasts, osteocytes, and other accessory cells whose identity and functions are still obscure. Our current knowledge of the mechanics of bone remodeling has been reviewed in detail elsewhere [1–5] and will be considered only briefly here. The purpose of this chapter is to review some new concepts pertaining to the activation of BRUs, intercellular communication within the BRU, and the role of apoptosis in regulating the bone balance within each BRU and, ultimately, bone mass. We shall also consider why a reduction in remodeling activation frequency in osteoporotic patients may reduce the risk of fracture without substantial increments in bone mass. Dynamics of Bone and Cartilage Metabolism
Regional Bone Center, Helen Hayes Hospital, West Haverstraw, New York, USA
There are four distinct phases in the remodeling cycle: activation, resorption, reversal, and formation (Figs 1 and 2). Activation refers to the initiating event that converts a previously quiescent bone surface into a remodeling one. The main processes involved are recruitment of mononucleated osteoclast precursors from cells in the monocyte– macrophage lineage in the circulation [6], penetration of the bone lining cell layer, and fusion of the mononuclear cells to form multinucleated preosteoclasts. The extent to which the activation process is deterministic (purposeful) and the extent to which it is stochastic (random) is unclear. However, as will be discussed below, there is evidence that at least some remodeling is targeted towards the repair of fatigue damage, but most remodeling is probably stochastic and regulated by hormones [7, 8]. The formation, activation, and activity of osteoclasts are all regulated by local 377
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
378
Figure 1
Cross-sectional diagrams of evolving BRU in cancellous bone (upper) and cortical bone (lower). The arrow indicates the direction of movement through space. Note that the cancellous BRU is essentially one half of the cortical BRU.
Figure 2
Scanning electron micrograph of a BRU in human cancellous bone. Note the eroded surface (ES) underlying an area of new bone formation (BF), and the forming osteocyte lacunae (arrowheads) in the area of new bone formation.
DAVID W. DEMPSTER AND HUA ZHOU
cytokines such as receptor activator of nuclear factor-κB ligand (RANKL), interleukins-1 and -6 (IL-1 and IL-6), colony-stimulating factors (CSFs), and systemic hormones such as parathyroid hormone, 1,25-dihydroxyvitamin D3 and calcitonin [6, 9–12]. Osteoclasts are ultimately activated by membrane signaling through RANK, the cognate receptor for RANKL, a member of the family of tumor necrosis factors expressed on marrow stromal cells and osteoblasts. The RANKL/RANK interaction promotes the differentiation of osteoclast precursors and results in the activation and prolonged survival of mature osteoclasts. Osteoprotegerin (OPG), a member of the superfamily of tumor necrosis factor receptors, is a decoy receptor for RANKL. The binding of RANKL to OPG results in inhibition of the differentiation and activity of osteoclasts. The preosteoclasts bind tightly to the bone matrix via interaction between integrin receptors in their membranes and RGD-containing peptides in the organic matrix, creating an annular sealing zone. The resorption phase of the cycle begins when fully differentiated osteoclasts secrete protons and proteolytic enzymes into the hemivacuole created by the sealing zone (Fig. 3). The proteolytic enzymes are primarily cathepsins, the most important one being cathepsin K [13]. Initially, resorption is accomplished by multinucleated osteoclasts, but there is some evidence to suggest that the job is completed by mononucleated cells [14]. Recent in vitro studies have confirmed that mononuclear cells derived from human peripheral monocytes are capable of extensive resorption [15]. The ending of the resorption phase, signaled by apoptosis of osteoclasts [16], is followed by reversal. The mechanism of osteoclast apoptosis remains unclear. Studies indicate that TGF-β, sex steroids such as estradiol and testosterone promote osteoclast apoptosis [6], and Fas/Fas ligandmediated osteoclast apoptosis [17]. During reversal, resorption lacunae are occupied by mononuclear cells, including monocytes, osteocytes that have been released from bone, and osteoblasts that are moving in to begin the formation phase of the cycle [18]. It is during the reversal phase that coupling signals, derived both from the resorbed organic matrix and osteoblast lineage cells, summon osteoblast progenitors into the resorption site and promote the proliferation and differentiation of osteoblast progenitors. One attractive feature of the hypothesis that bone matrix serves as a source of osteoblast growth factors is that it provides a mechanism whereby the number of osteoblast precursors recruited could be quantitatively related to the amount of bone removed and, therefore, the amount that has to be replaced. The nature of the signals is not completely known, but osteoblast growth factors, such as transforming growth factor (TGF)-β, insulin-like growth factors I and II (IGF-I and -II), bone morphogenetic proteins (BMPs),
379
Chapter 22 New Concepts in Bone Remodeling
Figure 3
Human osteoclast (OC) fixed in the process of excavating a resorption lacuna (RL) in vitro. Note the sharpness of the edge (arrowheads) between unresorbed and resorbed areas indicating how tightly fitting the sealing zone must be. Reproduced with permission from Reference 107.
platelet-derived growth factor (PDGF) and fibroblast growth factor (FGF), are plausible candidates [19–24]. A crucial transcription factor for osteoblast development, the core binding factor 1 (Cbfa1 or Runx2), also plays an important role in the recruitment of osteoblasts to the remodeling site by directing the commitment of multipotential mesenchymal stem cells to osteoprogenitor cells and promoting the proliferation and differentiation of osteoprogenitors to mature osteoblasts [25]. TGF-β has also recently been shown to prolong osteoblast life-span in vitro by inhibiting apoptosis (see below). That TGF-β plays an important role in bone remodeling is supported by the demonstration that the concentration of TGF-β is positively correlated with histomorphometric indices of bone resorption and bone formation and with serum levels of osteocalcin and bone-specific alkaline phosphatase [26]. Formation is a two-step process in which the osteoblasts initially synthesize the organic matrix and then preside over its mineralization. As bone formation ensues, osteoblasts become incarcerated in the matrix as osteocytes. These cells, however, are not in “solitary confinement”. They maintain close contact with each other via gap junctions between cytoplasmic processes that course through the canaliculae. They also make contact with lining cells on the inactive bone surface and with osteoblasts and osteoclasts on the remodeling bone surface via these processes.
These cell–cell connections form a complex, threedimensional web (Fig. 4), which puts the osteocytes in an ideal position to “sense” a change in the mechanical properties of the surrounding bone and to communicate this to the cells on the surface to initiate or regulate bone remodeling if it is necessary [27–29]. This ultimately results in bone structure adapting to changes in mechanical usage. The intercellular link could occur by wiring transmission (WT) and/or volume transmission (VT), these being two mechanisms of communication that have been defined in the central nervous system [30]. WT refers to signaling through synapses, gap junctions, and cell–cell adhesion molecules, such as connexin43 [31]. VT occurs by diffusion of soluble factors (hormones, cytokines, and growth factors) through the canalicular network. When formation is completed, the inactive osteoblasts become flattened and elongated and are referred to as lining cells. These cells have long been thought to regulate the flow of ions in and out of the bone extracellular fluid. Furthermore, it has recently been suggested that under certain conditions, for example stimulation by PTH or mechanical force, bone-lining cells can revert to boneforming osteoblasts [32, 33]. The newly formed piece of bone is referred to as a “basic structural unit” (BSU) of bone (Fig. 5). In cancellous bone, it is generally called a “packet” or a “trabecular osteon”; in cortical bone it is an “Haversian system” or a “cortical osteon”. Note that
380
DAVID W. DEMPSTER AND HUA ZHOU
Figure 5
Basic structural units in cancellous bone (A) and cortical bone (B). The arrowheads delineate reversal lines. Reproduced with permission from Reference 4.
Figure 4 Transmission (A) and scanning (B) electron micrographs showing osteocyte processes communicating with cells on the bone surface. Reproduced with permission from Reference 30.
III. FUNCTIONS OF BONE REMODELING
the process of bone remodeling is essentially identical in cancellous and cortical bone. The BRU in cortical bone is equivalent to two apposed cancellous BRUs (Fig. 1) [34]. The difference between the volume of bone removed by the osteoclasts and that replaced by the osteoblasts is referred to as the “bone balance”. BRUs on the periosteal surface of cortical bone produce a slightly positive bone balance so that, with aging, the periosteal circumference increases as the effect of the small positive balance in each BRU accumulates. Remodeling units on the endosteal surface of cortical bone are in negative balance so that the marrow cavity enlarges with age. Furthermore, the balance is more negative on the endosteal surface than it is on the periosteal surface and, as a result, cortical thickness decreases with age. The bone balance on cancellous surfaces is also negative, resulting in a gradual thinning of the trabecular plates with the passage of time [1].
Two principal functions of bone remodeling are known, although some have argued convincingly that there must be other functions, of which we are as yet unaware, or other reasons why the human skeleton undergoes such extensive remodeling [35]. The main functions of bone remodeling are presumed to be the preventive maintenance of mechanical strength by continuously replacing fatigued bone by new, mechanically sound bone, and, secondly, an important role in mineral homeostasis by providing access to the skeletal stores of calcium and phosphorus. While the turnover rate of 2–3% per year in cortical bone is consistent with maintenance of mechanical properties, the turnover rate in cancellous bone is much higher than would be required for this purpose, suggesting that this is driven more by its role in mineral metabolism [5]. In extreme situations, however, cortical bone can participate in mineral homeostasis. A striking example of this is the increase in cortical remodeling that occurs in Rocky Mountain deer during the antler-forming
381
Chapter 22 New Concepts in Bone Remodeling
season [36]. Cortical porosity increases dramatically, but temporarily, during this time to provide the mineral necessary for antler growth. In less extreme situations, it has been proposed [5] that activation of new remodeling units is not required for short-term mineral homeostasis, but that this may only require the presence of a minimum number of osteoclasts whose activity can be regulated. Continual remodeling activity therefore ensures the presence of this contingent of osteoclasts. Remodeling also ensures the continuous supply of young bone of low mineral density, which is necessary for ionic exchange at quiescent surfaces [5]. There are also a number of other postulated functions related to the skeleton’s role as a depository. One of these is the skeleton’s role in acid–base balance with bone serving as a source of buffer in the form of bicarbonate [37]. The skeleton is also a depot for growth factors and cytokines, some of which, for example transforming growth factor β, are activated by the acidic microenvironment created by the osteoclast. Remodeling may therefore function to supply hematopoietic marrow with these factors. This may account for some of the excess remodeling that occurs at cancellous sites where marrow and bone co-exist in a symbiotic manner [5]. The best experimental evidence for a role of bone remodeling in the maintenance of the mechanical integrity of the skeleton has come from studies in dogs [38, 39]. When long bones were loaded to induce fatigue damage and then studied histologically, a statistical association between microcracks and resorption spaces was demonstrated, with as much as six times as many cracks associated with resorption spaces as predicted by chance alone (Fig. 6) [40]. This could be explained by a propensity of microcracks to form close to resorption spaces. However, further experiments showed that the activation of remodeling was in response to the appearance of microcracks [39]. This was demonstrated by loading the left forelimbs of dogs 8 days prior to sacrifice and loading the right forelimbs immediately prior to sacrifice. If cracks simply localize at sites of resorption, then the number of cracks associated with resorption spaces should have been the same in each limb. The data showed that there was the same number of microcracks in each limb, but that the limb that was loaded first contained more resorption spaces and a greater association between microcracks and resorption than the limb that was loaded immediately prior to sacrifice. The conclusion from this observation is that microcracks cause the appearance of resorption spaces. One proposed mechanism for this deterministic remodeling is that microcracks may result in the debonding of a cortical osteon (Fig. 7) [41]. Consequently, there is a reduction in stress and strain in that aspect of the osteon, which could be “sensed” by the osteocytes and communicated to the
Figure 6 Images of a resorption space (Rc) associated with a microcrack (arrows) and another resorption space (R) that has no associated microcrack. A is a secondary electron image and B is backscattered electron image of the same sample. Reproduced with permission from Reference 38.
surface-lining cells by means of the lacuno-canalicular network. The lining cells could then initiate activation of a bone remodeling unit from the Haversian canal and this would move in the direction of the damage. Once the crack was reached the bone remodeling unit would turn to move in a longitudinal direction, thus repairing the crack.
IV. THE ROLE OF APOPTOSIS IN REGULATING BONE BALANCE One of the most interesting concepts to be applied to the study of bone biology in recent years is that of apoptosis or programmed cell death. The biological significance of programmed cell death was first recognized by Kerr, Wyllie, and Currie, who also coined the term apoptosis to describe it [42, 43]. In contrast to necrosis, cells dying by apoptosis do so in a purposeful and orderly manner;
382
DAVID W. DEMPSTER AND HUA ZHOU
Figure 7 Possible mechanism whereby microcracks may initiate their own repair. Reproduced with permission from Reference 41.
they shrink rather than swell, and do not spill their cytoplasmic contents, thus avoiding a local inflammatory response. Apoptosis plays a crucial role in the size and shape of organs and limbs during embryonic development and a failure of apoptosis is responsible for the unbridled growth of solid tumors. Similarly, apoptosis of bone cells is integral to skeletal development and physiological bone turnover [44–47]. One of the earliest descriptions of apoptosis in bone was in osteogenic cells involved in the formation of calvarial sutures [48]. The leading osteoblasts appeared
to die by apoptosis in regions where there was a failure of the normal pattern of overlap between two approaching segments of bone. In the postnatal and adult skeletons, apoptosis ultimately regulates bone volume and architecture by controlling the balance among proliferation, differentiation, and death of osteoblasts and osteoclasts [49, 50]. After the termination of bone formation, a proportion of osteoblasts undergo apoptosis and the remainder differentiate into either osteocytes or bone-lining cells. Apoptosis has also been observed in osteoclasts (Fig. 8) and osteocytes.
Figure 8 A series of transmission electron micrographs illustrating different stages of apoptosis in cultured neonatal rat osteoclasts. (A) Normal osteoclast. (B) Two osteoclasts in an early stage of apoptosis. Note cell shrinkage and compaction of nuclear chromatin against the nuclear membrane. (C) A later stage; the chromatin is condensed and fragmented and there is blebbing of the plasma membrane. These blebs contain recognizable cytoplasmic structures and are termed “apoptotic bodies”. (D) Final stage; the cell has assumed a small spherical shape and chromatin is highly condensed. Original magnification: A = ×2925; B = ×5010; C = ×5000; D = ×11 000. Reproduced with permission from Reference 46.
Chapter 22 New Concepts in Bone Remodeling
Because of its low frequency and transient nature, apoptosis is rarely observed in bone under normal conditions [44]. However, it is readily observed in the presence of agents that perturb the remodeling cycle. For example, osteoclast apoptosis is enhanced in the presence of estrogen [51, 52], selective estrogen receptor modulators (SERMs) [51, 53–55], bisphosphonates [56], and glucocorticoids [46], and is suppressed in the presence of calcitonin [57], and macrophage colony-stimulating factor [58]. Many factors regulate apoptosis in osteoblasts, including FGF, FGF2, IGF-I, IGF-II, and PDGF. Apoptosis of osteoblasts is induced by glucocorticoids [59, 60–62], human T-cell leukemia virus type I tax protein [63] and tumor necrosis factor [59, 63, 64], and is inhibited by various factors, such as transforming growth factor-β and interleukin-6-type cytokines, that are present in the bone/bone marrow environment [59]. Enhanced apoptosis of osteoclasts in response to estrogen, SERMs, and bisphosphonates may account for part of the therapeutic efficacy of these agents in the prevention and treatment of osteoporosis. With regard to estrogen, it has been suggested that a reduction in osteoclast apoptosis after menopause may contribute to the enhanced resorption and increased erosion depth that occurs at this time [44, 51]. Reducing apoptosis would extend the lifespan of individual osteoclasts, perhaps allowing them to penetrate deeper into the trabecular plates and thus shifting bone balance in a negative direction. A negative shift in bone balance would also occur in situations in which apoptosis is enhanced in osteoblasts, e.g. in the presence of glucocorticoid excess [65]. This could be an important component of the mechanism underlying the reduction in the wall thickness of trabecular packets that is observed in glucocorticoid-induced osteoporosis [66]. Osteocyte apoptosis is also enhanced by glucocorticoids [65], which may be a contributing factor to the aseptic or avascular osteonecrosis that is sometimes seen following high-dose glucocorticoid therapy. The finding that glucocorticoids enhanced apoptosis in rat osteoclasts [46] initially appeared to be at odds with their well-known action to enhance bone loss. However, several studies have revealed that, in the rat, glucocorticoid excess actually inhibits bone resorption, increases bone mass, and protects against bone loss induced by ovariectomy, dietary calcium deficiency, and immobilization [67–69]. It has been shown that apoptosis in osteocytes appears to be influenced by microdamage induced by mechanical loading [70–72]. Noble et al. and Verborgt et al. found that large-scale osteocyte apoptosis was associated with cortical remodeling induced by mechanical loading in a rat model of fatigue fracture [70–74]. Furthermore, Bronckers et al. [75] observed enhanced osteocyte apoptosis in deeper
383 layers of bone, i.e. in older cells. This was mainly seen in locations where intense bone resorption was occurring. However, mechanical loading to produce strains within the physiological range appeared to reduce osteocyte apoptosis [70]. Consequently, it has been proposed that, while undergoing apoptosis, osteocytes may transmit signals that are conveyed through the lacuno-canalicular system to enhance osteoclast recruitment or activation. Whatever the exact mechanism, it seems clear that osteocyte apoptosis plays an important role in the targeting of remodeling to replace microdamaged bone tissue with new, healthy bone.
V. POSSIBLE MECHANISMS WHEREBY A REDUCTION IN ACTIVATION FREQUENCY MAY PROTECT AGAINST FRACTURE At first glance, the postulate that a reduction in remodeling rate may reduce fracture risk seems counterintuitive if, as discussed above, we believe that an important function of remodeling is to maintain bone strength. In fact, there are relatively few data showing that reducing bone turnover results in an increase in fractures. The work that is most often cited is Flora et al.’s study [76] of the effects of etidronate in dogs, which showed an increase in spontaneous fractures at doses which severely depressed bone turnover. It should be noted, however, that this was a very dramatic inhibition of turnover, and, secondly, at these high doses, etidronate may have adversely affected mineralization, which could also increase skeletal fragility. More recently, treatment of dogs with alendronate or risedronate was shown to increase the density and length of microcracks in bone and this was associated with a decrease in the intrinsic toughness of the bone tissue [77]. However, again the doses of bisphosphonates were five times those used clinically and, despite the increase in microdamage, whole bone strength was not affected, presumably because changes in the mechanical properties of the tissue itself were compensated for by an increase in bone mass [77]. In contrast to this finding there is a growing body of data to support the view that high turnover rates are associated with increased fracture risk. Some of these data have come from two large prospective studies. The EPIDOS study [78] followed 7500 women for 22 months, and the Rotterdam study [79] followed 8000 men and women for almost 30 months. The EPIDOS study found an odds ratio (or risk of fracture) of 2.7 for patients with low bone mass and an odds ratio (2.2) that was almost as high for patients with normal bone mass who had high levels of urinary c-terminal telopeptide (CTx), a bone resorption marker.
384 Patients who had both low bone mass and high urinary CTx were at the highest risk of fracture (odds ratio of 4.8). Similar results were obtained in the Rotterdam study with a significant association between a resorption marker level and fractures in women, even when adjusted for bone mineral density at the femur [79]. It has been known for many years that the degree of reduction in bone mass is a good predictor of fracture risk but it is also true that the magnitude of the increase in bone mass upon treatment is not always a good predictor of the degree of fracture protection. One obvious example is sodium fluoride, which has been shown to produce significant increases in bone mass without affording fracture protection [80, 81]. This is due to the fact that the matrix formed in the presence of sodium fluoride is mechanically inferior [82]. From an analysis of the data from the Fracture Intervention Trial, Cummings et al. [83] found that the observed reduction in fracture incidence following treatment with alendronate was much greater than that predicted from the change in bone density alone. This was also the case for a number of other antiresorptive treatments in which the proportion of fracture protection that could be explained by the observed increase in bone mass ranged from about 70% in the case of alendronate to as low as 5% in the case of calcitonin (Fig. 9). Thus, there are now compelling data to support the hypothesis that a reduction in remodeling activation frequency reduces fracture risk by mechanisms other than those associated with an increase in bone mass. What are some of the possible mechanisms for this phenomenon? First, reducing activation frequency has the effect of reducing the size of the remodeling space, which results in only a modest increase in bone mass [2]. Perhaps more important, however, it also reduces the number of resorptive sites that are active at any one time. In a trabecular network
Figure 9
The increments in bone mass achieved by antiresorptive, or, more correctly, antiremodeling, agents account for only a part, and in some cases a small part, of the degree of fracture protection observed. Drawn from data presented in Reference 83.
DAVID W. DEMPSTER AND HUA ZHOU
that is already compromised by bone loss, the resorptive phase of the remodeling cycle may weaken crucial, but unsupported, surviving trabeculae to the point of failure without causing a significant decrease in bone mass [84]. Furthermore, high turnover rates increase the probability that resorption will occur simultaneously at more than one point along the length of such trabeculae, rendering them vulnerable to buckling and failure (Figs. 10 and 11). By analogy, a v-cut in the shaft of a thin walking stick has little effect on its mass but greatly reduces its ability to support a load, and several v-cuts have an even greater deleterious effect. In high turnover situations, the probability of trabecular plates being perforated is also increased.
Figure 10 Scanning electron micrographs of human cancellous bone. In (A) the entire circumference of a thinned trabecular rod in an osteoporotic patient rod is covered in resorption bays. Note that the rod has cracked at three points along its length. While this most likely occurred ex vivo it nevertheless indicates how fragile the rod was. In (B) a trabecular plate in a postmenopausal woman has been subjected to extensive osteoclastic resorption and has been perforated in the area between the arrowheads. A few millimeters to the left the plate has been breached by osteoclasts working from the other side (single arrowhead). Reproduced with permission from References 108 and 4, respectively.
385
Chapter 22 New Concepts in Bone Remodeling
Figure 11 Theoretical explanation for the increase in bone fragility associated with enhanced remodeling and why treatment with an agent that reduces bone turnover may be beneficial without changing bone mass significantly. In a normal subject (N) the two resorption cavities do not represent a threat to the stability of the trabecula because it is buttressed by the two horizontal trabeculae. In an osteoporotic individual (OP) the same two cavities significantly increase the chance of failure because the horizontal trabeculae have disappeared and the trabecula is thinner. Reducing the number of resorption cavities by treatment (Rx) decreases the risk of failure. Reproduced with permission from Reference 84.
Loss of vital trabeculae in this manner could dramatically reduce bone strength, again with relatively little effect on bone mass. This is supported by histomorphometric studies [85–88] showing that, when biopsies from fracture and nonfracture cases are matched for cancellous bone volume, the former show lower values for trabecular number. The importance of architecture in determining bone fragility has recently been confirmed by microcomputed tomography of the proximal femurs from hip fracture patients and controls [89]. Cancellous bone in specimens from fracture victims was found to have significantly greater anisotropy than in controls, even when the two groups were matched for bone mass. The greater anisotropy in the fracture cases was the result of preferential loss of trabeculae perpendicular to the principal loading axis. Several recent studies have shown that agents that lower fracture risk either maintain [90–92] or improve [93, 94] the microarchitecture of human cancellous bone. Secondly, higher turnover rates are accompanied by a shorter bone formation period in each remodeling cycle and also a lower mean age of the matrix. Consequently, mineralization density of the matrix is decreased. This has recently been demonstrated by microradiography of bone samples from baboons that had been ovariectomized and either left untreated or treated with alendronate (Fig. 12) [95]. The degree of mineralization was reduced in the ovariectomized animals and this was partially reversed by alendronate treatment. Furthermore, two recent studies have shown differences in the mineralization density in fracture patients compared to nonfractured controls [96, 97].
Figure 12
Microradiographs of cortical bone from control (CTRL), ovariectomized (OVX) and alendronate-treated (ALN) baboons. The darker osteons are of lower mineral density. Reproduced with permission from Reference 95.
It should be noted that the relationship between the degree of mineralization and the mechanical properties of bone is complex and that the optimal mineralization density value for bone strength has not yet been determined [98]. Increasing mineralization density up to an ash content of 60% by weight increases the bending strength of cortical bone [99], but increasing it beyond that point decreases the ability of the bone to withstand impacts. Bone that is too highly mineralized is brittle and more prone to failure because microfractures propagate more easily through the highly mineralized matrix [100]. In contrast to the studies mentioned above in ovariectomized animals, Boyde et al. [98] found that estrogen withdrawal in young women treated with gonadotrophin-releasing hormone analogs was accompanied by an increase in mineralization density, which the authors felt could contribute to increased skeletal fragility. While it remains to be determined exactly how changes in mineralization density affect bone quality in humans, one can hypothesize that, if the initial mineralization density is low, an increase due to a reduction in remodeling activation could have a beneficial effect on
386
DAVID W. DEMPSTER AND HUA ZHOU
bone strength. In support of this concept, treatment with antiresorptive drugs has been shown to reverse the decrease in mineralization density seen in postmenopausal women with osteoporosis [101–103]. A third potential mechanism for a beneficial effect of decreasing bone turnover is a reduction of the probability of a positive feedback loop being established between damage and the porosity associated with repair. Excessive repair of any structure has the potential to be self-defeating. This is particularly true if, as in bone, the repair process initially involves removal of material. As noted above, there is now good evidence that fatigue damage stimulates a remodeling response. With this in mind, it has been hypothesized [104, 105] that the possibility exists for a positive feedback loop between damage and porosity. In this postulate it is suggested that each time bone repairs a microcrack, it creates additional porosity and reduces bone mass. This reduces bone strength, resulting in more microdamage, which in turn initiates another wave of remodeling, which will result in further microdamage and so on and so forth. Martin has used a computer model to simulate this kind of scenario [104]. The model predicts that an increase in loading results in an increase in remodeling activation frequency to repair the increased damage. If loading is below a certain threshold the system is stable and reaches a new steady state. However, if the threshold is exceeded the porosity associated with increased remodeling causes the system to become unstable with damage, porosity, and strain all increasing at a very high rate and without limit. This instability has been suggested to be the theoretical equivalent of a stress fracture. One could propose, therefore, that a reduction in remodeling activation frequency would decrease the probability of this type of vicious circle being set up. In conclusion, although the study of bone remodeling began over 300 years ago with the microscopical observations of Antonie van Leeuwenhoek [106], we still have much to learn about the intricacies and regulation of this fascinating process. In recent years, clinical studies have tended to focus on bone mass, partly because it can be measured noninvasively and with relative ease. It is clear, however, that bone mass gives us only part of the picture. In order to understand how drugs, both those currently available and those under development, prevent fractures we must turn again to techniques that provide information on the remodeling cycle.
References
Acknowledgment This work was supported in part by NIH grants AR 39101. 39191 and 41331.
1. Frost, H. M. (1986). Intermediary Organization Of The Skeleton. CRC Press, Boca Raton, FL. 2. Parfitt, A. M. (2002). Physiologic and pathogenetic significance of bone histomorphometric data. In: “Disorders of Bone and Mineral Metabolism,” F. L. Coe and M. J. Favus (eds), Second Edition, pp. 469–485, Lippincott Williams and Wilkins, Philadelphia. 3. Dempster, D. W. (2002). Bone remodeling. In: “Disorders of Bone and Mineral Metabolism” F. L. Coe and M. J. Favus (eds), Second Edition, Lippincott Williams and Wilkins, Philadelphia, pp. 315–343. 4. Dempster, D. W. (1995). Bone remodeling. In: Osteoporosis: Etiology, Diagnosis, and Management, B. L. Riggs, L. J. Melton III (eds), Second Edition, Raven Press, New York, pp. 67–91. 5. Parfitt, A. M. (1995). Problems in the application of in vitro systems to the study of human bone remodeling. Calcified Tissue Int. 56 (Suppl. 1), S5–7. 6. Roodman, G. D. (1999). Cell biology of the osteoclast. Exp. Hematol. 27, 1229–1241. 7. Burr, D. B. (2002). Targeted and nontargeted remodeling. Bone 30, 2–4. 8. Parfitt, A. M. (2002). Targeted and nontargeted bone remodeling: Relationship to basic multicellular unit origination and progression. Bone 30, 5–7. 9. Roodman, G. D. (2004). Mechanisms of bone metastasis. N. Engl. J. Med. 350, 1655–1664. 10. Khosla, S. (2001). Minireview: The OPG/RANKL/RANK system. Endocrinology 142, 5050–5055. 11. Aubin, J. E., and Bonnelye, E. (2000). Osteoprotegerin and its ligand: a new paradigm for regulation of osteoclastogenesis and bone resorption. Osteoporos. Int. 11, 905–913. 12. Blair, H. C., and Athanasou, N. A. (2004). Recent advances in osteoclast biology and pathological bone resorption. Histol. Histopathol. 19, 189–199. 13. Littlewood-Evans, A., Kokubo, T., Ishibashi, O., Inaoka, T., Wlodarski, B., Gallagher, J. A., and Bilbe, G. (1997). Localization of cathepsin K in human osteoclasts by in situ hybridization and immunohistochemistry. Bone 20, 81–86. 14. Eriksen, E. F. (1986). Normal and pathological remodeling of human trabecular bone: three-dimensional reconstruction of the remodeling sequence in normals and in metabolic bone disease. Endocrine Rev. 7, 379–408. 15. Dempster, D. W., Hughes-Begos, C., Plavetic-Chee, K., Brandao-Burch, A., Cosman, F., Nieves, J., Neubort, S., Lu, S., Iida-Klein, A., Arnett, T., and Lindsay, R. (2005). Normal human osteoclasts formed from peripheral blood monocytes express PTH type 1 receptors and are stimulated by PTH in the absence of osteoblasts. J. Cell. Biochem. 95, 139–148. 16. Wright, K. R., McMillan, P. J., Hughes, D. F., and Boyce, B. F. (1995). Calcium deficiency/repletion in rats as an in vivo model for the study of osteoclast apoptosis. J. Bone Miner. Res. 10, S328. 17. Wu, X., Mckenna, M. A., Feng, X., Nagy, T. R., and Mcdonald, J. M. (2003). Osteoclast apoptosis: The role of Fas in vivo and in vitro. Endocrinology 144, 5545–5555. 18. Baron, R., Vignery, A., and Tran Van, P. (1980). The significance of lacunar erosion without osteoclasts: studies on the reversal phase of the remodeling sequence. Metab. Bone Dis. Rel. Res. 2S, 35–40. 19. Bonewald, L. F., and Mundy, G. R. (1990). Role of transforming growth factor beta in bone remodeling. Clin. Orthop. Rel. Res. 250, 261–276. 20. Mohan, S., and Baylink, D. (1991). Bone growth factors. Clin. Orthop. Rel. Res. 263, 30–48. 21. Hock, J. M., Centrella, M., and Canalis, E. (1998). Insulin-like growth factor I (IGF-I) has independent effects on bone matrix formation and cell replication. Endocrinology 122, 254–260. 22. Fiedler, J., Roderer, G., Gunther, K. P., and Brenner, R. E. (2002). BMP-2, BMP-4, and PDGF-bb stimulate chemotactic migration of
Chapter 22 New Concepts in Bone Remodeling
23.
24.
25.
26.
27. 28.
29.
30. 31.
32.
33.
34.
35. 36.
37.
38.
39. 40.
41. 42.
43.
primary human mesenchymal progenitor cells. J. Cell. Biochem. 87, 306–312. Tanaka, H., Wakisaka, A., Ogasa, H., Kawai, S., and Liang, C. T. (2003). Effects of basic fibroblast growth factor on osteoblast-related gene expression in the process of medullary bone formation induced in rat femur. J. Bone Miner. Metab. 21, 74–79. Locklin, R. M., Oreffo, R. O., and Triffitt, J. T. (1999). Effects of TGFbeta and bFGF on the differentiation of human bone marrow stromal fibroblasts. Cell Biol. Int. 23, 185–194. Lian, J. B., Javed, A., Zaidi, S. K., Lengner, C., Montecino, M., van Wijnen, A. J., Stein, J. L., and Stein, G. S. (2004). Regulatory controls for osteoblast growth and differentiation: role of Runx/Cbfa/AML factors. Crit. Rev. Eukaryot. Gene. Expr. 14, 1–41. Pfeilschifter, J., Diel, I., Scheppach, B., Bretz, A., Krempien, R., Erdmann, J., Schmidd, G., Reske, N., Bismar, H., Seck, T., Krempien, B., and Ziegler, R. (1998). Concentration of transforming growth factor beta in human bone tissue: relationship to age, menopause, bone turnover, and bone volume. J. Bone Miner. Res. 13, 716–730. Smit, T. H., and Burger, E. H. (2000). Is BMU-coupling a strain-regulated phenomenon? A finite element analysis. J. Bone Miner. Res. 15, 301–307. Huiskes, R., Ruimerman, R., van Lenthe, G. H., and Janssen, J. D. (2000). Effects of mechanical forces on maintenance and adaptation of form in trabecular bone. Nature 405, 704–706. Burger, E. H., Klein-Nulend, J., and Smit, T. H. (2003). Strain-derived canalicular fluid flow regulates osteoclast activity in a remodeling osteon – a proposal. J. Biomech. 36, 1452–1459. Marotti, G. (1996). The structure of bone tissues and the cellular control of their deposition. Ital. J. Anat. Embryol. 101, 25–79. Civitelli, R., Beyer, E. C., Warlow, P. M., Roberstson, A. J., Geist, S. T., and Steinberg, T. H. (1993). Connexin 43 mediates direct intercellular communication in human osteoblastic cell networks. J. Clin. Invest. 91, 1888–1896. Dobnig, H., and Turner, R. T. (1995). Evidence that intermittent treatment with parathyroid hormone increases bone formation in adult rats by activation of bone lining cells. Endocrinology 136, 3632–3638. Chow, J. W., Wilson, A. J., Chambers, T. J., and Fox, S. W. (1998). Mechanical loading stimulates bone formation by reactivation of bone lining cells in 13-week-old rats. J. Bone Miner Res. 13, 1760–1767. Parfitt, A. M. (1994). Osteonal and hemiosteonal remodeling: the spatial and temporal framework for signal traffic in adult bone. J. Cell. Biochem. 55, 273–276. Currey, J. D. (2002). Bones: Structure and Mechanics. Princeton University Press, Princeton, NJ. Banks, W. J., Epling, G. P., Kainer, R. A., and Davio, R. W. (1968). Antler growth and osteoporosis. I. Morphological and morphometric changes in the costal compacta during the antler growth cycle. Anat. Rec. 162, 387–398. Barzel, U. S. (1995). The skeleton as an ion exchange system: implications for the role of acid-base imbalance in the genesis of osteoporosis. J. Bone Miner. Res. 10, 1431–1436. Burr, D.B., Martin, R.B., Schaffler, M.B., and Radin, E.L. (1985). Bone remodeling in response to in vivo fatigue microdamage. J. Biomech. 18, 189–200. Mori, S., Burr, D. B. (1993). Increased intracortical remodeling following fatigue damage. Bone 14, 103–109. Burr, D. B., Forwood, M. R., Fyhrie, D. P., Martin, R. B., Schaffler, M. B., and Turner, C. H. (1997). Perspective: Bone microdamage and skeletal fragility in osteoporotic and stress fractures. J. Bone Miner. Res. 12, 6–15. Martin, R. B., and Burr, D. B. (1989). Structure, Function, and Adaptation of Compact Bone. Raven Press, New York. Kerr, J. F. R., Wyllie, A. H., and Currie, A. R. (1972). Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Brit. J. Cancer 26, 239–257. Wyllie, A. H., Kerr, J. F. R., and Currie, A. R. (1980). Cell death: the significance of apoptosis. Int. Rev. Cytol. 68, 258–306.
387 44. Parfitt, A. M., Mundy, G. R., Roodman, G. D., Hughes, D. E., and Boyce, B. F. (1996). Theoretical perspective: a new model for the regulation of bone resorption, with particular reference to the effects of bisphosphonates. J. Bone Miner. Res. 11, 150–159. 45. Boyce, B. F., Hughes, D. E., and Wright, K. R. (1997). Methods for studying cell death in bone. In: “Methods in Bone Biology,” T. R. Arnett, B. Henderson (eds), pp. 127–148, Chapman and Hall, London. 46. Dempster, D. W., Moonga, B. S., Stein, L. S., Horbert, W. R., and Antakly, T. (1997). Glucocorticoids inhibit bone resorption by isolated rat osteoclasts by enhancing apoptosis. J. Endocrinol. 154, 397–406. 47. Hughes, D. E., and Boyce, B. F. (1997). Apoptosis in bone physiology and disease. J. Clin. Pathol. Mol. Pathol. 50, 132–137. 48. Furtwangler, J. A., Hall, S. H., and Koskine-Moffett, L. K. (1985). Sutural morphogenesis in the mouse calvaria: the role of apoptosis. Acta. Anat. 124, 74–80. 49. Hock, J. M., Krishnan, V., Onyia, J. E., Bidwell, J. P., Milas, J., and Stanislaus, D. (2001). Perspective: Osteoblast apoptosis and bone turnover. J. Bone Miner. Res. 16, 975–984. 50. Weinstein, R. S., and Manolagas, S. C. (2000). Apoptosis and osteoporosis. Am. J. Med. 108, 153–164. 51. Hughes, D. E., Dai, A., Tiffee, J. C., Li, H. H., Mundy, G. R., and Boyce, B. F. (1996). Estrogen promotes apoptosis of murine osteoclasts mediated by TGF-β. Nat. Med. 2, 1132–1136. 52. Kameda, T., Mano, H., Yuasa, T., Mori, Y., Miyazawa, K., Shiokawa, M., Nakamura, Y., Hiroi, E., Hiura, K., Kameda, A., Yang, N. N., Hakeda, Y., and Kumegawa, M. (1997). Estrogen inhibits bone resorption by directly inducing apoptosis of the bone resorbing osteoclasts. J. Exp. Med. 186, 489–495. 53. Arnett, T., Lindsay, R., Kilb, L., Moonga, B. S., Spowage, M., and Dempster, D. W. (1996). Selective toxic effects of tamoxifen on osteoclasts: comparison with the effects of oestrogen. J. Endocrinol. 149, 503–508. 54. Lutton, J. D., Moonga, B. S., and Dempster, D. W. (1996). Osteoclast demise: physiological versus degenerative cell death. Experimental Physiology 81, 251–260. 55. Grasser, W. A., Pan, L. C., Thompson, D. D., and Paralkar, V. M. (1997). Common mechanism for the estrogen agonist and antagonist activities of droloxifene. J. Cell. Biochem. 65, 159–171. 56. Hughes, D. E., Wright, K. R., Uy, H. L., Sasaki, A., Yoneda, T., Roodman, G. D., Mundy, G. R., and Boyce, B. F. (1995). Bisphosphonates promote apoptosis in murine osteoclasts in vitro and in vivo. J. Bone Miner. Res. 10, 1478–1487. 57. Selander, K. S., Harkonen, P. L., Valve, E., Mönkkönen, J., Hannuniemi, R., and Väänänen, H. K. (1996). Calcitonin promotes osteoclast survival in vitro. Mol. Cell. Endocrinol. 122, 119–129. 58. Fuller, K., Owens, J. M., Jagger, C. J., Wilson, A., Moss, R., Chambers, T. J. (1993). Macrophage colony-stimulating factor stimulates survival and chemotactic behavior in isolated osteoclasts. J. Exp. Med. 178, 1733–1744. 59. Jilka, R. L., Weinstein, R. S., Bellido, T., Parfitt, A. M., Manolagas, S. C. (1998). Osteoblast programmed cell death (apoptosis): modulation by growth factors and cytokines. J. Bone Miner. Res. 13, 793–802. 60. Manusukhani, A., Bellosta, P., Sahni, M., and Basilico, C. (2000). Signaling by fibroblast growth factors (FGF) and fibroblast growth factor 2 (FGFR2)-activating mutations blocks mineralization and induces apoptosis in osteoblasts. J. Cell. Biol. 149, 1297–1308. 61. Hill, P., Tumber, A., and Meikle, M. (1997). Multiple extracellular signals promote osteoblast survival and apoptosis. Endocrinology 138, 3849–3858. 62. Gronowicz, G. A., McCarthy, M. B., Zhang, H., and Zhang, W. (2004). Insulin-like growth factor II induces apoptosis inosteoblasts. Bone 35, 621–628. 63. Kitajima, I., Nakajima, T., Imamura, T., Takasaki, I., Kawahara, K., Okano, T., Tokioka, T., Soejima, Y., Abeyama, K., and Murayama, I. (1996). Induction of apoptosis in murine clonal osteoblasts expressed
388
64.
65.
66. 67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
by human T-cell leukemia virus type I tax by NF-κB and TNF-α. J. Bone Miner. Res. 11, 200–210. Tsuboi, M., Kawakami, A., Nakashima, T., Matsuoka, N., Urayama, S., Kawabe, Y., Fujiyama, K., Kiriyama, T., Aoyagi, T., Maeda, K., and Euguchi, K. (1999). Tumor necrosis factor-alpha and interleukin-1beta increase the Fas-mediated apoptosis of human osteoblasts. J. Lab. Clin. Med. 134, 222–231. Weinstein, R. S., Jilka, R. L., Parfitt, A. M., and Manolagas, S. C. (1998). Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. J. Clin. Invest. 102, 274–282. Dempster, D.W. (1989). Bone histomorphometry in glucocorticoidinduced osteoporosis. J. Bone Min. Res. 4, 137–141. King, C. S., Weir, C. E., Gundberg, C. W., Fox, J., and Insogna, K. L. (1996). Effects of continuous glucocorticoid infusion on bone metabolism in the rat. Calcif. Tissue Int. 59, 184–191. Li, M., Shen, Y., Halloran, B. P., Baumann, B. D., Miller, K., and Wronski, T. J. (1996). Skeleton response to corticosteroid deficiency and excess in growing male rats. Bone 19, 81–88. Shen, V., Birchman, R., Liang, X. G., Wu, D. D., Lindsay, R., and Dempster, D. W. (1997). Prednisolone alone, or in combination with estrogen or dietary calcium deficiency or immobilization, inhibits bone formation but does not induce bone loss in mature rats. Bone 21, 345–351. Noble, B. S., Peet, N., Stevens, H. Y., Brabbs, A., Mosley, J. R., Reilly, G. C., Reeve, J., Skerry, T. M., and Lanyon, L. E. (2003). Mechanical loading: biphasic osteocyte survival and targeting of osteoclasts for bone destruction in rat cortical bone. Am. J. Physiol. Cell Physiol. 284, C934–C943. Verborgt, O., Tatton, N. A., Majeska, R. J., and Schaffler, M. B. (2002). Spatial distribution of Bax and Bcl-2 in osteocytes after bone fatigue: complementary roles in bone remodeling regulation? J. Bone Miner. Res. 17, 907–914. Verborgt, O., Gibson, G. J., and Schaffler, M. B. (2000). Loss of osteocyte integrity in associated with microdamage and bone remodeling after fatigue in vivo. J. Bone Miner. Res. 15, 60–67. Noble, B. S., Stevens, H., Loveridge, N., and Reeve, J. (1997). Identification of apoptotic changes in osteocytes in normal and pathological bone. Bone 20, 273–282. Ehrlich, P. J., Noble, B. S., Jessop, H. L., Stevens, H. Y., Mosley, J. R., Lanyon, L. E. The effect of in vivo mechanical loading on estrogen receptor alpha expression in rat ulnar osteocytes. J. Bone Miner. Res. 2002, 17:1646–1655. Bronckers, A. L. J. J., Goei, W., Luo, G., Karsenty, G., D’Sousa, R. N., Lyaruu, D. M., and Burger, E. H. (1996). DNA fragmentation during bone formation in neonatal rodents assessed by transferase-mediated end labeling. J. Bone Miner. Res. 11, 1281–1291. Flora, L., Hassing, G. S., Cloyd, G. G., Bevan, J. A., Parfitt, A. M., and Villanueva, A. R. (1981). The long term skeletal effects of EHDP in dogs. Metab. Bone Dis. Relat. Res. 3, 289–300. Mashiba, T., Hirano, T., Turner, C. H., Forwood, M. R., Johnston, C. C., and Burr, D. B. (2000). Suppressed bone turnover by bisphosphonates increases microdamage accumulation and reduces some biomechanical properties in dog rib. J. Bone Miner. Res. 15, 613–620. Garnero, P., Hausherr, E., Chapuy, M-C., Marcelli, C., Grandjean, H., Muller, C., Cormier, C., Breart, G., Meunier, P. J., and Delmas, P. D. (1996). Markers of bone resorption predict hip fracture in elderly women: the EPIDOS prospective study. J. Bone Miner. Res. 11, 1531–1538. van Daele, P. L. A., Seibel, M. J., Burger, H., Hofman, A., Grobbee, D. E., van Leeuwen, J. P. T. M., Birkenhäger, J. C., and Pols, H. A. P. (1996). Case-control analysis of bone resorption markers, disability and hip fracture risk: the Rotterdam study. Br. Med. J. 312, 482–483. Riggs, B. L., Hodgson, S. F., O’Fallon, W. M., Chao, E. Y. S., Wahner, H. W., Muhs, J. M., Cedel, S. L., and Melton, L. J. III (1990).
DAVID W. DEMPSTER AND HUA ZHOU
81.
82.
83.
84. 85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
Effect of fluoride treatment on the fracture rate in postmenopausal women with osteoporosis. N. Engl. J. Med. 322, 802–809. Kleerekoper, M., Peterson, E. L., Nelson, D. A., Phillips, E, Schork, M. A., Tilley, B. C., and Parfitt, A. M. (1991). A randomized trial of sodium fluoride as a treatment for postmenopausal osteoporosis. Osteoporosis Int. 1, 155–161. Sogaard, C. H., Mosekilde, L, Richards, A, and Mosekilde, L. (1994). Marked decrease in trabecular bone quality after five years of sodium fluoride therapy; assessed by biomechanical testing of iliac crest bone biopsies in osteoporotic patients. Bone 15, 393–399. Cummings, S. R., Karpf, D. B., Harris, F., Genant, H. K., Ensrud, K., LaCroix, A. Z., and Black, D. M. (2002). Improvement in spine bone density and reduction in risk of vertebral fractures during treatment with antiresorptive drugs. Am. J. Med. 112, 281–289. Parfitt, A. M. (1991). Use of bisphosphonates in the prevention of bone loss and fractures. Am. J. Med. 91, (Suppl 5B), 42S–46S. Kleerekoper, M., Villanueva, A. R., Stanciu, J., Rao, D. S., and Parfitt, A. M. (1985). The role of three dimensional trabecular microstructure in the pathogenesis of vertebral compression fractures. Calcified Tissue Int. 37, 594–597. Legrand, E., Chappard, D., Pascaretti, C., Duquenne, M., Krebs, S., Rohmer, V., Basle, M. F., and Audran, M. (2000). Trabecular bone microarchitecture, bone mineral density, and vertebral fractures in male osteoporosis. J. Bone Miner. Res. 15, 13–19. Aaron, J. E., Shore, P. A., Shore, R. C., Beneton, M., and Kanis, J. A. (2000). Trabecular architecture in women and men of similar bone mass with and without vertebral fracture: II. Three-dimensional histology. Bone 27, 277–282. Oleksik, A., Ott, S. M., Vedi, S., Bravenboer, N., Compston, J., and Lips, P. (2000). Bone structure in patients with low bone mineral density with or without vertebral fractures. J. Bone Miner. Res. 15, 1368–1375. Ciarelli, T. E., Fyhrie, D. P., Schaffler, M. B., and Goldstein, S. A. (2000). Variations in three-dimensional cancellous bone architecture of the proximal femur in female hip fractures and controls. J. Bone Miner. Res. 15, 32–40. Borah, B., Dufresne, T. E., Chmielewski, P. A., Johnson, T. D., Chines, A., Manhart, M. D. (2004). Risedronate preserves bone architecture in postmenopausal women with osteoporosis as measured by threedimensional microcomputed tomography. Bone 34, 736–746. Dufresne, T. E., Chmielewski, P. A., Manhart, M. D., Johnson, T. D., and Borah, B. (2003). Risedronate preserves bone architecture in early postmenopausal women in 1 year as measured by three-dimensional microcomputed tomography. Calcif. Tissue Int. 73, 423–432. Recker, R., Masarachia, P., Santora, A., Howard, T., Chavassieux, P., Arlot, M., Rodan, G., Wehren, L., Kimmel, D. (2005). Trabecular bone microarchitecture after alendronate treatment of osteoporotic women. Curr. Med. Res. Opin. 21, 185–194. Dempster, D. W., Cosman, F., Kurland, E. S., Zhou, H., Nieves, J., Woelfert, L., Shane, E., Plavetic, K., Müller, R., Bilezikian, J. P., and Lindsay, R. (2001). Effects of daily treatment with parathyroid hormone on bone microarchitecture and turnover in patients with osteoporosis: a paired biopsy study. J. Bone Miner. Res. 16, 1846–1853. Jiang, Y., Zhao, J. J., Mitlak, B. H., Wang, O., Genant, H. K., and Eriksen, E. F. (2003). Recombinant human parathyroid hormone (1–34) [teriparatide] improves both cortical and cancellous bone structure. J. Bone Miner. Res. 18, 1932–1941. Meunier, P. J., and Boivin, G. (1997). Bone mineral density reflects bone mass but also the degree of mineralization of bone: therapeutic implications. Bone 21, 373–377. Ciarelli, T. E., Fyhrie, D. P., and Parfitt, A. M. (2003). Effects of vertebral bone fragility and bone formation rate on the mineralization levels of cancellous bone from white females. Bone 32, 311–315. Loveridge, N., Power, J., Reeve, J., and Boyde, A. (2004). Bone mineralization density and femoral neck fragility. Bone 35, 929–941.
389
Chapter 22 New Concepts in Bone Remodeling 98. Boyde, A., Compston, J. E., Reeve, J., Bell, K. L., Noble, B. S., Jones, S. J., and Loveridge, N. (1998). Effect of estrogen suppression on the mineralization density of iliac crest biopsies in young women as assessed by backscattered electron imaging. Bone 22, 241–250. 99. Currey, J. D. (1969). The mechanical consequences of variation in the mineral content of bone. J. Biomech. 2, 1–11. 100. Currey, J. D., Brear, K., and Zioupos, P. (1995). The effects of aging and changes in the mineral content in degrading the toughness of human femora. J. Biomech. 29, 257–260. 101. Roschger, P., Rinnerthaler, S., Yates, J., Rodan, G. A., Fratzl, P., and Klaushofer, K. (2001). Alendronate increases degree and uniformity of mineralization in cancellous bone and decreases the porosity in cortical bone of osteoporotic women. Bone 29, 185–191. 102. Boivin, G., Vedi, S., Purdie, D. W., Compston, J. E., and Meunier, P. J. (2005). Influence of estrogen therapy at conventional and high doses on the degree of mineralization of iliac bone tissue: a quantitative microradiographic analysis in postmenopausal women. Bone 36, 562–567. 103. Boivin, G., Lips, P., Ott, S. M., Harper, K. D., Sarkar, S., Pinette, K. V., and Meunier, P. J. (2003). Contribution of raloxifene and calcium and
104. 105.
106.
107.
108.
vitamin D3 supplementation to the increase of the degree of mineralization of bone in postmenopausal women. J. Clin. Endocrinol. Metab. 88, 4199–4205. Martin, B. (1995). Mathematical model for repair of fatigue damage and stress fracture in osteonal bone. J. Orthop. Res. 13, 309–316. Burr, D. B., Forwood, M. R., Fyhrie, D. P., Martin, R. B., Schaffler, M. B., and Turner, C. H. (1997). Perspective: Bone microdamage and skeletal fragility in osteoporosis and stress fractures. J. Bone Miner. Res. 12, 6–15. Leeuwenhoek, A. (1678). Microscopical observations on the structure of teeth and other bones. Philos. Trans. R. Soc. Lond. 12, 1002–1003. Murrills, R. J., Shane, E., Lindsay, R., and Dempster, D. W. (1989). Bone resorption by isolated human osteoclasts in vitro: effects of calcitonin. J. Bone Miner. Res. 4, 259–268. Dempster, D. W., Shane, E. S., Horbert, W., and Lindsay, R. (1986). A simple method for correlative light and scanning electron microscopy of human iliac crest bone biopsies: qualitative observations in normal and osteoporotic subjects. J. Bone Miner. Res. 1, 15–21.
This Page Intentionally Left Blank
Chapter 23
Products of Bone Collagen Metabolism Juha Risteli and Leila Risteli
Department of Clinical Chemistry, FI-90014 University of Oulu, Oulu, Finland
I. Introduction II. Products of Bone Collagen Synthesis, the Procollagen Propeptides
III. Degradation Products of Type I Collagen IV. Closing Remarks References
I. INTRODUCTION
on chromosome 7. The basic pathways of collagen synthesis and degradation are shared by all tissues, but there are certain features that are more pronounced, either in the hard or the soft tissues, allowing some relative distinction to be made between these two sources of type I collagen metabolites. The main cells synthesizing the type I collagen in soft tissues are fibroblasts, which also produce significant amounts of type III collagen. Bone collagen is synthesized by osteoblasts, which normally do not express type III collagen [6]. However, certain established cell lines with an apparent osteoblastic nature (e.g., the human osteosarcoma line MG-63) produce large amounts of type III collagen [7]. Several other cell types (e.g., smooth muscle cells) also synthesize both type I and III collagens. They also produce several other collagens, the amounts of which are far less than those of the type I and III collagens. The assembly of type I collagen molecules into fibrils differs somewhat between bones and skin and this arrangement affects the cross-linking of the collagen molecules in collagen fibers, in particular at their carboxytermini [8]. However, the fibrillar collagens of many soft tissues, such as tendons, fasciae, vessel walls, and internal organs, contain cross-links similar to those present in bone.
Most of the collagen in the organic matrix of bone is type I collagen, which provides a well-organized and insoluble scaffold for the deposition of the mineral. Although the mineral is even more abundant than collagen, it is difficult and tedious to reliably assess its metabolism. This leaves type I collagen as the best target for quantification of the metabolic turnover in the skeleton [1–3]. Τhe synthesis of type I collagen involves the production of specific by-products that, among other things, provide the possibility of elegantly assessing the rate of the synthesis of this collagen. Another set of metabolic products is related to the degradation of this collagen, which can occur either together with the dissolution of the mineral phase or independently, e.g., in situations like rickets, when there is increased breakdown of a nonmineralized matrix [4–5]. Although most of the type I collagen in the human body is located in the skeleton, this protein is also the most abundant collagen in soft tissues. Unfortunately, there are no good quantitative estimates for the distribution of type I collagen between these two pools, either in health or in disease. In all locations the protein is encoded by the same genes, the COL1A1 on chromosome 17 and the COL1A2 Dynamics of Bone and Cartilage Metabolism
391
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
392
JUHA RISTELI AND LEILA RISTELI
The mineralization of bone also affects the maturation of bone collagen cross-links; thus many of them remain immature, divalent in nature, and a significant part of the α1-chains in the bone collagen molecules that contain potential cross-linking sites even remain uncross-linked [9]. Another difference between the soft tissues and bones is the fact that bone metabolites can be directly released into the circulation, whereas most soft tissues are first drained via lymphatics into larger vessels. Since there is much microheterogeneity in type I collagen, it is not possible using only biochemical means to predict whether a certain metabolic product, present either in the blood or in the urine, will predominantly reflect bone metabolism. More well-focused studies are still needed, e.g., experimental work on the structure and physiology of the collagens of various tissues, as well as clinical studies of different diseases. Detection of most of the metabolic products of collagen metabolism depends on the use of well-characterized immunological reagents, and proper quantification, in principle, can be achieved with a number of techniques, including RIAs, ELISAs, IRMAs, etc. However, proper immunochemical and biological validation is always necessary; it is particularly relevant if the assay is just based on a linear synthetic peptide, as the natural collagen metabolites are relatively complex structures. Conclusions based on assays not properly validated may introduce confusion in medical literature.
II. PRODUCTS OF BONE COLLAGEN SYNTHESIS, THE PROCOLLAGEN PROPEPTIDES A. Biochemical Basis The biosynthesis of fibrillar collagens is a complex chain of events that includes the release of two additional proteins, known as the propeptides of the respective procollagen. These proteins seem to be largely removed from their sites of origin without further degradation and thus offer the potential of directly measuring the rate of collagen synthesis in a manner analogous to the use of the C-peptide of proinsulin as an indicator of endogenous insulin production. Interestingly, the two proteins removed from the two ends of the rod-like collagen molecule differ from each other profoundly in their chemical natures and their further metabolic fates (see also Chapter 1). The type I collagen molecule is a long, thin, rigid rod – a shape necessary for its function as part of a collagen fiber in the tissue. The best known and most abundant form of this collagen, the “classical” type I collagen, is a
heterotrimer of two α1(I) chains and one α2(I) chain, which are wrapped around each other into the triple helix. The original gene products, the proα1(I) and proα2(I) chains of type I procollagen, are about 50% longer than the corresponding final products. The two additional, bulky domains at both ends of the molecule are usually called the amino-terminal (abbreviation PINP) and the carboxyterminal (PICP) propeptides of type I procollagen (Fig. 1) [10], despite their relatively large sizes, which are clearly outside the ordinary definition of a peptide (see Sections II.B and C). Once the molecule has reached the extracellular space, these parts are removed en bloc from the procollagen by two specific endoproteinases, the N- and C-proteinase, the latter of which is identical to BMP-1 [11]. Several functions have been proposed for the propeptide domains in the procollagen molecule (for N-propeptide, see reference [12]). The individual polypeptide chains are expressed separately, with the most amino-terminal signal peptide and propeptide parts being synthesized first, followed by the part that will form the collagen molecule itself, and followed finally by the carboxy-terminal propeptide moiety. As the chains pass into the cisternae of the endoplasmic reticulum, they are, at the same time, being post-translationally modified by several enzymes [10]. The carboxy-terminal propeptide sequences are needed for the proper selection and association of the three proαchains of type I procollagen within the cisternae of the endoplasmic reticulum. The collagen helix is formed starting from the carboxy-terminus and proceeding in the amino-terminal direction. The amino-terminal propeptide is thus the last part of the molecule to attain its native conformation, which includes an additional short collagenous triple helix within this domain. During the further transit of the procollagen, its propeptide domains are believed to prevent premature fiber formation in the cells and in the extracellular space, since there is a 1000-fold difference in solubility between procollagen and collagen. The carboxy-terminal propeptide can accomplish this on the basis of its bulky, globular form, whereas, in the amino-terminal propeptide, a similar function is probably due to its negative charge, analogous to the situation in blood fibrinogen (see Section II.C). When the carboxy-terminal propeptide has been removed, the molecules still containing the amino-terminal propeptide (so-called type I pN-collagen) can be layered onto the collagen fibrils, but just on the surface of the fibril. A delayed cleavage of the amino-terminal propeptide is believed to regulate the lateral growth of the type I collagen fibers [13]. In addition to the classical type I collagen that is most abundant both in the skeleton and in the soft tissues, there is some evidence for the existence of other molecular forms
393
Chapter 23 Products of Bone Collagen Metabolism
Figure 1
Schematic presentation of a procollagen molecule. The parts of the N-terminal propeptide are Col 1 (globular domain), Col 2 (nontriple-helical domain) and Col 3 (triple-helical domain). The nontriple-helical domains at both ends of the collagen molecule are the telopeptides. Reproduced from Prockop et al. [10], Copyright © 1979 Massachusetts Medical Society. All rights reserved.
of type I collagen. A type I α1-trimer collagen, which consists of three α1(I) chains, is the main collagen in the rare cases of osteogenesis imperfecta in which this disease is due to a genetic lack of the α2(I) chain [14]. Similar trimeric collagen has also been found in osteoarthritic bone [15]. There is evidence that some type I α1(I)-trimer collagen is also formed in people having a G to T substitution in the Sp1 binding site in the COL1A1 gene [16] (see also Chapter 30).
B. Carboxy-terminal Propeptide of Type I Procollagen (PICP) The carboxy-terminal propeptide – both in free form and as part of the type I procollagen molecule – has a globular shape and an amino acid composition of a noncollagenous protein. The parts corresponding to the propeptide are encoded by exons 49 through 52 of the COL1A1 and
Table I.
COL1A2 genes. Its overall molecular mass is about 100 000 (Table I), and those of the subunits about 33 000 each. All the three component polypeptide chains contain an N-glycosylation site, which is normally occupied by an oligosaccharide of the high-mannose type [17]. For this reason, the propeptide is bound to an affinity column of immobilized concanavalin A, a property that can be used in isolating the propeptide [18]. There are also interchain disulfide bridges between its three polypeptide chains; these are needed in the processes of chain selection and association. There also seems to be a free SH-group, since the propeptide can also be purified using an affinity column binding such groups [19]. It is relatively difficult to isolate the PICP antigen, but it has been purified from the culture media of human fibroblasts, either in the form of type I procollagen, from which the PICP has been liberated by bacterial collagenase digestion [18, 20], or directly as a propeptide cleaved in cell culture, starting from a serumfree culture medium [19].
Comparison of the Amino-terminal (PINP) and Carboxy-terminal (PICP) Propeptides of the Classical Human Type I Procollagen PINP
Location Relative molecular mass Shape Chemical nature Related serum antigen Homogeneity Size Concentration (µg/L) Clearance from blood Blocked by
PICP
Amino-terminal 35 000 Elongated Phosphorylated, partially collagenous
Carboxy-terminal 100 000 Globular Glycoprotein, oligosaccharides of the high-mannose type
One major and one minor form Same as intact PINP (major) or Col1-like (minor) Men: 20–76 Women: 19–84 Scavenger receptor of liver endothelial cells Formaldehyde-treated albumin, acetylated LDL, PIIINP
One form Same as PICP 38–202 50–170 Mannose receptor of liver endothelial cells Ovalbumin, mannan
394
JUHA RISTELI AND LEILA RISTELI
C. Amino-terminal Propeptide of Type I Procollagen (PINP) The aminoterminal propeptide contains a collagenous domain and thus has the shape of a short rod ending with a more globular part (Fig. 1). The subdomains of the propeptide have been named on the basis of their order of elution from a size-exclusion column after bacterial collagenase digestion. From the largest to the smallest, they are the following: Col 1 (the globular amino-terminal end with intrachain disulfide bridges); Col 2 (the short noncollagenous part joining the propeptide to the most amino-terminal part of the collagen proper); and Col 3 (the central collagenous domain that is broken down by bacterial collagenase). The parts corresponding to the propeptide are encoded by exons 1 through 6 of the COL1A1 and COL1A2 genes. The molecular mass of the propeptide is about 35 000 daltons (Table I), and masses of the subunits are about 14 250 daltons for the proα1 chain and 5500 daltons for the proα2 chain. The proα2 chain of the propeptide contains the collagenous domain (Col3) and the short carboxyterminal noncollagenous sequence (Col 2), but lacks the globular amino-terminal end (Col 1 domain). The length of the triple helix in the propeptide is 16 Gly-X-Y triplets – this is less than 5% of the length of the helical domain in the collagen molecule itself, which contains 338 such triplets. The globular Col 1 domain is phosphorylated in the type I procollagens produced by both fibroblasts and osteoblasts, the phosphate being linked to serine as phosphoserine [21]. The acidic nature of the propeptide is advantageous when the protein is being isolated from various biological starting materials, such as pleural effusions or ascitic fluids, which often contain high concentrations of the propeptide [22]. Because the amino-terminal propeptide of the α1-trimer type I procollagen contains three globular Col 1 domains, it is more acidic than the propeptide of the classical type I procollagen with its two Col 1 domains and thus elutes later in anion exchange chromatography [23].
D. Clearances of the Circulating PICP and PINP In comparison with the large amounts of the propeptides set free in the body, their circulating concentrations are surprisingly low, only 20–200 µg/L, which is in the nanomolar range (Table I). This fact suggests the existence of efficient removal mechanisms for this material. In fact, the further metabolic fates of both PICP and PINP are now known better than those of most other analytes used routinely in clinical medicine. The propeptides are
not passively lost in the urine because they are either too large (PICP) or have such an elongated shape (PINP) that they cannot pass through the glomerular filter. In addition, the negative charge of the latter, due to the covalently attached phosphate groups, obviously repels the propeptide from the negatively charged membrane. The metabolic fates of the free propeptides have been clarified by several types of experiments, such as by following the disappearance of radioactively labeled or fluorescent propeptides from the circulation in rats, by feeding the proteins to cultured cells in the presence of known competing ligands for various membrane receptors, by simultaneous measurement of the propeptide concentrations in different vascular beds, and, recently, also by studying the effect of receptor knockout on the circulating propeptide concentration in mice (Fig. 2) [24]. The propeptides are actively taken up and metabolized by a specialized part of the reticulo-endothelial system, the endothelial cells of the liver [25]. These cells bind and internalize the propeptides by receptor-mediated endocytosis using two different receptor systems, the mannose receptor for PICP [26] and the scavenger receptor for PINP [27]. The clearance follows biphasic kinetics, with half-lives measured in minutes in the rat. The endocytosed material is delivered to the lysosomes and degraded, with the resulting amino acids most probably being passed to the hepatocytes. The evolution of such a clearance mechanism could be related to the importance of having an effective recycling system for the large amounts of amino acids in the propeptides. The amount of type I collagen synthesized per year in adult humans has been estimated to be at least 1 kg [28], and for each kilogram of type I collagen, as much as 500 g of propeptide material is also produced. The oligosaccharides on PICP are of the high-mannose type that is considered relatively primitive; a special function related to such recycling could explain this feature, which has no effect on the functions or processing of the procollagen [29]. The two clearance systems, although present in the same cells, are independent of each other and also seem to be regulated separately. Some hormones, e.g., thyroxine, can affect the activities of these endocytotic systems and even sometimes cause opposite changes in the circulating concentrations of the procollagen propeptides [30]. However, the effects of such endocrinological regulation are relatively small. In contrast, at least one family is known that has an obvious genetic defect in the uptake mechanism of PICP [31]. The circulating PICP concentrations of the affected family members tend to be several standard deviations above the upper limits of the age- and sex-adjusted reference intervals, with no other abnormalities in either type I collagen metabolites or other biochemical markers of bone metabolism. The feature is inherited in an autosomal
395
Chapter 23 Products of Bone Collagen Metabolism
Figure 2
Proteomic analysis of total serum proteins in mice, two-dimensional polyacrylamidegel electrophoresis under reducing conditions. The gels were stained with SYPRO Ruby and scanned on a laser fluorescence scanner. WT = wild-type mice; KO = mannose receptor knockout mice; pI = isoelectric point; MW = molecular weight. The spot nos. 1–4 represent the proteins that were constantly found to be elevated in the knockout samples. The spots nos. 1–3 were identified by peptide mass fingerprinting as the carboxy-terminal propeptide domains of the proα1(I) and proα2(I) chains of type I procollagen and the proα1(III) chain of type III procollagen. Reproduced from Lee et al. [24], with permission.
dominant manner (Fig. 3) and is not associated with any obvious signs or symptoms. Interestingly, such grossly elevated PICP concentrations also react to factors known otherwise to increase or decrease this concentration, such as pregnancy [32] or glucocorticoid therapy [31]. Studies have shown also that the local PICP concentration in suction blisters of the skin (see Section II.E) is similarly elevated in these individuals, suggesting that there normally may also be some local uptake of PICP, at least in the soft tissues, possibly by tissue macrophages carrying mannose receptors [33].
practically noninvasively testing the validity for any new test assumed to detect type I collagen synthesis. The relative contribution of soft-tissue type I collagen synthesis to the circulating concentration of PICP has been studied in lymph flow experiments with conscious pigs [36]. There is a gradient of 10:1 from lymph to peripheral
E. Comparison of PICP and PINP in Physiological and Clinical Situations The local synthesis of soft-tissue type I procollagen can be assessed in human skin, either by collecting the interstitial fluid in surgical wounds [34] or by a suction blister method [35]. Such studies have shown that the propeptides of type I procollagen are efficiently set free from the collagen in the extracellular fluid. Their concentrations increase dramatically (up to 1000-fold within 1 week) during wound healing, when the expression of type I procollagen is induced [22, 34]. On the other hand, local treatment with glucocorticoids suppresses the synthesis of this collagen, leading to often dramatic decreases in the concentrations of the propeptides measured in the fluid in suction blisters that are made in the treated area of the skin [35]. Such experiments provide excellent means of reproducibly and
Figure 3
Family with inherited high circulating concentrations of the carboxy-terminal propeptide (PICP) of type I procollagen. The circles and squares indicate women and men, respectively. The open symbols show unaffected family members, whereas the solid symbols present individuals with high PICP concentrations. The arrow indicates the proband. AP, alkaline phosphatase; BAP, bone-specific alkaline phosphatase; OC, osteocalcin. Reproduced from Sorva et al. [31], with permission.
396 blood in the concentration of the amino-terminal propeptide of type III procollagen (PIIINP; soft-tissue origin), whereas there are no significant differences between the concentrations of PICP in lymph and blood. Stopping and restarting the lymph flow produces dramatic changes in the concentration of PIIINP in the blood, whereas there are no changes in the corresponding concentration of PICP. This study demonstrates that even large changes in the access of the products of soft-tissue type I collagen synthesis into the circulation have little effect on the blood concentration of PICP. This is not surprising because we are, in fact, not measuring the absolute amounts of the proteins in any pool in the body, but just their concentrations, and before mixing the lymph into the blood there is no gradient in this case. Because soft tissues have little effect on blood, the bone formation rate is obviously the main determinant of the concentrations of the type I procollagen propeptides in the blood (Fig. 4). This view is also supported by
Figure 4
Schematic presentation of the distribution between various pools and of the clearances of the type I procollagen propeptides. PICP, the carboxy-terminal propeptide; Intact PINP, the intact triple-helical aminoterminal propeptide of type I procollagen; PINP Col 1, the smaller, singlechain antigen related to the Col 1 domain of the proα1 chain, recognized by some assays for PINP. Thick arrows indicate the major routes. Reproduced from Risteli et al. [2], with permission.
JUHA RISTELI AND LEILA RISTELI
histomorphometric and calcium kinetic studies in patients [37, 38]. In the blood, the carboxy-terminal propeptide (PICP) is always found as the free antigen, with the size of the authentic propeptide [18], whereas the circulating antigenicity related to the amino-terminal propeptide (PINP) can be resolved into two peaks with different molecular sizes. The first is identical to the free, authentic trimeric antigen, and the second resembles part of a single polypeptide chain, probably containing at least the globular Col 1 domain of the proα1(I) chain of PINP [22]. PINP can be isolated as an intact, trimeric propeptide from pleural and ascitic fluids [22], whereas the corresponding propeptide isolated from amniotic fluid always seems to fall apart, into single, possibly truncated, chains [39, 40]. Different PINP assays may vary with respect to the extent to which their results are influenced by the smaller antigens. The corresponding propeptide of type III procollagen, PIIINP, is also present in the blood as a series of antigens with partially different metabolic origins and differing metabolic fates [1, 3]. The origin of the smaller antigenic forms of PINP is most probably the tissue degradation of newly synthesized type I procollagen or of the type I pN-collagen with the retained propeptide (Section III.A). Thus the exclusive assay of the intact PINP seems to be more sensitive than that of total PINP for detecting changes in the rate of bone collagen synthesis (e.g., estrogen treatment in postmenopausal women leads to a 42% decrease in the circulating concentration of intact PINP, compared with the 30% decrease observed with an assay measuring both antigenic forms [41]). There is also an important difference in the elimination routes, since intact PINP is cleared by the liver (see Section II.D), whereas the small PINP Col 1-like peptides can also be filtered by the kidneys (Fig. 4) [1]. Because PICP and PINP are produced in equimolar amounts during the synthesis of type I procollagen, their molar concentrations in any relevant body fluids should, in principle, be almost identical. This is true, for example, in the fluid collected from a healing wound, where both the propeptides give similar estimates for the production rate of type I procollagen [22]. The ratio of PICP:PINP in adult blood is also approximately the expected 2–3:1, when the concentrations of both propeptides are expressed as protein concentrations (µg/L). However, there are both physiological and pathological situations in which this ratio differs from equimolarity. In the blood of infants and children (up until puberty), the molar concentration of PINP is higher than that of PICP, the ratio of PICP/PINP (in µg/L) being even clearly lower than 1:1 [42]. An even more dramatic discrepancy is found in patients with active Paget’s disease of bone
Chapter 23 Products of Bone Collagen Metabolism
[43, 44], or with aggressive breast cancer [45], or ovarian cancer [46]. In these cases, there can be an up to six-fold molar excess of PINP in the blood. The physiological basis for this finding is not yet known.
III. DEGRADATION PRODUCTS OF TYPE I COLLAGEN A. Formation, Cross-linking, and Aging of Collagen Fibers The rod-like collagen molecules possess a capacity for self-assembly and spontaneous formation of fibrils and fibers (see Chapter 2). The amino-terminal propeptide often remains transiently attached to some of the molecules and can be visualized on the surface of thin type I collagen fibers in tissues undergoing active collagen synthesis (see Section II.A) [13]. One characteristic location for such fibers is the epithelial–stromal junction below a metabolically active epithelium [47]. In the newly formed osteoid, some amino-terminal propeptides of type I procollagen are retained for the time of matrix maturation, since the collagenous matrix can be stained with anti-PINP antibodies, but no substantial amounts of the propeptide are any longer present in the mineralized matrix [T. Taube, I. Elomaa, L. Risteli, J. Risteli, unpublished]. The aminoterminal propeptide can be isolated as a 24-kDa phosphoprotein from the organic matrix of fetal bone, but not from the adult bone that has a much lower rate of synthesis of type I collagen [21]. In the case of type III collagen, a major collagen of soft tissues, pN collagen molecules seem to cover the surface of virtually all fibers, although they only represent a 5% fraction of the total amount of type III collagen [48]. The biological function of a fibrillar collagen is to provide the tissue with tensile strength. This is due to covalent bonds, the cross-links, that are formed between the individual collagen molecules in a collagen fiber. The cross-links result from a complicated series of partially alternative chemical reactions that gradually lead through divalent cross-links, joining two polypeptide chains, to multivalent (tri- or even tetravalent) cross-links (see Chapter 2). Continuing cross-linking explains the fact that older collagen in soft tissues is progressively more insoluble when studied in various ways, e.g., by digestion with pepsin or bacterial collagenase. The trivalent pyridinoline cross-links, which have been much studied, are abundant in the bone, but are even more so in the soft tissues where the maturation process is mostly complete and occurs within a relatively short time after the assembly of the collagen fiber. At least in some
397 tissues, the appearance of pyridinoline cross-links seems to be related to the development of an irreversible fibrosis, e.g., in the skin [49] and in the liver [50]. There is a wide variation in the cross-links, depending on whether the precursor lysyl residues have been hydroxylated to hydroxylysines or not [51]. Bone is the major connective tissue in which the collagen cross-links remain immature to a significant extent, the content of the divalent cross-links being about 2–4 times the concentration of the pyridinolines [52]. There are two obvious explanations for this finding: first, that mineralization slows down the cross-linking process and, second, that a significant part of the mature cross-links in bone have some other structure than pyridinoline [9]. The concentration of the divalent cross-links has been found to specifically decrease in osteoporosis [52, 53]. Unfortunately, analysis of the maturity of the cross-links in bones or other tissues suffers from the lack of easy, rapid, and reliable quantitative methods that would be suitable for a large number of small tissue samples. The development of immunoassays based either on synthetic peptides or on naturally cross-linked telopeptides may help in this respect, provided that the specificity of such assays is carefully tested. The application of such assays that detect the polypeptide chains involved in the cross-linked telopeptide structures has already revealed that, in addition to the pyridinolines and the relatively recently described pyrrolic cross-links [54, 55], there must still exist other, uncharacterized mature cross-links [9, 56]. In soft tissues, immaturely cross-linked collagens are transiently encountered during normal collagen turnover and in the granulation tissues. Defective cross-linking can also be a major feature of the stroma of malignant tumor tissue [48]. Interestingly, it has been found that lysyl oxidase, the only enzyme involved in the cross-link process, is identical to the tumor suppressor protein known as rrg [57] and that its synthesis is decreased in malignant breast tumor tissue [58]. Some further modifications take place during the aging of the collagen fibers, such as the cross-linking through nonenzymatically attached sugar residues [59] and the racemization and β-isomerization of some aspartic acid residues (see below). It is not known if these modifications proceed similarly in the soft tissues and in the skeleton. A β-isomerization of an aspartic acid residue as the result of a nonenzymatic, physiological aging reaction is known to take place both in the carboxy-terminal [56] and the amino-terminal telopeptide of the α1-chain of type I collagen [60]. This occurs via the formation of an unstable succinimide ring, which will spontaneously hydrolyze either back to a normal L-aspartyl or to an L-βaspartyl residue (Fig. 5). The factors determining which
398
JUHA RISTELI AND LEILA RISTELI
Figure 5
The β-isomerization and racemization reactions of aspartate residues in
proteins.
Asp residues are susceptible to such isomerization are the nature of the adjacent amino acid residues and the constraints imposed by the surrounding secondary and tertiary structures [61]. On the basis of specific immunoassays, the ratio of the α:β forms of aspartic acid in the carboxy-terminal telopeptide has been reported to be 0.40 in cortical and 0.48 in trabecular bone collagen and as high as 1.33 in pagetic bone [62]. Thus, even 70–80 % of the L-Asp in this location in normal adult bone collagen could be isomerized into the β-configuration. It would also be interesting to know if a similar development takes place in soft-tissue type I collagen and if the mineralization process in bone affects the rate of β-isomerization. Several cross-linked carboxy-terminal telopeptides of type I collagen have been isolated from urine that contains either only two α or two β forms or one of both forms [56]. In addition to the isomerization, a racemization of the α-carbon of the aspartate to give the corresponding D-amino acid derivative is also possible (Fig. 5). D-Asp seems to accumulate at a rate of about 0.1% per year during aging in the human bone fraction, containing mostly collagen [63]; thus, at the age of 60 years, about 6% of the total amount of the L-Asp residues in bone collagen have been racemized into D-Asp.
as a covalently cross-linked assembly of large numbers of molecules. In particular, the helical region of a collagen molecule can only be attacked by a few proteolytic enzymes before the cross-linking ends of the molecule have been cleaved. In soft tissues, there seem to be two routes of collagen breakdown, one being active in conditions associated with rapid remodeling of the tissue and the other under steady-state conditions [64]. In bones, the corresponding pathways are most probably due to the activities of matrix metalloproteinases and cathepsin K, respectively [65]. Because of the structural complexity and the presence of alternative degradation pathways, the pattern of crosslinked peptides that arises from the degradation of a certain collagen type in a certain tissue is difficult to predict. The existence of different enzymes with differing cleaving specificities and functioning in different metabolic situations adds to the complexity. Assessment of bone collagen degradation can be based either on determining small breakdown products, such as modified amino acids (including cross-links), or on immunological quantification of peptides.
C. Products Derived from the Amino-terminal Telopeptide of Type I Collagen B. Degradation Pathways The collagen in many soft tissues and particularly that in the skeleton is actively turned over throughout the life of an individual, although the overall turnover rate is naturally at its highest in infants and children. A number of enzymes can degrade soluble forms of type I collagen in the test tube. These enzymes include matrix metalloproteinases, cysteine proteinases, and serine proteinases [64]. However, it is more difficult to elucidate the real biological significance of such an activity because, in the tissues, the collagen is not present as soluble single molecules but
The amino-terminal telopeptide regions of both the α1(I) chain (17 amino acids) and the α2(I) chain (11 amino acids) of type I collagen contain lysine (or hydroxylysine) that can be oxidized to allysine (hydroxyallysine). The further events at this location can, in principle, lead to either intermolecular or intramolecular cross-links, resulting in a complex pattern of possible structures. One class of multivalent cross-links, those with pyridinoline structures, can be found both in blood and in urine. When quantified in such fluids, they do not reveal their specific origins, however, although the lysyl pyridinoline variant is more abundant
399
Chapter 23 Products of Bone Collagen Metabolism
in bone than in soft-tissue collagen. In addition to the Nterminal telopeptide of type I collagen, the cross-links can originate in the C-terminal telopeptide of the same protein or in other collagen types. Measurement of collagen cross-links is a more specific way of assessing the breakdown of fibrillar collagen than the traditional hydroxyproline measurement, however, because the cross-links can only be derived from the degradation of collagen molecules that have participated in collagen fibers. Furthermore, the pyridinoline cross-links are not absorbed from the diet, so their excretion is only related to their endogenous formation. The N-terminal telopeptide of type I collagen also contains other multivalent cross-link structures than pyridinolines. In particular, the pyrrole cross-links are predominantly present in the N-terminus [54, 55]. An immunological assay method has been described for the N-terminal telopeptide of type I collagen that detects a related antigen in blood [66] and in urine [67]. The antigen measured has been called NTx, and it represents a neoepitope, i.e. the corresponding sequence is not recognized by the antibody if the collagen molecule is intact. The structure of the Ntx peptide and its antigenic epitope are shown in Figure 6. NTx comprises one peptide derived from the α1(I) chain and another from the α2(I) chain, incorporating a single (hydroxy)lysine residue from the helical domain of a third chain. In urine samples, the apparent concentration of the NTx peptide has been found
Figure 6 Antigenic epitope of the NTx assay for the amino-terminal telopeptide of the human type I collagen molecule. The epitope depends on a specific proteolytic cleavage of the α2(I)N-telopeptide to the sequence shown where J is pyroglutamate and K is the original lysyl residue embodied in a cross-link. Reproduced from Apone et al. [90], with permission.
to increase when the samples are exposed to UV light [68]. Such a finding could be explained by the presence of pyrrolic cross-links in the small peptides; these cross-links are known to easily aggregate, and UV light could dissolve such aggregates. An aspartate residue that is located in the α2(I) part of NTx has been shown to undergo β-isomerization [60]. Although it is possible to set up assays that differentiate between the α- and β-isomers in this location, the specificity of the commonly used NTx assay in this regard has not been published.
D. Products Derived from the Helical Domain of Type I Collagen According to a traditional view of collagen breakdown, the major triple-helical domain of type I collagen contains the cleavage site for the mammalian collagenase enzyme that is unique in its ability to degrade this protein under physiological conditions. This enzyme is a member of the group of matrix metalloproteinases. Once the collagen molecule has been cut in two, the resulting shorter helices are no longer stable at body temperature and the polypeptide chains thus become susceptible to other proteindegrading enzymes as well. However, cathepsin K, which is the major collagen-degrading enzyme of osteoclasts, seems also to cleave the helical domain of native type I collagen in many locations in a reproducible manner [65]. Peptide antigens derived from the central, helical part of type I collagen have been reported to exist in blood [69] and in urine [70–72]. These findings indicate that some of the degradation products must be relatively long peptides, in order to contain an antigenic determinant (epitope). The circulating antigens have been detected by using antibodies raised to purified human collagen [69] and the urinary ones by using monoclonal antibodies to synthetic peptides [70–72]. The collagenous helix is generally considered an extremely weak immunogen both in its native and denatured state [73]. Thus it is essential that the true chemical natures of the physiological, immunoreactive substances are carefully characterized. The collagenous nature of the circulating antigen has been verified by its sensitivity to highly purified bacterial collagenase [69]. The sequence of the immunogen that has been used in developing the urinary assay has, on the other hand, been derived from a peptide isolated from the urine of a patient with Paget’s disease [70]. Despite this, it has not yet been verified beyond doubt that what is measured in the urine by such an assay under normal circumstances is indeed derived from bone tissue, particularly as there is a 90% cross-reaction with the corresponding part of type III
400 collagen [74]. The assay has been used for some clinical studies in osteoporosis [75]. 4-hydroxyproline is necessary for the correct triplehelical conformation of collagen at body temperature. This modified imino acid makes up about 12% of the weight of the type I collagen molecule and can thus be used as a measure of the collagen content of a tissue. The acid hydrolysis, followed by a chemical color reaction used to assess this imino acid, can be applied both for tissue samples and for urine. This method has traditionally been used as a measure of bone collagen metabolism [28]. Once set free from the helical part of a collagen molecule, hydroxyproline can no longer be incorporated into new protein; however, exogenous hydroxyproline is absorbed from the diet. Most of the imino acid (up to 90%) is metabolized in the liver, leading to the formation of pyruvate and glyoxalate, but some is always passed into the urine. Excretion of hydroxyproline correlates with growth velocity in children and with the presence of bone degrading processes, e.g., bone metastases of cancers, in adults [28]. Also, a generally enhanced bone metabolic rate increases urinary hydroxyproline excretion. For this reason, hydroxyproline served as an indirect test of thyroid hormone action before the time of specific thyroid hormone assays. Some hydroxyproline is directly derived from collagen synthesis, since the helical domain of PINP contains it. A substantial part of the urinary hydroxyproline can also be derived from the C1q component of the complement, particularly in inflammatory conditions [76]. The 4-hydroxyproline content is similar in all type I collagens. The extent of other post-translational modifications in the helical domain varies from one situation and tissue to another. Such modifications include the hydroxylation of lysine, the subsequent glycosylation of hydroxylysine – first by galactose and then by glucose – and the hydroxylation of proline at position 3, leading to the formation of 3-hydroxyproline. The presence of a halfway glycosylated hydroxylysine, galactosylhydroxylysine, has been suggested to be characteristic of adult bone tissue, and the excretion of this amino acid in the urine has been used as a marker of bone collagen degradation [77]. An immunological assay has also been introduced for galactosylhydroxylysine [78].
E. Products Derived from the Carboxy-terminal Telopeptide of Type I Collagen In the carboxy-terminal telopeptides of type I collagen, lysine/hydroxylysine is only present in the α1(I) chains, making the number of possible alternative cross-linked structures smaller than the corresponding number at the
JUHA RISTELI AND LEILA RISTELI
amino-terminus where both α1(I) and α2(I) contain lysine/ hydroxylysine. Since the divalent cross-link formed first binds one telopeptide to the helical region of another collagen molecule, there is a possibility that the telopeptide of the second α1(I) chain remains free. Thus, several degradation products can be formed from the carboxyterminal part of type I collagen that differ with respect to the α1-chain; this either has no cross-link or participates in a divalent or a trivalent cross-link. The latter may be hydroxylysyl pyridinoline, lysyl pyridinoline, or even pyrrole. There is some evidence also for the presence of an as yet uncharacterized cross-link [9, 56]. When pyridinoline cross-links as such are being measured, they can be derived both from the carboxy-terminus and from the amino-terminus of the type I collagen molecule. More specifically, several immunoassays have been developed for structures involving the carboxy-terminal telopeptides and they can give different results, depending on their immunochemical specificity with respect to the length of the antigen, the maturity of the cross-link and the possible modifications of an aspartic acid residue in the telopeptide. The ICTP antigen is a trivalently cross-linked structure that was originally isolated from human femoral bone after digestion with trypsin or bacterial collagenase and shown to contain the carboxy-terminal telopeptides of two α1(I) chains and material from the helical part of a third chain [79]. The immunological assay for ICTP detects relatively large degradation products of mature type I collagen since the epitope requires the presence of two phenylalaninerich regions [80] (Fig. 7). Such antigens are detectable in the blood and in several other physiological fluids, but not in the urine. Specificity tests have indicated that cathepsin K cleaves the carboxy-terminal telopeptide of the α1-chain of type I collagen on both sides of the sequence LPQPPQE, which is located between the phenylalanine-rich region and the cross-link site; thus, the degradation product liberated by cathepsin K is not detected by the ICTP assay (Fig. 7). A series of assays called CrossLaps or CTx have been developed using synthetic peptides related to the carboxyterminal telopeptide as immunogen. The respective part of the telopeptide is situated more in C-terminal direction than the ICTP epitope (Fig. 7). The first of these could recognize related antigens in the urine; the immunogen contained a stretch of eight amino acids, including the cross-link site of the α1(I) chain as an unmodified lysine [81]. Later, variants of the assay were introduced that can differentiate between the α- and β-forms of the Asp located in the epitope [82]. The CrossLaps α- or β-assays may detect both uncross-linked and di- or trivalently crosslinked species since the epitope is a linear sequence of only
401
Chapter 23 Products of Bone Collagen Metabolism
Figure 7
Antigenic epitopes in the carboxy-terminal telopeptide of human type I collagen (ICTP and CrossLaps). The ICTP assay demands two adjacent streches that both contain a phenylalanine-rich domain FDFSF. The CrossLaps assays detect linear, eight amino acidlong sequences (bold) where the aspartic acid residue is either in an α- or a β-configuration. Cat K = cathepsin K cleavage site. Reproduced from Garnero et al. [92], with permission.
six amino acids (Fig. 7). However, the serum CrossLaps β-β-assay in which a double antibody technique is used [83] does not detect immature degradation products (Fig. 7). In addition to α to β isomerization, the aspartic acid residue in the CrossLaps epitope can undergo L to D racemization. All the four alternative structures (αL, αD, βL, and βD) have been found in the carboxy-terminal telopeptide of type I collagen in human tissues [84]. A number of clinical studies have indicated that the ICTP and CrossLaps methods reflect different processes in type I collagen metabolism, i.e. pathological states involving breakdown of bone tissue and normal turnover of the skeleton, respectively.
pathological situations, and thus to liberate the ICTP antigen into the blood. Such situations include tumor metastases, myeloma lesions, and rheumatoid arthritis (Fig. 8) [87]. In the hereditary disease pycnodysostosis, there is a genetic defect in cathepsin K, with lower than normal NTx and CrossLaps excretions and a compensatorily highly increased circulating ICTP concentration [88]. Degradation products of bone collagen have been immunochemically identified both in osteoclast cultures on bone slices [89–92] and in vitro, by degrading demineralized bone matrix with proteolytic enzymes. In the former system, peptides are liberated that react in assays for the amino-terminal [89] and the carboxy-terminal [89, 91, 92] telopeptides of type I collagen, although the exact sizes of these degradation products have not been reported.
F. The Physiological Degradation Products of Type I Collagen It first became evident from clinical studies that cathepsin K is active in the major pathway of normal bone turnover, a process that is enhanced in estrogen deficiency, for instance. The degradation phase of the bone metabolic cycle involves the specific attachment of osteoclasts to the mineralized surface, the dissolution of the inorganic mineral component (acidic microenvironment) and the degradation of matrix proteins (protease activity). Bone collagen degradation occurs mainly in the resorption lacuna and is analogous to the intracellular route of soft-tissue collagen degradation [85]. Cathepsin K, with its ability to cut the collagen molecule in both the telopeptides and in the helical domain [65, 86], liberates, among other things, NTx and CrossLaps (CTx) antigens from the bone. Another class of proteolytic enzymes, the matrix metalloproteinases, seems to be more actively involved in the localized breakdown of bone that takes place in
Figure 8 Size distribution of the ICTP and CrossLaps antigens in human serum. A 2.5-ml sample of serum from a 55-year-old woman with rheumatoid arthritis was chromatographed on a Sephacryl S-100 column at room temperature in phosphate-buffered saline containing 0.04% Tween20. The dotted line indicates the detection limit of the CrossLaps assay based on the absorbance of the blank (standard A = 0 pmol/l) of the kit. Before analyses the 20-min fractions were concentrated by lyophilization. Reproduced from Sassi et al. [87], with permission.
402 Interestingly, no free hydroxylysyl pyridinoline or lysyl pyridinoline is set free from the bone [90]. The probable liberation of peptide products originating from the helical parts of the collagen molecules or hydroxyproline and galactosylhydroxylysine have yet to be characterized in this system. When demineralized bone matrix is treated with proteolytic enzymes in the test tube, the kinetics of the appearance of the different degradation products can be followed more closely. Cathepsin K liberates an NTx antigen and generates a neoepitope that is necessary for reactivity in this immunoassay [86]. The enzyme can degrade up to 100% of a purified, decalcified bone matrix [65], cutting the collagen molecules in telopeptide areas as well as in the helical domain. The number of studies on the exact mechanism of bone collagen degradation is still quite limited. However, studies on the urinary excretion of presumed degradation products of bone collagen abound (see Chapter 35). Both hydroxyproline and the pyridinoline cross-links are present partially in free form and partially in peptide-bound form in the urine; when the turnover rate of bone increases, there is typically a concomitant increase in the total excretion of the cross-links, but also a disproportionate increase in their peptide-bound fraction. The urinary antigens that are recognized either by the NTx or the CrossLaps assay behave in a manner resembling the peptide-bound crosslinks in this respect. Furthermore, different antiresorptive pharmaceuticals seem to affect the free and peptidebound fractions differently. For example, estrogen therapy
JUHA RISTELI AND LEILA RISTELI
decreases the urinary excretion of both the free and peptide-bound cross-links in postmenopausal women, whereas bisphosphonate treatment of patients with Paget’s disease has a marked effect only on the cross-linked peptides without a change in the free cross-links [93]. It can be speculated that the bisphosphonates may be specific for osteoclasts, which do not seem to produce free pyridinolines, whereas estrogen has a more general effect on both bone and nonmineralized tissues. This reasoning suggests that the free pyridinolines in the urine would largely be derived from the degradation of soft-tissue collagens. While it is known that type III collagen contains a lot of hydroxylysyl pyridinoline, this most probably is not the whole explanation for the interesting finding, in particular because there is no significant difference between hydroxylysyl and lysyl pyridinolines in this respect. The degradation of extraskeletal type I collagen and the further metabolism of the degradation products of bone collagen are not known in detail (Fig. 9). The majority of the hydroxyproline set free from collagens is known to be metabolized in the liver. The endothelial cells of the liver possess specific gelatin receptors for the internalization of large collagenous peptides [25]. If such peptides still contain the cross-links, it is possible that the formation of free pyridinolines could partially take place in the liver. Another site for the further handling of the cross-linked peptides is the kidneys. In particular, the tubuli contain several exo- and endoproteinases, which digest small peptides to amino acids that can be reabsorbed. Most probably also the original degradation products of type I collagen are further shortened in the kidneys, for example the main urinary degradation product detected by the CrossLaps assay only contains the core sequence EKAHDGGR [56]. Also autoradiographic methods have demonstrated that the kidneys participate in the degradation of collagenous peptides [94]. The renal clearance of the free pyridinoline cross-links has been shown to be four-fold higher than that of the cross-links still present in peptides [95]. Since the fractional clearance of the former is greater than one and that of the latter less than one, some free pyridinolines excreted in urine are produced by the kidney (Fig. 9). The possible effects of pathological processes affecting the kidneys on cross-link excretion – or even on the excretion of any type I collagen telopeptide antigens – have not been studied.
IV. CLOSING REMARKS Figure 9
Schematic presentation of the distribution between various pools and of the clearances of type I collagen telopeptides.
It has been a challenge for research to assess the balance between bone collagen synthesis and degradation by the simultaneous measurement of metabolic products related
403
Chapter 23 Products of Bone Collagen Metabolism
to these two processes. There are two possible analytes related to the synthesis of type I collagen, the carboxyterminal (PICP) and the amino-terminal (PINP) propeptides of type I procollagen, and the respective assay methods are already quite well established. There is a great deal of clinical experience on their performance in different physiological and pathological situations. Although in general both propeptides give relevant information of type I collagen synthesis, in some situations the assay for the intact PINP seems to be superior to that for PICP. The general picture of the degradation of bone type I collagen has been cleared by recent research, with the clarification of the roles of the cathepsin K and matrix metalloproteinase pathways. However, a number of more specific questions remain, related to the further metabolic fates of the primary degradation products, to the breakdown products of the helical domain or to the origins of the free forms of cross-links, for instance. Much clinical experience has been gained using urine samples, but the present trend is towards serum assays. The respective contributions of the skeleton and the soft tissues to the different degradation products have so far not been seriously studied.
10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
References 20. 1. Risteli, J., and Risteli, L. (1995). Analysing connective tissue metabolites in human serum. Biochemical, physiological and methodological aspects. J. Hepatol. (Suppl. 22), 77–81. 2. Risteli, J., and Risteli, L. (1997). Assays of type I procollagen domains and collagen fragments: problems to be solved and future trends. Scand. J. Clin. Lab. Invest. (Suppl. 227), 107–113. 3. Risteli, L., and Risteli, J. (2002). Extracellular matrix metabolites in body fluids. In “Connective Tissue and its Heritable Disorders: Molecular, Genetic and Medical Aspects” (P. M. Royce, and B. Steinmann, eds.), 2nd edition, pp. 1141–1160. John Wiley & Sons, New York. 4. Scariano, J. K., Walter, E. A., Glew, R. H., Hollis, B. W., Henry, A., Ocheke, I., and Isicei C. O. (1995). Serum levels of pyridinoline crosslinked carboxyterminal telopeptide of type I collagen (ICTP) and osteocalcin in rachitic children in Nigeria. Clin. Biochem. 28, 541–545. 5. Oginni, L. M., Sharp, C. A., Badru, O. S., Risteli, J., Davie M. W. J., Worsfold, M. (2003). Radiological and biochemical resolution of nutritional rickets with calcium. Arch. Dis. Child. 88, 812–817. 6. Leskelä, H.-V., Risteli, J., Niskanen, S., Koivunen, J., Ivaska, K. K., and Lehenkari, P. (2003). Osteoblast recruitment from stem cells does not decrease by age at late childhood. Biochem. Biophys. Res. Commun. 311, 1008–1013. 7. Jukkola, A., Risteli, L., Melkko, J., and Risteli, J. (1993). Procollagen synthesis and extracellular matrix deposition in MG-63 osteosarcoma cellls. J. Bone Miner. Res. 8, 651–657. 8. Mechanic, G. L., Katz, E. P., Henmi, M., Noyes, C. and Yamauchi, M. (1987). Locus of a histidine-based, stable trifunctional, helix to helix collagen cross-link: stereospecific collagen structure of type I skin fibrils. Biochemistry 26, 3500–3509. 9. Eriksen, H., Sharp, C. A., Robins, S. P., Sassi, M.-L., Risteli, L., Risteli, J. (2004). Differently cross-linked and uncross-linked carboxy-terminal
21.
22.
23.
24.
25.
26.
27.
28. 29.
telopeptides of type I collagen in human mineralised bone. Bone 34, 720–727. Prockop, D. J., Kivirikko, K. I., Tuderman, L., and Guzman, N. A. (1979). The biosynthesis of collagen and its disorders. N. Engl. J. Med. 301, 13–23, 77–85. Kessler, E., Takahara, K., Biniaminov, L., Brusel, M., and Greenspan, D. S. (1996). Bone morphogenetic protein-1: the type I procollagen C-proteinase. Science 271, 360–362. Bornstein, P. (2002). The NH2-terminal propeptides of fibrillar collagens: highly conserved domains with poorly understood functions. Matrix Biol. 21, 217–226. Fleischmajer, R., Perlish, J. S., and Timpl, R. (1985). Collagen fibrillogenesis in human skin. Ann. N. Y. Acad. Sci. 460, 246–257. Byers, P. H., and Cola, W. G. (2002). Osteogenesis imperfecta. In “Connective Tissue and its Heritable Disorders” (P. M. Royce, and B. Steinmann, eds.), 2nd edition, pp. 385–430. Wiley-Liss, New York. Bailey, A. J., Sims, T. J., and Knott, L. (2002). Phenotypic expression of osteoblast collagen in osteoarthritic bone: production of type I homotrimer. Int. J. Biochem. Cell Biol. 34, 176–182. Mann, V., Hobson, E. E., Stewart, T. L., Grant, S. F., Robins, S. P., Aspden, R. M., and Ralston, S. H. (2001). A COL1A1 Sp1 binding site polymorphism predisposes to osteoporotic fracture by affecting bone density and quality. J. Clin. Invest. 107, 899–907. Clark, C. C. (1979). The distributional initial characterization of oligosaccharide units on the COOH-terminal propeptide extensions of the pro-α1 and pro-α2 chains of type I procollagen. J. Biol. Chem. 254, 10798–10802. Melkko, J., Niemi, S., Risteli, L., and Risteli, J. (1990). Radioimmunoassay of the carboxyterminal propeptide of human type I procollagen. Clin. Chem. 36, 1328–1332. Pedersen, B. J., and Bonde, M. (1994). Purification of human procollagen type I carboxyl-terminal propeptide cleaved as in vivo from procollagen and used to calibrate a radioimmunoassay of the propeptide. Clin. Chem. 40, 811–816. Taubman, M. B., Goldberg, B., and Sherr, C. J. (1974). Radioimmunoassay for human procollagen. Science 186, 1115–1117. Fischer, L. W., Robey, P. G., Tuross, N., Otsuka, A. S. O., Tepen, D. A., Esch, F. S., Shimasaki, S., and Termine, J. D. (1987). The Mr 24 000 phosphoprotein from developing bone is the NH2-terminal propeptide of the α1 chain of type I collagen. J. Biol. Chem. 37, 13457–13463. Melkko, J., Kauppila, S., Risteli, L., Niemi, S., Haukipuro, K., Jukkola, A., and Risteli, J. (1996). Immunoassay for intact aminoterminal propeptide of human type I procollagen. Clin. Chem. 42, 947–954. Kauppila, S., Jukkola, A., Melkko, J., Risteli, L., TurpeenniemiHujanen, T., Vuorinen, K., and Risteli, J. (2001). Aminoterminal propeptide of α1-homotrimer variant of human type I collagen (hotPINP) in malignant pleural effusion. Anticancer Res. 21, 2293–2296. Lee, S. J., Evers, S., Roeder, D., Parlow, A. F., Risteli, J., Risteli, L., Lee, Y. C., Feizi, T., Langen, H., and Nussenzweig, M. C. (2002). Mannose receptor-mediated regulation of serum glycoprotein homeostasis. Science 295, 1898–1901. Smedsrød, B., Pertoft, H., Gustafsson, S., and Laurent, T. C. (1990). Scavenger functions of the liver endothelial cells. Biochem. J. 266, 313–327. Smedsrød, B., Melkko, J., Risteli, L., and Risteli, J. (1990). Circulating C-terminal propeptide of type I procollagen is cleared mainly via the mannose receptor in liver endothelial cells. Biochem. J. 271, 345–350. Melkko, J., Hellevik, T., Risteli, L., Risteli, J., and Smedsrød, B. (1994). Clearance of NH2-terminal propeptides of types I and III procollagen is a physiological function of the scavenger receptor in liver endothelial cells. J. Exp. Med. 179, 405–412. Kivirikko, K. I. (1970). Urinary excretion of hydroxyproline in health and disease. Int. Rev. Conn. Tissue Res. 5, 93–163. Lamandé, S. R., and Bateman, J. F. (1995). The type I collagen proα1(I) COOH-terminal propeptide N-linked oligosaccharide. Functional analysis by site-directed mutagenesis. J. Biol. Chem. 270, 17858–17865.
404 30. Toivonen, J., Tähtelä, R., Laitinen, K., Risteli, J., and Välimäki, M. J. (1998). Markers of bone turnover in patients with differentiated thyroid cancer on and off thyroxine suppressive therapy. Eur. J. Endocrinol. 138, 667–673. 31. Sorva, A., Tähtelä, R., Risteli, J., Risteli, L., Laitinen, K., JuntunenBackman, K., and Sorva, R. (1994). Familial high serum concentrations of the carboxyl-terminal propeptide of type I procollagen. Clin. Chem. 40, 1591–1593. 32. Puistola, U., Risteli, L., Kauppila, A., Knip, M., and Risteli, J. (1993). Markers of type I and type III collagen synthesis in serum as indicators of tissue growth during pregnancy. J. Clin. Endocrinol. Metab. 77, 178–182. 33. Sorva, A., Sorva, R., Risteli, J., Risteli, L., and Autio, P. (2001). Familial high serum PICP: high even in skin interstitial fluid, suppressible by topical corticosteroid treatment. Clin. Chim. Acta 314, 153–157. 34. Haukipuro, K., Melkko, J., Risteli, L., Kairaluoma, M. I., and Risteli, J. (1991). Synthesis of type I collagen in healing wounds in humans. Ann. Surg. 213, 75–80. 35. Oikarinen, A., Autio, P., Kiistala, U., Risteli, L., and Risteli, J. (1992). A new method to measure type I and III collagen synthesis in human skin in vivo: demonstration of decreased collagen synthesis after topical glucocorticoid treatment. J. Invest. Dermatol. 98, 220–225. 36. Jensen, L. T., Olesen, H. P., Risteli, J., and Lorenzen, I. (1990). External thoracic duct-venous shunt in conscious pigs for long term studies of connective tissue metabolism in lymph. Lab. Anim. Sci. 40, 620–624. 37. Eriksen, E. F., Charles, P., Melsen, F., Mosekilde, L., Risteli, L., and Risteli, J. (1993). Serum markers of type I collagen formation and degradation in metabolic bone disease: correlation to bone histomorphometry. J. Bone Miner. Res. 8, 127–132. 38. Charles, P., Mosekilde, L., Risteli, L., Risteli, J., and Eriksen, E. F. (1994). Assessment of bone remodelling using biochemical indicators of type I collagen synthesis and degradation: relation to calcium kinetics. Bone Miner. 24, 81–94. 39. Price, K. M., Silman, R., Armstrong, P., and Grudzinskas, J. G. (1994). Development of a radioimmunoassay for fetal antigen 2. Clin. Chim. Acta 224, 95–102. 40. Ørum, O., Hansen, M., Jensen, C. H., Sørensen, H. A., Jensen, L. B., Hørslev-Petersen, K., and Teisner, B. (1996). Procollagen type I N-terminal propeptide (PINP) as an indicator of type I collagen metabolism: ELISA development, reference interval, and hypovitaminosis D induced hyperparathyroidism. Bone 19, 157–163. 41. Suvanto-Luukkonen, E., Risteli, L., Sundström, H., Penttinen, J., Kauppila, A., and Risteli, J. (1997). Comparison of three serum assays for bone collagen formation during postmenopausal estrogen-progestin therapy. Clin. Chim. Acta 266, 105–116. 42. Tähtelä, R., Turpeinen, M., Sorva, R., and Karonen, S.-L. (1997). The amino-terminal propeptide of type I procollagen: Evaluation of a commercial radioimmunoassay kit and values in healthy subjects. Clin. Biochem. 30, 35–40. 43. Sharp, C. A., Davie, M. W. J., Worsfold, M., Risteli, L., and Risteli, J. (1996). Discrepant blood concentrations of type I procollagen propeptides in active Paget disease of bone. Clin. Chem. 42, 1121–1122. 44. Sharp, C. A., Risteli, J., Risteli, L., Davie, M. W. J. (1997). Type I procollagen propeptide ratio (PICP/PINP) responds rapidly to pamidronate in Paget´s disease of bone. J. Bone Min. Res. 12(Suppl 1), S269. 45. Jukkola, A., Tähtelä, R., Thölix, E., Vuorinen, K., Blanco, G., Risteli, L., and Risteli, J. (1997). Aggressive breast cancer leads to discrepant serum levels of the type I procollagen propeptides PINP and PICP. Cancer Res. 57, 5517–5520. 46. Simojoki, M., Santala, M., Risteli, J., and Kauppila, A. (2003). Discrepant expression of carboxy- and aminoterminal propeptides of type I procollagen predicts poor clinical outcome in epithelial ovarian cancer. Gynecol. Oncol. 88, 358–362.
JUHA RISTELI AND LEILA RISTELI 47. Rasmussen, H. B., Teisner, B., Andersen, J. A., Brandrup, F., Purkis, T., and Leigh, I. (1991). Immunohistochemical studies on the localization of fetal antigen 2 (FA 2), laminin and type IV collagen in basal cell carcinoma. J. Cutaneous Pathol. 18, 215–219. 48. Kauppila, S., Bode, M., Stenbäck. F., Risteli, L., and Risteli, J. (1999). Cross-linked telopeptides of type I and III collagens in malignant ovarian tumors in vivo. Brit. J. Cancer 81, 654–661. 49. Ricard-Blum, S., Esterre, P., and Grimaud, J. A. (1993). Collagen cross-linking by pyridinoline occurs in non-reversible skin fibrosis. Cell. Mol. Biol. 39, 723–727. 50. Ricard-Blum, S., Bresson-Hadni, S., Guerret, S., Grenard, P., Volle, P.-J., Risteli, L., and Grimaud, J. A. (1996). Mechanism of collagen network stabilization in human irreversible granulomatous liver fibrosis. Gastroenterology 111, 172–182. 51. Pornprasertsuk, S., Duarte, W. R., Mochida, Y., and Yamauchi, M. (2004). Lysyl hydroxylase-2b directs collagen cross-linking pathways in MC3T3-E1 cells. J. Bone Miner. Res. 19, 1349–1355. 52. Bailey, A. J., Wotton, S. F., Sims, T. J., and Thompson, P. W. (1993). Biochemical changes in the collagen of human osteoporotic bone matrix. Connect. Tissue Res. 29, 119–132. 53. Oxlund, H., Mosekilde, L., and Øxtoft, G. (1996). Reduced concentration of collagen reducible cross links in human trabecular bone with respect to age and and osteoporosis. Bone 19, 479–484. 54. Hanson, D. A., and Eyre, D. R. (1996). Molecular site specificity of pyridinoline and pyrrole cross-linking in type I collagen in human bone. J. Biol. Chem. 271, 26508–26516. 55. Brady, J. D., and Robins, S. P. (2001). Structural characterization of pyrrolic cross-links in collagen using a biotinylated Ehrlich’s reagent. J. Biol. Chem. 276, 18812–18818. 56. Fledelius, C., Johnsen, A. H., Cloos, P. A. C., Bonde, M., and Qvist, P. (1997). Characterization of urinary degradation products derived from type I collagen. Identification of a β-isomerized Asp-Gly sequence within the C-terminal telopeptide (α1) region. J. Biol. Chem. 272, 9755–9763. 57. Kenyon, K., Contente, S., Trackman, P. C., Tang, J., Kagan, H. M., and Friedman, R. M. (1991). Lysyl oxidase and rrg messenger RNA? Science 253, 802. 58. Peyrol, S., Raccurt, M., Gerard, F., Gleyzal, C., Grimaud, J. A., and Sommer, P. (1997). Lysyl oxidase gene expression in the stromal reactions to in situ and invasive ductal breast carcinoma. Am. J. Pathol. 150, 497–507. 59. Sell, D. R., and Monnier, V. M. (1989). Structure elucidation of a senescence cross-link from human extracellular matrix. J. Biol. Chem. 264, 21597–21602. 60. Brady, J. D., Ju, J., and Robins, S. P. (1999). Isoaspartyl bond formation within N-terminal sequences of collagen type I: implications for their use as markers of collagen degradation. Clin. Science 96, 209–215. 61. Lowenson, J., and Clarke, S. (1988). Does the chemical instability of aspartyl and asparaginyl residues in proteins contribute to erythrocyte aging. Blood Cells 14, 103–117. 62. Garnero, P., Fledelius, C., Gineyts, E., Serre, C.-M., Vignot, E., and Delmas, P. D. (1997). Decreased β-isomerization of the C-terminal telopeptide of type I collagen α1 chain in Paget’s disease of bone. J. Bone Miner. Res. 12, 1407–1415. 63. Helfman, P. M., and Bada, J. L. (1976). Aspartic acid racemisation in dentine as a measure of ageing. Nature 262, 279–281. 64. Everts, V., van der Zee, E., Creemers, L., and Beertsen, W. (1996). Phagocytosis and intracellular digestion of collagen, its role in turnover and remodelling. Histochem. J. 28, 229–245. 65. Garnero, P., Borel, O., Byrjalsen, I., Ferreras, M., Drake, F. H., McQueney, M. S., Foged, N. T., Delmas, P. D., and Delaissé, J.-M. (1998). The collagenolytic activity of cathepsin K is unique among mammalian proteinases. J. Biol. Chem. 273, 32347–32352. 66. Clemens, J. D., Herrick, M. V., Singer, F. R., and Eyre, D. R. (1997). Evidence that serum NTx (collagen-type I N-telopeptides) can act as an immunochemical marker of bone resorption. Clin. Chem. 43, 2058–2063.
405
Chapter 23 Products of Bone Collagen Metabolism 67. Hanson, D. A., Weis, M. A. E., Bollen, A.-M., Maslan, S. L., Singer, F. R., and Eyre, D. R. (1992). A specific immunoassay for monitoring human bone resorption: quantitation of type I collagen cross-linked N-telopeptides in urine. J. Bone Miner. Res. 7, 1251–1258. 68. Blumsohn, A. C., Naylor, K., and Eastell, R. (1995). Effect of light and γ-irridiation on pyridinolines and telopeptide of type I collagen in urine. Clin. Chem. 41, 1195–1197. 69. Hartmann, D., Trinchet, J. C., Ricard-Blum, S., Beaugrand, M., Callard, P., and Ville, G. (1990). Radioimmunoassay of type I collagen that mainly detects degradation products in serum: application to patients with liver diseases. Clin. Chem. 36, 421–426. 70. Ju, J., Leigh, S., Byrne, M., van der Rest, M., Liu, V., and Krane, S. (1997). A new marker for measuring bone resorption using an ELISA for a 14 amino acid peptide (residues 620–633) derived from the collagen α1(I) chain. J. Bone Miner. Res.12 (Suppl), S178. 71. Ju, H.-S. J., Maskrey, J., Krane, S., Riggs, B. L., and Allen, G. (1998). An immunoassay for measuring helical peptide/ α1(I) 620-633 derived from type I collagen α1 chain. Bone 23 (Suppl), S629. 72. Garnero, P., and Delmas, P. D. (2003). An immunoassay for type I collagen α1 helicoidal peptide 620–633, a new marker of bone resorption in osteoporosis. Bone 32, 20–26. 73. Furthmayr, H. (1982). Immunization procedures, isolation by affinity chromatography, and serological and immunochemical characterization of collagen specific antibodies. In “Immunochemistry of the Extracellular Matrix Volume I Methods” (H. Furthmayr, ed.), pp. 143–178. CRC Press, Boca Raton, FL. 74. Kit insert (2001). Metra Helical Peptide EIA kit. Quidel, 0760A (5/01) Catalog number 8022, p. 1–3 (www.quidel.com, read August 17, 2004). 75. Taxel, P., Fall, P. M., Prestwood, K. M., Dulipsingh, L., Ohannessian, C., and Raisz, L. G. (2004). Changes in urinary excretion of helical peptide during therapy for osteoporosis in older adults. Clin. Chem. 50, 747–750. 76. Robins, S. P. (1982). Turnover and cross-linking of collagen. In “Collagen in Health and Disease” (J. B. Weiss, and M. I. V. Jayson, eds.), pp. 160–178. Churchill Livingstone, Edinburgh. 77. Bettica, P., Taylor, A. K., Talbot, J., Moro, L., Talamini, R., and Baylink, D. J. (1996). Clinical performances of galactosyl hydroxylysine, pyridinoline, and deoxypyridinoline in postmenopausal osteoporosis. J. Clin. Endocrinol. Metab. 81, 542–546. 78. Leigh, S. D., Ju, H. S., Lundgard, R., Daniloff, G. Y., and Liu, V. (1998). Development of an immunoassay for urinary galactosylhydroxylysine. J. Immunol. Methods 220, 169–178. 79. Risteli, J., Elomaa, I., Niemi, S., Novamo, A., and Risteli, L. (1993). Radioimmunoassay for the pyridinoline cross-linked carboxyterminal telopeptide of type I collagen: a new serum marker of bone collagen degradation. Clin. Chem. 39, 635–640. 80. Sassi, M.-L., Eriksen, H., Risteli, L., Niemi, S., Mansell, J., Gowen, M., and Risteli, J. (2000). Immunochemical characterization of assay for carboxyterminal telopeptide of human type I collagen: Loss of antigenicity by treatment with cathepsin K. Bone 26, 367–373. 81. Bonde, M., Qvist, P., Fledelius, C., Riis, B. J., and Christiansen, C. (1994). Immunoassay for quantifying type I collagen degradation products in urine. Clin. Chem. 40, 2022–2025. 82. Bonde, M., Garnero, P., Fledelius, C., Qvist, P., Delmas, P. D., and Christiansen, C. (1997). Measurement of bone degradation products in
83.
84.
85. 86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
serum using antibodies reactive with an isomerized form of an 8 amino acid sequence of the C-telopeptide of type I collagen. J. Bone Miner. Res. 12, 1028–1034. Rosenqvist, C., Fledelius, C., Christgau, S., Pedersen, B. J., Bonde, M., Qvist, P., and Christiansen, C. (1998). Serum CrossLaps One Step ELISA. First application of monoclonal antibodies for measurement in serum of bone-related degradation products of type I collagen. Clin. Chem. 44, 2281–2289. Gineyts, E., Cloos, P. A. C., Borel, O., Grimaud, L., Delmas, P. D., and Garnero, P. (2000). Racemization and isomerization of type I collagen C-telopeptides in human bone and soft tissues: assessment of tissue turnover. Biochem. J. 345, 481–485. Väänänen, H. K., Zhao, H., Mulari, M., and Halleen, J. M. (2000). The cell biology of osteoclast function. J. Cell Sci. 113, 377–381. Atley, L. M., Mort, J. S., Lalumiere, M., and Eyre, D. R. (2000). Proteolysis of human bone collagen by cathepsin K: Characterization of the cleavage sites generating the cross-linked N-telopeptide neoepitope. Bone 26, 241–247. Sassi, M.-L., Åman, S., Hakala, M., Luukkainen, R., and Risteli, J. (2003). Assay for cross-linked carboxyterminal telopeptide of type I collagen (ICTP) unlike CrossLaps assay reflects increased pathological degradation of type I collagen in rheumatoid arthritis. Clin. Chem. Lab. Med. 41, 1038–1044. Nishi, Y., Atley, L., Wyre, D. E., Edelson, J. G., Superti-Furga, A., Yasuda, T., Desnick, R. J., and Gelb, B. D. (1999). Determination of bone markers in pycnodysostosis: effects of cathepsin K deficiency on bone matrix degradation. J. Bone Min. Res. 14, 1902–1908. Foged, N. T., Delaissé, J.-M., Hou, P., Lou, H., Sato, T., Winding, B., and Bonde, M. (1996). Quantification of the collagenolytic activity of isolated osteoclasts by enzyme-linked immunosorbent assay. J. Bone Miner. Res. 11, 226–237. Apone, S., Lee, M. Y., and Eyre, D. R. (1997). Osteoclasts generate cross-linked collagen N-telopeptides (Ntx) but not free pyridinolines when cultured on human bone. Bone 21, 129–136. Parikka, V., Lehenkari, P., Sassi, M.-L., Halleen, J., Risteli, J., Härkönen, P., and Väänänen, H. K. (2001). Estrogen reduces the depth of resorption pits by disturbing the organic bone matrix degradation activity of mature osteoclasts. Endocrinology 142, 5371–5378. Garnero, P., Ferreras, M., Karsdal, K. A., Nicamhlaoibh, R., Risteli, J., Borel, O., Qvist, P., Delmas, P. D., Foged, N. T., and Delaissé, J. M. (2003). The type I collagen fragments ICTP and CTX reveal distinct enzymatic pathways of bone collagen degradation. J. Bone. Miner. Res. 18, 859–867. Garnero, P., Gineyts, E., Arbault, P., Christiansen, C., and Delmas, P. D. (1995). Different effects of bisphosphonate and estrogen therapy on free and peptide-bound bone cross-links excretion. J. Bone Miner. Res. 10, 641–649. Ruckidge, G. J., Riddoch, G. I., Williams, L. M., and Robins, S. P. (1988). Autoradiographic studies of the renal clearance of circulating Type I collagen fragments in the rat. Collagen Rel. Res. 8, 339–348. Colwell, A., and Eastell, R. (1996). The renal clearance of free and conjugated pyridinium cross-links of collagen. J. Bone Miner. Res. 11, 1976–1980.
This Page Intentionally Left Blank
Chapter 24
Supramolecular Structure of Cartilage Matrix Peter Bruckner
Department of Physiological Chemistry and Pathobiochemistry, University of Münster, Münster, Germany
IX. Studies of Transgenic Mice and of Human Genetic Matrix Diseases X. Correlating Structure with the Biomechanical Role of Articular Cartilage XI. Models of Cartilage Fibril Structure XII. Future Perspectives References
I. Summary II. Introduction III. Light and Electron Micrography IV. Biochemistry of Cartilage V. Studies of Fibril Structures by X-ray Diffraction VI. Structure of Fibril Fragments Obtained by Mechanical Disruption of Tissue VII. Studies of Collagen Cross-linking in Cartilage Fibrils VIII. Reconstitution of Aggregates from Soluble Collagens and other Macromolecules
I. SUMMARY
which is rich in polyanionic carbohydrates and, thus, water. Implications of this concept are discussed for connective tissues other than cartilage and for extracellular matrices in general.
The extracellular matrix of hyaline cartilage comprises the great majority of the volume of this tissue and is responsible for its properties. The matrix consists of two basic suprastructural components: fibrils with a D = 67 nm periodic banding pattern and an electron-lucent extrafibrillar matrix with no conspicuous features. The structural properties of these matrix constituents and their functional implications are discussed. The heterotypic fibrils contain not only several types of collagen but also noncollagenous macromolecules. These molecular components specifically self-assemble, either into distinct domains within the fibrils or, more attractively, into aggregates with characteristics of metal alloys in many respects. Thus, specific properties of the fibrils are indirectly determined by the mass proportions of their molecular components. Similar considerations apply to the extrafibrillar matrix Dynamics of Bone and Cartilage Metabolism
II. INTRODUCTION Three types of cartilage have been distinguished on the basis of histological criteria and biomechanical properties: hyaline, elastic, and fibrous cartilages. The most prevalent type is hyaline cartilage, which is a visually uniform, translucent tissue found in the skeleton of all vertebrates. Articular cartilage, the most familiar hyaline cartilage, forms the smooth gliding surfaces of joints, such as the knee and hip, that permit locomotion in animals. Injuries to this tissue and common diseases such as osteoarthritis, impair joint mobility and constitute a major 407
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
408
PETER BRUCKNER
focus of modern medicine. Hyaline cartilage also comprises growth plate, the transient template required for endochondral bone formation in fetal development, skeletal growth, and repair. In addition, hyaline cartilage occurs as a permanent structural tissue in costal cartilages and tracheal reinforcing rings. Elastic and fibrous cartilage are less widely distributed. Flexibility is the hallmark of elastic cartilage, found for example in the outer ear, and is a consequence of its large content of elastic fibers. Meniscus and annulus fibrosus are fibrous cartilages in which bundles of banded fibers, similar to those in tendon, resist tensile forces generated by the load-bearing functions of these tissues. These classes of cartilage will not be considered further in this review. The biomechanical requirements of cartilage are met by the extracellular matrix that constitutes the predominant fraction of the tissue. By contrast, the cellular component, the chondrocytes, occupy a relatively small volume, with each cell embedded into extracellular matrix with no direct cell-to-cell contacts. The matrix properties are determined in turn by tissue-specific supramolecular structures, such as periodically banded fibrils, filaments, and proteoglycan aggregates. This review will outline the current frontiers of our knowledge in this field. Cartilage morphology has been studied by light microscopy, electron microscopy, and X-ray diffraction. These techniques have been complemented by biochemical investigation of isolated aggregates and their self-assembly in vitro, and by genetic studies of human diseases and transgenic animals.
III. LIGHT AND ELECTRON MICROGRAPHY At the level of light microscopy, articular cartilage matrix is an almost homogeneous mass of negatively charged material well stained by basic dyes such as alcian blue or toluidine blue. However, staining discontinuities do highlight specific matrix domains of adult articular cartilage. Such domains are much less conspicuous in immature cartilage, e.g. in fetal or neonatal epiphyseal cartilage. Territorial regions are seen near the cell surfaces which are distinct from the interterritorial matrix in the large intervening spaces. In his early investigations of adult articular cartilage by light microscopy, Benninghoff postulated a reinforcement of the tissue by fibers forming arcades with their bases in the subchondral bone and arches at the articular surface [1]. This arcade-like organization of the fibrils buttresses the tissue against shear forces in the superficial layers and secures the tissue to the underlying bone.
Near the osteochondral junction, a prominent lightmicroscopic feature is the sharply defined limit of calcification of the tissue which is termed the tide-mark. Growth plate, the cartilage responsible for the longitudinal growth of long bones, does not show a tide-mark, but otherwise the salient features of its matrix structure are similar to those of adult articular cartilage. Electron microscopy has been extensively used to visualize the supramolecular structures in hyaline cartilage matrix. General features include a network of banded fibrils and an extrafibrillar matrix with no conspicuous ultrastructure that was once referred to as “amorphous ground substance”. In the immediate vicinity of the cells, there is a small space apparently lacking fibrils altogether. This region is known as the pericellular matrix and mainly contains a specialized extrafibrillar matrix that is easily extractable from the tissue during conventional histological fixation (see below). As apparent already by light microscopy, the matrix structure in immature cartilage is simpler than in adult tissue. Fibrils in fetal or neonatal cartilage have a round cross-section with a uniformly small diameter of about 20 nm which exhibit a faint longitudinal banding pattern [2]. They are almost randomly orientated throughout the tissue, but tend to be parallel to tissue or cell surfaces near to those boundaries. The architecture of adult hyaline cartilages is more complex. The fibril diameters are heterogeneous and depend on the precise anatomical localization within the tissue. In articular cartilage, for example, the fibrils strongly align with joint surfaces, thereby forming conspicuous two-dimensional mats [2]. In the intermediate zones between joint surfaces and the junctures with the subchondral bone, fibrils are more randomly distributed in three dimensions. Reconstructions from thick sections show fibrils with a kinked morphology woven together in a threedimensional knit [3]. In deep zones, fibrils are nearly parallel to the long-bone axis. This distribution of fibril diameters and orientations corresponds well to the Benninghoff arcade model described above. The structure of adult cartilage fibrils also varies between the territorial and interterritorial regions defined by light microscopy. As in immature cartilage, the immediate surrounding of the cell consists of an essentially fibril-free pericellular matrix. Further removed from the cell surface there is a weave of 20-nm fibrils forming a basket around the cells. Such basket structures can be isolated after mechanical disruption of the tissue and have been termed chondrons to designate a functional unit comprising a chondrocyte in its immediate extracellular environment [4]. Chondrons may contain more than one cell, in which case individual cells are separated by a thin
Chapter 24 Supramolecular Structure of Cartilage Matrix
409
septum of territorial matrix. Moving away from the cell surface, the territorial matrix gradually becomes the interterritorial region, in which fibrils progressively aquire diameters of 50–100 nm and have striking longitudinal cross-banding patterns. The period of the banding pattern is D, the characteristic periodicity of collagen fibrils, which is typically about 65 nm in electron microscopy, but varies with the degree of shrinkage encountered in the preparation of the specimen (see below). With traditional chemical fixation and embedding in hydrophobic resins, a number of features are observed that are now thought to be artifacts. Among these is the formation of lacunae inside the territorial matrix showing cells with apparent pseudopodia anchored in the fibril network [5]. This feature appears to be a consequence of shrinking of the cells and a loss of matrix macromolecules from the cellular environment. Also, in the extrafibrillar matrix, condensations of material into electron dense globules result from the artifactual precipitation of polyanionic proteoglycans [6, 7]. The recent development of improved chemical fixation, cryopreservation and freeze substitution techniques appears to have overcome these problems and allowed a more detailed analysis of cartilage matrix suprastructures. Under these conditions, the original round shape of chondrocytes is preserved and only small tube-like cellsurface projections are visible which may be part of the secretory machinery of the cells. Lacunae, however, are no longer observed. Further, filaments are seen in the pericellular matrix with a diameter of 10–15 nm and a very faint D-periodic banding pattern [7]. Because this feature is a hallmark of collagen-containing fibrils, the filaments may represent precursors of fibrils in the territorial matrix. Such filaments are also seen throughout the matrix, including the interterritorial regions, where they are intermingled in random orientation with the large fibrils. In the interterritorial matrix, the thicker fibrils often fan out into smaller fibrils closely resembling those of the territorial matrix, giving the impression that the well-banded interterritorial fibrils arise by fusion of archetypal 20-nm fibrils [8]. This is supported by observations of fibrils in cross-section: larger fibrils have an irregular outline and are surrounded by round fibrils that closely resemble the 20-nm fibrils. Verification that large fibrils are assembled by fusion of smaller ones rather than by accretion of individual molecules presents a challenge for future studies in cartilage structural biology. Another detail visualized by the observation of unstained sections of cartilage vitrified at high pressure is the presence of a thin water-rich layer surrounding the fibrils [9]. This implies the existence of a previously unknown transition zone between the fibril surface and
the extrafibrillar matrix, perhaps 5 nm in width, which has a lower density than either of these components. It will be of great interest to find the biochemical correlate of this zone. More recent studies have revealed considerable differences in the fibrillar organization in articular cartilage. For example, tube-like structures, consisting of parallelly aligned, thin fibrils are found extensively in the territorial zones in rabbit articular cartilage [10, 11] whereas fibrils are organized into parallel sheets in the proximal joint cartilage of mouse tibiae [12]. The tubes and the sheets of fibrillar matrix in these tissues have been related to the viscoelastic properties and fluid flow upon load bearing. The biomechanical properties of bovine articular cartilage change with age, diverge within different matrix compartments, and correlate with the density of the fibrilar networks [13].
IV. BIOCHEMISTRY OF CARTILAGE A vast amount of information is available on cartilage matrix macromolecules, which is discussed in detail in Chapters 1-5 of this volume. The extrafibrillar matrix is a complex of carbohydrates and proteins with a high negative charge density which leads to the binding of large amounts of water. Its main macromolecules are aggrecan, hyaluronan, and link protein (LP) which, together, form bottle-brush-like aggregates that have been reconstituted from the purified components and observed by electron microscopy of rotary shadowed preparations [14–16]. Such complexes are known to bind matrilin-1 (also called cartilage matrix protein, CMP) as a further component [17]. Although a considerable degree of structural organization has thus been demonstrated for isolated extrafibrillar matrix components, electron microscopy has failed to visualize the ultrastructure of these components in tissues. Cartilage fibrils are mostly proteinaceous and contain collagens II and XI. Fibrils initially deposited by chondrocytes also contain collagens IX and XVI in a mutually exclusive manner [18]. Such fibrils, here termed prototypic cartilage fibrils, preferentially or exclusively occur in the territorial matrix zones, have a uniform diameter of ca. 20 nm and exhibit a weak D-periodical banding pattern. The collagens also interact with noncollagenous components, most notably with leucine-rich proteins or proteoglycans. Decorin is the most abundant representative of this protein family [19, 20]. Furthermore, binding to collagens has also been reported for fibromodulin, biglycan, matrilins-1 and -3, and cartilage oligomeric matrix protein (COMP; also called thrombospondin-5).
410 COMP directly binds to collagens in a zinc-dependent manner [21] and has been shown to be a component of the fibril periphery in situ (Budde et al., unpublished). Matrilins-1 and-3 are particularly interesting components of the fibril periphery since they have a strong binding affinity for aggrecan [17] and also bind to fibrillar components, either directly to collagens or to COMP (Budde et al., unpublished). Other cartilage components described in Chapter 4 are likely to reside within the extrafibrillar matrix, but their localization also remains to be established. The major collagenous component of hyaline cartilage is collagen II which comprises up to 50% of the dry weight of the tissue. Collagens II and XI are defined as fibril-forming collagens in that they can form fibrils as pure proteins. The other fibril-forming collagens, types I, III, and V, were conventionally thought to be absent from cartilage, but type I is reported to comprise as much as 10% of the collagenous component of normal hyaline cartilage [22]. Collagen III is a very minor component and occurs together with collagen II in the same cartilage fibrils [23]. Collagen V with the chain composition α1(V)2α2(V) is not a cartilage protein but, in adult articular cartilage, about one-half of the α1(XI)-chains in collagen XI are replaced by structurally similar α1(V)chains [24]. The levels of the minor collagens are reported to increase in osteoarthritic cartilage [25–27] but this may be a consequence of increased solubility of collagens in general in diseased tissues. Minor nonfibril-forming collagens that have been reported in cartilage include type VI [22, 28], and types XII and XIV [29]. Collagen VI predominantly occurs in the territorial regions or in isolated chondrons [30]. The protein forms filamentous structures with a periodicity of 100 nm [31] that are independent of D-periodically banded fibrils. Collagen VI, however, also can be organized into hexagonal lattices by co-polymerizing with biglycan or, to a lesser extent, with decorin [32]. By contrast, collagens XII and XIV are likely to be constituents of D-periodic fibrils in skin [33] but their exact suprastructural organization in cartilage still remains to be determined. The polypeptides of fibril-forming collagens consist of uninterrupted long sequences of repeating Gly-X-Y amino acid residue triplets, where X and Y are arbitrary, but are frequently proline and hydroxyproline, respectively. Three chains of such a repeating primary sequence fold into a triple-helical collagen molecule with a length of 300 nm and a diameter of 1.5 nm [34]. Molecules of collagen pack laterally into “quarter-staggered” arrays as shown in Figure 1, in which adjacent molecules are staggered longitudinally by a multiple of D = 67 nm, which
PETER BRUCKNER
Figure 1
Changes in cartilage fibril structure with diameter. (A) Uniform diameter 20-nm fibrils of quarter-staggered collagen molecules illustrating the gap–overlap structure. Collagen XI molecules (gray) are shown with amino-terminal propeptides protruding from the surface of the fibril, which may act to control diameter. Collagen IX (hatched) is shown at the surface with the COL3 and NC4 (spherical) domains projecting from the fibril axis. The dark zig-zag represents the glycosaminoglycan (GAG) component covalently attached to the type IX molecules. (B) Intermediatediameter fibril showing both collagen IX and decorin incorporated into the surface. GAG presented at the surface derives from both collagen IX and decorin. (C) Large-diameter fibril showing only decorin, and its attendant GAG on the surface. Collagen IX, perhaps processed, may remain in the interior of the fibril.
corresponds to 234 amino acid residues [35]. The molecules thus are about 4.4 D in length, which gives rise to the well-known gap–overlap longitudinal structure of collagen, where the gap, or hole zone, is about 0.6 D in length [36]. The longitudinal D period is the most obvious feature of collagenous fibrils both in electron microscopy and by X-ray diffraction, the latter technique giving the precise value of D for fully hydrated native specimens. Dermatan sulfate and keratan sulfate proteoglycans are important structural components of fibrils in cartilage and have been visualized in tissues and isolated fibrils by staining with the cationic dye cupromeronic blue under conditions of critical electrolyte concentration [37]. Filamentous chains of proteoglycan are seen with highly regular D-periodic distribution. These have been interpreted as integral components of cartilage fibrils and serve to tether fibrils in position within the matrix. Decorin has been shown to bind to individual collagen I and II molecules near the N-terminal end [38]. This location would place it at the “d” band within the gap of the fibril structure, a location to which proteoglycan has been mapped by cupromeronic blue staining [39]. The proteinaceous part of decorin is a member of a large family of proteins containing leucine-rich repeating (LRR) sequences. The structure of ribonuclease inhibitor,
Chapter 24 Supramolecular Structure of Cartilage Matrix
411
another member of the LRR family, though not obviously related to decorin, has been solved by X-ray crystallography [40]. A detailed structural model of decorin binding based on homology to the horseshoe-shaped ribonuclease inhibitor structure has been proposed [41]. Energy calculations of the model support the assignment of the binding to the “d” band of collagen fibrils, but other nearby sites are also possible. Fibromodulin, like decorin, another LRR proteoglycan, has been shown to bind specifically to collagen II [42]. Decorin and fibromodulin have one binding site each per collagen and do not compete with each other [43, 44]. Hence, they must occupy separate specific sites on the collagen molecule. Thus, proteoglycans, including decorin, fibromodulin, and collagen IX, appear to have important specific roles in establishing and maintaining the fibrils of cartilage.
notochord, the collagen II-containing fibrils are apparently crystalline, but the uniform diameter fibrils are too small, ca. 20 nm, to give high-resolution diffraction [53]. Both in this tissue and in mammalian cartilages the average center-to-center intermolecular spacing in the fully hydrated state is ca. 1.7 nm, significantly higher than in collagen I-containing tissues [54]. This is consistent with a higher number of glucosyl-galactosyl-hydroxylysine residues found in collagen II: the levels of this posttranslational modification are 50–100% higher than in type I collagen. This results in a higher intrafibrillar water content in cartilage [55, 56]. Dried cartilage shows approximately the same intermolecular spacing (ca. 1.1 nm) as tendon, demonstrating that the collagen II molecular diameter is not significantly increased by the additional sugar content. The other fibril-forming collagen of cartilage, type XI, has an even higher degree of glycosylation than collagen II; however, there are no X-ray diffraction data specific for this collagen type. Glycation, the nonenzymatic reaction by glucose with protein amino groups that occurs naturally in tissues with slow protein turnover, has shed light on the lateral organization of collagen in fibrils. Rat tail tendon was subjected to cross-linking by ribose as a model for the age-related effects of glycation [57]. It was found that the average intermolecular spacing increased with the degree of cross-linking. This was interpreted as showing that the separation of collagen molecules is a weakly constrained equilibrium such that there is little energetic cost involved in swelling or shrinking the structure by the exchange of water [53]. This supports the idea that collagen II has increased intrafibrillar water because of its increased post-translational hydroxylysyl glycosylation. The biomechanical effects of age-related glycation in cartilage fibrils have also been modeled by ribosylation, but resulting changes in intrafibrillar water content were not reported [58].
V. STUDIES OF FIBRIL STRUCTURES BY X-RAY DIFFRACTION X-ray diffraction has been used to determine the structure of collagen in rat tail tendon nearly to atomic resolution [45, 46]. For this reason and because of the biochemical similarities in fibril-forming collagens, the fibrils in this tissue have often been considered as model structures of collagen-containing fibrils in general, a notion that later has been revised [47]. Rat tail tendon is almost unique among connective tissues in that collagen molecules are packed in molecular crystals that diffract to high resolution. A analysis is consistent with a tissue comprised of microfibrils of five-triple helical collagen molecules twisted into left-handed strands distorted by crystal packing forces into a quasi-hexagonal array [48, 49]. These arrays exhibit coherently diffracting domains up to 100 nm in lateral size, considerably smaller than the size of the fibrils (up to 600 nm) in which they reside. Thus, the large fibrils seen by electron microscopy are not single crystals, but mosaics of smaller crystallites [50]. The average center-to-center intermolecular spacing in hydrated rat tail tendon is ca. 1.5 nm. Most other tendons do not give crystalline patterns, but X-ray diffraction can still be used in these specimens to evaluate the intermolecular spacing. In other fully hydrated, collagen I-containing tissues, this spacing is also near 1.5 nm [51]. In the dry state, this distance collapses to about 1.1 nm, demonstrating that the collagen molecules in fibrils are separated from each other by substantial quantities of intrafibrillar water [52]. Cartilage fibrils have also been extensively studied by X-ray diffraction. In one case, the sheath of the lamprey
VI. STRUCTURE OF FIBRIL FRAGMENTS OBTAINED BY MECHANICAL DISRUPTION OF TISSUE Another approach to study the structure of cartilage fibrils has been to prepare fragments of fibrils from cartilage or from chondrocyte cultures. Such fibril preparations, contrasted by negative or positive stain with or without immuno-gold labeling, or by rotary shadowing, have yielded considerable information. Both cell cultures and tissues yield the thin fibrils characteristic of cartilage which have been shown by immunostaining to be heterotypic, in that individual fibrils contain collagens II,
412 IX, and XI. Biochemical analysis of fragments isolated from chick embryo cartilage revealed a mass proportion of 8:1:1 for these collagen types, respectively [59]. Interestingly, collagen XI in fibril fragments from chick embryo sterna exhibited a strong immunochemical masking toward antibodies raised against the long triple helix of the protein. Upon disruption of the collagen packing by partial digestion with pepsin, however, this immunochemical masking was released [60]. Employing immunogold labeling with antibodies directed to the aminoterminal nonhelical domain, collagen XI was localized to fibrils only of small diameter in tissue sections of human rib and growth plate cartilage [61]. By contrast, undigested fibril fragments were readily labeled with antibodies to the pepsin-treated forms of both collagens II and IX [60]. Taken together, these results raise the possibility that collagen XI is buried within the fibrillar body, and that, by forming a core structure, collagen XI regulates apposition of types II and IX onto the fibrils. Rotary shadowed fibrils exhibit a D-periodic occurrence of collagen IX, with the amino-terminal domains NC4 and COL3 projecting from the fibril axis (see Chapter 1 for a discussion of these domains). This has been interpreted as showing collagen IX to be a surface component [59]. However, fibrils of older mammalian cartilage are biochemically heterogeneous, with the detailed composition depending on which tissue layer is sampled. Small, uniform fibrils characteristically have collagen IX at their surface, whereas large fibrils have decorin. Some intermediate diameter fibrils stain for both collagen IX and decorin [20] (Fig. 1). Since both of these molecules are glycoproteins carrying dermatan sulfate chains in cartilage, the fibrils are studded with anionic groups which could help to secure the extrafibrillar matrix to the fibrils.
VII. STUDIES OF COLLAGEN CROSSLINKING IN CARTILAGE FIBRILS Naturally occurring cross-links between collagen polypeptides in cartilage have been used as indicators for the fibrillar organization (the occurrence and chemistry of such cross-links is discussed in Chapter 2). Collagen IX purified from steer cartilage was found to have 1.7–2.3 mol of cross-linking residues to collagen II per mol of collagen, consistent with the heterotypic nature of cartilage fibrils [62]. If it is assumed that the collagen IX molecules lie parallel to the fibril axis, then the locations of these links indicate that the collagen IX is oriented antiparallel to the type II and covering both the gap and the overlap region (Fig. 1). However, it is also possible to
PETER BRUCKNER
constuct models in which the type IX wraps around, or partially penetrates, the fibril, in which cases the molecular configuration cannot be specified. Collagen IX is also cross-linked to other molecules of collagen IX. Collagen XI is cross-linked as well, largely to other type XI molecules, but to a small degree also to collagen II [63]. This observation has been interpreted as supporting the idea that collagen XI forms a core on which collagen II is deposited. Cross-links to other fibrillar and nonfibrillar components have not yet been described.
VIII. RECONSTITUTION OF AGGREGATES FROM SOLUBLE COLLAGENS AND OTHER MACROMOLECULES One of the fascinating questions concerning cartilage matrix structure is what mechanism controls the diameter of the thin, uniform fibrils. Fibril reconstitution experiments with purified components have been used to approach this question. In such experiments, collagens in solution are subjected to temperature and buffer conditions under which they spontaneously aggregate into fibrils. Chicken sternal chondrocytes from 17-day embryos grown in agarose maintain the cartilage phenotype and synthesize uniform 20-nm fibrils. Collagens II, XI, and IX have been purified from these cultures by salt fractionation and ion exchange chromatography, yielding essentially pure components in fibril-competent form [64]. Fibrils reconstituted from the mixture of the three collagen types directly extracted from the cultures gave uniform D-periodic fibrils closely resembling authentic 20-nm fibrils by their appearance in the electron microscope after negative contrasting. Collagen XI alone also formed small, uniform fibrils of about the same diameter. The kinetics of in vitro fibrillogenesis was followed by turbidometry and by electron microscopy. As reported previously [65], collagen II alone developed turbidity only after a lag time of 90–120 min. After this lag phase, the protein formed wide tactoids which, in the electron microscope, appear as relatively short, D-periodically banded aggregates with tapered ends and without apparent diameter control. Aggregation of collagen XI alone resulted in a hyperbolic increase in turbidity with no discernible lag phase at all. Immediately after initiation of in vitro fibrillogenesis, within the limits of time resolution of about 1 min, the protein assembled into long and flexible, small-diameter filaments. As turbidity developed to moderate amplitudes, the filaments turned into very weakly cross-banded fibrils with tightly controlled diameters of ca. 20 nm.
Chapter 24 Supramolecular Structure of Cartilage Matrix
413
Mixtures of collagens II and XI, at mass proportions of up to 8:1, respectively, produced turbidity characteristics without lag phase similar to those of pure collagen XI and the final fibrils also had a closely similar appearance. Therefore, the diameter control is inherent to these two components alone. In addition, the absence of a lag phase in the binary reconstitution system of collagens II and XI indicated that collagen XI was a powerful nucleator of fibril formation in cartilage. Furthermore, the 8:1 limit of collagen proportions and the 20-nm diameter match the corresponding properties of authentic chick embryo sternal cartilage fibrils. Thus, fibrils jointly formed by collagens II and XI have characteristics that vastly differ from those of aggregates formed by both individual collagens and, therefore, represent macromolecular alloys [66]. At larger proportions of collagen II, two-phase turbidity curves were observed and 20-nm fibrils and tactoids coexisted in the final reconstitution products. However, unlike the case of pure collagen II undergoing in vitro fibrillogenesis, tactoids formed in these supercritical mixtures of collagens II and XI were already present within the lag period of the second increase in turbidity. This necessitates that such tactoids were heterotypic, in that they contain both collagen types, and that the decision on the shape of the final aggregate occurs during the early phase of assembly and is determined by the biochemical composition of the early aggregates. This conclusion is illustrated by the drawing in Figure 2. In addition, many tactoids were continuous with the 20-nm
fibrils, suggesting that tactoid formation nucleated on fibrils [64]. Unlike collagen II extracted from natural sources, collagen I purified from tendons or skin can be reconstituted into large, strongly banded fibrils. This self-assembly system, again, is associated with a considerable lag phase of turbidity development and displays little if any diameter control. However, the reconstitution products share many morphological features which are characteristic for natural collagen I-containing fibrils. They are rigid ropelike objects, with each individual fibril maintaining a constant diameter over very long distances. Ends are very rarely observed, since the length of the reconstituted fibrils usually far exceeds the dimensions of the apertures of commonly used EM grids [67]. However, even exceedingly small quantities of minor fibrillar collagens, such as collagen XI, can profoundly alter both the aggregation kinetics and the morphology of the reconstitution products. The length of the lag phase of turbidity development is strongly reduced by collagen XI but, unlike in mixtures of collagens II and XI, occurs even when collagen XI represents a large or even predominant mass fraction. Therefore, collagen XI molecules can interact with those of collagen I and strongly enhance the aggregation efficiency of collagen I. The strict diameter control characteristic for the collagen II/XI system, however, is not observed in mixtures of collagens I and XI. In addition, fibrils reconstituted from collagens I and XI are composites containing alloyed cores of collagens I and XI and homotypic collagen I sheaths varying considerably
Figure 2
Illustration of the different aggregates resulting from fibril formation with different stoichiometries of collagens II and XI. (A) Tactoids are formed in mixtures with an excess of collagen II (greater than 8:1), probably as a result of separation of a collagen II-rich phase characterized by tight lateral packing of identical molecules. (B) The native proportion of collagens II and XI result in uniform 20-nm fibrils.
414
PETER BRUCKNER
in thickness. All these differences between mixtures of collagens I and XI or collagens II and XI occur in spite of the extraordinary sequence similarity in the collagenous regions of all polypeptides constituting collagens I and II [68]. The causes underlying these differences remain to be identified.
IX. STUDIES OF TRANSGENIC MICE AND OF HUMAN GENETIC MATRIX DISEASES Studies of mutations in genes encoding matrix macromolecules have also contributed to our current understanding of cartilage matrix organization. A large repertoire of mutations in human and animal genes for almost all macromolecules discussed above has been reported. Information comes from experimentally induced mutations in transgenic mice as well as from naturally occurring mutations in human and animal diseases. Consequences of mutations in collagen genes to cartilage suprastructure will primarily be discussed here. The mutations and their pathological consequences have been catalogued in recent reviews [69, 70]. In-frame deletions resulting in a loss of peptide sequences within the triple-helical domain in the collagen II gene (COL2A1 in man or Col2a1 in mice) lead to chondrodysplasia of variable severity, depending on the extent and the location of the deletion within the gene. Such mutations result in shortened α1(II) chains that are incorporated into homotrimeric collagen II molecules, together with either normal or other mutated chains. This explains the dominant phenotype of the mutations. However, the penetrance of the mutations varies not only between mice with different genetic backgrounds but even between individual inbred animals. The ability of cells to eliminate mutated mRNA or polypeptide species may be subject to variation between individual animals, which may explain the heterogeneous severity of the consequences of a given mutation. Triple-helix formation is impaired in affected molecules which, thereby, frequently have a lowered thermal stability and a compromised competence of incorporation into fibrillar aggregates. Due to the D-periodic regularity of collagen II molecules within the fibrils, incorporation of shortened molecules, even though they are secreted by the cells in reduced amounts, will destroy the entire fibrillar organization. This translates into a generalized injury to the structure of the cartilage matrix and its functions. Point mutations, particularly those exchanging glycine residues in the Gly-X-Y repeats of the triple-helical domain, have similar consequences. Such mutations have shown that the formation of
cartilaginous tissue does not depend on the presence of collagen II: in patients with hypochondrogenesis or achondrogenesis, collagen II, though synthesized, was not secreted by chondrocytes, but an abnormal mixture of collagens I and III was present [71, 72]. The resulting cartilage even matured sufficiently to produce collagen X. Bone formation, however, was severely disrupted in these lethal mutations. In general, abnormalities within carboxyterminal regions of the triple helix tend to result in more severe phenotypes, since collagen folding proceeds from the carboxyl- towards the amino-terminal end in a zipperlike fashion [73]. Interestingly, transgenic mice harboring extra copies of the normal Col2a1 gene have a perinatally lethal phenotype. Although no gross abnormalities were found in their cartilages, wide and strongly banded fibrils occurred in the interterritorial regions. In these animals, the balance in the amounts of collagens II and XI is altered due to an overexpression of collagen II resulting from transcription and translation of the transgenes. This agrees well with the formation in vitro of tactoids rather than normal cartilage-like fibrils from solutions containing collagen II above physiological levels (see above). Mice homozygous in a nonfunctional collagen II gene (type II knockout mice) also have been reported [74, 75]. These animals show extensive abnormalities of their endoskeleton but, surprisingly, survive until birth. They extensively develop cartilage tissue in which, however, collagen II is substituted, at least in part, by type I. Poorly organized fibrils occur in much smaller numbers than in normal cartilage and lack discernible longitudinal banding patterns as well as clearly defined lateral limits. The reasons for this fibrillar disorganization are not immediately clear. However, since the α3(XI) chains are the same gene product as the α1(II) chains (see Chapter 1), collagen XI production is also affected by the knockout of the Col2a1 gene. The thermal stability of collagen XI is lowered and, hence, the protein occurs in lower quantities due to degradation in the tissue. However, collagen XI very strongly nucleates formation of heterotypic fibrils containing collagens I and XI, whereas collagen I alone only slowly forms fibrils at relatively high monomer concentrations [68]. Thus, in cartilage of mice with homozygously inactivated Col2a1 alleles, the virtual absence of banded fibrils is a consequence of collagen XI deficiency rather than a lack of the major cartilage collagen type II. Consistent with this notion, the abnormally produced collagen I, due to a lack of incorporation into supramolecular aggregates, is extractable from cartilage in markedly increased quantities [75]. Heterozygous mice with only one inactivated Col2a1 allele survive into adulthood and
Chapter 24 Supramolecular Structure of Cartilage Matrix
415
also can reproduce. However, the lowered dose of active Col2a1 genes results in premature vertebral endplate ossification and mild disk degeneration [76] and weakens the fibrillar organization in articular cartilage [77]. If heterozygous inactivation of the Col2a1 gene results in reduced translation of α1(II) polypeptide chains in articular chondrocytes collagens II and XI would occur at molar fractions different from 8:1. Hence, prototypic cartilage fibrils would be formed at abnormally low levels [64], which may well explain the cartilage phenotype of heterozygous collagen II knockout mice. A variety of mutations affecting the genes of collagen XI chains also have been discovered and, again, disturbed formation of fibrils as a consequence of collagen XI abnormalities well explains the phenotypes observed. A single base deletion in the gene encoding the α1(XI) chain occurs naturally in the cho/cho mouse strain [65]. This represents a frame-shift mutation leading to a prematurely terminated α1(XI) polypeptide which either is not efficiently synthesized or is prematurely degraded. Collagen XI protein is absent in the cartilage of these mice and, again in keeping with in vitro fibrillogenesis of cartilage collagen mixtures, the cartilage of homzygous cho/cho mice contains disorganized fibrils with exceptionally large diameters. Analogous consequences are seen in patients with Stickler syndrome where a splice-site mutation probably causes the production of shortened α2(XI) chains [66]. Taken together, these observations are consistent with the notion that collagen XI is required as a powerful nucleating agent in the formation of cartilage fibrils (see above). If production of collagen XI is compromised, collagen II not only remains unincorporated into supramolecular aggregates but, as a consequence of this lack of suprastructural integration, is degraded and largely absent from the tissues [78]. The deletion of collagen IX in mice [67] by inactivation of the Col9a1 gene [68] results in a remarkably mild phenotype. The mice show no extensive abnormality in skeletal development and the fibrils in cartilage of affected animals also are close to normal. However, the birth rate of homozygous mice from heterozygous parents appears to be reduced to about one-half of the expected Mendelian proportion, which may be related to serious skeletal malformations incompatible with further progression of fetal development (Failing, Jongebloed, and Bruckner, unpublished). In addition, cartilage integrity appears to be impaired only in older animals that show premature onset of osteoarthritis-like cartilage degeneration. Again, this is consistent with in vitro fibrillogenesis showing a reduced stability of collagen IXdeficient fibrils [64] and suggests that the main role of collagen IX is at the level of tissue organization of fibrils
in cartilage, as originally proposed by Müller Glauser et al. [79] and revisited in greater detail in cross-linking studies by Eyre et al. [24, 80]. Nevertheless, it is surprising that bone development is almost normal in these animals in view of the fact that it is the thin fibrils arising in fetal and young cartilage which are rich in collagen IX [81]. It is tempting to speculate that the function of collagen IX is redundant, at least in part, in cartilage fibrils. Likely candidates are collagen XVI [18, 82] or other FACITs, including collagens XII, XIV, or XXII [83]. An exon-skipping mutation in the COL9A2 gene of patients with multiple epiphyseal dysplasia [84] leads to the transcription of a shortened α2(IX) mRNA expected to result in α2(IX) chains lacking 12 amino acids within the Col3 domain which protrudes from the fibril surface [59]. It still remains to be determined whether, in analogy to the Col9a1 knockout mice, the production of shortened α2(IX) chains leads to an overall absence of collagen IX. However, fully functional protein is not likely to be synthesized in the affected tissues. Consistent with mice lacking collagen IX, the consequences of this mutation do not include severe skeletal abnormalities. Again, the genotype–phenotype correlation agrees with the notion that collagen IX is not essential for fibrillogenesis in cartilage but that the protein is required for long-term stability of the tissue. If an abnormal variant of collagen IX exists in the patient cartilage, it will be interesting to determine the mode in which a shortened Col3 domain causes impairment of the interaction between fibrils and the extrafibrillar matrix.
X. CORRELATING STRUCTURE WITH THE BIOMECHANICAL ROLE OF ARTICULAR CARTILAGE A major function of articular cartilage is to enable unhindered motion of joints by providing a smooth, low-friction gliding surface, and by cushioning the bone against stress and impact. These properties are achieved by prestressing the cartilagenous material through the interaction of the fibrils with their proteoglycan-rich matrix. The matrix has a high fixed-charge density, deriving from the negatively charged groups of its constituent glycosaminoglycans, chiefly those of aggrecan. This charge is responsible for the high degree of hydration of the tissue, which in turn causes the matrix component to swell strongly. Swelling is counteracted by the cartilage fibrils, which act to keep the glycosaminoglycans compressed and relatively dehydrated in comparison to their free, equilibrium hydration (see also Chapter 5). The collagen fibril network is firmly
416 attached to the extrafibrillar matrix by a high density of noncovalent interactions between proteins, such as between fibrils and decorin, and between glycosaminoglycans. In turn, some of the glycosaminoglycans are covalently bound to proteins. The arrangement of fibrils into arcades is precisely the architecture needed to counteract the stress. The physical properties of cartilage have been extensively studied by mechanical testing, with the observed properties being modeled as a viscoelastic [85], or poroviscoelastic [86] material. Most such modeling investigations have the drawback that the tissue is regarded as a homogeneous substance, but the latter study did investigate samples with the surface layer, the collagen-rich top of the arcades, removed. In this case it was found that exudation of fluid under pressure was more rapid than in intact tissue, demonstrating the role of the superficial zone in controlling interstitial movement of fluid. The intratissue pressure in cartilage has been estimated by subjecting samples to various osmotic pressures and measuring changes in hydration [87]. Osmotic pressure was controlled by use of calibrated solutions of polyethylene glycol (PEG) at various concentrations. It was found that the unperturbed internal pressure of cartilage was ca. 3.9 bar. The mass of normal tissue varied surprisingly little with osmotic pressure changes: normal specimens transferred from 0.15 M NaCl to 0.015 M NaCl swelled by only 1–2 %. The collagen fibrils too respond to changes in osmotic pressure: increases in osmotic pressure reduce the hydration, and thus the intermolecular spacing, of the collagen triple-helices [48]. The tension in the fibrils has been calculated and shown to be easily within the known tensile limits of collagen [88]. Very large increases in osmotic pressure by PEG, up to 10 bar with a concomitant 20% mass loss, were required to make the collagen fibrillar network go “limp” in normal cartilage [87]. The continuous tension to which cartilage fibrils are subjected is an unusal role for collagen. In other tissues collagens are called on to resist tension only transiently. Further, the fibrils must be synthesized and repaired while maintaining tension, requiring a mechanism that is not well understood. It is perhaps not surprising that a number of medically important conditions can develop in such a system. Ameliorating the biological problem of tissue maintenance is the exceptionally long life of cartilage components under normal circumstances. Collagen II has a half-life of more than 200 years in humans [58]. Consistent with this is the rate of accumulation of glycationderived cross-links in cartilage collagen. Such cross-links increase linearly with age after 20, when the organism stops growing, and show no evidence of reaching a plateau
PETER BRUCKNER
level at advanced age [58]. By these measures, then, the collagen is never turned over. The rate of aspartic acid residue racemization has been used to show that the half-life of different species of aggrecan ranges from 3.4 to 25 years in man [89]. Thus, in the absence of disease or injury, cartilage is highly stable. The mechanical properties of cartilage are drastically affected in osteoarthritis. An osteoarthritic specimen transferred from 0.15 M NaCl to 0.015 M NaCl swelled by more than 10%, demonstrating the effects of destruction of the fibrillar net in this disease [87]. In this condition, the tissue has not the means to resist water uptake by the glycosaminoglycans. Thus, therapies that aim to make up the observed loss of glycosaminoglycans in osteoarthritis probably will not work as expected; the fibrillar network needs to be restored.
XI. MODELS OF CARTILAGE FIBRIL STRUCTURE What models of fibril structure are compatible with these observations? The classical scheme has collagen XI forming a core or scaffold on which the collagen II is deposited. Collagen IX then serves to coat the structure and arrest the diametral growth. We have recognized that collagen IX is not necessary to achieve the morphology of the prototypic 20-nm cartilage fibril, but the model is still attractive. First, it appears to account for the observed greater difficulty in extracting type XI from cartilage as compared to type II. Second, the cross-link data showing that collagen XI is cross-linked mostly with other homologous molecules suggests that the protein forms a filamentous core of the fibrils. The core model for cartilage fibrils, unfortunately, does not easily provide a mechanism for diameter control in the absence of type IX: the fibrils should be indefinitely large. However, collagen IX at the surface of cartilage fibrils may be suitable to stabilize interfibrillar connections by integration of its triple helical domain into more than one fibril. This concept has initially been derived from a conspicuous concentration of collagen IX immunogold labeling at fibrillar intersections [60] and has since been substantiated by studies on cross-linking of collagens within cartilage fibrils [80] (Fig. 3). The possible existence of a core filament has also been discussed in conjunction with the analogous collagen I– collagen V system in cornea. Fibrils in cornea too have tightly controlled diameters, a property that is responsible for corneal transparency. One suggestion for the control of diameter in these fibrils, and the relative inaccessibility of their collagen V component to antibody labeling, is
417
Chapter 24 Supramolecular Structure of Cartilage Matrix
Figure 4
Illustration of the core model vs. the alloy model of uniform cartilage fibrils. (A) A cross-section of the core model has collagen XI (gray) at its center, collagen II (open circles) apposed around it, and collagen IX (filled) capping the structure to limit diametral growth. It is now known that collagen IX is not required for diameter control. (B) The alloy model posits a random commingling of collagen II and XI, with either molecule being able to occupy any lattice position. Diameter control could be achieved by the necessity to accommodate the collagen XI propeptides at the surface (cf. Fig. 1). Lattice faults (*), arising from incommensurate packing of collagens II and XI, are important in both models, either providing a binding site for collagen IX in the core model, or destabilizing largediameter fibrils in the alloy model.
Figure 3 Model of the fibrillar network in cartilage and the role of collagen IX and its known cross-links between collagen II and IX and between IX and IX collagen molecules. (A) Covalent cross-links between the C-terminal, nonhelical domain NC1 and a triple-helical portion of collagen II initiate network formation, followed by other cross-links between collagens II and IX or two collagen IX molecules. (B) A model illustrating a possible cross-over of collagen IX molecules between neighboring collagen II/XI fibrils. Reproduced with permission from Eyre, D. R. et al. J. Biol. Chem. 2004; 279, 2568–2574.
could incorporate a certain amount of collagen II up to approximately 8:1, the most favorable composition for the alloy, without structural modification. Beyond that ratio, the structure doesn’t form and phase separation takes place. Collagen XI would be distributed throughout the fibril in this model. This is not necessarily contradicted by the type XI cross-link data because of the relatively great length and flexibility of the molecules. The resistance to extraction of type XI in tissues can be a simple consequence of the cross-linking, rather than of any structural model.
XII. FUTURE PERSPECTIVES the existence of a core fiber of type V [90]. As in the case of cartilage fibrils, this model does not easily provide a mechanism for diameter control in the absence of other components. A different model is that of the biological alloy, in which collagens II and XI randomly commingle to form diameter-controlled fibrils as long as the ratio is 8:1 or less [64] (Fig. 4). This model is attractive because it explains why fibril diameter does not depend on composition. With the core model, one would expect mixed fibrils to be larger than the core material alone, which is not observed. Diameter control is an innate property of the alloy, mediated by collagen XI. It is known that most of the N-propeptides of type XI are retained in the tissue [91]. The requirement to accommodate these bulky moieties, probably on the exterior of the fibrils, could inhibit diametral growth [92, 93]. This class of alloys
A great deal of information has been accumulated on the identity and structure of cartilage matrix macromolecules. However, much remains to be learned about one of the quintessential characteristics of these components: their assembly into insoluble suprastructural aggregates. The fibrillar organization of cartilage collagens is gradually emerging and we are beginning to understand the assembly of the bulk components of the extrafibrillar matrix, aggrecan, and hyaluronan, but the aggregate structures of most of the other macromolecules identified as cartilage matrix components is either unknown or controversial. As is true for most extracellular matrices, even less is known about the formation of structure at the next hierarchic level, i.e. the principles directing the assembly of cartilage matrix from fibrils and the extrafibrillar matrix. As our knowledge expands about additional molecular
418
PETER BRUCKNER
components of fibrils or the extrafibrillar matrix our chances to formulate viable working hypotheses will increase. For example, the definition of fibrillar components other than collagens, particularly those at surfaces, will be one of the challenges in the immediate future. Likewise, it will be interesting to learn about the identity of molecular components of the extrafibrillar matrix which occur in the immediate vicinity of fibrils. Methods will have to be devised to reconstitute matrix domains from suprastructural subunits to understand the forces driving matrix assembly. These efforts will be strongly supported by expanding our knowledge of the function of matrix macromolecules in situ. This goal will be achieved by defining in greater detail the connection between the genotype and the phenotype in human or animal genetic disorders or in transgenic animals. Concurrently, the refinement of techniques such as X-ray diffraction and electron and atomic force microscopy, combined with specific labeling, will provide additional opportunities to elucidate cartilage matrix structure at high resolution. The combined information will lay the groundwork of our understanding of the biomechanics of cartilage and its metabolic regulation through cell–matrix interactions under normal conditions, as well as its abnormalities under pathological conditions.
References 1. Benninghoff, A. (1922). Über den funktionellen Bau des Knorpels. Anat. Anz. 55, 250–267. 2. Hunziker, E. B. (1992). Articular cartilage structure in humans and experimental animals. In “Articular Cartilage and Osteoarthritis” (K. E. Kuettner, R. Schleyerbach, J. G. Peyron, and V. C. Hascall, eds.), pp. 183–199. Raven Press, New York. 3. Broom, N. D., and Silyn-Roberts, H. (1989). The 3-dimensional knit of collagen fibrils in articular cartilage. Connect. Tissue Res. 23, 261–277. 4. Poole, C. A. (1997). Articular cartilage chondrons: Form, function and failure. J. Anat. 191, 1–13. 5. Hunziker, E. B., Herrmann, W., and Schenk, R. K. (1983). Ruthenium hexamine trichloride (RHT)-mediated interaction between plasmalemmal components and pericellular matrix proteoglycans is responsible for the preservation of chondrocyte plasma membranes in situ during cartilage fixation. J. Histochem. Cytochem. 31, 717–727. 6. Studer, D., Michel, M., Wohlwend, M., Hunziker, E. B., and Buschmann, M. D. (1995). Vitrification of articular cartilage by highpressure freezing. J. Microsc. 179, 321–332. 7. Hunziker, E. B., Wagner, J., and Studer, D. (1996). Vitrified articular cartilage reveals novel ultra-structural features respecting extracellular matrix architecture. Histochemical Cell Biology 106, 375–382. 8. Hunziker, E. B., Michel, M., and Studer, D. (1997). Ultrastructure of adult human articular cartilage matrix after cryotechnical processing. Microsc. Res. Tech. 37, 271–284. 9. Studer, D., Chiquet, M., and Hunziker, E. B. (1996). Evidence for a distinct water-rich layer surrounding collagen fibrils in articular cartilage extracellular matrix. J. Struct. Biol. 117, 81–85.
10. ap Gwynn, I., Wade, S., Kääb, M. J., Owen, G. R., and Richards, R. G. (2000). Freeze-substitution of rabbit tibial articular cartilage reveals that radial zone collagen fibres are tubules. J. Microsc. 197, 159–172. 11. ap Gwynn, I., Wade, S., Ito, K., and Richards, R. G. (2002). Novel aspects to the structure of rabbit articular cartilage. Eur. Cell. Mater. 4, 18–29. 12. Hughes, L. C., Archer, C. W., and ap Gwynn, I. (2005). The ultrastructure of mouse articular cartilage: collagen orientation and implications for tissue functionality. A polarised light and scanning electron microscope study and review. Eur. Cell. Mater. 9, 68–84. 13. Williamson, A. K., Chen, A. C., Masuda, K., Thonar, E. J. M. A., and Sah, R. L. (2003). Tensile mechanical properties of bovine articular cartilage: Variations with growth and relationships to collagen network components. J. Orthop. Res. 21, 872–880. 14. Hascall, V. C., and Heinegård, D. (1974). Aggregation of cartilage proteoglycans. 1. The role of hyaluronic acid. J. Biol. Chem. 249, 4232–4241. 15. Rosenberg, L., Hellmann, W., and Kleinschmidt, A. K. (1975). Electron microscopic studies of proteoglycan aggegates from bovine articular cartilage. J. Biol. Chem. 250, 1877–1883. 16. Mörgelin, M., Paulsson, M., Heinegård, D., Aebi, U., and Engel, J. (1995). Evidence of a defined spatial arrangement of hyaluronate in the central filament of cartilage proteoglycan aggregates. Biochem. J. 307, 595–601. 17. Hauser, N., Paulsson, M., Heinegård, D., and Mörgelin, M. (1996). Interaction of cartilage matrix protein with aggrecan. Increased covalent cross-linking with tissue maturation. J. Biol. Chem. 271, 32247–32252. 18. Kassner, A., Hansen, U., Miosge, N., Reinhardt, D. P., Aigner, T., Bruckner-Tuderman, L., Bruckner, P., and Grässel, S. (2003). Supramolecular association of collagen XVI to different fibrillar systems in human skin and cartilage. Matrix Biol. 22, 131–143. 19. Miosge, N., Flachsbart, K., Götz, W., Schultz, W., Kresse, H., and Herken, R. (1994). Light and electron microscopical immunohistochemical localization of the small proteoglycan core proteins decorin and biglycan in human knee joint cartilage. Histochem. J. 26, 939–945. 20. Hagg, R., Bruckner, P., and Hedbom, E. (1998). Cartilage fibrils of mammals are biochemically heterogeneous: Differential distribution of decorin and collagen IX. J. Cell Biol. 142, 285–294. 21. Rosenberg, K., Olsson, H., Mörgelin, M., and Heinegård, D. (1998). Cartilage oligomeric matrix protein shows high affinity zinc-dependent interaction with triple helical collagen. J. Biol. Chem. 273, 20397–20403. 22. Wardale, R. J., and Duance, V. C. (1993). Quantification and immunolocalisation of porcine articular and growth plate cartilage collagens. J. Cell Sci. 105, 975–984. 23. Young, R. D., Lawrence, P. A., Duance, V. C., Aigner, T., and Monaghan, P. (2000). Immunolocalization of collagen types II and III in single fibrils of human articular cartilage. J. Histochem. Cytochem. 48, 423–432. 24. Eyre, D. R. (2004). Collagens and cartilage matrix homeostasis. Clin. Orthop. Rel. Res. 427S, S118–S122. 25. Adam, M., and Deyl, Z. (1983). Altered expression of collagen phenotype in osteoarthrosis. Clin. Chim. Acta 12, 25–32. 26. Aigner, T., Bertling, W., Stöss, H., Weseloh, G., and von der Mark, K. (1993). Independent expression of fibril-forming collagens I, II, and III in chondrocytes of human osteoarthritic cartilage. J. Clin. Invest. 91, 829–837. 27. Aigner, T., Glückert, K., and von der Mark, K. (1997). Activation of fibrillar collagen synthesis and phenotypic modulation of chondrocytes in early human osteoarthritic cartilage lesions. Osteoarthritis Cart. 5, 183–189. 28. Hagiwara, H., Schröter-Kermani, C., and Merker, H.-J. (1993). Localization of collagen type VI in articular cartilage of young and adult mice. Cell Tissue Res. 272, 155–160.
Chapter 24 Supramolecular Structure of Cartilage Matrix
419
29. Watt, S. L., Lunstrum, G. P., McDonough, A. M., Keene, D. R., Burgeson, R. E., and Morris, N. P. (1992). Characterization of collagen type XII and type XIV from fetal bovine cartilage. J. Biol. Chem. 267, 20093–20099. 30. Poole, C. A., Ayad, S., and Gilbert, R. T. (1992). Chondrons from articular cartilage. V. Immunohistochemical evaluation of type VI collagen organisation in isolated chondrons by light, confocal and electron microscopy. J. Cell Sci. 103, 1101–1110. 31. Furthmayr, H., Wiedemann, H., Timpl, R., Odermatt, E., and Engel, J. (1983). Electron-microscopical approach to a structural model of intima collagen. Biochem. J. 211, 303–311. 32. Wiberg, C., Heinegård, D., Wenglén, C., Timpl, R., and Mörgelin, M. (2002). Biglycan organizes collagen VI into hexagonal-like networks resembling tissue structures. J. Biol. Chem. 277, 49120–49126. 33. Keene, D. R., Lunstrum, G. P., Morris, N. P., Stoddard, D. W., and Burgeson, R. E. (1991). Two type XII-like collagens localize to the surface of banded collagen fibrils. J. Cell Biol. 113, 971–978. 34. Rich, A., and Crick, F. H. C. (1961). The molecular structure of collagen. J. Mol. Biol. 3, 483–506. 35. Hulmes, D. J. S., Miller, A., Parry, D. A. D., Piez, K. A., and Woodhead-Galloway, J. (1973). Analysis of the primary structure of collagen for the origins of molecular packing. J. Mol. Biol. 79, 137–148. 36. Hodge, A. J., and Petruska, J. A. (1963). Recent studies with the electron microscope on ordered aggregates of the tropocollagen macromolecule. In “Aspects of Protein Structure” (G. N. Ramachandran, ed.), pp. 289–300. Academic Press, New York. 37. Scott, J. E. (1990). Proteoglycan-collagen interactions and subfibrillar structure in collagen fibrils. Implications in the development and ageing of connective tissues. J. Anat. 169, 23–35. 38. Yu, L., Cummings, C., Sheehan, J. K., Kadler, K. E., Holmes, D. F., and Chapman, J. A. (1993). Visualization of individual proteoglycancollagen interactions. In “Dermatan Sulfate Proteoglycans” (J. E. Scott, ed.), pp. 183–188. Portland Press, London.. 39. Scott, J. E. and Orford, C. R. (1981). Dermatan sulphate-rich proteoglycan associates with rat tail-tendon collagen at the d band in the gap region. Biochem. J. 197, 213–216. 40. Kobe, B., and Deisenhofer, J. (1993). Crystal structure of procine ribonuclease inhibitor, a protein with leucine-rich repeats. Nature 366, 751–756. 41. Weber, I. T., Harrison, R. W., and Iozzo, R. V. (1996). Model structure of decorin and implications for collagen fibrillogenesis. J. Biol. Chem. 271, 31767–31770. 42. Hedlund, H., Mengarelli-Widholm, S., Heinegård, D., Reinholt, F. P., and Svensson, O. (1994). Fibromodulin distribution and association with collagen. Matrix Biol. 14, 227–232. 43. Hedbom, E., and Heinegård, D. (1993). Binding of fibromodulin and decorin to separate sites on fibrillar collagens. J. Biol. Chem. 268, 27307–27312. 44. Schönherr, E., Witsch-Prehm, P., Harrach, B., Robenek, H., Rauterberg, J., and Kresse, H. (1995). Interaction of biglycan with type I collagen. J. Biol. Chem. 270, 2776–2783. 45. Fraser, R. D., and MacRae, T. P. (1979). The crystalline structure of collagen fibrils in tendon. J. Mol. Biol. 127, 129–133. 46. Fraser, R. D., MacRae, T. P., Miller, A., and Suzuki, E. (1983). Molecular conformation and packing in collagen fibrils. J. Mol. Biol. 167, 497–521. 47. Brodsky, B., and Eikenberry, E. F. (1985). Supramolecular collagen assemblies. In “Biology, Chemistry, and Pathology of Collagen” (R. Fleischmajer, B. R. Olsen, and K. Kühn, eds.), Vol. 460, pp. 73–84. The New York Academy of Sciences, New York. 48. Hulmes, D. J. S., and Miller, A. (1979). Quasi-hexagonal packing in collagen fibrils. Nature 282, 878–880. 49. Wess, T. J., Hammersley, A. P., Wess, L., and Miller, A. (1998). Molecular packing of type I collagen in tendon. J. Mol. Biol. 275, 255–267.
50. Hulmes, D. J. S., Holmes, D. F., and Cummings, C. (1985). Crystalline regions in collagen fibrils. J. Mol. Biol. 184, 473–477. 51. Brodsky, B., and Eikenberry, E. F. (1982). Characterization of fibrous forms of collagen. Methods of Enzymology 82, 127–174. 52. Katz, E. P., Wachtel, E. J., and Maroudas, A. (1986). Extrafibrillar proteoglycans osmotically regulate the molecular packing of collagen in cartilage. Biochim. Biophys. Acta 882, 136–139. 53. Tanaka, S., Avigad, G., Eikenberry, E. F., and Brodsky, B. (1988). Isolation and partial characterization of collagen chains dimerized by sugar-derived cross-links. J. Biol. Chem. 263, 17650–17657. 54. Grynpas, M. D., Eyre, D. R., and Kirschner, D. A. (1980). Collagen type II differs from type I in native molecular packing. Biochim. Biophys. Acta 626, 346–355. 55. Wachtel, E., Maroudas, A., and Schneiderman, R. (1995). Age-related changes in collagen packing of human articular cartilage. Biochim. Biophys. Acta 1243, 239–243. 56. Price, R. I., Lees, S., and Kirschner, D. A. (1997). X-ray diffraction analysis of tendon collagen at ambient and cryogenic temperatures: role of hydration. Int. J. Biol. Macromol. 20, 23–33. 57. Tanaka, S., Avigad, G., Brodsky, B., and Eikenberry, E. F. (1988). Glycation induces expansion of the molecular packing of collagen. J. Mol. Biol. 203, 495–505. 58. Bank, R. A., Bayliss, M. T., Lafeber, F. P. J. G., Maroudas, A., and Tekoppele, J. M. (1998). Ageing and zonal variation in post-translational modification of collagen in normal human articular cartilage. The age-related increase in non-enzymatic glycation affects biomechanical properties of cartilage. Biochem. J. 330, 345–351. 59. Vaughan, L., Mendler, M., Huber, S., Bruckner, P., Winterhalter, K. H., Irwin, M. I., and Mayne, R. (1988). D-periodic distribution of collagen IX along cartilage fibrils. J. Cell Biol. 106, 991–997. 60. Mendler, M., Eich-Bender, S. G., Vaughan, L., Winterhalter, K. H., and Bruckner, P. (1989). Cartilage contains mixed fibrils of collagen types II, IX, and XI. J. Cell Biol. 108, 191–197. 61. Keene, D. R., Oxford, J. T., and Morris, N. P. (1995). Ultrastructural localization of collagen types II, IX, and XI in the growth plate of human rib and fetal bovine epiphyseal cartilage: Type XI collagen is restricted to thin fibrils. J. Histochem. Cytochem. 43, 967–979. 62. Wu, J. J., Woods, P. E., and Eyre, D. R. (1992). Identification of crosslinking sites in bovine cartilage type IX collagen reveals an antiparallel type II-type IX molecular relationship and type IX to type IX bonding. J. Biol. Chem. 267, 23007–23014. 63. Wu, J.-J., and Eyre, D. R. (1995). Structural analysis of cross-linking domains in cartilage type XI collagen. Insights on polymeric assembly. J. Biol. Chem. 270, 18865–18870. 64. Blaschke, U. K., Eikenberry, E. F., Hulmes, D. J. S., Galla, H. J., and Bruckner, P. (2000). Collagen XI nucleates assembly and limits lateral growth of cartilage fibrils. J. Biol. Chem. 275, 10370–10378. 65. Lee, S. L., and Piez, K. A. (1983). Type II collagen from lathyritic rat chondrosarcoma: Preparation and in vitro fibril formation. Collagen Relat. Res. 3, 89–103. 66. Bruckner, P., and van der Rest, M. (1994). Structure and function of cartilage collagens. Microsc. Res. Tech. 28, 378–384. 67. Gelman, R. A., Williams, B. R., and Piez, K. A. (1979). Collagen fibril formation. Evidence for a multistep process. J. Biol. Chem. 254, 180–186. 68. Hansen, U., and Bruckner, P. (2003). Macromolecular specificity of collagen fibrillogenesis: Fibrils of collagens I and XI contain a heterotypic alloyed core and a collagen I sheath. J. Biol. Chem. 278, 37352–37359. 69. Li, Y. F., and Olsen, B. R. (1997). Murine models of human genetic skeletal disorders. Matrix Biol. 16, 49–52. 70. Aszódi, A., Pfeifer, A., Wendel, M., Hiripi, L., and Fässler, R. (1998). Mouse models for extracellular matrix diseases. J. Mol. Med. 76, 238–252. 71. Chan, D., Cole, W. G., Chow, C. W., Mundlos, S., and Bateman, J. F. (1995). A COL2A1 mutation in achondrogenesis type II results in the
420
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
replacement of type II collagen by type I and III collagens in cartilage. J. Biol. Chem. 270, 1747–1753. Mundlos, S., Chan, D., McGill, J., and Bateman, J. F. (1995). An a1(II) Gly913 to Cys substitution prevents the matrix incorporation of type II collagen which is replaced with type I and II collagens in cartilage from a patient with hypochondrogenesis. Am. J. Med. Genet. 63, 129–136. Bruckner, P., and Eikenberry, E. F. (1984). Formation of the triple helix of type I procollagen in cellulo. Temperature-dependent kinetics support a model based on cis-trans isomerization of peptide bonds. Eur. J. Biochem. 140, 391–395. Li, S. W., Prockop, D. J., Helminen, H., Fässler, R., Lapveteläinen, T., Kiraly, K., Peltarri, A., Arokoski, J., Lui, H., Arita, M., and Khillan, J. (1995). Transgenic mice with targeted inactivation of the COL2A1 gene for collagen II develop a skeleton with membranous and periosteal bone but no endochondral bone. Genes Dev. 9, 2821–2830. Aszódi, A., Chan, D., Hunziker, E. B., Bateman, J. F., and Fässler, R. (1998). Collagen II is essential for the removal of the notochord and the formation of intervertebral discs. J. Cell Biol. 143, 1399–1412. Sahlman, J., Inkinen, R., Hirvonen, T., Lammi, M. J., Lammi, P. E., Nieminen, J., Lapveteläinen, T., Prockop, D. J., Arita, M., Li, S. W., Hyttinen, M. M., Helminen, H. J., and Puustjärvi, K. (2001). Premature vertebral endplate ossification and mild disc degeneration in mice after inactivation of one allele belonging to the Col2a1 gene for type II collagen. Spine 26, 2558–2565. Hyttinen, M. M., Toyras, J., Lapveteläinen, T., Lindblom, J., Prockop, D. J., Li, S. W., Arita, M., Jurvelin, J. S., and Helminen, H. J. (2001). Inactivation of one allele of the type II collagen gene alters the collagen network in murine articular cartilage and makes cartilage softer. Ann. Rheum. Dis. 60, 262–268. Rodriguez, R. R., Seegmiller, R. E., Stark, M. R., and Bridgewater, L. C. (2004). A type XI collagen mutation leads to increased degradation of type II collagen in articular cartilage. Osteoarthritis Cart. 12, 314–320. Müller-Glauser, W., Humbel, B., Glatt, M., Sträuli, P., Winterhalter, K. H., and Bruckner, P. (1986). On the role of type IX collagen in the extracellular matrix of cartilage: Type IX collagen is localized to intersections of collagen fibrils. J. Cell Biol. 102, 1931–1939. Eyre, D. R., Pietka, T., Weis, M. A., and Wu, J.-J. (2004). Covalent cross-linking of the NC1 domain of collagen type IX to collagen type II in cartilage. J. Biol. Chem. 279, 2568–2574. Hagg, R., Hedbom, E., Möllers, U., Aszódi, A., Fässler, R., and Bruckner, P. (1997). Absence of the α1(IX) chain leads to a functional knock-out of the entire collagen IX protein in mice. J. Biol. Chem. 272, 20650–20654.
PETER BRUCKNER 82. Kassner, A., Tiedemann, K., Notbohm, H., Ludwig, T., Mörgelin, M., Reinhardt, D. P., Chu, M.-L., Bruckner, P., and Grässel, S. (2004). Molecular structure and interaction of recombinant human type XVI collagen. J. Mol. Biol. 339, 835–853. 83. Koch, M., Schulze, J., Hansen, U., Ashwodt, T., Keene, D. R., Brunken, W. J., Burgeson, R. E., Bruckner, P., and BrucknerTuderman, L. (2004). A novel marker of tissue junctions, collagen XXII. J. Biol. Chem. 279, 22514–22521. 84. Muragaki, Y., Mariman, E. C. M., Van Beersum, S. E. C., Perälä, M., Van Mourik, J. B. A., Warman, M. L., Olsen, B. R., and Hamel, B. C. J. (1996). A mutation in the gene encoding the α2 chain of the fibrilassociated collagen IX, COL9A2, causes multiple epiphyseal dysplasia (EDM2). Nat. Genet. 12, 103–105. 85. Parsons, J. R., and Black, J. (1977). The viscoelastic shear behavior of normal rabbit articular cartilage. J. Biomech. 10, 21–29. 86. Setton, L. A., Zhu, W., and Mow, V. C. (1993). The biphasic poroviscoelastic behavior of articular cartilage: role of the surface zone in governing the compressive behavior. J. Biomech. 26, 581–592. 87. Basser, P. J., Schneiderman, R., Bank, R. A., Wachtel, E., and Maroudas, A. (1998). Mechanical properties of the collagen network in human articular cartilage as measured by osmotic stress technique. Arch. Biochem. Biophys. 351, 207–219. 88. Aspden, R. M., and Hukins, D. W. L. (1990). Stress in collagen fibrils of articular cartilage calculated from their measured orientations. Matrix 9, 486–488. 89. Maroudas, A., Bayliss, M. T., Uchitel-Kaushansky, N., Schneiderman, R., and Gilav, E. (1998). Aggrecan turnover in human articular cartilage: Use of aspartic acid racemization as a marker of molecular age. Arch. Biochem. Biophys. 350, 61–71. 90. Birk, D. E., Fitch, J. M., Babiarz, J. P., Doane, K. J., and Linsenmayer, T. F. (1990). Collagen fibrillogenesis in vitro. Interaction of type I and type V collagen regulates fibril diameter. J. Cell Sci. 95, 649–657. 91. Rousseau, J. C., Farjanel, J., Boutillon, M. M., Hartmann, D. J., van der Rest, M., and Moradi-Améli, M. (1996). Processing of type XI collagen. Processing of the matrix forms of the α1(XI) chain. J. Biol. Chem. 271, 23743–23748. 92. Chapman, J. A. (1989). The regulation of size and form in the assembly of collagen fibrils in vivo. Biopolymers 28, 1367–1382. Published erratum: Biopolymers 28, 2201–2205. 93. Hulmes, D. J. S. (1983). A possible mechanism for the regulation of collagen fibril diameter in vivo. Collagen Relat. Res. 3, 317–321.
Chapter 25
Products of Cartilage Metabolism Daniel-Henri Manicourt
Jean-Pierre Devogelaer
Laboratoire de Chimie Physiologique (Metabolic Research Group, Connective Tissue Section), Christian de Duve Institute of Cellular Pathology and Department of Rheumatology, Saint Luc University Hospital, Université Catholique de Louvain, 1200 Brussels, Belgium Department of Rheumatology, Saint Luc University Hospital, Université Catholique de Louvain, 1200 Brussels, Belgium
Eugene J.-M. A. Thonar
Departments of Biochemistry, Internal Medicine and Orthopedic Surgery, Rush Medical College, Rush-Presbyterian-St. Luke’s Medical Center, Chicago, Illinois, 60612 USA
I. Introduction II. The Chondrocyte and its Extracellular Matrix III. Products of Collagen Metabolism IV. Products of Aggrecan Metabolism V. Products of the Metabolism of other Proteoglycans
VI. Products of the Metabolism of Link Protein and Hyaluronan VII. Other Products of Chondrocyte Metabolism VIII. Concluding Statement References
I. INTRODUCTION
worth pointing out that articular cartilage in diarthrodial (synovial) joints makes up only approximately 10% of the total cartilage mass in the body of an adult dog [1]. Joint diseases and related conditions involving abnormalities in the metabolism of cartilaginous tissues cause widespread disability. While articular cartilage failure was first regarded as a passive degenerative process, modern studies have provided a large body of evidence that this is not so [2]. In osteoarthritis (OA), for example, the destruction of the articular surface is now thought to result from an imbalance between dynamic anabolic and catabolic processes that normally proceed in a harmonized and strictly regulated manner [2]. Current research on cartilage includes attempts to define the mechanisms that regulate
Cartilage comprises approximately 1% of the dry body weight of an adult dog [1]. Distributed throughout the body, this hard but resilient tissue performs a variety of functions. During development, cartilage acts as a growing scaffold for limb elongation. Articular cartilage, which covers the ends of bones in diarthrodial joints, as well as the softer and more compressible nucleus pulposus of the intervertebral disk, allows the joints in the rigid skeleton to articulate smoothly. Cartilage is the supporting structure of the respiratory tract (nose, larynx, and trachea) and, as costal cartilage, contributes to the protection of the lungs, heart, liver, and spleen from external physical stresses. It is Dynamics of Bone and Cartilage Metabolism
421
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
422
DANIEL-HENRI MANICOURT ET AL.
the expression of cartilage-specific genes, to shed light upon the structural and functional properties of individual matrix components and to better understand how these matrix-building blocks are turned over in both health and disease. This chapter will primarily discuss the major products of articular cartilage metabolism with special emphasis upon those that are relevant as body fluid markers. The structure and supramolecular organization of major cartilage components as well as the main regulatory pathways of chondrocyte metabolism have been detailed in the preceding chapters of this volume. The following section provides a brief review of the structure–function relationships of the major matrix-building blocks.
II. THE CHONDROCYTE AND ITS EXTRACELLULAR MATRIX The prime function of articular cartilage is physical, with water, ions and anionic aggrecan molecules within the collagenous meshwork all playing key roles in endowing the tissue with its load-bearing properties. The collagenous meshwork rich in type II collagen molecules gives the tissue tensile strength and hinders the expansion of the viscoelastic under-hydrated aggrecan molecules that provide compressive stiffness. The highly sulfated aggrecan molecules interact noncovalently with a single strand of hyaluronan (HA) and link protein (LP) molecules to form supramolecular aggregates of very large size that become firmly entrapped within the collagenous meshwork (see Chapters 2 and 5). The resulting high fixed negative charge density, balanced by mobile counter ions, gives rise to high osmotic pressures (up to 2–3 atmospheres), which are counteracted by the constant tension that is developed within the meshwork of cross-linked collagenous fibrils [3]. Cartilage is thus made stiff by being swollen with water. As aggrecan molecules oppose the fluid loss and the redistribution of water within the tissue, cartilage shows resilience upon loading, with the tissue rapidly recovering its elasticity when the load is removed. During cycle loading, the fluid flowing through the fine pores of the tissue not only dissipates energy and absorbs shocks but also participates in joint lubrication, contributes to chondrocyte nutrition, and acts as a mechanical signal transducer that regulates chondrocyte synthetic/catabolic activities [4]. It is obvious, therefore, that the structural integrity of the collagenous meshwork and aggrecan molecules is critically important for the maintenance of the biomechanical properties of the tissue. The extracellular matrix (ECM) of cartilage contains many additional components that contribute to the cohesiveness of the matrix and to the regulation of
chondrocyte function. For example, types IX and XI collagen (Chapters 1 and 2) are essential players in the organization and functions of the major collagenous fibrils, whereas type VI collagen forms distinct microfibrils that appear concentrated in the capsular matrix surrounding individual chondrocytes or groups of chondrocytes; this protective structure, termed the chondron, dampens the osmotic and physicomechanical changes induced by joint loading [5, 6]. By interacting with growth factors and other macromolecules of the ECM, including collagens and fibronectin (Chapter 5), small nonaggregating proteoglycans (PGs) (including decorin, biglycan, perlecan, lumican, glypican, and fibromodulin) contribute to the regulation of chondrocyte function and to the maintenance of the supramolecular organization of the matrix, a contention strengthened by the observation that mutations in the gene encoding for several of these molecules lead to chondrodysplasias [7, 8]. An ill-defined number of noncollagenous proteins (Chapter 4) also participate in the complex interactions that together form a network in which chondrocytes and most molecules of the ECM are involved. The relevance of some of these proteins is illustrated by their involvement in genetic disorders [7]. Cartilage matrix-building blocks are synthesized, organized, and maintained by a sparse population of chondrocytes. These cells are protected from the potentially damaging forces of mechanical function by the ECM they produce. As the properties of cartilage are critically dependent upon the structure and integrity of the ECM, normal turnover is a conservative process in which the rate of degradation of each molecular species does not exceed the rate at which it is replaced by newly synthesized products. The degradation of matrix molecules involves primarily proteinases (Chapter 10) but probably also free radicals secreted by chondrocytes. In disease states, the rate of degradation of matrix molecules often exceeds the rate of synthesis and, as a consequence, the tissue becomes thin and mechanically weak. Because cartilage is avascular, chondrocyte nutrition in a diarthrodial joint has to rely upon the diffusion of nutrients from the synovial fluid that bathes the tissues and links together the different components within the joint. The oxygen tension in cartilage may be as low as 1–3%, compared to 24% in the normal atmosphere [9]; this low oxygen tension appears to be beneficial for maintenance of the chondrocyte-specific phenotype in vitro [10]. This remarkable adaptation of chondrocytes to the hypoxic environment prevailing in the cartilage tissue might be related to their relatively high constitutive expression levels of the transcriptional complex hypoxia-inducible factor-1, which regulates genes encoding growth factors, glycolytic enzymes, and glucose transporters [11, 12].
Chapter 25 Products of Cartilage Metabolism
While the majority of animal cells derive their energy by using oxygen for mitochondrial oxidative phosphorylation, mammalian articular chondrocytes use oxygen to stimulate the pathway of glycolysis, a sequence of reaction that produces lactate, does not consume oxygen and is the principal source of ATP for metabolic processes, including the synthesis of the extracellular matrix [13, 14]. To maintain intracellular pH, and hence glycolysis, the large quantities of lactate− and H+ ions produced by the glycolytic pathway are extruded into the extracellular environment via the monocarboxylate transporters and other H+ active transporters, such as the sodium–hydrogen exchanger [15, 16]. By inhibiting glycolysis, iodoacetate reduces ATP levels and lactate production markedly, and, in so doing, leads to degenerative changes in articular cartilage. Since aggrecan molecules are highly sulfated, the Donnan equilibrium ensures that chondrocytes in vivo also survive in a relatively acidic environment and at a higher osmolarity than other cell types (350–450 mOsm, compared with 280 mOsm) [17]. Osmolarity is a powerful regulator of matrix turnover. In vitro, chondrocytes respond readily to alterations in their ionic environment with maximal rates of synthesis of aggrecan and proteins at osmolalities close to those experienced in situ [17]. When cells are faced with osmotic challenges, the activation of signaling pathways (SAPK2, ERK, and others) enhances the activity of a variety of membrane transport proteins, including ionic channels and carriers as well as amino acid transport systems, which all contribute to restore the intracellular ionic milieu and, hence, cell function [18, 19]. Changes in the rate of synthesis of matrix macromolecules also occur in vivo when PG concentrations are altered during fluid loss under load and when cartilage swells because of a decrease in the tensile stiffness of the collagenous meshwork, as observed in the initial stages of OA [19]. Clinical observations, as well as in vitro and in vivo studies, point out that cartilage loading markedly influences the biosynthetic activity of chondrocytes, which then can adapt the composition and organization of the ECM in response to changes in functional demand. In postnatal cartilage, the high weight-bearing regions have a higher aggrecan content and a stiffer collagenous meshwork than low weight-bearing regions [3]. This is in contrast to neonatal cartilage which does not exhibit any topographical variation in tissue composition [20, 21]. Cartilage explants are sensitive to applied loads. Fluctuating loads stimulate matrix production, the extent of the response being influenced by the frequency, amplitude, and wave forms, whereas static pressure causes a reduction in the rate of synthesis of both aggrecan and LP molecules, without affecting HA synthesis [22–25].
423 Joint immobilization also causes a marked loss of PGs from articular cartilage [26]. This depletion, reflecting both a decline in the rate of PG synthesis and an increase in PG catabolism, is, however, completely reversible when the joints are remobilized [27]. The absence of motion appears more critical than the lack of loading because even small amounts of motion of the unloaded joint reduces the severity of the PG depletion, suppresses the signal transduction pathways of pro-inflammatory/catabolic mediators, and stimulates anabolic pathways; a finding that justifies the clinical use of continuous passive motion [28]. On the other hand, strenuous joint loading can cause disruption of the collagenous meshwork and a chondrocyte-mediated loss of matrix PGs as well as cell necrosis and apoptosis, which can all lead to osteoarthritic cartilage changes [29–31]. Further, after an injurious compression, the remaining viable chondrocytes are no longer able to respond to the stimulatory effects of moderate dynamic compression as seen in normal cartilage [32]. How joint loading affects the biosynthetic activity of chondrocytes remains poorly understood. There is, however, evidence that the intracellular composition, the rate of glycolysis, and the pattern of gene expression are all influenced by changes in chondrocyte membrane permeability, which are generated upon joint loading by changes in pericellular osmolarity and pressure, as well as by membrane stretch and shear stresses [33, 34]. Further, by deforming the intracellular organelles and nucleus, loads might change the rate of synthesis, secretion, and/or quality of matrix molecules [35]. It is also clear that, by deforming the ECM and producing local and temporal changes in gradient pressure, loads initiate intratissue fluid flows, which not only generate electrical streaming potentials and currents but also affect the transport of soluble factors and nutrients and might liberate locally basic fibroblast growth factor from an extracellular heparan sulfate-bound pool [35–38]. All these mechanical, chemical, and electrical signals are sensed by the chondrocyte and are thought to be mediated, at least in part, by integrins (Chapter 8) and extracellular signal-regulated kinase (ERK), a member of the ubiquitous family of mitogen-activated kinases (MAPs), which contributes to the transmission of these stimuli from outside the cell to the nucleus and influences a variety of cell processes, including transcriptional regulation [39]. The effects of mechanical stresses are also modulated by the quality of the matrix and more particularly its aggrecan content, which regulates swelling pressure, and the properties of the collagenous meshwork, which influences the degree of deformation. Thus, under the same load, the extent of cartilage deformation, the increase of fluid flow, the hydrostatic pressure and the change in fluid content can be very different in different areas of the same joint,
424
DANIEL-HENRI MANICOURT ET AL.
in young relative to old cartilage, or in normal relative to arthritic tissue. The biosynthetic response of chondrocytes to mechanical stress is also affected by their origin [40]. Indeed, with distance from the articular surface, chondrocytes exhibit heterogeneity with respect to shape, surface receptors, and metabolic activities [41]. In culture, chondrocytes from the deeper layers of articular cartilage exhibit a faster rate of PG synthesis and a slower rate of PG turnover than cells from the superficial layers, a finding that is consistent with the observation that net matrix accumulation is more pronounced around deep but superficial chondrocytes. On the other hand, chondrocytes from the superficial layers produce greater amounts of nitric oxide when stimulated with cytokines and are even more sensitive to the catabolic effect of interleukin-1 [42]. The precise reasons for this type of heterogeneity among chondrocytes are not obvious. However, it is clear that the shape and magnitude of the biomechanical stresses are unevenly distributed throughout
Figure 1
the cartilage matrix. It also is clear that chondrocytes in different layers are subjected to different concentrations of nutrients and metabolites.
III. PRODUCTS OF COLLAGEN METABOLISM As stated in Chapter 1, type II collagen is synthesized as a large precursor molecule, termed procollagen, which exhibits a globular extension propeptide in both its N- and C-terminal ends (Fig. 1) [43]. Depending upon alternative splicing of exon 2, the N-propeptide either contains (form IIA) or does not contain (form IIB) a 69 aminoacid cysteine-rich domain [44]. Although the function of the domain encoded by exon 2 remains poorly understood, there is clear evidence that normal adult articular cartilage expresses high levels of type IIB procollagen, whereas type IIA procollagen is expressed during embryogenesis,
Type II collagen fibrils are made of tropocollagen molecules interconnected by cross-links between the nonhelical telopeptide regions of individual tropocollagen molecules and the helical region of adjacent molecules. Each tropocollagen molecule is composed of three identical α chains. After a first cleavage by any one of the three mammalian collagenases (MMP-1, MMP-8, and MMP-13), the cleaved triple helix unwinds at physiological temperature and exposes hidden epitopes on α chains that are not recognized on the native triple helix. Epitopes on nonhelical telopeptides and conformational triple helical dependent epitopes are also indicated. Adapted from reference [43].
Chapter 25 Products of Cartilage Metabolism
at the onset of cartilage hypertrophy and in human OA cartilage as well [45–47]. Soon after its release extracellularly, the type II procollagen molecule is processed by specific proteinases which remove its two globular propeptides: the disintegrin and metalloproteinase with thrombospondin-like motifs (ADAMTS)-3 cleaves the N-propeptide at A181 ↓ Q182 [48] whereas bone morphogenic protein-1 (BMP-1) cleaves the C-propeptide at A1241 ↓ D1242 [49]. This proteolytic cleavage allows the resulting tropocollagen molecule, consisting of a long compact triple-helical region with nontriplehelical short telopeptides in both N- and C-terminal ends, to become incorporated into a fibril (Fig. 1). Because they cannot be detected in cell layers and culture media of chondrocytes, the free N-propeptide molecules of type II collagen are thought to be quickly metabolized [50]. In contrast, the free C-propeptide of the pro [α(II)] chain (CPII), also referred to as chondrocalcin, is readily detected in the pericellular matrix of cartilage explants [47, 51]. In healthy adult articular cartilage, the concentrations of CPII are low, appear relatively constant throughout the tissue and do not change from 30 to 75 years of age whereas, in OA cartilage, CPII levels are usually elevated in the middle and deeper zones of the matrix but not in the superficial more damaged zone, therefore suggesting that the less degenerated cartilage is capable of mounting a reparative response characterized by increased synthesis of type II procollagen. Most CPII molecules are thought to diffuse out of the cartilage matrix after they are produced and their concentration in body fluids can be measured by specific immunoassays [51, 52]. Support for the contention that the level of this body fluid marker provides a measure of the rate of type II collagen synthesis in cartilage comes from the observation that the rate of synthesis of type II collagen in articular cartilage is directly proportional to the tissue content of C-propeptide: the latter was found to have a halflife of 14–20 hours [51]. This contention is further strengthened by the observation that, during longitudinal bone growth, the enhanced rate of synthesis of type II collagen in the growth plate is associated with elevated serum levels of CPII, which then drop as growth ceases [53]. Synovial fluid levels of CPII are found in higher levels in traumatic and primary osteoarthritis (OA) than in wellestablished rheumatoid arthritis (RA) or infectious arthritis [54–57], confirming the widely held view that the degradative events in OA are balanced, at least during the early stages, by an up-regulation of the biosynthetic processes. In contrast, the synovial fluid levels of CPII are low in early RA suggesting that, in the early stages of this disease, down-regulation of the rate of type II collagen synthesis might contribute to joint destruction [57]. Interestingly, the
425 enhanced levels of CPII found in the articular cartilage and synovial fluid from both RA and OA joints are reflected by an increase in the serum concentrations of CPII in RA patients, but not in OA patients who instead exhibit a significant reduction in their circulating levels of type II collagen propeptides [51, 52, 58]. It is possible that in OA patients the altered serum levels of CPII reflect a reduction of type II collagen synthesis in cartilages not affected by OA. Once tropocollagen molecules have spontaneously aggregated into fibrils (Fig. 1), complex covalent crosslinks form within and between the collagen molecules to enhance the cohesiveness of the collagenous meshwork (Chapter 2). The only step of the crosslinking process that is known to be cell-controlled is the oxidative deamination (by lysyl oxidase) of the ε-amino group in telopeptidyl lysine and hydroxylysine residues to form the aldehydes, termed allysine and hydroxyallysine, respectively [59]. The hydroxyallysine-derived cross-links predominate in cartilage and lead almost exclusively to the formation of the stable, nonreducible and trivalent 3-hydroxypyridinium cross-link termed hydroxylysyl pyridinoline (also termed pyridinoline) whose concentrations change little during adulthood. On the other hand, with increasing age, the nonenzymatic glycation of lysine and arginine residues leads to the formation and accumulation of advanced glycation end products (AGEs) and crosslinks, such as pentosidine, methylglyoxallysine dimer, and threosidine [60, 61]. AGE crosslinking of the collagen network is believed to contribute to the age-related increase in stiffness of the collagen network, which, by decreasing the resistance of this network to mechanical damage, might predispose to the development of OA. Since type II collagen is the most abundant component of the collagenous meshwork in cartilage (90–95% of the total collagen content), the proteolytic degradation of this fibrillar collagen is likely an important rate-limiting step in tissue remodeling both in health and disease. Proteinases can contribute to the extracellular degradation of type II collagen in several ways. First, by cleaving the telopeptides of collagen molecules, proteinases separate the intact triple-helical domain from the telopeptide crosslinks and thus depolymerize the collagen fibrils without actually damaging the triple helix: the latter can be recognized by monoclonal antibodies that only react with the intact native triple-helical molecule [62]. Second, proteinases can also cleave the triple-helical domain of the native and fully wound collagen molecules. At physiologic temperature the cleaved triple helix unwinds (Fig. 1) to yield denatured collagen or gelatin and exposes neoepitopes that are normally hidden and not recognized in the native triple helix but are recognizable on cyanogen bromide (CB)
426
DANIEL-HENRI MANICOURT ET AL.
peptides of type II collagen [62]. Third, proteinases can also degrade gelatin and/or activate zymogens which hereby acquire the ability to cleave collagen molecules. Although neutrophil elastase and stromelysin-1 or metalloproteinase-3 (MMP-3) are the only mammalian proteinases known to cleave telopeptides in vitro [63, 64], studies conducted in cartilage explants [65] suggest that cleavage of the N-telopeptide by MMP-3 is unlikely to be a major mechanism in the chondrocyte-mediated degradation of collagen (see also Chapter 11). This notwithstanding, MMP-3 might play a pivotal role in collagen damage by activating collagenase-1 (MMP-1), collagenase-3 (MMP-13), and gelatinase B (MMP-9) [66, 67]. It also
should be noted that neutrophil elastase also can activate MMP-3 and MMP-8 [66]. Thus far, MMP-1, MMP-8 (neutrophil collagenase), MMP-13, and MMP-14 (membrane type I MMP or MT1MMP), which all are synthesized by chondrocytes, are the major mammalian proteinases known to be capable of hydrolyzing the intrahelical domain of native fibrillar collagens at neutral pH [68]. These collagenases cleave initially all three α chains of type II collagen at the classic G775 ↓ L776 bond, three-quarters of the way from the N-terminal end (Fig. 1). This produces denatured α chain fragments having approximately 3/4 length (TCA fragment) and 1/4 length (TCB fragment) of the native molecule.
Table I. Neoepitopes of Type II Collagen Cleavage. Summary of the Type II Collagen Neoepitopes and their Molecular Origin in the Triple Helical and C-telopeptide Domains of Type II Collagen Alpha Chains. 1. Schematic of type II collagen alpha chain cleavage by collagenases and matrix metalloproteinases First collagenase cleavage in triple helical domain -------GPPOHGPQG775 Ø LAGQRGICG-----Second collagenase cleavage in triple helical domain LAG778 Ø QRGICG-----Matrix metalloproteinase cleavage in C-telopeptide domain ------PREKGPDP Ø LQYMRA-----2. Neoepitopes located in the triple helical domain Identification
Immunizing antigen
Characteristics
COL2-1/4N1
LAGQRG
COL2-1/4N2
QRGIVG (LP)
COL2-3/4Cshort C1,2C
GPPOHGPQG GPPGPQG
COL2-3/4Clong C2C
GGEGPPOHGPQG
COL2-3/4m
APOHGEDGRPOHGP
COLL2-1 COLL2-1-NO2
HRGYPGLDG HRGY(NO2)PGLDG
Located at the N-terminus of the C-terminal α chain fragment after collagenase first cleavage of the triple helical domain of type II tropocollagen molecules This epitope is rapidly destroyed by secondary collagenase cleavage Located at the N-terminus of the C-terminal α chain fragment after collagenase second cleavage of the triple helical domain of type II tropocollagen molecules Located at the C-terminus of the N-terminal α chain fragment after collagenase first cleavage of the triple helical domain of type II tropocollagen molecules. Detected in body fluids Also present in a collagenase cleavage fragment generated from type I collagen Located at the C-terminus of the N-terminal α chain fragment after collagenase first cleavage of the triple helical domain of type II tropocollagen molecules. Detected in body fluids Digested by alpha chymotrypsin Residues 511–523 located in the N-portion of the N-terminal α chain fragment after collagenase first cleavage of the triple helical domain of type II tropocollagen molecules This neoepitope signs collagen denaturation: it is resistant to cleavage by alpha-chymotrypsin which is used to solubilize denatured collagen The sequence is also present in alpha 3 (XI) chain Residues 108–116 located in the N-terminal portion of the triple helical domain of type II tropocollagen molecules Residue Y may be nitrated by oxygen-derived free radicals Detected in body fluids The sequence is also found in type XI collagen
3. C-telopeptide neoepitope Identification
Immunizing antigen
Characteristics
CTX-II CartiLaps COL2-CTx
EKGPDP
Sequence found exclusively in human type II collagen C-telopeptide domain K indicate the lysine residue involved in intramolecular cross-linking Generated by matrix metalloproteinases. Detected in urine
Chapter 25 Products of Cartilage Metabolism
Immuno-histochemical staining and immunoassays conducted with antibodies that react specifically with denatured but not with native type II α chains (neoepitope COL2-3/4m; Table I), have established that adult normal human cartilage from large joints such as the hip or the knee, but not the ankle, may contain up to 3% of its total collagen in the denatured form: after 35 years of age, this can be detected around chondrocytes at and near the articular surface, and, with increasing age this damage may extend progressively into the cartilage [69–71]. Although denaturation of type II collagen is enhanced in cartilage obtained from joints with OA and RA, the topographical distribution of the collagen cleavage differs markedly between the two arthritides [72]. In OA cartilage, damage to collagen is first detected at and near the articular surface in territorial and interterritorial sites around chondrocytes and from there on extends into deeper cartilage layers, whereas in RA, denaturation of type II collagen is much more pronounced in the territorial matrix surrounding chondrocytes of the deeper layers [69, 70, 72]. These findings strongly suggest that cytokines derived from subchondral bone might enhance chondrocyte-derived degradation of the collagen meshwork in RA. After cleavage of the α chains by any one of the four collagenases (MMP-1, MMP-8, MMP-13, and MMP-14),
Figure 2 The three mammalian collagenases (MMP-1, MMP-8, and MMP-13) first cleave the triple helical domain of type II procollagen molecules at a site located three-quarters of the way from the N-terminus of the molecule. This cleavage unwinds the triple helix and exposes two neoepitopes, i.e. the C-terminus of the N-terminal α chain fragment, termed COL2-3/4 C, and the N-terminus of the C-terminal α chain fragment, termed COL2-1/4 N1. The N-terminal neoepitope COL2-1/4 N1 is rapidly destroyed by secondary cleavage sites of collagenases which also release the C-terminal neoepitope COL2-3/4 C that diffuses then out of the matrix to reach biological fluids where it can be detected. The amino acid sequence of the neoepitopes is also given. Adapted from reference [73].
427 unwinding of the triple helix exposes neoepitopes at the C-terminus of the TCA fragments and at the N-terminus of the TCB fragments (Fig. 2). Immunoassays conducted with antibodies reacting specifically with the C-terminal (COL2-3/4) and N-terminal (COL2-1/4 N1) neoepitopes (Table I) have demonstrated that, in both normal and OA cartilages, the content of COL2-3/4 neoepitope correlates strongly with the tissue content of denatured type II collagen [73]. Further, in OA cartilage, the enhanced denaturation of type II collagen becomes inversely proportional to the cartilage tissue content of this type of collagen [68]. These findings support the contention that collagenase activity is responsible for the cleavage of type II collagen in human articular cartilage and that any effective therapeutic intervention in OA must encompass a potent control of collagen degradation. When compared to other collagenases, MMP-13 hydrolyzes type II collagen more efficiently and also has a broader substrate specificity, cleaving type I, II, III, IV, IX, X, and XIV collagen, gelatin, fibronectin, and tenascin [74]. Studies of cartilage explants exposed to synthetic inhibitors of collagenases [75, 76] as well as the semi-quantification of MMP RNA transcripts [77], provide strong evidence that MMP-13 might be the most important collagenase involved in the degradation of type II collagen both in normal and OA cartilage, a contention further strengthened by the observation that transgenic mice over-expressing active MMP-13 exhibit joint pathology that mimics OA [78]. Further, during growth plate development, the synthesis and activity of MMP-13, but not MMP-1, by hypertrophic chondrocytes is associated with increased degradation of type II collagen [79]. On the other hand, proteolysis of newly synthesized collagen molecules might involve MMP-1 rather than MMP-13 [75], whereas other members of the MMP family such as MMP-2 (gelatinase A), MMP-3, and MMP-9 are very efficient at cleaving denatured α chains and at removing damaged collagen from the extracellular matrix [80]. It is thought that this subsequent clearance of unwound α chains may be a necessary event to facilitate repair but it is possible that the gelatinolytic enzymes cause additional damage to the cartilage matrix. Another enzyme likely to contribute to the turnover of type II collagen both in health and disease is cathepsin K. OA chondrocytes up-regulate the expression of this cysteine protease, which not only can cleave within the N-terminal portion of the helical domain of native type II collagen at near neutral pH, but also exhibits a high gelatinolytic activity in the pH range 4.0–7.0 [81]. Cathepsin K might be involved in the retinoic acid-induced type II collagen degradation which has been shown to be mainly independent of MMP activity in cartilage explants [82, 83].
428 The two epitopes present at the carboxy-terminus of the TCA 3/4 fragment (Table I and Fig. 2) and termed C2C (known previously as COL2-3/4Clong) and C1,2C (known previously as COL2-3/4Cshort) seem to be a better marker of collagen cleavage than the COL2-1/4 N1 neoepitope, since the N-terminal neoepitope is readily destroyed by a secondary collagenase-mediated cleavage. In contrast, as collagenases produce further cleavage of the TCA fragment, albeit at a slower rate, the C2C and C1,2C neoepitopes are released and can be readily detected in the conditioned medium of cartilage explants and in body fluids as well [62, 68]. Thus, in experimental canine OA and as early as at 12 weeks after joint destabilization, levels of the C2C neoepitope increase approximately 32-fold in synovial fluid and about 1.5-fold both in serum and urine [84] whereas the release of the C1,2C neoepitope is enhanced in the culture medium of human osteoarthritic cartilage explants and this increased release is selectively inhibited by a synthetic inhibitor that spares MMP-1, but not MMP-8 or MMP-13 [75]. Synovial fluid levels of the C2C neoepitope are also increased in experimental inflammatory arthritis [85] and in the early stages of human RA [57]. It is worth noting that in human RA the synovial levels of C2C correlate strongly not only with the synovial fluid levels of tumor necrosis factor alpha (TNF-α) but also with the serum levels of C-reactive protein [57]. The baseline serum ratio of C1,2C to C2C, but not the levels of individual markers, is predictive of disease progression in patients with knee OA [86]. The neoepitope COLL2-1 located near the N-terminus of the TCA fragment (Table I) and its nitrated form (COLL2-1-NO2) are also released into the synovial fluid and from there on reach the systemic circulation [87]. Serum levels of COLL-1 are increased 1.5-fold in both OA and RA whereas serum levels of COLL-1-NO2 are enhanced about two-fold in OA and three-fold in RA [88]. Since chondrocytes and other cells of the joint organ produce nitric oxide and superoxide anion, which then react to yield peroxynitrite anion, these neoepitopes might be promising tools to evaluate the contribution of oxygenderived species to the degradation of type II collagen [89–91]. Some degradation products of type II collagen may, acting in a positive-feedback mechanism, stimulate collagen gene expression and promote the differentiation of cartilage cells [68]. There is indeed evidence that cleaved, denatured type II collagen not only can induce the production of MMP-1, MMP-13, interleukin-1 (IL-1), and TNF-α by chondrocytes but also, and most importantly, can induce hypertrophy of cartilage cells, a process characterized by the expression of type X collagen and the up-regulation of cbfa1, the MMP-13 transcription factor [68].
DANIEL-HENRI MANICOURT ET AL.
Therefore, chondrocyte apoptosis and calcification of cartilage matrix, which are both commonly observed in OA cartilage, might result, at least in part, from chondrocyte hypertrophy [92]. On the other hand, an evolutionary conserved domain of type II collagen, located between amino acids 359 and 369, might be a target of pathogenic autoimmune responses in the induction or perpetuation of joint inflammation in RA [93]. Little is known about the degradation of other collagens in cartilage. Type IX collagen can be cleaved by MMP-3 in its NC2 domain [64] and by an unknown protease in its NC4 domain [94]. Loss of the NC4 domain is an early event in experimental inflammatory arthritis: it accompanies proteoglycan loss and precedes detectable damage to type II collagen within the cartilage [85]. Type IX collagen degradation might decrease the tensile stiffness of cartilage since this collagen protrudes from the surface of the type II collagen fibril where it apparently functions as a covalent bridge between collagenous fibrils [59]. On the other hand, the degradation of type IX collagen also might allow the diameter of collagen fibrils to increase during growth and maturation since the type IX molecules must be removed before fibril growth can occur, either by accretion of additional type II molecules and/or by lateral fusion of thin fibrils (see Chapter 25). When cartilage collagens are degraded, crosslinked N- and C-telopeptide fragments bearing pyridinoline crosslinks are released into proximal fluids, i.e. synovial fluid in the case of a diarthrodial joint, and from there on enter the systemic circulation before being most effectively cleared by filtration through the kidneys. Pioneering work has detected pyridinoline, but not deoxypyridinoline (the “bone-specific” crosslink), in OA and RA synovial fluids [95]. Although the urinary excretion of pyridinoline increased approximately two-fold in patients with OA and four-fold in patients with RA [96–98], it was not clear how much of this increase could be attributed to degradative changes occurring in cartilaginous structures, since the increase in urinary pyridinoline originating from cartilage is likely to be blurred by the concomitant increase in the turnover of other structures of the diseased joints, including bone and synovium [99–101], a contention strengthened by the observation that patients with knee and hip OA exhibit high urinary levels of the glycosylated pyridinoline derivative, glucosyl-galactosyl-pyridinoline, which is found in large amounts in human synovium, and in very low levels in the cartilage and other soft tissues [102]. The development of immunoassays capable of quantifying the neoepitope EKGPDP bound to the crosslinks and generated by proteolytic cleavage of the C-telopeptide domain of type II collagen fibrils has dramatically simplified
429
Chapter 25 Products of Cartilage Metabolism
the interpretation of analyses of pyridinium crosslinks in body fluids [103, 104]. Several MMPs, including MMP-8, MMP-13, and MMP-14, can generate this neoepitope termed CTX-II, COL2CTX, or CartiLaps, but MMP-7 (matrilysin) is the most potent [103]. Enhanced levels of CTX-II have been found in the culture medium of human articular cartilage explants stimulated with the catabolic cytokine oncostatin M [105] as well as in the synovial fluid of patients with joint injury, OA, pseudogout, and septic arthritis [106]. In the early stages of canine experimental OA, levels are increased by an order of magnitude of six in synovial fluid and of two in serum [84]. The urinary excretion of CTX-II increases approximately 1.5-fold in patients with OA and 2.3-fold in patients with RA [52, 104]. Further, in patients with knee and hip OA, levels of the urinary excretion of CTX-II are associated with both the prevalence and the progression of radiographic OA changes, and this independently of known clinical risk factors for osteoarthritis [107]. Further, in patients with early RA, the magnitude of the changes of urinary CTX-II levels at 3 months after induction of therapy are predictive, independent of changes in disease activity, of the changes in radiologic score of joint destruction after 1 year [108]. Combining urinary CTX-II excretion, a marker of type II collagen degradation, with serum levels of N-propeptide of type IIA procollagen, a marker of type II collagen synthesis, might be more effective in predicting
Figure 3
cartilage destruction than measurements of a single marker [52].
IV. PRODUCTS OF AGGRECAN METABOLISM The numerous chondroitin sulfate (CS) and keratan sulfate (KS) chains carried by the long-extended interglobular domain (IGD) of the aggrecan core protein confer to this PG a bottle-brush structure with viscoelastic properties (Fig. 3). This elaborate architecture, however, leaves vulnerable to proteolytic attack the glycosaminoglycan (GAG)-poor N-terminal end of the molecule that separates the glycosaminoglycan (GAG) attachment region from the first globular domain (G1) that interacts with HA and LP and thus anchors the aggrecan molecule in the matrix. Newly secreted aggrecan molecules have low affinity for HA and their maturation into the high-affinity form probably involves the formation of disulfide bonds in the G1 domain (see Chapter 5). This, coupled to an accelerated transport, termed “rapid polymer transport”, may enable newly secreted molecules to diffuse away from the metabolically active pericellular environment and enter the metabolically inactive interterritorial matrix compartment before aggregating [109]. In contrast, LP molecules are fully capable of binding to both HA and aggrecan molecules
Through its globular domain G1, aggrecan interacts with hyaluronan (HA) and link protein (LP) to form supramolecular aggregates of very large size. The function of G2 is unknown and G3 has a role in intracellular translocation. Some of the antibodies that react with the core protein and the chondroitin sulfate (CS) and keratan sulfate (KS) chains are indicated. Other monoclonal antibodies (Mabs) react with the chondroitin sulfate chain that remains attached to the core protein after chondroitinase ABC lyase digestion. Adapted from reference [43].
430
DANIEL-HENRI MANICOURT ET AL.
when they are released extracellularly. Binding of LP to newly synthesized aggrecan molecules might provide the stability the latter need for interacting early with HA [110]: this may play a key role in determining what proportion of newly synthesized aggrecan molecules become established in the pericellular/cell-associated matrix. While studies of the racemization of aspartic acid have suggested that aggrecan molecules turn over much more rapidly (average half-life = approximately 2 years) than collagen molecules within the fibrillar network [111, 112], the results of both in vivo [113] and in vitro [109, 114, 115] studies have suggested that articular cartilage contains at least two pools of aggrecan molecules with different rates of turnover. It is thought that the two pools of aggrecan molecules reside in different compartments of the matrix: a cell-associated matrix compartment in which they turn over rapidly and a further removed, more abundant, interterritorial matrix compartment in which the PGs turn over much more slowly [109, 115]. The catabolism of aggrecan is a proteolytic process and multiple cleavage sites have been observed along the protein core [116–119] (Fig. 4). GAG-rich fragments generated by proteolytic cleavage of the core protein are lost relatively rapidly from the cartilage matrix. In the case of articular cartilage, they diffuse into the synovial fluid
Figure 4
where they can be quantified by chemical assays of their sulfated glycosaminoglycans [120, 121] or by immunoassays capable of measuring specific carbohydrate epitopes [84, 122–125]. They also can be characterized by peptide sequence analysis after isolation and deglycosylation [126, 127]. The development of monospecific antibodies that recognize enzyme-generated neoepitopes has proved most useful for identifying and quantifying specific proteolytic cleavage products [128–134]. Proteinases of different classes and hydroxyl radicals [135] have the capacity to cleave the aggrecan core protein in vitro. Importantly, the use of antibodies against these types of neoepitopes has now provided strong evidence that at least two families of zinc-dependent metalloenzymes, the metalloproteinases (MMPs) and the ADAMTS or aggrecanases, play major roles in vivo in the catabolism of aggrecan molecules both in health and disease (Fig. 4). When proteolytic processing is confined to the C-terminus of the GAG-rich region, the aggrecan molecules remaining within cartilage matrix retain some GAGs and thus are able to continue to exert their physicochemical properties. The contention that progressive C-terminal trimming of aggrecan molecules does occur in vivo is supported by N-terminal sequencing of aggrecan fragments released into the conditioned medium of cartilage explants
Major sites of cleavage of aggrecan core protein by tissue proteinases. The two major sites of cleavage are located in the short extended interglobular domain between the globular domain G1 and the globular domain G2: the first one occurs between residues Asn341– Phe342 whereas the second one is located between residues Glu 373 and Ala374. Both cleavage sites separate the anchoring G1 domain from the bulk of the aggrecan molecule carrying the glycosaminoglycan side chains and thus deprive cartilage from its load-bearing properties. The N-terminal and C-terminal amino acid sequences of the neoepitopes resulting from the different cleavage sites as well as the nature of the proteinases acting at these sites are also indicated. The correct size of different domains is not respected. A disintegrin and metalloproteinase with thrombospondin-like motifs (ADAMTS); cathepsin B (cat B); matrix metalloproteinase (MMP).
Chapter 25 Products of Cartilage Metabolism
[126, 136, 137], as well as by physiochemical and electron microscopic analysis of aggrecan molecules extracted from cartilage. These studies have identified in adult articular cartilage an abundance of polydisperse molecules lacking the G3 domain and exhibiting a great variability in the length of their CS-rich domain [116, 138–143]. This C-terminal processing of aggrecan also occurs early in growth and might be viewed as a “nondestructive” tailoring of aggrecan that promotes cartilage matrix assembly and organization [144]. On the other hand, too extensive C-terminal processing of aggrecan with significant loss of CS-containing regions of the molecule might contribute significantly to loss of tissue function both in aging and disease. Although the C-termini of the majority of truncated aggrecans are apparently located in regions of the core protein with known MMP-sensitive cleavage sites, their amino acid sequence remains unknown, and thus the nature of the proteinase(s) responsible for the generation of these products is still unsolved. On the other hand, cleavage of the core protein near the G1 domain is more “destructive” as it results in rapid loss of the whole GAG-attachment region and thus has more adverse consequences, at least from the functional standpoint [145]. Two critical cleavage sites have been identified in the short extended IGD that separates the G1 and G2 domains: one occurs at the N341 ↓ F342 bond and the other site is located at the E373 ↓ A374 bond (Fig. 4). The first critical cleavage site generates a G1-containing fragment with the C-terminal neoepitope VDIPEN341 and a larger GAG-bearing fragment with the N-terminal neoepitope 342FFGVGGE [132, 133, 146–148]. It is the major cleavage site of several metalloproteinases including stromelysin-1 (MMP-3 ), the three collagenases (MMP-1, MMP- 8, and MMP-13), the two gelatinases (MMP-2 and MMP-9) as well as MMP-7 and MMP-14 (MT1-MMP). It can also be cleaved by the combined endopeptidase and carboxypeptidase activity of cathepsin B, but at low pH [149]. MMPs appear to be involved in the conservative baseline turnover of aggrecan: MMP-derived 342FFGVG fragments are found in small amounts in the culture medium of pig articular cartilage where they represent about 1% of the total amount of aggrecan released [150, 151]. Further, and as observed in human synovial fluids, these fragments recovered in culture medium usually have a small size, and, because they do not carry the NITEGE373 C-terminus generated by aggrecanases, they are likely to be the product of more extensive MMP processing [131, 133]. However, while most degraded aggrecan components are quickly released in culture, the majority of 342FFGVG fragments are retained in cartilage tissue. Even upon stimulation with IL-1, the 342FFGVG fragments increase significantly
431 in the tissue without a corresponding increase in 342FFGVG neoepitope in the medium. Because the 342FFGVG products of MMP catabolism are large, resistant to destruction by aggrecanases and selectively retained in cartilage, they may have a role in conveying signals or organizing a microenvironment [151]. The 342FFGVG fragments remaining in cartilage matrix could represent a non-negligible proportion of the “nonaggregating” large proteoglycans of cartilage and/or contribute to the metabolic pool of aggrecans which exhibit a long half-life [109, 152]. MMPs have the capacity to cleave at other sites within the short IGD of aggrecan molecules, but this requires higher enzyme concentrations. Thus, MMP-1, MMP-7, MMP-8, and MMP-13 all cleave the D441 ↓ L442 bond (Fig. 4), whereas MMP-3 also cleaves at S377 ↓ V378, MMP-8 at Q373 ↓ A374, and MMP-13 at P384 ↓ V385 [148]. Although it may be generated by both MMP-8 and MMP-14 [153], the second critical cleavage at the E373 ↓ A374 bond is believed to be the signature of the glutamyl endopeptidases termed ADAMTS [153–155]. This produces a large GAG-rich aggrecan fragment with the N-terminal neoepitope 374ARGSVI and a G1 fragment with the C-terminal neoepitope NITEGE373 (Fig. 4). The bulk of aggrecan fragments found in synovial fluid of patients with joint injury, OA, rheumatoid arthritis (RA), and other types of inflammatory arthritis all have the N-terminal sequence ARGSVI [127, 156, 157]. These fragments are also found in enhanced amounts in the culture media of both chondrocytes and cartilage explants stimulated to undergo matrix degradation with IL-1 or retinoic acid [154, 158, 159]. However, the cleavage site at the E373 ↓ A374 bond within the short IGD of aggrecan molecules is not the preferential site of recombinant ADAMTS-4 and ADAMTS-5: the enzymes first cleave within the CS-rich G2–G3 domain at the KEEE1714 ↓ 1715GLGS bond to release the COOH terminus of the molecule (Fig. 4) before contributing to the removal of an additional portion of the COOH terminus by cleaving at the ASELE1539 ↓ 1540GRGT bond [119, 155]. While the C-terminus portion of the molecule released by the first cleavage is further processed at two additional cleavage sites, at TAQE1819 ↓ 1820AGEG bond and at ISQE 1919 ↓ 1920LGQR bond, the cleavage within the short IGD at the E373 ↓ A374 bond occurs more slowly. Thus far, it is not clear whether the first cleavages in the C-terminal portion of the core protein are required and/or facilitate the cleavage in the short IGD domain of the aggrecan molecule. On the other hand, it is clear that the successive cleavages of the protein core within the GAG-rich region are facilitated by the binding of the C-terminal thrombospondin motif of ADAMTS to the GAG side chains [118] and, accordingly, exogenous chondroitin sulfate and
432 heparin both inhibit aggrecanase activity [160], whereas changes in glycosylation with age may alter susceptibility to cleavage by ADAMTS [161]. It is now obvious that ADAMTS-mediated aggrecan cleavages occur both in vivo and in vitro: the neoepitopes generated by the four cleavage sites located within the GAG-rich domain (E1539 ↓ G1540, E1714 ↓ G1715, E1819 ↓ A1820, and E1919 ↓ L1920) have been found in both human and bovine articular cartilage [137, 144], in rat chondrosarcoma [162], in the culture medium of cartilage explants stimulated with either IL-1 and/or retinoic acid [116, 119, 137] as well as in high-density aggrecan fragments recovered from human synovial fluids [144]. After cleavage at the two critical sites within the short IGD, the majority of VDIPEN341 and NITEGE373 neoepitopes resident on G1 fragments remain in tissue bound to hyaluronan. Both neoepitopes are detected not only in cartilage from joints with OA and RA, two conditions exhibiting quite contrasting pathological and clinical features, but also in cartilage from normal adult joints where they appear to accumulate with age [134, 163]. The generation of these neoepitopes is not necessarily coordinated since both the NITEGE373 and VDIPEN341 neoepitopes can map predominantly to different regions within a single normal joint, can show a temporal and spatial separation in arthritis models and in developing cartilage [67, 132, 134, 164, 165], and since turnover of aggrecan by cultured rat chondrosarcoma cells and primary bovine chondrocytes can be mediated almost exclusively by aggrecanase [130]. On the other hand, although there is evidence that MMP cleavage of the core protein at the N341 ↓ F342 bond and ADAMTS cleavage at the E373 ↓ A374 bond are independent [134, 166] and mutually exclusive [148, 167], the detection of the VDIPEN341 neoepitope does not prove that the primary cleavage of the core protein results from MMP activity as MMPs and ADAMTS4 both can cleave the N341 ↓ F342 bond of a G1NITEGE373 fragment to yield a G1- VDIPEN341 fragment [167, 168]. It is generally believed that, in vivo, ADAMTS play a major role in the degradation of aggrecan during the early stages of cartilage degradation whereas MMP-mediated catabolism of aggrecan occurs as a late event in cartilage matrix breakdown, at the time when active degradation of collagen is occurring [124, 169]. Indeed, in both models culture systems and murine arthritis, the NITEGE373 neoepitope appears very early, well before the appearance of the VDIPEN341 neoepitope, and importantly in proportion to GAG loss from the tissue [124, 170–172]. The ADAMTS-mediated early loss of aggrecan from cartilage could facilitate collagen proteolytic cleavage [173]. On the other hand, studies conducted in MMP-3-deficient mice [67] have pointed out that, although MMP-3 plays
DANIEL-HENRI MANICOURT ET AL.
no role in early aggrecan loss, this MMP is essential for generating both aggrecan and collagen neoepitopes, and that these epitopes co-localize and correlate with disease progression. Although ADAMTS engender cartilage aggrecan catabolism during the primary phases of arthritic diseases, it is still not known which iso-form(s) play(s) a major role. Regulation of the ADAMTS genes is still poorly understood, though there is evidence that the effect of growth factors, hormones, and inflammatory cytokines are dependent on age, species, and stimulus [124]. In bovine cartilage ADAMTS4 is not constitutively expressed but is up-regulated by IL-1, TNF-α, and fibronectin fragments whereas ADAMTS5 is constitutively expressed with little up-regulation in response to cytokines [154]. In another study, 90% of the aggrecanase activity in media from bovine articular cartilage stimulated with IL-1 could be inhibited by immunoprecipitation with a combination of antibodies to ADAMTS4 and ADAMTS5, indicating that these proteins are primarily responsible for the aggrecanase activity stimulated by IL-1 in this system [119]. On the other hand, retinoic acid induces a marked increase in the release of ADAMTS-generated aggrecan catabolites from human cartilage with no concomitant increase in the mRNA levels for any of ADAMTS1, 2, 3, 4, and 5 [174], whereas ADAMTS1, 4, and 5 mRNAs are all regulated by a combination of IL-1 and oncostatin M, though with differing magnitude and kinetics [175]. IL-17 also up-regulates ADAMTS4 via phosphorylation of ERK, p38, and JNK MAP kinases [176]. In human OA cartilage samples obtained at late stages of the disease process, ADAMTS1, 5, 9, and 15 were down-regulated while ADAMTS2, 12, 14, and 16 were up-regulated by comparison with phenotypically normal cartilage obtained from fracture patients [177]. Interestingly, the incorporation of n-3 fatty acids into chondrocyte membranes causes a dosedependent reduction in both the expression and activity of ADAMTS [178]. As observed for MMPs, control of ADAMTS activity is exerted at multiple points including zymogen activation, substrate accessibility, and inhibition by naturally occurring inhibitors. In the case of ADAMTS4, which is the best characterized of the aggrecanases, C-terminal processing of the 75-kDa full-length active form produces isoforms of 60 and 50 kDa [179]: this results in the release of the enzyme from the ECM and alteration of its profile activity [180]. The C-terminal spacer region of the 75-kDa isoform can act to inhibit the aggrecanase activity at the E373 ↓ A374 bond because the 75-kDa isoform cleaves at E1480 ↓ G 1481 but not at E373 ↓ A374: this is in contrast to the 60- and 50-kDa isoforms, which both lack the spacer region and cleave at the E373 ↓ A374 bond. Further, binding of the C-terminal spacer region to fibronectin could anchor
433
Chapter 25 Products of Cartilage Metabolism
the 75-kDa isoform to the ECM and/or modulate aggrecanase substrate specificity [181, 182]. On the other hand, although the tissue inhibitors of MMPs (TIMPs) are broadly effective inhibitors of the MMPs [183], it is clear that they display much greater selectivity towards ADAMTS. ADAMTS4 and ADAMTS5 are both potentially inhibited by TIMP-3, while being insensitive to TIMP-1, 2, and 4 [184], whereas ADAMTS1 is partially inhibited by TIMP-2 and TIMP-3 [185]. Alpha-2macroglobulin is another endogenous inhibitor of both ADAMTS4 and 5 [186]. As stated in Chapter 5, several glycosyl- and sulfotransferase enzymes are implicated in the biosynthetic pathways of aggrecan CS chains whose disaccharide units consist of glucuronic acid (GlcA) and N-acetyl-galactosamine (GalNAc). The ratio of 6- to 4-sulfation of the GalNAc residues present in the internal disaccharides increases markedly with age and within the successive cartilage zones with distance from the articular surface [187, 188]. Because the distinct 6-sulfotransferase activity, with specificity for the nonreducing terminal GalNAc, appears after skeletal maturity, the majority of terminal GalNAc residues are 4-sulfated in juvenile aggrecans and 4,6disulfated in adult aggrecans [187]. In both normal adult
Table II.
and OA articular cartilage, the size of CS chains on aggrecan is similar but smaller to that of juvenile aggrecans; however, alterations in the GAG biosynthesis pathways of OA chondrocytes lower the capacity for 4,6-disulfation of the nonreducing termini of CS chains, thus resulting in the synthesis and deposition of CS chains with predominantly nonsulfated termini. Therefore, based on CS chain length and pattern of sulfation of nonreducing termini, it is tempting to speculate that OA chondrocytes synthesize aggrecans substituted with CS chains similar to those found in juvenile aggrecans [189]. The fine structure of CS chains is sensitive to the physicochemical environment (pH, ionic strength) of the biosynthetically active chondrocytes [17] and can also be modulated by growth factors [190], cytokines [191], and mechanical factors [192]. Dynamic compression of cartilage tissue increases not only the length of CS chains but also decreases the abundance of the 4,6-disulfated nonreducing terminal N-acetylgalactosamine residues [192]. Presumably, these marked changes in the sulfation pattern of CS aggrecan chains initiated by chondrocytes enable the cartilage tissue to respond more efficiently to biological changes in the cell environment, and in turn, the physiochemical properties of aggrecans are influenced by these structural events.
Monoclonal Antibodies Commenly Used to Analyze Aggrecan and Other Components of Proteoglycan Aggregates
Monoclonal antoibody
Antibody specificity
5-D-4 and 1-B-4
Skeletal and corneal keratan sulfate. Highly sulfated sequence of serveral repeats of 6-sulfated N-acetylglucosamine-b1, 3-6sulfated galactose-b1, 4 Keratan sulfate. Sulfated poly (N-acetyl lactosamine) sequence of hepta- or larger oligosaccharides of 6-sulfated galactose and N-acetylglucosamine Generated by chondroitinase digestion of aggrecans. Delta-unsaturated disaccharides of chondroitin-4-sulfate, but not chondroitin-6-sulfate and unsulfated chondroitin Intact chondroitin-6-sulfate chains but not unsaturated disaccharides of chondroitin-6-sulfate Nonreducing termini of native chondroitin sulfate chains containing a terminal hexuronate residue adjacent to a 6-sulfated N-acetylglucosamine residue. Delta-unsaturated or saturated oligosaccharides of chondroitin-6-sulfate generated by chondroitinase or tessticular hyaluronidase digestion Nonreducing termini of native chondroitin sulfate chains containing a terminal hexuronate residue adjacent to a 4-sulfated N-acetylglucosamine residue. Delta-unsatutated or saturated oligosaccharides of chondroitin-4-sulfate generated by chondroitinase or testicular hyaluronidase digestion Delta-unsaturated or saturated oligosaccharides of chondroitin-4-sulfate generated by chondroitinase or testicular hyaluronidase digestion Delta-unsaturated or saturated oligosaccharides of unsulfated chondroitin generated by chondroitinase or testicular hyaluronidase digestion Chondroitin sulfate oligosaccharide containing 10–12 sugar residues. Probably contains several of the eight possible CS-isomer disaccharide combinations Complex epitopes residing within native chondroitin sulfate chains
AN9P1 2-B-4 (B) and 9-A-2 (B) 846 (A) 3-B-3 (C)
3-D-5 (C)
2-B-6 (B) 1-B-5 (B) 7-D-4 (A) 4-C-3 (A) 4-D-3 (A) 6-C-3 (A) 8-A-4 8-A-5
Determinant comment to link protein (LP) isolated from aggregates recovered from cartilage of different species Determinant present in the three molecular forms of LP (LP1, LP2, and LP3) Similar to that of 8-A-4. Reduction and carboxymethylation of the link protein does not alter the epitope
Note. The list is not exhaustive. Monoclonal antibodies directed against chondroitin sulfate epitopes can be divided into two categories: (1) those that recognize epitopes present on the native chondroitin sulfate glycosaminoglycan chains (A) and (2) those that require predigestion of the chondroitin sulfate glycosaminoglycan with endo- or exoglycosidases to generate their reducing teminal epitopes (B). Some antibodies (C) recognize epitopes that fall into both of these categories.
434 Monoclonal antibodies have been produced with specificities against structural carbohydrate epitopes on the CS chains (Table II and Fig. 3). With the exception of the 3-B-3 epitope, which corresponds to the nonreducing terminal sequence GluAβ1, 3GalNac-6-sulfate, the precise structure of these (neo)epitopes remains to be defined and it is not known to what extent their reactivity detected by Western blots or immuno-histochemistry could be affected by the CS chain length and/or degree of sulfation [187]. Two epitopes, termed 3-D-5 and 7-D-4, are present in normal mature articular cartilage [193]. The content of the 7-D-4 epitope in aggrecan increases markedly in diseased cartilage from different species [193–195] as well as in the synovial fluid from human knees with acute traumatic cruciate ligament and meniscal injuries [196], but it is decreased in RA synovial fluids where it shows a negative association with the degree of joint inflammation [197]. The CS epitope termed 846 does not contain unsaturated disaccharides of chondroitin-6-sulfate [84, 198]: it is found at highest concentrations in fetal cartilage; it is barely detectable in adult cartilage but reappears in enhanced amounts in cartilage from OA and RA joints [195, 198, 199]. As it is present on intact CS chains located in close proximity to the G3 domain of the core protein of newly synthesized aggrecan molecules, it may be a useful marker of aggrecan synthesis during inflammation and disease processes [84, 198]. While the percentage of CS chains bearing the 3-B-3 epitope in articular cartilage remains constant, at about 9%, throughout life [187], its immunoreactivity varies widely with the stage of development, maturity, and pathology [194, 195, 200]. This apparent discrepancy might be related, at least in part, to the observation that the recognition of CS chain termination by the 3-B-3 antibody on solid phase assay could be influenced by the sulfation and/or length of the chains ending with 3-B-3 epitope [187, 189]. Accordingly, the enhanced levels of 3-B-3 reactive aggrecan fragments detected in human synovial fluid after traumatic knee injury [196] might just reflect the release of degraded aggrecans that are resident in the healthy adult cartilage, whereas the increased levels observed in human OA synovial fluid [197, 201] might relate to a change in metabolic activity of chondrocytes, with alterations in the sulfation and length of the CS chains modulating the immunoreactivity of the chains bearing the 3-B-3 epitope. In OA patients, synovial fluid levels of the 3-B-3 epitope cannot distinguish between progressive and nonprogressive OA [202], and do not change significantly after a 12-week program of walking exercise [201]. Measurement of GAG epitopes can also provide a measure of the turnover of PGs. Immunoassays that use
DANIEL-HENRI MANICOURT ET AL.
monoclonal antibodies (Table II and Fig. 3) capable of recognizing KS epitope(s) in intact aggrecan molecules or fragments thereof [203, 204], have shown that levels of antigenic KS in the blood circulation correlate well with the concentration of aggrecan core protein-related epitopes in serum, suggesting that the serum levels of agKS offer a good measure of the level of aggrecan-derived fragments in body fluid (Chapter 32). Although the function of KS is still unclear, the rate of synthesis of this molecule appears to be inversely related to the ambient oxygen tension [205], which may explain why the the KS content of aggrecan increases with distance from the articular surface [206]. On the other hand, aggrecans exhibit an agerelated increase in the number, size, and sulfation of their KS chains whereas the KS content of aggrecans recovered from OA cartilage is markedly reduced [207, 208].
V. PRODUCTS OF THE METABOLISM OF OTHER PROTEOGLYCANS Members of the family of Small Leucine-Rich Proteoglycans (PGs), termed the SLRPs, have been identified in adult articular cartilage [209] (see also Chapter 4). Biglycan, decorin, and fibromodulin all have the capacity to bind to transforming growth factor betas (TGF-β) via their core protein with both high-affinity and low-affinity binding sites, and hence these small PGs may play an important role in regulating the activity of TGF-β and its impact on the synthesis of matrix components. On the other hand, studies conducted in cartilage explants have pointed out that TGF-β enhances the rate of synthesis of both decorin and fibromodulin, but not biglycan [210]. The surface of collagen fibers is decorated by decorin and fibromodulin, and the message level for these two members of the SLRPs family shows increases with increasing age in human articular cartilage [211]. Although fibromodulin might regulate collagen fibril diameter, this does not seem to be true for decorin as no effects on type II collagen fiber diameter in the decorin knockout mice have been reported as of yet [212]. Because of its specific localization in the pericellular space as well as on the chondrocyte cell surface, biglycan may have important functions in modulating morphogenesis and differentiation. Mice with a targeted disruption of the biglycan gene are apparently normal at birth, with no skeletal patterning defects [213]. However, the biglycan-null mice, that express unaffected levels of decorin, showed decreased postnatal skeletal growth, indicating that biglycan is a positive regulator of bone formation and bone mass. Fibromodulin and decorin can be cleaved by ADAMTS-4 [214]. Although Poole et al. [215] reported
435
Chapter 25 Products of Cartilage Metabolism
no overall differences in biglycan and decorin contents in human knee OA, other studies have reported an increase in the content of intact biglycan, decorin, and fibromodulin as well as fragments of these PGs in both human OA and RA cartilage [216–218], suggesting they are involved in the attempt at matrix repair by chondrocytes within diseased cartilage. On the other hand, patients with RA expressed an increased humoral immunity to biglycan whereas patients with seronegative spondyl-arthropathies exhibit enhanced decorin-specific antibodies [219]. Lumican and the protein known as PRELP (Proline arginine-Rich End Leucine-rich repeat Protein) are two other members of the SLRPs found in cartilage matrix. The glycosylation pattern of lumican by chondrocytes can be modulated by interactions of the cell with surrounding matrix components as well as by growth factors and cytokines: thus, while basic fibroblast factor, insulin-like growth factor and TGF-β all induce the secretion of a lumican form bearing KS chains, IL-1 promotes the secretion of a KS-deficient form of lumican [220]. On the other hand, PRELP has four potential N-linked glycosylation sites and apparently functions as a cartilage matrix protein with the capacity for matrix organization [221]. Since it has the capacity to simultaneously bind collagen and perlecan, PRELP might function as a linker between the matrix close to the chondrocyte and the more distant matrix via perlecan [222]. In adult articular cartilage, perlecan is a CS/heparin sulfate hybrid PG that is found in the territorial matrix with highest intensity in the pericellular region [223]. Human perlecan deficiencies have been found in chondrodysplasias [224, 225]. Mice lacking perlecan develop a severe cartilage phenotype characterized by a reduced collagen network and shorter collagen fibrils, implying an important role for perlecan in the cartilage structure [225, 226]. Perlecan might also maintain chondrocyte differentiation, perhaps working in concert with SOX9, and/or influence chondrocyte metabolism via signaling receptors or by modulating the expression of fibroblast growth factors [227, 228].
VI. PRODUCTS OF THE METABOLISM OF LINK PROTEIN AND HYALURONAN The synthesis of LP follows the biosynthetic pathway of other secretory proteins through the endoplasmic reticulum and the Golgi. SOX9 and its known transcriptional co-activators L-SOX5 and SOX6 are potent inducers of LP expression [229] whereas TNF-α suppresses LP expression through the MEK1/2 and NF-κB pathways [230]. Although the rate of LP synthesis is up-regulated by mechanical compression of cartilage [231], the
age-related decrease in the rate of LP synthesis might regulate the rate at which newly synthesized aggrecan acquires affinity for HA [232]. Human articular cartilage LP can be separated by SDS/PAGE into three components of molecular mass 48, 44, and 41 kDa, referred to as LP-1, LP-2, and LP-3, respectively [233]. LP-1 and LP-2 are different glycosylated forms of the same intact protein core, while LP-3 is proteolytically derived from either LP-1 or LP-2. In cartilage of newborns, LP-3 is a minor component, while, with age, LP-3 accumulates at the expense of LP-1 and LP-2. Three major cleavage sites have been identified in the N-terminal region of intact LPs (Fig. 5). The first one occurs at the H18 ↓ I19 bond: it is generated by stromelysins 1 and 2, gelatinases A and B as well as by collagenase [234]. Before being inactivated by rapid specific cleavage mediated by membrane-associated proteinases [235], the cleaved N-terminal peptide (DHLSDNYTLDHDRAIH or Link N) produced thereof stimulates the biosynthesis of both PG and type II collagen synthesis by human articular cartilage [236, 237] and hence this peptide may have an important role in the feedback control of cartilage matrix synthesis. On the other hand, LP3 from mature cartilage exhibits two additional amino termini: one is compatible with
Figure 5
Sites of cleavage of link protein (LP) in human articular cartilage. In vitro many proteolytic agents are able to cleave the link protein within the N-terminal region between amino acid residues 14 and 29 to produce the LP3 forms of the molecule. In contrast, only a few proteolytic agents can cleave the link protein in vitro within the disulfide-bonded loop between amino acid residues 56 and 87 to produce LP fragments upon reduction. The site for LP3 generation (1 between residues 16–17; 2 between residues 18–19 and 3 between residues 23–24) and LP fragment generation (4 between residues 65–66 and 5 between residues 72–73) in vivo occur within the same regions that are susceptible to in vitro proteolysis. The sites cleaved by metalloproteinases are also shown. Adapted from reference [233].
436 cleavage at the E18 ↓ A19 bond, a property of cathepsins B and G, whereas the other one occurs at the P23 ↓ H24 bond; the enzyme(s) responsible for this site of cleavage has not been identified. As the three forms of LP possess the three disulfide-bonded loops that are associated with LP function, LP3 generation is unlikely to affect the stability of PG aggregates. In vitro studies have also pointed out that the N-terminal disulfide-bonded loop of LP molecules can be cleaved at one site by cathepsin L [233], and at two sites by hydroxyl radicals [238]; however, none of these corresponds to the sites of cleavage observed in vivo, i.e. at the K65 ↓ W66 bond and the D72 ↓ Y73 bond. Whether these native cleavage sites result from the action of a yet to be identified enzyme remains to be established. On the other hand, the continued ability of fragmented LP molecules to interact with HA probably reflects the lack of proteolytic modification within the C-terminal disulfide-bonded loops. Very little is known about the half-life and turnover of LP in the cartilage matrix. During arthritides, the loss of LP from cartilage matrix into the synovial fluid may elicit the local production of antibodies and the expression of cellular immunity to LP: patients with RA, juvenile RA (JRA), ankylosing spondylitis, and OA express cellular immunity to LP, which is correlated with disease activity and severity in JRA [239–242]. In canine experimental OA, the synovial fluid levels of LP and the rate of loss of this glycoprotein from cultured cartilage explants increases several-fold within 3 months of disease induction [243, 244]. In contrast, the synovial fluid levels of LP are markedly decreased in disused joints [243]. These findings are in close agreement with the progressive disappearance of the link-rich, more saturated, aggregates from the cartilage matrix in experimental OA and with the maintenance of these fast-sedimenting aggregates in disuse atrophy [26]. Although articular cartilage from canine OA joints and from disused joints both exhibit enhanced levels of matrixdegrading enzymes [27], these cartilages differ in their HA content: the latter is markedly reduced in OA and unchanged in disuse. The same holds true for human OA cartilage. Since, in contrast to the disuse process, the OA process is apparently irreversible, one might argue that the loss of both HA and link-rich aggregates prognosticates articular cartilage destruction. Three mammalian HA synthase (HAS) genes have been cloned and referred to as HAS1, HAS2, and HAS3 [245]; the different HA isoforms might have different enzymatic properties as the size of HA chains secreted by HAS2 transfectants is larger than that produced following transfection with HAS1 or HAS3. While HAS2 is the major isoform expressed in chondrocytes, the presence of HAS1 is controversial and that of HAS3 could vary with age and in response to cytokine/growth factors [246–248].
DANIEL-HENRI MANICOURT ET AL.
This notwithstanding, HA is synthesized on the inner side of the plasma membrane: chain elongation occurs at the nonreducing end of the growing HA chain [249], which is extruded into the extracellular space [245, 250]. This unusual mode of synthesis allows unconstrained polymer growth and explains why HA molecules can reach exceptionally large sizes. Half of the HA molecules synthesized by cultured explants of normal articular cartilage diffuses rapidly into the culture medium [251]. The newly synthesized HA molecules remaining within the matrix are quite large, but with time they are gradually depolymerized to give a molecular size distribution similar to that observed for endogenous HA [251, 252]. This depolymerization could result from the action of either a recently discovered hyaluronidase active at neutral pH [253] and/or oxygenderived free radicals [254, 255]. In contrast to their native HA molecules, HA degradation fragments induce the expression of inflammatory genes in different cell lines [256]. By binding to the cell surface receptor CD44, the HA molecules (and their bound aggrecan molecules) contribute to the formation of the pericellular matrix and maintenance of matrix homeostasis [257]. Indeed, reduction of HAS2 expression by antisense inhibition reduces the size of the cell-associated matrix and the capacity of the chondrocytes to retain PGs [246], whereas disruption of chondrocyte CD44 receptor and HA interaction by either HA oligosaccharides and/or CD44 antisense oligonucleotides induces a cascade of events resulting in the activation of both catabolic and anabolic gene products [258]. It is worth noting that osteogenic protein-1, also known as bone morphogenic protein-7, inhibits the HA hexosaccharide-induced depletion of HA and PGs within cartilage matrix, thus providing further support for the emerging paradigm that cell–matrix interactions modulate the cellular response to morphogens or growth factors [259]. On the other hand, chondrocytes also use the CD44 receptor to internalize some of the HA molecules, which are then degraded to small oligosaccharides within a low pH lysosomal compartment [257]. Factors that up-regulate chondrocyte catabolism, such as interleukin-1 and the 29-kDa fibronectin fragment, also cause an increased expression of CD44 at the cell surface. As cartilage is bathed by synovial fluid, many investigators have studied the effects of exogenous HA upon matrix metabolism. Addition of high-molecular-weight HA to the culture medium of cartilage explants blocks PG release caused by catabolic cytokines and fibronectin fragments [260–263]. Whether high-molecular-weight HA molecules can reduce in vivo the extent of cartilage matrix degradation in OA and other arthritides remains a matter for debate [264, 265].
Chapter 25 Products of Cartilage Metabolism
VII. OTHER PRODUCTS OF CHONDROCYTE METABOLISM The 35-kDa glycoprotein termed tumor necrosis factor stimulated gene-6 (TSG-6) is not found in normal joints but is markedly expressed by the synoviocytes and chondrocytes in both OA and RA joints [266]. Because, on the one hand, the link module present in its N-terminal domain binds to both HA and the G1 domain of aggrecan, and, on the other hand, the CUB module present in its C-terminus can bind to collagen-like structures, members of the TGF-β superfamily and GAGs as well, TSG-6 might cross-link components of the ECM [267–269]. On the other hand, TGS-6 might protect cartilage from extensive cartilage degradation, even in the presence of acute inflammation: by forming a covalent complex with the inter-αinhibitor (IαI), TSG-6 enhances 100-fold the inhibitory activity of IαI against plasmin which then becomes unable to activate latent MMPs, including MT-MMPs, which are apparently required for the activation of aggrecanase activity [270]. In response to partial oxygen pressure variations, mechanical stress and inflammatory mediators, chondrocytes produce abnormal levels of reactive oxygen species (ROS): nitric oxide (NO) and superoxide anion generate derivative radicals, including peroxynitrite and hydrogen peroxide (as reviewed in reference [271]). Chondrocytes constitutively express the nitric oxide synthase-3 (NOS-3), a low-output enzyme, but when stimulated by either bacterial lipopolysaccharides (LPS), and/or pro-inflammatory cytokines such as IL-1 and TNF-α, the cartilage cells synthesize an inducible isoform of nitric oxide synthase (iNOS, NOS-2) [271, 272], an induction that can be markedly inhibited by transforming growth factor-β, antiinflammatory cytokines such as IL-4, IL-10, and IL-13 [271, 273], as well as by nonsteroidal anti-inflammatory drugs [274]. Once induced, this iNOS generates large quantities of NO which sustains the nuclear translocation of NF-κB and thus keeps NF-κB-dependent transcription persistently switched on [275]. It has been estimated that chondrocyte-derived NO contributes significantly to the total NO produced in the joint cavity in inflammatory arthritides [276–280]. On the other hand, NO inhibits the enzyme complex NADPH-oxidase that produces superoxide anion radicals [281]. As chondrocytes also synthesize myeloperoxidase whose mRNA levels are enhanced in OA, the cartilage cells are likely to produce hypochlorous acid [282]. There is increasing evidence that cellular responses to ROS generation are dependent on the cellular redox status that results from a subtle equilibrium between ROS production and the intracellular antioxidant level. When they are produced at low levels in articular chondrocytes,
437 ROS are likely to contribute to the maintenance of cartilage homeostasis. In addition to the activation of different members of signaling cascades involved in cell growth and differentiation, ROS regulate the DNA-binding activity of transcription factors, and hence modulate gene expression [271]. On the other hand, when they are produced in greater amounts, ROS become deleterious for joint tissues: NO up-regulates the expression of genes coding for MMPs [283], activates latent MMPs [284], and inhibits the synthesis of both PGs and type II collagen [285, 286]. Adding strength to this contention are the results of in vivo studies pointing out that a specific inhibitor of iNOS reduces the severity of cartilage lesions in canine experimental OA [287], whereas the addition of anti-oxidative vitamins to the diet of STR/1R mice diminishes the development of mechanically induced OA [288]. Further studies conducted in iNOS knockout mice strongly suggest that, by inhibiting insulin-like growth factor-(IGF)-1 receptor autophosphorylation, NO might contribute to the resistance of arthritic chondrocytes to IGF-1 stimulation [289]. On the other hand, NO by itself cannot initiate chondrocyte apoptosis, the programmed cell death requiring the concomitant production of superoxide anion [290]. With aging and with the development of OA, 3nitrotyrosine, a product of the reaction of peroxynitrite with tyrosine residues in nearby proteins [291], accumulates within cartilage matrix and hereby provides further indirect evidence of the production of ROS in the articular tissue [292]. The potential mechanism of matrix degradation of ROS cannot be discounted. Even at low concentrations, ROS inhibit glycolysis and induce a rapid depletion in ATP; the depletion in ATP is further compounded by a direct nonenzymatic hydrolysis of ATP leading to a rapid increase in inorganic pyrophosphate (PPi) levels [293]. This effect of ROS on ATP depletion is reversible as long as the exposure of chondrocytes to ROS does not last more than 1–2 hours. Long-term exposure to ROS leads to peroxidation of membrane lipids and cell death. ROS also promote the degradation or denaturation of various cartilaginous matrix components. Thus, ROS depolymerize HA and can degrade collagen, LP and PG either directly or indirectly by activating latent proteinases and/or inactivating inhibitors of proteinase. Articular chondrocytes have the unique ability to constitutively elaborate large amounts of extracellular inorganic pyrophosphate (ePPi) whose levels play an important role in the formation of both calcium pyrophosphate dehydrate (CPPD) crystals and basic calcium phosphate (BCP) crystals, the latter being related to hydroxyapatite, carbonate-substituted apatite and octacalcium phosphate [294, 295]. Specifically, excess of ePPi promotes CPPD crystal formation whereas ePPi deficiency leads to hydroxyapatite crystal deposition. PPi itself is the byproduct of
438 many biosynthetic processes and biochemical reactions in the cell, and the direct product of nucleoside triphosphate pyrophosphohydrolases (NTPPPHs) [296]; it is degraded by several inorganic pyrophosphatases, including alkaline phosphatase. Because PPi does not readily diffuse across healthy biomembranes, levels of ePPi are controlled by the activity of chondrocyte ecto-NTPPHs, such as plasma cell membrane glycoprotein-1 (PC-1) and cartilage intermediate-layer protein (CILP), as well as by the activity of ANKH, a multipass transmembrane protein controlling the egress of iPPi and intracellular substrates for NTPPPHs [297]. Numerous soluble factors modulate ePPi production, as does aging. By enhancing the expression of PC-1 and ANK, TGF-β is unique in its capacity to markedly up-regulate extracellular elaboration of PPi by the chondrocytes [294, 298], a response that is further increased by epidermal growth factor but suppressed by IL-1, parathyroid hormone-related peptide and insulin-like growth factor [299]. Further, aged chondrocytes are more responsive to TGF-β than chondrocytes from young subjects [300]. On the other hand, transglutaminases (type II and factor XIIIA) activate latent TGF-β to increase chondrocyte ePPi production [301, 302], and, in turn, NO and pro-inflammatory cytokines induce increased chondrocyte transglutaminase activity [303]. The linkage between calcium-containing crystal deposition and changes in ePPi has been documented in several animal models and human diseases. In mice, the reduced levels of ePPi secondary to the recessive mutation of ANK leads to spontaneous hydroxyapatite crystal deposition in articular cartilage [304, 305] whereas, in humans, the dominant mutations of ANKH are associated with a slower increase in the levels of ePPi and with deposition of CPPD crystals in cartilage matrix [306, 307]. Hypophosphatasia, a deficiency of the tissue-nonspecific form of alkaline phosphatase, that hydrolyzes PPi to inorganic phosphate, also results in accumulation of ePPi and CPPD deposition in cartilage [308]. A member of the thrombospondin family termed cartilage oligomeric matrix protein (COMP), is an enigmatic pentameric protein that despite its high expression in cartilage has not been assigned a biological function (see Chapter 4). COMP binds collagens I/II and the noncollagenous domain of collagen IX, and, accordingly, this abundant cartilage matrix protein might have several roles in cartilage tissue homeostasis, including regulating collagen fibril formation and maintenance of the integrity and properties of the collagen network [309], a contention strengthened by the observation that mutations of the gene encoding COMP cause pseudo-achondroplasia and multiple epiphyseal dysplasia [310]. The rate of synthesis of
DANIEL-HENRI MANICOURT ET AL.
COMP is enhanced in human cartilage with early OA lesions [311]. In patients with RA and OA, serum COMP levels of COMP could reflect the extent of cartilage matrix turnover [312–315]. As observed for other biomarkers of cartilage metabolism [316], the serum levels of COMP are increased after physiological cyclic loading [317]. Several components of the complement system, including C1q, have been reported to be secreted by normal and OA chondrocytes but their potential roles in matrix homeostasis and chondrocyte metabolism are still largely hypothetical [318]. Likewise, nothing is known about the possible function(s) of a glycoprotein termed YLK-40 that appears to be secreted by both chondrocytes and synoviocytes from patients with OA and RA [319]. Although IL-1 and TGF-β both suppress markedly YKL-40 production by articular chondrocytes [319], sustained enhanced levels of the glycoprotein are apparently related to progression of joint destruction in early RA patients [320]. By interacting with high affinity with decorin, the cartilage-specific (V+C)− fibronectin (Fn) isoform not only inhibits PG degradation by binding ADAMTS4 [181], but also might help stabilize the differentiated phenotype of chondrocytes [321]. Fn synthesis is selectively affected by the frequency of intermittent loads applied to cartilage [322]. In both OA and RA, the cartilage matrix exhibits a 10–20-fold increase in content of this ubiquitous protein [323]. Although TGF-β markedly increases Fn synthesis by normal chondrocytes, this growth factor alone is not sufficient to explain such an accumulation in diseased cartilage; therefore, either intact or degraded Fn molecules are likely to originate from the synovial fluid. In contrast to intact molecules, Fn fragments (fn-f) markedly affect the metabolism of chondrocytes. Indeed, at the high concentrations found in synovial fluids from diseased joints, these fragments can bind to the alpha[5]beta[1] integrin receptor subunit of chondrocytes [324] and up-regulate cartilage catabolism: they stimulate the liberation of IL-1, up-regulate iNOS [325], and cause not only the loss of PGs but also induce the release and/or degradation of COMP and chondroadherin [326]. It is most interesting that, at lower concentrations, the same fragments have an anabolic effect, thought to be mediated by IGF-1, upon matrix metabolism [327].
VIII. CONCLUDING STATEMENT The cartilage matrix contains several molecules that are found in much higher concentrations than in other tissues. The attention that aggrecan and collagen type II are currently receiving as potential markers of specific metabolic alterations in cartilage disease will likely extend to
Chapter 25 Products of Cartilage Metabolism
less abundant molecules found to play key roles in the maintenance of matrix homeostasis. As our understanding of the function and metabolism of these molecules improves, so will their potential as markers of the irreversible destabilization of the collagenous meshwork or of other specific alterations in cartilage matrix organization.
Acknowledgments The preparation of the manuscript was supported in part by the FSR of the UCL.
References 1. Atencia, L. J., McDevitt, C. A., Nile, W. B., and Sokoloff, L. (1989). Cartilage content of an immature dog. Connect Tissue Res. 18, 235–242. 2. Poole, A. R. (2005). Cartilage in health and disease. In “Arthritis and Allied Conditions. A Textbook of Rheumatology” (15th ed) (W. J. Koopman and L. W. Moreland, eds), pp. 223–269, Lippincott Williams and Wilkins Philadelphia, PA. 3. Mow, V. C., and Ratcliffe, A. (1997). Structure and function of articular cartilage and meniscus. In “Basic Orthopaedic Biomechanics” (2nd ed) (V. C. Mow and W. C. Hayes, eds), pp. 113–177, Lippincott-Raven Press Publishers, Philadelphia, PA. 4. Mow, V. C., Wang, C. C., and Hung, C. T. (1999). The extracellular matrix, interstitial fluid and ions as a mechanical signal transducer in articular cartilage. Osteoarthritis Cartilage 7, 41–58. 5. Hing, W. A., Sherwin, A. F., and Poole, C. A. (2002). The influence of the pericellular microenvironment on the chondrocyte response to osmotic challenge. Osteoarthritis Cartilage 10, 297–307. 6. Horikawa, O., Nakajima, H., Kikuchi, T., Ichimura, S., Yamada, H., Fujikawa, K., and Toyama, Y. (2004). Distribution of type VI collagen in chondrocyte microenvironment: study of chondrons isolated from human normal and degenerative articular cartilage and cultured chondrocytes. J. Orthop. Sci. 9, 29–36. 7. Roughley, P. J. (2001). Articular cartilage and changes in arthritis: noncollagenous proteins and proteoglycans in the extracellular matrix of cartilage. Arthritis Res. 3, 342–347. 8. Schwartz, N. B., and Domowicz, M. (2002). Chondrodysplasias due to proteoglycan defects. Glycobiology 12, 57R–68R. 9. Grimshaw, M. J., and Mason, R. M. (2001). Modulation of bovine articular chondrocyte gene expression in vitro by oxygen tension. Osteoarthritis Cartilage 9, 357–364. 10. Domm, C., Schunke, M., Christesen, K., and Kurz, B. (2002). Redifferentiation of dedifferentiated bovine articular chondrocytes in alginate culture under low oxygen tension. Osteoarthritis Cartilage 10, 13–22. 11. Semenza, G. L. (2001). Hypoxia-inducible factor 1: oxygen homeostasis and disease pathophysiology. Trends Mol. Med. 7, 345–350. 12. Coimbra, I. B., Jimenez, S. A., Hawkins, D. F., Piera-Velazquez, S., and Stokes, D.G. (2004). Hypoxia inducible factor-1 alpha expression in human normal and osteoarthritic chondrocytes. Osteoarthritis Cartilage 12, 336–345. 13. Mason, R. M., and Sweeney, C. (1994). The relationship between proteoglycan synthesis in Swarm chondrocytes and pathways of cellular energy and UDP-sugar metabolism. Carbohydrate Res. 255, 255–270. 14. Lee, R. B., and Urban, J. P. (2002). Functional replacement of oxygen by other oxidants in articular cartilage. Arthritis Rheum. 46, 3190–3200.
439 15. Lee, R. B., Wilkins, R. J., Razaq, S., and Urban, J. P. (2002). The effect of mechanical stress on cartilage energy metabolism. Biorheology 39, 133–143. 16. Browning, J. A., and Wilkins, R. J. (2004). Mechanisms contributing to intracellular pH homeostasis in an immortalised human chondrocyte cell line. Comp. Biochem. Physiol. A. Mol. Integr. Physiol. 137, 409–418. 17. Urban, J. P. G., Hall, A. C., and Ghel, K. A. (1993). Regulation of matrix synthesis rates by the ionic and osmotic environment of articular chondrocytes. J. Cell. Physiol. 154, 262–270. 18. Hall, A. C., Starks, I., Shoults, C. L., and Rashidbigi, S. (1996). Pathways for K+ transport across the bovine articular chondrocyte membrane and their sensitivity to cell volume. Am. J. Physiol. 270, (5 Pt 1), C1300–C1310. 19. Hopewell, B., and Urban, J. P. (2003). Adaptation of articular chondrocytes to changes in osmolality. Biorheology 40, 73–77. 20. Aydelotte, M. B., and Kuettner, K. E. (1993). Heterogeneity of articular chondrocytes and cartilage matrix. In “Joint Cartilage Degradation. Basic and Clinical Aspects” (J. F. Woessner, Jr. and D. S. Howell, eds), pp. 37–65, M. Dekker Inc., New York. 21. Little, C. B., Ghosh, P., and Bellenger, C. R. (1996). Topographic variation in biglycan and decorin synthesis by articular cartilage in the early stages of osteoarthritis: an experimental study in sheep. J. Orthop. Res. 14, 433–444. 22. Kim, Y. J., Grodzinsky, A. J., and Plaas, A. H. (1996). Compression of cartilage results in differential effects on biosynthetic pathways for aggrecan, link protein, and hyaluronan. Arch. Biochem. Biophys. 328, 331–340. 23. Wong, M., Siegrist, M., and Cao, X. (1999). Cyclic compression of articular cartilage explants is associated with progressive consolidation and altered expression pattern of extracellular matrix proteins. Matrix Biol. 18, 391–399. 24. Jin, M., Frank, E. H., Quinn, T. M., Hunziker, E. B., and Grodzinsky, A. J. (2001). Tissue shear deformation stimulates proteoglycan and protein biosynthesis in bovine cartilage explants. Arch. Biochem. Biophys. 395, 41–48. 25. Sauerland, K., Raiss, R. X., and Steimeyer, J. (2003). Proteoglycan metabolism and viability of articular cartilage explants as modulated by the frequency of intermittent loading. Osteoarthritis Cartilage 11(5), 343–350. 26. Muller, F. J., Setton, L. A., Manicourt, D. H., Mow, V. C., Howell, D. S., and Pita, J. C. (1994). Centrifugal and biochemical comparison of proteoglycan aggregates from articular cartilage in experimental joint disuse and joint instability. J. Orthop. Res. 12, 498–508. 27. Grumbles, R. M., Howell, D. S., Howard G. A., Roos, B. A., Setton, L. A., Mow, V. C., Ratcliffe A, Muller, F. J., and Altman, R. D. (1995). Cartilage metalloproteases in disuse atrophy. J. Rheumatol. Suppl. 43, 146–148. 28. Deschner, J., Hofman, C. R., Piesco, N. P., and Agarwal, S. (2003). Signal transduction by mechanical strain in chondrocytes. Curr. Opin. Clin. Nutr. Metab. Care 6, 289–293. 29. Borrelli, J. Jr, Tinsley, K., Ricci, W. M., Burns M., Karl, I. E., and Hotchkiss R. (2003). Induction of chondrocyte apoptosis following impact load. J. Orthop. Trauma. 17, 635–641. 30. Chen, C. T., Bhargava, M., Lin, P. M., and Torzilli, P. A. (2003). Time, stress, and location dependent chondrocyte death and collagen damage in cyclically loaded articular cartilage. J. Orthop. Res. 21, 888–898. 31. Kuhn, K., D’Lima, D. D., Hashimoto, S., and Lotz, M. (2004). Cell death in cartilage. Osteoarthritis Cartilage 12, 1–16. 32. Kurz, B., Jin, M., Patwari, P., Cheng, D. M., Lark, M. W., and Grodzinsky, A. J. (2001). Biosynthetic response and mechanical properties of articular cartilage after injurious compression. J. Orthop. Res. 19, 1140–1146. 33. Wilkins, R. J., Browning, J. A., and Urban, J. P. (2000). Chondrocyte regulation by mechanical load. Biorheology 37(1-2), 67–74. 34. Lee, R. B., Wilkins, R. J., Razaq, S., and Urban, J. P. (2002). The effect of mechanical stress on cartilage energy metabolism. Biorheology 39, 133–143.
440 35. Mow, V. C., Wang, C. C., and Hung, C. T. (1999). The extracellular matrix, interstitial fluid and ions as a mechanical signal transducer in articular cartilage. Osteoarthritis Cartilage 7, 41–58. 36. Guilak, F., and Mow, V. C. (2000). The mechanical environment of the chondrocyte: a biphasic finite element model of cell-matrix interactions in articular cartilage. J. Biomech. 33, 1663–1673. 37. Sun, D. D., Guo, X. E., Likhitpanichkul, M., Lai, W. M., and Mow, V. C. (2004). The influence of the fixed negative charges on mechanical and electrical behaviors of articular cartilage under unconfined compression. J. Biomech. Eng. 126, 6–16. 38. Vincent, T. L., Hermannsson, M. A., Hansen, U. N., Amis, A. A., and Saklatvala, J. (2004). Basic fibroblast growth factor mediates transduction of mechanical signals when articular cartilage is loaded. Arthritis Rheum. 50, 526–533. 39. Pearson, G., Robinson, F., Beers Gibson, T., Xu, B. E., Karandikar, M., Berman, K., and Cobb, M. H. (2001). Mitogen-activated protein (MAP) kinase pathways: regulation and physiological functions. Endocr. Rev. 22, 153–183. 40. Lee, D. A., Noguchi, T., Frean, S. P., Lees, P., and Bader, D. L. (2000). The influence of mechanical loading on isolated chondrocytes seeded in agarose constructs. Biorheology 37(1-2), 149–161. 41. Hayashi, T., Abe, E., Yamate, T., Taguchi, Y., and Jasin, H. (1997). Nitric oxide production by superficial and deep articular chondrocytes. Arthritis Rheum. 40, 261–269. 42. Hauselmann, H. J., Stefanovic-Racic, M., Michel, B. A., and Evans, C. H. (1998). Differences in nitric oxide production by superficial and deep human articular chondrocytes: implications for proteoglycan turnover in inflammatory joint diseases. J. Immunol. 160, 1444–1448. 43. Poole, A. R. (1992). Immunology of Cartilage. In “Osteoarthritis: Diagnosis and Medical / Surgical Management” (R. Moskowitz, D. S. Howell, V. M. Goldberg and, H. J. Mankin, eds), pp. 155–189, W. B. Saunders Company, Philadelphia, PA. 44. Ryan, M. C., and Sandell, L. J. (1990). Differential expression of a cysteine-rich domain in the amino-terminal propeptide of type II (cartilage) procollagen by alternative splicing of mRNA. J. Biol. Chem. 265, 10334–10339. 45. Aigner, T., Zhu, Y., Chansky, H. H., Matsen, F.A. 3rd., Maloney, W. J., and Sandell, L. J. (1999). Reexpression of type IIA procollagen by adult articular chondrocytes in osteoarthritic cartilage. Arthritis Rheum. 42, 1443–1450. 46. Nah, H. D., Swoboda, B., Birk, D. E., and Kirsch, T. (2001). Type IIA procollagen: expression in developing chicken limb cartilage and human osteoarthritic articular cartilage. Dev. Dyn. 220, 307–322. 47. Aurich, M., Poole, A. R., Reiner, A., Mollenhauer, C., Margulis, A., Kuettner, K. E., and Cole, A. A. (2002). Matrix homeostasis in aging normal human ankle cartilage. Arthritis Rheum. 46, 2903–2910. 48. Fernandes, R. J., Hirohata, S., Engle, J. M., Colige A., Cohn, D. H., Eyre, D. R., and Apte, S. S. (2001). Procollagen II amino propeptide processing by ADAMTS-3. Insights on dermatosparaxis. J. Biol. Chem. 276, 31502–31509. 49. Li, S-W., Sieron, A. L., Fertala, A., Hojima, Y., Arnold, W. V., and Prockop, D. J. (1996). The C-proteinase that processes procollagens to fibrillar collagens is identical to the protein previously identified as bone morphogenic protein-1. Proc. Natl Acad. Sci. USA 93, 5127-5130. 50. Rebuck, N., Croucher, L. J., and Hollander, A. P. (1999). Distribution of two alternatively spliced variants of the type II collagen N-propeptide compared with the C-propeptide in bovine chondrocyte pellet cultures. J. Cell. Biochem. 75, 13–21. 51. Nelson, F., Dahlberg, L., Laverty, S., Reiner, A., Pidoux, I., Ionescu, M., Fraser, G. L., Brooks, E., Tanzer, M., Rosenberg, L. C., Dieppe, P., and Robin Poole, A. (1998). Evidence for altered synthesis of type II collagen in patients with osteoarthritis. J. Clin. Invest. 102, 2115–2125. 52. Garnero, P., Ayral, X., Rousseau, J. C., Christgau, S., Sandell, L. J., Dougados, M., and Delmas, P. D. (2002). Uncoupling of type II collagen
DANIEL-HENRI MANICOURT ET AL.
53.
54.
55.
56.
57.
58.
59. 60.
61.
62.
63. 64.
65.
66. 67.
68.
69.
synthesis and degradation predicts progression of joint damage in patients with knee osteoarthritis. Arthritis Rheum. 46, 2613–2624. Carey, D. E., Alini, M., Ionescu, M., Hyams, J. S., Rowe, J. C., Rosenberg, L. C., and Poole, A. R. (1997). Serum content of the C-propeptide of the cartilage molecule type II collagen in children. Clin. Exp. Rheumatol. 15, 325–328. Shinmei, M., Ito, K., Matsuyama, S., Yoshihara, Y., and Matsuzawa, K. (1993). Joint fluid carboxy-terminal type II procollagen peptide as a marker of cartilage collagen biosynthesis. Osteoarthritis Cartilage 1, 121–128. Lohmander, L. S., Yoshihara, Y., Roos, H., Kobayashi, T., Yamada, H., and Shinmei, M. (1996). Procollagen II C-propeptide in joint fluid: changes in concentration with age, time after knee injury, and osteoarthritis. J. Rheumatol. 23, 1765–1769. Kobayashi, T., Yoshihara, Y., Samura, A., Yamada, H., Shinmei, M., Roos, H., and Lohmander, L. S. (1997). Synovial fluid concentrations of the C-propeptide of type II collagen correlate with body mass index in primary knee osteoarthritis. Ann. Rheum. Dis. 56, 500–503. Fraser, A., Fearon, U., Billinghurst, R. C., Ionescu, M., Reece, R., Barwick, T., Emery, P., Poole, A. R., and Veale, D. J. (2003). Turnover of type II collagen and aggrecan in cartilage matrix at the onset of inflammatory arthritis in humans: relationship to mediators of systemic and local inflammation. Arthritis Rheum. 48, 3085–3095. Mansson, B., Carey, D., Alini, M., Ionescu, M., Rosenberg, L. C., Poole, A. R., Heinegard, D., and Saxne, T. (1995). Cartilage and bone metabolism in rheumatoid arthritis. Differences between rapid and slow progression of disease identified by serum markers of cartilage metabolism. J. Clin. Invest. 95, 1071–1077. Eyre, D. (2002). Collagen of articular cartilage. Arthritis Res. 4, 30–35. Verzijl, N., DeGroot, J., Ben, Z. C., Brau-Benjamin, O., Maroudas, A., Bank, R. A., Mizrahi, J., Schalkwijk, C. G., Thorpe, S. R., Baynes, J. W., Bijlsma, J. W., Lafeber, F. P., and TeKoppele, J. M. (2002). Crosslinking by advanced glycation end products increases the stiffness of the collagen network in human articular cartilage: a possible mechanism through which age is a risk factor for osteoarthritis. Arthritis Rheum. 46, 114–123. DeGroot, J., Verzijl, N., Wenting-van Wijk, M. J., Jacobs, K. M., Van El, B., Van Roermund, P. M., Bank, R. A., Bijlsma, J. W., TeKoppele, J. M., and Lafeber, F. P. (2004). Accumulation of advanced glycation end products as a molecular mechanism for aging as a risk factor in osteoarthritis. Arthritis Rheum. 50, 1207–1215. Billinghurst, R. C., Mwale, F., Hollander, A., Ionescu, M., and Poole, A. R. (2004). Immunoassays for collagens in chondrocyte and cartilage explant cultures. Meth. Mol. Med. 100, 251–274. Barrett, A. J. (1978). The possible role of neutrophil proteinases in damage to articular cartilage. Agents Actions 8, 15–18. Wu, J-J., Lark, M. W., Chun, L. E., and Eyre, D .R. (1991). Sites of stromelysin cleavage in collagen, II, IX, X and, XI. J. Biol. Chem. 266, 5625–5628. Kozaci, L. D., Buttle, D. J., and Hollander, A. P. (1997). Degradation of type II collagen, but not proteoglycan correlates with matrix metalloproteinase activity in cartilage explant cultures. Arthritis Rheum. 40, 164–174. Nagase, H. (1997). Activation mechanisms of matrix metalloproteinases. Biol. Chem. 378, 151–160. van Meurs, J., van Lent, P., Holthuysen, A., Lambrou, D., Bayne, E., Singer, I., and van den Berg W. (1999). Active matrix metalloproteinases are present in cartilage during immune complex-mediated arthritis: a pivotal role for stromelysin-1 in cartilage destruction. J. Immunol. 163, 5633–5639. Poole, A. R., Nelson, F., Dahlberg, L., Tchetina, E., Kobayashi, M., Yasuda, T., Laverty, S., Squires, G., Kojima, T., and Wu, W., and Billinghurst, R. C. (2003). Proteolysis of the collagen fibril in osteoarthritis. Biochem. Soc. Symp. 70, 115–123. Hollander, A. P., Pidoux, I., Reiner, A., Rorabeck, C., Bourne, R., and Poole, A. R. (1995). Damage to type II collagen in aging and
Chapter 25 Products of Cartilage Metabolism
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
osteoarthritis starts at the articular surface, originates around chondrocytes, and extends into the cartilage with progressive degeneration. J. Clin. Invest. 96, 2859–2869. Wu, W., Billinghurst, R. C., Pidoux, I., Antoniou, J., Zukor, D., Tanzer, M., and Poole, A. R. (2002). Sites of collagenase cleavage and denaturation of type II collagen in aging and osteoarthritic articular cartilage and their relationship to the distribution of matrix metalloproteinase 1 and matrix metalloproteinase 13. Arthritis Rheum. 46, 2087–2094. Aurich, M., Poole, A. R., Reiner, A., Mollenhauer, C., Margulis, A., Kuettner, K. E., and Cole, A. A. (2002). Matrix homeostasis in aging normal human ankle cartilage. Arthritis Rheum. 46, 2903–2910. Dodge, G. R., and Poole, A. R. (1989). Immunohistochemical detection and immunochemical analysis of type II collagen degradation in human normal, rheumatoid and osteoarthritic articular cartilages and in explants of bovine articular cartilage cultured with interleukin-1. J. Clin. Invest. 83, 647–661. Billinghurst, R. C., Dahlberg, L., Ionescu, M., Reiner, A., Bourne, R., Rorabeck, C., Mitchell, P., Hambor, J., Diekmann, O., Tschesche, H., Chen, J., Van Wart, H., and Poole, A. R. (1997). Enhanced cleavage of type II collagen by collagenases in osteoarthritic articular cartilage. J. Clin. Invest. 99, 1534–1545. Knauper, V., Lopez-Otin, C., Smith, B., Knight, G., and Murphy, G. (1996). Biochemical characterization of human collagenase-3. J. Biol. Chem. 271, 1544–1550. Dahlberg, L., Billinghurst, R. C., Manner, P., Nelson, F., Webb, G., Ionescu, M., Reiner, A., Tanzer, M., Zukor, D., Chen, J., van Wart, H. E., and Poole, A. R. (2000). Selective enhancement of collagenase-mediated cleavage of resident type II collagen in cultured osteoarthritic cartilage and arrest with a synthetic inhibitor that spares collagenase 1 (matrix metalloproteinase 1). Arthritis Rheum. 43, 673–682. Billinghurst, R. C., Wu, W., Ionescu, M., Reiner, A., Dahlberg, L., Chen, J., van Wart, H., and Poole, A. R. (2000). Comparison of the degradation of type II collagen and proteoglycan in nasal and articular cartilages induced by interleukin-1 and the selective inhibition of type II collagen cleavage by collagenase. Arthritis Rheum. 43, 664–672. Bau, B., Gebhard, P. M., Haag, J., Knorr, T., Bartnik, E., and Aigner, T. (2002). Relative messenger RNA expression profiling of collagenases and aggrecanases in human articular chondrocytes in vivo and in vitro. Arthritis Rheum. 46, 2648–2657. Neuhold, L. A., Killar, L., Zhao, W., Sung, M. L., Warner, L., Kulik, J., Turner, J., Wu, W., Billinghurst, C., Meijers, T., Poole, A. R., Babij, P., and DeGennaro, L. J. (2001). Postnatal expression in hyaline cartilage of constitutively active human collagenase-3 (MMP-13) induces osteoarthritis in mice. J. Clin. Invest. 107, 35–44. Wu, C. W., Tchetina, E. V., Mwale, F., Hasty, K., Pidoux, I., Reiner, A., Chen, J., Van Wart, H. E., and Poole, A. R. (2002). Proteolysis involving matrix metalloproteinase 13 (collagenase-3) is required for chondrocyte differentiation that is associated with matrix mineralization. J. Bone Miner. Res. 17, 639–651. Murphy, G., and Reynolds, J. J. (1993). Extracellular matrix degradation. In “Connective Tissue and Its Heritable Disorders. Molecular Genetic and Medical Aspects” (P. M. Royce and B. Steinmann, eds), pp. 287–316, Wiley-Liss. New York. Kafienah, W., Bromme, D., Buttle, D. J., Croucher, L. J., and Hollander, A. P. (1998). Human cathepsin K cleaves native type I and II collagens at the N-terminal end of the triple helix. Biochem. J. 331 (Pt 3), 727–732. Price, J. S., Wang-Weigand, S., Bohne, R., Kozaci, L. D., and Hollander, A. P. (1999). Retinoic acid-induced type II collagen degradation does not correlate with matrix metalloproteinase activity in cartilage explant cultures. Arthritis Rheum. 42, 137–147. Brown, N. L., Knott, L., Halligan, E., Yarram, S. J., Mansell, J. P., and Sandy, J. R. (2003). Microarray analysis of murine palatogenesis: temporal expression of genes during normal palate development. Dev. Growth. Differ. 45, 153–165.
441 84. Matyas, J. R., Atley, L., Ionescu, M., Eyre, D. R., and Poole, A. R. (2004). Analysis of cartilage biomarkers in the early phases of canine experimental osteoarthritis. Arthritis Rheum. 50, 543–552. 85. Kojima, T., Mwale, F., Yasuda, T., Girard, C., Poole, A. R., and Laverty, S. (2001). Early degradation of type IX and type II collagen with the onset of experimental inflammatory arthritis. Arthritis Rheum. 44, 120–127. 86. Cerejo, R., Poole, A. R., Ionescu, M., Lobanok, T., Song, J., Cahue, S., Dunlop, D., and Sharma, L. (2002). The association between cartilage collagenase activity measured in serum and progression of knee osteoarthritis in patients with and without evidence of generalized disease. Arthritis Rheum. 48, (suppl 9), S144. 87. Henrotin, Y., Deberg, M., Dubuc, J. E., Quettier, E., Christgau, S., and Reginster, J. Y. (2004). Type II collagen peptides for measuring cartilage degradation. Biorheology 41, 543–547. 88. Deberg, M., Labasse, A., Christgau, S., Cloos, P., Bang Henriksen, D., Chapelle, J. P., Zegels, B., Reginster, J. Y., and Henrotin, Y. (2005). New serum biochemical markers (Coll 2-1 and Coll 2-1 NO2) for studying oxidative-related type II collagen network degradation in patients with osteoarthritis and rheumatoid arthritis. Osteoarthritis Cartilage 13, 258–265. 89. Farrell, A. J., Blake, D. R., Palmer, R. M., and Moncada, S. (1992). Increased concentrations of nitrite in synovial fluid and serum samples suggest increased nitric oxide synthesis in rheumatic diseases. Ann. Rheum. Dis. 51, 1219–1222. 90. Hiran, T. S., Moulton, P. J., and Hancock, J.T. (1998). In situ detection of superoxide anions within porcine articular cartilage. Br. J. Biomed. Sci. 55, 199–203. 91. Henrotin, Y. E., Zheng, S. X., Deby, G. P., Labasse, A. H., Crielaard, J. M., and Reginster, J. Y. (1998). Nitric oxide downregulates interleukin 1beta (IL-1beta) stimulated IL-6, IL-8, and prostaglandin E2 production by human chondrocytes. J. Rheumatol. 25, 1595–1601. 92. Hashimoto, S., Ochs, R. L., Komiya, S., and Lotz, M. (1998). Linkage of chondrocyte apoptosis and cartilage degradation in human osteoarthritis. Arthritis Rheum. 41, 1632–1638. 93. Burkhardt, H., Koller, T., Engstrom, A., Nandakumar, K. S., Turnay, J., Kraetsch, H. G., Kalden, J. R., and Holmdahl, R. (2002). Epitopespecific recognition of type II collagen by rheumatoid arthritis antibodies is shared with recognition by antibodies that are arthritogenic in collagen-induced arthritis in the mouse. Arthritis Rheum. 46, 2339–2348. 94. Mwale, F., Billinghurst, C., Wu, W., Alini, M., Webber, C., Reiner, A., Ionescu, M., Poole, J., and Poole, A. R. (2000). Selective assembly and remodelling of collagens II and IX associated with expression of the chondrocyte hypertrophic phenotype. Dev. Dyn. 218, 648–662. 95. Sinigaglia, L., Varenna, M., Binelli, L., Bartucci, F., Arrigoni, M., Ferrara, R., and Abbiati, G. (1995). Urinary and synovial pyridinium crosslink concentrations in patients with rheumatoid arthritis and osteoarthritis. Ann. Rheum. Dis. 54, 144–147. 96. Black, D., Marabani, M., Sturrock, R. D., and Robins, S. P. (1989). Urinary excretion of hydroxypyridinium crosslinks of collagen in patients with rheumatoid arthritis. Ann. Rheum. Dis. 48, 641–644. 97. Seibel, M. J., Duncan, A., and Robins, S. P., (1989). Urinary hydroxypyridinium crosslinks provide indices of cartilage and bone involvement in arthritic diseases. J. Rheumatol. 16, 964–970. 98. Astbury, C., Bird, H. A., McLaren, A. M., and Robins, S. P. (1994). Urinary excretion of pyridinium crosslinks of collagen correlated with joint damage in arthritis. Br. J. Rheumatol. 33, 11–15. 99. Eyre, D. R., Dickson, I. R., and Van Ness, K. P. (1988). Collagen cross-linking in human bone and articular cartilage. Age-related changes in the content of mature hydroxypyridinium residues. Biochem. J. 252, 495–500. 100. MacDonald, A. G., McHenry, P., Robins, S. P., and Reid, D. M. (1994). Relationship of urinary pyridinium crosslinks to disease extent and activity in osteoarthritis. Br. J. Rheumatol. 33, 16–19.
442 101. Gough, A. K. S., Peel, N. F. A., Eastell, R., Holder, R. L., Lilley, J., and Emery, P. (1994). Excretion of pyridinium crosslinks correlates with disease activity and appendicular bone loss in early rheumatoid arthritis. Ann. Rheum. Dis. 53, 14–17. 102. Garnero, P., and Delmas, P. D. (2003). Biomarkers in osteoarthritis. Curr. Opin. Rheumatol. 15, 641–646. 103. Eyre, D. R., Atley, L. M., Shao, P., Vosberg-Smith, K., Ochs, V., and Clemens, J. D. (2000). Biochemical markers of bone and cartilage collagen degradation. In “Chemistry and Biology of Mineralized Tissues” (M. Goldberg, A. Boskey, and C. Robinson, eds), pp. 347–350, American Academy of Orthopedic Surgeons, Rosemoat, IL. 104. Christgau, S., Garnero, P., Fledelius, C., Moniz, C., Ensig, M., Gineyts, E., Rosenquist, C., and Qvist, P. (2001). Collagen type II C-telopeptide fragments as an index of cartilage degradation. Bone 29, 209–215. 105. Atley, L. M., Carlson, K., Pietka, T. A., and Eyre, D. R. (2001). Oncostatin M induces collagen degradation in human articular cartilage. Osteoarthritis Research Society International, Washington DC. 106. Lohmander, L. S., Atley, L. M., Pietka, T. A., and Eyre, D. R. (2003). The release of crosslinked peptides from type II collagen into human synovial fluid is increased soon after joint injury and in osteoarthritis. Arthritis Rheum. 48, 3130–3139. 107. Reijman, M., Hazes, J. M., Bierma-Zeinstra, S. M., Koes, B. W., Christgau, S., Christiansen, C., Uitterlinden, A. G., and Pols, H. A. (2004). A new marker for osteoarthritis: cross-sectional and longitudinal approach. Arthritis Rheum. 50, 2471–2478. 108. Landewe, R., Geusens, P., Boers, M., van der Heijde, D., Lems, W., te Koppele, J., van der Linden, S., and Garnero, P. (2004). Markers for type II collagen breakdown predict the effect of disease-modifying treatment on long-term radiographic progression in patients with rheumatoid arthritis. Arthritis Rheum. 50, 1390–1399. 109. Mok, S. S., Masuda, K., Hauselmann, H. J., Aydelotte, M. B., and Thonar, E. J. (1994). Aggrecan synthesized by mature bovine chondrocytes suspended in alginate. Identification of two distinct metabolic matrix pools. J. Biol. Chem. 269, 33021–33027. 110. Heinegard, D., Lorenzo, P., and Sommarin, Y. (1995). Articular cartilage matrix proteins. In “Osteoarthritic Disorders” (K. E. Kuettner and V. M. Goldberg, eds), pp. 229–237. American Academy of Orthopedic Surgeons Rosemont, LI. 111. Maroudas, A., Bayliss, M. T., Uchitel-Kaushansky, N., Schneiderman, R., and Gilav, E. (1998). Aggrecan turnover in human articular cartilage: use of aspartic acid racemization as a marker of molecular age. Arch. Biochem. Biophys. 350, 61–71. 112. Ritz-Timme, S., and Collins, M. J. (2002). Racemization of aspartic acid in human proteins. Ageing Res. Rev. 1, 43–59. 113. van Kampen, G. P., van de Stadt, R. J., van de Laar, M. A. F. J., and van der Korst, J. K. (1992). Two distinct metabolic pools of proteoglycans in articular cartilage In “Articular Cartilage and Osteoarthritis” (K. E. Kuettner, R. Schleyerbach, J. G. Peyron, V. C. Hascall, eds), pp. 281–289, Raven Press, New York. 114. Bayliss, M. T. (1992). Metabolism of animal and human osteoarthritic cartilage. In “Articular Cartilage and Osteoarthritis” (K. E. Kuettner, R. Schleyerbach, J. G. Peyron, V. C. Hascall, eds), pp. 487–500, Raven Press, New York. 115. Hauselmann, H. J., Masuda, K., Hunziker, E. B., Neidhart, M., Mok, S. S., Michel, B. A., and Thonar, E. J. (1996). Adult human chondrocytes cultured in alginate form a matrix similar to native human articular cartilage. Am. J. Physiol. 271 (3 Pt 1), C742–C752. 116. Ilic, M. Z., Robinson, H. C., and Handley, C. J. (1998). Characterization of aggrecan retained and lost from the extracellular matrix of articular cartilage. Involvement of carboxyl-terminal processing in the catabolism of aggrecan. J. Biol. Chem. 273, 17451–17458. 117. Loulakis, P., Shrikhande, A., Davis, G., and Maniglia, C. A. (1992). N-terminal sequence of proteoglycan fragments isolated from
DANIEL-HENRI MANICOURT ET AL.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
medium of interleukin-1-treated articular-cartilage cultures. Putative site(s) of enzymic cleavage. Biochem. J. 284 (Pt 2), 589–593. Tortorella, M., Pratta, M., Liu, R. Q., Abbaszade, I., Ross, H., Burn, T., and Arner, E. (2000). The thrombospondin motif of aggrecanase-1 (ADAMTS-4) is critical for aggrecan substrate recognition and cleavage. J. Biol. Chem. 275, 25791–25797. Tortorella, M. D., Malfait, A. M., Deccico, C., and Arner, E. (2001). The role of ADAM-TS4 (aggrecanase-1) and ADAM-TS5 (aggrecanase-2) in a model of cartilage degradation. Osteoarthritis Cartilage 9, 539–552. Ratcliffe, A. M., Doherty, M., Maini, R. N., and Hardingham, T. E. (1988). Increased levels of proteoglycan components in the synovial fluids of patients with acute joint disease. Ann. Rheum. Dis. 47, 826–832. Caroll, G. (1989). Measurement of sulfated glycosaminoglycans and proteoglycan fragments in arthritic synovial fluid. Ann. Rheum. Dis. 48, 17–24. Thonar, E. J-M. A., Manicourt, D. H., Williams, J. M., Fukuda, K., Campion, G. V., Sweet, B. M. E., Lenz, M. E., Schnitzer, T. J., and Kuettner, K.E. (1992). Serum keratan sulfate: a measure of cartilage proteoglycan metabolism. In “Articular Cartilage and Osteoarthritis” (K. E. Kuettner, R. Schleyerbach, J. G. Peyron, V. C. Hascall, eds), pp. 429–445, Raven Press, New York. Dahlberg, L., Roos, H., Saxne, T., Heinegard, D., Lark, M. W., Hoerrner, L. A., and Lohmander, L. S. (1994). Cartilage metabolism in the injured and uninjured knee of the same patient. Ann. Rheum. Dis. 53, 823–827. Caterson, B., Flannery, C. R., Hughes, C. E., and Little, C. B. (2000). Mechanisms involved in cartilage proteoglycan catabolism. Matrix Biol. 19, 333–344. Elsaid, K. A., Jay, G. D., and Chichester, C. O. (2003). Detection of collagen type II and proteoglycans in the synovial fluids of patients diagnosed with non-infectious knee joint synovitis indicates early damage to the articular cartilage matrix. Osteoarthritis Cartilage 11, 673–680. Sandy, J. D., Neame, P. J., Boynton, R. E., and Flannery, C. R. (1991). Catabolism of aggrecan in cartilage explants. identification of a major cleavage site within the interglobular domain. J. Biol. Chem. 266, 8683–8685. Lohmander, L. S., Neame, P. J., and Sandy, J. D. (1993). The structure of aggrecan fragments in human synovial fluid. Evidence that aggrecanase mediates cartilage degradation in inflammatory joint disease, joint injury and osteoarthritis. Arthritis Rheum. 36, 1214–1222. Hughes, C. E., Caterson, B., Fosang, A. J., Roughley P. J., and Mort, J. S. (1995). Monoclonal antibodies that specifically recognize neoepitope sequences generated by “aggrecanase” and matrix metalloproteinase cleavage of aggrecan: application to catabolism in situ and in vitro. Biochem. J. 305, 799–804. Lark, M. W., Williams, H., Hoernner, L. A., Weidner, J., Ayala, J. M., Harper, C. F., Christen, A., Olszewski, J., Konteatis, Z., Webber, R., and Mumford, R. A. (1995). Quantification of a matrix metalloproteinasegenerated aggrecan G1 fragment using monospecific anti-peptide serum. Biochem. J. 307, 245–252. Lark, M. W., Gordy, J. T., Weidner, J. R., Ayala, J., Kimura, J. H., Williams, H. R., Mumford, R. A., Flannery, C. R., Carlson, S. S., Iwata, M., and Sandy, J. D. (1995). Cell-mediated catabolism of aggrecan. Evidence that cleavage of the “aggrecanase” site (Glu 373Ala 374) is a primary event in proteolysis of the interglobular domain. J. Biol. Chem. 270, 2550–2556. Fosang, A. J., Last, K., Gardiner, P., Jackson, D. C., and Brown, L. (1995). Development of a cleavage-site-specific monoclonal antibody for detecting metalloproteinase-derived aggrecan fragments: detection of fragments in human synovial fluid. Biochem. J. 310, 337–343.
Chapter 25 Products of Cartilage Metabolism 132. Singer, I. I., Kawka, D. W., Bayne, E. K., Donatelli, S. A., Weidner, J. R., Williams, H. A., Ayala, J. M., Mumford, R. A., Lark, M. W., Glant, T. T., Nabozny, G. H., and David, C. S. (1995). VDIPEN, A metalloproteinasegenerated neoepitope is induced and immunolocalized in articular cartilage during inflammatory arthritis. J. Clin. Invest. 95, 2178–2186. 133. Fosang, A. J., Last, K., and Maciewicz, R. A. (1996). Aggrecan is degraded by matrix metalloproteinases in human arthritis. Evidence that matrix metalloproteinases and aggrecanase activities can be independent. J. Clin. Invest. 98, 2292–2299. 134. Lark, M.W., Bayne, E.K., Flanagan, J., Harper, C. F., Hoerrner, L. A., Hutchinson, N. I., Singer, I. I., Donatelli, S. A., Weidner, J. R., Williams, H. R., Mumford, R. A., and Lohmander, L. S. (1997). Aggrecan degradation in human cartilage. Evidence for both metalloproteinase and aggrecanase activity in normal, osteoarthritic and rheumatoid joints. J. Clin. Invest. 100, 93–106. 135. Roberts, C. R., Roughley, P. J., and Mort, J. S. (1989). Degradation of human proteoglycan aggregate induced by hydrogen peroxide. Protein fragmentation, amino acid modification and hyaluronic acid cleavage. Biochem. J. 259, 805–811. 136. Ilic, M. Z., Handley, C. H., Robinson, H. C., and Mok, M. T. (1992). Mechanism of catabolism of aggrecan by articular cartilage. Arch. Biochem. Biophys. 294, 115–122. 137. Loulakis, P., Shrikhande, A., Davis, G., and Maniglia, C. A. (1992). N-terminal sequence of proteoglycan fragments isolated from medium of interleukin-1-treated articular-cartilage cultures. Putative site(s) of enzymic cleavage. Biochem J. 284 (Pt 2), 589–593. 138. Rosenberg, L., Wolfenstein-Todel, C., Margolis, R., Pal, S., and Strider, W. (1976). Proteoglycans from bovine proximal humeral articular cartilage. Structural basis for the polydispersity of proteoglycan subunits. J. Biol. Chem. 251, 6439–6444. 139. Heinegard, D. (1977). Polydispersity of cartilage proteoglycans. Structural variation with size and buoyant density of the molecules. J. Biol. Chem. 252, 1980–1989. 140. Bayliss, M. T., and Ali, S. Y. (1978). Age-related differences in the composition and structure of human articular cartilage proteoglycans. Biochem. J. 176, 683–693. 141. Morgelin, M., Paulsson, M., Hardingham, T. E., Heinegard, D., and Engel, J. (1988). Cartilage proteoglycans. Assembly with hyaluronate and link protein as studied by electron microscopy. Biochem. J. 253, 175–185. 142. Sandy, J. D., Plaas, A. H., and Koob, T. J. (1995). Pathways of aggrecan processing in joint tissues. Implications for disease mechanism and monitoring. Acta. Orthop. Scand. Suppl. 266, 26–32. 143. Dudhia, J., Davidson, C. M., Wells, T. M., Vynios, D. H., Hardingham, T. E., and Bayliss, M. T. (1996). Age-related changes in the content of the C-terminal region of aggrecan in human articular cartilage. Biochem J. 313 (Pt 3), 933–940. 144. Sandy, J. D., and Verscharen, C. (2001). Analysis of aggrecan in human knee cartilage and synovial fluid indicates that aggrecanase (ADAMTS) activity is responsible for the catabolic turnover and loss of whole aggrecan whereas other protease activity is required for C-terminal processing in vivo. Biochem. J. 358 (Pt 3), 615–626. 145. Patwari, P., Kurz, B., Sandy, J. D., and Grodzinsky, A. J. (2000). Mannosamine inhibits aggrecanase-mediated changes in the physical properties and biochemical composition of articular cartilage. Arch. Biochem. Biophys. 374, 79–85. 146. Flannery, C. R., Lark, M. W., and Sandy, J. D. (1992). Identification of a stromelysin cleavage site within the interglobular domain of human aggrecan. Evidence for proteolysis at this site in vivo in human articular cartilage. J. Biol. Chem. 267(2), 1008–1014. 147. Fosang, A. J., Neame, P. J., Last, K., Hardingham, T. E., Murphy, G., and Hamilton, J. A. (1992). The interglobular domain of cartilage aggrecan is cleaved by Pump, gelatinases and cathepsin B. J. Biol. Chem. 267, 19470–19474.
443 148. Fosang, A. J., Last, K., Knauper, V., Murphy, G., and Neame, P. J. (1996). Degradation of cartilage aggrecan by collagenase-3 (MMP-13). FEBS Lett. 380(1-2), 17–20. 149. Mort, J. S., Magny, M. C., and Lee, E. R. (1998). Cathepsin B: an alternative protease for the generation of an aggrecan ‘metalloproteinase’ cleavage neoepitope. Biochem J. 335 (Pt 3), 491–494. 150. Little, C. B., Flannery, C. R., Hughes, C. E., Mort, J. S., Roughley, P. J., Dent, C., and Caterson, B. (1999). Aggrecanase versus matrix metalloproteinases in the catabolism of the interglobular domain of aggrecan in vitro. Biochem. J. 344 (Pt 1), 61–68. 151. Fosang, A. J., Last, K., Stanton, H., Weeks, D. B., Campbell, I. K., Hardingham, T. E., and Hembry, R. M. (2000). Generation and novel distribution of matrix metalloproteinase-derived aggrecan fragments in porcine cartilage explants. Generation and novel distribution of matrix metalloproteinase-derived aggrecan fragments in porcine cartilage explants. J. Biol. Chem. 275(42), 33027–33037. 152. Hascall, V. C., Sandy, J. D., and Handley, C. J. (1999). Regulation of proteoglycan metabolism in articular cartilage. In “Biology of the Synovial Joint” (C. W. Archer, B. Caterson, M. Benjamin, J. R. Ralphs, eds), pp. 101–120, Harwood Academic Publishers, The Netherlands. 153. Buttner, F. H., Hughes, C. E., Margerie, D., Lichte, A., Tschesche, H., Caterson, B., and Bartnik, E. (1998). Membrane type 1 matrix metalloproteinase (MT1-MMP) cleaves the recombinant aggrecan substrate rAgg1mut at the ‘aggrecanase’ and the MMP sites. Characterization of MT1-MMP catabolic activities on the interglobular domain of aggrecan. Biochem. J. 333 (Pt 1), 159–165. 154. Tortorella, M. D., Burn, T. C., Pratta, M. A., Abbaszade, I., Hollis, J. M., Liu, R., Rosenfeld, S. A., Copeland, R. A., Decicco, C. P., Wynn, R., Rockwell, A., Yang, F., Duke, J. L., Solomon, K., George, H., Bruckner, R., Nagase, H., Itoh, Y., Ellis, D. M., Ross, H., Wiswall, B. H., Murphy, K., Hillman M. C. Jr, Hollis, G. F., Newton, R. C., Magolda, R. L., Trzaskos, J. M., and Arner E. C. (1999). Purification and cloning of aggrecanase-1: a member of the ADAMTS family of proteins. Science 284, 1664–1666. 155. Tortorella, M. D., Pratta, M., Liu, R. Q., Austin, J., Ross, O. H., Abbaszade, I., Burn, T., and Arner, E. (2000). Sites of aggrecan cleavage by recombinant human aggrecanase-1 (ADAMTS-4). J. Biol. Chem. 275, 18566–18573. 156. Sandy, J. D., Flannery, C. R., Neame, P. J., and Lohmander, L. S. (1992). The structure of aggrecan fragments in human synovial fluid. Evidence for the involvement in osteoarthritis of a novel proteinase which cleaves the Glu 373-Ala 374 bond of the interglobular domain. J. Clin. Invest. 89, 1512–1516. 157. Hu, L. T., Eskildsen, M. A., Masgala, C., Steere, A. C., Arner, E. C., Pratta, M. A., Grodzinsky, A. J., Loening, A., and Perides, G. (2001). Host metalloproteinases in Lyme arthritis. Arthritis Rheum. 44, 1401–1410. 158. Lark, M. W., Gordy, J. T., Weidner, J. R., Ayala, J., Kimura, J. H., Williams, H. R., Mumford, R. A., Flannery, C. R., Carlson, S. S., Iwata, M., and Sandy, J. D. (1995). Cell-mediated catabolism of aggrecan. Evidence that cleavage of the « aggrecanase » site (Glu 373-Ala 374) is a primary event in proteolysis of the interglobular domain. J. Biol. Chem. 270, 2550–2556. 159. Arner, E. C., Pratta, M. A., Trzaskos, J. M., Decicco, C. P., and Tortorella, M. D. (1999). Generation and characterization of aggrecanase. A soluble, cartilage-derived aggrecan-degrading activity. J. Biol. Chem. 274, 6594–6601. 160. Sugimoto, K., Takahashi, M., Yamamoto, Y., Shimada, K., and Tanzawa, K. (1999). Identification of aggrecanase activity in medium of cartilage culture. J. Biochem. (Tokyo). 126, 449–455. 161. Pratta, M. A., Tortorella, M. D., and Arner, E. C. (2000). Age-related changes in aggrecan glycosylation affect cleavage by aggrecanase. J. Biol. Chem. 275, 39096–39102. 162. Sandy, J. D., Thompson, V., Doege, K., and Verscharen, C. (2000). The intermediates of aggrecanase-dependent cleavage of aggrecan
444
163.
164.
165.
166.
167.
168.
169.
170.
171.
172.
173.
174.
175.
DANIEL-HENRI MANICOURT ET AL. in rat chondrosarcoma cells treated with interleukin-1. Biochem. J. 351 (Pt 1), 161–166. Bayliss, M. T., Hutton, S., Hayward, J., and Maciewicz, R. A. (2001). Distribution of aggrecanase (ADAMts 4/5) cleavage products in normal and osteoarthritic human articular cartilage: the influence of age, topography and zone of tissue. Osteoarthritis Cartilage 9, 553–560. Chambers, M. G., Cox, L., Chong, L., Suri, N., Cover, P., Bayliss, M. T., and Mason, R. M. (2001). Matrix metalloproteinases and aggrecanases cleave aggrecan in different zones of normal cartilage but colocalize in the development of osteoarthritic lesions in STR/ort mice. Arthritis Rheum. 44, 1455–1465. Lee, E. R., Lamplugh, L., Davoli, M. A., Beauchemin, A., Chan, K., Mort, J. S., and Leblond, C. P. (2001). Enzymes active in the areas undergoing cartilage resorption during the development of the secondary ossification center in the tibiae of rats ages 0-21 days: I. Two groups of proteinases cleave the core protein of aggrecan. Dev Dyn. 222, 52–70. Sztrolovics, R., Alini, M., Roughley, P. J., and Mort, J. S. (1997). Aggrecan degradation in human intervertebral disc and articular cartilage. Biochem. J. 326 ( Pt 1), 235–241. Mercuri, F. A., Maciewicz, R. A., Tart, J., Last, K., and Fosang, A. J. (2000). Mutations in the interglobular domain of aggrecan alter matrix metalloproteinase and aggrecanase cleavage patterns. Evidence that matrix metalloproteinase cleavage interferes with aggrecanase activity. J. Biol. Chem. 275, 33038–33045. Westling, J., Fosang, A. J., Last, K., Thompson, V. P., Tomkinson, K. N., Hebert, T., McDonagh, T., Collins-Racie, L. A., LaVallie, E. R., Morris, E. A., and Sandy, J. D. (2002). ADAMTS4 cleaves at the aggrecanase site (Glu373-Ala374) and secondarily at the matrix metalloproteinase site (Asn341-Phe342) in the aggrecan interglobular domain. J. Biol. Chem. 277, 16059–16066. Epub 2002 Feb 19. Little, C. B., Hughes, C. E., Curtis, C. L., Janusz, M. J., Bohne, R., Wang-Weigand, S., Taiwo, Y. O., Mitchell, P. G., Otterness, I. G., Flannery, C. R., and Caterson, B. (2002). Matrix metalloproteinases are involved in C-terminal and interglobular domain processing of cartilage aggrecan in late stage cartilage degradation. Matrix Biol. 21, 271–288. van Meurs, J. B., van Lent, P. L., Holthuysen, A. E., Singer, I. I., Bayne, E. K., and van den Berg, W. B. (1999). Kinetics of aggrecanaseand metalloproteinase-induced neoepitopes in various stages of cartilage destruction in murine arthritis. Arthritis Rheum. 42, 1128–1139. van Meurs, J., van Lent, P., Stoop, R., Holthuysen, A., Singer, I., Bayne, E., Mudgett, J., Poole, R., Billinghurst, C., van der Kraan, P., Buma, P., and van den Berg, W. (1999). Cleavage of aggrecan at the Asn341-Phe342 site coincides with the initiation of collagen damage in murine antigen-induced arthritis: a pivotal role for stromelysin 1 in matrix metalloproteinase activity. Arthritis Rheum. 42, 2074–2084. van Meurs, J. B., van Lent, P. L., van de Loo, A. A., Holthuysen, A. E., Bayne, E. K., Singer, I. I., and van den Berg, W. B. (1999). Increased vulnerability of postarthritic cartilage to a second arthritic insult: accelerated MMP activity in a flare up of arthritis. Ann. Rheum. Dis. 58, 350–356. Pratta, M. A., Yao, W., Decicco, C., Tortorella, M. D., Liu, R. Q., Copeland, R. A., Magolda, R., Newton, R. C., Trzaskos, J. M., and Arner, E. C. (2003). Aggrecan protects cartilage collagen from proteolytic cleavage. J. Biol. Chem. 278, 45539–45545. Epub 2003 Jul 30. Flannery, C. R., Little, C. B., Hughes, C. E., Caterson, B. (1999). Expression of ADAMTS homologues in articular cartilage. Biochem. Biophys. Res. Commun. 260, 318–322. Koshy, P. J., Lundy, C. J., Rowan, A. D., Porter, S., Edwards, D. R., Hogan, A., Clark, I. M., and Cawston, T. E. (2002). The modulation of matrix metalloproteinase and ADAM gene expression in human chondrocytes by interleukin-1 and oncostatin M: a time-course study using real-time quantitative reverse transcription-polymerase chain reaction. Arthritis Rheum. 46, 961–967.
176. Sylvester, J., Liacini, A., Li, W. Q., and Zafarullah, M. (2004). Interleukin-17 signal transduction pathways implicated in inducing matrix metalloproteinase-3, -13 and aggrecanase-1 genes in articular chondrocytes. Cell Signal. 16, 469–476. 177. Kevorkian, L., Young, D. A., Darrah, C., Donell, S. T., Shepstone, L., Porter, S., Brockbank, S. M., Edwards, D. R., Parker, A. E., and Clark, I. M. (2004). Expression profiling of metalloproteinases and their inhibitors in cartilage. Arthritis Rheum. 50, 131–141. 178. Curtis, C. L., Hughes, C. E., Flannery, C. R., Little, C. B., Harwood, J. L., and Caterson, B. (2000). n-3 fatty acids specifically modulate catabolic factors involved in articular cartilage degradation. J. Biol. Chem. 275, 721–724. 179. Gao, G., Plaas, A., Thompson, V. P., Jin, S., Zuo, F., and Sandy, J. D. (2004). ADAMTS4 (aggrecanase-1) activation on the cell surface involves C-terminal cleavage by glycosylphosphatidyl inositolanchored membrane type 4-matrix metalloproteinase and binding of the activated proteinase to chondroitin sulfate and heparan sulfate on syndecan-1. J. Biol. Chem. 279, 10042–10051. Epub 2003 Dec 29. 180. Gao, G., Westling, J., Thompson, V. P., Howell, T. D., Gottschall, P. E., and Sandy, J. D. (2002). Activation of the proteolytic activity of ADAMTS4 (aggrecanase-1) by C-terminal truncation. J. Biol. Chem. 277, 11034–11041. Epub 2002 Jan 16. 181. Hashimoto, G., Shimoda, M., and Okada, Y. (2004). ADAMTS4 (aggrecanase-1) interaction with the C-terminal domain of fibronectin inhibits proteolysis of aggrecan. J. Biol. Chem. 279, 32483–32491. Epub 2004 May 25. 182. Flannery, C. R., Zeng, W., Corcoran, C., Collins-Racie, L. A., Chockalingam, P. S., Hebert, T., Mackie, S. A., McDonagh, T., Crawford, T. K., Tomkinson, K. N., LaVallie, E. R., and Morris, E. A. (2002). Autocatalytic cleavage of ADAMTS-4 (Aggrecanase-1) reveals multiple glycosaminoglycan-binding sites. J. Biol. Chem. 277, 42775–42780. Epub 2002 Aug 28. 183. Baker, A. H., Edwards, D. R., and Murphy G. (2002). Metalloproteinase inhibitors: biological actions and therapeutic opportunities. J. Cell. Sci. 115(Pt 19), 3719–3727. 184. Kashiwagi, M., Tortorella, M., Nagase, H., and Brew, K. (2001). TIMP-3 is a potent inhibitor of aggrecanase 1 (ADAM-TS4) and aggrecanase 2 (ADAM-TS5). J. Biol. Chem. 276, 12501–12504. Epub 2001 Jan 23. 185. Rodriguez-Manzaneque, J. C., Westling, J., Thai, S. N., Luque, A., Knauper, V., Murphy, G., Sandy, J. D., and Iruela-Arispe M. L. (2002). ADAMTS1 cleaves aggrecan at multiple sites and is differentially inhibited by metalloproteinase inhibitors. Biochem. Biophys. Res. Commun. 293, 501–508. 186. Tortorella, M. D., Arner, E. C., Hills, R., Easton, A., Korte-Sarfaty, J., Fok, K., Wittwer, A. J., Liu, R. Q., and Malfait, A. M. (2004). Alpha2macroglobulin is a novel substrate for ADAMTS-4 and ADAMTS-5 and represents an endogenous inhibitor of these enzymes. J. Biol. Chem. 279, 17554–17561. Epub 2004 Jan 07. 187. Plaas, A. H., Wong-Palms, S., Roughley, P. J., Midura, R. J., and Hascall, V. C. (1997). Chemical and immunological assay of the nonreducing terminal residues of chondroitin sulfate from human aggrecan. J. Biol. Chem. 272, 20603–20610. 188. Bayliss, M. T., Osborne, D., Woodhouse, S., and Davidson, C. (1999). Sulfation of chondroitin sulfate in human articular cartilage. The effect of age, topographical position, and zone of cartilage on tissue composition. J. Biol. Chem. 274, 15892–15900. 189. Plaas, A. H., West, L. A., Wong-Palms, S., and Nelson, F. R. (1998). Glycosaminoglycan sulfation in human osteoarthritis. Disease-related alterations at the non-reducing termini of chondroitin and dermatan sulfate J. Biol. Chem. 273, 12642–12649. 190. van den Berg, W. B., van der Kraan, P. M., Scharstuhl, A., and van Beuningen, H. M. (2001). Growth factors and cartilage repair. Clin. Orthop. (391 Suppl): S244–S250.
Chapter 25 Products of Cartilage Metabolism 191. Martel-Pelletier, J., Alaaeddine, N., and Pelletier, J. P. (1999). Cytokines and their role in the pathophysiology of osteoarthritis. Front Biosci. 4, D694–D703. 192. Sauerland, K., Plaas, A. H., Raiss, R. X., and Steinmeyer, J. (2003). The sulfation pattern of chondroitin sulfate from articular cartilage explants in response to mechanical loading. Biochim. Biophys. Acta. 1638, 241–248. 193. Caterson, B., Hughes, C. E., Johnstone, B., Mort, J. S. (1992). Immunological markers of cartilage proteoglycan metabolism in animal and human osteoarthritis. In “Articular Cartilage and Osteoarthritis” (K. E. Kuettner, R. Schleyerbach, J. G. Peyron, and V. C. Hascall, eds), pp. 415–427, Raven Press, New York. 194. Visco, D. M., Johnstone, B., Hill, M. A., Jolly, G. A., and Caterson, B. (1993). Immunohistochemical analysis of 3-B-(-) and 7-D-4 epitope expression in canine osteoarthritis. Arthritis Rheum. 36, 1718–1725. 195. Slater, R. R., Jr, Bayliss, M. T., Lachiewicz, P. F., Visco, D. M., and Caterson, B. (1995). Monoclonal antibodies that detect biochemical markers of arthritis in humans. Arthritis Rheum. 38, 655–659. 196. Hazell, P. K., Dent, C., Fairclough, J. A., Bayliss, M. T., and Hardingham, T. E. (1995). Changes in glycosaminoglycan epitope levels in knee joint fluid following injury. Arthritis Rheum. 38, 953–959. 197. Bautch, J. C., Clayton, M. K., Chu, Q., and Johnson, K. A. (2000). Synovial fluid chondroitin sulphate epitopes 3B3 and 7D4, and glycosaminoglycan in human knee osteoarthritis after exercise. Ann. Rheum. Dis. 59, 887–891. 198. Poole, A. R. (2003). Biochemical/immunochemical biomarkers of osteoarthritis: utility for prediction of incident or progressive osteoarthritis. Rheum. Dis. Clin. North. Am. 29, 803–818. 199. Poole, A. R., Ionescu, M., Swan, A., and Dieppe, P. A. (1994). Changes in cartilage metabolism in arthritis are reflected by altered serum and synovial fluid levels of the cartilage proteoglycan aggrecan. J. Clin. Invest. 94, 25–33. 200. Carlson, C. S., Loeser, R. F., Johnstone, B., Tulli, H. M., Dobson, D. B., and Caterson, B. (1995). Osteoarthritis in cynomolgus macaques. II. Detection of modulated proteoglycan epitopes in cartilage and synovial fluid. J. Orthop. Res. 13, 399–409. 201. Belcher, C., Yaqub, R., Fawthrop, F., Bayliss, M., and Doherty, M. (1997). Synovial fluid chondroitin and keratan sulphate epitopes, glycosaminoglycans, and hyaluronan in arthritic and normal knees. Ann. Rheum. Dis. 56, 299–307. 202. Fawthrop, F., Yaqub, R., Belcher, C., Bayliss, M., Ledingham, J., and Doherty, M. (1997). Chondroitin and keratan sulphate epitopes, glycosaminoglycans, and hyaluronan in progressive versus nonprogressive osteoarthritis. Ann. Rheum. Dis. 56, 119–122. 203. Thonar, E. J-M. A., Manicourt, D. H., Williams, J. M., Fukuda, K., Campion, G. V., Sweet, B. M. E., Lenz, M. E., Schnitzer, T. J., and Kuettner, K. E. (1992). Serum keratan sulfate: a measure of cartilage proteoglycan metabolism. In “Articular Cartilage and Osteoarthritis” (K. E. Kuettner, R. Schleyerbach, J. G. Peyron, and V. C. Hascall, eds), pp. 429–445, Raven Press, New York. 204. Webber, C., Glant, T. T., Roughley, P. J., and Poole, A. R. (1987). The identification and characterization of two populations of aggregating proteoglycans of high buoyant density isolated from post-natal human articular cartilages of different ages. Biochem. J. 248, 735–740. 205. Scott, J. E. (1992). Oxygen and the connective tissues. T I B S 17, 340–343. 206. Manicourt, D-H., Pita, J. C., Thonar, E. J-M., and Howell, D. S. (1991). Proteoglycans non-dissociatively extracted from different zones of canine normal articular cartilage: variations in the sedimentation profile of aggrecan with degree of physiological stress. Connect. Tissue Res. 26, 231–246. 207. Sweet, M. B., Thonar, E. J., Immelman, A. R., and Solomon, L. (1977). Biochemical changes in progressive osteoarthrosis. Ann. Rheum. Dis. 36, 387–398.
445 208. Brown, G. M., Huckerby, T. N., Bayliss, M. T., and Nieduszynski, I. A. (1998). Human aggrecan keratan sulfate undergoes structural changes during adolescent development. J. Biol. Chem. 273, 26408–26414. 209. Iozzo R. V. (1999). The biology of the small leucine-rich proteoglycans. Functional network of interactive proteins J. Biol. Chem. 274, 18843–18846. 210. Burton-Wurster, N., Liu, W., Matthews, G. L., Lust, G., Roughley, P. J., and Glant, T. T. (2003). Cs-Szabo G. TGF beta 1 and biglycan, decorin, and fibromodulin metabolism in canine cartilage. Osteoarthritis Cartilage 11, 167–176. 211. Knudson, C. B., and Knudson, W. (2001). Cartilage proteoglycans. Semin. Cell Dev. Biol. 12, 69–78. 212. Danielson, K. G., Baribault, H., Holmes, D. F., Graham, H., Kadler, K. E., and Iozzo, R. V. (1997). Targeted disruption of decorin leads to abnormal collagen fibril morphology and skin fragility. J. Cell. Biol. 136, 729–743. 213. Xu, T., Bianco, P., Fisher, L. W., Longenecker, G., Smith, E., Goldstein, S., Bonadio, J., Boskey, A., Heegaard, A. M., Sommer, B., Satomura, K., Dominguez, P., Zhao, C., Kulkarni, A. B., Robey, P. G., and Young, M. F. (1998). Targeted disruption of the biglycan gene leads to an osteoporosis-like phenotype in mice. Nat Genet. 20, 78–82. 214. Kashiwagi, M., Enghild, J. J., Gendron, C., Hughes, C., Caterson, B., Itoh, Y., and Nagase, H. (2004). Altered proteolytic activities of ADAMTS-4 expressed by C-terminal processing. J. Biol. Chem. 279, 10109–10119. Epub 2003 Dec 08. 215. Poole, A. R., Rosenberg, L. C., Reiner, A., Ionescu, M., Bogoch, E., and Roughley, P. J. (1996). Contents and distributions of the proteoglycans decorin and biglycan in normal and osteoarthritic human articular cartilage. J. Orthop. Res. 14, 681–689. 216. Cs-Szabo, G., Roughley, P. J., Plaas, A. H., and Glant, T. T. (1995). Large and small proteoglycans of osteoarthritic and rheumatoid articular cartilage. Arthritis Rheum. 38, 660–668. 217. Cs-Szabo, G., Melching, L. I., Roughley, P. J., and Glant, T. T. (1997). Changes in messenger RNA and protein levels of proteoglycans and link protein in human osteoarthritic cartilage samples. Arthritis Rheum. 40, 1037–1045. 218. Bock, H. C., Michaeli, P., Bode, C., Schultz, W., Kresse, H., Herken, R., and Miosge, N. (2001). The small proteoglycans decorin and biglycan in human articular cartilage of late-stage osteoarthritis. Osteoarthritis Cartilage 9, 654–663. 219. Polgar, A., Falus, A., Koo, E., Ujfalussy, I., Sesztak, M., Szuts, I., Konrad, K., Hodinka, L., Bene, E., Meszaros, G., Ortutay, Z., Farkas, E., Paksy, A., and Buzas, E. I. (2003). Elevated levels of synovial fluid antibodies reactive with the small proteoglycans biglycan and decorin in patients with rheumatoid arthritis or other joint diseases. Rheumatology (Oxford) 42, 522–527. 220. Melching, L. I., and Roughley, P. J. (1999). Modulation of keratan sulfate synthesis on lumican by the action of cytokines on human articular chondrocytes. Matrix Biol. 18, 381–390. 221. Bengtsson, E., Neame, P. J., Heinegard, D., and Sommarin, Y. (1995). The primary structure of a basic leucine-rich repeat protein, PRELP, found in connective tissues. J. Biol. Chem. 270, 25639–25644. 222. Bengtsson, E., Morgelin, M., Sasaki, T., Timpl, R., Heinegard, D., and Aspberg, A. (2002). The leucine-rich repeat protein PRELP binds perlecan and collagens and may function as a basement membrane anchor. J. Biol. Chem. 277, 15061–15068. Epub 2002 Feb 14. 223. SundarRaj, N., Fite, D., Ledbetter, S., Chakravarti, S., and Hassell, J. R. (1995). Perlecan is a component of cartilage matrix and promotes chondrocyte attachment. J. Cell Sci. 108 (Pt 7), 2663–2672. 224. Nicole, S., Davoine, C. S., Topaloglu, H., Cattolico, L., Barral, D., Beighton, P., Hamida, C. B., Hammouda, H., Cruaud, C., White, P. S., Samson, D., Urtizberea, J. A., Lehmann-Horn, F., Weissenbach, J., Hentati, F., and Fontaine, B. (2000). Perlecan, the major proteoglycan
446
225.
226.
227.
228.
229.
230.
231.
232.
233.
234.
235.
236.
237.
238.
239.
240.
DANIEL-HENRI MANICOURT ET AL. of basement membranes, is altered in patients with Schwartz-Jampel syndrome (chondrodystrophic myotonia). Nat Genet. 26, 480–483. Arikawa-Hirasawa, E., Watanabe, H., Takami, H., Hassell, J. R., and Yamada, Y. (1999). Perlecan is essential for cartilage and cephalic development. Nat. Genet. 23, 354–358. Costell, M., Gustafsson, E., Aszodi, A., Morgelin, M., Bloch, W., Hunziker, E., Addicks, K., Timpl, R., and Fassler, R. (1999). Perlecan maintains the integrity of cartilage and some basement membranes. J. Cell Biol. 147, 1109–1122. Handler, M., Yurchenco, P. D., and Iozzo, R. V. (1997). Developmental expression of perlecan during murine embryogenesis. Dev. Dyn. 210, 130–145. French, M. M., Smith, S. E., Akanbi, K., Sanford, T., Hecht, J., FarachCarson, M. C., and Carson, D. D. (1999). Expression of the heparan sulfate proteoglycan, perlecan, during mouse embryogenesis and perlecan chondrogenic activity in vitro. J Cell Biol. 145, 1103–1115. Kou, I., and Ikegawa, S. (2004). SOX9-dependent and -independent transcriptional regulation of human cartilage link protein. J. Biol. Chem. 279, 50942–50948. Epub 2004 Sep 28. Seguin, C. A., Bernier, S. M. (2003). TNFalpha suppresses link protein and type II collagen expression in chondrocytes: Role of MEK1/2 and NF-kappaB signaling pathways. J. Cell. Physiol. 197, 356–369. Fitzgerald, J. B., Jin, M., Dean, D., Wood, D. J., Zheng, M. H., and Grodzinsky, A. J. (2004). Mechanical compression of cartilage explants induces multiple time-dependent gene expression patterns and involves intracellular calcium and cyclic AMP. J. Biol. Chem. 279, 19502–19511. Epub 2004 Feb 11. Bolton, M. C., Dudhia, J., and Bayliss, M. T. (1999). Age-related changes in the synthesis of link protein and aggrecan in human articular cartilage: implications for aggregate stability. Biochem. J. 337 (Pt 1), 77–82. Roughley, P. J., Nguyen, Q., and Mort, J. S. (1992). The role of proteinases and oxygen radicals in the degradation of human articular cartilage. In “Articular Cartilage and Osteoarthritis” (K. E. Kuettner, R. Schleyerbach, J. G. Peyron and, V. C. Hascall, eds), pp. 305–317, Raven Press, New York. Mwale, F., Demers, C. N., Petit, A., Roughley, P., Poole, A. R., Steffen, T., Aebi, M., and Antoniou, J. (2003). A synthetic peptide of link protein stimulates the biosynthesis of collagens, II, IX and proteoglycan by cells of the intervertebral disc. J. Cell. Biochem. 88, 1202–1213. Dean, M. F., and Sansom, P. (2000). Link peptide cartilage growth factor is degraded by membrane proteinases. Biochem J. 349 (Pt 2), 473–479. McKenna, L. A., Liu, H., Sansom, P. A., and Dean, M. F. (1998). An N-terminal peptide from link protein stimulates proteoglycan biosynthesis in human articular cartilage in vitro. Arthritis Rheum. 41, 157–162. Liu, H., McKenna, L. A., and Dean, M. F. (2000). An N-terminal peptide from link protein can stimulate biosynthesis of collagen by human articular cartilage. Arch. Biochem. Biophys. 378, 116–122. Ilic, M. Z., Handley., C. H., Robinson., H. C., and Mok, M. T. (1992). Mechanism of catabolism of aggrecan by articular cartilage. Arch. Biochem. Biophys. 294, 115–122. Goodstone, N. J., Doran, M. C., Hobbs, R. N., Butler, R. C., Dixey, J. J., and Ashton, B. A. (1996). Cellular immunity to cartilage aggrecan core protein in patients with rheumatoid arthritis and non-arthritic controls. Ann. Rheum. Dis. 55, 40–46. Guerassimov, A., Duffy, C., Zhang, Y., Banerjee, S., Leroux, J. Y., Reimann, A., Webber, C., Delaunay, N., Vipparti, V., Ronbeck, L., Cartman, A., Arsenault, L., Rosenberg, L. C., and Poole, A. R. (1997). Immunity to cartilage link protein in patients with juvenile rheumatoid arthritis. J. Rheumatol. 24, 959–964.
241. Guerassimov, A., Zhang, Y., Banerjee, S., Cartman, A., Webber, C., Esdaile, J., Fitzcharles, M. A., and Poole, A. R. (1998). Autoimmunity to cartilage link protein in patients with rheumatoid arthritis and ankylosing spondylitis. J. Rheumatol. 25, 1480–1484. 242. Guerassimov, A., Zhang, Y., Cartman, A., Rosenberg, L. C., Esdaile, J., Fitzcharles, M. A., and Poole, A. R. (1999). Immune responses to cartilage link protein and the G1 domain of proteoglycan aggrecan in patients with osteoarthritis. Arthritis Rheum. 42, 527–533. 243. Ratcliffe, A., Beauvais, P. J., Saed-Nejad, F. (1994). Differential levels of synovial fluid aggrecan aggregate components in experimental osteoarthritis and joint disuse. J. Othop. Res. 12, 464–473. 244. Ratcliffe, A., Billingham, M. E., Saed-Nejad, F., Muir, H., and Hardingham, T. E. (1992). Increased release of matrix components from articular cartilage in experimental canine osteoarthritis. J. Orthop. Res. 10, 350–358. 245. Itano, N., and Kimata, K. (2002). Mammalian hyaluronan synthases. IUBMB Life 54, 195–199. 246. Nishida, Y., Knudson, C. B., Nietfeld, J. J., Margulis, A., and Knudson, W. (1999). Antisense inhibition of hyaluronan synthase-2 in human articular chondrocytes inhibits proteoglycan retention and matrix assembly. J. Biol. Chem. 274, 21893–21899. 247. Hiscock, D. R., Caterson, B., and Flannery, C. R. (2000). Expression of hyaluronan synthases in articular cartilage. Osteoarthritis Cartilage. 8, 120–126. 248. Recklies, A. D., White, C., Melching, L., and Roughley, P. J. (2001). Differential regulation and expression of hyaluronan synthases in human articular chondrocytes, synovial cells and osteosarcoma cells. Biochem. J. 354 (Pt 1), 17–24. 249. Bodevin-Authelet, S., Kusche-Gullberg, M., Pummill, P. E., DeAngelis, P. L., and Lindahl, U. (2005). Biosynthesis of hyaluronan: direction of chain elongation. J. Biol. Chem. 280, 8813–8818. Epub 2004 Dec 28. 250. Itano, N., Sawai, T., Yoshida, M., Lenas, P., Yamada, Y., Imagawa, M., Shinomura, T., Hamaguchi, M., Yoshida, Y., Ohnuki, Y., Miyauchi, S., Spicer, A. P., McDonald, J. A., and Kimata K. (1999). Three isoforms of mammalian hyaluronan synthases have distinct enzymatic properties. J. Biol. Chem. 274, 25085–25092. 251. Ng, C. K., Handley, C. J., Mason, R. M., and Robinson, H. C. (1989). Synthesis of hyaluronate in cultured bovine articular cartilage. Biochem. J. 263, 761–767. 252. Holmes, M. W. A., Bayliss, M. T., and Muir, H. (1988). Hyaluronic acid in human articular cartilage. Age-related changes in content and size. Biochem. J. 250, 435–441. 253. El Hajjaji, H., Cole, A. A., and Manicourt, D. H. (2005). Chondrocytes, synoviocytes and dermal fibroblasts all express PH-20, a hyaluronidase active at neutral pH. Arthritis Res. Ther. 7, R756–R768. 254. Kvam, B. J., Fragonas, E., Degrassi, A., Kvam, C., Matulova, M., Pollesello, P., Zanetti, F., and Vittur, F. (1995). Oxygen-derived free radical (ODFR) action on hyaluronan (HA), on two HA ester derivatives, and on the metabolism of articular chondrocytes. Exp. Cell Res. 218, 79–86. 255. Parsons, B., Al-Assaf, S., Navaratnam, S., and Phillips, G. (2002). Comparison of the reactivity of different reactive oxidative species towards hyaluronan. In “Hyaluronan.” Volume 1 (J. Kennedy, G. Phillips, P. Williams, V. Hascall, eds), pp. 141–150, MFK Group Ltd. Cambridge, England. 256. Noble, P. W. (2002). Hyaluronan and its catabolic products in tissue injury and repair. Matrix Biol. 21, 25–29. 257. Knudson, C. B., and Knudson, W. (2004). Hyaluronan and CD44: modulators of chondrocyte metabolism. Clin. Orthop. (427 Suppl), S152–S162. 258. Knudson, W., Casey, B., Nishida, Y., Eger, W., Kuettner, K. E., and Knudson, C. B. (2000). Hyaluronan oligosaccharides perturb cartilage matrix homeostasis and induce chondrocytic chondrolysis. Arthritis Rheum. 43, 1165–1174.
Chapter 25 Products of Cartilage Metabolism 259. Nishida, Y., Knudson, C. B., and Knudson, W. (2004). Osteogenic Protein-1 inhibits matrix depletion in a hyaluronan hexasaccharideinduced model of osteoarthritis. Osteoarthritis Cartilage 12, 374–382. 260. Fukuda, K., Dan, H., Takayama, M., Kumano, F., Saitoh, M., and Tanaka, S. (1996). Hyaluronic acid increases proteoglycan synthesis in bovine articular cartilage in the presence of interleukin-1. J. Pharmacol. Exp. Ther. 277, 1672–1675. 261. Fukuda, K., Takayama, M., Ueno, M., Oh, M., Asada, S., Kumano, F., and Tanaka, S. (1997). Hyaluronic acid inhibits interleukin-1-induced superoxide anion in bovine chondrocytes. Inflamm. Res. 46, 114–117. 262. Homandberg, G. A., Hui, F., Wen, C., Kuettner, K. E., and Williams, J. M. (1997). Hyaluronic acid suppresses fibronectin fragment mediated cartilage chondrolysis: I. in vitro. Osteoarthritis Cartilage 5, 309–319. 263. Williams, J. M., Zhang, J., Kang, H., Ummadi, V., and Homandberg, G. A. (2003). The effects of hyaluronic acid on fibronectin fragment mediated cartilage chondrolysis in skeletally mature rabbits. Osteoarthritis Cartilage 11, 44–49. 264. Uebelhart, D., and Williams, J. M. (1999). Effects of hyaluronic acid on cartilage degradation. Curr. Opin. Rheumatol. 11, 427–435. 265. Altman, R. D., Akermark, C., Beaulieu, A. D., and Schnitzer, T., Durolane International Study Group. (2004). Efficacy and safety of a single intra-articular injection of non-animal stabilized hyaluronic acid (NASHA) in patients with osteoarthritis of the knee. Osteoarthritis Cartilage 12, 642–649. 266. Bayliss, M. T., Howat, S. L., Dudhia, J., Murphy, J. M., Barry, F. P., Edwards, J. C., and Day, A. J. (2001). Up-regulation and differential expression of the hyaluronan-binding protein TSG-6 in cartilage and synovium in rheumatoid arthritis and osteoarthritis. Osteoarthritis Cartilage 9, 42–48. 267. Bork, P., and Beckmann, G. (1999). The CUB domain. A widespread module in developmentally regulated proteins. J. Mol. Biol. 231, 539–545. 268. Wisniewski, H. G., and Vilcek, J. (1997). TSG-6: an IL-1/TNFinducible protein with anti-inflammatory activity. Cytokine Growth Factor Rev. 8, 143–156. 269. Solis, D., Romero, A., Jimenez, M., Diaz-Maurino, T., and Calvete, J. J. (1998). Binding of mannose-6-phosphate and heparin by boar seminal plasma PSP-II., a member of the spermadhesin protein family. FEBS Lett. 431, 273–278. 270. Glant, T. T., Kamath, R. V., Bardos, T., Gal, I., Szanto, S., Murad, Y. M., Sandy, J. D., Mort, J. S., Roughley, P. J., and Mikecz, K. (2002). Cartilage-specific constitutive expression of TSG-6 protein (product of tumor necrosis factor alpha-stimulated gene 6) provides a chondroprotective, but not antiinflammatory, effect in antigen-induced arthritis. Arthritis Rheum. 46, 2207–2218. 271. Henrotin, Y. E., Bruckner, P., and Pujol, J. P. (2003). The role of reactive oxygen species in homeostasis and degradation of cartilage. Osteoarthritis Cartilage 11, 747–755. 272. Amin, A. R., and Abramson, S. B. (1998). The role of nitric oxide in articular cartilage breakdown in osteoarthritis. Curr. Opin. Rheumatol. 10, 263–268. 273. Lyons, C. R. (1995). The role of nitric oxide in inflammation. Adv. Immunol. 60, 323-371. 274. Aeberhard, E. E., Henderson, S. A., Arabolos, N. S., Griscavage, J. M., Castro, F. E., Barrett, C. T., and Ignarro, L. J. (1995). Nonsteroidal anti-inflammatory drugs inhibit expression of the inducible nitric oxide synthase gene. Biochem. Biophys. Res. Commun. 208, 1053–1059. 275. Clancy, R. M., Gomez, P. F., and Abramson, S. B. (2004). Nitric oxide sustains nuclear factor kappaB activation in cytokine-stimulated chondrocytes. Osteoarthritis Cartilage 12, 552–558. 276. Stefanovic-Racic, M., Stadler, J., and Evans, C. H. (1993). Nitric oxide and arthritis. Arthritis Rheum. 36, 1036–1044.
447 277. Baddour, V. T., and Bradley, J. D. (1999). Clinical assessment and significance of inflammation in knee osteoarthritis. Curr. Rheumatol. Rep. 1, 59–63. 278. Spreng, D., Sigrist, N., Schweighauser, A., Busato, A., and Schawalder, P. (2001). Endogenous nitric oxide production in canine osteoarthritis: Detection in urine, serum, and synovial fluid specimens. Vet. Surg. 30, 191–199. 279. Karan, A., Karan, M. A., Vural, P., Erten, N., Tascioglu, C., Aksoy, C., Canbaz, M., and Oncel, A. (2003). Synovial fluid nitric oxide levels in patients with knee osteoarthritis. Clin. Rheumatol. 22, 397–399. Epub 2003 Oct 18. 280. Kobayashi, K., Matsuzaka, S., Yoshida, Y., Miyauchi, S., Wada, Y., and Moriya, H. (2004). The effects of intraarticularly injected sodium hyaluronate on levels of intact aggrecan and nitric oxide in the joint fluid of patients with knee osteoarthritis. Osteoarthritis Cartilage 12, 536–542. 281. Moulton, P. J., Goldring, M. B., and Hancock, J. T. (1998). NADPH oxidase of chondrocytes contains an isoform of the gp91phox subunit. Biochem. J. 329 (Pt 3), 449–451. 282. Attur, M. G., Dave, M., Akamatsu, M., Katoh, M., and Amin, A. R. (2002). Osteoarthritis or osteoarthrosis: the definition of inflammation becomes a semantic issue in the genomic era of molecular medicine. Osteoarthritis Cartilage 10, 1–4. 283. Sasaki, K., Hattori, T., Fujisawa, T., Takahashi, K., Inoue, H., and Takigawa, M. (1995). Nitric oxide mediates interleukin-1-induced gene expression of matrix metalloproteinases and basic fibroblast growth factor in cultured rabbit articular chondrocytes. J. Biochem. (Tokyo) 123, 431–439. 284. Murrell, G. A., Jang, D., and Williams, R. J. (1995). Nitric oxide activates metalloprotease enzymes in articular cartilage. Biochem. Biophys. Res. Commun. 206, 15–21. 285. Hickery, M. S., and Bayliss, M. T. (1998). Interleukin-1 induced nitric oxide inhibits sulphation of glycosaminoglycan chains in human articular chondrocytes. Biochim. Biophys. Acta. 1425, 282–290. 286. Cao, M., Westerhausen-Larson, A., Niyibizi, C., Kavalkovich, K., Georgescu, H. I., Rizzo, C. F., Hebda, P. A., Stefanovic-Racic, M., and Evans, C. H. (1997). Nitric oxide inhibits the synthesis of type-II collagen without altering Col2A1 mRNA abundance: prolyl hydroxylase as a possible target. Biochem J. 324 (Pt 1), 305–310. 287. Pelletier, J. P., Jovanovic, D. V., Lascau-Coman, V., Fernandes, J. C., Manning, P. T., Connor, J. R., and Currie, M. G. (1997). MartelPelletier J. Selective inhibition of inducible nitric oxide synthase reduces progression of experimental osteoarthritis in vivo: possible link with the reduction in chondrocyte apoptosis and caspase 3 level. Arthritis Rheum. 43, 1290–1299. 288. Kurz, B., Jost, B., and Schunke, M. (2002). Dietary vitamins and selenium diminish the development of mechanically induced osteoarthritis and increase the expression of antioxidative enzymes in the knee joint of STR/1N mice. Osteoarthritis Cartilage 10, 119–126. 289. Studer, R. K., Levicoff, E., Georgescu, H., Miller, L., Jaffurs, D., and Evans, C. H. (2000). Nitric oxide inhibits chondrocyte response to IGF-I: inhibition of IGF-IRbeta tyrosine phosphorylation. Am. J. Physiol. Cell. Physiol. 279, C961–C969. 290. Del Carlo, M. Jr, and Loeser, R. F. (2002). Nitric oxide-mediated chondrocyte cell death requires the generation of additional reactive oxygen species. Arthritis Rheum. 46, 394–403. 291. Reiter, C. D., Teng, R. J., and Beckman, J. S. (2000). Superoxide reacts with nitric oxide to nitrate tyrosine at physiological pH via peroxynitrite. J. Biol. Chem. 275, 32460–32466. 292. Loeser, R. F., Carlson, C. S., Del Carlo, M., and Cole, A. (2002). Detection of nitrotyrosine in aging and osteoarthritic cartilage: Correlation of oxidative damage with the presence of interleukin1beta and with chondrocyte resistance to insulin-like growth factor 1. Arthritis Rheum. 6, 2349–2357.
448 293. Kwam, B. J., Fragonas, E., Degrassi, A., Kwam, C., Matulova, M., Pollesello, P., Zanetti, F., and Vittur, F. (1995). Oxygen-derived free radical (ODFR) action on hyaluronan (HA), on two HA ester derivatives, and on the metabolism of articular chondrocytes. Exp. Cell. Res. 218, 79–86. 294. Johnson, K., Vaingankar, S., Chen, Y., Moffa, A., Goldring, M. B., Sano, K., Jin-Hua, P., Sali, A., Goding, J., and Terkeltaub, R. (1999). Differential mechanisms of inorganic pyrophosphate production by plasma cell membrane glycoprotein-1 and B10 in chondrocytes. Arthritis Rheum. 42, 1986–1997. 295. Ea, H. K., and Liote, F. (2004). Calcium pyrophosphate dihydrate and basic calcium phosphate crystal-induced arthropathies: update on pathogenesis, clinical features, and therapy. Curr. Rheumatol. Rep. 6, 221–227. 296. Howell, D. S., Martel-Pelletier, J., Pelletier, J-P., Morales, S., and Muniz, O. (1984). NTP pyrophosphohydrolase in human chondrocalcinotic and osteoarthritic cartilage II Further studies on histologic and subcellular distribution. Arthritis Rheum. 27, 193–199. 297. Hirose, J., Ryan, L. M., and Masuda, I. (2002). Up-regulated expression of cartilage intermediate-layer protein and ANK in articular hyaline cartilage from patients with calcium pyrophosphate dihydrate crystal deposition disease. Arthritis Rheum. 46, 3218–3229. 298. Sohn, P., Crowley, M., Slattery, E., and Serra, R. (2002). Developmental and TGF-beta-mediated regulation of Ank mRNA expression in cartilage and bone. Osteoarthritis Cartilage 10, 482–490. 299. Terkeltaub, R. A. (2001). Inorganic pyrophosphate generation and disposition in pathophysiology. Am. J. Physiol. Cell. Physiol. 281, C1-C11. 300. Hirose, J., Masuda, I., and Ryan, L. M. (2000). Expression of cartilage intermediate layer protein/nucleotide pyrophosphohydrolase parallels the production of extracellular inorganic pyrophosphate in response to growth factors and with aging. Arthritis Rheum. 43, 2703–2711. 301. Rosenthal, A. K., Masuda, I., Gohr, C. M., Derfus, B. A., and Le, M. (2001). The transglutaminase, factor XIIIA, is present in articular chondrocytes. Osteoarthritis Cartilage 9, 578–581. 302. Rosenthal, A. K., Mosesson, M. W., Gohr, C. M., Masuda, I., Heinkel, D., and Seibenlist, K. R. (2004). Regulation of transglutaminase activity in articular chondrocytes through thrombin receptor-mediated factor XIII synthesis. Thromb. Haemost. 91, 558–568. 303. Johnson, K., Hashimoto, S., Lotz, M., Pritzker, K., and Terkeltaub, R. (2001). Interleukin-1 induces pro-mineralizing activity of cartilage tissue transglutaminase and factor XIIIa. Am. J. Pathol. 159, 149–163. 304. Ho, A. M., Johnson, M. D., and Kingsley, D. M. (2000). Role of the mouse ank gene in control of tissue calcification and arthritis. Science 289 (5477), 265–270. 305. Okawa, A., Nakamura, I., Goto, S., Moriya, H., Nakamura, Y., and Ikegawa, S. (1998). Mutation in Npps in a mouse model of ossification of the posterior longitudinal ligament of the spine. Nat. Genet. 19, 271–273. 306. Pendleton, A., Johnson, M. D., Hughes, A., Gurley, K. A., Ho, A. M., Doherty, M., Dixey, J., Gillet, P., Loeuille, D., McGrath, R., Reginato, A., Shiang, R., Wright, G., Netter, P., Williams, C., and Kingsley, D. M. (2002). Mutations in ANKH cause chondrocalcinosis. Am. J. Hum. Genet. 71, 933–940. Epub 2002 Sep 20. 307. Williams, C. J., Pendleton, A., Bonavita, G., Reginato, A. J., Hughes, A. E., Peariso, S., Doherty, M., McCarty, D. J., and Ryan, L. M. (2003). Mutations in the amino terminus of ANKH in two US families with calcium pyrophosphate dihydrate crystal deposition disease. Arthritis Rheum. 48, 2627–2631.
DANIEL-HENRI MANICOURT ET AL. 308. Wendling, D., Jeannin-Louys, L., Kremer, P., Fellmann, F., Toussirot, E., and Mornet, E. (2001). Adult hypophosphatasia. Current Aspects Joint Bone Spine 68, 120–124. 309. Rosenberg, K., Olsson, H., Morgelin, M., and Heinegard, D. (1998). Cartilage oligomeric matrix protein shows high affinity zinc-dependent interaction with triple helical collagen. J. Biol. Chem. 273, 20397–20403. 310. Mabuchi, A., Manabe, N., Haga, N., Kitoh, H., Ikeda, T., Kawaji, H., Tamai, K., Hamada, J., Nakamura, S., Brunetti-Pierri, N., Kimizuka, M., Takatori, Y., Nakamura, K., Nishimura, G., Ohashi, H., and Ikegawa, S. (2003). Novel types of COMP mutations and genotype-phenotype association in pseudoachondroplasia and multiple epiphyseal dysplasia. Hum. Genet. 112, 84–90. Epub 2002 Oct 29. 311. Lorenzo, P., Bayliss, M. T., and Heinegard, D. (2004). Altered patterns and synthesis of extracellular matrix macromolecules in early osteoarthritis. Matrix Biol. 23, 381–391. 312. Vilim, V., Vytasek, R., Olejarova, M., Machacek, S., Gatterova, J., Prochazka, B., Kraus, V. B., and Pavelka, K. (2001). Serum cartilage oligomeric matrix protein reflects the presence of clinically diagnosed synovitis in patients with knee osteoarthritis. Osteoarthritis Cartilage 9, 612–618. 313. Skoumal, M., Kolarz, G., and Klingler, A. (2003). Serum levels of cartilage oligomeric matrix protein. A predicting factor and a valuable parameter for disease management in rheumatoid arthritis. Scand. J. Rheumatol. 32, 156–161. 314. Crnkic, M., Mansson, B., Larsson, L., Geborek, P., Heinegard, D., and Saxne, T. (2003). Serum cartilage oligomeric matrix protein (COMP) decreases in rheumatoid arthritis patients treated with infliximab or etanercept. Arthritis Res. Ther. 5, R181–R185. Epub 2003 Apr 29. 315. Jordan, J. M. (2004). Cartilage oligomeric matrix protein as a marker of osteoarthritis. Rheumatol. Suppl. 70, 45–49. 316. Manicourt, D. H., Poilvache, P., Nzeusseu, A., van Egeren, A., Devogelaer, J. P., Lenz, M. E., and Thonar, E. J. (1999). Serum levels of hyaluronan, antigenic keratan sulfate, matrix metalloproteinase 3, and tissue inhibitor of metalloproteinases 1 change predictably in rheumatoid arthritis patients who have begun activity after a night of bed rest. Arthritis Rheum. 42, 1861–1869. 317. Mundermann, A., Dyrby, C. O., Andriacchi, T. P., and King, K. B. (2005). Serum concentration of cartilage oligomeric matrix protein (COMP) is sensitive to physiological cyclic loading in healthy adults. Osteoarthritis Cartilage 13, 34–38. 318. Bradley, K., North, J., Saunders, D., Schwaeble, W., Jeziorska, M , Woolley, D. E. (1996). Synthesis of classical pathway complement components by chondrocytes. Immunology 88, 648–656. 319. Johansen, J. S., Olee, T., Price, P. A., Hashimoto, S., Ochs, R. L., and Lotz, M. (2001). Regulation of YKL-40 production by human articular chondrocytes. Arthritis Rheum. 44, 826–837. 320. Johansen, J. S., Kirwan, J. R., Price, P. A., and Sharif, M. (2001). Serum YKL-40 concentrations in patients with early rheumatoid arthritis: relation to joint destruction. Scand. J. Rheumatol. 30, 297–304. 321. Gendelman, R., Burton-Wurster, N. I., MacLeod, J. N., and Lust, G. (2003). The cartilage-specific fibronectin isoform has a high affinity binding site for the small proteoglycan decorin. J. Biol. Chem. 278, 11175–11181. Epub 2002 Dec. 322. Wolf, A., Raiss, R. X., and Steinmeyer, J. (2003). Fibronectin metabolism of cartilage explants in response to the frequency of intermittent loading. J. Orthop. Res. 21, 1081–1089. 323. Lust, G., and Burton-Wurster, N. (1992). Fibronectin in osteoarthritis: comparison of animal and human diseases. In “Articular Cartilage and Osteoarthritis” (K. E. Kuettner, R. Schleyerbach, J. G. Peyron, and V. C. Hascall, eds), pp. 447–500, Raven Press, New York.
Chapter 25 Products of Cartilage Metabolism 324. Homandberg, G. A., Costa, V., and Wen, C. (2002). Fibronectin fragments active in chondrocytic chondrolysis can be chemically crosslinked to the alpha5 integrin receptor subunit. Osteoarthritis Cartilage 10, 938–949. 325. Pichika, R., and Homandberg, G. A. (2004). Fibronectin fragments elevate nitric oxide (NO) and inducible NO synthetase (iNOS) levels in bovine cartilage and iNOS inhibitors block fibronectin fragment mediated damage and promote repair. Inflamm. Res. 53, 405–412. Epub 2004 Aug 10.
449 326. Johnson, A., Smith, R., Saxne, T., Hickery, M., and Heinegard, D. (2004). Fibronectin fragments cause release and degradation of collagen-binding molecules from equine explant cultures. Osteoarthritis Cartilage 12, 149–159. 327. Homandberg, G. A., Hui, F. (1994). High concentrations of fibronectin fragments cause short-term catabolic effects in cartilage tissue while lower concentrations cause continuous anabolic effects. Archives Biochem. Biophys. 311, 213–218.
This Page Intentionally Left Blank
Chapter 26
Fluid Dynamics of the Joint Space and Trafficking of Matrix Products Peter A. Simkin, M. D.
Professor of Medicine, Adjunct Professor of Orthopaedics, University of Washington
I. Introduction II. Interpretation of Marker Data and Strategies for Dealing with Them
III. Conclusion References
I. INTRODUCTION
subchondral bone. Although this assumption clearly deserves more intensive study, particularly in the setting of inflammatory joint disease, it will be accepted for the purpose of this chapter. It means that any product of cartilaginous injury or degradation must leave that tissue by passing into the overlying synovial fluid. Sampling of that fluid and measurements of its marker content may therefore provide insights into the extent of the lesion. Further, serial observations may help evaluate the healing response to injury and the effectiveness of therapeutic interventions. Since synovial fluid is not always accessible and markers are subsequently cleared from the joint into plasma, it has also been thought that simple measurements in blood may have even greater value. Whether in synovial fluid or in serum, measurements of concentration comprise virtually all of the published marker data. How useful are such static findings in this book about Dynamics of Bone and Cartilage Metabolism? All synovial fluid solutes are molecules in transit. Some move mainly from plasma into the joint (i.e. glucose and oxygen), others move mainly from the joint back into
A large and growing literature attests to the broad, continuing interest in cartilage-derived molecules as “markers” of damage and repair across a wide variety of joint diseases [1–5]. Specific markers and the rationale for their use are examined in detail elsewhere in this volume (see Chapters 24 and 25). This chapter will aim to put such data in the context of normal and pathologic joint physiology. In so doing, we will introduce important caveats in the interpretation of marker data and discuss potential strategies for dealing with them.
II. INTERPRETATION OF MARKER DATA AND STRATEGIES FOR DEALING WITH THEM The underlying assumptions are straightforward. Both in normal and in abnormal joints, articular cartilage is thought to be backed by an impermeable barrier of Dynamics of Bone and Cartilage Metabolism
451
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
452 plasma (i.e. lactate, carbon dioxide, and cartilaginous “markers”), and most simply move back and forth in the ongoing equilibrium which characterizes any specific compartment of the body’s extracellular space. When we aspirate an aliquot of synovial fluid, we arrest this process. Measurements in such aspirates characterize a moment in the joint just as a still frame freezes but a moment in a moving picture. To extend the analogy, a single photograph of a busy avenue would reveal measurable numbers of pedestrians and automobiles. No matter how carefully one counted, however, such a snapshot could tell little or nothing about how fast each individual moved, how great was the overall traffic flow, or where everyone was going. If one’s interest was narrower, as in police cars or men in overcoats, the photo would reveal even less. To begin to understand what is really going on, one must incorporate time in the measurements and evaluate the differing rates that characterize each participant in the scene. This limitation poses a fundamental problem for those interested in specific synovial fluid solutes and underscores the importance of interpreting the findings of aspirates within a dynamic context. In considering the dynamics of large synovial solutes, the most relevant data have come not from cartilaginous markers but from studies of serum albumin. Such data are useful primarily for the perspective they provide on synovial lymphatic outflow. Small molecules (roughly those having molecular weights less than 10 000) are thought to diffuse directly into plasma across the fenestrae of synovial blood vessels and are therefore cleared more rapidly from the joint. For larger molecules, however, the sole route of egress is thought to be convective transport through lymphatics. These valved vessels lie within the synovial interstitium. There, they take up interstitial fluid, including all of its dissolved solutes, and cyclic joint usage then provides the pump that sends it on its way back to the bloodstream. That path has classically been thought to end at the thoracic duct, but subsequent evidence suggests that vascular access may also occur through blood vessels perfusing regional lymph nodes [6]. Albumin kinetics have been studied in joints primarily through studies using radiolabeled molecules [7–9]. Briefly, serum albumin is tagged in vitro with radioiodide and a precisely quantified dose is injected into the joint space. Serial gamma counts are then obtained over the joint for a period of hours and their log values are plotted against time. Such studies usually yield a highly linear declining function whose slope can be readily expressed either as a simple rate constant (in min−1) or as a half-life (in minutes)(Fig. 1). At the conclusion of the experiment, the joint space is aspirated and the residual labeled albumin is quantified in a measured volume of synovial fluid.
Peter A. Simkin
Figure 1 Kinetics of albumin removal from a human knee. Radiolabeled albumin (RISA) was injected at time 0 and followed serially by external counting (∆) and in synovial fluid aspirates (䊐). After a brief equilibration period, both curves decline at the same constant rate which may be expressed as a half life or in min−1. The latter rate constant may be multiplied by the volume of distribution to provide a clearance value in ml/min. The more rapid curve labeled 123I illustrates removal of free iodide which was followed concurrently. (Wallis WJ, Simkin PA, Nelp WB, Foster DM. Intraarticular volume and clearance in human synovial effusions. Arthritis Rheum. 28: 441–449, 1985.)
It is then a simple matter to take the initial counts injected, correct that number for the fraction known to have left the joint and the physical decay of the isotope, and calculate the distribution volume of the remainder by mass balance. The product of that volume (in ml) and the removal rate constant (in min−1) provides a clearance rate for albumin (in ml/min) that is considered the best available measure of articular lymphatic flow [8–10]. Proteins up to the size of IgM macroglobulin have been found to leave human knees at the same rate as albumin [11–13]. This means that one may multiply the concentration of such a protein (in, for instance, mg/ml) by the albumin clearance (in ml/min) to calculate a flux rate for the protein of interest (mg/min). Since most joints would presumably be studied under steady-state condition, the flux out should equal the flux in. This influx rate is of great potential interest in converting marker studies from a static to a dynamic basis. The influx into the joint of a solely cartilage-derived molecule should provide a true indicator of “release” and thus of the rate of cartilage catabolism. Similarly, of course, the product of the distribution volume of albumin and the concentration of a comparable protein should provide an appropriate estimate of the intra-articular mass of that protein. Previous investigators interested in marker mass have based their calculations on the synovial fluid volume, either that recovered by aspiration, that measured by indicator dilution, or a combination of the two [14, 15]. It seems most likely
Chapter 26 Fluid Dynamics of the Joint Space and Trafficking of Matrix Products
however, that any marker found in the synovial fluid will also be distributed through the synovial interstitial tissue. Studies with radiolabeled albumin invariably show an articular distribution volume that is larger than that of synovial fluid. Our pilot studies also found that this larger volume did not change between 24 and 72 hours (see Fig. 1). These findings are consistent with the concept of a synovial interstitial volume that is limited by a functionally impermeable capsule and that remains in passive equilibrium with a smaller volume of included synovial fluid. It is the larger, interstitial volume that should be considered in estimating the mass and/or the dynamics of an intra-articular solute (Fig. 2). Figure 2 also illustrates some of the inherent problems when concentrations are measured not in synovial fluid but in the far more accessible plasma. If a given arthritic joint is indeed in the steady state, the rate of cartilage products released into the joint will be equalled by the rate at which those same products are passed on to the plasma pool. There, however, these products will be admixed with and diluted by the same products from all other joints and tissues of the body. This major problem limits the hope that plasma values will be useful in interpreting the degree of involvement in any single joint. The plasma pooling of markers could be seen as an advantage in assessments of therapeutic response in a polyarticular process such as rheumatoid arthritis. Two additional concerns must be addressed, however, in any
Figure 2 In a hypothetical knee (joint A), cartilage matrix turnover reflects the balance between ongoing synthesis and loss. “Lost” molecules enter the synovial fluid and equilibrate throughout the synovial interstitium. Degradation, which can potentially occur at many sites, produces putative “marker” molecules that are cleared from the joint into plasma by the lymphatics. In plasma, the markers mix with like molecules from other joints as well as from nonarticular sources. From plasma, markers may redistribute into other interstitial spaces, may be further processed (primarily in the liver), or may be excreted into the urine. The marker literature consists mainly of concentration measurements in synovial fluid and plasma. (Simkin PA, Bassett JE. Cartilage matrix molecules in serum and synovial fluid. Current Opin. Rheumatol. 7: 346–351, 1995.)
453
such attempt. First is the fact that most hyaline cartilage is nonarticular and many “cartilaginous” molecules are also found in tissues other than cartilage such as the synovial matrix. Input from these sources will obviously dilute the interpretation of plasma findings as indicators of wellbeing in articular cartilage. The second and more important concern is that concentrations in plasma, like those in synovial fluid, are just as dependent on removal as they are on input. Most of this removal is thought to be hepatic but some of it may be renal, some may be redistribution into other tissue compartments, and we presently know very little about how these rates vary between individuals or within the same person over time. This concern is potentially addressable, but plasma concentrations will remain of limited value until we can place them in a meaningful kinetic context. These concerns can be illustrated by some of the findings to date on hyaluronan. This polymeric disaccharide enters the plasma from many different tissues with indirect evidence indicating that the peritoneum may be quantitatively the most important. Local kinetics vary greatly among tissues with the posterior chamber of the eye having perhaps the slowest turnover (t1/2 ≅ 70 h), the anterior chamber the fastest (t1/2 ≅ 1 h) and other tissues, including the synovium, falling somewhere in between. Entry into plasma from joints is accelerated by local inflammation and is also related clearly to the timing and extent of physical activity. Hyaluronan degradation may occur locally in tissues or subsequently in the liver. There, the rate of removal from the blood is size-dependent with larger polymers being taken up more rapidly than shorter ones. The complexity of these factors illustrates some of the formidable difficulties inherent in any attempt to use plasma measurements of hyaluronan to provide meaningful information about what is happening in any specific joint. Unfortunately, the same concern applies to most plasma measurements of markers derived from cartilage. In practice, the articular methodology shows clearly that the clearance of radiolabeled albumin is significantly faster from the knees of patients with rheumatoid arthritis than it is from patients with osteoarthritis. This was true both in the initial series from Seattle and in a large study from London that focused on corticosteroid effects on hyaluronan levels [7, 8, 16]. Mean values were the same in both studies: 0.07 ml/min in RA (and in ankylosing spondylitis) and 0.04 ml/min in OA. In the British study, intra-articular corticosteroids were administered to all subjects and the study was repeated after 2 weeks and again after 2 months. Albumin clearance fell to mean rates of 0.03 and 0.04 ml/min at these intervals in RA but remained steady at 0.04 ml/min in the OA patients.
454
Peter A. Simkin Table I.
Clearance of Labeled Albumin from Arthritic Knees of Human Patients*
Ref
Conditions
n
CRISA
CRISA 2 wks/steroid
7 7 16 16 16
OA RA OA RA AS
11 9 8 8 6
0.039 ± 0.010 0.071 ± 0.008 0.04 ± 0.01 0.07 ± 0.01 0.07 ± 0.02
n.d. n.d. 0.04 ± 0.01 0.03 ± 0.01 0.04 ± 0.01
CRISA 2 months/steroid n.d. n.d. 0.04 ± 0.01 0.04 ± 0.01 0.03 ± 0.01
*CRISA is clearance of radioiodinated serum albumin in ml/min. OA, RA, and AS are osteoarthritis, rheumatoid arthritis, and ankylosing spondylitis. Values are in ml/min (mean SEM).
Patients with ankylosing spondylitis were also studied and their albumin clearance also fell markedly after corticosteroid injections. Thus, the available evidence in humans indicates that albumin clearance (which can be considered as a measure of lymphatic drainage) is accelerated in inflammatory disease and is markedly diminished when the inflammation is effectively treated (Table I). Comparable findings have been shown in experimental animals. Using both intra-articular pyrophosphate crystals and anterior cruciate ligament resection, Myers et al. have found that inflammation markedly accelerates the clearance of albumin from canine knees [17, 18]. This increase (estimated by comparison with studies of the unmanipulated, contralateral knee) was by approximately 20-fold in the intense inflammation induced by large amounts of crystals. It was also highly significant, however, when the inflammatory response was low grade as it was with small amounts of crystals and in a common surgically induced model of osteoarthritis (Table II). In unpublished work, we have used IL-1α to induce synovitis
in caprine carpal and tarsal joints and in each case the clearance of albumin from inflamed joints was substantially greater than when the same joints were studied under control conditions, both before and after the acute experimental inflammation [19]. Thus, in animals as well as in studies of people with arthritis, the articular inflammatory response includes a significant acceleration of protein clearance. These rates of protein clearance are essential determinants of the concentration of every synovial solute. In fact, the rate of removal from the joint is every bit as critical to the intrasynovial concentration as is the rate of entry. This simple fact has obvious applications for the interpretation of marker studies. A number of these investigations have, for instance, compared synovial fluid concentrations found in rheumatoid arthritis with those found in osteoarthritis. Since the clearance from RA knees is approximately twice that found in OA, a finding of comparable SF concentrations would mean that the release rate into the RA knee was not the same but twice
Table II. Albumin Clearance and Volume Determinations in Canine Knees with and without Experimental Joint Disease* Ref
Condition
n
Va (ml)
Vd (ml)
17
500 µg CPPD 0.5 µg CPPD 0.05 µg CPPD contra cruciate contra
3
33.7
2.2 ± 0.6
36.6 ± 11.0
3
6.7
1.4 ± 1.1
6.4 ± 2.1
6
2.7
0.8 ± 0.6
3.8 ± 1.1
13 6 6
1.5 3.8 1.4
0.2 ± 0.2 1.1 ± 0.9 0.1 ± 0.1
2.7 ± 0.1 9.2 ± 3.6 3.7 ± 0.9
17 17 17 18 18
*CRISA is the clearance of radioiodinated serum albumin. CPPD is the mass of calcium phyrophosphate dihydrate crystals injected. cruciate – indicates surgical transection of the anterior cruciate ligament. contra is the contralateral knee in each series of animals. Va is the aspirated volume of synovial fluid. Vd is the distribution volume of labeled albumin. Values are mean ± S.D.
CRISA (ml/min)
Chapter 26 Fluid Dynamics of the Joint Space and Trafficking of Matrix Products
as great. Similarly, an RA marker concentration after a corticosteroid injection that was unchanged from the preinjection value would mean that the release of that marker had doubled. This is true because the articular clearance after injections is halved. These simple illustrations show that it is, and will always be, extremely dangerous to use marker concentrations to draw conclusions about differences between diseases, responses to therapy or any other comparison between joints, unless the clearance kinetics have also been assessed. Note that the critical determinants are the clearance rates (in ml/min). The flux rate (mg/min) is not critical because the flux in will always equal the flux out in the steady state. The intrasynovial volume is also not critical in determining concentration. A constant input of a marker molecule will rapidly lead to a steady concentration when the SF volume is small and will take considerably longer when the volume is large. The end-point steady-state concentration will be the same, however, regardless of the effusion volume (Fig. 3). One of the most disappointing aspects of marker studies to date has been the wide range of SF concentrations observed for virtually every cartilaginous marker in aspirates from virtually every diagnostic category. Values spanning two or more orders of magnitude are more the rule than the exception and this broad range has greatly limited the practical application of the method. Much of this variation, of course, reflects lack of homogeneity in
Figure 3 Effect of volume (ml) and clearance rate (ml per unit of time) on the solute concentration (U/ml) in a well-mixed knee receiving solute at a constant rate of 1 U per unit of time. The knee of patient A has a volume of 10 ml, whereas the knee of patient B has a volume of 100 ml. When both knees are cleared at a constant rate of 2 ml per unit of time, they both attain the same solute concentration of 0.5 U/ml, although the larger volume requires much more time to reach equilibrium. The knee of patient C also has a volume of 100 ml and receives material at the same rate. However, it is cleared at twice the rate in patient B (4 ml per unit of time) and therefore goes to an equilibrium solute concentration half as great. This figure illustrates that steady-state concentrations of any solute reflect the balance of input and output and are independent of volume. (Simkin PA, Bassett JE. Cartilage matrix molecules in serum and synovial fluid. Current Opin. Rheumatol. 7: 346–351, 1995.)
455
any patient group. Among rheumatoid subjects, for instance, both sexes are likely to be represented and there will always be a significant range of patient age, of disease duration, of effusion duration, of therapeutic measures in use, and in the intensity of the inflammatory response. Each of these factors may play a role in the intrasynovial input or clearance of specific marker molecules. Some of the observed variation must reflect real differences in marker release that carry important information about the condition and the prognosis of the studied joint. One may hope that quantification of lymphatic outflow, by allowing conversion of static SF concentrations to dynamic marker release rates, will help to sort out the logical mechanisms that underlie the current diversity of findings. The foregoing discussion has focused on proteins “comparable” to albumin. Many of the intrasynovial molecules of greater current interest do not fit this description. As already mentioned, solutes with molecular weights of 10 000 or less will be cleared not only by lymphatics but will also pass through microvascular pores and be cleared directly into the venous circulation. Since the rate of synovial plasma flow greatly exceeds that of lymphatic drainage, venous clearance could easily accelerate the clearance of IL-1 or TNF-α, for instance, by an order of magnitude compared to that of albumin. As yet, there are no available data that allow us to estimate the passive transport rates for such solutes. Obviously, the specific cellular uptake of such cytokine molecules adds huge additional complications to any consideration of their articular dynamics. At the other end of the size spectrum, markers released from cartilage may not resemble the compact, globular structure of albumin but may instead be much larger and more linear strands of aggrecan (or its catabolities), of type II collagen, or of other cartilage constituents. It is now clear that such macromolecules do not leave the joint as readily as does albumin. This issue was addressed by Page-Thomas et al. who found the mean clearance t1/2 for large proteoglycan molecules (Mr = 2.5 × 106) leaving the rabbit knee to be 12.5 hours, in contrast to 3.9 hours for albumin in the same joints [20]. A more striking difference was reported by Coleman et al., who also used rabbit knees but studied hyaluronan kinetics with a different experimental approach. Their mean t1/2 values were 26.3 h for hyaluronan and 1.23 h for albumin [21]. These findings lead to an inescapable conclusion: marker molecules, which are large in mass and/or configuration, are partially reflected back into the joint space as fluid flows into terminal lymphatics through the interstitial matrix of the synovium. More work remains to be done in defining just what are the critical molecular dimensions, in evaluating
456
Peter A. Simkin
the effects of local inflammation on this sieving function of the synovium and in distinguishing between the fractions that are cleared into lymphatics, reflected back into the joint, or locally degraded. That work will be facilitated by Levick’s simple approach comparing half-lives of markers to that of albumin in order to quantify the extent of reflection [23].
III. CONCLUSION To date, studies of cartilaginous markers have been limited primarily to measurements of concentration in synovial fluid and in plasma. When such data are used to compare different conditions or to follow the same condition serially, the work necessarily assumes that the clearance does not vary between conditions, between individuals, or within the same individual over time. Unfortunately, those assumptions are demonstrably false. To fulfill the marker promise of information regarding “release” from cartilage, the kinetics of these molecules must be examined together with their concentration. Fortunately, relatively simple techniques, largely based on concurrent studies of serum albumin as a reference solute, may go a long way to remedy this situation.
References 1. Garnero, P., and Delmas, P. (2003). Biomarkers in osteoarthritis. Curr. Opin. Rheumatol. 15, 641–646. 2. Poole, A. R. (2003). Biochemical/immunochemical biomarkers of osteoarthritis: utility for prediction of incident or progressive osteoarthritis. Rheum. Dis. Clin. North Am. 29, 803–818. 3. Bruyere, O., Collette J. H., Ethgen, O., Rovati, L. C., Giacovelli, G., Henrotin, Y. E., Seidel, L., and Reginster, J. Y. (2003). Biochemical markers of bone and cartilage remodeling in prediction of longterm progression of knee osteoarthritis. J. Rheumatol. 30, 1043–1050. 4. Lohmander, L.S., and Felson, D. (2004). Can we identify a ‘high risk’ patient profile to determine who will experience rapid progression of osteoarthritis? Osteoarthritis Cartilage 12(Suppl A), S49–S52. 5. Simkin, P. A., and Bassett, J. E. (1995). Cartilage matrix molecules in serum and synovial fluid. Curr. Opin. Rheumatol. 7, 346–351.
6. Rayan, I., Thonar, E. J. M. A., Chen, L. M., Lenz, M. E., and Williams, J. M. (1998). Regional differences in the rise in blood levels of antigenic keratan sulfate and hyaluronan after chymopapain induced knee joint injury. J. Rheumatol. 25, 521–526. 7. Wallis, W. J., Simkin, P. A., Nelp, W. B., and Foster, D. M. (1985). Intraarticular volume and clearance in human synovial effusions. Arthritis Rheum. 28, 441–449. 8. Wallis, W. J., Simkin, P. A., and Nelp, W. B. (1987). Protein traffic in human synovial effusions. Arthritis Rheum. 30, 57–63. 9. Simkin, P. A., and Benedict, R. S. (1990). Iodide and albumin kinetics in normal canine wrists and knees. Arthritis Rheum. 33, 73–79. 10. Levick, J. R., and McDonald, J. N. (1995). Fluid movement across synovium in healthy joints: role of synovial fluid macromolecules. Ann. Rheum. Dis. 54, 417–423. 11. Rodnan, G. P., and MacLachlan, M. J. (1960). The absorption of serum albumin and gammaglobulin from the knee joint of man and rabbit. Arthritis Rheum. 3, 152–157. 12. Sliwinski, A. F., and Zvaifler, N. J. (1969). The removal of aggregated and nonaggregated autologous gamma globulin from rheumatoid joints. Arthritis Rheum. 12, 504–514. 13. Weinberger, A., and Simkin, P. A. (1989). Plasma proteins in synovial fluids of normal human joints. Semin. Arthritis Rheum. 19, 66–76. 14. Geborek, P., Saxne, T., Heinegard, D., and Wollheim, F. A. (1988). Measurement of synovial fluid volume using albumin dilution upon intraarticular saline injection. J. Rheumatol. 15, 91–94. 15. Delecrin, J., Oka, M., Kumar, P., Takahashi, S., Kotoura, Y., Yamamuro, T., and Daculsi, G. (1992). Measurement of synovial fluid volume: a new dilution method adapted to fluid permeation from the synovial cavity. J. Rheumatol. 19, 1746–1752. 16. Pitsillides, A. A., Will, R. K., Bayliss, M. T., and Edwards, J. C. W. (1994). Circulating and synovial fluid hyaluronan levels: effects of intraarticular corticosteroid on the concentration and the rate of turnover. Arthritis Rheum. 37, 1030–1038. 17. Myers, S. L., Brandt, K. D., and Eilam, O. (1995). Even low-grade synovitis significantly accelerates the clearance of protein from the canine knee. Implications for measurement of synovial fluid “markers” of osteoarthritis. Arthritis Rheum. 38, 1085–1091. 18. Myers, S. L., O’Connor, B. L., and Brandt, K. D. (1996). Accelerated clearance of albumin from the osteoarthritic knee: implications for interpretation of concentrations of “cartilage markers” in synovial fluid. J. Rheumatol. 23, 1744–1748. 19. Simkin, P. A., and Bassett, J. E. (unpublished). 20. Page-Thomas, D. P., Bard, D., King, B., and Dingle, J. T. (1987). Clearance of proteoglycan from joint cavities. Ann. Rheum. Dis. 46, 934–937. 21. Coleman, P. J., Scott, D., Ray, J., Mason, R. M., and Levick, J. R. (1997). Hyaluronan secretion into the synovial cavity of rabbit knees and comparison with albumin turnover. J. Physiol. 503(Pt 3), 645–656. 22. Levick, J. R. (1998). A method for estimating macromolecular reflection by human synovium, using measurements of intra-articular half lives. Ann. Rheum. Dis. 57, 339–344.
Chapter 27
Transgenic Models of Bone Disease Barbara E. Kream John R. Harrison
Departments of Medicine and Genetics and Developmental Biology, University of Connecticut Health Center, Farmington, CT 06030 Division of Orthodontics, University of Connecticut Health Center, Farmington, CT 06030
I. Introduction II. Generation of Mouse Models III. Transgenic Models in Bone Biology
IV. Perspectives and Future Directions References
I. INTRODUCTION
II. GENERATION OF MOUSE MODELS
The ability to manipulate the mouse genome has proven invaluable for the development of models to understand gene function in vivo [1]. Several excellent and comprehensive reviews have been published on the development and use of transgenic models of bone and mineral metabolism and diseases [2–5]. The intent of this chapter is to focus on recent work in this area, including a discussion of methods currently used to manipulate gene expression in the mouse. We will highlight selected examples of how transgenic models have provided insights into the roles of transcription factors and hormone signaling pathways in normal bone biology and the pathophysiology of bone diseases. Finally, new technological developments are discussed that will influence the future of functional genomics in the mouse.
A. Transgenesis
Dynamics of Bone and Cartilage Metabolism
The overexpression of genes in transgenic mice has become a widely used technique for understanding the role of genes in vivo. Transgenesis in the mouse provides a means of over-expressing genes of interest to generate gain- and loss-of-function models. Transgenesis has been traditionally accomplished by the microinjection of DNA constructs into the single-cell fertilized embryos [6]. Transgenes typically contain a fragment of a gene promoter fragment positioned upstream of a gene of interest, which can be a reporter, a functional gene, or a dominant negative molecule. The transgene can be epitope-tagged to distinguish it from its endogenous counterpart. Alternatively, the availability of species-specific antibodies to functionally 457
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
458 equivalent transgene products derived from a different species can allow their detection in mouse tissues and blood. Transgenes integrate randomly into the host genome, in one or a few different sites, usually as concatamers in which multiple copies are all oriented in the same direction. Transgene expression is typically dependent on the integration site but not on the copy number unless sequences are present in the transgenic construct to insulate it from the influence of its surrounding chromosomal environment. Both ubiquitous and tissue-specific promoters have been used for targeting transgenes to osteoblasts and osteoclasts. Transgenes driven by ubiquitous promoters may have global effects in the mouse model. Therefore, discernible bone phenotypes may be caused directly by transgene expression in bone cells or indirectly by effects on other tissues. Even with relatively tissue-specific promoters, low-level expression in other cell types may be sufficient to perturb normal function of various organs. Some transgenes, particularly those for secreted proteins, may have non-cell autonomous effects on neighboring nontargeted cells. The overexpression of an endogenous gene provides a gain-of-function model in which there is an exaggeration of the normal activity of a gene product. Expression of a dominant negative molecule, on the other hand, can be used to generate a loss-of-function model by disrupting the role of the normal gene product. The disadvantages of overexpression of genes by transgenesis include generating levels of a gene product that far exceeds its physiological level. This can cause phenotypes that are unexpected and/or do not represent normal physiological function. Moreover, transgenes may be misexpressed both temporally and spatially. Because the promoters used to target transgenes usually contain only a small fragment of the regulatory sequences in the gene, the transgene can be expressed in cells that normally do not express the endogenous gene. Ideally, to build a bone cell-specific transgenic model, expression of the promoter should closely mimic expression of the endogenous gene. Moreover, the transgene should be specifically expressed in the targeted cell and at a level at least as high as physiological. The promoters used to target transgene expression to bone cells have typically been those of matrix genes, which are selectively or specifically expressed in bone [5]. Type I collagen promoters have been used frequently to drive transgenes in cells of the osteoblast lineage. The endogenous Col1a1 gene is highly expressed in cells of the osteoblast lineage and serves as an early marker of osteoblast differentiation. The isolation and characterization of the rat [7] and mouse [8] Col1a1 promoters has provided the tools to drive transgenes in osteoblasts. Promoter fragments that have been commonly used include a rat 3.6-kb fragment of the rat Col1a1 gene and a 2.3-kb fragment of the rat or mouse
BARBARA E. KREAM AND JOHN R. HARRISON
Col1a1 genes [9, 10]. The rat 3.6-kb Col1a1 promoter, while highly expressed in cells of the osteoblast lineage, is not specific for bone since it is activated in a number of other tissues such as tendon, skin, and lung [11]. Studies of the spatial and temporal pattern of Col1a1 promoter– reporter constructs have provided crucial information for developing transgenic models in which genes are targeted to bone cells. These studies showed that a 3.6-kb Col1a1 promoter fragment (Col3.6) is highly expressed in osteoblasts. Deletion of the promoter to 2.3-kb (Col2.3) does not affect promoter activity in vivo, although there is a reduction in activity in vitro [11]. Recently, the green fluorescent protein (GFP) reporter has been cloned downstream of these promoter fragments to determine the spatial activation of the Col3.6 and Col2.3 promoters in vivo [9]. The Col3.6 promoter is broadly expressed in cells of the osteoblast lineage, while the Col2.3 promoter is expressed in mature osteoblasts both in vivo and in primary calvarial and bone marrow stromal cells. Both promoters drive strong transgene expression in osteoblasts and odontoblasts [12]. The Col3.6 promoter is expressed in tendon, lung, and aorta, while the Col2.3 promoter has a more restricted pattern of expression in other tissues. The advantages of these Col1a1 promoter fragments include strong expression in osteoblasts and the ability to distinguish subpopulations of osteoblast lineage cells. Because of the fidelity of its bone-specific expression, the osteocalcin promoter provides an excellent tool for targeting genes to mature osteoblasts [13]. The proximal osteocalcin promoter contains sequences that confer tissuespecific and developmental stage-specific expression in mature osteoblasts [14, 15]. The expression of osteocalcin in skeletal tissues is highly restricted to differentiated osteoblasts and is not seen in preosteoblasts or early osteoblast progenitors. Osteocalcin expression is activated late in development in the perinatal period. However, the strength and hormonal regulation of osteocalcin promoter fragments can differ due to species differences [16] and can be affected by age and sex of the animals [17]. Lowlevel transgene expression may be an advantage for some studies in which it is critical to limit the degree of overexpression. However, for other systems such as the Cre-loxP model for generating conditional knockouts, Cre expression must be driven above a threshold level that allows the rearrangement of a floxed gene [18]. Recently, the dental matrix protein-1 (DMP-1) promoter has been fused to GFP and introduced into transgenic mice [19]. The DMP-1 promoter is strongly expressed in late-stage osteoblasts and osteocytes and will provide a tool for targeting genes to these cells. There have been fewer transgenic models in which genes have been targeted to osteoclasts; this may be due
459
Chapter 27 Transgenic Models of Bone Disease
to the relative paucity of cloned promoter fragments that specifically target osteoclasts and their progenitors. The tartrate-resistant acid phosphatase and cathepsin K promoters are expressed in osteoclasts and other hematopoietic cells and have been used to drive Cre recombinase in osteoclasts [20]. The CD11b promoter has also been used to target Cre recombinase to cells of the hematopoietic myeloid-osteoclast lineage [21]. The CD11b promoter is expressed in mature osteoclasts arising from bone marrow and spleen progenitors [21]. Both models may be useful tools for generating conditional knockout models in cells of the osteoclast lineage.
B. Gene Knockout Models Genes can be inactivated by homologous recombination of targeting vectors in embryonic stem cells [1]. A critical region of the gene is replaced or disrupted by a cassette containing both positive and negative selection markers for the selection of homologous and nonhomologous recombinants, respectively. At the advent of this technology, the positive selection marker (usually the neomycin resistance gene) was often left in the targeted allele. However, since the neo gene can sometimes interfere with the function of nearby genes [22], it is now customary to remove it by flanking it with sites that allow it to be removed by bacterial recombinases. The advantages of a global gene knockout are that the function of the gene can be demonstrated in the target cells of interest. However, major disadvantages include the early embryonic lethality that often accompany global gene knockouts, compensation by another gene product during development and alterations of other organ systems that may indirectly affect bone. Considering these issues, the ability to ablate genes both temporally and spatially has been an important tool for understanding tissue-specific gene function [22]. Conditional gene knockouts can be accomplished using Cre-loxP technology [18]. Two different mouse strains are required: a transgenic line expressing Cre recombinase in the tissue of interest and a line in which the gene of interest has been engineered with lox P sites flanking a critical region. The breeding of the two mouse lines will lead to progeny having gene rearrangement in the cells expressing Cre. Several criteria should be met when developing a conditional knockout. Cre expression should be above the minimum threshold to cause gene rearrangement in the tissue of interest. The efficiency of Cre-mediated gene rearrangement should be such that gene expression is knocked down enough to cause a phenotype. To ensure sufficient gene knockdown in a conditional model, mice
can be developed that contain one floxed allele and one null allele for a gene of interest. Thus, prior to Cre-mediated gene rearrangement, expression of the gene of interest should be decreased by 50%. The temporal and spatial pattern of gene rearrangement can be determined using several Cre reporter strains, such as ROSA26 [23], Z/EG [24], and Z/AP [25]. These reporter strains provide historical marking of Cre expression and are useful tools for characterizing a newly developed Cre transgenic line. Disadvantages with the currently available Cre transgenic lines are the early and widespread activation of Cre recombinase in the embryo and expression of Cre that is so low or sporadic that gene rearrangement is inefficient. To circumvent detrimental effects during embryonic development and to precisely time gene disruption, inducible Cre transgenic systems can be generated [22]. A popular model is the tamoxifen-inducible system in which Cre recombinase is fused to a mutated ligand-binding domain of the ER and then cloned downstream of a targeting promoter [26, 27]. The Cre fusion protein, which is initially sequestered in the cytoplasm, becomes activated and translocates to the nucleus when 4-hydroxytamoxifen is given the animal (or to cells if an ex vivo experiment is performed). An inducible model has recently developed in which the murine 2.3-kb Col1a1 promoter has been used to drive a tamoxifen-inducible Cre transgene for temporally regulated gene ablation in osteoblasts and odontoblasts [28]. The binary tetracycline-dependent systems have enjoyed widespread use [29]. With the Tet-Off system, removal of tetracycline induces Cre expression. By contrast, with the Tet-On system, addition of tetracycline induces Cre expression. There are several limitations of inducible systems. It is important to test several transgenic lines to find one that confers high and inducible Cre expression. It is critical to determine whether there is leaky expression of Cre in the uninduced state. Finally, it should be determined whether the inducer itself has any unexpected or adverse effects on bone (this can be tested in wild-type mice).
C. Phenotypic Analysis A careful phenotypic analysis is required to fully understand and interpret results in a transgenic mouse model. Because transgenes can integrate randomly into the genome, it is important to examine multiple lines of transgenic mice to exclude the possibility that the observed phenotype is not caused by disruption of an endogenous gene at a fortuitous transgene insertion site. Validity of the phenotype is strengthened if a transgene dose-dependent phenotype can be demonstrated. Often, transgenic mice are studied
460
BARBARA E. KREAM AND JOHN R. HARRISON
as hemizygotes. Subtle phenotypes can be enhanced if hemizygotes are bred together to produce homozygous progeny. Although producing a homozygous line reduces the need for frequent genotyping, transgene homozygosity can often lead to inbreeding and poor breeding performance. It is important to determine whether transgenic mice are obtained in Mendelian ratios. The under-representation of transgenic offspring may indicate embryonic lethality, while the converse may indicate multiple transgene insertion sites. Finally, it is important to consider the background strain used to generate transgenic mouse lines. Phenotypes may be altered among different inbred strains due to the presence of modifier genes. On the other hand, measurements of phenotypic traits may be more variable in an outbred strain, thus requiring a larger sample size. It may be advantageous to develop a transgenic in an inbred background for initial characterization; then, the line can be easily outbred to examine the impact of the transgene in a heterogeneous genetic background. Southern blot analysis of a transgenic line will serve to unequivocally identify the line.
III. TRANSGENIC MODELS IN BONE BIOLOGY A. Hormones that Regulate Bone Formation A number of hormonal signaling pathways are critical for bone formation including the BMP [30] and Wnt/LRP [31–33] pathways. Various components of their signaling pathways, including receptors, ligands, and downstream signaling molecules, have been genetically modified in the mouse and the reader is directed to the reviews mentioned above. We have chosen to discuss in detail the insulin like-growth factor-I (IGF-I) system, as manipulation of the system in mice has led to a greater understanding of how they may regulate bone mass acquisition in humans. The insulin/IGF system consists of the polypeptides insulin, IGF-I and IGF-II, their cognate receptors, six IGF-binding proteins (IGFBP 1–6) and IGFBP proteases [34, 35]. IGF-I and IGF-II signaling through tyrosine kinase type I IGF receptors (IGF1R) leads to enhanced cell proliferation, differentiation, and survival in many tissues. In postnatal life, IGF-I is produced in liver under the control of growth hormone and then secreted into the blood to provide an endocrine source of growth factor. In addition, most other tissues produce IGF-I, and this component plays an important local role in tissue homeostasis [35]. There has been much interest in elucidating the role of IGF-I in bone remodeling since many human and animal
studies show an association between circulating IGF-I and bone mineral density [36]. To delineate the role of IGF-I in embryonic development, the murine Igf1 gene has been inactivated by targeted mutagenesis [37, 38]. Fetal mice with a complete knockout of the Igf1 gene are nearly 40% smaller than wild-type littermates and display developmental abnormalities of muscle, lung, skeleton, and other organ systems [37, 38]. Mice with total IGF-I deficiency have varying degrees of perinatal lethality depending on the genetic background [39]. The few Igf1 knockout mice that survive into adulthood show severe growth impairment along with abnormalities in many organ systems. Trabecular bone mass in the proximal tibia of Igf1 knockout mice is increased [40], although this might be due to increased circulating growth hormone or the lack of IGF-I-mediated resorption. We have found that mice heterozygous null for the Igf1 gene (HET) in an outbred strain show seemingly normal perinatal survival and fertility. HET mice had reduced body weight, IGF-I levels, femur length, and bone mineral density compared to their sex-matched wild-type littermates between 1 and 18 months of age. Analysis by microcomputed tomography showed reduced cortical bone width and periosteal circumference at 2 and 8 months of age [40a]. Similar results in young HET mice have been published [41]. To distinguish between the endocrine and paracrine roles of IGF-I, liver-specific Igf1 knockout models (LID) have been generated using Cre-loxP technology [42, 43]. LID mice do not have detectable IGF-I expression in liver and at least 75% less circulating IGF-I than control littermates. Surprisingly, LID mice have apparently normal fetal development, do not show perinatal lethality and are similar in size to control littermates [42, 43]. In additional studies, a thorough inspection of the skeletons showed LID mice to have shorter femurs and reduced periosteal circumference, indicative of impaired longitudinal bone growth and cortical bone apposition [44]. Interestingly, these changes are similar to those in HET mice (Kream, unpublished), which have a more modest, yet global, reduction in circulating IGF-I. These data may imply that local as well as circulating IGF-I is important for bone acquisition. A particularly informative study was the conditional knockout of the type 1 IGF receptor gene (Igf1r) in osteoblasts using an OC-Cre transgene. These mice had increased osteoid surface indicating that IGF-I plays an important role in mineralization [45]. Several transgenic models have been generated in which IGF-I overexpression is targeted to osteoblasts. The osteocalcin promoter was first used to express a human IGF-I cDNA to mature osteoblasts [46]. Osteocalcin-IGF
Chapter 27 Transgenic Models of Bone Disease
transgenic mice have increased matrix apposition rate at 3 weeks and increased trabecular bone volume at 6 weeks of age. There were no differences in cortical bone width, osteoblast number, or osteoclast number at these time points and no changes in static or dynamic bone histomorphometric parameters at 24 weeks. In another model developed by our group, the Col3.6 promoter (including most of the Col1a1 first intron) was used to drive a FLAGtagged murine IGF-I cDNA more broadly in osteoblast lineage cells (Kream, unpublished). Transgenic calvaria had increased width, marrow spaces, and osteoclast number and surface. Notably, the calvarial phenotype was dependent on transgene expression and persisted with age. One line of mice had increased circulating IGF-I levels, presumably due to spillover from bone production. These mice showed an increase in femur length and mid-diaphyseal cortical width and area. In Col3.6-driven IGF-I transgenic mice, trabecular bone volume was not increased, which may have been due to increased resorption. The phenotypic differences between the two models (Col3.6-IGF and osteocalcin-IGF mice) are likely due to the distinct temporal and spatial expression pattern of the two promoters. The osteocalcin promoter is expressed in the perinatal period and is restricted to mature osteoblasts [13], while the Col3.6 promoter is expressed early during embryonic life and more broadly both spatially and temporally in osteoblast lineage cells [9]. Recently, several models have been produced in which various IGFBPs have been targeted to osteoblasts. Some IGFBPs act to sequester and inactivate IGFs, while others, such as IGFBP-5, may have IGF-independent effects on bone. Mice with constitutive transgenic IGFBP-3 overexpression are smaller and have reduced bone mineral density, reduced mineral apposition rates in cortical bone, and increased bone resorption markers [47]. Ex vivo osteoblast cultures showed that the decrease in bone formation may have been due to impaired proliferation [47]. Transgenic mice with constitutive overexpression of IGFBP-5 have decreased bone mineral density, with males being more severely affected than females [48]. Disparate effects of the transgene on bone formation rates and mineralizing surfaces were found on the periosteal (decreased) and endosteal (increased) surfaces [48]. Likewise, transgenic mice with osteocalcin promoter-driven IGFBP-5 overexpression had a transient decrease in femoral trabecular bone volume and mineral apposition rate [49, 50]. Since osteoblast number was not changed, osteoblast function may have been impaired; osteoclast number was not changed. These studies indicate that overexpression of IGFBPs decreases bone formation by impairing osteoblast number and/or function most likely by sequestering and
461 inactivating IGFs. However, some effects of transgenic IGFBP-5 may have been IGF-independent [48].
B. Cytokines that Regulate Bone Resorption Murine transgenic and knockout models have been integral in unraveling the complex network of cytokine signaling that controls osteoclast differentiation and bone resorption. Receptor activator of NF-κΒ (RΑΝΚL) is a member of the TNF ligand family that is expressed on the surface of osteoblast lineage cells and signals through the receptor activator of NF-κB (RANK) on the surface of hematopoietic osteoclast progenitors via a direct cell–cell interaction. RANK, a member of the TNF receptor family, signals the nuclear translocation of NF-κB, which activates a number of target genes that ultimately lead to osteoclast progenitor fusion, osteoclast differentiation, and survival. To allow this process to be tightly controlled, a novel secreted glycoprotein called osteoprotegerin (OPG), a soluble member of the TNF receptor family, is expressed by cells of the osteoblast lineage and acts as a decoy receptor for RANKL. The balance of RANKL and OPG expression in osteoblasts is therefore a major determinant of osteoclast differentiation and bone resorption, and a number of factors known to modulate bone resorption regulate their expression. OPG was the first member of this network to be discovered. A transgenic mouse model engineered for hepatic overexpression of OPG produced a nonlethal osteopetrotic phenotype in which osteoclasts failed to undergo terminal differentiation [51]. Conversely, OPG-null mice exhibited osteoporosis due to excessive osteoclastogenesis, resulting in a severe loss of trabecular bone, increased cortical porosity, and diminished mechanical bone strength [52, 53]. Min et al. [54] demonstrated that this osteoporotic phenotype could be rescued by systemic transgenic overexpression of OPG, which restored osteoclastogenesis. Similar genetic approaches subsequently established the roles of RANKL and its cognate receptor, RANK, in this system. Both RANKL- and RANK-deficient mice were found to be severely osteopetrotic, with a complete absence of osteoclasts, leading to profound accumulation of bone and a failure of tooth eruption [55, 56]. After the major cytokines responsible for osteoclast differentiation had been identified, attention shifted to the downstream cellular signaling pathways mediating RANK signaling. As its name implies, RANK is capable of activating NF-κB by inducing degradation of cytoplasmic IκB, allowing translocation of NF-κB complexes to the nucleus. The essential nature of NF-κB signaling in osteoclasts was
462
BARBARA E. KREAM AND JOHN R. HARRISON
underscored by the observation that mice deficient in the p50 and p52 subunits of NF-κB exhibited osteopetrosis [57, 58]. Subsequent studies showed that RANK-expressing osteoclast progenitors are present in p50/52 doubleknockout mice, but terminal differentiation into TRAPpositive osteoclasts was blocked [59]. The RANK receptor is coupled to NF-κB signaling by the TNF receptorassociated factor (TRAF) family of adapter proteins. Not long after the observation that NF-κB signaling was essential for osteoclastogenesis, a requirement for TRAF6 in this process was demonstrated by the finding that TRAF5deficient mice likewise exhibited the osteopetrotic/tooth eruption phenotype [60]. There is, however, both longstanding and new evidence that other signaling pathways are essential for osteoclast differentiation. For example, it has been known for more than 10 years that c-fos-null mice are osteopetrotic secondary to a lack of differentiated osteoclasts. New studies have shown that a nuclear factor of activated T cells (NFAT) is a target for Fos and Jun signaling in the osteoclast progenitor, and that expression of an activated form of NFAT can rescue osteoclast differentiation in c-fos-null progenitor cells exposed to RANKL [61, 62]. While much progress has been made on understanding the regulation of osteoclastogenesis using global gene ablation and overexpression strategies, such studies are complicated by a lack of specificity of many target genes to the osteoclast lineage. Clearly, more sophisticated models are required for targeted overexpression and conditional gene knockout strategies in osteoclasts and their progenitors. Toward this end, transgenic mouse models with osteoclastspecific Cre recombinase are being developed. Chiu et al. [20] have recently generated mice driving Cre expression under control of the osteoclast-specific TRAP and cathepsin K promoters and demonstrated expression specifically in bones and cartilage of a Cre-reporter mouse strain.
C. Nuclear Receptors Many steroid hormones, including estrogens, vitamin D, glucocorticoids, and several orphan receptors affect bone development and remodeling. Transgenic approaches to understanding the role of these hormones have included knockout and overexpression of their cognate receptors. The functions of estrogen in bone have been highly studied due to their role in postmenopausal osteoporosis. To determine the role of estrogens in bone metabolism, the estrogen receptors, ERα and ERβ, have been targeted in mice by several laboratories producing either complete or incomplete knockouts [63, 64]. Although there have been some discrepant results, there appear to be complex gender- and
species-specific phenotypes [64]. It seems that ERα primarily mediates the effects on bone development in males and females and regulates the suppressive effect of estrogens on bone resorption [64]. ERβ may counteract some of the effects of ERα on bone formation. Since estrogens are formed by aromatization of androgens, genetic perturbations of the aromatase gene have been of particular interest. Aromatase knockout female and male mice have low bone mass, indicating that the gene is functional in both sexes [65]. By contrast, transgenic mice with human ubiquitin C promoter-driven aromatase expression have increased bone mass [66]. Male mice with transgenic expression of the androgen receptor driven by the rat 3.6-kb Col1a1 promoter had an altered skeleton characterized by shorter femurs, thickened calvaria, and increased trabecular bone volume. The phenotype was sexually dimorphic since females were not affected [67]. The vitamin D receptor (VDR) has been both ablated and overexpressed in mice. VDR knockout mice have rickets [68] and they can be cured by normalization of mineral homeostasis [69]. Interestingly, the rickets in VDR knockout mice appear to be secondary to the apoptosis of hypertrophic chondrocytes [70]. Human osteocalcin promoter-targeted VDR transgenic mice had increased bone formation and decreased bone resorption [71]. Although glucocorticoid excess has deleterious effects on bone, the endogenous function of glucocorticoids in bone has been elusive. There is unequivocal evidence that glucocorticoid excess in humans and most animal species leads to glucocorticoid-induced osteoporosis, characterized by a decrease in bone formation due to decreased matrix protein production, decreased preosteoblast replication and increased osteoblast apoptosis; yet, many in vitro studies indicate that glucocorticoids facilitate osteoblast differentiation, particularly in human and rat osteoblast cultures [72]. To determine the role of endogenous glucocorticoids in bone, transgenic ligand inactivation models employing 11β-hydroxysteroid dehydrogenase 2 (11β-HSD2) have been developed [73, 74]. This enzyme metabolizes active glucocorticoids to inactive metabolites, thereby producing a barrier to glucocorticoid action in target cells [75]. This approach was undertaken because global knockout of the glucocorticoid receptor (GR) results in perinatal lethality, precluding an analysis of the bone phenotype in postnatal life [76]. Moreover, while a global or conditional GR knockout would not preclude glucocorticoids from signaling through the mineralocorticoid receptor (MR), a ligand inactivation strategy would block glucocorticoid signaling through both the GR and MR. Both the osteocalcin [74] and the 2.3-kb rat Col1a1 [73] promoters have been used to drive 11β-HSD2 in osteoblasts. Transgenic mice with Col2.3-driven 11β-HSD2 showed a sexually dimorphic bone phenotype consisting of decreased vertebral bone
Chapter 27 Transgenic Models of Bone Disease
mass in females, with no change in femoral trabecular bone in males or females [73]. However, both sexes have decreased femoral cortical bone mass and decreased latestage osteoblast differentiation in primary calvarial and bone marrow stromal cell cultures (Kream, unpublished). Interestingly, osteoid volume was increased in female vertebral bone [73]. An osteocalcin-driven 11β-HSD2 transgene blocked glucocorticoid-induced osteoblast and osteocyte apoptosis [74]. These data indicate that glucocorticoids are required for bone formation and mineralization. Recent data indicate that 11β-HSD1 expression is high in undifferentiated human osteoblast cultures and decreases as the cells [77]. Thus, it can be inferred that local glucocorticoid production in the bone microenvironment plays a role osteoblast differentiation.
D. Transcription Factors Identification of Runx2 (originally Cbfa1) as a master regulator of osteoblast differentiation and osteogenesis relied heavily on genetic mouse models and shed light on the etiology of a heritable human skeletal disorder. Homozygous mice with targeted deletion of the Runx2 gene die shortly after birth due to respiratory failure and show a complete absence of skeletal mineralization due to inhibition of osteoblast differentiation [78]. Expression of Runx2 is restricted to skeletal elements, and is both necessary and sufficient for expression of a number of osteoblast lineage-specific or enriched genes, including bone sialoprotein, type I collagen, and osteocalcin [79]. Heterozygous-null mice, on the other hand, show a nonlethal skeletal phenotype that includes hypoplasia of the clavicles and delayed development of the intramembranous cranial bones [80]. Strikingly, this phenotype parallels that seen in human patients with cleidocranial dysplasia (CCD), who suffer from short stature, supernumerary teeth, hypoor aplastic clavicles, and patent fontanelles. This autosomal dominant condition was shown to arise by heterozygous loss of Runx2 resulting from deletions, insertions, or mis-sense mutations within the Runx2 gene locus [80]. Similarly, a radiation-induced mouse mutant (Ccd), with a phenotype indistinguishable from the Runx2 heterozygous-null mouse, was found to have a deletion affecting the Runx2 gene. To determine whether Runx2 plays a role not only in development but in the postnatal skeleton as well, Ducy et al. [81] have used a targeted transgenic loss-of-function strategy. In this model, a dominant-negative version of Runx2 containing only the DNA-binding domain was overexpressed in postnatal osteoblasts using a 1.3-kb osteocalcin promoter fragment. These mice showed a normal skeleton at birth but developed a progressive osteopenia
463 due to a reduced rate of bone formation that was associated with reduced expression of osteoblast extracellular matrix components. In some cases, transgenic models have yielded important insights in the field of bone biology without bone-specific targeting of the transgene. For example, two members of the AP-1 transcription factor family that lack trans-activation domains, Fra-1, and deltaFosB, cause dramatic effects on bone mass in mice when overexpressed under the control of ubiquitous promoters. Fra-1 expression, driven by the major histocompatibility complex class I antigen H2-Kb promoter, caused increases in cortical width as well as trabecular number and thickness beginning at 4 weeks of age [82]. This osteosclerotic phenotype results from an increase in bone formation that is caused, in turn, by enhanced osteoblast differentiation and an increase in the osteoblast and mineralizing surfaces. No effects on osteoblast proliferation were observed in this model. While Fra-1 lacks transcription activation domains and is generally thought to be a dominant negative regulator of Fos function, it should be noted that a “knock-in” of Fra-1 into the c-fos locus (effectively replacing c-fos) rescued of some abnormalities observed in c-fos-null mice, including the lack of osteoclasts. These observations suggest that transcription factors devoid of transactivation domains may not act solely as dominant negative inhibitors of their transcriptionally active counterparts. A somewhat similar osteosclerotic phenotype was seen in transgenic mice expressing a naturally occurring truncated splice variant of the Fos family known as deltaFosB. Expression of this transcription factor, directed by the nonspecific enolase promoter, caused an increase in bone formation and a progressive osteosclerosis with no effects on bone resorption [83]. Unlike the Fra-1 model, the increase in bone mass seen in these mice is accompanied by a reduction in adipogenesis both in vivo and in vitro, indicating a cell autonomous effect on fat. To determine whether deltaFosB could regulate bone formation in an adult mouse independently of any developmental effects, a model was developed in which deltaFosB expression was regulated using the inducible Tet-Off expression system. In these mice, postnatal induction of deltaFosB expression by withdrawal of tetracycline causes a progressive increase in bone mass that was reversed by the subsequent re-administration of tetracycline and silencing of deltaFosB expression, demonstrating the ability of deltaFosB to regulate bone mass in the adult animal [84]. Since fat mass was reduced in deltaFosB transgenic mice, it is possible that some effects of the transgene on bone mass were secondary to reduced adipogenesis and low serum leptin levels. However, long-term administration of leptin to these mice did not rescue the bone phenotype [85].
464 Alternatively, since osteoblasts and adipocytes share a common progenitor cell in bone marrow, it is possible that deltaFosB was promoting osteoblast differentiation at the expense of adipogenesis. This question was answered by targeting deltaFosB expression specifically to bone using an osteocalcin promoter fragment, which is expressed only in mature osteoblasts. These mice showed enhanced bone mass with no effect on adiposity, indicating that deltaFosB exerts independent cell autonomous effects on both the osteoblast and adipocyte lineages [86]. It has long been recognized that a reciprocal relationship exists between bone marrow adiposity and trabecular bone volume, leading to a search for transcription factors that might act as a molecular switch between these closely related lineages. One such gene candidate is PPARg, a member of the peroxisome proliferator activated receptor family that is known to play a critical role in adipocyte differentiation. Its expression is stimulated by C/EBP transcription factors, including C/EBPα, with which it syergistically activates expression of terminal adipocyte marker genes. Akune et al. [87] recently reported that embryonic stem cells from PPARγ-deficient mice failed to differentiate into adipocytes, but rather demonstrated spontaneous differentiation into osteoblasts. Mice heterozygous for PPARγ showed increased osteoblastogenesis and exhibited high bone mass. The cell-autonomous nature of this haploinsufficiency was demonstrated using bone marrow stromal cell cultures, which displayed an increase in osteoblast differentiation at the expense of adipogenesis. Cock et al. [88], in a study of a lipodystrophic PPARγ hyp/hyp mouse model, reported a marked increase in bone mass and consequent reduction in the size of the bone-marrow cavity, resulting in compensatory extramedullary hematopoiesis in the spleen. These gene disruption studies in mice raise the possibility that polymorphisms in the PPARg gene, already the subject of intense scrutiny for their possible contribution to obesity and the metabolic syndrome, might also play a role in the genetic specification of bone mass. As mentioned previously, C/EBP transcription factors regulate adipocyte differentiation, and recent evidence suggests that osteoblasts and adipocytes share a common pluripotent progenitor in bone marrow. However, little is known about the role of C/EBP transcription factors in the control of osteoblast differentiation or function. To assess the role of C/EBP transcription factors in osteogenesis, we targeted a naturally occurring dominant negative C/EBP isoform, called p20C/EBPβ or LIP, using the 3.6-kb Col1a1 promoter/first intron construct [89]. All four transgenic lines generated showed osteopenia, ranging from mild to severe, as evidenced by reduced trabecular bone volume. Severely affected lines also showed reduced cortical width. Dynamic histomorphometry demonstrated a decrease in
BARBARA E. KREAM AND JOHN R. HARRISON
the mineral apposition and bone formation rates, which was associated with reduced Col1a1 and osteocalcin mRNA levels. These data, using a dominant negative overexpression strategy to block all C/EBP function, suggest that C/EBP transcription factors may be required for osteoblast differentiation or maintenance of osteoblast function, and may be important determinants of bone mass. The cAMP signaling pathway is particularly important to bone biology since cAMP production is stimulated by the osteotropic hormones parathyroid hormone (PTH) and parathyroid hormone-related peptide (PTHrP). The classical cAMP signaling pathway includes receptor-mediated cAMP production by adenylate cyclase, cAMP binding, and activation of protein kinase A, translocation of PKA into the nucleus, phosphorylation of the cAMP response element binding protein (CREB), and regulation of gene transcription. Various approaches have been used to understand the role of the cAMP pathway in bone cells. A transgenic approach to understanding the role of the cAMP signaling pathway in endochondral bone formation has involved driving expression of a dominant negative CREB molecule (A-CREB) in chondrocytes by the type II collagen promoter [90]. The A-CREB transgene led to a shortlimbed dwarfism due to a delay in chondrocyte hypertrophy [90]. We have used a similar strategy to explore the role of cAMP signaling in osteoblasts. The inducible cAMP early repressor (ICER) is a naturally occurring dominant negative isoform of the cAMP responsive element modulator (Crem) gene [91]. ICER is strongly and transiently induced by cAMP and PTH in osteoblasts [92]. To gain further insight into the role of ICER in bone, and to determine the importance of the cAMP signaling pathway in osteoblasts, we have used the 3.6-kb Col1a1 promoter to drive ICER overexpression in osteoblasts. Mice in the most affected transgenic lines were small and had almost no trabecular bone (Kream, unpublished). There was impairment of bone formation and an increase in osteoclast formation and resorption. Thus, the bone phenotype of ICER transgenic mice suggests the cAMP pathway is critical for bone formation and remodeling. Genetic mouse models have also been instrumental in the recent demonstration that ATF4, a member of the CREB/ATF transcription factor family, is another key regulator of osteoblast function and bone mass. Using an ATF4-null mouse model, Yang et al. [93] reported a delay in bone formation during embryogenesis and persistence of a reduced bone mass phenotype through adulthood. This osteopenia was associated with reduced expression of osteocalcin and an inhibition of type I collagen synthesis at a post-transcriptional level. Interestingly, these authors demonstrated that ATF4 is a phosphorylation target for Rsk2, a growth factor-regulated protein kinase. An Rsk2
465
Chapter 27 Transgenic Models of Bone Disease
gene ablation model showed a reduction of femur length, hypoplasia of calvaria bones with failure to close the fontanelles, and a reduced rate of bone formation. This phenotype appears to mirror the bone phenotype seen in human patients with Coffin–Lowry syndrome, an X-linked genetic disorder characterized by mental retardation and skeletal dysplasia. ATF4 phosphorylation may also play a role in the neural control of bone remodeling mediated by the sympathetic nervous system. Sympathetic stimulation of preosteoblastic cells via the β2-adrenergic receptor stimulates expression of receptor activator of NF-κB ligand, or RANKL, thereby promoting osteoclast differentiation and bone remodeling [94].
IV. PERSPECTIVES AND FUTURE DIRECTIONS Many new tools have recently become available for studying gene expression in transgenic models. Visual markers based on GFP, used singly or in combination, will allow assessment of promoter activity in real time in vivo [95]. It is possible to follow in real time the intracellular trafficking of chimeric proteins containing GFP domains. The knock-in of GFP reporters into endogenous genes will enable investigators to observe the spatial and temporal pattern of transgene expression in vivo. GFP can be used as a reporter to examine the induction or repression of gene transcription in vivo in distinct populations of cells. Finally, it will be possible to use GFP-marked cells to demonstrate the homing of cells to different tissues in vivo [95]. Newly developed technologies should result in transgenic models in which the spatial and temporal pattern of expression of the gene of interest better recapitulates the endogenous gene. Bacterial artificial chromosomes (BAC) and BAC recombineering methodology have been used to develop transgenes that are expressed in a developmental stage-specific and tissue-specific manner [96, 97]. The large chromosomal segments in BACs contain multiple genes with their full spectrum of regulatory elements. A gene of interest within a BAC is engineered with a reporter or gene of interest and used for transgenesis. Finally, the use of small interfering RNAs (siRNA) will allow investigators to develop models with hypomorphic gene expression. High levels of siRNA can be driven with viral vectors to accomplish sustained and heritable knockdown of gene expression in vivo [98–100]. Since it is relatively easy to make siRNA vectors and they can be obtained commercially, this technology may become an efficient means of modulating gene expression that is easier and faster than producing a complex targeting construct for each gene. Thus, these recent technological advances provide
remarkable potential for the future of functional genomic studies in the mouse.
References 1. Capecchi, M. R. (2005). Gene targeting in mice: functional analysis of the mammalian genome for the twenty-first century. Nat. Rev. Genet. 6, 507–512. 2. Davey, R. A., MacLean, H. E., McManus, J. F., Findlay, D. M., and Zajac, J. D. (2004). Genetically modified animal models as tools for studying bone and mineral metabolism. J. Bone Miner. Res. 19, 882–892. 3. McCauley, L. K. (2001). Transgenic mouse models of metabolic bone disease. Curr. Opin. Rheumatol. 13, 316–325. 4. Schilling, A. F., Priemel, M., Timo Beil, F., Haberland, M., Holzmann, T., Catala-Lehnen, P., Pogoda, P., Blicharski, D., Muldner, C., Locherbach, C., Rueger, J. M., and Amling, M. (2001). Transgenic and knock out mice in skeletal research. Towards a molecular understanding of the mammalian skeleton. J. Musculoskelet. Neuronal. Interact. 1, 275–289. 5. Clark, S., and Rowe, D. (2002). Application of transgenic mice to problems of skeletal biology. In “Principles of Bone Biology” (J. P. Bilezikian, L. G. Raisz, and G. A. Rodan, eds.), Vol. 2, pp. 1491–1502. Academic Press, San Diego, CA. 6. Jaenisch, R. (1988). Transgenic animals. Science 240, 1468–1474. 7. Lichtler, A., Stover, M. L., Angilly, J., Kream, B., and Rowe, D. W. (1989). Isolation and characterization of the rat alpha 1(I) collagen promoter. Regulation by 1, 25-dihydroxyvitamin D. J. Biol. Chem. 264, 3072–3077. 8. Rossert, J., Eberspaecher, H., and de Crombrugghe, B. (1995). Separate cis-acting DNA elements of the mouse pro-a1(I) collagen promoter direct expression of reporter genes to different type I collagenproducing cells in transgenic mice. J. Cell Biol. 129, 1421–1432. 9. Kalajzic, I., Kalajzic, Z., Kaliterna, M., Gronowicz, G., Clark, S. H., Lichtler, A. C., and Rowe, D. (2002). Use of type I collagen green fluorescent protein transgenes to identify subpopulations of cells at different stages of the osteoblast lineage. J. Bone Miner. Res. 17, 15–25. 10. Dacquin, R., Starbuck, M., Schinke, T., and Karsenty, G. (2002). Mouse alpha1(I)-collagen promoter is the best known promoter to drive efficient Cre recombinase expression in osteoblast. Dev. Dyn. 224, 245–251. 11. Bogdanovic, Z., Bedalov, A., Krebsbach, P. H., Pavlin, D., Woody, C. O., Clark, S. H., Thomas, H. F., Rowe, D. W., Kream, B. E., and Lichtler, A. C. (1994). Upstream regulatory elements necessary for expression of the rat COL1A1 promoter in transgenic mice. J. Bone Miner. Res. 9, 285–292. 12. Braut, A., Kalajzic, I., Kalajzic, Z., Rowe, D. W., Kollar, E. J., and Mina, M. (2002). Col1a1-GFP transgene expression in developing incisors. Connect Tissue Res. 43, 216–219. 13. Kesterson, R. A., Stanley, L., DeMayo, F., Finegold, M., and Pike, J. W. (1993). The human osteocalcin promoter directs bone-specific vitamin D-regulatable gene expression in transgenic mice. Mol. Endocrinol. 7, 462–467. 14. Frendo, J. L., Xiao, G., Fuchs, S., Franceschi, R. T., Karsenty, G., and Ducy, P. (1998). Functional hierarchy between two OSE2 elements in the control of osteocalcin gene expression in vivo. J. Biol. Chem. 273, 30509–30516. 15. Schinke, T., and Karsenty, G. (1999). Characterization of Osf1, an osteoblast-specific transcription factor binding to a critical cis-acting element in the mouse Osteocalcin promoters. J. Biol. Chem. 274, 30182–30189. 16. Clemens, T. L., Tang, H., Maeda, S., Kesterson, R. A., Demayo, F., Pike, J. W., and Gundberg, C. M. (1997). Analysis of osteocalcin
466
17.
18. 19.
20.
21. 22. 23. 24.
25.
26.
27.
28.
29.
30. 31.
32. 33. 34.
35.
36.
37.
expression in transgenic mice reveals a species difference in vitamin D regulation of mouse and human osteocalcin genes. J. Bone Miner. Res. 12, 1570–1576. Frenkel, B., Capparelli, C., Van Auken, M., Baran, D., Bryan, J., Stein, J. L., Stein, G. S., and Lian, J. B. (1997). Activity of the osteocalcin promoter in skeletal sites of transgenic mice and during osteoblast differentiation in bone marrow-derived stromal cell cultures: effects of age and sex. Endocrinology 138, 2109–2116. Nagy, A. (2000). Cre recombinase: the universal reagent for genome tailoring. Genesis 26, 99–109. Kalajzic, I., Braut, A., Guo, D., Jiang, X., Kronenberg, M. S., Mina, M., Harris, M. A., Harris, S. E., and Rowe, D. W. (2004). Dentin matrix protein 1 expression during osteoblastic differentiation, generation of an osteocyte GFP-transgene. Bone 35, 74–82. Chiu, W. S., McManus, J. F., Notini, A. J., Cassady, A. I., Zajac, J. D., and Davey, R. A. (2004). Transgenic mice that express Cre recombinase in osteoclasts. Genesis 39, 178–185. Ferron, M., and Vacher, J. (2005). Targeted expression of Cre recombinase in macrophages and osteoclasts in transgenic mice. Genesis 41, 138–145. Lewandoski, M. (2001). Conditional control of gene expression in the mouse. Nat. Rev. Genet. 2, 743–755. Soriano, P. (1999). Generalized lacZ expression with the ROSA26 Cre reporter strain [letter]. Nat. Genet. 21, 70–71. Novak, A., Guo, C., Yang, W., Nagy, A., and Lobe, C. G. (2000). Z/EG, a double reporter mouse line that expresses enhanced green fluorescent protein upon Cre-mediated excision. Genesis 28, 147–155. Lobe, C. G., Koop, K. E., Kreppner, W., Lomeli, H., Gertsenstein, M., and Nagy, A. (1999). Z/AP, a double reporter for cre-mediated recombination. Dev. Biol. 208, 281–292. Indra, A. K., Warot, X., Brocard, J., Bornert, J. M., Xiao, J. H., Chambon, P., and Metzger, D. (1999). Temporally-controlled sitespecific mutagenesis in the basal layer of the epidermis: comparison of the recombinase activity of the tamoxifen-inducible Cre-ER(T) and Cre-ER(T2) recombinases. Nucleic Acids Res. 27, 4324–4327. Feil, R., Brocard, J., Mascrez, B., LeMeur, M., Metzger, D., and Chambon, P. (1996). Ligand-activated site-specific recombination in mice. Proc. Natl Acad. Sci. USA 93, 10887–10890. Kim, J. E., Nakashima, K., and de Crombrugghe, B. (2004). Transgenic mice expressing a ligand-inducible cre recombinase in osteoblasts and odontoblasts: a new tool to examine physiology and disease of postnatal bone and tooth. Am. J. Pathol. 165, 1875–1882. Gossen, M., Freundlieb, S., Bender, G., Muller, G., Hillen, W., and Bujard, H. (1995). Transcriptional activation by tetracyclines in mammalian cells. Science 268, 1766–1769. Wan, M., and Cao, X. (2005). BMP signaling in skeletal development. Biochem. Biophys. Res. Commun. 328, 651–657. Johnson, M. L. (2004). The high bone mass family – the role of Wnt/Lrp5 signaling in the regulation of bone mass. J. Musculoskelet. Neuronal. Interact. 4, 135–138. Johnson, M. L., Harnish, K., Nusse, R., and Van Hul, W. (2004). LRP5 and Wnt signaling: a union made for bone. J. Bone Miner. Res. 19, 1749–1757. Westendorf, J. J., Kahler, R. A., and Schroeder, T. M. (2004). Wnt signaling in osteoblasts and bone diseases. Gene 341, 19–39. Rosen, C. J., and Donahue, L. R. (1995). Insulin-like growth factorspotential therapeutic options for osteoporosis. Trends Endocrinol. Metab. 6, 235–240. Van Wyk, J. J., and Smith, E. P. (1999). Insulin-like growth factors and skeletal growth: possibilities for therapeutic interventions. J. Clin. Endocrinol. Metab. 84, 4349–4354. Rosen, C. J. (2004). Insulin-like growth factor I and bone mineral density: experience from animal models and human observational studies. Best Pract. Res. Clin. Endocrinol. Metab. 18, 423–435. Baker, J., Liu, J. P., Robertson, E. J., and Efstratiadis, A. (1993). Role of insulin-like growth factors in embryonic and postnatal growth. Cell 75, 73–82.
BARBARA E. KREAM AND JOHN R. HARRISON 38. Powell-Braxton, L., Hollingshead, P., Warburton, C., Dowd, M., Pitts-Meek, S., Dalton, D., Gillett, N., and Stewart, T. A. (1993). IGF-I is required for normal embryonic growth in mice. Genes Dev. 7, 2609–2617. 39. Liu, J. L., Yakar, S., and LeRoith, D. (2000). Mice deficient in liver production of insulin-like growth factor I display sexual dimorphism in growth hormone-stimulated postnatal growth. Endocrinology 141, 4436–4441. 40. Bikle, D., Majumdar, S., Laib, A., Powell-Braxton, L., Rosen, C., Beamer, W., Nauman, E., Leary, C., and Halloran, B. (2001). The skeletal structure of insulin-like growth factor I-deficient mice. J. Bone Miner. Res. 16, 2320-2329. 40a.He, J., Rosen, C. J., Adams, D. J., and Kream, B. E. (2006). Postnatal growth and bone mass in mice with IGF-I haploinsufficiency. Bone, Jan 18 [Epub ahead of print]. 41. Mohan, S., and Baylink, D. J. (2005). Impaired skeletal growth in mice with haploinsufficiency of IGF-I: genetic evidence that differences in IGF-I expression could contribute to peak bone mineral density differences. J. Endocrinol. 185, 415–420. 42. Sjogren, K., Liu, J. L., Blad, K., Skrtic, S., Vidal, O., Wallenius, V., LeRoith, D., Tornell, J., Isaksson, O. G., Jansson, J. O., and Ohlsson, C. (1999). Liver-derived insulin-like growth factor I (IGF-I) is the principal source of IGF-I in blood but is not required for postnatal body growth in mice. Proc. Natl Acad. Sci. USA 96, 7088–7092. 43. Yakar, S., Liu, J. L., Stannard, B., Butler, A., Accili, D., Sauer, B., and LeRoith, D. (1999). Normal growth and development in the absence of hepatic insulin-like growth factor I. Proc. Natl Acad. Sci. USA 96, 7324–7329. 44. Ueki, I., Ooi, G. T., Tremblay, M. L., Hurst, K. R., Bach, L. A., and Boisclair, Y. R. (2000). Inactivation of the acid labile subunit gene in mice results in mild retardation of postnatal growth despite profound disruptions in the circulating insulin-like growth factor system. Proc. Natl Acad. Sci. USA 97, 6868–6873. 45. Zhang, M., Xuan, S., Bouxsein, M. L., von Stechow, D., Akeno, N., Faugere, M. C., Malluche, H., Zhao, G., Rosen, C. J., Efstratiadis, A., and Clemens, T. L. (2002). Osteoblast-specific knockout of the insulin-like growth factor (IGF) receptor gene reveals an essential role of IGF signaling in bone matrix mineralization. J. Biol. Chem. 277, 44005–44012. 46. Zhao, G., Monier-Faugere, M. C., Langub, M. C., Geng, Z., Nakayama, T., Pike, J. W., Chernausek, S. D., Rosen, C. J., Donahue, L. R., Malluche, H. H., Fagin, J. A., and Clemens, T. L. (2000). Targeted overexpression of insulin-like growth factor I to osteoblasts of transgenic mice: increased trabecular bone volume without increased osteoblast proliferation. Endocrinology 141, 2674–2682 47. Silha, J. V., Mishra, S., Rosen, C. J., Beamer, W. G., Turner, R. T., Powell, D. R., and Murphy, L. J. (2003). Perturbations in bone formation and resorption in insulin-like growth factor binding protein-3 transgenic mice. J. Bone Miner. Res. 18, 1834–1841. 48. Salih, D. A., Mohan, S., Kasukawa, Y., Tripathi, G., Lovett, F. A., Anderson, N. F., Carter, E. J., Wergedal, J. E., Baylink, D. J., and Pell, J. M. (2005). Insulin-like growth factor-binding protein-5 induces a gender-related decrease in bone mineral density in transgenic mice. Endocrinology 146, 931–940. 49. Devlin, R. D., Du, Z., Buccilli, V., Jorgetti, V., and Canalis, E. (2002). Transgenic mice overexpressing insulin-like growth factor binding protein-5 display transiently decreased osteoblastic function and osteopenia. Endocrinology 143, 3955–3962. 50. Atti, E., Boskey, A. L., and Canalis, E. (2005). Overexpression of IGFbinding protein 5 alters mineral and matrix properties in mouse femora: an infrared imaging study. Calcif. Tissue Int. 76, 187–193. 51. Simonet, W. S., Lacey, D. L., Dunstan, C. R., Kelley, M., Chang, M. S., Luthy, R., Nguyen, H. Q., Wooden, S., Bennett, L., Boone, T., Shimamoto, G., DeRose, M., Elliott, R., Colombero, A., Tan, H. L., Trail, G., Sullivan, J., Davy, E., Bucay, N., Renshaw-Gegg, L., Hughes, T. M., Hill, D., Pattison, W., Campbell, P., Sander, S., Van, G.,
Chapter 27 Transgenic Models of Bone Disease
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
Tarpley, J., Derby, P., Lee, R., and Boyle, W. J. (1997). Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89, 309–319. Bucay, N., Sarosi, I., Dunstan, C. R., Morony, S., Tarpley, J., Capparelli, C., Scully, S., Tan, H. L., Xu, W., Lacey, D. L., Boyle, W. J., and Simonet, W. S. (1998). Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification. Genes Dev. 12, 1260–1268. Mizuno, A., Amizuka, N., Irie, K., Murakami, A., Fujise, N., Kanno, T., Sato, Y., Nakagawa, N., Yasuda, H., Mochizuki, S., Gomibuchi, T., Yano, K., Shima, N., Washida, N., Tsuda, E., Morinaga, T., Higashio, K., and Ozawa, H. (1998). Severe osteoporosis in mice lacking osteoclastogenesis inhibitory factor/osteoprotegerin. Biochem. Biophys. Res. Commun. 247, 610–615. Min, H., Morony, S., Sarosi, I., Dunstan, C. R., Capparelli, C., Scully, S., Van, G., Kaufman, S., Kostenuik, P. J., Lacey, D. L., Boyle, W. J., and Simonet, W. S. (2000). Osteoprotegerin reverses osteoporosis by inhibiting endosteal osteoclasts and prevents vascular calcification by blocking a process resembling osteoclastogenesis. J. Exp. Med. 192, 463–474. Kong, Y. Y., Yoshida, H., Sarosi, I., Tan, H. L., Timms, E., Capparelli, C., Morony, S., Oliveira-dos-Santos, A. J., Van, G., Itie, A., Khoo, W., Wakeham, A., Dunstan, C. R., Lacey, D. L., Mak, T. W., Boyle, W. J., and Penninger, J. M. (1999). OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397, 315–323. Dougall, W. C., Glaccum, M., Charrier, K., Rohrbach, K., Brasel, K., De Smedt, T., Daro, E., Smith, J., Tometsko, M. E., Maliszewski, C. R., Armstrong, A., Shen, V., Bain, S., Cosman, D., Anderson, D., Morrissey, P. J., Peschon, J. J., and Schuh, J. (1999). RANK is essential for osteoclast and lymph node development. Genes Dev. 13, 2412–2424. Franzoso, G., Carlson, L., Xing, L., Poljak, L., Shores, E. W., Brown, K. D., Leonardi, A., Tran, T., Boyce, B. F., and Siebenlist, U. (1997). Requirement for NF-kappaB in osteoclast and B-cell development. Genes Dev. 11, 3482–3496. Iotsova, V., Caamano, J., Loy, J., Yang, Y., Lewin, A., and Bravo, R. (1997). Osteopetrosis in mice lacking NF-kappaB1 and NF-kappaB2. Nat. Med. 3, 1285–1289. Xing, L., Bushnell, T. P., Carlson, L., Tai, Z., Tondravi, M., Siebenlist, U., Young, F., and Boyce, B. F. (2002). NF-kappaB p50 and p52 expression is not required for RANK-expressing osteoclast progenitor formation but is essential for RANK- and cytokine-mediated osteoclastogenesis. J. Bone Miner. Res. 17, 1200–1210. Lomaga, M. A., Yeh, W. C., Sarosi, I., Duncan, G. S., Furlonger, C., Ho, A., Morony, S., Capparelli, C., Van, G., Kaufman, S., van der Heiden, A., Itie, A., Wakeham, A., Khoo, W., Sasaki, T., Cao, Z., Penninger, J. M., Paige, C. J., Lacey, D. L., Dunstan, C. R., Boyle, W. J., Goeddel, D. V., and Mak, T. W. (1999). TRAF6 deficiency results in osteopetrosis and defective interleukin-1, CD40, and LPS signaling. Genes Dev. 13, 1015–1024. Ikeda, F., Nishimura, R., Matsubara, T., Tanaka, S., Inoue, J., Reddy, S. V., Hata, K., Yamashita, K., Hiraga, T., Watanabe, T., Kukita, T., Yoshioka, K., Rao, A., and Yoneda, T. (2004). Critical roles of c-Jun signaling in regulation of NFAT family and RANKL-regulated osteoclast differentiation. J. Clin. Invest. 114, 475–484. Matsuo, K., Galson, D. L., Zhao, C., Peng, L., Laplace, C., Wang, K. Z., Bachler, M. A., Amano, H., Aburatani, H., Ishikawa, H., and Wagner, E. F. (2004). Nuclear factor of activated T-cells (NFAT) rescues osteoclastogenesis in precursors lacking c-Fos. J. Biol. Chem. 279, 26475–26480. McCauley, L. K., Tozum, T. F., and Rosol, T. J. (2002). Estrogen receptors in skeletal metabolism: lessons from genetically modified models of receptor function. Crit. Rev. Eukaryot. Gene. Expr. 12, 89–100. Windahl, S. H., Andersson, G., and Gustafsson, J. A. (2002). Elucidation of estrogen receptor function in bone with the use of mouse models. Trends Endocrinol. Metab. 13, 195–200.
467 65. Oz, O. K., Hirasawa, G., Lawson, J., Nanu, L., Constantinescu, A., Antich, P. P., Mason, R. P., Tsyganov, E., Parkey, R. W., Zerwekh, J. E., and Simpson, E. R. (2001). Bone phenotype of the aromatase deficient mouse. J. Steroid Biochem. Mol. Biol. 79, 49–59. 66. Peng, Z., Li, X., Makela, S., Vaananen, H. K., and Poutanen, M. (2004). Skeletal changes in transgenic male mice expressing human cytochrome p450 aromatase. J. Bone Miner. Res. 19, 1320–1328. 67. Wiren, K. M., Zhang, X. W., Toombs, A. R., Kasparcova, V., Gentile, M. A., Harada, S., and Jepsen, K. J. (2004). Targeted overexpression of androgen receptor in osteoblasts: unexpected complex bone phenotype in growing animals. Endocrinology 145, 3507–3522. 68. Kato, S., Takeyama, K., Kitanaka, S., Murayama, A., Sekine, K., and Yoshizawa, T. (1999). In vivo function of VDR in gene expressionVDR knock-out mice. J. Steroid Biochem. Mol. Biol. 69, 247–251. 69. Amling, M., Priemel, M., Holzmann, T., Chapin, K., Rueger, J. M., Baron, R., and Demay, M. B. (1999). Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: formal histomorphometric and biomechanical analyses. Endocrinology 140, 4982–4987. 70. Donohue, M. M., and Demay, M. B. (2002). Rickets in VDR null mice is secondary to decreased apoptosis of hypertrophic chondrocytes. Endocrinology 143, 3691–3694. 71. Gardiner, E. M., Baldock, P. A., Thomas, G. P., Sims, N. A., Henderson, N. K., Hollis, B., White, C. P., Sunn, K. L., Morrison, N. A., Walsh, W. R., and Eisman, J. A. (2000). Increased formation and decreased resorption of bone in mice with elevated vitamin D receptor in mature cells of the osteoblastic lineage. Faseb. J. 14, 1908–1916. 72. Harrison, J. R., Woitge, H. W., and Kream, B. E. (2002). Genetic approaches to determine the role of glucocorticoid signaling in osteoblasts. Endocrine 17, 37–42. 73. Sher, L. B., Woitge, H. W., Adams, D. J., Gronowicz, G. A., Krozowski, Z., Harrison, J. R., and Kream, B. E. (2004). Transgenic expression of 11beta-hydroxysteroid dehydrogenase type 2 in osteoblasts reveals an anabolic role for endogenous glucocorticoids in bone. Endocrinology 145, 922–929. 74. O’Brien, C. A., Jia, D., Plotkin, L. I., Bellido, T., Powers, C. C., Stewart, S. A., Manolagas, S. C., and Weinstein, R. S. (2004). Glucocorticoids act directly on osteoblasts and osteocytes to induce their apoptosis and reduce bone formation and strength. Endocrinology 145, 1835–1841. 75. Krozowski, Z., Li, K. X., Koyama, K., Smith, R. E., Obeyesekere, V. R., Stein-Oakley, A., Sasano, H., Coulter, C., Cole, T., and Sheppard, K. E. (1999). The type I and type II 11beta-hydroxysteroid dehydrogenase enzymes. J. Steroid Biochem. Mol. Biol. 69, 391–401. 76. Kellendonk, C., Tronche, F., Reichardt, H. M., and Schutz, G. (1999). Mutagenesis of the glucocorticoid receptor in mice. J. Steroid Biochem. Mol. Biol. 69, 253–259. 77. Eijken, M., Hewison, M., Cooper, M. S., de Jong, F. H., Chiba, H., Stewart, P. M., Uitterlinden, A. G., Pols, H. A., and van Leeuwen, J. P. (2005). 11beta-Hydroxysteroid dehydrogenase expression and glucocorticoid synthesis are directed by a molecular switch during osteoblast differentiation. Mol. Endocrinol. 19, 621–631. 78. Komori, T., Yagi, H., Nomura, S., Yamaguchi, A., Sasaki, K., Deguchi, K., Shimizu, Y., Bronson, R. T., Gao, Y. H., Inada, M., Sato, M., Okamoto, R., Kitamura, Y., Yoshiki, S., and Kishimoto, T. (1997). Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89, 755–764. 79. Ducy, P., Zhang, R., Geoffroy, V., Ridall, A. L., and Karsenty, G. (1997). Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89, 747–754. 80. Mundlos, S., Otto, F., Mundlos, C., Mulliken, J. B., Aylsworth, A. S., Albright, S., Lindhout, D., Cole, W. G., Henn, W., Knoll, J. H., Owen, M. J., Mertelsmann, R., Zabel, B. U., and Olsen, B. R. (1997). Mutations involving the transcription factor CBFA1 cause cleidocranial dysplasia. Cell 89, 773–779.
468 81. Ducy, P., Starbuck, M., Priemel, M., Shen, J., Pinero, G., Geoffroy, V., Amling, M., and Karsenty, G. (1999). A Cbfa1-dependent genetic pathway controls bone formation beyond embryonic development. Genes Dev. 13, 1025–1036. 82. Jochum, W., David, J. P., Elliott, C., Wutz, A., Plenk, H., Jr., Matsuo, K., and Wagner, E. F. (2000). Increased bone formation and osteosclerosis in mice overexpressing the transcription factor Fra-1. Nat. Med. 6, 980–984. 83. Sabatakos, G., Sims, N. A., Chen, J., Aoki, K., Kelz, M. B., Amling, M., Bouali, Y., Mukhopadhyay, K., Ford, K., Nestler, E. J., and Baron, R. (2000). Overexpression of DeltaFosB transcription factor(s) increases bone formation and inhibits adipogenesis. Nat. Med. 6, 985–990. 84. Sims, N. A., Sabatakos, G., Chen, J. S., Kelz, M. B., Nestler, E. J., and Baron, R. (2002). Regulating DeltaFosB expression in adult Tet-OffDeltaFosB transgenic mice alters bone formation and bone mass. Bone 30, 32–39. 85. Kveiborg, M., Chiusaroli, R., Sims, N. A., Wu, M., Sabatakos, G., Horne, W. C., and Baron, R. (2002). The increased bone mass in deltaFosB transgenic mice is independent of circulating leptin levels. Endocrinology 143, 4304–4309. 86. Kveiborg, M., Sabatakos, G., Chiusaroli, R., Wu, M., Philbrick, W. M., Horne, W. C., and Baron, R. (2004). DeltaFosB induces osteosclerosis and decreases adipogenesis by two independent cell-autonomous mechanisms. Mol. Cell Biol. 24, 2820–2830. 87. Akune, T., Ohba, S., Kamekura, S., Yamaguchi, M., Chung, U. I., Kubota, N., Terauchi, Y., Harada, Y., Azuma, Y., Nakamura, K., Kadowaki, T., and Kawaguchi, H. (2004). PPARgamma insufficiency enhances osteogenesis through osteoblast formation from bone marrow progenitors. J. Clin. Invest. 113, 846–855. 88. Cock, T. A., Back, J., Elefteriou, F., Karsenty, G., Kastner, P., Chan, S., and Auwerx, J. (2004). Enhanced bone formation in lipodystrophic PPARgamma(hyp/hyp) mice relocates haematopoiesis to the spleen. EMBO Rep. 5, 1007–1012. 89. Harrison, J. R., Huang, Y. F., Wilson, K. A., Kelly, P. L., Adams, D. J., Gronowicz, G. A., and Clark, S. H. (2005). Col1a1 promoter-targeted expression of p20 CCAAT enhancer-binding protein beta (C/EBPbeta), a truncated C/EBPbeta isoform, causes osteopenia in transgenic mice. J. Biol. Chem. 280, 8117–8124.
BARBARA E. KREAM AND JOHN R. HARRISON 90. Long, F., Schipani, E., Asahara, H., Kronenberg, H., and Montminy, M. (2001). The CREB family of activators is required for endochondral bone development. Development 128, 541–550. 91. Molina, C. A., Foulkes, N. S., Lalli, E., and Sassone-Corsi, P. (1993). Inducibility and negative autoregulation of CREM: an alternative promoter directs the expression of ICER, an early response repressor. Cell 75, 875–886. 92. Tetradis, S., Nervina, J. M., Nemoto, K., and Kream, B. E. (1998). Parathyroid hormone induces expression of the inducible cAMP early repressor in osteoblastic MC3T3-E1 cells and mouse calvariae [In Process Citation]. J. Bone Miner. Res. 13, 1846–1851. 93. Yang, X., Matsuda, K., Bialek, P., Jacquot, S., Masuoka, H. C., Schinke, T., Li, L., Brancorsini, S., Sassone-Corsi, P., Townes, T. M., Hanauer, A., and Karsenty, G. (2004). ATF4 is a substrate of RSK2 and an essential regulator of osteoblast biology; implication for Coffin-Lowry Syndrome. Cell 117, 387–398. 94. Elefteriou, F., Ahn, J. D., Takeda, S., Starbuck, M., Yang, X., Liu, X., Kondo, H., Richards, W. G., Bannon, T. W., Noda, M., Clement, K., Vaisse, C., and Karsenty, G. (2005). Leptin regulation of bone resorption by the sympathetic nervous system and CART. Nature 434, 514–520. 95. Hadjantonakis, A. K., Dickinson, M. E., Fraser, S. E., and Papaioannou, V. E. (2003). Technicolour transgenics: imaging tools for functional genomics in the mouse. Nat. Rev. Genet. 4, 613–625. 96. Sparwasser, T., Gong, S., Li, J. Y., and Eberl, G. (2004). General method for the modification of different BAC types and the rapid generation of BAC transgenic mice. Genesis 38, 39–50. 97. Copeland, N. G., Jenkins, N. A., and Court, D. L. (2001). Recombineering: a powerful new tool for mouse functional genomics. Nat. Rev. Genet. 2, 769–779. 98. Carpenter, A. E., and Sabatini, D. M. (2004). Systematic genomewide screens of gene function. Nat. Rev. Genet. 5, 11–22. 99. Voinnet, O. (2005). Induction and suppression of RNA silencing: insights from viral infections. Nat. Rev. Genet. 6, 206–220. 100. Shashikant, C. S., and Ruddle, F. H. (2003). Impact of transgenic technologies on functional genomics. Curr. Issues Mol. Biol. 5, 75–98.
Part III
Markers of Bone and Cartilage Metabolism
This Page Intentionally Left Blank
Chapter 28
The Role of Genetic Variation in Osteoporosis André G. Uitterlinden, Joyce B.J. van Meurs, Fernando Rivadeneira, Johannes P.T.M.van Leeuwen, Huibert A.P. Pols
I. II. III. IV. V.
Department of Internal Medicine, Department of Epidemiology & Biostatistics, Erasmus Medical Centre, Rotterdam, The Netherlands
Abstract Osteoporosis has Genetic Influences Genome-wide Approaches to Find the Genes Association Analysis of Candidate Gene Polymorphisms Haplotypes
VI. Meta-analyses VII. Osteoporosis Candidate Genes: Collagen Type Iα1 and the Vitamin D Receptor VIII. Summary References
I. ABSTRACT
association analysis has much better possibilities, as is illustrated by the successful identification of risk alleles for several complex diseases. Candidate gene association analysis followed by replication and prospective multicentered meta-analysis, is currently the best way forward to identify genetic markers for complex traits, such as osteoporosis. To accomplish this, we need large (global) collaborative studies using standardized methodology and definitions, to quantify by meta-analysis the subtle effects of the responsible gene variants and assess heterogeneity in these associations between populations.
Over the past decades epidemiological research of so-called “complex” diseases, i.e., common age-related disorders such as cancer, cardiovascular disease, diabetes, and osteoporosis, has identified anthropometric, behavioral, and serum parameters as risk factors. Recently, genetic polymorphisms have gained considerable interest, propelled by the Human Genome Project and its sequela that have identified most genes and uncovered a plethora of polymorphic variants, some of which embody the genetic risk factors. In all fields of complex disease genetics (including osteoporosis) progress in identifying these genetic factors has been hampered by often controversial results. Because of the small effect size for each individual risk polymorphism, this is mostly due to low statistical power and limitations of analytical methods. Genome-wide scanning approaches can be used to find the responsible genes. It is by now clear that linkage analysis is not suitable for this, but genome-wide Dynamics of Bone and Cartilage Metabolism
II. OSTEOPOROSIS HAS GENETIC INFLUENCES Certain aspects of osteoporosis have been found to have strong genetic influences. This can be derived, for example, from genetic epidemiological analyses which showed that, 471
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
472 in women, a maternal family history of fracture is positively related to fracture risk [1, 2]. Most evidence, however, has come from twin studies on bone mineral density (BMD) [3–5]. For BMD the heritability has been estimated to be high: 50–80% [3–5]. Thus, although twin studies can overestimate the heritability, a considerable part of the variance in BMD values might be explained by genetic factors while the remaining part could be due to environmental factors and to gene–environment interactions. This also implicates that there are “bone density” genes, variants of which will result in BMD levels that are different between individuals. These genetic differences can become apparent in different ways, for example, as peak BMD or as differences in the rates of bone loss at advanced age. While this notion has resulted in much attention being paid to the genetics of BMD in the field of osteoporosis, it is likely that this attention is also due simply to the widespread availability of devices to measure BMD. At the same time it is important to realize that (low) BMD is but one of many risk factors for osteoporotic fracture, the clinically most relevant endpoint of the disease. In addition, several other phenotypes related to osteoporosis have been shown to have heritable components including biochemical markers, heel ultrasound measurements, and skeletal geometry. Heritability estimates of fracture risk have been – understandably – much more limited, due to the scarcity of good studies allowing somewhat precise estimates. Collecting large collections of related subjects with accurate standardized fracture data is notoriously difficult in view of the advanced age at which they occur. While documenting a fracture event is now possible in several longitudinal studies, excluding a fracture event in those who report no fracture is more difficult because they could still suffer a fracture later in life. One option to overcome this might be to take controls which are (much) older. In the case of hip fracture patients (with a mean age of 80 years) this would require control subjects of 90–100 years. It is questionable whether such healthy survivors are proper controls for fracture cases. In general, studies on fracture have shown low heritability in the range of 20–50% [6–9]. Initial studies on Colles’ fracture of the wrist in 2471 white midwestern USA proband women aged 65 years, 3803 sisters of the probands, and their mothers, reported a heritability of 25% [6]. A similar estimate for osteoporotic fractures was derived from data on 2308 Finnish monozygotic twins and 5241 dizygotic twins [7, 8]. Andrew et al. [9] recently studied 6570 white healthy UK female volunteer twins between 18 and 80 years of age, and identified and validated 220 nontraumatic wrist fracture cases. They estimated a heritability of 54% for the genetic contribution to liability of wrist fracture in
ANDRÉ G. UITTERLINDEN ET AL.
these women. Interestingly, while BMD was also highly heritable, the statistical models showed very little overlap of shared genes between the two traits in this study. An additional difficulty in these studies is, of course, the multifactorial and age-related etiology of fracture. This could implicate that perhaps it is not so much the heritability of bone strength which underlies the fracture heritability, but rather, for example, heritability of falls which by themselves also appear to have a heritable component [10]. Finally, it can be assumed that expression of genetic predispositions will not be constant during life, further complicating a precise estimate of heritability of (a certain type of) fracture. While it might be difficult to demonstrate fracture risk is heritable, one can also argue that it follows from simple logic reasoning that aspects of osteoporosis, including fracture risk, must have a genetic influence. We know that DNA is the blueprint of life, and that the genotype differs between individuals, and that phenotypes differ between individuals. Thus, the difficulties in demonstrating heritability of fracture risk are probably due to limitations of our methods and approaches of measuring it. The heritability estimates of osteoporosis indicate a considerable influence of environmental factors which can modify the effect of genetic predisposition. Gene– environment interactions in this respect include diet, exercise, and exposure to sunlight (for vitamin D metabolism), for example. While genetic predisposition itself (in the form of DNA sequence variations) will be constant during life, environmental factors tend to change during the different periods of life, resulting in different “expression levels” of the genetic susceptibility. Aging is associated with a general functional decline resulting in, for example, less exercise, less time spent outdoors, changes in diet, etc. This can result in particular genetic susceptibilities being revealed only later on in life after a period when they went unnoticed due to sufficient exposure to one or more environmental factors. Taking all this into account it becomes evident that osteoporosis is, not very surprisingly, considered a truly “complex” genetic trait. This complex character is shared with other common and often age-related traits with genetic influences such as diabetes, schizophrenia, Alzheimer’s disease, osteoarthritis, cancer, etc. “Complex” means that a trait is multifactorial as well as multigenic. Thus, genetic risk factors (i.e., certain alleles or gene variants) will be transmitted from one generation to the next, but the expression of these genotype factors in the final phenotype will be dependent on interaction with other gene variants and with environmental factors. Given that the Human Genome Project has now resulted in the identification of nearly all genes in the
473
Chapter 28 Genetic Variation in Osteoporosis Table I. Allele
Haplotype kbp Linkage
Linkage disequilibrium Locus Mbp Microsatellite
Minisatellite
Mutation Polymorphism QTL RFLP SNP VNTR
Brief Glossary of Genetic Terms One of several alternative forms of a DNA sequence at a specific chromosomal location (locus). At each autosomal chromosomal locus in a cell two alleles are present, one inherited from the mother, the other from the father A series of alleles found at linked loci on a single chromosome (phase) Kilobasepairs (1 × 103 bp) The tendency of DNA sequences to be inherited together as a consequence of their close proximity on a chromosome Nonrandom association of alleles at linked loci A unique chromosomal location defining the position of a particular DNA sequence Megabasepairs (1 × 106 bp) A locus consisting of tandemly repetitive sequence units the size of which is (arbitrarily) defined as 1–5 bp A locus consisting of tandemly repetitive sequence units the size of which is (arbitrarily) defined as 6 bp or more An alteration in the DNA sequence The existence of two or more alleles at a frequency of at least 1% in the population Quantitative trait locus; a gene that influences quantitative variation in a trait Restriction fragment length polymorphism Single Nucleotide Polymorphism Variable number of tandem repeats; a polymorphic micro- or minisatellite
human genome, it is not very surprising that most attention in the analysis of gene–environment interactions has gone to the genes, also referred to as the “genocentric” approach. The idea behind this is that, once we know which gene variants are involved, it will be more straightforward to analyze the contribution of environmental factors and their interplay with genetic factors. Table I lists a few terms frequently used when discussing genetics of complex diseases.
III. GENOME-WIDE APPROACHES TO FIND THE GENES To identify and disentangle the genetic factors underlying risk for osteoporosis, basically two approaches can be applied: the “genome-wide analysis” approach, which is free of any hypothesis regarding which gene(s) are involved, and the “candidate gene” approach, which requires an a priori hypothesis as to which gene is involved (see Fig. 1). The idea was and is that genome-wide approaches are preferred because they will identify true and major genetic effects (the “low hanging fruit”) while candidate gene approaches are prone to heavy bias and thus not to be pursued with great priority. While genome-wide approaches have of course great appeal the results obtained so far have been (very) disappointing. This is mainly due to methodological limitations of linkage analysis as indicated below. However, novel and much better possibilities are now offered by the so-called genome-wide association strategies, which are also discussed below.
Figure 1 Methods of dissection of a complex disease/trait to identify the “risk” alleles of candidate genes that explain the genetic contribution to the trait. Resolution refers to the size of the genomic area that can be identified by the approach.
474
ANDRÉ G. UITTERLINDEN ET AL.
In the meantime, several of the more classical candidate genes have already been analyzed to establish what the role of polymorphisms in these genes is in conferring osteoporosis risk. Even in the absence of such genome-wide scans, this is a valid approach to determine their contribution to the genetic risk for osteoporosis. Indeed, candidate gene analyses have identified genetic risk factors for osteoporosis, albeit of modest effect size. In addition, the outcome of any genome-wide analysis is the subsequent study of a particular candidate gene, so this approach will be among us for the coming years in any case. A paradigm in human genetics is already for a long time to scan the whole genome, free of any hypothesis and to look for co-segregation of the disease gene(s) with a DNA marker, so-called linkage. With the advent of the Human Genome Project a multitude of single nucleotide polymorphism (SNP) markers has now become available which, coupled with the discovery of so-called “haplotype blocks”, in the genome, has led to a new genome-wide approach based on the old-fashioned association analysis in cases and controls: the genome-wide association analysis. Thus, two genome-wide approaches can now be distinguished: those based on linkage analysis and those based on association analysis.
A. Genome-wide Linkage Analysis
of relatives (siblings, pedigrees, etc.) are genotyped for hundreds of DNA markers (mostly microsatellites but now also thousands of SNPs can be used, e.g., the Affymetrix 10K SNP chip) evenly spread over the genome. Most genome searches focus on humans, although several mice genome searches have also been performed. Such genome searches are based on the assumption that relatives who share a certain phenotype will also share one or more chromosomal areas, identical-by-descent, containing one or more gene variants causing (to a certain extent) the phenotype of interest (e.g., low BMD). The gene is then said to be linked with the DNA marker used to “flag” a certain chromosomal region, but this area is usually several million base pairs in size. Upon positive linkage, subsequent research will then have to analyze dozens of genes in the chromosomal area to determine which one(s) is (are) the one(s) involved in bone metabolism, and then identify the particular sequence variant in that (those) gene(s) giving rise to (aspects of) osteoporosis. Up to the stage of finding linkage to one or more chromosomal regions this approach has been quite successful in the past and linkage results from several genome searches have indeed been published (e.g., reference [14]). However, so far the subsequent step to identify the gene variant causing the linkage peak has been difficult and has not resulted in many – if any – “osteoporosis risk genes”. Why is that?
Finding the responsible gene for monogenic disorders (caused by rare mutations in a single gene) is a straightforward routine exercise for specialized laboratories. This is based on linkage analysis in pedigrees (in which the disease is segregating according to Mendelian laws) whereby a standardized set of hundreds of well-characterized DNA markers are analyzed for co-segregation with a phenotypic endpoint. An example of such an approach is the identification of LRP5 gene mutations being responsible for osteoporosis pseudoglioma as well as for a trait called High Bone Mass [11, 12]. Many other examples in the bone field exist and so for single-gene diseases this approach works very well to identify “bone genes”. However, the complex (and nonMendelian) character of osteoporosis makes it quite resistant to the methods of analysis which in the past decades have worked so well for the monogenic diseases. Therefore different and often more cumbersome approaches have to be applied which can be broadly defined as top–down and bottom–up approaches (see, e.g. reference [13]). In top–down approaches whole-genome searches are performed which indicate which chromosomal areas might contain osteoporosis genes. For linkage analysis hundreds
1. Weak statistical linkage evidence. It has proven difficult to find statistically significant linkage with LOD scores above 3.6 for genome-wide significance. Typically only “suggestive” linkage is found with LOD scores of 1–3. This indicates that there are not a few major genes for osteoporosis but rather many subtle genes. 2. Lack of replication of linkage peaks. There is hardly any single region which has been identified convincingly (and statistically significant) by more than one genome search (with perhaps some possible exceptions such as 1p36). Replication has also proven difficult because the families/pedigrees/sibling pairs between different genome searches have differences in ethnicity, environmental factors, gender, age, etc. 3. Lack of power to find a gene. The effects per polymorphism are too weak to be detected with the typical number of sibling pairs available. It has therefore been difficult to go beyond “linkage” and to demonstrate that a certain gene variant is causing the linkage peak observed in the genome search. Chromosomal regions showing linkage are typically 1–10 million base pairs wide containing dozens of candidate genes. In these candidate genes hundreds of polymorphisms occur, some of which are organized in linkage disequilibrium blocks of 10–20 kb,
475
Chapter 28 Genetic Variation in Osteoporosis
making it virtually impossible (by statistical genetics alone) to pinpoint the causative variant using the linkage design. 4. Choice of endpoint. Most genome searches have focused on BMD as an endpoint. BMD, however, explains only a part of osteoporotic fracture risk. It is difficult to “switch” between major outcomes during or after the study because the families/sib-pairs are selected on the basis of such an endpoint. It is also noteworthy that all linkage scans have identified few linkage peaks, suggesting few genes to explain the genetics of osteoporosis. It is by now widely assumed however, that many, maybe hundreds of gene variants are implicated. Given the very limited population-attributable risk of the claimed genes identified so far by this approach, this probably also reflects the very limited power of genome-wide linkage scans. Recently, an example of the identification of an allegedly “major osteoporosis gene” through a genome search was published. It was the identification of BMP2 (20p12.3) as a risk factor for osteoporotic fracture by analysis of Icelandic pedigrees and a Danish cohort by the company Decode from Iceland [15]. Although one might interpret this as proof of the success of the genome search approach, several notions would preclude the following. 1. BMP2 was already well-known as an important gene for bone metabolism for several decades and as such represents a good candidate gene. So far, however, nobody had looked for polymorphisms in this gene in relation to osteoporosis. 2. The effect size of the BMP2 gene variants on fracture risk in the two samples (Icelandic and Danish) is modest and in line with what has been found for other candidate genes. It is rather premature to call this a “major” risk gene for osteoporosis and, especially given the low population frequency of a risk allele (37Ser, f1 million haplotype tagging SNPs [18]. Illumina (www.illumina.com) is using glass arrays which are spotted at high density with very selected SNPs, such as coding SNPs (100 k), and will shortly offer such arrays with haplotype tagging SNPs. The first successful use of a genome-wide association analysis using such high numbers of SNPs was reported by Ozaki et al. [19] who, by means of a large-scale case (n = 94)- control (n = 658) association study using 92 788 gene-based SNPs, identified significant associations between myocardial infarction and two SNPs in LTA (encoding lymphotoxin-alpha): one SNP changed an aminoacid residue from threonine to asparagine (Thr26Asn) while another SNP in intron 1 influenced the transcription level of LTA. More recently, Klein et al. [20] reported a GWA study of 96 cases and 50 controls for polymorphisms associated with age-related macular degeneration (AMD), a major cause of blindness in the elderly. Among 116 204 singlenucleotide polymorphisms genotyped (using Affymetrix chips), a tyrosine–histidine change at amino acid position 402 (T402H) in the complement factor H gene (CFH) was strongly associated with AMD. This polymorphism is in a region of CFH that binds heparin and C-reactive protein. The CFH gene is located on chromosome 1 in a region repeatedly linked to AMD in family-based studies.
476
ANDRÉ G. UITTERLINDEN ET AL.
The relatively low density of SNPs used in these studies, in combination with the limited genetic complexity of diseases studied (of AMD in particular), could explain why only a few associated regions were observed in these studies, while one would expect (many) more gene regions to show up. Indeed ,very recent press releases from Affymetrix (www.microarraybulletin.com; April 2005) indicated similar successes for multiple sclerosis (MS; Cohen et al.), graft vs. host disease (GVHD; Ogawa et al.), and cardiovascular disease (CVD; Salonen et al.), where more cases/ controls were analyzed (1800 for MS, 2000 for GVHD, and with higher density SNP arrays (100–500 K) and more gene regions were found to be associated with the disease (80 for multiple sclerosis, 400 for CVD). However, we will have to await the detailed report in the scientific literature of these studies to know the exact methods used and results found. Nevertheless, these results from GWA studies show the great potential they have for elucidating complex disease and it will only be a matter of time before similar approaches will be reported for osteoporosis. The only current limitation might be that they are quite expensive: they cost roughly 1000 euros per DNA sample. Yet, the same statistical requirements apply as in a given case-control or population-based study of a candidate gene polymorphism. This means that for truly complex traits a minimum of several hundred cases/controls, if not
Figure 2
thousands of subjects have to be studied, thus requiring 1–10 million euros per GWA study. For less complex traits, e.g. those with a few major genes such as might be the case for AMD, this might be less. It should be noted that the size of the haplotype blocks/ chromosomal regions identified through the GWA approach is much smaller (10–50 kb) than what is usually found in genome-wide linkage analyses (which is 1–10 million(!) base pairs). This offers major advantages for the subsequent research. Yet, even when one or more such haplotype blocks are found associated, these blocks need further scrutiny to identify the one or more polymorphisms driving the association and functionality has to be established. So this approach could (also) end up with a candidate gene analysis. After that, the associations have to be replicated in other populations and, finally, meta-analysis can be used to assess heterogeneity and quantify effect size (see later).
IV. ASSOCIATION ANALYSIS OF CANDIDATE GENE POLYMORPHISMS The bottom–up approach to identify genetic risk factors for osteoporosis builds upon biology, i.e., the known involvement of a particular gene in aspects of osteoporosis, e.g., bone metabolism (see Fig. 2). This gene is then referred
A schematic flow-diagram depicts the different steps in a candidate gene polymorphism analysis. On top genome-wide association analysis is indicated that will identify multiple areas across the genome as LD blocks within candidate genes. This is used in concordance with biological evidence based on three independent sources, to implicate a gene in the disease of interest.
477
Chapter 28 Genetic Variation in Osteoporosis
to as a “candidate gene”. The candidacy of such a gene can be established by several lines of evidence. 1. Cell biological and molecular biological experiments, indicating for example bone cell-specific expression of the gene. 2. Animal models in which a gene has been mutated (e.g., natural mouse mutants), overexpressed (transgenic mice), or deleted (knockout mice) and which result in a bone phenotype. 3. Naturally occurring mutations of the human gene resulting in monogenic Mendelian diseases with a bone phenotype. Subsequently, in the candidate gene frequently occurring sequence variants (polymorphisms) have to be identified which supposedly lead to subtle differences in function of the encoded protein. We distinguish mutations from polymorphisms purely on the basis of frequency: polymorphisms occur in at least 1% of the population, mutations in less. Sequence analysis of a “candidate” osteoporosis gene in a number of different individuals will identify sequence variants, but also several databases are now available which contain this information (e.g., NCBI, Celera, HapMap, and several more specialized databases). Some DNA sequence variations will be just polymorphic (anonymous polymorphisms) while others will have consequences for the activity of the protein encoded (functional polymorphisms). These can include, e.g., sequence variations leading to alterations in the amino acid composition of the protein, changes in the 5′ promoter region leading to differences in mRNA expression, and/or polymorphisms in the 3′ region leading to differences in mRNA degradation. Clearly, it depends on the gene how many and what kind of polymorphisms will be present in the population. Some genes will have been, for example, under more evolutionary pressure and will not display much variation. Other genes, however, might be part of a pathway with sufficient redundancy to allow for more genetic variation to occur. Polymorphisms of interest are usually first tested in population-based and/or case-control “association studies”, to evaluate their contribution to the phenotype of interest at the population level. However, association studies do not establish cause and effect; they just show correlation or co-occurrence of one with the other. Table II lists some of the pitfalls in interpreting candidate gene association studies. Yet, it is also important to realize in this respect that it is of uncertain value to test functionality of a certain polymorphism in the absence of an association at the population level. Cause and effect has to be established in truly functional cellular and molecular biological experiments involving,
Table II.
Pitfalls in Genetic Association Studies of Candidate Genes
Epidemiological/statistical 1. Sample size is too small, leading to chance findings 2. Population is biased due to selection, admixture, inbreeding, etc. 3. Environmental factors differ between populations 4. Lack of standardized phenotypes Genetic 1. Allelic heterogeneity: different alleles are associated in different populations 2. Locus heterogeneity: gene effects differ between populations due to genetic drift and founder effect 3. Linkage disequilibrium: one or more adjacent genes are the true susceptibility loci instead of the locus being tested Molecular genetic 1. Lack of standardized genotyping techniques 2. Ill-defined choice of polymorphisms: there is no known functional effect of the polymorphism to provide a direct biological explanation of the association Other 1. Publication bias
e.g. transfection of cell lines with allelic constructs and testing activities of the different alleles. This can occur at different levels of organization (see Fig. 3) and depends on the type of protein analyzed, e.g., enzymes vs. matrix molecules vs. transcription factors. Acknowledging these complexities will remain a challenge, once an association has been observed, to identify the correct test of functionality; and vice versa once functionality has been established, to identify the correct endpoint in an epidemiological study. Because functional polymorphisms lead to meaningful biological differences in function of the encoded “osteoporosis” protein this also makes the interpretation of association analyses using these variants quite straightforward. For example, for functional polymorphisms it is expected that the same allele will be associated with the same phenotype in different populations. This can even be extended to similar associations being present in different ethnic groups, although allele frequencies can of course differ by ethnicity [21]. Out of the three lines of evidence mentioned above, numerous candidate genes have emerged and Table III lists only a few of these. These include “classical” candidate genes such as collagen type I, the vitamin D receptor, and the estrogen receptors, but also more novel genes of which their involvement in bone biology has only recently become known. A good example of the latter is the identification of LRP5 gene mutations being responsible for osteoporosis pseudoglioma as well as for a trait called high bone mass [11, 12]. These studies have identified LRP5 as a candidate gene, but, of course, not established its role as a genetic risk factor for osteoporosis.
478
ANDRÉ G. UITTERLINDEN ET AL.
Figure 3 Depiction of how “functional” DNA polymorphisms might affect physiological processes at different levels of organization, that ultimately result in an association that is seen after many years (for age-related disorders this can be 80 years) of “exposure” to the risk factor.
With this plethora of candidate genes it is difficult to decide where to start. Initially this happened somewhat randomly but current choices are guided by increasing insights in the metabolic pathways in which the genes play a pivotal role. For example, the identification of LRP5 as a candidate gene has put the complete Wnt-signaling pathway on the map as a target. It can therefore now be expected that multiple genes from this pathway will be tested as candidate genes for osteoporosis. The current focus in genetic studies of osteoporosis is quite strongly on common variants which are expected to explain a substantial portion of population variance, simply due to their frequency in the population (10–50%). However, also more rare variants (0.1–10% or even less frequent) can contribute to population variance with stronger effects, and perhaps can play an important role in certain populations but not in others. An example of this was described by Cohen et al. [22] who tested whether rare DNA sequence variants collectively contribute to variation in plasma levels of high-density lipoprotein cholesterol (HDL-C). They sequenced three candidate genes (ABCA1, APOA1, and LCAT) that cause Mendelian forms of low HDL-C levels in individuals from a population-based study. Nonsynonymous sequence variants were significantly more common (16% versus 2%) in individuals with low HDL-C (below the fifth percentile) than in those with high HDL-C (above the 95th percentile). Similar findings were obtained in an independent population, and biochemical studies indicated that most sequence variants in the low HDL-C group were functionally important. Thus, rare alleles with major phenotypic
effects contribute significantly to low plasma HDL-C levels in the general population. Similarly, such rare alleles of bone genes might contribute to variation in BMD and other bone parameters, and even fracture risk in the general population. Thus, when we compare the genome-wide linkage approach to the candidate gene approach, the latter approach is now clearly the more promising one [16–18, 23]. Genome-wide linkage searches are not designed and not statistically powered to detect the many subtle gene effects which underlie osteoporosis following the “common variant–common disease” hypothesis. The genome-wide association analysis seems to be a better alternative, but has not been used so far and has not identified any risk genes yet. Thus, the current way forward is to simply test individual candidate genes to establish what their contribution is to osteoporosis risk. Once that has been established, the interaction or multiplicative effects of several genes will be analyzed and, finally, gene–environment interactions can be studied.
V. HAPLOTYPES From resequencing studies for the HapMap project (www.hapmap.org) it has become evident that, on average, one out of every 300 base pairs is varying in the population. Given an average size of 100 kb of a gene this means there are hundreds of polymorphisms in a given gene. Thus, candidate gene analyses will have to focus on which of the many variant nucleotides are the ones that actually matter.
Table III. Selected Osteoporosis Candidate Genes Gene name Matrix protein molecules Osteocalcin α2HS Glycoprotein Osteopontin Osteonectin Collagen type Iα1 Collagen type Iα2 Matrix-associated enzymes Cathepsin K Alkaline phosphatase Carbonic anhydrase II Matrix metalloproteinase 3 Lysyl oxidase Lysine hydroxylase Lysine hydroxylase 2 Lysine hydroxylase 3 Calciotropic (steroid) hormone/receptors/enzymes Estrogen receptor α Estrogen receptor β Aromatase Androgen receptor Glucocorticoid receptor Vitamin D receptor Vitamin D binding protein B3-adrenergic receptor Peroxisome proliferator-activated receptor-gamma Calcium sensing receptor Calcitonin receptor Parathyroid hormone PTH receptor Epidermal growth factor Gonadotropin releasing hormone 1 Gonadotropin releasing hormone receptor Luteinizing hormone beta peptide LH-choriogonadotropin receptor Growth factors/cytokines/receptors Interleukin-1β receptor antagonist Interleukin-4 Interleukin-6 Transforming growth factor β1 Transforming growth factor β2 Growth hormone Growth hormone receptor Insulin-like growth factor I Insulin-like growth factor I receptor Insulin-like growth factor 2 Insulin-like growth factor II receptor IGF-binding protein 3 Tumor necrosis factor α TNF receptor superfamily/1β Bone morphogenetic protein 2 Bone morphogenetic protein 3 Sclerostin Osteoprotegerin RANK RANKL-ligand/OPG-ligand Wnt-signaling pathway Low density lipoprotein receptor-related protein 5* Homocystein pathway Methylene TetraHydroFolate reductase Cystathionine beta-synthase Methionine synthase reductase Methyltetrahydrofolate-homocysteine s-methyltransferase Thymidylate synthetase Miscellaneous Major histocompatibility complex Apolipoprotein E
Gene symbol
Chromosomal location
BGLAP AHSG SPP1 SPOCK COL1A1 COL1A2
1q25–q31 3q27 4q21–q25 5q31.3–q32 17q21.3–q22.1 7q22.1
CTSK ALPL CA2 MMP3 LOX PLOD PLOD2 PLOD3
1q21 1p36.1–p34 8q22 11q22.3 5q23.3–q31.2 1p36 3q23–q24 7q36
ESR1 ESR2 CYP19 AR GR/NR3C1 VDR DBP/GC ADRB3 PPARG CASR CALCR PTH PTHR1 EGF GNRH1 GNRHR LHB LHCGR
6q25.1 14q23 15q21.1 Xq11 5q31 12q13 19q13.3 8p12–p11.2 3p25 3q21–q24 7q21.3 11p15.3–p15.1 3p22–p21.1 4q25 8p21–p11.2 4q21.2 19q13.32 2p21
IL-1RN IL-4 IL-6 TGFB1 TGFB2 GH1 GHR IGF1 IGF1R IGF2 IGF2R IGFBP3 TNF TNFRG5 BMP2 BMP3 SOST OPG/TNFRSF11B RANK/TNFRSF11A RANKL/TNFSF11
2q14.2 5q31.3 7p21 19q13.2 1q41 17q22–q24 5p13–p12 12q22–q23 15q25–q26 11p15.5 6q26 7p14–p12 6p21.3 1p36.3–p36.2 20p12.3 4p14–q21 17q12–q21 8q24 18q22.1 13q14
LRP5
11q12
MTHFR CBS MTRR MTR TYMS
1p36.3 21q22.3 5p15.3–p15.2 1q43 18p11.32
MHC/HLA APOE
6p21.3 19q13.2
480 That is, which sequence variation is functionally relevant by changing expression levels, changing codons, etc. Given the average size of a gene and the relatively young age of human populations, it can be predicted that several sequence variations “that matter” will co-exist in a gene in a given number of subjects from a study population. A major challenge of fundamental research will therefore be to unravel the functionality of these variations and how they interact with each other within the gene. More recently, it has become clear that some of these neighboring polymorphisms are not independent from each other in genetic terms, that is to say some of them tend to “travel together” in so-called haplotypes. Haplotypes are strings of coupled or linked variants which occur, on average, over a distance of 10–30 kb in the human genome. With polymorphisms occurring in roughly one out of 300 base pairs this means there will be dozens of polymorphisms within these “haplotype blocks”. An important aspect of association analyses in this respect is then to establish which common haplotype alleles are occurring in the candidate gene, which has two important practical consequences. 1. If association is found of a particular allele of an individual polymorphism with a certain phenotype/disease, this can also be explained by an adjacent polymorphism within the haplotype block. Thus, one can never be sure what causes the association until the haplotype structure at that position within the gene has been resolved. 2. When, for example, 20 polymorphisms are located within a haplotype block only a fraction (typically only 30%) has to be genotyped to identify the haplotype alleles. This saves on time and money to perform the association analyses while obtaining maximal information relevant for point 1. Once such haplotype blocks have been identified, it becomes important how they are organized in the gene of interest. A typical gene can have one or several haplotype blocks covering the promoter region, another block covering the coding region and yet another block covering regulatory regions 3′ of the gene. For the functioning of a complete gene in a given cell of a given subject, it is then important to know which combination of haplotype alleles is present in that subject. In Figure 4 a hypothetical example is given, of the functional relevance of gene-wide combinations of genotypes (based on single SNPs or on haplotypes). The figure describes the situation when two subjects have identical genotypes for three adjacent polymorphic sites when analyzed independently. Yet, they differ in their combination of alleles on one chromosome, and this will result in different effects at the cellular level. This example illustrates that the effects of single polymorphisms might
ANDRÉ G. UITTERLINDEN ET AL.
be difficult to interpret when ignoring the polymorphisms in the rest of the haplotype block and the other haplotype blocks in the gene. Alternatively, if an association is driven by a single SNP rather than by a haplotype, the association will be diluted if haplotyes are analyzed. In the case of high-density SNP analyses of a candidate gene both approaches, i.e., haplotypes and single SNP analysis, should therefore be compared to determine the driving force of the strongest association.
VI. META-ANALYSES In the coming years we can expect to see more and more association analyses to be performed of an ever-increasing list of candidate gene polymorphisms. It will therefore be necessary to put all these data in perspective by performing meta-analyses of the individual association analyses. Meta-analysis can quantify the results of various studies on the same topic and by assessing heterogeneity estimate and explain their diversity. Recent evidence indicates that a systematic meta-analysis approach can estimate populationwide effects of genetic risk factors for human disease [24] and that large studies are more conservative in these estimates and should preferably be used [25]. An analysis of 301 studies on genetic associations (on many different diseases) concluded that there are many common variants in the human genome with modest but real effects on common disease risk, and that studies using large samples will be able to convincingly identify such variants [26]. Several meta-analyses of osteoporosis candidate gene polymorphisms have been published in the past few years, in particular for VDR polymorphisms, the COL1A1 Sp1 polymorphism, and ESR1 polymorphisms. Although these have to some extent all confirmed association of these polymorphisms with bone-related endpoints, including fracture, considerable caution must be taken in interpreting these results. For example, these meta-analyses were all based on published data, thereby raising the possibility of publication bias influencing the outcome of the analysis, since some (perhaps mostly negative) studies were not included. Another important problem can be lack of standardization of methods to determine genotype and phenotype in each of the component populations. For example, there are differences in accuracy of genotyping technologies, BMD can be measured in different ways by different machines, and “osteoporotic fracture” is a quite heterogeneous phenotype. In addition, such meta-analysis treats studies to some extent as equal whereby subtle differences might not be properly accounted for, such as for example differences between populations in environmental factors. In addition, some alleles might be quite specific for some
481
Chapter 28 Genetic Variation in Osteoporosis
Figure 4
Hypothetical example of the importance of gene-wide genotype combinations. Three adjacent SNPs in different parts of a gene are shown for two individuals (A and B indicated at the bottom). The subjects A and B have identical genotypes, i.e., they are both heterozygous for all three SNPs. However, they have different allele combinations on the same chromosome (numbered 1–4): 1 + 2 for subject A and 3 + 4 for subject B. The promoter area regulates production of mRNA while the 3′ UTR is involved in degradation of mRNA and their interaction/combined effects regulate the net availability of the mRNA for translation into the protein. In this case the example is shown for a promoter polymorphism which has two alleles + and −, of which the + allele is the high-producer variant in certain target cells. Of the two different 3′ UTR variants + and −, the + is the more stable 3′ UTR resulting in more mRNA being maintained. Hence, a “good” promoter allele and a “good” 3′ UTR allele on the same chromosome, result in more protein being produced. The protein itself can occur in two variants: a less active “risk” form (−) and a more active form (+), and both A and B are again heterozygous for this polymorphism. The combined result of the particular allele combinations is that individual A has less of the “risk” protein than individual B in the target cell. This could not have been predicted by analyzing single SNPs and/or only looking at genotypes of individual SNPs, but is only evident upon analysis of the gene-wide genotype combinations.
populations but not for others (such as the lactose deficiency polymorphism LPH C-13170T) and show specific gene– environment interactions and gene–gene interactions. One might for these reasons therefore still prefer to perform a single, large, well-powered study on candidate gene associations and seek replication, rather than rely on meta-analysis results based on published data. To overcome the potential drawbacks of the classical meta-analysis approach in the field of osteoporosis, the EUsponsored GENOMOS (Genetic Markers for Osteoporosis) consortium attempts to perform prospective meta-analyses of osteoporosis candidate gene polymorphisms using standardized methods of genotyping and phenotyping. The GENOMOS project involves the large-scale study of several candidate gene polymorphisms in relation to osteoporosisrelated outcomes in subjects drawn from several European centers. Its main outcomes are femoral neck and lumbar spine BMD and fractures, and design details are described in the first meta-analysis of individual-level data on the
ESR1 gene performed by the GENOMOS team [27]. The GENOMOS meta-analysis of three polymorphisms in the ESR1 gene (intron 1 polymorphisms XbaI [dbSNP: rs9340799] and PvuII [dbSNP: rs2234693] and promoter TA repeats microsatellite) and haplotypes thereof, among 18 917 individuals in eight European centers, demonstrated no effects on BMD but a modest effect on fracture risk (19–35% risk reduction for XbaI homozygotes), independent of BMD [27]. An important aspect of this study is its prospective multicenter design. This means the genotype data are generated for all centers only after which the association analysis is done, thereby rendering it practically immune to possible publication bias. The targets of the study are polymorphisms for which some a priori evidence for involvement in osteoporosis is present already; it is not designed to be a risk gene-discovery tool and currently therefore cannot assess all genetic diversity across a gene. While fracture has been debated as an endpoint in genetics
482
ANDRÉ G. UITTERLINDEN ET AL.
of osteoporosis studies, this was chosen in the GENOMOS study because it is clinically the most relevant endpoint. Statistical power of the GENOMOS study to detect genetic effects on fracture risk is high with >5000 fractures in the dataset. With such a diverse set of populations included in the GENOMOS study, possible population stratification could be a problem. This is not likely, however, because GENOMOS involves almost exclusively white Caucasians, who – in addition – come from relatively stable populations with little recent immigration/emigration. Indeed, so far the tested allele frequencies for ESR1 [27] and COL1A1 [27a] are remarkably similar between populations supporting the absence of major population stratification. Importantly, functional SNPs are expected to show similar effects across different ethnic groups in spite of the different genetic backgrounds of the ethnic groups. In this respect it has recently been demonstrated that genetic markers for proposed gene–disease associations can vary in frequency across populations, but their biological impact on the risk for common diseases may usually be consistent across traditional “racial” boundaries [21]. Thus, such a meta-analysis approach will identify individual genetic risk factors but it will probably also be instrumental in estimating presence and effect size of genetic interactions (gene–gene) and gene–environment interactions. This approach will be followed for genes in a certain pathway, for which we know that interaction is likely, and can be extended to explore unexpected interactions. However, even with large studies of, e.g., 20 000 subjects, this might be difficult to convincingly demonstrate. This stresses the need for even larger studies.
VII. OSTEOPOROSIS CANDIDATE GENES: COLLAGEN TYPE Ia1 AND THE VITAMIN D RECEPTOR The vitamin D receptor (VDR) is the candidate gene that initiated the molecular genetic studies on osteoporosis in 1994 by the work of Morrison, Eisman et al. [28]. They initially reported anonymous 3′ polymorphisms, detected as BsmI, ApaI, and TaqI RFLPs, to be associated with differences in BMD and claimed that up to 80% of the population variance in BMD could be explained by this single gene. Although some of their results were later retracted [29], it left behind the idea that there could be such a thing as a single osteoporosis gene. For a complex disease such as osteoporosis this is very unlikely and, indeed, by now several large population-based association analyses [30] and including several meta-analyses [31–33] have
indicated that – at most – the VDR variants explain only a few percent of the population variance of BMD, if any. Thus, for prediction of BMD the VDR does not seem to be a marker. Whether it is associated with fracture risk remains to be analyzed by meta-analysis of the scarce published data so far and by the prospective type of metaanalysis as has been discussed above. However, in view of its pleiotropic involvement in a number of biological pathways, the vitamin D receptor polymorphisms have also been analyzed in relation to a number of other diseases. For example, we have shown VDR polymorphisms to be associated with (bony) aspects of osteoarthritis, i.e., osteophytosis [34]. Yet, a major problem with the vitamin D receptor gene polymorphisms is still the lack of insight into the functional consequences of the polymorphisms used so far and which other polymorphisms there are in this gene. Therefore a major effort in our own laboratory is focused on finding additional polymorphic sites but now in functionally relevant areas of the gene, such as the 3′ untranslated region and the 5′ promotor area of this gene. One such functional VDR polymorphisms is the Cdx2 polymorphism in the 1a/1e promoter region of the VDR gene. The Cdx2 polymorphism has been well-characterized by the studies of Arai et al. [35] and Yamamoto et al. [36]. The G to A polymorphism is located in a Cdx2-binding site in the 1e promoter region and this site is suggested to play an important role in intestinal-specific transcription of the VDR gene. As this is the site where the calcium absorption predominantly takes place, the Cdx2 site is thought to influence the vitamin D regulation of calcium absorption. The A allele has been demonstrated to be more “active” than the G-allele by binding the Cdx2 transcription factor more strongly, and by having more transcriptional activity [35]. Thus, the A allele is thought to cause increased VDR expression in the intestine and, thereby, can increase the transcription of calcium transport proteins (such as calbindin 9K and -28K, TRPV5, TRPV6). This could thereby enhance the intestinal absorption of calcium and result in increased BMD. Although this increased BMD has indeed been demonstrated for Japanese women who carry the A-allele [35], this was not found in a much larger study of Dutch Caucasian women [37]. Yet, the A-allele of this polymorphism was indeed found to be associated with decreased fracture risk (as would be expected from having an increased BMD) in the large study of Dutch Caucasian women, but independently of BMD [37]. Therefore, although the functionality of this polymorphism has indeed been convincingly demonstrated, the exact mechanism whereby the A-allele would confer lower risk for fracture has not been elucidated yet and requires further study.
Chapter 28 Genetic Variation in Osteoporosis
Another example of a functional sequence variation in an osteoporosis candidate gene that has quickly shown promising associations, is the G to T substitution in the Sp1binding site in the first intron of the collagen type Iα1 gene. After its discovery and initial association analysis by Grant et al. [38], a large-scale population analysis in the Rotterdam Study showed the T allele to be associated with low BMD and increased fracture risk [39]. Interesting observations in this cohort of 1782 elderly women were the increase with age of the genotype-dependent differences and the fact that the genotype-dependent fracture risk was independent of the differences in BMD. Meanwhile, several other studies have confirmed these observations as is demonstrated in meta-analyses of published studies [40–42], making this a promising osteoporosis candidate gene polymorphism with consistent, albeit modest, effects. The molecular way in which this polymorphism is influencing BMD and fracture risk is becoming more and more clear. Work by Ralston and colleagues [40] has shown that the T-allele has increased affinity for the Sp1-binding factor, has increased mRNA production, increased protein production, and, in biomechanic experiments on bone biopsies, has been shown to be associated with decreased bone strength. It has been suggested that the over-representation of collagen type Iα1 homo-trimers might explain the decreased bone strength although this still has to be proven. The COLIA1 and VDR gene are “classical” osteoporosis candidate genes and indeed have been shown to be associated with real, albeit modest, effects. Therefore there must be other osteoporosis genes carrying risk alleles. An overview of possible candidate genes is given in Table III. These include more classical genes (such as the ESR1 gene) but also more unexpected genes (such as the IL genes) and more novel genes such as SOST, BMP2, and LRP5. Systematic analysis of these pathways, and the candidate genes in them, will provide an ever-increasing clearer picture of the individual components of the genetic risk factors. Association analysis in population-based studies is the most promising way forward in this respect.
VIII. SUMMARY So, in summary, if people were to embark on an association study of a candidate gene to identify genetic markers for osteoporosis, what would be the crucial issues to address? A few suggestions. 1. Take a large population. Bigger is better, to make your initial observations statistically robust. 2. Identify proper endpoints upfront. Fractures are clinically the most relevant but you need substantial
483 numbers to make your finding statistically robust. BMD is only one of the risk factors but it is a continuous trait and gives more statistical power. Population-based studies have the advantage of being able to switch phenotypes during analysis very easily, for case-controls this possibility is very limited. 3. Cover all relevant genetic variation within the gene. Focus on functionally relevant variants within a gene. A clear-cut functional variant can be analyzed in isolation, ignoring the rest of the genetic variation in the gene. However, determine the haplotype structure to understand how the complete gene is functioning. 4. P-values: rather seek replication of your finding. Simple adjustment for multiple testing is regarded as not appropriate (where to start and stop counting?). Rather, formulate a proper a priori hypothesis and seek replication(s) of the observed association in similar populations. 5. Perform a meta-analysis to quantify effect size and assess heterogeneity. Join consortia with your population and datasets to standardize genotype and phenotype definition and estimate effect size of polymorphisms and assess heterogeneity of the association in different populations, preferably by prospective meta-analysis rather than meta-analysis of published data. Although still in its infancy with respect to having clinical implications, the field of genetics of osteoporosis (or any complex disease for that matter) is expected to eventually find applications in three main areas. 1. Prediction of response-to-treatment (pharmacogenetics). Polymorphisms in, e.g., drug-metabolizing enzymes, will result in different efficiencies with which drugs can exert their effect. The same holds true for receptors of hormones and growth factors, analogs of which are currently prescribed as treatment. Genotype analysis can identify those subjects expected to profit most from a particular treatment or exclude those subjects which will suffer more from side-effects (personalized medicine). 2. Identification of subjects-at-risk. Subjects carrying risk alleles are more likely to develop osteoporosis. Genotype analysis will allow to take preventive measures, targeted at the individual at an early stage. 3. Identification of novel disease pathways and therapeutic targets. Such novel pathways can be targeted for drug development and investigated further for key components. So far only one polymorphism is currently being considered as an osteoporosis risk factor (the COL1A1 Sp1 polymorphism) and commercial parties have taken up interest in this genetic marker. Its utility in clinical practice has to be considered with considerable caution, however. For example, analyses in different ethnic populations have
484
ANDRÉ G. UITTERLINDEN ET AL.
shown it to be present mostly in Caucasian subjects [43]. Furthermore, interaction of this variant with another polymorphism (the VDR 3′ variants) has been demonstrated [44]. This latter study also highlights the complex and multigenic nature of osteoporosis. It underlines the need to identify additional osteoporosis risk alleles to better understand how particular genetic markers are expressed and result in a phenotype. Another spin-off of genetic research of osteoporosis is the discovery of new and/or unexpected genes and pathways to be involved in determining, e.g., BMD. A good example is the recent discovery of LRP5 mutations to influence BMD [5, 6] and by this the identification of the Wnt-signaling pathway to be involved in bone metabolism. Such discoveries led to new possibilities to develop drugs to treat osteoporosis. In addition, such genes become candidate osteoporosis risk genes and will be searched for polymorphisms. Risk alleles resulting from such analyses can then be added to the still-growing list of osteoporosis gene variants. Thus, in spite of complicating factors, genetic research will contribute to a further understanding of complex diseases, including osteoporosis. The identification of new genes, or new roles of already-known genes, will allow insights in mechanistic pathways which might help in designing therapeutic protocols. Finally, the description of genetic variation underlying phenotypic variation can be used, in concert with existing easy-to-assess risk factors, in prediction of risk for aspects of osteoporosis. In this respect, novel therapeutic protocols but also insights in gene–environment interactions allow for ways to further improve treatment of patients
7.
8.
9.
10.
11.
12.
13. 14.
15.
16.
17.
References 1. Cummings, S. R., Nevitt, M. C., Browner, W. S., Stone, K., Fox, K. M., Ensrud, K. E., Cauley, J., Black, D., and Vogt, T. M. (1995). Risk factors for hip fracture in white women. New Engl. J. Med. 332, 767–773. 2. Kanis, J. A., Johansson, H., Oden, A., Johnell, O., de Laet, C., Eisman, J. A., McClsokey, E. V., Mellstrom, D., Melton III, L. J., Pols, H. A. P., Reeve, J., Silman, A. J., and Tenenhouse, A. (2004). A family history of fracture and fracture risk: a meta-analysis. Bone 35, 1029–1037. 3. Smith, D. M., Nance, W. E., Kang, K. W., Christian, J. C., and Johnston, C. C. (1973). Genetic factors in determining bone mass. J. Clin. Invest. 52, 2800–2808. 4. Pocock, N. A., Eisman, J. A., Hopper, J. L., Yeates, G. M., Sambrook, P. N., and Ebert, S. (1987). Genetic determinants of bone mass in adults: a twin study. J. Clin. Invest. 80, 706–710. 5. Flicker, L., Hopper, J. L., Rodgers, L., Kaymakci, B., Green, R. M., and Wark, J. D. (1995). Bone density determinants in elderly women: a twin study. J. Bone Miner. Res. 10, 1607–1613. 6. Deng, H. W., Chen, W. M., Recker, S., Stegman, M. R., Li, J. L., Davies, K. M., Zhou, Y., Deng, H., Heaney, R., and Recker, R. R. (2000). Genetic determination of Colles’ fracture and differential bone mass
18.
19.
20.
21.
22.
23.
in women with and without Colles’ fracture. J. Bone Miner. Res. 15, 1243–1252. Kannus, P., Palvanen, M., Kaprio, J., Parkkari, J., and Koskenvuo, M. (1999). Genetic factors and osteoporotic fractures in elderly people: prospective 25 year follow up of a nationwide cohort of elderly Finnish twins. BMJ 319, 1334–1337. MacGregor, A., Snieder, H., and Spector, T. D. (2000). Genetic factors and osteoporotic fractures in elderly people. Twin data support genetic contribution to risk of fracture. BMJ 320, 1669–1670. Andrew, T., Antioniades, L., Scurrah, K. J., Macgregor, A. J., and Spector, T. D. (2005). Risk of wrist fracture in women is heritable and is influenced by genes that are largely independent of those influencing BMD. J. Bone Miner. Res. 20, 67–74. Carmelli, D., Kelly-Hayes, M., Wolf, P. A., Swan, G. E., Jack, L. M., Reed, T., and Guralnik, J. M. (2000). The contribution of genetic influences to measures of lower-extremity function in older male twins. J. Gerontol. A Biol. Sci. Med. Sci. 55, B49–B53. Gong, Y., Slee, R. B., Fukai, N., Rawadi, G., Roman-Roman, S., Reginato, A. M., and 56 others. (2001). LDL Receptor-related protein 5 (LRP5) affects bone accrual and eye development. Cell 107, 513–523. Little, R. D., Carulli, J. P., Del Mastro, R. G., Dupuis, J., Osborne, M., Folz, C., and 29 others. (2002). A mutation in the LDL Receptor related protein 5 gene results in the autosomal dominant high bone mass trait. Am. J. Hum. Genet. 70, 11–19. Lander, E. S., and Schork, N. J. (1994). Genetic dissection of complex traits. Science 265, 2037–2048. Ralston, S. H., Galwey, N., MacKay, I., Albagha, O. M., Cardon, L., Compston, J. E., Cooper, C., Duncan, E., Keen, R., Langdahl, B., McLellan, A., O’Riordan, J., Pols, H. A., Reid, D. M., Uitterlinden, A. G., Wass, J., and Bennett, S. T. (2005). Loci for regulation of bone mineral density in men and women identified by genome wide linkage scan: the FAMOS study. Hum. Mol. Genet. 14, 943–951. Styrkarsdottir, U., Cazier, J. B., Kong, A., Rolfsson, O., Larsen, H., Bjarnadottir, E., Johannsdottir, V. D., Sigurdardottir, M. S., Bagger, Y., Christiansen, C., Reynisdottir, I., Grant, S. F., Jonasson, K., Frigge, M. L., Gulcher, J. R., Sigurdsson, G., and Stefansson, K. (2003). Linkage of Osteoporosis to Chromosome 20p12 and Association to BMP2. PLoS Biol. 1(3), E69. Hirschhorn, J. N., and Daly, M. J. (2005). Genome-wide association studies for common diseases and complex traits. Nat. Rev. Genet. 6, 95–108. Wang, W. Y., Barratt, B. J., Clayton, D. G., and Todd, J. A. (2005). Genome-wide association studies: theoretical and practical concerns. Nat. Rev. Genet. 6, 109–118. Hinds, D. A., Stuve, L. L., Nilsen, G. B., Halperin, E., Eskin, E., Ballinger, D. G., Frazer, K. A., and Cox, D. R. (2005). Whole-genome patterns of common DNA variation in three human populations. Science 307, 1072–1079. Ozaki, K., Ohnishi, Y., Iida, A., Sekine, A., Yamada, R., Tsunoda, T., Sato, H., Sato, H., Hori, M., Nakamura, Y., and Tanaka, T. (2002). Functional SNPs in the lymphotoxin-alpha gene that are associated with susceptibility to myocardial infarction. Nat. Genet. 32, 650–654. Klein, R. J., Zeiss, C., Chew, E. Y., Tsai, J. Y., Sackler, R. S., Haynes, C., Henning, A. K., Sangiovanni, J. P., Mane, S. M., Mayne, S. T., Bracken, M. B., Ferris, F. L., Ott, J., Barnstable, C., and Hoh, J. (2005). Complement factor H polymorphism in age-related macular degeneration. Science 308, 385–389. Ioannidis, J. P., Ntzani, E. E., and Trikalinos, T. A. (2004). ‘Racial’ differences in genetic effects for complex diseases. Nat. Genet. 36, 1312–1318. Cohen, J. C., Kiss, R. S., Pertsemlidis, A., Marcel, Y. L., McPherson, R., and Hobbs, H. H. (2004). Multiple rare alleles contribute to low plasma levels of HDL cholesterol. Science 305, 869–872. Risch, N., and Merikangas, K. (1999). The future of genetic studies of complex human diseases. Science 273, 1516–1517.
Chapter 28 Genetic Variation in Osteoporosis 24. Ioannidis, D. G., Ntzani, E. E., Trikalinos, T. A., and ContopoulosIoannidis, D. G. (2001) Replication validity of genetic association studies. Nat. Genet. 29, 306–309. 25. Ioannidis, J. P., Trikalinos, T. A., Ntzani, E. E., and ContopoulosIoannidis, D. G. (2003). Genetic associations in large versus small studies: an empirical assessment. Lancet 361, 567–571. 26. Lohmueller, K. E., Pearce, C. L., Pike, M., Lander, E. S., and Hirschhorn, J. N. (2003) Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat. Genet. 33, 177–182. 27. Ioannidis, J. P., Ralston, S. H., Bennett, S. T., Brandi, M. L., Grinberg, D., Karassa, F. B., Langdahl, B., van Meurs, J. B., Mosekilde, L., Scollen, S., Albagha, O. M., Bustamante, M., Carey, A. H., Dunning, A. M., Enjuanes, A., van Leeuwen, J. P., Mavilia, C., Masi, L., McGuigan, F. E., Nogues, X., Pols, H. A., Reid, D. M., Schuit, S. C., Sherlock, R. E., and Uitterlinden, A. G. (2004) GENOMOS Study. Differential genetic effects of ESR1 gene polymorphisms on osteoporosis outcomes. JAMA 292, 2105–2114. 27a.Ralston, S. H., Uitterlinden, A. G., Brandi, M. L., Balcells, S., Langdahl, B. L., Lips, P., Lorenc, R., Obermayer-Pietsch, B., Scollen, S., Bustamante, M., Husted, L. B., Carey, A. H., Diez-Perez, A., Dunning, A. M., Falchetti, A., Karczmarewicz, E., Kruk, M., van Leeuwen, J. P. T. M., van Meurs, J. B. J., Mangion, J., McGuigan, F. E. A., Mellibovsky, L., del Monte, F., Pols, H. A. P., Reeve, J., Reid, D. M., Renner, W., Rivadeneira, F., van Schoor, N., Sherlock, R. E., Ioannidis, J. P. A., APOSS investigators, EPOS investigators, EPOLOS investigators, FAMOS investigators, LASA investigators, and Rotterdam Study investigators; for the GENOMOS Study. Large-scale evidence for the effect of the COL1A1 Sp1 polymorphism on osteoporosis outcomes: The GENOMOS Study. PloS Medicine 2006 (in press). 28. Morrison, N. A., Qi, J. C., Tokita, A., Kelly, P. J., Crofts, L., Nguyen, T. V., Sambrook, P. N., and Eisman, J. A. (1994). Prediction of bone density from vitamin D receptor alleles. Nature 367, 284–287. 29. Morrison, N. A., Qi, J. C., Tokita, A., Kelly, P. J., Crofts, L., Nguyen, T. V., Sambrook, P. N., and Eisman, J. A. (1997). Prediction of bone density from vitamin D receptor alleles (correction). Nature 387, 106. 30. Uitterlinden, A. G., Pols, H. A. P., Burger, H., Huang, Q., van Daele, P. L. A., van Duijn, C. M., Hofman, A., Birkenhäger, J. C., and van Leeuwen, J. P .T .M. (1996). A large scale population based study of the association of vitamin D receptor gene polymorphisms with bone mineral density. J. Bone Miner. Res. 11, 1242–1248. 31. Cooper, G. S., and Umbach, D. M. (1996). Are vitamin D receptor polymorphisms associated with bone mineral density? A meta-analysis. J. Bone Miner. Res. 11, 1841–1849. 32. Gong, G., Stern, H. S., Cheng, S. C., Fong, N., Mordeson, J., Deng, H. W., and Recker, R. R. (1999). The association of bone mineral density with vitamin D receptor gene polymorphisms. Osteoporos. Int. 9, 55–64.
485 33. Thakkinstian, A., D’Este, C., Eisman, J., Nguyen, T., and Attia, J. (2004). Meta-analysis of molecular association studies: vitamin D receptor gene polymorphisms and BMD as a case study. J. Bone Miner. Res. 19, 419–428. 34. Uitterlinden, A. G., Burger, H., Huang, Q., Odding, E., van Duijn, C. M., Hofman, A., Birkenhäger, J. C., van Leeuwen, J. P. T. M., and Pols, H. A. P. (1997). Vitamin D receptor genotype is associated with osteoarthritis. J. Clin. Invest. 100, 259–263. 35. Arai, H., Miyamoto, K. I., Yoshida, M., Yamamoto, H., Taketani, Y., Morita, K., Kubota, M., Yoshida, S., Ikeda, M., Watabe, F., Kanemasa, Y., and Takeda, E. (2001). The polymorphism in the caudal-related homeodomain protein Cdx-2 binding element in the human vitamin D receptor gene. J. Bone Miner. Res. 16, 1256–1264. 36. Yamamoto, H., Miyamoto, K-I., Bailing, L., Taketani, Y., Kitano, M., Inoue, Y., Morita, K., Pike, J. W., and Takeda, E. (1999). The caudalrelated homeodomain protein Cdx-2 regulates vitamin D receptor gene expression in the small intestine. J. Bone Miner. Res. 14, 240–247. 37. Fang, Y., van Meurs, J. B., Bergink, A. P., Hofman, A., van Duijn, C. M., van Leeuwen, J. P., Pols, H. A., and Uitterlinden, A. G. (2003). Cdx-2 polymorphism in the promoter region of the human vitamin D receptor gene determines susceptibility to fracture in the elderly. J. Bone Miner. Res. 18, 1632–1641. 38. Grant, S. F. A., Reid, D. M., Blake, G., Herd, R., Fogelman, I., and Ralston, S. H. (1996). Reduced bone density and osteoporotic vertebral fracture associated with a polymorphic Sp1 binding site in the collagen type Iα1 gene. Nature Genet. 14, 203–205. 39. Uitterlinden, A. G., Burger, H., Huang, Q., Yue, F., McGuigan, F. E. A., Grant, S. F. A., Hofman, A., van Leeuwen, J. P. T. M., Pols, H. A. P., and Ralston, S. H. (1998). Relation of alleles at the collagen type Iα1 gene to bone density and risk of osteoporotic fractures in postmenopausal women. New Engl. J. Med. 338, 1016–1021. 40. Mann, V., Hobson, E. E., Li, B., Stewart, T. L., Grant, S. F. A., Robins, S. P., Aspden, R. M., and Ralston, S. H. (2001). A COLIA1 Sp1 binding site polymorphism predisposes to osteoporotic fracture by affecting bone density and quality. J. Clin. Invest. 107, 899–907. 41. Efstathiadou, Z., Tsatsoulis, A., and Ioannidis, J. P. A. (2001). Association of collagen Iα1 Sp1 polymorphism with the risk of prevalent fractures: a meta-analysis. J. Bone Miner. Res. 16, 1586–1592. 42. Mann, V., and Ralston, S. H. (2003). Meta-analysis of COL1A1 Sp1 polymorphism in relation to bone mineral density and osteoporotic fracture. Bone 32, 711–717. 43. Beavan, S., Prentice, A., Bakary, D., Yan, L., Cooper, C., and Ralston, S. H. (1998). Polymorphism of the collagen type Iα1 gene and ethnic differences in hip fractures rates. N. Engl. J. Med. 339, 351–352. 44. Uitterlinden, A. G., Weel, A. E. A. M., Burger, H., Fang, Y., van Duijn, C. M., Hofman, A., van Leeuwen, J. P. T. M., and Pols, H. A. P. (2001). Interaction between the vitamin D receptor gene and collagen type Iα1 gene in susceptibility for fracture. J. Bone Miner. Res. 16, 379–385.
This Page Intentionally Left Blank
Chapter 29
Measurement of Calcium, Phosphate and Magnesium Heinrich Schmidt-Gayk
Department of Endocrinology, Laboratory Group, Im Breitspiel 15, 69126 Heidelberg, Germany
I. Measurement of Calcium II. Measurement of Phosphate
III. Measurement of Magnesium References
I. MEASUREMENT OF CALCIUM
the ionized calcium concentration is increased. For each 0.1 decrease in pH, the ionized calcium rises by about 0.05 mmol/L. Fetuin-A prevents extraosseous precipitation of calcium–phosphate complexes. a. Choice of Analyte: Total Calcium or Ionized Calcium For routine purposes in not severely ill patients it is often sufficient to determine the concentration of total calcium together with the concentration of total protein and/or albumin. The concentration of total protein or albumin might be necessary for the calculation of corrected serum calcium. Therefore, the most frequently used methods for the determination of serum or plasma calcium measure total calcium, and less often ionized calcium is measured using an ion-selective electrode (see below). However, the corrections for total protein, albumin, and pH are in many cases poor substitutes for measuring ionized calcium. Therefore, as described below, in severely ill patients, in transfused patients and in patients with an unclear diagnosis, the ionized calcium should be used whenever possible to guide therapy. b. Reference Range The reference range of calcium in serum or plasma is given in Table I.
A. Calcium in Serum 1. INTRODUCTION
Calcium in serum (or plasma) is found in three fractions, namely the ionized fraction (Ca2+), the complexed fraction, and the protein-bound fraction. The sum of the three fractions is the total serum (or plasma) calcium, which is fairly constant in a healthy person. In a normal person, approximately 45–50% is ionized, 10–15% is complexed to anions such as bicarbonate, citrate, sulfate, phosphate, and lactate and 40% is protein-bound (about 32% to albumin, about 8% to globulins) [1]. Some calcium is bound as a calcium–phosphate complex to α2-Heremann-Schmid globulin (α2HSG, Fetuin-A) (see Fig. 1). In alkalosis, hydrogen ions dissociate from albumin and globulins, and calcium binding to albumin and globulins increases. In addition, in alkalosis there is an increase in calcium complex formation. As a result, the concentration of ionized calcium decreases and this may result in clinical symptoms of hypocalcemia, although total plasma calcium concentration is unchanged. On the contrary, in acidosis Dynamics of Bone and Cartilage Metabolism
487
Copyright © 2006 by Academic Press. All rights of reproduction in any form reserved.
488
HEINRICH SCHMIDT-GAYK
Figure 1
The equilibrium between protein-bound, complexed, and ionized calcium in blood is shown. Some calcium is bound as a calcium-phosphate complex to α2-HeremannSchmid globulin (α2HSG, Fetuin-A).
The reference range of ionized calcium in venous blood in mothers is as in adults (1.20 ± 0.035 (mean ± SD)), however, neonates show higher levels (1.48± 0.055 mmol/L, venous cord blood) [3]. Within 24 hours, the ionized calcium levels drop to 1.20 mmol/L. From day 3 to 11, the levels in newborns increase and approach 1.38 mmol/L after day 7 [3]. The so-called physiological hypocalcemia in the first 3 days is no true hypocalcemia, if compared to adults. It is suggested that it takes some days until the parathyroid glands are fully active, as these are suppressed by the fairly high calcium levels at birth.
Table I. Reference Range of Calcium in Serum (or Plasma, Ammonium Heparinate) [1, 2]) Age Total calcium Children
