Doctoral dissertation

To be presented by permission of the Faculty of Medicine of the University of Kuopio

for public examination in Auditorium L2, Canthia building, University of Kuopio,

on Wednesday 3rd June 2009, at 12 noon

Institute of Biomedicine,Department of Anatomy and Institute of Clinical Medicine,

Department of SurgeryUniversity of Kuopio

JYRKI NIEMINEN

Effect of Functional Loading onRemodelling in Canine, and Normal and Collagen Type II Transgenic Murine Bone

JOKAKUOPIO 2009

KUOPION YLIOPISTON JULKAISUJA D. LÄÄKETIEDE 448KUOPIO UNIVERSITY PUBLICATIONS D. MEDICAL SCIENCES 448

Distributor : Kuopio University Library P.O. Box 1627 FI-70211 KUOPIO FINLAND Tel. +358 40 355 3430 Fax +358 17 163 410 www.uku.fi/kirjasto/julkaisutoiminta/julkmyyn.shtml

Series Editors: Professor Esko Alhava, M.D., Ph.D. Institute of Clinical Medicine, Department of Surgery Professor Raimo Sulkava, M.D., Ph.D. School of Public Health and Clinical Nutrition Professor Markku Tammi, M.D., Ph.D. Institute of Biomedicine, Department of Anatomy

Author´s address: Coxa Hospital for Joint Replacement P.O.Box 652 FI-33101 TAMPERE FINLAND E-mail : jyrki .nieminen@coxa.fi

Supervisors: Professor Heikki J . Helminen, M.D., Ph.D. Institute of Biomedicine, Department of Anatomy University of Kuopio

Professor Heikki Kröger, M.D., Ph.D. Department of Orthopaedics and Traumatology Kuopio University Hospital

Docent Kaija Puustjärvi-Sunabacka, M.D., Ph.D. Department of Physical Medicine and Rehabil itation Helsinki University Hospital

Reviewers: Professor Ari Heinonen, Ph.D. Department of Health Sciences University of Jyväskylä

Docent Harri Sievänen, Sc.D. UKK Institute, Tampere, Finland

Opponent: Professor Harri Suominen, Ph.D. Department of Health Sciences University of Jyväskylä

ISBN 978-951-27-1168-0ISBN 978-951-27-1205-2 (PDF)ISSN 1235-0303

KopijyväKuopio 2009Finland

Nieminen, Jyrki. Effect of functional loading on remodelling in canine, and normal and collagen type II transgenic murine bone. Kuopio University Publications D. Medical Sciences 448. 2009. 105 p.ISBN 978-951-27-1168-0ISBN 978-051-27-1205-2 (PDF)ISSN 1235-0303

ABSTRACT

These experiments were carried out to investigate the effects of some potential key factors contributing to the development of osteoporosis (OP) and osteoarthrosis (OA) in animal models. These factors include aging, gender, functional joint loading and unloading, and different types of gene defects that affect the synthesis and assembly of collagen type II. The effects of long-term running on canine bone were studied in 20 young female beagle dogs. Half of them ran on a treadmill for 55 weeks, beginning at the age of 15 weeks. The daily running distance increased progressively up to 40 km/day. The unloading effects on bone were studied in 34 young beagle dogs. The right hind limb of half of them was immobilized by splinting for 11 weeks at the age of 29 weeks. The casts were removed at the age of 40 weeks and the dogs were then remobilized for 50 weeks. The effects of limb malalignment on bone mineral density and remodelling was studied in 44 young female beagle dogs. In 15 dogs, a 30º valgus angulation of the right tibia was surgically created, 15 dogs were sham operated and the rest served as controls. The effect of knockout inactivation of one allele of the Col2a1 gene on bone remodelling and on bone properties after functional loading was studied in 205 C57/BL male and female mice, half of which had access to a running wheel. Also, transgenic mice carrying an internally deleted human type COL2A1 gene were used to study bone development and growth with 100 FVB/N mice. Long-term running exercise resulted in a significant reduction in bone mineral density (BMD) in female beagles. However, biomechanical testing showed no differences in the bone strength. The collagen fibres of bone were organized in a more parallel manner in the runner dogs, although there were no differences in total collagen amount or collagen cross-links. Immobilization of right hind limb for 11 weeks induced a site dependent decrease in BMD of the casted limb in association with a reduction of BMD in the contralateral limb. Bone histomorphometry revealed decreased bone volume, increased trabecular bone formation and resorption. These changes were almost completely reversible after one year of remobilization. Valgus osteotomy of the proximal tibia and the joint malalignment led to an increased turnover in the subchondral bone and susceptibility to OA changes in articular cartilage. The altered biomechanics of the operated limb induced also reduction in the BMD of the proximal femur. Defects in the gene encoding collagen type II induced a marked decrease in synthesis of normal collagen type II and disturbed normal bone growth and remodelling in mice. This led to a reduced bone formation, increased bone resorption, and increased percentage of woven bone when studied with histomorphometry. The transgenic animals had lower BMD and their bones were biomechanically weaker than their wild-type littermates. The BMD did not correlate with the mechanical properties in association with this type of collagen gene defect. In this study, we demonstrated the positive effect of functional loading on bones. Short-term immobilization induced reversible bone loss in growing animals. The important role of collagen type II in bone homeostasis, mechanical integrity, and remodelling was also revealed.

National Library of Medicine Classification: QU 450, QY 60.D6, QY 60.R6, WE 250, WE 348

Medical Subject Headings: Biomechanics; Bone and Bones; Bone Density; Bone Remodeling; Collagen; Collagen Type II; Disease Models, Animal; Dogs; Femur; Genetic Diseases, Inborn; Immobilization; Mice; Mice, Transgenic; Mice, Knockout; Models, Animal; Osteoarthritis; Osteoporosis

Inter nos, haec habui quae dixi

ACKNOWLEDGEMENTS

The present study was carried out in the Department of Biomedicine, Anatomy, University of Kuopio between the years 1993-2008.

I wish to express my deepest gratitude to my supervisor, emeritus Professor Heikki J. Helminen, M.D., Ph.D., the former Head of the Department of Anatomy, University of Kuopio. I admire his broad knowledge of basic biological science and scientific vision. In addition, he never lost his faith in the work despite the major delays occuring during these years. This thesis could not have been conducted without his immeasurable personal contribution.

My special thanks go to my second supervisor Docent Kaija Puustjärvi-Sunabacka, M.D., Ph.D., with whom this work was originally planned, and who as a supervisor of the study introduced me to scientific work. Her guidance and unfailing support made this study possible.

I am grateful for my third supervisor Professor Heikki Kröger, M.D., Ph.D., for helping me to find a new perspective to bone research through dynamic histomorphometry. Without his patient teaching and guidance from the very beginning, motivating discussions during difficulties, and exemplary enthusiasm and devotion to research, this thesis would not exist.

My warm thanks belong to Professor Markku Tammi for providing me with the use of the facilities of the Department of Anatomy and for the advice on the preparation of the manuscripts during these years.

I am very thankful to Professor Esko Alhava, M.D., Ph.D. for valuable advice and constructive criticism, especially during the editorial phase of this work. I wish to thank the late Docent Paavo Karjalainen, Ph.D. for introducing me to the fascinating world of medical physics.

My special thanks go to Professor Juha Tuukkanen, DDS, Ph.D., and Professor Timo Jämsä, Ph.D., for the co-operation with the pQCT and biomechanical analyses of murine bone samples. I also thank Docent Vuokko Kovanen, Ph.D., for the biochemical analyses of collagen. I owe my gratitude to Professor Darwin J. Prockop, M.D., Ph.D., Machiko Arita, M.Sc., and Shi-Wu Li, Ph.D., for providing mouse models for this work.

To my many co-authors and co-workers, Professor Ilkka Kiviranta, M.D., Ph.D., Docent Ilkka Arnala, M.D., Ph.D., Docent Jari Arokoski, M.D., Ph.D., Professor Jukka Jurvelin, Ph.D., Docent Jyrki Parkkinen, M.D., Ph.D., Harri Panula, M.D., Ph.D., Jussi Haapala, M.D., Ph.D., Mika Hyttinen, M.D., Professor Mikko Lammi, Ph.D., Tuomo Lapveteläinen, M.D., Ph.D., Professor Markku Miettinen, M.D., Ph.D., Professor Hartmut H. Malluche, M.D., Ph.D., Ritva Inkinen, Ph.D., Tuomas Räsänen, M.Sc., Janne Sahlman, M.D., Ph.D., I offer sincere thanks for enjoyable co-operation.

I am very indebted to Docent Harri Sievänen, Sc.D., and Professor Ari Heinonen, Ph.D., the official reviewers of this thesis, for their advice and critical reading of the manuscript.

I owe my warm thanks to the entire staff of the Department of Anatomy. Especially, I would like to thank Mrs. Eija Rahunen and Mr. Kari Kotikumpu for their skilful technical assistance with bone samples. Also the invaluable help of Mrs. Eila Koski, RN, with DXA measurements is acknowledged.

I owe my sincere thanks to Ewen MacDonald, D. Pharm., for revising the English language of the manuscript.

I further wish to express my gratitude to all my colleagues and the staff of the Department of Surgery, Mikkeli Central Hospital, the Department of Orthopaedic Surgery, University Hospital of Tampere and

Coxa Hospital for Joint Replacement, Tampere, for their encouraging and patient attitude towards my work. I owe my special thanks to Docent Minna Laitinen, M.D., Ph.D., who has had an essential role in pushing me to finish this never-ending story. I appreciate our friendship, as well as the close collaboration in clinical practise.

I am thankful also for the support of my parents and family during these years. The last, the most, and the greatest thanks go to my dear wife Heta. Your endless patience and understanding made this work possible. I dedicate this book to my beloved daughter Sanni, the sunshine of my life, who has bravely tried to understand my absence from many important moments. You really are my best achievement.

The University of Kuopio, the Coxa Hospital for Joint Replacement, the Finnish Research Foundation of Orthopaedics and Traumatology, the South-Savo Fund of Finnish Cultural Foundation and the Finnish Medical Foundation have financially supported this study. The support of these foundations is gratefully acknowledged.

Pelisalmi, Kangasala, 1st May, 2009

Jyrki Nieminen

ABBREVIATIONS

AIR Area-adjusted integrated retardation value of polarized lightALP Alkaline phosphataseBFR Bone formation rateBGP Bone gla-protein, see osteocalcin BMC Bone mineral contentBMD Bone mineral densityBMP Bone morphogenetic proteinBMU Basic multicellular unitBS Bone surfaceBV Bone volumeCCD Charge-coupled deviceCOL2A1 Human gene encoding for type II procollagenCol2a1 Murine gene encoding for type II procollagenCSA Cross-sectional areaCSMI Cross-sectional moment of inertiaCT Computed tomographyCV CoefficientofvariationDPA Dual-photon absorptiometrydpi Dots per inch DXA Dual-energy X-ray absorptiometryECM Extracellular matrixFE Finite elementFACIT Fibril associated collagens with interrupted triple-helixGAG GlycosaminoglycanHA Hyaluronic acid, hyaluronanHPLC High performace liquid chromatographyHP HydroxyprolineHU HounsfieldunitHZ HeterozygousHZK Heterozygous knockoutMAR Mineral apposition rateMMA MethylmetacrylateMMP Matrix metalloproteinaseMRI Magnetic resonance imagingNCP Non-collagenous proteinns. Notsignificantn/a Not availableOA Osteoarthrosis, osteoarthritisOC OsteocalcinOP OsteoporosisOS Osteoid surfaceOV Osteoid volumePCR Polymerase chain reactionPG ProteoglycanPICP Carboxyterminal propepide of type I procollagenPINP Aminoterminal propepide of type I procollagenPLM Polarized light microscopypQCT Peripheral quantitative computed tomographyPTH Parathyroid hormoneROI Region of interest

SD Standard deviationSE Standard error (of the mean)SPA Single-photon absorptiometrySSI Strength strain indexTG TransgenicTRAP Tartrate-resistant acid phosphataseTV Total bone volumeUS Ultrasoundv(Ct)BMD Volumetric (cortical) bone mineral densitywt Wild type (normal genetics)

LIST OF ORIGINAL PUBLICATIONS

The thesis is based on the following reports, which are referred to in the text by their Roman numerals:

I Puustjärvi K, Nieminen J, Räsanen T, Hyttinen M, Helminen HJ, Kröger H, Huuskonen J, Alhava E, Kovanen V: Do more highly organized collagen fibrils increase bone mechanical strength in loss of mineral density after one-year running training? J Bone Miner Res 14(3):321-329, 1999.

II Nieminen J, Puustjärvi K, Lappalainen R, Laitinen M, Kröger H, Helminen HJ, Haapala J: Reversibilty of immobilization induced osteopenia in dogs. Submitted for publication.

III Panula HE, Nieminen J, Parkkinen JJ, Arnala I, Kröger H, Alhava E: Subchondral bone remodelling increases in early experimental osteoarthrosis in young beagle dogs. Acta Orthop Scand 69(6):627-632, 1998.

IV Nieminen J, Sahlman J, Hirvonen T, Jämsä T, Tuukkanen J, Kovanen V, Kröger H, Jurvelin J, Arita M, Li S-W, Prockop DJ, Hyttinen MM, Helminen HJ, Lapveteläinen T, Puustjärvi K: Abnormal response to physical activity in femurs after heterozygous inactivation of one allele of the Col2a1 gene for type II collagen in mice. Calcif Tissue Int 77(2):104-112, 2005.

V Nieminen J, Sahlman J, Hirvonen T, Lapveteläinen T, Miettinen M, Arnala I, Malluche HH, Helminen HJ: Disturbed synthesis of type II collagen interferes with rate of bone formation and growth and increases bone resorption in transgenic mice. Calcif Tissue Int 82(3):229-237, 2008.

This thesis contains also unpublished data. The original papers in this thesis have been reproduced with the permission of the copyright holders.

Table of Contents

ABBREVIATIONS 9

LIST OF ORIGINAL PUBLICATIONS 11

1 INTRODUCTION 17

2 REVIEW OF THE LITERATURE 19

2.1 Bone biology 192.1.1 Bone development and growth 192.1.2 Bone structure and chemical composition 20 2.1.2.1 Macroscopic structure 20 2.1.2.2 Microscopic structure and bone cells 22 2.1.2.3 Bone minerals and chemical composition 232.1.3 Bone matrix proteins 23 2.1.3.1 Collagens 23 2.1.3.1.1 Fibril-forming collagens (I-III, V, and XI) 24 2.1.3.1.2 Other collagens 27 2.1.3.2 Non-collagenous proteins 282.1.4 Bone modelling and remodelling 312.1.5 Bone turnover and chemical markers of bone formation and resorption 332.1.6 Skeletal adaptation to functional loading 35

2.2 Biomechanical properties of bone 362.2.1 Stress and strain 372.2.2 Stiffness and elastic modulus 38

2.3 Biomechanical testing of bone 382.3.1 Compression and tension 382.3.2 Torsion and bending 39

2.4 Bone densitometry 402.4.1 History of bone densitometry 402.4.2 Dual-energy x-ray absorptiometry 412.4.3 Peripheral quantitative computed tomography 422.4.4 Quantitative magnetic resonance imaging 432.4.5 Bone ultrasound 43

2.5 Osteoporosis 44

2.6 Models of experimental osteoporosis 452.6.1 Effect of exercise on bone 452.6.2 Effect of immobilization on bone 47

2.6.2.1 Disuse osteopenia 47 2.6.2.2 Weightlessness 48 2.6.2.3 Neurectomy or paralysis 482.6.3 Collagen gene defects and bone 492.6.4 Effect of estrogen deficiency on bone 49

2.7 Osteoarthrosis 50

2.8 Animal models of experimental osteoarthrosis 51

3 AIMS OF THE STUDY 52

4 MATERIALS AND METHODS 53

4.1 Animal models 534.1.1 Canine long-term running training (I) 544.1.2 Canine immobilization and remobilization (II) 554.1.3 Canine osteotomy of proximal tibia (III) 564.1.4 Murine heterozygous knockout inactivation of one allele of the Col2a1

gene (IV) 574.1.5 Transgenic mice harbouring truncated human COL2A1 gene (V) 58

4.2 Methods 594.2.1 Bone densitometry 594.2.1.1 DXA (I-IV) 594.2.1.2 QCT and pQCT (I, IV) 594.2.2 Bone mechanical testing (I, IV) 604.2.3 Quantitative X-ray imaging (I, IV-V) 614.2.4 Bone histomorphometry (I-III, V) 624.2.5 Quantitative polarized light microscopy (PLM) (I, IV) 624.2.6 Elemental composition of bone samples (I, II) 634.2.7 Statistical analyses (I-V) 63

5 RESULTS 65

5.1 Effect of long-term training on canine bone (I) 65

5.2 Effects of immobilization and remobilization on canine bone (II) 70

5.3 Effects of proximal tibial osteotomy on canine bone (III) 71

5.4 Effects of heterozygous knockout of one allele of the Col2a1 gene on murine

bone (IV) 72

5.5 Effects of human COL2A1 transgene on murine bone (V) 74

6 DISCUSSION 78

6.1 Effect of functional loading on bone (I) 78

6.2 Effect of immobilization on bone (II) 80

6.3 Effect of proximal tibial osteotomy on bone (III) 81

6.4 Effect of altered or disturbed synthesis of type II collagen on bone (IV, V) 82

6.5 Methodological discussion 83

7 SUMMARY AND CONCLUSIONS 85

REFERENCES 86

LIST OF ORIGINAL PUBLICATIONS 107

17

1 INTRODUCTION

Osteoporosis (OP) together with osteoarthrosis (OA) of joints and spine, associated with ageing,

are an enormous and an ever-growing health problem in the industrialized countries. The diseases cause

pain, loss of capacity to work and reduced musculoskeletal activity and function.

OP is characterized by low bone mass and microarchitectural deterioration of bone tissue, with

a consequent increase in bone fragility and susceptibility to fractures. Although in OP bone loss is a

systemic condition, the most frequent fractures occur at the hips, vertebrae, shoulder and distal forearms.

Hip fracture is the most common cause of mortality and morbidity in the osteoporotic fractures; 12–20%

of hip fracture victims die within the first year after the fracture and more than 50% of the survivors

become incapacitated (Consensus Development Conference 1991, 1993). There were 7 083 hip fractures

in Finland in 2004 (Kannus et al. 2006). The costs of the first post-fracture year have been estimated

to be 14 410 € per patient, with the sum increasing to 35 700 € if the patient needs to be permanently

institutionalized due to an inability to cope with activities of daily living (Nurmi et al. 2003).

OA is another rapidly growing health problem in our aging population of the western world.

Idiopathic OA is a debilitating progressive disease leading to damage of the articular cartilage. OA

affects 60% of men and 70% of women over the age of 65 years with enormous social costs, almost as

much as that of ischemic heart disease (Roach and Tilley 2007). The onset of OA is multifactorial and

the initial pathomechanisms are still poorly understood. In 2006, 8.4% of all the disability pensions in

Finland were granted due to limb OA or rheumatoid arthritis, corresponding to about 21 200 pensioners

(The Social Insurance Institution of Finland 2007). More than 15 800 total joint replacements needs to

be performed in Finland every year, mainly on account of primary OA (Finnish Arthroplasty Register

2008).

It is crucial that we understand better these disorders if we are to find an effective cure for the

diseases, or even to find ways to prevent them. The use of animal models is the one reliable way to

control many environmental, nutritional or activity-related factors when investigating the complex

cascade of different factors that eventually induce the typical appearance of OA or OP. It is known that

18

genetic factors can predispose to OA and OP. Collagen, the major structural macromolecule of bone

and cartilage, may be defective due to genetical reasons and this can cause structural alterations in the

connective tissues leading to the pathogenesis of OA and OP. Genomic studies have revealed the clear

homology between the human and the murine genomes (Waterston et al. 2002). Thus, murine transgenic

models can be considered to provide a new means to study the role of alterations in collagen amounts,

structure or fibril network in association with a variety of musculoskeletal disorders (Aszódi et al. 1998b,

Kuivaniemi et al. 1991, Mann and Ralston 2003, Sokolov et al. 1991).

The objective of this study was to investigate the association of functional loading, and, also

disturbance of collagen structure on bone in the development of OA and OP. The effects of functional

joint loading and immobilization, and the effects of proximal tibial valgus osteotomy, on bone mechanical

properties, architecture and metabolism were studied in canine models. The influence of altered type II

collagen structure or synthesis on OA and OP was studied using transgenic mouse models.

19

2 REVIEW OF THE LITERATURE

2.1 Bone biology

2.1.1 Bone development and growth

Bone is a specialized connective tissue that confers mechanical support to the body and also

represents the sites of muscle attachment. This allows locomotion and mechanical protection of vital

organs and the bone marrow. Finally, bone is a metabolic reservoir for the entire organism of important

ions, especially calcium and phosphate (Baron 1993, Recker 1992).

During embryogenesis, the skeletal tissue in humans and higher vertebrates is developed from the

mesenchymal cells of the mesoderm. The mesenchymal cells can develop into osteoblasts, chondroblasts,

fibroblasts, and other connective tissue cells. By the gestation time of six weeks in humans, some

mesenchymal cells have differentiated into chondrocytes and these cells initiate their protein synthesis

leading to an entire cartilaginous model of the appendicular skeleton being formed (Sadler 1985).

Bone formation can occur via two different pathways: endochondral ossification or

intramembranous ossification. The cartilaginous bone template, formed by chondroblasts and

chondrocytes, ossifies by endochondral ossification (Gerstenfeld and Shapiro 1996, Wuthier 1982).

Undifferentiated mesenchymal cells migrate, condense and differentiate into chondroblasts and

chondrocytes. The cells proliferate and secrete extracellular matrix components and form the cartilaginous

bone primordium. The cartilaginous matrix eventually calcifies. Normally the primary ossification centre

develops in the middle of the bone primordium, the diaphysis. In this process, osteoclasts appear in the

calcifying tissue. These cells are needed to remove and remodel the calcified cartilage. Subsequently,

osteoblasts begin to form the lamellar bone. This process continues toward the epiphyses, i.e. the bone

ends. Endochondral ossification is by far the most common mechanism of bone formation and involves

the vertebral column, pelvis, base of the skull and the long bones. The intramembranous ossification

covers the development of the cranial and facial bones.

20Intramembranous ossification consists of the direct formation of bone from a sheet of mesenchymal

cells or membrane without any intermediate cartilaginous state. Some mesenchymal cells differentiate

directly into bone-forming osteoblasts along the membrane and start synthesizing the collagenous bone

matrix, the osteoid. The collagen fibrils in the newly formed osteoid matrix are arranged randomly. The

bone tissue produced by this process is called woven bone. During the maturation process, the collagen

fibrils of the woven bone are replaced with more organized collagen fibrils, thus forming so-called

lamellar bone.

After birth, secondary ossification centres develop in the epiphyses of long bones. Simultaneously,

the cartilaginous, metaphyseal growth plate is formed between the primary and secondary ossification

centres. The growth plate is a layer of proliferating cells expressing cartilage matrix molecules. This

tissue gradually undergoes transformation to bone tissue as described above. This process allows bone to

elongate longitudinally. The growth period comes to end, when the entire growth plate becomes ossified

and remodeled in early adult life. At this stage the long bones have reached their final length.

2.1.2 Bone structure and chemical composition

2.1.2.1 Macroscopic structure

Bone is a matrix-mineral composite. The matrix and minerals endow bone with its mechanical

properties. The mineral phase provides stiffness to resist compressive forces, and the protein matrix

supplies tensional resistance. The macroscopic appearance of the cut surface of a bone near the epiphyseal

region reveals both trabecular (spongy or cancellous) and dense compact bone. The trabecular bone

consists of intercommunicating trabeculae forming a porous meshwork-like structure. The bones can

be divided into two groups: (i) the flat bones including the skull bones, clavicle and mandible, which

develop through intramembranous ossification, and (ii) the long bones, e.g., tibia, femur, humerus, and

vertebrae that develop through endochondral ossification. In the long bones, three distinct parts can be

displayed (Figure 1). The epiphyses of both ends of long bones are the wide parts beneath the articulating

surfaces. The epiphysis consists primarily of trabecular bone with a thin cortex of compact bone. The

diaphysis consists of a tubular, more or less cylindrical part between the bone ends. The metaphysis

21forms the transitional zone between diaphysis and epiphysis and during the growth period this contains

the epiphysial or growth plate (Gray 1980).

Figure 1. Macroscopic features of sectioned human proximal femur.

The long bones are covered by a thick and dense layer of calcified tissue, the cortex (compact

bone). At the metaphysis and epiphysis, the cortex becomes progressively thinner and the internal bone

space is filled with the trabecular bone. The articulating surfaces at the ends of long bones are covered

with a relatively thin layer of hyaline cartilage. The spaces in the hollow diaphysis, as well as the spaces

between the bone trabeculae in cancellous bone are filled with haematopoietic bone marrow and/or fat

tissue.

With the exception of the articular surface, bone is covered by two distinguishable surfaces,

where the bone is in contact with other tissue types. The endosteal surface is the inner surface facing

22the fatty bone marrow. The surface facing outwards to muscles or soft tissue is the periosteal surface.

These surfaces are lined with osteogenic cells organized in layers, the endosteum and the periosteum,

respectively.

There are structural and functional differences between the compact and trabecular bones. In

all, about 80–90% of the total volume of compact bone is calcified as compared with only 15–25% in

trabecular bone. The endosteal surface consists of 70–85% of the total interface with bone marrow. The

functional differences are a consequence of these structural differences and vice versa. The cortical bone

fulfils mainly the mechanical and protective function while the trabecular bone is more responsible for

the metabolic properties.

2.1.2.2 Microscopic structure and bone cells

The functional unit of the cortical bone is the osteon. Each osteon contains a central Haversian

canal complete with blood vessels, nerves and connective tissue. At the surface of these canals, bone-

resorbing and –forming cells may be seen reflecting bone modelling or remodelling activity within the

unit. The active remodelling units are also referred to as basic multicellular units (BMU) (Malluche and

Faugere 1986).

Cortical and trabecular bones are comprised of the same cells types; osteoprogenitor cells,

osteoblasts, osteocytes, and osteoclasts. Osteoprogenitor cells are mesenchymal, undifferentiated stem

cells of bone, which during bone formation, undergo division and transformation into bone-forming

cells, osteoblasts (Kassem et al. 2008). Osteoblasts form the bone lining cells and are important in bone

formation, producing the osteoid. The osteoid is the uncalcified collagenous bone matrix. Some of the

osteoblasts, approximately 10% in all, are totally entrapped in the surrounding matrix. These cells are

transformed into osteocytes, true bone cells, and thus become the main cellular component of mature

bone (Malluche and Faugere 1986). Osteoclasts are the giant multinucleated cells derived from the

haematopoietic stem cells. Osteoclasts are active in bone resorption (Väänänen and Laitala-Leinonen

2008). Osteocytes are derived from osteoblasts and are embedded deep within the calcified bone matrix.

These cells, however, are in contact with each other, as well as with cell processes from other osteocytes

or bone lining cells (Civitelli 2008, Skerry 2008). Their functions are not completely understood, but

there is an emerging consensus that osteocytes act as mechanosensitive cells translating mechanical

23strain into signals of bone formation or resorption. They also may act as microenviromental sensors

modifying the properties of bone and the magnitude of shear stress in the bone fluid, and as regulators

of phosphate metabolism and bone mineralization (Bonewald 2006, Datta et al. 2008, Fukumoto 2008,

Noble 2008, Teti and Zallone 2008, You et al. 2008).

2.1.2.3 Bone minerals and chemical composition

Bone minerals and inorganic matter account for about 70% of total bone weight, water for 5–8%,

with organic matrix making up the remaining 20–25%. Approximately 95% of the mineral phase is

composed of needle-shaped crystals of hydroxyapatite [Ca10(PO4)6(OH)2] (Gray 1980). These crystals

are impregnated with impurities accounting for the remaining 5% of the inorganic phase. The crystals

are 100 nm in length, 20–40 nm in width and 3–6 nm in thickness (Currey 2002). In general, the

hydroxyapatite crystals are arranged with their long axis in parallel to those of the collagen fibrils and

they lie partly within the fibrils or thicker fibres. In addition to calcium and phosphorus, bone contains

also some trace elements including zinc, fluoride, copper, aluminium, and magnesium (Abbott and Rude

1993, Jonas et al. 1993, Quarles et al. 1990, Rahn et al. 1991, Reis et al. 1988, Root and Diamond Jr.

1993).

2.1.3 Bone matrix proteins

Bone and cartilage contain relatively few cells. A major part of their volume is extracellular

material, which is largely composed of a complex network of macromolecules, forming the extracellular

matrix (ECM). This matrix is composed of a variety of proteins and polysaccharides that are secreted

locally and assembled into an organized meshwork in close association with the surface of the cells

that produced them. ECM proteins of bone can be roughly divided into the collagens and the non-

collagenous proteins.

2.1.3.1 Collagens

Collagens are the most abundant ECM proteins in mammals, accounting for about 30% of the

total protein mass (Burgeson and Nimni 1992). The collagen fibrils and their larger bundles, i.e. fibres,

are flexible and have a remarkably high tensile strength. The importance of these proteins is most obvious

24in connective tissues such as bone, cartilage, ligament, tendon, dermis and dentin, where they provide a

highly organized fibrous matrix. The collagens also confer structural integrity on the capsules, septae and

laminae of the gastrointestinal, cardiovascular, urogenital, respiratory, and nervous system (Cole 1994,

Rahkonen et al. 2003). The collagens are products of a superfamily of closely related genes with several

characteristic features and a wide range of functions (Myllyharju and Kivirikko 2001). A typical collagen

molecule has a long and stiff triple helical structure, in which three collagen polypeptide chains (α

chains) are wound around each other in a rope-like left-handed superhelix. Collagens are distinguished

from other ECM proteins by their triple-helical conformation (Fields 1995) and they are extremely rich

in proline and glycine, both of which are important in the formation and stabilization of the triple-helix.

So far, at least 27 distinct collagen α chains have been identified, each encoded by a separate gene

(Alberts et al. 2002, Mayne and Ala-Kokko 2005). Collagen types I-III, V and XI are classified as fibril-

forming collagens. Another group of collagens includes types IX, XII and XIV, which are assigned to

the fibril-associated collagens (FACIT). These collagens have an interrupted triple-helix. Collagen types

VIII and X appear to form another class of ECM collagens, and they can form regular hexagonal lattice

structures (Thomas et al. 1994). Rather than summarizing the extensive literature on all of these collagen

types, this review will focus on the major collagen types of bone and cartilage.

Collagen accounts for about 2/3 of the dry weight of adult articular cartilage. The supportive

heteropolymeric collagen fibril network in articular cartilage matrix consists of collagens II, IX, and XI.

The fibril diameter varies between 15 and 60 nm. Minor amounts of type III, VI, XII and XIV collagens

have been detected in articular cartilage. The proportion of different collagen types in the fibrils varies

with age. Especially in the young cartilage, the proportion of the type IX and XI collagens can account

for as much as 20% of the total collagens, while in mature cartilage the amount of type II collagen

exceeds 90% (Eyre 2002).

Bone collagen consists almost entirely of type I collagen. The other types are not present in

anything like the levels found in cartilage and soft tissue. However, several FACITs have been detected

in bone, as well as low levels of type III and V collagens (Gehron Robey and Boskey 1996).

2.1.3.1.1 Fibril-forming collagens (I-III, V, and XI)

Collagen type I is the most important collagen in bone, accounting for over 90% of the total bone

25collagen (Bätge et al. 1992). It is a heterotrimer composed of two α1(I) chains and one α2(I) chain. It

forms the major scaffold of virtually all connective tissues with the exception of cartilage. Two genes

encode these polypeptides, COL1A1 and COL1A2, with the corresponding peptide sizes being 18 kb

and 40 kb (Kream et al. 1995). Type I collagen is also found in tendons, intervertebral discs and dentin,

and it can also be detected in the sclera and cornea of the eye, lungs and the skin (Gehron Robey and

Boskey 1996). In different tissues, the collagens are not exactly identical, differing in the extent of the

post-translational modifications. Each triple-helical collagen molecule forms a 300 nm long and 1.4 nm

wide semiflexible rod that spontaneously assembles into microfibrils (Figure 2). In bone, the osteoblasts

are the predominant cells expressing the COL1A1 and COL1A2 genes (Gerstenfeld et al. 1993).

Type II collagen is the predominant protein in cartilage matrix. This collagen was first described

in 1969 (Miller and Matukas 1969). Structurally it is related to the other fibril forming collagens. It

is a homotrimeric molecule consisting of three identical α1(II) chains encoded by the COL2A1 gene,

approximately 30 kb in size (Cole 1994). Like other fibrillar collagens, type II collagen is synthesized

as a procollagen, containing carboxy- and aminoterminal extensions. In the endoplasmic reticulum, the

polypeptide chains undergo several modifications including hydroxylation of many of the proline and

lysine residues and glycosylation of hydroxylysines. The propeptides are cleaved off in the ECM. Type

II collagen is also present in the nucleus pulposus of the intervertebral discs and in the vitreus humour

of the eye (Aszódi et al. 1998a, Savontaus et al. 1997). The chondrocytes are the predominant cells

expressing the COL2A1 gene (Cole 1994, Sandell et al. 1994).

Collagen type III is a delicate thin, widely distributed fibrillar collagen, which is present in dermis,

blood vessels, and fibrous structures of viscera. It usually emerges in heterotypic fibrils in combination

with collagens type I and XI. Type III collagen is synthesized or secreted by fibroblasts and smooth

muscle cells. This protein consists of three identical α1(III) chains. Type III collagen is often considered

to be a fetal collagen, due to its abundant expression in fetal tissues. However, little is known about its

role in the development of skeletal tissues (Liu et al. 1997, Niederreither et al. 1994).

Type V collagen is present in annulus fibrosus of the intervertebral disc. The distribution of this

collagen in skeletal tissues seems to follow the pattern of type I and II collagens (Eyre 1988). Type V

collagen resides within the collagen fibril, suggesting that type V collagen may be involved in the initial

events of fibrillogenesis (Burgeson and Nimni 1992). Type XI collagen is buried in the collagen fibrils

26

Ext

race

llula

r

In

trac

ellu

lar

-OH

OH OHOH

-OH

-OH

-OH

-OH-OH -OH

-O-Gal-GlcGal-O-

HO-

OH OH OH OH

OH OH

OGal

GlcGalO

(Man)nGlcNAc

Figure 2. Biosynthesis of a fibril-forming collagen. Procollagen polypeptide chains are synthesized on the ribosomes of the rough endoplasmic reticulum and secreted into the lumen, where they are modified by hydroxylation of certain proline and lysine residues and glycosylation before chain association and triple helix formation. The procollagen molecules are secreted into the extracellular space where the N-terminal and C-terminal propeptides are cleaved off by specific proteases. The collagen molecules then assemble into fibrils, which are stabilized by the formation of covalent crosslinks. Reproduced from Myllyharju and Kivirikko 2001.

27of cartilage, where it forms a heteropolymer. It eventually copolymerizes with collagen type II and it has

been postulated to provide a mechanism for fibril diameter regulation in cartilaginous tissue (Aszódi et

al. 1998a, Tsumaki et al. 1996).

Collagen fibrils are stabilized by crosslinks. The crosslinks of bone and cartilage collagens

are attributable to the lysine and hydroxylysine residues (Eyre 1988). The total amount of crosslinks

correlates positively with the mechanical properties of bone (Oxlund et al. 1995). It is generally agreed

that the stability of collagen fibres increases with advancing age (Kim et al. 1994). A variety of human

conditions, normal and pathological, are based on the ability or failure of tissues to repair and regenerate

their collagenous mainframe (Burgeson and Nimni 1992).

2.1.3.1.2 Other collagens

Type IV collagen is the main collagen of basal lamina. Type VI collagen is a trimer of three

different chains, α1(VI), α2(VI), and α3(VI). It interacts with a number of other ECM molecules, such as

type I-III collagens, biglycan, decorin, and hyaluronan (HA) (Braghetta et al. 1996). In ECM, the type

VI collagen forms a network of microfibrillar aggregates with the banded fibers of type I collagen. A

high ratio of type VI collagen to type I collagen is found in intervertebral discs. The opposite situation

exists in bone (Braghetta et al. 1996). Type VI collagen is also present in the territorial matrix of hyaline

cartilage. However, its significance in skeletal tissues is uncertain.

Collagen type IX is a 190 nm long heterotrimer of disulfide-bonded α1(IX), α2(IX), and α3(IX)

chains. The α2(IX) chain of the collagen type IX contains a covalently bound glycosaminoglycan (GAG)

chain. This collagen usually co-exists in a periodic fashion along the surface of type II collagen (Wu

et al. 1992). Type IX collagen may be detected in hyaline cartilage, vitreous humour, sclera and neural

retina of the eye, inner ear, and notochordal sheath (Cole 1994). In hyaline cartilage, type IX collagen

seems to be more abundant during development than in adult mature tissue (Eyre 1988).

Collagen type X is a homotrimer of three identical α1(X) chains encoded by the COL10A1 gene.

The gene expression occurs only in the hypertrophic chondrocytes. The role of collagen type X in the

development of skeletal tissue is somewhat putative, as no gross abnormalities in long bone growth or

development have been observed in experimental animals with deficient Col10a1 expression (Paschalis

et al. 1996, Rosati et al. 1994).

28

2.1.3.2 Non-collagenous proteins

Non-collagenous proteins (NCPs) comprise about 10–15% of the total bone protein content.

Several NCPs have been identified in bone, including osteopontin, bone sialoprotein, osteonectin,

osteocalcin, proteoglycans, alkaline phosphatase, bone morphogenetic proteins (BMPs), matrix

metalloproteins and growth factors (Ingram et al. 1993). About one quarter of the amount of NCPs is

exogenously derived, being adsorbed or entrapped into the bone matrix (Termine 1993). These proteins

include various growth factors and albumin.

Osteopontin and bone sialoprotein belong to a family of phosphorylated glycoproteins. They

both contain the tripeptide sequence Arg-Gly-Asp common to cell attachment type proteins that bind

the cell surface receptors called integrins (Shinar et al. 1993). Integrins are expressed in osteoblasts and

osteoclasts and are required in the attachment of these cells to bone (Hughes et al. 1993). Osteopontin

and bone sialoprotein are believed to influence the cell-matrix interactions and the activity of the cells.

They have high affinity for calcium and may therefore regulate the deposition of mineral in the bone

matrix (Gehron Robey et al. 1993, Hultenby et al. 1994, Ingram et al. 1993, Rittling et al. 1998).

Osteonectin is the most abundant NCP produced by bone cells. It is also a phosphorylated

glycoprotein, accounting for about 2% of the total protein of the developing bone in most animal species.

It has a high affinity for binding ionic calcium and hydroxyapatite. It is believed to possess multiple

functions in bone, and appears to be associated with osteoblast growth and/or proliferation and matrix

mineralization (Termine 1993).

Bone contains a number of proteins that are post-translationally modified by vitamin K dependent

enzymes to form the amino acid, γ-carboxy glutamic acid (gla). These proteins include osteocalcin (OC,

previously called bone gla-protein (BGP)) (Gehron Robey et al. 1993). The function of osteocalcin is

described in more detail in Chapter 2.1.5 below.

Proteoglycans (PGs) are macromolecules characterized by the attachment to the protein core of

long chains of repeating disaccharides that are often sulphated, these are termed glycosaminoglycans.

The glycosaminoglycans include chondroitin sulphate, dermatan sulphate, keratan sulphate, and HA

(Forriol and Shapiro 2005). Large aggregating PGs can be found in bone. These include aggrecan

and versican. Also smaller leucine-rich low molecular weight proteoglycans, decorin, biglycan and

29fibromodulin, are distributed throughout most connective tissues (Hardingham and Fosang 1992). The

PGs are important proteins in cartilage, where they bind ions and water and are the major determinants of

the biomechanical properties of hyaline cartilage in addition to type II collagen. The PGs of bone differ

substantially from those present in cartilage, consisting primarily of small glycoconjugates with only

one or two chondroitin sulphate chains. It is proposed that PGs can provide a strain-related stimulus to

the various cells, including osteocytes (Bonewald and Johnson 2008, Skerry et al. 1988, Tarbell et al.

2005).

Alkaline phosphatase (ALP) is not a typical matrix protein. However, a few reports have been

published, where ALP activity has been isolated from mineralized matrix extracts (Gehron Robey and

Boskey 1996, Sakano et al. 1993). The specific function of ALP in bone has been debated for more than

60 years. As knowledge of the structure and function of ALP has increased greatly in recent years, the

enzyme has been found to play a crucial role in bone mineralization (Hui and Tenenbaum 1998, Millán

2006).

Bone and cartilage tissues are rich in different growth factors, which participate in the regulation

of skeletal development and growth. Bone morphogenetic proteins (BMPs) belong to the TGF-β

superfamily and though they are known to have important function in the differentiation of bone cells

their exact role in bone remodelling and healing is still a matter of debate (Dimitriou et al. 2006, Elima

1993, Lind et al. 1996).

Matrix metalloproteinases (MMPs) are a large family of Ca2+-requiring zinc metalloendopeptidases

that collectively degrade most ECM components (Hayakawa 2002). They play a key role in bone

development and remodelling, but they are also involved in many pathological conditions, such as

cancer invasion and metastasis development (John and Tuszynski 2001), arthritis, glomerulonephritis,

and atherosclerosis. A summary of the proteinases of bone and cartilage is given in Table 1. The activity

of MMPs is controlled by their natural inhibitors, the tissue inhibitors of metalloproteinases (TIMPs).

To date, four members of the human TIMP family (TIMP-1, TIMP-2, TIMP-3, and TIMP-4) have been

characterized (Hayakawa 2002).

30

Proteinase Subtypes Major target constituents

MetalloproteinasesCollagenases MMP-1 (interstitial collagenase) Collagens II, III, X, aggrecan

MMP-8 (neutrophil collagenase) PG-link proteinMMP-13 (collagenase 3)

Gelatinases MMP-2 (72 kDa collaginase) Collagens I, X, XI, XIVMMP-9 (92 kDa collaginase) Aggrecan, PG-link protein

Stromelysins MMP-3 (stromelysin 1) Collagens III, IX, X, aggrecanMMP-7 (matrilysin, PUMP-1)MMP-10 (stromelysin 2)MMP-11 (stromelysin 3)

Membrane type MMP-14 (MT-MMP-1) Collagens II, III, aggrecan

AggrecanasesDisintegrin and a disintegrin and metalloproteinase Aggrecanmetalloproteinase type with thrombospondin motifs 4 and 5

(ADAM-TS4, ADAM-TS5)

Other proteinasesElastase cathepsins Cathepsin B & L (cysteine proteases) Collagen telopeptides

Collagen IX, XI, aggrecan, elastinCathepsin G (serine protease) Aggrecan, elastin, collagen II

pro-MMPsCathepsin D (asparate protease) Phagocytosed extracellular matrix

components

Proteinases of the Tissue plasminogen activator (tPa),coagulation system Urokinase type plasminogen activator (uPa)

Plasmin Fibrin, fibronectin, aggrecan,

pro-MMPs

Table 1. Proteinases affecting the metabolism of extracellular matrix moleculules of articular cartilage and bone.

312.1.4 Bone modelling and remodelling

Bone modelling and remodelling are important aspects of bone metabolism. In the modelling,

which takes place principally during childhood, bone undergoes changes in bone size and shape; the

skeleton grows longitudinally, and bones expand also in diameter. Modelling is the main process by

which the bone can increase the size and mass of both cortical and trabecular bone. In humans, the peak

bone mass is usually achieved at the age of 20 years (Currey 2002). During the growth and modelling

period, the osteoblasts and osteoclasts work independently of each other on different bone surfaces.

Usually new bone is formed at the periosteal surface, while active resorption occurs at the endosteal

envelope (Parfitt 1984). The modelling processes are mainly influenced by mechanical loading as well

as by nutritional and hormonal factors.

In remodelling, which is the main process in the mature tissue, bone formation and resorption are

linked in space and time. Bone remodelling is important for at least two main reasons. First, it regulates

the mineral balance of the body by exchanging calcium and other ions from the skeletal reservoir. Second,

bone remodelling can renew old bone and thus improve the mechanical properties of micro-damaged

bone. The remodelling process is thought to be site-specific, though the actual factors regulating this

process are still poorly understood (Currey 2002).

Bone remodelling takes places in BMUs in both cortical and trabecular bone (Frost 1969).

Ultimately, the basic mechanisms of remodelling are the same for cortical and trabecular bone, even

though the final three-dimensional structures are totally different for the two bone types.

Bone remodelling can be divided into four consecutive sequences, namely activation phase, bone

resorption, reversal and formation phases (Figure 3). In the activation phase, osteoclasts are recruited,

stimulated to differentiate and guided to the resorption site. The regulatory mechanisms of the activation

phase are multifactorial and partly unknown. However, the osteocytes have a leading role in this process

which would also mean that they are important in determining the remodelling site (Civitelli 2008, Datta

et al. 2008, Mann et al. 2006, Noble 2008). A new activation phase in BMU is initiated about once every

10 seconds (Dempster 1992).

The resorption phase follows the activation phase, in which activated multinuclear osteoclasts

migrate to the resorption area and become attached to the resorption surface. In the resorption lacuna,

bone mineral is solubilized using the acidification process before the degradation of bone organic

32matrix. Degradation of the organic bone matrix utilizes a large variation of proteinases including

collagenases (Vaes 1988). Table 1 describes the proteinases and their substrates during bone resorption

in both physiological or pathological conditions. While the ultimate regulatory mechanisms of bone

resorption remains to be elucidated, it seems logical that during the resorption process, active mediators

are liberated from the bone matrix or osteoclasts themselves to activate or induce differentiation of

osteoblasts precursors to the resorption lacuna.

The reversal phase follows the resorption phase and lasts 1–2 weeks (Dempster 1992, Väänänen

1993). It reverses the processes acting on the bone from destruction to repair and during this time, the

resorption lacuna is covered by collagen-deficient cement line substance, which will bind the newly

synthesized bone material to the old surface. The reversal phase is presumably a delicate and important

period for the coupling phenomenon of bone remodelling (Väänänen 1993). However, the ultimate

regulatory mechanism has remained unclear.

Figure 3. Bone is remodelled throughout life by a process in which old bone is resorbed and replaced by new bone. In postmenopausal osteoporosis, there is typically an increased frequency of activation of remodelling sites. New bone formation by osteoblasts is not sufficient to fill the remodelling cavities fully. Adapted from IBMS 2008. http://www.bonekey-ibms.org/education/slides/clinical_osteoporosis/

33Bone formation starts with the activation of preosteoblasts which differentiate into osteoblasts.

The active bone-forming osteoblasts produce osteoid, organic bone matrix consisting mainly of type I

collagen and non-collagenous proteins and later on, the osteoid becomes mineralized. Complete filling

of the resorption cavity determines that the bone mass will stay unchanged. If more bone is formed

than that resorbed, the net balance will be positive. However, usually the histomorphometric analysis

of bones of old people have shown increased bone resorption and decreased bone formation, creating a

negative net loss (Reeve et al. 1993). This defect explains the negative bone balance occurring in OP.

After the formation phase is completed, the remodelling site returns to its quiescent state.

2.1.5 Bone turnover and chemical markers of bone formation and resorption

The mature bone of a full-grown individual is continuously replaced by newly formed bone

during the remodelling process. Bone turnover depends on biological and geometrical factors (Parfitt

2002a). Many biological factors are involved in determining the bone formation rate. The bone surface-

to-volume ratio is a geometrical factor that is site dependent. The overall turnover rate is estimated to

be about 10 % a year (Väänänen 1993). It is highest in the central trabecular bone and lowest at the

peripheral cortical bone, where the turnover rate difference can be even ten-fold. In general, there are

marked differences in bone turnover rates in different skeletal sites at different ages and even between

different races (Brockstedt et al. 1993, Dempster 1992, Podenphant and Engel 1987). Most of the

turnover occurs at the endosteal surface of bone close to the bone marrow tissue (Baron 1993, Parfitt

2002a, b).

Bone turnover is regulated by a complex cascade of hormones and active peptides, including not

only the so-called “calcitropic” agents such as parathyroid hormone (PTH), vitamin D, calcitonin but

also matrix metalloproteinases, prostaglandins and growth factors (Nagai et al. 1993, Niederreither et

al. 1992).

Efforts have been made during recent years to create reliable biochemical tests to assay the

bone formation and turnover. There is one bone-specific isoenzyme of alkaline phosphatase (ALP). It

is localized at the cell membrane of the osteoblasts and is also released into the circulation though the

mechanism of its release is unclear. ALP is also released into circulation from the liver, intestine and

placenta. The wide distribution of the ALP sources limits its clinical use, because it lacks sensitivity and

34specificity. However, serum total ALP activity is the most commonly used marker of bone formation

(Delmas and Garnero 1996, Seibel 1993).

Serum osteocalcin (OC) is a small non-collagenous protein, which is specific for bone tissue and

dentin and again, its precise function remains unclear. OC is predominantly synthesized by osteoblasts

and is incorporated into the extracellular matrix. A fraction of this newly synthesized OC is released into

the circulation, where it can be measured by radioimmunoassay (Delmas and Garnero 1996, Risteli and

Risteli 1993). Under most conditions, OC is a valid test for bone turnover when bone resorption and

formation are coupled. OC is also a good marker for bone formation when formation and resorption are

uncoupled (Delmas and Garnero 1996).

Markers reflecting the synthesis of type I collagen can also be used as biochemical indicators

of osteoblast function. These include a test for the carboxyterminal propeptide of type I procollagen

(PICP) and the aminoterminal propeptide of type I procollagen (PINP) (de la Piedra et al. 1994, Risteli

et al. 1991, Risteli and Risteli 1993). During extracellular processing of type I collagen, the amino- and

carboxyterminal propeptides are cleaved off prior to fibril formation. The peptides circulate in blood,

where they serve as useful markers of bone formation, since collagen type I is a major component of the

newly formed bone matrix.

Tartrate-resistant acid phosphatase (TRAP) is an enzyme secreted by osteoclasts during bone

resorption. TRAP activity can also be measured in serum and its activity increases in conditions where

bone resorption is accelerated. As with ALP, its use is limited by the lack of specificity (Risteli and

Risteli 1993). Human serum contains two forms of TRAPs, 5a and 5b, of which 5b is derived from

osteoclasts, and 5a from some other, still unidentified, source (Halleen et al. 2000).

Hydroxyproline (HP) is found only in collagen where it represents about 13% of the amino acid

content of the molecule. The breakdown of the collagen matrix can serve as an indirect method to analyze

the extent of bone resorption. HP can be determined from urine samples by high performance liquid

chromatography (HPLC). Another amino acid unique to collagens and proteins containing collagen-like

peptides is hydroxylysine. It can also be detected by HPLC (Delmas and Garnero 1996, Gough et al.

1994, Grazioli et al. 1993).

A major development improving the specificity of the assessment of bone resorption came about

with the introduction of assays for collagen crosslinks containing pyridinium residues. Pyridinoline

35crosslinks have been found in fibrillar collagens from several tissues, but they are particularly

characteristic of the type I and II collagens. Since the crosslinks are specific for fibrillar collagen, the

pyridinoline crosslinks cannot be derived from the degradation of recently synthesized collagen. The

levels of pyridinoline crosslinks are usually analyzed using HPLC (Colwell et al. 1993, Delmas and

Garnero 1996, Risteli and Risteli 1993, Seibel et al. 1993, Seyedin et al. 1993).

2.1.6 Skeletal adaptation to functional loading

There is known to be a structure-function relationship maintained in the bone tissue. According to

the Wolff’s law all changes in the biology and function of a bone are followed by changes in the internal

structure and external conformation of the bone in accordance with a mathematical model (Kushner 1940,

Prendergast and Huiskes 1995). The generation of bone properties are achieved through mechanically

adaptive modelling and remodelling. In principle, bone is removed from sites where the mechanical

demands from the skeleton are marginal and formed at sites experiencing increased mechanical loading.

Skeletal adaptation to mechanical loading is homeostatically regulated (Figure 4) (Frost 1998).

The orthopaedic surgeon Harold M. Frost (1921–2004) stated in his mechanostat theory that increased

mechanical loading increases bone modelling and reduces remodelling. The modelling occurs in the

periosteal and endosteal surfaces. The remodelling, however, takes place in the cortical bone and endosteal

surface (Frost 1990a, b). Modelling or remodelling continues until a new steady state is attained.

Another theory involved in the structure-function adaptation of bone is based on

mechanotransduction. The central concept of this theory is the transfer of the mechanical loading-induced

stimuli to chemical and/or electrical signals and eventually to cell and tissue responses. According to this

theory, the viscoelastic properties of bone induce fluid flow through the interstices of bone, either directly

by local deformation or by electrical effects related to streaming potentials, mediating the mechanical

signal (Beck et al. 2002). However, details of this phenomenon are still poorly understood.

36

Figure 4. Schematic presentation of the mechanostat theory. The mechano-biologic feedback mechanism functions under the control of the subject’s mechanical usage, adjusting skeletal ar-chitecture in the way that tends to prevent structural failures of skeletal tissues and organs. MFL means the mechanical feedback loop, IM stands for intermediary mechanisms, in other words the mechanostat’s central unit or decision-making centre. Adapted from Frost (1998).

2.2 Biomechanical properties of bone

Bone is a unique composite material. It can withstand a considerable mechanical load especially

in comparison to its slender structure. Its lightness makes bone an ideal structure for biological

organisms. The differently shaped and sized bones of the skeletal system are under strict genetic control.

In addition, endogenous hormonal, nutritional and environmental factors influence the bone architecture

and mass (Howard et al. 1998). During physical activity, complex distributions of forces are applied to

the skeleton. Due to the imposition of these forces, bones undergo deformation and remodelling.

Since bone is an anisotropic material, its elastic properties depend on the orientation of the

inherent material with respect to the loading direction, in contrast with isotropic materials (e.g., steel and

rubber), where the elastic and material properties are the same in all directions.

Any treatment that can change the structure or composition can affect the mechanical properties

of bone. Even changes in the water content of bone, or different sample storage methods, can alter its

mechanical properties (Callaghan and McGill 1995, Linde and Sørensen 1993, Turner and Burr 1993).

The material properties can be defined by its tissue-level qualities, independent of the macroscopic

structure, size or geometry of bone. This method can be used in the determination of the mechanical

properties with prepared bone samples. On the other hand, the gross structural properties of bone describe

the tissue as a whole anatomical unit (Einhorn 1992, Martin 1991).

LOADS SKELETON SIGNALS (IM) BIOLOGIC MECHANISMS/ACTIVITIES

MFL

Systemic and nonmechanical agentsAmino acids Hormones VitaminsBlood pH Blood osmolality Blood gasesMinerals Cytokines DrugsFluoride Apoptosis ToxinsCarbonhydrates Proteins LipidsCell-cell interactions Cell-intercellular matrix interactionsOther synthetic agents Gene expression Disease

372.2.1 Stress and strain

During mechanical testing, the bone samples generate an internal force to resist the testing load.

The original shape of bone becomes deformed. The force and the deformation can be expressed by two

basic quantities: stress and strain. Stress is defined as force per unit area. In qualitative terms, it can

further be divided into compressive, tensile, or shear stress (Turner and Burr 1993). Compressive stress

is produced when a compressive force causes bone to become shorter. A tensile stress occurs when the

forces act in opposite directions. Shear stress is produced when two forces are acting in parallel with

each other, but in different directions (Figure 5). The paper-cutting ability of the scissors is an everyday

example of shear stress. In real life, these three basic stress types may act alone, or in combination.

Actually bone bending is a combination of compressive and tensile stress. Torsion force adds a shear

stress component along with the longitudinal axis of the bone.

Strain describes the resulting deformation in shape and size, when the bone is subjected to force.

Unlike stress, strain is a scalar and dimensionless quantity and is usually recorded as percentage change

from the original bone dimension (Turner and Burr 1993). The proportionality of stress and strain is

called Hooke’s law in physics after Robert Hooke, a contemporary of Sir Isaac Newton in the 17th

century (Young and Freedman 2004).

Figure 5. Typical deformation curve from the three-point bending test of the murine femoral diaphysis (A). Fu = ultimate load to break the bone, K = elastic stiffness of bone, B = ultimate displacement before break, and A = ultimate energy needed for break. During the test, the ends of the femoral diaphysis are supported by two fulcra separated by a fixed distance of, e.g., five mm (B).

382.2.2 Stiffness and elastic modulus

The load-deformation curve (Figure 5) can be divided into two regions; elastic and plastic

deformation regions. The slope of the load-deformation curve within the elastic deformation region

reflects the extrinsic stiffness or rigidity of the structure. When the load-deformation curve is converted

into a stress-strain curve, the slope in the elastic region represents the elastic modulus, which is also

called the Young’s modulus of elasticity. Young’s modulus is a measure of the intrinsic stiffness of the

material (Turner and Burr 1993, Young and Freedman 2004).

The elastic and plastic strain regions of the stress-strain curve are divided by the yield-point.

Up till this point, the load and deformation are linearly related. If the load is further increased, then the

specimen begins to undergo permanent deformation and eventually fails at the point of ultimate stress

or strain.

2.3 Biomechanical testing of bone

The history of the mechanical testing of bone goes back more than a century. One of the first

scientists who measured effects of various loads on bone dimensions, density, and displacement was

Wertheim in 1847 (Jämsä 1998). He observed that the deformation was proportional to the load applied

to the sample. Since then, several different laboratory methods have been developed to determine the

mechanical properties of bone. Today, mechanical testing is usually performed using screw-driven,

pneumatic electrical, or servo-hydraulic testing machines (Turner and Burr 1993). Some of the most

frequently used regimes include the methods of compression, three- or four-point bending, torsion, and

tension testing (Jämsä et al. 1998a, Jämsä et al. 1996, Turner and Burr 1993). The mechanical properties

of whole bones as well as prepared bone samples are assessed with these testing methods.

2.3.1 Compression and tension

The compression test is suitable for bone samples, which are exposed to compressive forces in

vivo. A typical anatomical site is the vertebral body, the mechanical properties of which can be tested

by compression. Cylindrical and cubic specimens from cortical or trabecular bone can also be tested by

compression. Compression is typically performed using a specimen with trimmed parallel surfaces and

compressing it between two steel plates (Turner and Burr 1993).

39The validity of this test depends critically on the sample preparation and geometry, and the

results can be altered by variations in the specimen architecture at its trimmed ends, interface conditions

between the specimen and the plate, the Poisson’s ratio, and other factors such as the preconditioning

protocol prior to the test (Keaveny and Hayes 1993, Keaveny et al. 1997, Linde et al. 1992). This is

extremely important when trabecular bone specimens are to be analyzed (Keaveny et al. 1997, Odgaard

et al. 1989, Odgaard and Linde 1991).

Tensile strength testing has also been applied to bone samples (McCalden et al. 1993, Wang et

al. 1998). Tensile strength test is a sensitive indicator of fracture healing (Walsh et al. 1997). However,

tensile tests are more often used to test the mechanical properties of ligaments and tendons (Lin et al.

1999, Meikle et al. 1984).

2.3.2 Torsion and bending

In a torsion test, the whole diaphyseal bone is twisted from the ends of the specimen. Torsion

testing is regarded as the best testing method for long bones, because a torsional load subjects the bone

to equally heavy loading at every section along its length, thus making it possible to identify the weakest

bone site (Burstein and Frankel 1971). Torsion loading consists of tensile, compressive, and shear forces,

each acting at different planes in the same region of bone (Lepola et al. 1993). The bone ends are usually

secured by embedding them in blocks of plaster, epoxy, fibreglass resin, a low melting point metal alloy

or some other embedding material, though fixation with clamps has also been employed (Jämsä and

Jalovaara 1996).

Bending loading of long bones is actually a combination of compressive and tensile forces acting

on the object of interest. The compressive force vector is situated at the site of intender contact and the

support fulcra in three- and four-point testing devices (Figure 5). The tensile forces produce their effect

on the opposite side of the specimen. There is a point within the sample where these forces neutralize

each other. Bone is weaker in resisting tension than compression, and therefore a bending failure usually

occurs on the tensile side of the bone (Martens et al. 1986, Turner and Burr 1993). The applied force is

recorded at the time of fracture.

Methodological problems may arise in the three-point bending test, as there a remarkable shear

stress component may be present in the predetermined fracture point that is situated directly under the

40indenter head. On the other hand, four-point testing is more difficult to perform accurately, due to the

difficulty of achieving equal forces in all loading points when an entire long bone is tested (Turner

and Burr 1993). With the irregularly shaped murine bones, three-point bending is the preferred method

(Turner and Burr 1993). There is also a method to test the properties of the femoral neck bone. This

technique mimics the events occurring in the femoral neck fracture of a patient (Jämsä et al. 1998b, Peng

et al. 1994).

2.4 Bone densitometry

2.4.1 History of bone densitometry

Historically, before the introduction of bone densitometry, bone mineral content and bone mass

were determined by burning dried and pulverized bones at a high temperature and then weighing the

remaining ash, the inorganic mineral content (Chang and Bloom 1983, Martin 1967).

The history of bone densitometry actually was initiated by the physicist and later Nobel Prize

winner Wilhelm Conrad Röntgen (1845–1923), Dean at the Julius Maximilian University of Würzburg

and holder of the chair of physics, who discovered X-rays in 1895. X-ray images provided an early

opportunity to assess bone quality, although due to various sources of error, the estimations were often

rather inaccurate (Nilsson et al. 1990).

Although bone ashing provided an accurate measure of bone mass, this method has gradually

been replaced by bone densitometry. The first innovation was the use of single- (SPA) and dual photon

absorptiometry (DPA) in the 1980’s to assess the bone mass in a non-invasive manner (Cameron and

Sorenson 1963, Castelo-Branco et al. 1993, Eriksson et al. 1988, 1989, Kelly et al. 1988, Li et al. 1990,

Peppler and Mazess 1981, Shipp et al. 1988, Ørtoft et al. 1993). The next step forward, the dual-energy

X-ray absorptiometry (DXA), became available at the beginning of the 1990’s (Mazess et al. 1990).

The invention of the computed tomography (CT) has been considered to be the greatest innovation

in the field of radiology since the discovery of X-rays. This cross-sectional imaging technique provides

diagnostic radiology with better three-dimensional resolution. In 1979, G.N. Hounsfield and A.M.

Cormack were awarded the Nobel Prize in medicine for the invention of the CT. Since CT-imaging is

based on X-ray absorption in imaged objects, it was also found to be suitable for estimation of tissue

41densities (Cann and Genant 1980, Genant and Boyd 1977).

The single energy X-ray attenuates exponentially in a homogenous medium as follows:

(1),

in which I0 is the intensity of incoming photon radiation, x is the thickness of the homogenous

medium, and μ is the medium-specific coefficient of attenuation. From this equation, the unknown

variable μ can be computed as follows:

(2).

2.4.2 Dual-energy x-ray absorptiometry

Currently, DXA is a widely used method to evaluate bone mineral density and to determine which

patients are at risk of osteoporotic fractures (Ito et al. 1993, Mazess et al. 1990). Furthermore, it has also

been used in many studies to evaluate new therapies for osteoporosis and in animal models predicting

the mechanical properties of bone (Järvinen et al. 1998). The advantages of the DXA technique include

the low radiation dose (typically < 1µSv per scan), short scanning time, low cost, high precision and low

coefficient of variation (Johnson and Dawson-Hughes 1991, Mazess et al. 1990).

The technical principle of DXA-scan is based on the use of two different X-ray energy bands

(typically 38 and 70 keV). The two energies utilized allow discrimination of the soft and hard (bone) tissues

in the scan area. This minimizes errors due to an irregular body contour and soft tissue inhomogenities.

According to equation 1, the attenuation of the dual energy x-ray can be calculated as follows:

(3)

(4),

where I0 is the intensity of original x-ray radiation, I is the intensity of attenuated radiation, M mass of

medium per area unit (g/cm2), and μ is the attenuation constant of the medium (cm2/g). Subscripts 38

and 70 determine the attenuation at specific energy levels (keV). The subscripts s and b refer to mass of

soft tissue and bone, respectively. Furthermore, the mass and ratio of the attenuation constants (R) can

be determined as follows:

€

I = I0e−μx

€

μ =1x

ln I0

I

€

I38 = I0(38)e−(μ(s38 )M s +μ(b38 )M b )

€

I70 = I0(70)e−(μ(s70 )M s +μ(b70 )M b )

42

(5)

(6)

(7)

(8).

The R-value is the ratio of soft tissue attenuation at 38 keV compared to that at 70 keV. A typical ratio

is 1.40. High values indicate that there is less subcutaneous fat in a lean subject. Lower values reflect

obesity.

The amount of bone in the beam path is calculated typically as bone mineral content in density

(BMD). BMD (g/cm2) is usually multiplied by the scanned bone area (cm2), and the calculated bone

mineral content (BMC) in the region of interest (ROI) is expressed in grams and often, also the scanned

bone area will be reported (Koo et al. 1995). With the introduction of high resolution software, DXA

has been successfully applied in animal studies, including bovine (Oden et al. 1998), monkey (Jayo et

al. 1994), dog (Markel and Bogdanske 1994, Puustjärvi et al. 1992), swine (Yang et al. 1998), and rat

(Bateman et al. 1998, Järvinen et al. 1998, Kastl et al. 2002, Vanderschueren et al. 1993) bones, and even

histological bone sections (Denissen et al. 1996).

2.4.3 Peripheral quantitative computed tomography

Today peripheral quantitative computed tomography (pQCT) can be used to obtain high-resolution

three-dimensional images (Mueller et al. 2008). The tomographic X-ray information is quantified pixel

by pixel. Thus, by using pQCT, the true volumetric BMD (vBMD) can be determined. This method

€

Ms = Rb ln I70

I0(70)

⎛

⎝ ⎜

⎞

⎠ ⎟ −

ln I38

I0(38)

⎛

⎝ ⎜

⎞

⎠ ⎟

μs(38)μs(70)Rb

⎛

⎝

⎜ ⎜ ⎜ ⎜ ⎜

⎞

⎠

⎟ ⎟ ⎟ ⎟ ⎟

€

Mb = Rs ln I70

I0(70)

⎛

⎝ ⎜

⎞

⎠ ⎟ −

ln I38

I0(38)

⎛

⎝ ⎜

⎞

⎠ ⎟

μb(38)μb(70)Rs

⎛

⎝

⎜ ⎜ ⎜ ⎜ ⎜

⎞

⎠

⎟ ⎟ ⎟ ⎟ ⎟

€

Rs =μs(38)

μs(70)

€

Rb =μb(38)

μb(70)

43allows also an accurate and precise determination of the cross-sectional geometry of a bone section,

including the cross-sectional moment of inertia (CSMI) of long bones.

The bone strength index (BSI) determined by pQCT, calculated from formula CSMI x vCtBMD,

correlates closely with the breaking force of the femoral shaft in experimental studies (Ferretti et al.

1996). The elastic modulus of bone, however, correlates strongly but negatively with CSMI, while the

ultimate fracture force and stiffness show a positive correlation with the pQCT parameters (Ferretti

1995).

2.4.4 Quantitative magnetic resonance imaging

The growing awareness of the role of parameters other than BMD in determining the bone strength

and fracture susceptibility has spurred the search to find new methods to characterize bone quality, in

particular techniques that would be non-invasive. High-resolution magnetic resonance imaging (MRI),

in conjunction with advanced image processing and analysis techniques, which are capable of extracting

structural information at the limited spatial resolution achievable in vivo, has demonstrated significant

potential to achieve this goal. MRI reveals bone architecture, relating to both trabecular and cortical

bone. However, the current limitations relate to availability and cost, but these are likely to be overcome

in future with the development of dedicated instruments (Kröger et al. 1995c, Wehrli 2007).

2.4.5 Bone ultrasound

Ultrasound (US) has been investigated as another alternative method free of ionizing radiation for

achieving a non-invasive assessment of the biomechanical competence of bone. The rationale of using

US rests on two basic properties of US. First, its interaction with the material can be used to determine

the US velocity and the density of the tissue. This makes it possible to characterize the elasticity of bone

and to estimate its strength. Second, US may be used to characterize tissue properties over a wide range

of spatial dimensions and organization levels (Hans et al. 1999, Kröger et al. 1995b). Furthermore, it is

an economic and safe way to estimate bone quality.

Bone US measurements in clinical practice are usually performed from calcaneus, because it is

easily available. Calcaneal US correlates fairly well (r = 0.6–0.8) with BMD of the lower lumbar spine

and femoral neck (Kröger et al. 1995b, Yamada et al. 1993) and predicts osteoporotic fractures (Huopio

44et al. 2004). US signal information is dependent on trabecular orientation and thus it does provide some

information about bone architecture. However, ultrasonic bone measurements are a relatively recent

development, and for many of the clinical devices in use today, interpretation of the values produced by

the device remains a challenge. Therefore, further research is needed into the clinical relevance of the

numerous parameters provided by US measurement (Glüer et al. 1993, Glüer et al. 1994, Kröger et al.

1995b, Nicholson 2008).

2.5 Osteoporosis

Osteoporosis is recognized as a major public health problem in many countries, particularly in the

United States and northern Europe. Together with another general musculoskeletal disorder, osteoarthrosis,

OP is responsible for significant health care costs (Holroyd et al. 2008). The term osteoporosis refers to

a group of conditions where the bone mass and structure are altered, increasing the risk for fracture. The

current view is that the age-dependent (primary type II or senile) OP is caused by a gradual loss of bone,

leading to osteopenia. In the criteria published by the World Health Organization (WHO), OP is defined

as a state in which the BMD of a person is at least 2.5 SD (standard deviation of mean) below the BMD

of an average healthy individual of the same gender aged 20 to 40 years. Severe OP is a condition where

osteoporotic BMD exists with one or more osteoporotic bone fractures. Postmenopausal (primary type I)

OP is associated with increased activity of bone resorption, while in type II OP, reduced bone formation

is a typical finding. Secondary OP is caused by administration of pharmaceutical agents, by diet, diseases

or due to other related conditions (Marcus 1992). The relative proportions of the organic and mineral

phases of bone remain constant even in the trabeculae of osteoporotic bone (Bätge et al. 1992, Oxlund

et al. 1995). As well as the bone mass, the mechanical properties of bone depend on the architecture and

structure of the inorganic and organic component (Boonen et al. 1995).

Though, it is believed that genetic factors contribute to the risk for OP (Ferrari 2008), there is

a lack of information about the mechanisms and specific genes contributing to the development of OP

(Brown and Duncan 1999, Kung and Huang 2007). Heredity is an important determinant of the bone

mass determined by densitometry (Evans et al. 1988, Kröger et al. 1994). According to data from family

studies, hip fractures in a woman’s mother, sister or maternal grandmother at least doubles her risk

for a fracture in the future (Cummings et al. 1995). The lower incidence of OP in men is due to their

45higher peak bone mass, shorter life expectancy and the absence of a distinct menopause-equivalent,

andropause, which would cause marked acceleration of bone loss. However, 20–25% of all hip fractures

are found in men as are one-seventh of all osteoporotic vertebral compression fractures (Jackson 1993,

Kaufman and Goemaere 2008).

OP incurs expenses on the public health budget by increasing osteoporotic fractures of hip,

proximal humerus, distal radius and thoracolumbar spine in elderly citizens in context with low-energy

trauma. These fractures represent one of the most important causes of disability and death in these

individuals. The incidence of hip fractures increases exponentially with age. In Finland, the number of

hip fractures has increased steeply over the last 30 years, in 1970 these were 1857 hip fractures reported

and in 1997 the number had risen to 7122 (Kannus et al. 1999, Kannus et al. 2000). If the prediction

of Kannus et al. holds true, the number of hip fractures will be almost three times higher in the year

2030 (Kannus et al. 1999). This would create an almost overwhelming financial burden due to OP and

related fractures. However, recent statistics about the number and incidence of hip fractures indicate that

a steady state, or even a decline in incidence, has been achieved (Kannus et al. 2006). If the incidence

of hip fractures were to become stabilized at the year 2004 level, the number of hip fractures in Finland

would still be about 12 600 in the year 2030 (Kannus et al. 2006).

2.6 Models of experimental osteoporosis

2.6.1 Effect of exercise on bone

Bones undergo continuous modelling and remodelling in response to mechanical and metabolic

demands (Currey 2002).The positive effects of physical exercise on a healthy bone have been well-

documented in several clinical and experimental studies (Chen et al. 1994, Mosekilde et al. 1994,

Nordsletten et al. 1993, Puustjärvi et al. 1995, Shimegi et al. 1994, Søgaard et al. 1994). There is evidence

that physical exercise can inhibit, or reverse, the involutional bone loss from the lumbar vertebrae in

normal postmenopausal women and can also prevent spinal OP (Krølner et al. 1983). However, the

threshold levels of activity associated with either positive or negative changes are still unknown (Currey

2002, Forwood and Burr 1993).

Growing bone has a greater capacity to form new bone than the adult tissue (Yeh et al. 1993).

46Thus, exercise that increases bone strength and peak bone mass in children and adolescents is able to

improve bone quality when these individuals are elderly adults (Frost 1999, Welten et al. 1994).

The bone gain in response to exercise seems to depend on certain conditions and types of exercise

(Haapasalo et al. 1994). Weight-bearing exercise such as running (Lane et al. 1986), weight-lifting

(Karlsson et al. 1995), or ballet dancing (Karlsson et al. 1993) are more effective than non-weight-

bearing exercise (e.g., swimming) (Matsumoto et al. 1997). This emphasizes the importance of impact-

type training as a way to enhance bone mass (Grimston et al. 1993, Heinonen et al. 1993, Nordström et

al. 1998, Pettersson et al. 1999).

Negative effects of strenuous physical training have been documented in humans. In female

marathon runners, the decrease of BMD in the lumbar spine was associated with amenorrhea (Drinkwater

et al. 1984, Hetland et al. 1993, Jonnavithula et al. 1993, Marcus et al. 1985, Myburgh et al. 1993).

Strenuous training reduces longitudinal bone growth and induces bone loss in growing rats. The bone

loss seems to be caused by decreased osteoblastic activity rather than a more general adaptation of the

bone to remodelling (Bourrin et al. 1994). It has been suggested that excessively vigorous mechanical

usage increases bone remodelling due to an elevated level of bone microdamage (Bentolila et al. 1998,

Frost 1990b).

The exact mechanism by which physical activity increases bone mass is still unknown and there

is a need for better understanding of the adaptive mechanisms of bone to mechanical stress and to

determine the optimal level and modality of physical activity (Rockwell et al. 1990, Suominen 2006).

The response of bone to weight-bearing exercise appears to be site-specific (Tommerup et al. 1993).

Cessation of physical exercise reduces the gained bone mass to normal so that long-term benefits are

only maintained with continuing exercise (Karlsson et al. 1995).

It should be remembered that there are also other benefits to be obtained from improved fitness

and muscle strength in addition to bone mass gain. These benefits include improvements in the body

balance and stability (Karinkanta et al. 2008). In addition, the reaction and muscular response time are

shortened, which can be more important in preventing falls and reducing fracture risk than any increment

of bone mass per se.

472.6.2 Effect of immobilization on bone

BMD is one of the several methods, that can be used to study the effect of physiological unloading

on bone architecture, biomechanical properties, and markers of bone metabolism. The unloading effect

can be induced by casts of plaster of Paris or other material, tail suspension apparatus, utilizing a

microgravity environment, and by neural damage or neurectomy.

2.6.2.1 Disuse osteopenia

The immobilization-induced osteopenia has been studied intensively (for review see Kannus et al.

(1992a, b)) and atrophic changes in tendons, ligaments and cartilages induced by immobilization, have

also been reported. In the bones of dogs, three consecutive steps have been described in experimental

studies (Jaworski et al. 1980). First, a rapid bone loss takes place reaching a maximum at six weeks.

This is followed by a rapid recovery, with contributions of periosteal, endosteal and Haversian surfaces.

During the second stage, a slow but long lasting bone loss occurs over a time period of 24 to 32 weeks.

The loss occurs mostly at the periosteal envelope. In the third stage, the bone mass is maintained at a

level corresponding to 30 to 50% of its original amount (Uhthoff and Jaworski 1978).

The first two stages of immobilization may represent a stage of reversible osteopenia. There is

a decrease of the volume of mineralized bone due to the expansion of the bone remodelling space. The

third stage may be a reflection of permanent osteopenia, i.e. true OP. In this stage, total bone volume and

other components of bone, e.g., trabeculae have been irreversibly lost (Jaworski et al. 1980). During the

first stage rapid bone resorption occurs by the action of the increased number of osteoclasts, the enlarged

osteoclast surface and the sustained decline in the bone mineral apposition rate (MAR) and the reduced

bone formation rate (BFR) (Weinreb et al. 1989).

Since trabecular bone turnover is more rapid by a factor ranging from 10 to 100 than the turnover

of cortical bone, the immobilization-induced skeletal alterations are probably more rapid and are more

striking in trabecular than in compact bone (Frost 1969).

482.6.2.2 Weightlessness

Studies from space flights over the past three decades have revealed changes in mineral

metabolism and bone mechanical properties during prolonged weightlessness. Long-term exposure to a

microgravity environment causes 10 to 20% loss of BMD, up to 25% loss of muscle mass, cardiovascular

deconditioning, fluid shift to the upper body, anaemia, and reduced immune function (Hughes-Fulford

1993, Lane et al. 1993). Continuous negative calcium balance (140–200 mg/day) induce monthly losses

of BMD in lumbar spine, femoral neck, trochanter, pelvis, and legs (1.1%, 1.2%, 1.6%, 1.4%, and 0.3%,

respectively) (LeBlanc et al. 2007). Interestingly, the distal radius seems to be resistant to mineral loss.

After the space flight, bone recovery occurs, but at a slower rate than the rate of loss during space flight.

The final recovery of bone may require 1 to 3 years after the flight, and even then the recovery may not

be complete.

Prolonged bed rest has been used as ground-based model of a space flight to study the nature of

disuse osteopenia and to test possible countermeasures. The bed-rest induced changes in BMD, calcium

excretion, calcium balance, and bone metabolic markers are qualitatively similar to those documented

in space flights, even though quantitatively the changes are of a lesser degree (Krølner and Toft 1983,

LeBlanc et al. 1987, LeBlanc et al. 2007, Lueken et al. 1993, Uebelhart et al. 2000). The femoral arterial

blood flow decreases as much as 40% in the rat tail-suspension model of microgravity pointing to a role

for changes in blood flow in the etiology of muscular atrophy and bone loss (Roer and Dillaman 1994).

2.6.2.3 Neurectomy or paralysis

The effects of long-term immobilization of the lower limbs have been studied on patients with

hemiplegia (del Puente et al. 1996) and spinal cord injury (Dionyssiotis et al. 2007). Hemi- or paraplegia

creates a sudden immobilization of the lower limbs. However, the pathophysiology of OP in spinal

cord-injured patients is complex. There is some evidence that the nervous system would participate

in the skeletal development and turnover (Takeda 2005). Dionyssiotis et al. reported a loss of BMD

in the epiphyseal region of tibia of up to 57.5% in high and 51% in low paraplegics as compared with

their age-matched controls. The cortical thickness of tibia decreased by 21.4% and 23.2%, respectively

(Dionyssiotis et al. 2007).

Cutting of the sciatic nerve has been used to provoke immobilization-induced bone loss of one

49lower limb in experimental animals (Garcés and Santandreu 1988, Iwamoto et al. 2002, Suyama et al.

2002). However, the results of neurectomy on BMD have proved quite difficult to interpret, because

there are several confounding factors. The role of the nervous system is not fully understood and it

is possible there are also systemic factors associated with regulation of mineral homeostasis. After a

tibial shaft fracture, the sciatic nerve resection protects the femoral shaft against normally occurring

post-traumatic bone loss (Madsen et al. 1996). Neurectomy of a peripheral nerve does not either affect

the longitudinal bone growth in rats (Garcés and Santandreu 1988). However, Iwamoto et al. reported

that briefly after neurectomy, bone resorption increased and bone formation decreased, resulting in a

reduction of trabecular bone mass (Iwamoto et al. 2002).

2.6.3 Collagen gene defects and bone

Defects in collagen type I create the typical appearance of osteogenesis imperfecta, in which

the bones are brittle and susceptible to fracture (Bonadio et al. 1990, Fratzl et al. 1996, Prockop et al.

1994, Saban et al. 1996). Mutations in type II collagen provoke various types of chondrodysplasias, OA

and Stickler’s syndrome, where ocular manifestations and spinal deformations are the most prominent

features (Sahlman et al. 2001, Savontaus et al. 1996). Ehlers-Danlos syndrome is a heritable generalized

connective tissue disorder that is characterized by skin hyperextensibility, fragile and soft skin, delayed

wound healing with the formation of atrophic scars, easy bruising and generalized joint hypermobility

(Sokolov et al. 1991). Recently, mutations in the COL5A1 and the COL5A2 gene were identified in

approximately 50% of patients with a clinical diagnosis of classic Ehlers-Danlos syndrome (Malfait and

De Paepe 2005). Mutations of the type III collagen gene have also been reported in the vascular variant

(type IV) of Ehlers-Danlos syndrome (Callewaert et al. 2008, Gdynia et al. 2008). For a review of

animal models in ECM diseases see Aszódi et al. (1998b).

2.6.4 Effect of estrogen deficiency on bone

Estrogen deficiency in postmenopausal females or after ovariectomy, may lead to an accelerated

bone loss and ultimately to OP (Järvinen et al. 2003b). However, the precise mechanism of action of

estrogen in bone is not clear. Estrogen can regulate bone metabolism by multiple mechanisms, having

direct effects on osteoblasts through estrogen receptors, but also it may attenuate the bone resorption-

50stimulating pathway (Majeska et al. 1994). Thus, estrogen plays an important role in the treatment of

postmenopausal OP (Rozenberg et al. 1994).

Postmenopausal bone loss is on average 2–3% per year and this can be divided into two separate

phases. First, an accelerated, though transient, phase occurs during the decade after the cessation

of the menstrual cycles accounting for 20–30% loss of trabecular and 5–10% loss of cortical bone.

Subsequently, a gradual and continuous phase takes place causing bone loss of up to 20–30% in both

trabecular and cortical bone (Nordin et al. 1990, Riggs et al. 1998).

2.7 Osteoarthrosis

OA is a common progressive degenerative disease of joint cartilages. It is defined by the American

College of Rheumatology as a “heterogenous group of conditions that lead to joint symptoms and signs

which are associated with defective integrity of articular cartilage, in addition to related changes in

the underlying bone at the joint margins”. OA is usually classified as a primary or idiopathic disease

when there are no obvious predisposing factors, and a secondary disorder when those factors exist, e.g.,

previous intra-articular fractures. Primary OA is the most common form of the disease, affecting 60%

of men and 70% of women over the age of 65 years. This incurs enormous socioeconomic costs (Sarzi-

Puttini et al. 2005). As the baby-boomers reach middle age and the obesity continues to increase in the

general population, OA will have an ever greater impact on the populations in western countries (Roach

and Tilley 2007).

OA can affect any joint of the body though the weight-bearing joints, hip and knee, are often

important sites for this disease. Aging is a major risk factor for the development of OA, the other risk

factors are abnormal loading conditions due to trauma, heavy manual labour and obesity (Hamerman

1993, Roach and Tilley 2007). In addition, genetic factors, e.g., a defect in type II collagen, may

predispose to OA (Vikkula et al. 1993). A breakdown of the collagen fibril network is considered to be

the critical and probably the irreversible step in the progression of OA (Eyre 2002).

Typical symptoms of OA are pain, joint stiffness and restriction of joint mobility. The clinical

diagnosis of OA is based on the patient history and clinical examination of affected joints. Plain X-ray

images are usually sufficient to confirm the diagnosis. Primary OA is a difficult disease for both the patient

and the clinician, because often neither can any true cause be found nor any effective cure initiated. Once

51the degenerative process has started, the disease cannot be halted and the only relief can be achieved by

palliative treatment. This includes treatment with analgesics, intra-articular injections of corticosteroids

or HA, physiotherapy and surgical interventions including osteotomies and eventually joint arthroplasty.

Although OA is generally considered as a cartilage-related disease, it affects periarticular bone

as well. The subchondral bone plate thickens during the OA process. It is still unclear whether these latter

changes are causative or if they are a consequence of cartilage degradation (Roach and Tilley 2007).

2.8 Animal models of experimental osteoarthrosis

Animal models provide one way to elucidate the early OA changes in vivo. Animal models allow

accurate timing of the follow-up of changes occurring in the OA progress. An optimal animal model

should reproduce both the pathogenesis and the pathology of human OA. Nonetheless the interpretations

of disease pathogenesis needs to be made with caution when results from an animal model are extrapolated

with the same disease in humans.

Some mouse strains develop spontaneous OA during their lifespan (Silberberg and Silberberg

1941, Sokoloff 1956). The onset of degenerative cartilage changes can be accelerated by utilizing weight-

bearing or strenuous exercise regimens (Arokoski et al. 1993a, Lapveteläinen et al. 1995). In most

species, however, the prevalence of spontaneous OA is too infrequent to be of value in experimental

studies.

The development of transgenic technology has made it possible to generate targeted gene-mutated

mouse lines suitable for use in experimental OA research (Helminen et al. 2002). The vast majority of

these studies have focused on the effect of collagen mutations, but also the role of other matrix proteins

in the pathogenesis has been studied (for review see Helminen et al. (2002)). In addition, studies using

surgical interventions to provoke OA changes have been reported (Lovász et al. 1995, Panula et al. 1997,

Pritzker 1994).

52

3 AIMS OF THE STUDY

The aims of the present study were to evaluate:

1. The effect of long-term running exercise on canine bone mineral density, me-

chanical properties and organization of collagen fibril network.

2. The effect of immobilization and remobilization on the canine bone mineral

density, composition and bone structure.

3. The effect of osteotomy of the canine proximal tibia and the development of

experimental osteoarthrosis on subchondral bone remodelling.

4. The impact of heterozygous knockout inactivation of one allele of the Col2a1

gene on the bone development and properties.

5. The effect of the truncated human COL2A1 transgene on bone development and

bone remodelling in mice.

53

4 MATERIALS AND METHODS

4.1 Animal models

This study utilizes material from five animal studies carried out by the KUVANI1-team in the

Department of Anatomy, University of Kuopio, Finland during years 1989–1995 (Table 2).

Table 2. Summary of the animals and study groups. HZK (heterozygous knock-out), TG (transgenic).

1KUVANI is an acronym for ”kuormituksen vaikutus nivelrustoon” (the effect of loading on articular cartilage)

StudyNumber and (age) of the

animals Study groups (gender and intervention)

I 10 (15-50 wk) female beagle dogs in progressive running training

(n=20) 10 (15-50 wk) female beagle dogs served as controls

8 (29-40 wk) female beagle dogs cast-immobilized (29-40 wk)

II 9 (29-90 wk) female beagle dogs cast-immobilized (29-40 wk) and remobilized (40-90 wk)

(n=34) 8 (29-40 wk) female beagle dogs served as controls

9 (29-90 wk) female beagle dogs served as controls

8 (12 wk - 7 mo) female beagle dogs with valgus osteotomy of proximal tibia

7 (12 wk - 18 mo) female beagle dogs with valgus osteotomy of proximal tibia

III 7 (12 wk - 7 mo) female beagle dogs with sham operation

(n=44) 7 (12 wk - 18 mo) female beagle dogs with sham operation

8 (12 wk - 7 mo) female beagle dogs served as controls

7 (12 wk - 18 mo) female beagle dogs served as controls

27 (4 wk - 15 mo) female HZK C57BL mice

22 (4 wk - 15 mo) female C57BL mice served as controls

IV 33 (4 wk - 15 mo) male HZK C57BL mice with running ability

(n=205) 30 (4 wk - 15 mo) male C57BL mice with running ability

31 (4 wk - 15 mo) male HZK C57BL mice served as controls

31 (4 wk - 15 mo) male C57BL mice served as controls

V 48 (1 wk - 12 wk) female TG FVB/N mice

(n=100) 52 (1 wk - 12 wk) female FVB/N mice served as controls

544.1.1 Canine long-term running training (I)

Twenty female purebred beagle dogs from the National Laboratory Animal Centre in Kuopio,

Finland and from Shamrock Ltd., UK, were used as experimental animals. The experimental design was

approved by the Animal Care and Use Committee of the University of Kuopio and it complied with the

principles of the Care and Use of Animals (National Research Council 1985). Ten dogs ran five days a

week according to a progressive program on a treadmill from the age of 15 weeks to the age of 70 weeks

(Figure 6). During the first ten weeks of the training program, the dogs were adapted to running whereafter

the running distance was gradually increased up to 40 km/day with the dogs running this distance, five

days a week for the last 15 weeks of the experiment (Arokoski et al. 1991). The speed of the treadmill

was 5.5–6.8 km/h and the inclination of the treadmill belt was 15 degrees uphill. Ten age-matched sister

dogs were housed in individual cages during the study period. The bottom areas of the cages were 0.9 x

1.2 m and the height 0.8 m. After exercise, the dogs rested in individual cages. During the rest period in

the cage, the moving activity was recorded by the rest shelf method (Arokoski et al. 1993b). While in the

cages, no statistically significant difference was detected between the moving activities of the exercised

and control animals. The dogs were fed with a commercial dog food (Hankkija, Kolppi, Finland) and

water was given ad libitum. The running dogs had three to four breaks for rest and drinking during each

running bout. The individual amount of food to dogs was calculated by weekly weight monitoring to

maintain the body weights at the same level in both groups. The food consumption of the runner dogs

was approximately 1.5 times higher that of the controls during the last 15 weeks of the experiment.

Within 60 hour of completing the running regime, the dogs were sacrificed with an overdose

of barbiturate (Mebunat, Orion Oy, Espoo, Finland) and their spines, femoral and pelvic bones, were

immediately dissected free of soft tissue. Before sacrifice, X-ray radiographs were taken from the spines

and the bones. Within 1.5 hours, all the samples were weighed, anthropometric measurements were

recorded and the samples were fixed or frozen for further analyses.

55

0

10

20

30

40

15 20 25 30 35 40 45 50 55 60 65 70

Dis

tan

ce (

km

/day

)

Age (weeks)Figure 6. Running protocol. The dogs ran from the age of 15 weeks to the age of 70 weeks, five days a week. The daily running distance was gradually increased to the final distance of 40 km/day.

4.1.2 Canine immobilization and remobilization (II)

Thirty-four female purebred beagle dogs obtained from Marshall Farms (North Rose, NY,

USA) were used for this experiment. The dogs were kept in steel cages (0.9 x 1.2 x 0.8 m) in the

National Laboratory Animal Centre of the University of Kuopio, Finland. The beagles were fed with

commercial dog food (Hankkija, Kolppi, Finland) and water was given ad libitum. The dogs were treated

in accordance with the principles given in the National Research Council’s Guide for the Care and Use

of Laboratory Animals (National Research Council 1985). The Animal Care and Use Committee of the

University of Kuopio approved the design of this experiment.

The right hind limbs of seventeen dogs were immobilized in 90° flexion of the knee joint with a

fibreglass cast (Dynacast Pro, Smith and Nephew Medical, Hull, UK) (Haapala et al. 2000, Haapala et

al. 1999). The immobilized foot was tied to the body of the dog to prevent loading. The cast was applied

56at the age of 29 weeks for an 11-week period of experiment. At the age of 40 weeks, eight dogs were

killed using an overdose of pentobarbital (Mebunat, Orion Oy, Espoo, Finland), and nine dogs were

remobilized. The remobilization period lasted 50 weeks. The dogs ambulated freely in their cages. Age-

matched littermate dogs formed the control group for the immobilized and remobilized dogs. After the

experiment, bone samples were collected and freed from soft-tissue and thereafter frozen at -20 °C for

further analyses.

4.1.3 Canine osteotomy of proximal tibia (III)

Forty-four female beagle dogs of pure breed were purchased from Marshall Farms (North Rose,

NY, USA) consisting of 14 sister triplets and one sister twins. Littermate dogs were divided into six

experimental groups. The experimental groups included valgus osteotomy -group, sham operated and

control group with two time points, 7 months and 18 months after the operation (Panula et al. 1997).

A 30° valgus angulation of the right tibia was achieved by surgical operation in 15 dogs at the age of

three months. Fourteen littermate dogs were sham operated (tibia and fibula were osteotomized without

angulation) and fifteen sister dogs served as controls. The dogs were fed with commercial dog food

(Vaasan Mylly Oy, Vaasa, Finland) 240 g per day. Water was given ad libitum. During the experiment, the

dogs were kept under kennel conditions in the National Laboratory Animal Center (Karttula, Finland).

The Animal Care and Use Committee of the University of Kuopio approved the experimental design.

The experimental protocol complied with the principles of the National Research Council’s Guide for

the Care and Use of Laboratory Animals (National Research Council 1985).

The surgical procedure of valgus osteotomy of proximal tibia was described in detail in a previous

report (Panula et al. 1997). The osteotomy line was carried out 10–12 mm distal to the growth plate of

proximal tibia in the direction of the tibial plateau. Through a separate skin incision, 5 mm of the fibular

bone was excised at its upper third to allow the tibial osteotomy surfaces to come in contact with each

other. The osteotomy was fixed using a prebent 30° T-shaped AO plate (Stratec Medical, Waldenburg,

Switzerland) on the medial aspect of the tibia with screws.

57The operated dogs received benzylpenicillin and streptomycin antibiotics intramuscularly for

2 days after the operation. Buprenorphine was used as the postoperative analgesic. No cast was used

and the dogs were allowed to use the operated limb immediately after recovery from the anaesthesia.

During the first week after operation, the dogs gradually increased weight-bearing and after four weeks

no limping could be observed. Dogs with 30° valgus osteotomy showed a slight external rotation of the

operated limb when standing. This was not observed in the sham-operated dogs.

After euthanasia with an overdose of anaesthetic, the hindlimbs were dissected free from soft

tissues. The stability of the osteotomy line was examined after AO plate removal. The bones including

femurs and tibiae were frozen at -20 °C for further analyses.

4.1.4 Murine heterozygous knockout inactivation of one allele of the Col2a1 gene (IV)

In this experiment, two genetically different inbred mouse lines were investigated. Conventional

C57BL/6JO1aHsd breeders were obtained from Harlan CBP (Rijswijk, The Netherlands) and knockout

breeders, C57BL/6-TGN (Col2a1 heterozygous knockout) originated from the laboratory Animal

Services of Thomas Jefferson University (Philadelphia, PA, USA). The knockout breeders were

produced using a homologous recombination technique (Li et al. 1995). Breeding and the experiment

took place in the National Laboratory Animal Centre (Kuopio, Finland). Breeders heterozygous (HZ)

for the Col2a1 allele, one allele inactivated, were mated with normal C57BL/6JO1aHsd breeders to

obtain either normal or heterozygous pups. The genotype was verified using polymerase chain reaction

(PCR) from the samples taken at the age of three days (toe tips) and, for confirmation, at the age of three

weeks (tail tips). The missing toe tips served also as identification codes. After weaning at the age of

three weeks, the pups lived with their littermates for one week and were then placed in individual cages

at the age of one month. The dimensions of the polycarbonate cages were 16 cm x 32 cm (512 cm2)

floor area and 12.5 cm height. The cages were equipped with steel wire lids. Cages of the runner mice

had also stainless steel running wheels, 8 cm wide and 23 cm of inner diameter. The runner mice could

access freely the running wheels 24 hours a day. A sensor and information storage system was used to

58record the movements of the running wheels (Lapveteläinen et al. 1997). Cages and running wheels

were commercially available (Techniplast Cazzada SARL., Varese, Italy). The room temperature was

kept between 19 and 23 °C, humidity 30–70%, ventilation rate 7–18 times per hour, noise levels 40–55

dB (20–50000 Hz), and the light cycle 12 hours light and 12 hours dark. Commercial R36 mouse food

(Lactamin Ab, Stockholm, Sweden) was given and tap water were available ad libitum.

From the HZ Col2a1-knockouts and controls, the following experimental groups were formed:

normal controls, normal runners, HZ knockout controls and HZ knockout runners. The groups of mice

were sacrificed at two time points; the first at the age of 9 months (average group size 20 mice) and

the second at the age of 15 months (average group size 15 mice). The consumptions of food and water

as well as the weight development were recorded weekly. After the first time point, only the weight

development was recorded at monthly intervals.

4.1.5 Transgenic mice harbouring truncated human COL2A1 gene (V)

In this study we utilized transgenic (TG) mice, which were generated using the human type II

procollagen construct described by Vandenberg et al. (Vandenberg et al. 1991) and Helminen et al.

(Helminen et al. 1993). The mice harboured a 291 codon-long deletion mutation in the COL2A1 transgene

of human collagen type II. This caused a deletion of 12 exons of the 54 exons of the gene. Therefore, the

internally deleted gene construct coded for a shortened proα1(II) chain with an in-frame deletion of 291

amino acids (α1-157 to 447) of the 1,014-residue-long triple-helical domain. The transgenic mice were

generated and propagated in the inbred FVB/N strain.

A total number of 100 female mice were used: 48 TG and 52 control littermates (wild-type) mice,

obtained from the Thomas Jefferson University, Jefferson Medical College, PA, USA, were utilized.

They were divided into three age groups, to be analyzed at the age of one week, one month and three

months.

594.2 Methods

4.2.1 Bone densitometry

4.2.1.1 DXA (I-IV)

For canine studies (I-III), Lunar DPX bone scanner (Lunar Corp., Madison, WI, U.S.A.) was

utilized to determine the BMD of thoracic (segments C2-4) and lumbar spine (segments L6-7), iliac,

third metatarsal and radius bones (I). Left and right proximal femurs were also measured (II and III).

Two X-ray energies were used in the DXA analysis (38 keV and 70 keV). Prepared bones were dissected

free of surrounding soft tissue and frozen at -80 °C prior to the measurement. The scanner was calibrated

according to the operator’s manual provided by the manufacturer. The samples were positioned in a

plastic container filled with water to simulate soft tissue density. The depth of the water layer was 12 cm

with canine samples and 2 cm with murine samples. All measurements were performed in triplicate and

the mean value of the measurements was recorded.

The BMD of the mice (IV) was determined using the second generation Lunar Expert (Lunar

Corp., Madison, WI, U.S.A.) DXA-scanner. The densitometer was calibrated with a bone phantom every

4 hours according to the guidelines provided by the manufacturer. The results were expressed in grams

(g) per square centimetre (cm2).

4.2.1.2 QCT and pQCT (I, IV)

Trabecular BMD (I) was measured using a Siemens DR 1 QCT-scanner (Siemens AG, Karlsruhe,

Germany). The consecutive slices of 4 mm thickness were scanned with energy of 96 kV, 300 mA. For

calibration, phantoms of known density (KMnO4, H2O, and air) were also scanned. The correction of

density measurement was calculated using water (set as 0 Hounsfield units (HU)) and air (set as -1000

HU) phantoms in the equation:

60

(9)

in which HUb represents the bone density, HUw and HUa are the measured densities of water and

air phantoms, respectively. The density of a predetermined ROI was presented as the corrected HU. For

bone, the measured densities are usually over +400 HU.

In study IV, a Stratec 960A pQCT scanner was used (Norland Stratec Medizintechnik GmbH,

Birkenfeld, Germany). The femurs were aligned axially with the anterior surface upward in a custom-

made sample holder. The exact mid-point of the diaphyseal segment was ensured using the scout view.

Mean volumetric BMD, CSMI, mean cortical thickness, periosteal and endosteal circumferences were

determined using the pQCT analyzing software at an attenuation threshold value of 0.815 cm-1. Routine

calibration of the system was performed daily with a phantom sample provided by the manufacturer.

4.2.2 Bone mechanical testing (I, IV)

Biomechanical parameters of canine trabecular bone were estimated by using the compression

test (I). The test was performed using two standard locations in vertebral trabecular bone and three

locations in iliac bone (Figure 1, I). The material test apparatus and the measurement settings have been

described earlier (Räsänen et al. 1991). For the testing, a cylindrical plane-ended indenter, with a contact

area of 3.02 mm2, was employed. After the indenter was adjusted to come into contact with the sample

surface, a compression test was performed using an indenter speed of 0.166 mm/s, until the fracture

threshold was achieved. The diameter of the indenter covered up several trabeculae of the specimen.

The bending strength of mice femora (IV) was determined by the three-point bending test. The

testing device was described earlier (Jämsä et al. 1996). The whole bone samples were set on two fulcra

lying 6.5 mm apart from each other. The indenter head as well as the two fulcra were rounded to avoid

shear load and cutting of the bone during the test. The loading force was directed vertically to midshaft

of the bone. The compression continued at constant speed of 0.155 mm/s until failure.

€

HU =1000(HUb − HUw )

HUw − HUa

61The load-time curves were recorded using a standard x-y plotter (IV) or a custom-made software

application (I). The plotter curves were digitized, converted to force-deformation curves and analyzed

using an image analysis software (UTHCA ImageTools for Windows v.1.28, University of Texas Health

Science Center, San Antonio, TX, U.S.A.). The stress, strain and elastic modulus (Young’s modulus)

were calculated as follows (Turner and Burr 1993):

(10)

(11)

(12),

where σ is stress, ε is strain, c is the distance from the center of mass, F is applied force and L is the

distance between the fulcra in the three-point bending test. I refers to the CSMI determined by pQCT.

The stiffness of the sample was determined from the slope of the force-deformation curve at its elastic

region (Figure 5).

4.2.3 Quantitative X-ray imaging (I, IV-V)

X-ray images of the vertebral column, forefoot and hind limb in canine studies (I-III) were taken

after the experiment. The measurements were performed using a scaling device at a transillumination

board.

Total body X-ray images were taken from the mice in studies IV and V. The standard anterior-

posterior and lateral images were taken with 50 kV tube voltage (15mAs). The images were digitized

at 600 dpi (dots per inch) resolution using a computer equipped with a scanner (Microtek international

Inc., Düsseldorf, Germany). The anthropometric parameters including individual bone length and width

were measured with Prism (Analytical Vision Inc., Raleigh, NC, U.S.A.) and NIH Image v.1.62 (NIH,

€

σ =FLc4I

€

ε =12cd

L2

€

E =FL3

d48I

62Bethesda, MD, U.S.A.) image analysis software.

4.2.4 Bone histomorphometry (I-III, V)

Static (I-III) and dynamic (V) histomorphometry was performed using a semiautomatic

Kontron Osteoplan II image analysis system (Kontron GmbH, Munich, Germany) with Zeiss Axioskop

microscope (Zeiss, Thornwood, NY, U.S.A.). The bone samples were dissected free of soft tissue, and

fixed in 70% ethanol. After dehydration in graded alcohol solutions, the specimens were embedded in

methyl methacrylate. Sections of 5 to 7 μm in thickness were cut using a Reichert- Jung K polycut S

microtome with HK knives (Jung, Heidelberg, Germany). After dissolving the methyl methacrylate, the

sections were stained with modified Masson-Goldner trichrome stain. For dynamic histomorphometry,

fluorescent double labelling with calcein was performed by intraperitoneal administration of the agent

five and two days before sacrifice. The following data were collected: trabecular bone volume fraction

(BV/TV), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), trabecular number (Tb.N), percent

osteoid volume (OV/BV), specific osteoid surface (OS/BV), percent osteoid surface (OS/BS), osteoid

thickness (O.Th), percent eroded surface (ES/BS), specific eroded surface (ES/BV), erosion depth

(E.De), percent osteoid to eroded surface (OS/ES), osteoblast number (N.Ob/B.Le), osteoclast number

(N.Oc/B.Le). In dynamic bone histomorphometry, additional parameters including mineral apposition

rate (MAR) and bone formation rate (BFR/BV) were calculated.

4.2.5 Quantitative polarized light microscopy (PLM) (I, IV)

Quantitative polarized light microscopy was used to estimate the spatial arrangement and

organization of collagen fibrils in bone (Király et al. 1997). Five-μm-thick sections were cut from

decalcified bone samples. After deparaffinization, the PGs were removed with testicular hyaluronidase

digestion. The unstained sections were analyzed with a Leitz Ortholux II Pol BK2 microscope (Leitz,

Wezlar, Germany) using a 16/0.45 N.A. strain free objective. For PLM analysis, light with the wavelength

of 591.4 nm was used and adjusted by a monochromator filter (IL 591.4 nm, t 1/2–width 19.3 nm; Schott,

63Iserlohn, Germany). Digital images were taken for the birefringence analysis using a Photometrics CH-

250 CCD-camera (Photometrics, Tuscon, AZ, U.S.A.). The ROI was selected randomly by rotating the

stage of the microscope where the section was placed. Values for birefringence were determined as a

mean of the filtered light intensity (area-adjusted integrated retardation value of polarized light, AIR).

The measurements were performed twice and the mean values were used for calculation.

4.2.6 Elemental composition of bone samples (I, II)

The second thoracic vertebral body, a standard region of frontal bone and the proximal head of

the ulnar bone in study I, and a standard region of trochanter major in study II were assayed for their

elemental composition, carbonate and citrate. The bones were macerated in chloroform-methanol (2:1)

to remove fat. The samples were dried at 105 °C and pulverized. Citrate was released with 2 M H2SO4

for 1 h at +60 °C and the samples were analyzed after neutralization using a commercial kit (Boehringer

Mannheim, cat. no. 139076, Mannheim, Germany). Fifty mg of the bone sample was ashed at +500 °C

for 10 h. After ashing, samples were dissolved 4x250 μl 3 M HNO3 and diluted to 10 ml. Ca, Mg, Zn,

K, and Na were determined by atomic absorption spectrophotometry (Perkin Elmer model 372, Perkin

Elmer Corp., Norwalk, CT, USA). Perkin Elmer Intensitron hollow cathode lamps were used and the

operating parameters in the linear working range were followed as described in the manufacturer’s

operating manual. The reproducibility in the Ca and Zn determinations was within 2%. The recovery

rates varied from 99.1% to 99.7%.

4.2.7 Statistical analyses (I-V)

Data distributions for each parameter in this study are expressed as mean ± standard error of the

mean (SE) or SD. Non-parametric Wilcoxon matched-pairs signed-ranks test and the two-tailed Mann-

Whitney U-test were used to calculate statistical significance of differences between the experimental

and control groups (I-III). The Mann-Whitney U-test was also used in studies with mice (IV-V), because

the normal distribution of the data could not be assumed on account of the relatively small sample size.

64PLM results were tested using the Kolmogorov-Smirnov test, which is more sensitive to differences

in the general shapes of the data distributions between two samples. Pearson’s correlation coefficient

was calculated to evaluate the correlation between the biomechanical properties and the anthropometric

parameters in study IV. Various versions of statistical software package (SPSS for Windows, SPSS Inc,

Chicago, IL, U.S.A.) were used in calculation. P values less than 0.05 were considered statistically

significant.

65

5 RESULTS

5.1 Effect of long-term training on canine bone (I)

All dogs were able to complete the training regimen and the physical condition of the animals

remained good throughout the study. The growth plates of the long bones had closed by the end of

the study period in both groups. The X-rays of spines and extremities were normal in all animals. The

weight-bearing bones were larger while the vertebrae were smaller in the runner dogs as compared with

the controls (Table 3).

Table 3. Anthropometric measurements (mean ± SE).

Parameter Controls (n=10) Runners (n=10)

WeightBody (kg) 12.4 ± 0.3 12.3 ± 0.3Radius bone (g) 12.8 ± 0.4 14.4 ± 0.6Radius bone/body wt (g/kg) 1.05 ± 0.02 1.17 ± 0.04 *Ulnar bone (g) 14.4 ± 0.3 15.5 ± 0.6Ulnar bone/body wt (g/kg) 1.16 ± 0.03 1.26 ± 0.04 *Hip bone (hemipelvis) 37.5 ± 1.9 39.0 ± 2.6Th 2 vertebra 2.47 ± 0.10 2.45 ± 0.12

Length (cm)Radius 11.5 ± 0.2 12.0 ± 0.5Ulna 13.9 ± 0.3 14.4 ± 0.2

Height (mm)Th 2 14.3 ± 0.28 14.1 ± 0.28Th 3 14.0 ± 0.32 13.7 ± 0.26Th 4 14.0 ± 0.28 13.7 ± 0.15

Radius diaphysisCross-sectional area (cm2) 0.90 ± 0.03 0.96 ± 0.03Cortical bone area (cm2) 0.67 ± 0.01 0.72 ± 0.02Cortical index (%) 74.63 ± 1.71 74.56 ± 1.57* p < 0.05 between runners and controls

66The density distribution of vertebral bodies differed between the runners and the controls (Figure

7). The shapes of the curves in mid-frontal and mid-sagittal planes were similar in both groups in thoracic

and lumbar spines. Higher bone densities were observed in cortices and posterior parts of the vertebral

segments. The BMDs were significantly lower in thoracic and lumbar spine, ilium, radius and MT III in

the runner dogs as compared with their littermate sisters (Figure 8).

Interestingly, the biomechanical properties of trabecular bone of vertebrae or iliac bone revealed

no differences between the study groups. However, the ultimate fracture force needed to break the

vertebral bodies was lower (p < 0.05) than that of iliac crest in the runner dogs (Figure 9).

Zinc and sodium accumulated in the bones of the trained animals (Table 4). BMDs correlated

significantly with bone ash weight ( r= 0.62, p < 0.01) and zinc content (r = -0.54, p < 0.05), but not

with calcium concentration (r = 0.30). The amount of urinary calcium and the concentration of serum

alkaline phosphatase were significantly higher in the exercised group (Table 5). Other markers of bone

metabolism did not show any differences between groups. Serum estradiol levels were lower in the

runner dogs at the age of 55 weeks.

Thoracic vertebra 2 Ulnar bone Frontal boneControls (n=10)

Runners (n=10)

Controls (n=10)

Runners (n=10)

Controls (n=10)

Runners (n=10)

Dry wt/ash wt 1.97 ± 0.039 1.91 ± 0.036 1.60 ± 0.013 1.57 ± 0.007 1.56 ± 0.012 1.56 ± 0.018Ca/dry wt (mg/g) 167 ± 3 164 ± 4 170 ± 3 160 ± 4 215 ± 3 213 ± 2Ca/ash wt (mg/g) 319 ± 2 322 ± 3 303 ± 3 309 ± 4 336 ± 3 332 ± 2Mg/ash wt (mg/g) 5.90 ± 0.07 5.84 ± 0.07 5.27 ± 0.03 5.18 ± 0.05 5.28 ± 0.06 5.33 ± 0.07K/ash wt (mg/g) 5.73 ± 0.3 5.92 ± 0.1 6.91 ± 0.3 8.47 ± 0.6 * 4.86 ± 0.2 4.78 ± 0.2Na/ash wt (mg/g) 10.2 ± 0.15 11.5 ± 0.27 ** 8.8 ± 0.09 9.3 ± 0.14 * 9.6 ± 0.14 9.2 ± 0.06 **Zn/ash wt (mg/g) 0.23 ± 0.007 0.26 ± 0.007 ** 0.19 ± 0.008 0.22 ± 0.009 * 0.16 ± 0.005 0.18 ± 0.005 *F/ash wt (mg/g) 0.22 ± 0.005 0.22 ± 0.001 0.17 ± 0.004 0.17 ± 0.001 0.15 ± 0.006 0.14 ± 0.006Citrate/dry wt (μg/g) 7.4 ± 0.27 7.5 ± 0.23 9.5 ± 0.37 9.5 ± 0.44 6.9 ± 0.41 8.7 ± 0.31 *Carbonate (μg/g) 33.1 ± 0.7 31.2 ± 0.6 38.7 ± 0.7 34.5 ± 0.7 46.1 ± 0.9 46.5 ± 0.7*p < 0.05, **p < 0.01 between controls and runners (Wilcoxon matched-pairs signed-ranks test).

Table 4. Elemental composition of bones (mean ± SE)

67

Figure 7. The density distribution in thoracic and lumbar vertebrae measured from consecutive QCT sections data at the mid-frontal and mid-sagittal planes at right angles to each other (shown in picture). The values are expressed in adjusted Hounsfield’s units (HU) with standard gray-scale windowing of the actual HUs, in which +500 HU equals the adjusted value of 255 an 0 HU equals to value 0. The lengths of the planes were adjusted to be comparable with each other (percentage scale of the plane). Kolmogorov-Smirnov test was used to test the similarity of the density distributions. ***p < 0.001.

aHU

68

Figure 8. Comparison of bone mineral density in thoracic (Th) and lumbar (L) spine, radius, ulna, third metatarsal (MT III) and iliac bone. The runner values are given as percentage difference (mean ± SE) compared with their littermate sisters. The measurements were carried out with quantitative computed tomography (QCT) and dual-energy x-ray absorptiometry (DXA). *p < 0.05, **p < 0.01.

-15 -10 -5 05 10 15

QCT: Th

QCT: L

DXA: Th

DXA: L

DXA: Ulna

DXA: Radius

DXA: MT III

DXA: Ilium

Difference (%)

Site

*

*

**

**

There were no differences in bone histomorphometry (Table 3, I) or in the concentration of

collagen crosslinks (Table 2, I). However, quantitative PLM showed higher organization of the collagen

matrix in the trabecular bone of the vertebral bodies (Figure 2, I)

0

10

20

30

40

Runners Controls

Ultim

ate s

tres

s (M

Pa)

Lumbar vertebraIliac crest

Figure 9. Ultimate stress at fracture of canine trabecular bones of the lumbar vertebra and the iliac crest of runners and age-matched controls at the age of 70 weeks. Mean ± SEM. Wilcoxon matched pairs signed ranks test. *p < 0.05.

69Table 5. Serum and urine calcium, serum alkaline phosphatase (S-ALP), serum estradiol, osteocalcin (S-OC), carboxyterminal propeptide of type I procollagen (S-PICP) and cortisol in the beginning and at the end of the 40 km/day exercise period, i.e. at 55 and 70 weeks). Modified from Puustjärvi et al. (1995).

Parameter Runners (n=10) Controls (n=10)mean ± SEM mean ± SEM

S-Ca (mmol/l)55 weeks 2.66 ± 0.03 2.74 ± 0.0470 weeks 2.55 ± 0.03 2.54 ± 0.03

U-Ca (mmol/day)55 weeks 0.67 ± 0.08 0.39 ± 0.0870 weeks 1.21 ± 0.10 0.79 ± 0.12

S-ALP (U/l)55 weeks 236.6 ± 22.5** 109.7 ± 7.770 weeks 123.7 ± 8.7** 86.6 ± 9.2

S-estradiol (nmol/l)55 weeks 0.021 ± 0.003* 0.036 ± 0.00970 weeks 0.022 ± 0.004 0.027 ± 0.006

S-OC (ng/ml)55 weeks 0.29 ± 0.02 0.36 ± 0.0370 weeks 0.58 ± 0.07 0.64 ± 0.07

S-PICP (μg/l)55 weeks 5.3 ± 0.5 5.3 ± 0.570 weeks 7.7 ± 1.5 6.4 ± 0.8

S-Cortisol (nmol/l)55 weeks 6.3 ± 1.0 5.0 ± 1.570 weeks 5.4 ± 1.6 6.8 ± 1.7

*p < 0.05 and **p < 0.01.

705.2 Effects of immobilization and remobilization on canine bone (II)

After casting of the lower limb, the dogs soon ambulated actively in their cages. Splinting did

not reduce the range of motion of the knee joint. One week after the remobilization, the gait of the

remobilized dogs appeared similar to that of controls (Haapala et al. 2000). There were no changes in

overall growth or weight between the groups (Table 6). The cast-immobilization for a period of 11 weeks

caused a significant site-dependent 21.8–27.4% bone loss in the casted proximal femur (p < 0.01) and

also systemic 4.7–10.9% decrease of BMD in the contralateral limb (Table 3, II) in comparison with the

age-matched controls. These changes had recovered after the remobilization period of 50 weeks.

Table 6. The weight (kg ± SD) of dogs at the beginning of the immobilization (29 weeks), at the beginning of the remobilization (40 weeks), and at the end of the remobilization period (90 weeks).

Age (weeks)29 40 90

Immobilized (n=8) 9.8 ± 1.8 10.2 ± 2.1Controls (n=8) 9.7 ± 1.7 10.1 ± 1.9

Remobilized (n=9) 8.9 ± 1.9 9.2 ± 1.4 10.1 ± 2.1Controls (n=9) 9.1 ± 1.2 9.3 ± 1.3 9.8 ± 1.6

Immobilization induced accumulation of zinc not only in the splinted limb but also on the

opposite side (p < 0.05). However, the changes became normalized after remobilization. The calcium

concentration was 9.3% lower in immobilized bone than in the contralateral weight-bearing bone (p

< 0.05). After the remobilization period of 50 weeks, there were no differences in immobilized and

weight-bearing bones with regard to the concentrations of calcium, phosphorus, sodium and zinc (Table

2, II). There was a positive correlation between BMD and bone calcium content (r = 0.453, p < 0.001)

and Ca/P-ratio (r = 0.461, p < 0.001), while bone phosphorus, sodium and zinc concentrations showed

a negative correlation with BMD (r = -0.466, p < 0.001; r = -0.324, p < 0.01 and r = -0.263, p < 0.05,

respectively). The calcium concentration of bone correlated negatively with sodium (r = -0.610, p <

0.001) and zinc (r = -0.353, p < 0.01) contents.

71Bone histomorphometry revealed osteopenia in the bones of splinted dogs. The bone volume (BV/

TV) was 21.1% lower and the trabecular separation was 8.3% wider in the immobilized bone as compared

with the bone of the contralateral limb (Table 4, II). Bone turnover was increased in the immobilized

bone, since bone formation and resorption had both increased when compared with the weight-bearing

bone of the opposite side. Remobilization reversed the immobilization induced osteopenic changes in

bone histomorphometry.

5.3 Effects of proximal tibial osteotomy on canine bone (III)

The BMD of the proximal femur decreased in the right (operated) limb after osteotomy as

compared with the contralateral limb of age-matched controls (Figure 10). At the same time, the BMD of

the contralateral limb increased as compared with the controls. Most changes at 7 months after operation

were seen in the proximal part of the femur, in the caput and collum, where the percentage of the

trabecular bone is greatest. The decrease of BMD was seen later (18 months after the operation) also

in the metaphyseal and diaphyseal ROIs, where the rate of bone turnover is somewhat lower. Similar

changes in all ROIs were seen in the sham-operated dogs, although to a lesser extent.

Data about the structure and turnover of subchondral bone are presented in Table 1 (III). The

differences of bone structure parameters between operated and control dogs did not reach statistical

significance except for the number of trabeculae (Tb.N) in the medial femoral condyle, where the number

decreased in the operated animals as compared to controls (p < 0.05). This was associated with a non-

significant decrease in the bone volume (BV/TV). Most changes in the femoral bone histomorphometry

were seen in the subchondral region of patellar surface, where the osteoid thickness (O.Th), osteoblast

surface/bone surface (OS/BS), erosion depth (E.De) and active erosion depth (E.De[Oc+]) indicated

increased bone remodelling in subchondral bone in the operated animals as compared with the controls

(p < 0.05).

72

Figure 10. Bone mineral density measurements determined by dual-energy X-ray absorptiometry from caput, collum, metaphysis and diaphysis of femur at the age of 7 (left triplets) and 18 months (right triplets) after osteotomy. Wilcoxon matched-pairs signed-ranks test. *p < 0.05.

5.4 Effects of heterozygous knockout of one allele of the Col2a1 gene on murine

bone (IV)

The overall phenotype appearance of the HZ knockout (HZK) and wild-type (wt) mice was

similar. The body weights of the normal runners were 2–10% lower than those of the controls (p < 0.01

at the age of six months). No differences were seen in the lengths of the body or tail. The female mice

had to some extent longer and narrower femurs as measured from X-ray images (Sahlman et al. 2001).

The daily running distance increased continuously during the first three months in HZK and wt

runners in conjunction with the size of the mice. There was a marked difference in the running activity,

speed and cumulative running distance in HZK mice as compared with the wt runners (Figure 1, IV)

indicating that the HZK mice were less willing to run. The difference was greatest from the age of 9 to

11 months (p < 0.001). The average running speed was 6.9% lower in the HZK mice as compared with

7 months 18 months

7 months 18 months7 months 18 months

7 months 18 months

73their wt controls. The HZK runners spent 114 hours less in the running wheel during the 9 months as

compared with wt runners (p < 0.05). A remarkable decline in daily running activity was seen after the

age of 4 months in both groups. The overall running activity took place at the dark hours of the day as

would be expected from typical murine behaviour.

DXA showed 6.1% lower BMD in the male transgenic runners than in the group of normal

runners at the age of 9 months (p < 0.05). The measured BMD was higher in the normal male runners as

compared with their normal (non-runner) controls (p < 0.01), whereas there were no differences between

the HZK runners and HZK controls (Table 1a and 1b, IV).

The pQCT-results reflected the changes of DXA, even though they measured primarily the cortical

bone densitiy (Table 1a and 1b, IV). The maximal breaking forces were 14.8% and 23.5% lower at the

age of 9 and 15 months in the HZK male runners as compared with the control male runners (p < 0.05

and p < 0.01, respectively). The breaking forces were 12.6% and 12.1% higher in the group of wt male

runners at the age of 9 and 15 months than the corresponding values in the sedentary male wt controls (p

< 0.05 and p < 0.01, respectively). This adaptation to an increased mechanical load was not seen in the

HZK mice (Figure 2, IV). Anthropometric parameters seemed to reflect better the mechanical properties

of bone than the density measurements (Table 2, IV). However, if the HZK mice are omitted from the

analysis, the BMD measured using DXA correlated well with the breaking force (r = 0.419, p < 0.001).

Quantitative PLM revealed a 13.9% reduced degree of organization of the bone collagen fibril

network in HZK runners than in the wt runners at the age of 9 months (p < 0.05, Figure 3, IV). This

difference could not be observed anymore at the age of 15 months. Aging from 9 to 15 months decreased

the AIR values of the normal runners by 46%, and 37% of the normal controls. Similar changes in AIR

values were also seen in the HZK mice (32% and 28%, respectively).

The HPLC analyses of bone collagen did not detect any significant differences between the study

groups in total collagen amount or in the crosslinking of the collagen fibrils (Table 3, IV).

745.5 Effects of human COL2A1 transgene on murine bone (V)

TG mice carrying an internally deleted human type II collagen gene (COL2A1) were smaller in

terms of weight and body size at the end of 3-month-long observation period (Figure 11, Figure 1 and

Table 1, V). The cortical bone area of TG mice was thinner than that of the wt controls (p < 0.05, Figure

12). The human type II collagen was actively expressed in the mouse tissues (Figure 2, V).

Dynamic bone histomorphometry revealed a nearly two-fold increase of the trabecular bone

volume in the wt mice from the age of 1 week to the age of 3 months (Figure 4 and 5, V). The mean

trabecular thickness increased concurrently by almost three-fold. No normal bone maturation process

was seen in the TG mice, and the bone mass remained almost unchanged during the weeks of bone

maturation leading to a difference of 47% in trabecular bone volume (p = 0.012) and 40% in trabecular

thickness (p < 0.01) at the age of three months as compared with the control mice. The osteoid volume

was lower in the TG mice than in the wt controls at the age of one week and one month, but the difference

was not statistically significant. Bone erosion was greater at each time interval in the transgenic mice

(Figure 5, V). At the age of three months, the specific eroded surface per bone volume (ES/BV) was 31%

greater in the TG mice as compared with the wt littermates (p < 0.05).

A normal remodelling pattern and bone matrix mineralization was seen in both groups at all

age intervals. At the two youngest age points, the mice in both groups produced proportionally more

woven bone as compared with the pattern at the age of three months (p < 0.001). The qualitative analysis

revealed that the production of woven bone was more pronounced in the TG mice. The collagen network

in woven bone is immature and loosely organized. In contrast, normal lamellar bone contains highly

organized collagen fibers in a parallel manner when analyzed using PLM.

The trabecular thickness correlated positively with the body weight (r = 0.71, p < 0.001).

Interestingly, the body weight correlated with bone volume in the wt mice (r = 0.27, p < 0.01) but

no correlation was observed in the TG mice. Trabecular thickness decreased and trabecular separation

increased at the age of 3 months in TG mice as compared with the age-matched wt controls. The femoral

bones of the TG mice were weaker in the three-point bending test (Table 7).

75Table 7. Selected bone properties of three-month-old female mice.

3 months

Control (n=8-17) Transgenic (n=7-14)

Mean ± SD Mean ± SD

FEMUR - wholeWet weight (mg) 58.9 ± 1.5 49.9 ± 1.6 ***

Wet weight/10 g body weight (mg) 26.2 ± 1.5 24.9 ± 1.8Ash weight (mg) 22.7 ± 1.3 18.3 ± 0.9 ***Ash weight/10 g body weight (mg) 10.1 ± 0.8 9.1 ± 0.6 **

Ash weight/bone weight (%) 39.2 ± 2.8 38.2 ± 3.5

FEMUR - diaphysis

Breaking force (N) 15.5 ± 0.7 14.4 ± 1.0Displacement at break (mm) 0.75 ± 0.11 0.62 ± 0.05 ***

Bone Siffness (x103 N/m) 70.0 ± 8.2 64.5 ± 23.9Energy for bone break (x10-3 Nm) 8.2 ± 1.3 6.1 ± 0.3 ***

HUMERUS - wholeWeight (mg) 31.6 ± 1.5 24.1 ± 1.0 ***Weight/10 g body weight (mg) 14.1 ± 0.9 11.8 ± 1.1 ***Ash weight/10 g body weight (mg) 6.1 ± 0.4 5.3 ± 0.4 **

Ash weight/bone weight (%) 43.6 ± 2.1 44.2 ± 1.7

*p < 0.05, **p < 0.01, ***p < 0.001 Transgenic vs. Control.

76

Figure 11. Newborn pups (a) and three-month-old mice (b). Transgenic mice harbouring internally deleted human type II collagen gene COL2A1 (above) are smaller than their age-matched wild-type mice (below). Scale (cm).

77

Figure 12. Diameter, cortical bone area and thickness values from femoral diaphysis of three-month-old female mice determined after the three-point bending test. “Load” indicates direction of the indenter in relation to bone during three-point bending test.

Wild type Transgenic

78

6 DISCUSSION

The main objective of this study was to characterize the quality and mechanical strength of

bone subjected to loading and unloading. On the other hand, the effects of defective type II collagen

expression on mechanical loading of bone were studied using normal and transgenic animal models

harbouring altered type II collagen genes. Animal models are one way to reliably control the loading

effects of physical exercise or immobilization on bone. The material used in this study was a spin-off

from a research line focusing on the development and prevention of OA at the Department of Anatomy,

Institute of Biomedicine, University of Kuopio. In this study, we predefined that the material would

consist of bone tissue, not cartilage, and we did not attempt to study the effects of various medical agents

on bone metabolism.

OP and OA are two major and increasing health problems in the industrialized countries. The

diseases lead to pain and loss of ability to work and function in everyday life. Although bone loss is

a systemic condition in the skeleton in OP, the most frequent fractures occur at the hips, vertebrae,

shoulder and distal forearms. If one considers all osteoporotic fractures, then the hip fractures are the

most common cause of mortality and morbidity; 12—20% of hip fracture victims die within the first year

of the fracture event and more than 50% of the survivors will be incapacitated (Consensus Development

Conference 1991, 1993). Our possibilities for preventing bone loss and fractures are limited. In order to

develop more effective treatment modalities, we need more information about the pathophysiology and

pathogenesis of the diseases.

6.1 Effect of functional loading on bone (I)

The positive effects of physical training are well documented. However, strenuous long-

term exercise may evoke adverse effects on the skeleton, as we have reported here. In study I, long-

term strenuous exercise had a negative effect on BMD of the axial skeleton while the bone mass of

appendicular skeleton increased. The remodelling of bone in these dogs was increased in the subchondral

bone of the knee joint as was reported earlier (Oettmeier et al. 1992). Active recreational sports and

79gymnastics regimes often train youngsters with a relatively high intensity today while the athletes are

still in their adolescence. The effects of these programs on the developing locomotor system are not well

known. Very few of the reports are longitudinal or they are on intensive exercise regimes. Even though

exercise provides an effective way of enhancing bone mass, thus delaying the time taken to reach the

postmenopausal fracture threshold, the benefit of exercise is lost if the training intensity is excessive

leading to amenorrhoea or microdamage of bone and ultimately to loss of BMD (Bourrin et al. 1992,

Christo et al. 2008, Drinkwater et al. 1984, Marcus et al. 1985, Rockwell et al. 1990).

Amenorrhoea in athletes is associated with a reduction in trabecular, but not in cortical, BMD

(Drinkwater et al. 1984). In this experiment, the estradiol levels were lower in the trained beagles

at the age of 55 weeks, even though dogs in estrus were excluded from the sample collection. It is

possible that the lower estrogen levels at the age of 55 weeks reflected a maturational delay in the

trained animals as compared with their untrained littermate sisters. However, at the end of the study,

no significant differences were noted in the hormonal balance between the trained and control groups.

On the other hand, in two earlier studies it has been shown that ovariohysterectomy in beagles does

not induce significant alterations in bone mass or histomorphometry (Boyce et al. 1990, Shen et al.

1992). Therefore, mechanisms other than estrogen deficiency, such as the local adaptation of bones to

biomechanical forces, are suggested to be more important in causing the observed decrease in BMC and

alterations in the elemental composition. The type of exercise, its intensity and the anatomical site where

the stress is applied are important.

Bone turnover rate is considered to contribute to the variations of BMDs in the examined bones.

Trabecular bone responds more rapidly to metabolic changes than compact bone. Therefore QCT was

a more sensitive method than DXA for detecting changes in BMD, since the QCT measurements were

carried entirely out of the trabecular bone of the vertebrae, while the DXA results represented a mean

value for the total bone image under the scanner, which also included the cortical bone. The mean

turnover rate of canine trabecular bone is reported to be 134% ± 94% per year (Kimmel and Jee 1982).

In this study, a decrease in BMD was found in canine bones with a high turnover, e.g., os ileum and

80vertebral bodies.

The inverse correlation between BMD and fracture risk is reported to be high (Alhava 1991, Cody

et al. 1991, Hansson et al. 1987, Kröger et al. 1995a, Ott 1993). However, there are reports suggesting

that the density measurements do not always relate to the biomechanical properties of trabecular or

cortical bone (Carbon et al. 1990, Hui et al. 1988) and that other factors in addition to BMD contribute

to the mechanical behaviour of bone. For example, most studies ignore the contribution of the collagen

component in the determination of bone strength (Boskey et al. 1999).

In this study, PLM of bone revealed a more organized collagen network structure in the runner

dogs than in their sedentary littermates. An improved organization may also affect mineralization pattern

of the bones, since the hydroxyapatite crystals are arranged parallel to the long axis of the collagen

fibrils (Bagambisa et al. 1993). However, bone histomorphometry detected no significant differences

between the two groups in this respect. This study suggests that the higher organization of collagen fibrils

maintained the biomechanical properties of the bone despite the lower BMD in the group of runner dogs.

Bone histomorphometry showed no significant changes in the cellular response of the vertebral

bodies to long-term exercise. Dynamic analysis using tetracycline labelling would probably have

provided more information on the mineralization of bone, but unfortunately this was not originally

included in the study protocol.

Since the training took place when the dogs were still growing, the exercise may have caused

alterations in the maturation of the different tissues. However, the dogs had ceased to grow when the

most intense 40 km/day run started at the age of 55 weeks. Therefore the effect should be comparable to

those encountered in young adolescents in modern sports.

6.2 Effect of immobilization on bone (II)

Immobilization-induced osteopenia is a common adverse effect of fracture treatment. This

results in site-specific bone loss in the unloaded limb. However, our study showed also bone loss in the

contralateral limb in conjunction of the cast-immobilization. This finding is in accordance with earlier

81reports (Mueller et al. 1992). Nonetheless, after a relatively long remobilization period (50 weeks), the

difference in BMD between the splinted and weight-bearing contralateral limb could still be seen.

Bone histomorphometry showed increased bone formation and increased bone resorption of

trabecular bone after immobilization. The higher bone turnover led to a remarkable bone loss during

the immobilization period of 11 weeks. Remobilization reversed the situation and there were no

differences in bone histomorphometry between the remobilized and control dogs. However, dynamic

bone histomorphometry would perhaps have provided more detailed information about the rate of

bone mineralization, but, unfortunately, the administration of fluorescent labels was not included in the

original study protocol. Our results of bone histomorphometry after immobilization are in agreement with

previous findings. Decreased bone volume was associated with the increased trabecular bone separation

and a higher bone turnover, mainly due to unbalanced increase in bone resorption (Thomsen et al. 2005).

The accumulation of zinc in bone tissues of immobilized animals probably reflects the alterations

in bone turnover, since zinc acts as a cofactor of ALP. Furthermore, zinc is required as a catalytic factor

for MMPs that degrade matrix proteins. Thus both bone formation and resorption appear to eventually

increase the zinc concentration in bone tissue.

6.3 Effect of proximal tibial osteotomy on bone (III)

The effects of proximal tibial valgus osteotomy on bone density and subchondral bone parameters

were studied with a uniform group of beagles. The littermate sisters served as age- and sex-matched

control animals.

As a consequence of tibial valgus osteotomy, the tibiofemoral joint line was inclined laterally

and also the mechanical axis of the operated limb shifted laterally. These alterations, together with a

slight external rotation of paws of the operated limbs, resulted in a deranged pattern of joint loading in

the knee joint, and subsequently cartilage failure (Panula et al. 1997). The evolution of degenerative

cartilage defects due to altered joint biomechanics has been observed in several studies (Cooke 1985,

Roach and Tilley 2007, Takai et al. 1985). The mean valgus tibial angle was 8° ± 1° in group of controls,

8229° ± 5° in osteotomized and 15° ± 3° in sham operated dogs (Panula et al. 1997). Thus, in a few sham-

operated dogs, the valgus angle increased during the experiment, apparently on account of the fact that

the osteotomy fixation was not sufficiently rigid to maintain the original position.

The gait pattern and weight-bearing of the limbs of the operated and control dogs after the

operation were examined by using an electromechanical film technique to register the forces under the

paws (Heikkinen et al. 1997). One month after operation, the animals were bearing 70% of the original

weight with the operated limb. Three months after the operation almost no difference in weight-bearing

could be detected between the operated and the contralateral hindlimbs.

The histomorphometry of subchondral bone of femur showed increased remodelling in this study

(III). The changes were the greatest in the patellofemoral subchondral bone area suggesting an altered

loading mechanism and/or a greater contact pressure at the patellofemoral joint. The changes in the

BMDs of proximal femur indicated that alterations in the bone microarchitecture had taken place. It has

been suggested that the primary OA can initially be also a bone disease rather than a cartilage disease

(Lajeunesse and Reboul 2007). Previously, relatively little attention has been paid to changes in the

subchondral bone. Bone changes have been considered to be secondary to the primary lesions observed

in the cartilage. However, mechanical stress on weight-bearing joints may cause microfractures in the

subchondral bone plate and in overlying cartilage. Progress of the sclerosis of subchondral bone plate

increases bone stiffness, which, in turn, may contribute to injury and erosion of the cartilage. Also,

healing of trabecular micro-fractures in the subchondral bone can result in a stiffer bone that no longer

functions as an effective shock absorber (Radin et al. 1970, Radin and Rose 1986).

6.4 Effect of altered or disturbed synthesis of type II collagen on bone (IV, V)

The effect of heterozygous inactivation of one allele of the type II collagen and incorporation to

the murine genome of a transgene with a truncated human COL2A1 gene, was studied to find out the kind

of role that type II collagen plays in bone development, modelling and remodelling.

Although type II collagen is present only in a tiny amount in mature bone, it does have an

83important role during bone development. During the maturation process, the cartilaginous model of bone

consists mainly of type II collagen and it ossifies through endochondral ossification.

The physiological adaptation of bone to increased mechanical load, by increasing the bone mass

or by arranging bone geometry, was not seen in the mice with HZ inactivation of Col2a1 gene. The

running exercise caused an increase in the trabecular bone turnover. The production of type II collagen

and probably also of PGs was disturbed. The PGs may have a specific role as a strain memory in the bone

tissue and, thus, having a guiding influence on bone remodelling and organization (Skerry et al. 1988).

No significant differences were seen in the biomechanical parameters between the sedentary HZ

Col2a1 knockout mice and their wt male controls. However, physical activity induced a significant

decrease in the bone fracture resistance and stiffness among the HZ Col2a1 knockout mice, pointing

to an important role for type II collagen not only in chondrogenesis but also in bone maturation

and remodelling. The synthesis of type II collagen may be inadequate in knockout mice with a high

remodellation rate, but the difference was further more balanced with advanced age.

Female HZ knockout mice and their wt sisters, on the other hand, had surprisingly high BMD and

resistance to fractures compared to the male mice at the age of 9 months. Nevertheless, the difference

seemed to diminish in older animals. This finding is in accordance with previous studies on female

rodents and may be related to the production of estrogens (Järvinen et al. 2003a, Järvinen et al. 2003b,

Wang et al. 2003). The production of estrogens by the female mice may contribute to the deposition of

more minerals to bones during the growth phase.

6.5 Methodological discussion

DXA has become the golden standard in bone densitometry. It has proven to be reliable and

precise method in the evaluation of the BMD. However, DXA measures bone mass in relation to the two-

dimensional area in the path of X-rays. Any structural anomaly in the scanned area, bone exostosis or

rotational malalignment may skew the BMD values. Thus, the settings of bone density measurement are

critical (Bolotin et al. 2003, Bolotin et al. 2001). In addition, the role of the non-mineralized matrix in

84the BMD measurement is not totally understood. The variation in the collagen content and arrangement

may possibly affect the results of bone densitometry. In this study, we revaluated the use of a second-

generation DXA device for the determination of BMD in small animals. However, the mineral content of

mouse bones was on the edge of the device sensitivity. Thus, when the number of specimens is limited,

the use of DXA to analyze small animal specimens is not advisable. In QCT, an increased collagen

concentration of bone may overestimate the BMD (Goodsitt et al. 1988).

In this study, we utilized the three-point bending test to determine the biomechanical properties

of murine bones. While this method is valid and well documented, it may give rise to methodological

problems. In nature, diaphyseal bending forces sufficient to fracture bone occur only rarely without

considerable shear stress. Usually the fractures emerge from a combination of bending, compressive and

torsional loads.

Animal models have been widely used to study basic phenomena involved in OP and OA.

Even though the results have species-specific features, these studies contribute to the knowledge of the

biology of bone and cartilage and may provide a fundamental basis for developing new treatments in the

disorders of these tissues in humans. The statistical power of the results of the present study could have

been stronger if the number of the experimental animals had been larger, especially in the experiments

studying the loading effects on canine bone.

The genetic variability among mouse strains may have a major effect on bone mineral density

(Beamer et al. 1996, Li et al. 2001). For example, the C57BL/6J mouse strain used in this study is a strain

with one of the lowest bone mineral densities. Interestingly, similar variations in BMD and incidence of

osteoporotic fractures have been observed also between different human races (Thomas 2007).

85

7 SUMMARY AND CONCLUSIONS

In this study, five canine and murine models were utilized to evaluate the effect of functional

loading on bone remodelling. This thesis work demonstrates that:

1. Long-term running training of female dogs resulted in reduction in BMD in association with a

slightly lower estradiol levels. However, the bone biomechanical properties remained unchanged

possibly due to a higher degree of organization of the collagen matrix.

2. Immobilization induced not only a site-specific reduction of BMD in the casted limb but also

a decrease in the contralateral limb. Osteopenia was associated with increased bone formation

and bone resorption in the immobilized limb, which reflected also the accumulation of zinc in

bone. The bone changes were almost completely reversible after one-year of remobilization, but

a difference in BMD between the immobilized and the contralateral limb could still be detected.

3. Valgus osteotomy of the proximal tibia led to an increased turnover in the subchondral bone.

Reduced BMD also in the proximal femur of the osteotomized limb was registered.

4. Heterozygous knockout inactivation of one allele of the Col2a1 gene in association with running

training of transgenic mice disturbed normal bone remodelling. The mutation also disturbed

growth and produced a less organized collagen network when examined with the polarized light

microscopy. The biomechanical properties of bone were also impaired.

5. Truncated human type II collagen transgene in mice increased bone resorption and decreased

bone formation. The transgenic mice were smaller and the biomechanical properties of bones

were impaired.

86

REFERENCES

Abbott LG and Rude RK. Clinical manifestations of magnesium deficiency. Miner Electrolyte Metab 1993;19:314-322.

Alberts B, Johnson A, Lewis J, Raff M, Roberts K and Walter P, Eds.: Molecular biology of the cell. Garland Science, New York, 2002.

Alhava EM. Bone density measurements. Calcif Tissue Int 1991;49 (Suppl):S21-S23.

Arokoski J, Kiviranta I, Jurvelin J, Tammi M and Helminen HJ. Long-distance running causes site-dependent decrease of cartilage glycosaminoglycans content in the knee joints of beagle dogs. Arthritis & Rheumatism 1993a;36:1451-1459.

Arokoski J, Kiviranta I, Laakkonen J, Tiihonen A, Sopanen E, Ikonen J, Nevalainen T, Haapanen K, Tammi M and Helminen HJ. A ten-track treadmill for running training of dogs. Lab Anim Sci 1991;41:246-250.

Arokoski J, Miettinen PVA, Säämänen A-M, Haapanen K, Parviainen M, Tammi M and Helminen HJ. Effects of aerobic long distance running training (up to 40 km/day) of 1-year duration on blood and endocrine parameters of female beagle dogs. Eur J Appl Physiol 1993b;67:321-329.

Aszódi A, Chan D, Hunziker E, Bateman JF and Fässler R. Collagen II is essential for the removal of the notochord and the formation of intervertebral discs. J Cell Biol 1998a;143:1399-1412.

Aszódi A, Pfeifer A, Wendel M, Hiripi L and Fässler R. Mouse models for extracellular matrix diseases. J Mol Med 1998b;76:238-252.

Bagambisa FB, Joos U and Schilli W. A scanning electron microscope study of the ultrastructural organization of bone mineral. Cells and Materials 1993;3:93-102.

Baron R. Anatomy and ultrastructure of bone. In: Favus MJ (Ed.) Primer on the metabolic bone diseases and disorders of mineral metabolism. 2nd ed. Raven Press, New York, 1993.

Bateman TA, Zimmerman RJ, Ayers RA, Ferguson VL, Chapes SK and Simske SJ. Histomorphometric, physical, and mechanical effects of spaceflight and insulin-like growth factor-I on rat long bones. Bone 1998;23:527-535.

Beamer WG, Donahue LR, Rosen CJ and Baylink DJ. Genetic variability in adult bone density among inbred strains of mice. Bone 1996;18:397-403.

Beck BR, Qin YX, McLeod KJ and Otter MW. On the relationship between streaming potential and strain in an in vivo bone preparation. Calcif Tissue Int 2002;71:335-343.

Bentolila V, Boyce TM, Fyhrie DP, Drumb R, Skerry TM and Schaffler MB. Intracortical remodeling in adult rat long bones after fatigue loading. Bone 1998;23:275-281.

Bolotin HH, Sievänen H and Grashuis JL. Patient-specific DXA bone mineral density inaccuracies: quantitative effects of nonuniform extraosseous fat distributions. J Bone Miner Res 2003;18:1020-1027.

87Bolotin HH, Sievänen H, Grashuis JL, Kuiper JW and Järvinen TL. Inaccuracies inherent in patient-specific dual-energy X-ray absorptiometry bone mineral density measurements: comprehensive phantom-based evaluation. J Bone Miner Res 2001;16:417-426.

Bonadio JF, Saunders TL, Tsai E, Goldstein SA, Morris-Wiman J, Brinkley L, Dolan DF, Altschuler RA, Hawkins JE, Bateman JF, Mascara T and Jaenisch R. Transgenic mouse model of the mild dominant form of osteogenesis imperfecta. Proc Natl Acad Sci 1990;87:7145-7149.

Bonewald L. Osteocytes as multifunctional cells. J Musculoskelet Neuronal Interact 2006;6:331-333.

Bonewald LF and Johnson ML. Osteocytes, mechanosensing and Wnt signaling. Bone 2008;42:606-615.

Boonen S, Koutri R, Dequeker J, Aerssens J, Lowet G, Nijs J, Verbeke G, Lesaffre E and Geusens P. Measurement of femoral geometry in type I and type II osteoporosis: differences in hip axis length consistent with heterogeneity in the pathogenesis of osteoporotic fractures. J Bone Miner Res 1995;10:1908-1912.

Boskey AL, Wright TM and Blank RD. Collagen and bone strength. J Bone Miner Res 1999;14:330-335.

Bourrin S, Genty C, Palle S, Gharib C and Alexandre C. Adverse effects of strenuous exercise: a densitometric and histomorphometric study in the rat. J Appl Physiol 1994;76:1999-2005.

Bourrin S, Zerath E, Vico L, Milhaud C and Alexandre C. Bone mass and bone cellular variations after five months of physical training in Rhesus monkeys: Histomorphometric study. Calcif Tissue Int 1992;50:404-410.

Boyce RW, Franks AF, Jankowsky ML, Orcutt CM, Piacquadio AM, White JM and Bevan JA. Sequential histomorphometric changes in cancellous bone from ovariohysterectomized dogs. J Bone Miner Res 1990;5:947-953.

Braghetta P, Fabbro C, Piccolo S, Marvulli D, Bonaldo P, Volpin D and Bressan GM. Distinct regions control transcriptional activation of the α1(VI) collagen promoter in different tissues of transgenic mice. J Cell Biol 1996;135:1163-1177.

Brockstedt H, Kassem M, Eriksen EF, Mosekilde L and Melsen F. Age- and sex-related changes in iliac cortical bone mass and remodeling. Bone 1993;14:681-691.

Brown MA and Duncan EL. Genetic studies of osteoporosis. Expert Rev Mol Med 1999;1999:1-18.

Burgeson RE and Nimni ME. Collagen types: molecular structure and tissue distribution. Clin Orthop Rel Res 1992;282:250-272.

Burstein AH and Frankel VH. A standard test for laboratory animal bone. J Biomech 1971;4:155-158.

Bätge B, Diebold J, Stein H, Bodo M and Müller PK. Compositional analysis of the collagenous bone matrix. A study on adult normal and osteopenic bone tissue. Eur J Clin Invest 1992;22:805-812.

Callaghan JP and McGill SM. Frozen storage increases the ultimate compressive load of porcine vertebrae. J Orthop Res 1995;13:809-812.

Callewaert B, Malfait F, Loeys B and De Paepe A. Ehlers-Danlos syndromes and Marfan syndrome. Best

88Pract Res Clin Rheumatol 2008;22:165-189.

Cameron JR and Sorenson J. Measurement of bone mineral in vivo: an improved method. Science 1963;142:230-232.

Cann CE and Genant HK. Precise measurement of vertebral mineral content using computed tomography. J Comput Assist Tomogr 1980;4:493-500.

Carbon R, Sambrook PN, Deakin V, Fricker P, Eisman JA, Kelly P, Maguire K and Yeates MG. Bone density of elite female athletes with stress fractures. Med J Aust 1990;153:373-376.

Castelo-Branco C, Pons F and Gonz lez-Merlo J. Bone mineral density in surgically postmenopausal women receiving hormonal replacement therapy as assessed by dual photon absorptiometry. Maturitas 1993;16:133-137.

Chang C and Bloom S. A simple ashing method for determination of Mg and Ca in laboratory animal feed and tissues. J Am Coll Nutr 1983;2:149-155.

Chen MM, Yeh JK, Aloia JF, Tierney JM and Sprintz S. Effect of treadmill exercise on tibial cortical bone in aged female rats: a histomorphometric and dual energy x-ray absorptiometry study. Bone 1994;15:313-319.

Christo K, Prabhakaran R, Lamparello B, Cord J, Miller KK, Goldstein MA, Gupta N, Herzog DB, Klibanski A and Misra M. Bone metabolism in adolescent athletes with amenorrhea, athletes with eumenorrhea, and control subjects. Pediatrics 2008;121:1127-1136.

Civitelli R. Cell-cell communication in the osteoblast/osteocyte lineage. Arch Biochem Biophys 2008;473:188-192.

Cody DD, Goldstein SA, Flynn MJ and Brown EB. Correlations between vertebral regional bone mineral density (rBMD) and whole bone fracture load. Spine 1991;16:146-154.

Cole WG. Collagen genes: mutations affecting collagen structure and expression. Progr Nucl Acid Res Mol Biol 1994;47:29-80.

Colwell A, Russell RGG and Eastell R. Factors affecting the assay of urinary 3-hydroxy pyridinium crosslinks of collagen as markers of bone resorption. Eur J Clin Invest 1993;23:341-349.

Consensus Development Conference. Prophylaxis and treatment of osteoporosis. Am J Med 1991;90:107-110.

Consensus Development Conference. Diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med 1993;94:646-650.

Cooke TD. Pathogenetic mechanisms in polyarticular osteoarthritis. Clin Rheum Dis 1985;11:203-238.

Cummings SR, Nevitt MC, Browner WS, Stone K, Fox KM, Ensrud KE, Cauley J, Black D and Vogt TM. Risk factors for hip fracture in white women. Study of Osteoporotic Fractures Research Group. N Engl J Med 1995;332:767-773.

Currey J. Bones: Structure and mechanics. Princeton University Press. 456 p., 2002.

89Datta HK, Ng WF, Walker JA, Tuck SP and Varanasi SS. The cell biology of bone metabolism. J Clin Pathol 2008;61:577-587.

de la Piedra C, Díaz Martin MA, Diaz Diego EM, López Gavilanes E, González Parra E, Caramelo C and Rapado A. Serum concentrations of carboxyterminal cross-linked telopeptide of type I collagen (ICTP), serum tartrate resistant acid phosphatase, and serum levels of intact parathyroid hormone in parathyroid hyperfunction. Scand J Clin Lab Invest 1994;54:11-15.

del Puente A, Pappone N, Mandes MG, Mantova D, Scarpa R and Oriente P. Determinants of bone mineral density in immobilization: a study on hemiplegic patients. Osteoporosis Int 1996;6:50-54.

Delmas PD and Garnero P. Utility of biochemical markers of bone turnover in osteoporosis. In: Marcus R, Feldman D and Kelsey J (Eds.) Osteoporosis. Academic Press, Inc., London, 1996.

Dempster DW. Bone remodeling. In: Coe FL and Favus MJ (Eds.) Disorders of Bone and Mineral Metabolism. 1st ed. Raven Press, New York, 1992.

Denissen H, de Blieck J, Verhey H, Klein C and van Lingen A. Dual-energy x-ray absorptiometry for histologic bone sections. J Bone Miner Res 1996;11:638-644.

Dimitriou R, Tsiridis E, Carr I, Simpson H and Giannoudis PV. The role of inhibitory molecules in fracture healing. Injury 2006;37 (Suppl 1):S20-29.

Dionyssiotis Y, Trovas G, Galanos A, Raptou P, Papaioannou N, Papagelopoulos P, Petropoulou K and Lyritis GP. Bone loss and mechanical properties of tibia in spinal cord injured men. J Musculoskelet Neuronal Interact 2007;7:62-68.

Drinkwater BL, Nilson K, Chesnut III CH, Bremer WJ, Shainholtz S and Southworth MB. Bone mineral content of amenorrheic and eumenorrheic athletes. N Engl J Med 1984;311:277-281.

Einhorn TA. Bone strength: the bottom line. Calcif Tissue Int 1992;51:333-339.

Elima K. Osteoinductive proteins. Ann Med 1993;25:395-402.

Eriksson SAV, Isberg BO and Lindgren JU. Vertebral bone mineral measurement using dual photon absorptiometry and computed tomography. Acta Radiol 1988;29:89-94.

Eriksson SAV, Isberg BO and Lindgren JU. Prediction of vertebral strength by dual photon absorptiometry and quantitative computed tomography. Calcif Tissue Int 1989;44:243-250.

Evans RA, Marel GM, Lancaster EK, Kos S, Evans M and Wong SYP. Bone mass is low in relatives of osteoporotic patients. Ann Intern Med 1988;109:870-873.

Eyre D. Collagen of articular cartilage. Arthritis Res 2002;4:30-35.

Eyre DR. Collagens of the disc. In: Ghosh P (Ed.) The biology of the intervertebral disc. CRC Press, Inc., Boca Raton, 1988.

Ferrari S. Human genetics of osteoporosis. Best Pract Res Clin Endocrinol Metab 2008;22:723-735.

Ferretti JL. Perspectives of pQCT technology associated to biomechanical studies in skeletal research employing rat models. Bone 1995;17:353S-364S.

90Ferretti JL, Capozza RF and Zanchetta JR. Mechanical validation of a tomographic (pQCT) index for noninvasive estimation of rat femur bending strength. Bone 1996;18:97-102.

Fields GB. The collagen triple-helix: correlation of conformation with biological activities. Connect Tiss Res 1995;31:235-243.

Finnish Arthroplasty Register. The 2006 implant yearbook on orthopaedic endoprostheses. Publications of the National Agency for Medicines. 51 p., 2008.

Forriol F and Shapiro F. Bone development: interaction of molecular components and biophysical forces. Clin Orthop Relat Res 2005:14-33.

Forwood MR and Burr DB. Physical activity and bone mass: exercises in futility? Bone Miner 1993;21:89-112.

Fratzl P, Paris O, Klaushofer K and Landis WJ. Bone mineralization in an osteogenesis imperfecta mouse model studied by small-angle x-ray scattering. J Clin Invest 1996;97:396-402.

Frost HM. Tetracycline-based histological analysis of bone remodeling. Calcif Tiss Res 1969;3:211-237.

Frost HM. Skeletal structural adaptations to mechanical usage (SATMU): 1. Redefining Wolff’s law: the bone modeling problem. Anat Rec 1990a;226:403-413.

Frost HM. Skeletal structural adaptations to mechanical usage (SATMU): 2. Redefining Wolff’s law: the remodeling problem. Anat Rec 1990b;226:414-422.

Frost HM. From Wolff’s law to the mechanostat: a new “face” of physiology. J Orthop Sci 1998;3:282-286.

Frost HM. Why do bone strength and “mass” in aging adults become unresponsive to vigorous exercise? Insights of the Utah paradigm. J Bone Miner Metab 1999;17:90-97.

Fukumoto S. The role of bone in phosphate metabolism. Mol Cell Endocrinol 2008. In press.

Garcés GL and Santandreu MaE. Longitudinal bone growth after sciatic denervation in rats. J Bone Joint Surg 1988;70-B:315-318.

Gdynia HJ, Kuhnlein P, Ludolph AC and Huber R. Connective tissue disorders in dissections of the carotid or vertebral arteries. J Clin Neurosci 2008;15:489-494.

Gehron Robey P and Boskey AL. The biochemistry of bone. In: Marcus R, Feldman D and Kelsey J (Eds.) Osteoporosis. Academic Press, Inc., London, 1996.

Gehron Robey P, Fedarko NS, Hefferan TE, Bianco P, Vetter UK, Grzesik W, Friedenstein A, van der Pluijm G, Mintz KP, Young MF, Kerr JM, Ibaraki K and Heegaard A-M. Structure and molecular regulation of bone matrix proteins. J Bone Miner Res 1993;8 (Suppl. 2):S483-S487.

Genant HK and Boyd D. Quantitative bone mineral analysis using dual energy computed tomography. Invest Radiol 1977;12:545-551.

Gerstenfeld LC, Riva A, Hodgens K, Eyre DR and Landis WJ. Post-translational control of collagen fibrillogenesis in mineralizing cultures of chick osteoblasts. J Bone Miner Res 1993;8:1031-1043.

91Gerstenfeld LC and Shapiro FD. Expression of bone-specific genes by hypertrophic chondrocytes: implications of the complex functions of the hypertrophic chondrocyte during endochondral bone development. J Cell Biochem 1996;62:1-9.

Glüer C-C, Wu CY and Genant HK. Broadband ultrasound attenuation signals depend on trabecular orientation: An in vitro study. Osteoporosis Int 1993;3:185-191.

Glüer C-C, Wu CY, Jergas M, Goldstein SA and Genant HK. Three quantitative ultrasound parameters reflect bone structure. Calcif Tissue Int 1994;55:46-52.

Goodsitt MM, Kilcoyne RF, Gutcheck RA, Richardson ML and Rosenthal DI. Effect of collagen on bone mineral analysis with CT. Radiology 1988;167:787-791.

Gough AKS, Peel NFA, Eastell R, Holder RL, Lilley J and Emery P. Excretion of pyridinium crosslinks correlates with disease activity and appendicular bone loss in early rheumatoid arthritis. Ann Rheum Dis 1994;53:14-17.

Gray H. Osteology. In: Williams PL and Warwick R (Eds.) Gray’s anatomy. 36th ed. Churchill Livingstone, Edinburgh, 1980.

Grazioli V, Casari E, Murone M and Bonini PA. High-performance liquid chromatographic method for measuring hydroxylysine glycosides and their ratio in urine as a possible marker of human bone collagen breakdown. J Chromatogr 1993;615:59-66.

Grimston SK, Willows ND and Hanley DA. Mechanical loading regime and its relationship to bone mineral density in children. Med Sci Sports Exerc 1993;25:1203-1210.

Haapala J, Arokoski J, Pirttimäki J, Lyyra T, Jurvelin J, Tammi M, Helminen HJ and Kiviranta I. Incomplete restoration of immobilization induced softening of young beagle knee articular cartilage after 50-week remobilization. Int J Sports Med 2000;21:76-81.

Haapala J, Arokoski JP, Hyttinen MM, Lammi M, Tammi M, Kovanen V, Helminen HJ and Kiviranta I. Remobilization does not fully restore immobilization induced articular cartilage atrophy. Clin Orthop Relat Res 1999:218-229.

Haapasalo H, Kannus P, Sievänen H, Heinonen A, Oja P and Vuori I. Long-term unilateral loading and bone mineral density and content in female squash players. Calcif Tissue Int 1994;54:249-255.

Halleen JM, Alatalo SL, Suominen H, Cheng S, Janckila AJ and Väänänen HK. Tartrate-resistant acid phosphatase 5b: a novel serum marker of bone resorption. J Bone Miner Res 2000;15:1337-1345.

Hamerman D. Aging and osteoarthritis: basic mechanisms. JAGS 1993;41:760-770.

Hans D, Wu C, Njeh CF, Zhao S, Augat P, Newitt D, Link T, Lu Y, Majumdar S and Genant HK. Ultrasound velocity of trabecular cubes reflects mainly bone density and elasticity. Calcif Tissue Int 1999;64:18-23.

Hansson TH, Keller TS and Panjabi MM. A study of the compressive properties of lumbar vertebral trabeculae: effects of tissue characteristics. Spine 1987;12:56-61.

Hardingham TE and Fosang AJ. Proteoglycans: many forms and many functions. FASEB J 1992;6:861-870.

92Hayakawa T. Multiple functions of tissue inhibitors of metalloproteinases (TIMPs): a new aspect involving osteoclastic bone resorption. J Bone Miner Metab 2002;20:1-13.

Heikkinen LM, Panula HE, Lyyra T, Olkkonen H, Kiviranta I, Nevalainen T and Helminen HJ. Electromechanical film sensor device for dynamic force recording from canine limbs. Scand J Lab Anim Sci 1997;2:85-92.

Heinonen A, Oja P, Kannus P, Sievänen H, Mänttäri A and Vuori I. Bone mineral density of female athletes in different sports. Bone Miner 1993;23:1-14.

Helminen HJ, Kiraly K, Pelttari A, Tammi MI, Vandenberg P, Pereira R, Dhulipata R, Khillan JS, Ala-Kokko L, Hume EL, Sokolov BP and Prockop DJ. An inbred line of transgenic mice expressing an internally deleted gene for type II procollagen (COL2A1). J Clin Invest 1993;92:582-595.

Helminen HJ, Säämänen AM, Salminen H and Hyttinen MM. Transgenic mouse models for studying the role of cartilage macromolecules in osteoarthritis. Rheumatology (Oxford) 2002;41:848-856.

Hetland ML, Haarbo J and Christiansen C. Running induces menstrual disturbances but bone mass is unaffected, except in amenorrheic women. Am J Med 1993;95:53-60.

Holroyd C, Cooper C and Dennison E. Epidemiology of osteoporosis. Best Pract Res Clin Endocrinol Metab 2008;22:671-685.

Howard GM, Nguyen TV, Harris M, Kelly PJ and Eisman JA. Genetic and environmental contributions to the association between quantitative ultrasound and bone mineral density measurements: a twin study. J Bone Miner Res 1998;13:1318-1327.

Hughes DE, Salter DM, Dedhar S and Simpson R. Integrin expression in human bone. J Bone Miner Res 1993;8:527-533.

Hughes-Fulford M. Review of the biological effects of weightlessness on the human endocrine system. Receptor 1993;3:145-154.

Hui M and Tenenbaum HC. New face of an old enzyme: alkaline phosphatase may contribute to human tissue aging by inducing tissue hardening and calcification. Anat Rec 1998;253:91-94.

Hui SL, Slemenda CW and Johnston CC, Jr. Age and bone mass as predictors of fracture in a prospective study. J Clin Invest 1988;81:1804-1809.

Hultenby K, Reinholt FP, Norgård M, Oldberg Å, Wendel M and Heinegård D. Distribution and synthesis of bone sialoprotein in metaphyseal bone of young rats show a distinctly different pattern from that of osteopontin. Eur J Cell Biol 1994;63:230-239.

Huopio J, Kroger H, Honkanen R, Jurvelin J, Saarikoski S and Alhava E. Calcaneal ultrasound predicts early postmenopausal fractures as well as axial BMD. A prospective study of 422 women. Osteoporos Int 2004;15:190-195.

Ingram RT, Clarke BL, Fisher LW and Fitzpatrick LA. Distribution of noncollagenous proteins in the matrix of adult human bone: evidence of anatomic and functional heterogenity. J Bone Miner Res 1993;8:1019-1029.

Ito M, Hayashi K, Kawahara Y, Uetani M and Imaizumi Y. The relationship of trabecular and cortical

93bone mineral density to spinal fractures. Invest Radiol 1993;28:573-580.

Iwamoto J, Takeda T, Katsumata T, Tanaka T, Ichimura S and Toyama Y. Effect of etidronate on bone in orchidectomized and sciatic neurectomized adult rats. Bone 2002;30:360-367.

Jackson JA. Osteoporosis in men. In: Favus MJ (Ed.) Primer on the metabolic bone diseases and disorders of mineral metabolism. 2nd ed. Raven Press, New York, 1993.

Jaworski ZFG, Liskova-Kiar M and Uhthoff HK. Effect of long-term immobilisation on the pattern of bone loss in older dogs. J Bone Joint Surg 1980;62-B:104-110.

Jayo MJ, Jerome CP, Lees CJ, Rankin SE and Weaver DS. Bone mass in female cynomolgus macaques: a cross-sectional and longitudinal study by age. Calcif Tissue Int 1994;54:231-236.

John A and Tuszynski G. The role of matrix metalloproteinases in tumor angiogenesis and tumor metastasis. Pathol Oncol Res 2001;7:14-23.

Johnson J and Dawson-Hughes B. Precision and stability of dual-energy x-ray absorptiometry measurements. Calcif Tissue Int 1991;49:174-178.

Jonas J, Burns J, Abel EW, Cresswell MJ, Strain JJ and Paterson CR. Impaired mechanical strength of bone in experimental copper deficiency. Ann Nutr Metab 1993;37:245-252.

Jonnavithula S, Warren MP, Fox RP and Lazaro MI. Bone density is compromised in amenorrheic women despite return of menses: a 2-year study. Obstet Gynecol 1993;81:669-674.

Jämsä T. Methods for evaluating bone strength in experimental studies. 72 p. Department of Anatomy, Department of Surgery, Technical Services Unit and Biomedical Engineering Program, University of Oulu, Finland.1998

Jämsä T and Jalovaara P. A cost-effective, accurate machine for testing the torsional strength of sheep long bones. Med Eng Phys 1996;18:433-435.

Jämsä T, Jalovaara P, Peng Z, Väänänen HK and Tuukkanen J. Comparison of three-point bending test and pQCT analysis in the evaluation of the strength of mouse femur and tibia. Bone 1998a;23:155-161.

Jämsä T, Jalovaara P, Peng Z and Väänänen K. Equipment for testing the biomechanical strength of bones. Med Biol Eng Comp 1996;34 (Supplement 1, Part 1):347-348.

Jämsä T, Tuukkanen J and Jalovaara P. Femoral neck strength of mouse in two loading configurations: method evaluation and fracture characteristics. J Biomech 1998b;31:723-729.

Järvinen TL, Kannus P, Pajamäki I, Vuohelainen T, Tuukkanen J, Järvinen M and Sievänen H. Estrogen deposits extra mineral into bones of female rats in puberty, but simultaneously seems to suppress the responsiveness of female skeleton to mechanical loading. Bone 2003a;32:642-651.

Järvinen TLN, Kannus P and Sievänen H. Estrogen and bone--a reproductive and locomotive perspective. J Bone Miner Res 2003b;18:1921-1931.

Järvinen TLN, Sievänen H, Kannus P and Järvinen M. Dual-energy x-ray absorptiometry in predicting mechanical characteristics of rat femur. Bone 1998;22:551-558.

94Kannus P, Jozsa L, Renström P, Järvinen M, Kvist M, Lehto M, Oja P and Vuori I. The effects of training, immobilization and remobilization on musculoskeletal tissue: 1. Training and immobilization. Scand J Med Sci Sports 1992a;2:100-118.

Kannus P, Jozsa L, Renström P, Järvinen M, Kvist M, Lehto M, Oja P and Vuori I. The effects of training, immobilization and remobilization on musculoskeletal tissue: 2. Remobilization and prevention of immobilization atrophy. Scand J Med Sci Sports 1992b;2:164-176.

Kannus P, Niemi S, Parkkari J, Palvanen M, Vuori I and Järvinen M. Hip fractures in Finland between 1970 and 1997 and predictions for the future. Lancet 1999;353:802-805.

Kannus P, Niemi S, Parkkari J, Palvanen M, Vuori I and Järvinen M. Nationwide decline in incidence of hip fracture. J Bone Miner Res 2006;21:1836-1838.

Kannus P, Palvanen M, Niemi S, Parkkari J and Järvinen M. Epidemiology of osteoporotic pelvic fractures in elderly people in Finland: sharp increase in 1970-1997 and alarming projections for the new millennium. Osteoporosis Int 2000;11:443-448.

Karinkanta S, Heinonen A, Sievänen H, Uusi-Rasi K, Fogelholm M and Kannus P. Maintenance of exercise-induced benefits in physical functioning and bone among elderly women. Osteoporos Int 2009;20;665-674.

Karlsson MK, Johnell O and Obrant KJ. Bone mineral density in professional ballet dancers. Bone Miner 1993;21:163-169.

Karlsson MK, Johnell O and Obrant KJ. Is bone mineral density advantage maintained long-term in previous weight lifters? Calcif Tissue Int 1995;57:325-328.

Kassem M, Abdallah BM and Saeed H. Osteoblastic cells: differentiation and trans-differentiation. Arch Biochem Biophys 2008;473:183-187.

Kastl S, Sommer T, Klein P, Hohenberger W and Engelke K. Accuracy and precision of bone mineral density and bone mineral content in excised rat humeri using fan beam dual-energy X-ray absorptiometry. Bone 2002;30:243-246.

Kaufman JM and Goemaere S. Osteoporosis in men. Best Pract Res Clin Endocrinol Metab 2008;22:787-812.

Keaveny TM and Hayes WC. A 20-year perspective on the mechanical properties of trabecular bone. Transactions of the ASME 1993;115:534-542.

Keaveny TM, Pinilla TP, Crawford RP, Kopperdahl DL and Lou A. Systematic and random errors in compression testing of trabecular bone. J Orthop Res 1997;15:101-110.

Kelly TL, Slovik DM, Schoenfeld DA and Neer RM. Quantitative digital radiography versus dual photon absorptiometry of the lumbar spine. J Clin Endocrinol Metab 1988;67:839-844.

Kim M, Otsuka M and Arakawa N. Age-related changes in the pyridinoline content of guinea pigs cartilage and achilles tendon collagen. J Nutr Sci Vitaminol 1994;40:95-103.

Kimmel DB and Jee WS. A quantitative histologic study of bone turnover in young adult beagles. Anat Rec 1982;203:31-45.

95Király K, Hyttinen MM, Lapveteläinen T, Elo M, Kiviranta I, Dobai J, Módis L, Helminen HJ and Arokoski JPA. Specimen preparation and quantification of collagen birefringence in unstained sections of articular cartilage using image analysis and polarizing light microscopy. Histochem J 1997;29:317-327.

Koo WWK, Walters J and Bush AJ. Technical considerations of dual-energy x-ray absorptiometry-based bone mineral measurements for pediatric studies. J Bone Miner Res 1995;10:1998-2004.

Kream BE, Harrison JR, Krebsbach PH, Bogdanovic Z, Bedalov A, Pavlin D, Woody CO, Clark SH, Rowe D and Lichtler AC. Regulation of type I collagen gene expression in bone. Connect Tiss Res 1995;31:261-264.

Kröger H, Huopio J, Honkanen R, Tuppurainen M, Puntila E, Alhava E and Saarikoski S. Prediction of fracture risk using axial bone mineral density in a perimenopausal population: a prospective study. J Bone Miner Res 1995a;10:302-306.

Kröger H, Jurvelin J, Arnala I, Penttilä K, Rask A, Vainio P and Alhava E. Ultrasound attenuation of the calcaneus in normal subjects and patients with wrist fracture. Acta Orthop Scand 1995b;66:47-52.

Kröger H, Tuppurainen M, Honkanen R, Alhava E and Saarikoski S. Bone mineral density and risk factors for osteoporosis - a population based study of 1600 perimenopausal women. Calcif Tissue Int 1994;55:1-7.

Kröger H, Vainio P and Nieminen J. Estimation of spinal bone density using conventional MRI. Comparison between MRI and DXA in 32 subjects. Acta Orthop Scand 1995c;66:532-534.

Krølner B and Toft B. Vertebral bone loss: an unheeded side effect of therapeutic bed rest. Clin Sci 1983;64:537-540.

Krølner B, Toft B, Nielsen SP and Tøndevold E. Physical exercise as prophylaxis against involutional vertebral bone loss: a controlled trial. Clin Sci 1983;64:541-546.

Kuivaniemi H, Tromp G and Prockop DJ. Mutations in collagen genes: causes of rare and some common diseases in humans. FASEB J 1991;5:2052-2060.

Kung AW and Huang QY. Genetic and environmental determinants of osteoporosis. J Musculoskelet Neuronal Interact 2007;7:26-32.

Kushner A. Evaluation of Wolff’s law of bone formation. J Bone Joint Surg Am 1940;22:589-596.

Lajeunesse D and Reboul P. The role of bone in the development of osteoarthritis. In: Bronner F and Farach-Carson MC (Eds.) Bone and osteoarthritis. Springer, London, 2007.

Lane HW, LeBlanc AD, Putcha L and Whitson PA. Nutrition and human physiological adaptations to space flight. Am J Clin Nutr 1993;58:583-588.

Lane NE, Bloch DA, Jones HH, Marshall WH, Jr., Wood PD and Fries JF. Long-distance running, bone density and osteoarhritis. JAMA 1986;255:1147-1151.

Lapveteläinen T, Nevalainen T, Parkkinen JJ, Arokoski J, Kiraly K, Hyttinen MM, Halonen P and Helminen HJ. Lifelong moderate running training increases the incidence and severity of osteoarthritis in the knee joint of C57BL mice. Anat Rec 1995;242:159-165.

96Lapveteläinen T, Tiihonen A, Koskela P, Nevalainen T, Lindblom J, Kiraly K, Halonen P and Helminen HJ. Training a large number of laboratory mice using running wheels and analyzing running behavior by use of a computer-assisted system. Lab Anim Sci 1997;47:172-179.

LeBlanc A, Schneider V, Krebs J, Evans H, Jhingran S and Johnson P. Spinal bone after 5 weeks of bed rest. Calcif Tissue Int 1987;41:259-261.

LeBlanc AD, Spector ER, Evans HJ and Sibonga JD. Skeletal responses to space flight and the bed rest analog: a review. J Musculoskelet Neuronal Interact 2007;7:33-47.

Lepola V, Väänänen K and Jalovaara P. The effect of immobilization on the torsional strength of the rat tibia. Clin Orthop Rel Res 1993;297:55-61.

Li S-W, Prockop DJ, Helminen H, Fässler R, Lapveteläinen T, Kiraly K, Pelttari A, Arokoski J, Lui H, Arita M and Khillan JS. 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 & Development 1995;9:2821-2830.

Li X, Gu W, Masinde G, Hamilton-Ulland M, Rundle CH, Mohan S and Baylink DJ. Genetic variation in bone-regenerative capacity among inbred strains of mice. Bone 2001;29:134-140.

Li XJ, Jee WSS, Chow S-Y and Woodbury DM. Adaptation of cancellous bone to aging and immobilization in the rat: a single photon absorptiometry and histomorphometry study. Anat Rec 1990;227:12-24.

Lin R, Chang G and Chang L. Biomechanical properties of muscle-tendon unit under high-speed passive stretch. Clin Biomech (Bristol, Avon) 1999;14:412-417.

Lind M, Eriksen EF and Bünger C. Bone morphogenetic protein-2 but not bone morphogenetic protein-4 and -6 stimulates chemotactic migration of human osteoblasts, human marrow osteoblasts, and U2-OS cells. Bone 1996;18:53-57.

Linde F, Hvid I and Madsen F. The effect of specimen geometry on the mechanical behaviour of trabecular bone specimens. J Biomech 1992;25:359-368.

Linde F and Sørensen HCF. The effect of different storage methods on the mechanical properties of trabecular bone. J Biomech 1993;26:1249-1252.

Liu X, Wu H, Byrne M, Krane S and Jaenisch R. Type III collagen is crucial for collagen I fibrillogenesis and for normal cardiovascular development. Proc Natl Acad Sci 1997;94:1852-1856.

Lovász G, Llinás A, Benya P, Bodey B, McKellop HA, Luck JV, Jr. and Sarmiento A. Effects of valgus tibial angulation on cartilage degeneration in the rabbit knee. J Orthop Res 1995;13:846-853.

Lueken SA, Arnaud SB, Taylor AK and Baylink DJ. Changes in markers of bone formation and resorption in a bed rest model of weightlessness. J Bone Miner Res 1993;8:1433-1438.

Madsen JE, Aune AK, Falch JA, Hukkanen M, Konttinen YT, Santavirta S and Nordsletten L. Neural involvement in post-traumatic osteopenia: an experimental study in the rat. Bone 1996;18:411-476.

Majeska RJ, Ryaby JT and Einhorn TA. Direct modulation of osteoblastic activity with estrogen. J Bone Joint Surg 1994;76-A:713-721.

97Malfait F and De Paepe A. Molecular genetics in classic Ehlers-Danlos syndrome. Am J Med Genet C Semin Med Genet 2005;139C:17-23.

Malluche HH and Faugere M-C. Atlas of mineralized bone histology. Karger, Basel, 1986.

Mann V, Huber C, Kogianni G, Jones D and Noble B. The influence of mechanical stimulation on osteocyte apoptosis and bone viability in human trabecular bone. J Musculoskelet Neuronal Interact 2006;6:408-417.

Mann V and Ralston SH. Meta-analysis of COL1A1 Sp1 polymorphism in relation to bone mineral density and osteoporotic fracture. Bone 2003;32:711-717.

Marcus R. Secondary forms of osteoporosis. In: Coe FL and Favus MJ (Eds.) Disorders of Bone and Mineral Metabolism. 1st ed. Raven Press, New York, 1992.

Marcus R, Cann CE, Madvig P, Minkoff J, Goddard M, Bayer M, Martin M, Gaudiani L, Haskell W and Genant HK. Menstrual function and bone mass in elite women distance runners. Ann Intern Med 1985;102:158-163.

Markel MD and Bogdanske JJ. The effect of increasing gap width on localized densitometric changes within tibial ostectomies in a canine model. Calcif Tissue Int 1994;54:155-159.

Martens M, van Audekercke R, de Meester P and Mulier JC. Mechanical behaviour of femoral bones in bending loading. J Biomech 1986;19:443-454.

Martin A. A comparison of some bone ashing techniques. Health Phys 1967;13:1348-1351.

Martin RB. Determinants of the mechanical properties of bones. J Biomech 1991;24 Suppl 1:79-88.

Matsumoto T, Nakagawa S, Nishida S and Hirota R. Bone density and bone metabolic markers in active collegiate athletes: findings in long-distance runners, judoists, and swimmers. Int J Sports Med 1997;18:408-412.

Mayne R and Ala-Kokko L. Collagen structure and function. In: Koopman WJ and Moreland LW (Eds.) Arthritis and allied conditions. 15 ed. Lippincott Williams & Wilkins, Philadelphia, 2005.

Mazess RB, Barden HS, Bisek JP and Hanson J. Dual-energy x-ray absorptiometry for total-body and regional bone-mineral and soft-tissue composition. Am J Clin Nutr 1990;51:1106-1112.

McCalden RW, McGeough JA, Barker MB and Court-Brown CM. Age-related changes in the tensile properties of cortical bone. J Bone Joint Surg 1993;75A:1193-1205.

Meikle MC, Heath JK and Reynolds JJ. The use of in vitro models for investigating the response of fibrous joints to tensile mechanical stress. Am J Orthodont 1984;85:141-153.

Millán JL. Alkaline Phosphatases: Structure, substrate specificity and functional relatedness to other members of a large superfamily of enzymes. Purinergic Signal 2006;2:335-341.

Miller EJ and Matukas VJ. Chick cartilage collagen: a new type of α1 chain not present in bone or skin of the species. Proc Natl Acad Sci USA 1969;64:1264-1268.

Mosekilde L, Danielsen CC, Søgaard CH and Thorling E. The effect of long-term exercise on vertebral

98and femoral bone mass, dimensions, and strength - assessed in a rat model. Bone 1994;15:293-301.

Mueller M, Schilling T, Minne HW and Ziegler R. Does immobilization influence the systemic acceleratory phenomenon that accompanies local bone repair? J Bone Miner Res 1992;7 (Suppl 2):S425-S427.

Mueller TL, Stauber M, Kohler T, Eckstein F, Müller R and van Lenthe GH. Non-invasive bone competence analysis by high-resolution pQCT: An in vitro reproducibility study on structural and mechanical properties at the human radius. Bone 2009;44:364-271.

Myburgh KH, Bachrach LK, Lewis B, Kent K and Marcus R. Low bone mineral density at axial and apendicular sites in amenorrheic athletes. Med Sci Sports Exerc 1993;25:1197-1202.

Myllyharju J and Kivirikko KI. Collagens and collagen-related diseases. Ann Med 2001;33:7-21.

Nagai M, Suzuki Y and Ota M. Systematic assessment of bone resorption, collagen synthesis, and calcification in chick embryonic calvaria in vitro: effects of prostaglandin E2. Bone 1993;14:655-659.

National Research Council. Guide for the care and use of laboratory animals. NIH Publication, 1985.

Nicholson PF. Ultrasound and the biomechanical competence of bone. IEEE Trans Ultrason Ferroelectr Freq Control 2008;55:1539-1545.

Niederreither K, D’Souza RN and de Crombrugghe B. Minimal DNA sequences that control the cell lineage-specific expression of the proα2(I) collagen promoter in transgenic mice. J Cell Biol 1992;119:1361-1370.

Niederreither K, D'Souza RN, Metsäranta M, Eberspaecher H, Toman PD, Vuorio E and de Crombrugghe B. Coordinate patterns of expression of type I and III collagens during mouse development. Matrix Biol 1994;14:705-713.

Nilsson BE, Johnell O and Petersson C. In vivo bone-mineral measurement: how and why. Acta Orthop Scand 1990;61:275-281.

Noble BS. The osteocyte lineage. Arch Biochem Biophys 2008;473:106-111.

Nordin BEC, Need AG, Chatterton BE, Horowitz M and Morris HA. The relative contributions of age and years since menopause to postmenopausal bone loss. J Clin Endocrinol Metab 1990;70:83-88.

Nordsletten L, Kaastad TS, Skjeldal S, Kirkeby OJ, Reikerås O and Ekeland A. Training increases the in vivo strength of the lower leg: an experimental study in the rat. J Bone Miner Res 1993;8:1089-1095.

Nordström P, Pettersson U and Lorentzon R. Type of physical activity, muscle strength, and pubertal stage as determinants of bone mineral density and bone area in adolescent boys. J Bone Miner Res 1998;13:1141-1148.

Nurmi I, Narinen A, Luthje P and Tanninen S. Cost analysis of hip fracture treatment among the elderly for the public health services: a 1-year prospective study in 106 consecutive patients. Arch Orthop Trauma Surg 2003;123:551-554.

Oden ZM, Selvitelli DM, Hayes WC and Myers ER. The effect of trabecular structure on DXA-based predictions of bovine bone failure. Calcif Tissue Int 1998;63:67-73.

99Odgaard A, Hvid I and Linde F. Compressive axial strain distributions in cancellous bone specimens. J Biomech 1989;22:829-835.

Odgaard A and Linde F. The underestimation of Young’s modulus in compressive testing of cancellous bone specimens. J Biomech 1991;24:691-698.

Oettmeier R, Arokoski J, Roth AJ, Helminen HJ, Tammi M and Abendroth K. Quantitative study of articular cartilage and subchondral bone remodeling in the knee joint of dogs after strenuous running training. J Bone Miner Res 1992;7 (Suppl 2):S419-S424.

Ott SM. When bone mass fails to predict bone failure. Calcif Tissue Int 1993;53 Suppl.:S7-S13.

Oxlund H, Barckman M, ørtoft G and Andreassen TT. Reduced concentrations of collagen cross-links are associated with reduced strength of bone. Bone 1995;17(Suppl):365S-371S.

Panula HE, Helminen HJ and Kiviranta I. Slowly progressive osteoarthritis after tibial valgus osteotomy in young beagle dogs. Clin Orthop Rel Res 1997;343:192-202.

Parfitt AM. Age-related structural changes in trabecular and cortical bone: cellular mechanisms and biomechanical consequences. Calcif Tissue Int 1984;36 Suppl 1:S123-128.

Parfitt AM. Misconceptions (2): turnover is always higher in cancellous than in cortical bone. Bone 2002a;30:807-809.

Parfitt AM. Targeted and nontargeted bone remodeling: relationship to basic multicellular unit origination and progression. Bone 2002b;30:5-7.

Paschalis EP, Jacenko O, Olsen B, de Crombrugghe B and Boskey AL. The role of type X collagen in endochondral ossification as reduced by Fourier transform infrared microscopy analysis. Connect Tiss Res 1996;35:371-377.

Peng Z, Tuukkanen J, Zhang H, Jämsä T and Väänänen HK. The mechanical strength of bone in different rat models of experimental osteoporosis. Bone 1994;15:523-532.

Peppler WW and Mazess RB. Total body mineral and lean body mass by dual-photon absorptiometry. I. Theory and measurement procedure. Calcif Tissue Int 1981;33:353-359.

Pettersson U, Nordström P and Lorenzon R. A comparison of bone mineral density and muscle strength in young male adults with different exercise level. Calcif Tissue Int 1999;64:490-498.

Podenphant J and Engel U. Regional variations in histomorphometric bone dynamics from the skeleton of an osteoporotic woman. Calcif Tissue Int 1987;40:184-188.

Prendergast PJ and Huiskes R. The biomechanics of Wolff’s law: recent advances. Ir J Med Sci 1995;164:152-154.

Pritzker KPH. Animal models for osteoarthritis: processes, problems and prospects. Ann Rheum Dis 1994;53:406-420.

Prockop DJ, Kuivaniemi H and Tromp G. Molecular basis of osteogenesis imperfecta and related disorders of bone. Clinics in Plastic Surgery 1994;21:407-413.

100Puustjärvi K, Karjalainen P, Nieminen J, Arokoski J, Parviainen M, Helminen HJ and Soimakallio S. Endurance training associated with slightly lowered serum estradiol levels decreases mineral density of canine skeleton. J Bone Miner Res 1992;7:619-624.

Puustjärvi K, Lappalainen R, Niemitukia L, Arnala I, Nieminen J, Tammi M and Helminen HJ. Long-distance running alters bone anthropometry, elemental composition and mineral density of young dogs. Scand J Med Sci Sports 1995;5:17-23.

Quarles LD, Murphy G, Vogler JB and Drezner MK. Aluminum-induced neo-osteogenesis: a generalized process affecting trabecular networking in the axial skeleton. J Bone Miner Res 1990;5:625-635.

Radin EL, Paul IL and Lowy M. A comparison of the dynamic force transmitting properties of subchondral bone and articular cartilage. J Bone Joint Surg Am 1970;52:444-456.

Radin EL and Rose RM. Role of subchondral bone in the initiation and progression of cartilage damage. Clin Orthop Relat Res 1986;213:34-40.

Rahkonen O, Savontaus M, Abdelwahid E, Vuorio E and Jokinen E. Expression patterns of cartilage collagens and Sox9 during mouse heart development. Histochem Cell Biol 2003;120:103-110.

Rahn KA, Vanderby R, Kohles SS, Kiratli BJ, Thielke RJ, Clay AB and Suttie JW. Mechanical effect of sodium fluoride on bovine cortical bone. Clin Biomech 1991;6:185-189.

Recker RR. Embryology, anatomy, and microstructure of bone. In: Coe FL and Favus MJ (Eds.) Disorders of Bone and Mineral Metabolism. 1st ed. Raven Press, New York, 1992.

Reeve J, Zanelli JM, Garrahan N, Bradbeer JN, Wand JS, Moyes ST, Roux J-P and Smith T. Bone remodeling in hip fracture. Calcif Tissue Int 1993;53 Suppl.:S108-S112.

Reis BL, Keen CL, Lönnerdal B and Hurley LS. Mineral composition and zinc metabolism in female mice of varying age and reproductive status. J Nutr 1988;118:349-361.

Riggs BL, Khosla S and Melton LJ, 3rd. 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 1998;13:763-773.

Risteli J, Melkko J, Niemi S and Risteli L. Use of a marker of collagen formation in osteoporosis studies. Calcif Tissue Int 1991;49 (Suppl):S24-S25.

Risteli L and Risteli J. Biochemical markers of bone metabolism. Ann Med 1993;25:385-393.

Rittling SR, Matsumoto HN, McKee MD, Nanci A, An X-R, Novick KE, Kowalski AJ, Noda M and Denhardt DT. Mice lacking osteopontin show normal development and bone structure but display altered osteoclast formation in vitro. J Bone Miner Res 1998;13:1101-1111.

Roach HI and Tilley S. The pathogenesis of osteoarthritis. In: Bronner F and Farach-Carson MC (Eds.) Bone and osteoarthritis. Springer-Verlag, London, 2007.

Rockwell JC, Sorensen AM, Baker S, Leahey D, Stock JL, Michaels J and Baran DT. Weight training decreases vertebral density in premenopausal women: a prospective study. J Clin Endocrinol Metab 1990;71:988-993.

101Roer RD and Dillaman RM. Decreased femoral arterial flow during simulated microgravity in the rat. J Appl Physiol 1994;76:2125-2129.

Root AW and Diamond Jr. FB. Disorders of calcium and phosphorus metabolism in adolescents. Endocrinol Metab Clin North Amer 1993;22:573-592.

Rosati R, Horan GSB, Pinero GJ, Garofalo S, Keene DR, Horton WA, Vuorio E, de Crombrugghe B and Behringer RR. Normal long bone growth and development in type X collagen-null mice. Nature Genetics 1994;8:129-135.

Rozenberg S, Vandromme J, Kroll M, Pastijn A and Degueldre M. Osteoporosis prevention with sex hormone replacement therapy. Int J Fertil 1994;39:262-271.

Räsänen T, Jurvelin J and Helminen HJ. Indentation and shear tests of bovine knee. Articular Cartilage Biomechanical Seminar 1991. Center for Biomechanics, Gothenburg, Sweden.

Saban J, Zussman MA, Havey R, Patwardhan AG, Schneider GB and King D. Heterozygous oim mice exhibit a mild form of osteogenesis imperfecta. Bone 1996;19:575-579.

Sadler T. Skeletal system. In: Sadler T (Ed.) Langman’s medical embryology. 5th ed. Williams & Wilkins, Baltimore, 1985.

Sahlman J, Inkinen R, Hirvonen T, Lammi MJ, Lammi PE, Nieminen J, Lapveteläinen T, Prockop DJ, Arita M, Li SW, Hyttinen MM, Helminen HJ and Puustjärvi K. 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 2001;26:2558-2565.

Sakano S, Murata Y, Miura T, Iwata H, Sato K, Matsui N and Seo H. Collagen and alkaline phosphatase gene expression during bone morphogenetic protein (BMP)-induced cartilage and bone differentation. Clin Orthop Rel Res 1993;292:337-344.

Sandell LJ, Sugai JV and Trippel SB. Expression of collagens I, II, X, and XI and aggrecan mRNAs by bovine growth plate chondrocytes in situ. J Bone Joint Surg 1994;12:1-14.

Sarzi-Puttini P, Cimmino MA, Scarpa R, Caporali R, Parazzini F, Zaninelli A, Atzeni F and Canesi B. Osteoarthritis: an overview of the disease and its treatment strategies. Semin Arthritis Rheum 2005;35:1-10.

Savontaus M, Ihanamäki T, Metsäranta M, Vuorio E and Sandberg-Lall M. Localization of type II collagen mRNA isoforms in the developing eyes of normal and transgenic mice with a mutation in type II collagen gene. Invest Ophtalmol Vis Sci 1997;38:930-942.

Savontaus M, Metsäranta M and Vuorio E. Retarded skeletal development in transgenic mice with a type II collagen mutation. Am J Pathol 1996;149:2169-2182.

Seibel MJ. Biochemical markers of bone formation and bone resorption. In: Galteau M-M, Siest G and Henny J (Eds.) Biologie Prospective: Comptes rendus du 8E Colloque de Pont- -Moussan. 1st ed. John Libbey Eurotext, Paris, 1993.

Seibel MJ, Cosman F, Shen V, Gordon SE, Dempster DW, Ratcliffe A and Lindsay R. Urinary hydroxypyridinium crosslinks of collagen as markers of bone resorption and estrogen efficacy in postmenopausal osteoporosis. J Bone Miner Res 1993;8:881-889.

102Seyedin SM, Kung VT, Daniloff YN, Hesley RP, Gomez B, Nielsen LA, Rosen HN and Zuk RF. Immunoassay for urinary pyridoline: The new marker of bone resorption. J Bone Miner Res 1993;8:635-641.

Shen V, Dempster DW, Birchman R, Mellish RW, Church E, Kohn D and Lindsay R. Lack of changes in histomorphometric, bone mass, and biochemical parameters in ovariohysterectomized dogs. Bone 1992;13:311-316.

Shimegi S, Yanagita M, Okano H, Yamada M, Fukui H, Fukumura Y, Ibuki Y and Kojima I. Physical exercise increases bone mineral density in postmenopausal women. Endocrine J 1994;41:49-56.

Shinar DM, Schmidt A, Halperin D, Rodan GA and Weinreb M. Expression of αv and α3 integrin subunits in rat osteoclasts in situ. J Bone Miner Res 1993;8:403-414.

Shipp CC, Berger PS, Deehr MS and Dawson-Hughes B. Precision of dual-photon absorptiometry. Calcif Tissue Int 1988;42:287-292.

Silberberg M and Silberberg R. Age changes of bones and joints in various strains of mice. Am J Anat 1941;68:69-95.

Skerry TM. The response of bone to mechanical loading and disuse: fundamental principles and influences on osteoblast/osteocyte homeostasis. Arch Biochem Biophys 2008;473:117-123.

Skerry TM, Bitensky L, Chayen J and Lanyon LE. Loading-related reorientation of bone proteoglycan in vivo. Strain memory in bone tissue? J Orthop Res 1988;6:547-551.

Sokoloff L. Natural history of degenerative joint disease in small laboratory animals. AMA Arch Path 1956;0:118-128.

Sokolov BP, Prytkov AN, Tromp G, Knowlton RG and Prockop DJ. Exclusion of COL1A1, COL1A2, and COL3A1 genes as candidate genes for Ehlers-Danlos syndrome type I in one large family. Human Genetics 1991;88:125-129.

Suominen H. Muscle training for bone strength. Aging Clin Exp Res 2006;18:85-93.

Suyama H, Moriwaki K, Niida S, Maehara Y, Kawamoto M and Yuge O. Osteoporosis following chronic constriction injury of sciatic nerve in rats. J Bone Miner Metab 2002;20:91-97.

Søgaard CH, Danielsen CC, Thorling EB and Mosekilde L. Long-term exercise of young and adult female rats: effect on femoral neck biomechanical competence and bone structure. J Bone Miner Res 1994;9:409-416.

Takai S, Sakakida K, Yamashita F, Suzu F and Izuta F. Rotational alignment of the lower limb in osteoarthritis of the knee. Int Orthop 1985;9:209-215.

Takeda S. Central control of bone remodeling. Biochem Biophys Res Commun 2005;328:697-699.

Tarbell JM, Weinbaum S and Kamm RD. Cellular fluid mechanics and mechanotransduction. Ann Biomed Eng 2005;33:1719-1723.

Termine JD. Bone matrix proteins and the mineralization process. In: Favus MJ (Ed.) Primer on the metabolic bone diseases and disorders of mineral metabolism. 2nd ed. Raven Press, New York, 1993.

103Teti A and Zallone A. Do osteocytes contribute to bone mineral homeostasis? Osteocytic osteolysis revisited. Bone 2009;44:11-16.

The Social Insurance Institution of Finland. Statistical yearbook of pensioners in Finland 2006. Hakapaino Oy, 2007.

Thomas JT, Ayad S and Grant ME. Cartilage collagens: strategies for the study of their organization and expression in the extracellular matrix. Ann Rheum Dis 1994;53:488-496.

Thomas PA. Racial and ethnic differences in osteoporosis. J Am Acad Orthop Surg 2007;15 (Suppl 1):S26-30.

Thomsen JS, Morukov BV, Vico L, Alexandre C, Saparin PI and Gowin W. Cancellous bone structure of iliac crest biopsies following 370 days of head-down bed rest. Aviat Space Environ Med 2005;76:915-922.

Tommerup LJ, Raab DM, Crenshaw TD and Smith EL. Does weight-bearing exercise affect non-weight-bearing bone? J Bone Miner Res 1993;8:1053-1058.

Tsumaki N, Kimuri T, Matsui Y, Nakata K and Ochi T. Separable cis-regulatory elements that contribute to tissue- and site-specific α2(XI) collagen gene expression in the embryonic mouse cartilage. J Cell Biol 1996;134:1573-1582.

Turner CH and Burr DB. Basic biomechanical measurements of bone: a tutorial. Bone 1993;14:595-608.

Uebelhart D, Bernard J, Hartmann DJ, Moro L, Uebelhart B, Rehailia M, Mauco G, Schmitt DA, Alexandre C and Vico L. Modifications of bone and connective tissue after orthostatic bedrest. Osteoporosis Int 2000;11:59-67.

Uhthoff HK and Jaworski ZFG. Bone loss in response to long-term immobilisation. J Bone Joint Surg 1978;60-B:420-429.

Vaes G. Cellular biology and biochemical mechanism of bone resorption. A review of recent developments on the formation, activation, and mode of action of osteoclasts. Clin Orthop Relat Res 1988:239-271.

Vandenberg P, Khillan JS, Prockop DJ, Helminen H, Kontusaari S and Ala-Kokko L. Expression of a partially deleted gene of human type II procollagen (COL2A1) in transgenic mice produces a chondrodysplasia. Proc Natl Acad Sci 1991;88:7640-7644.

Vanderschueren D, Van Herck E, Schot P, Rush E, Einhorn T, Geusens P and Bouillon R. The aged male rat as a model for human osteoporosis: Evaluation by nondestructive measurements and biomechanical testing. Calcif Tissue Int 1993;53:342-347.

Vikkula M, Palotie A, Ritvaniemi P, Ott J, Ala-Kokko L, Sievers U, Aho K and Peltonen L. Early-onset osteoarthritis linked to the type II procollagen gene. Arthritis & Rheumatism 1993;36:401-409.

Väänänen HK. Mechanism of bone turnover. Ann Med 1993;25:353-359.

Väänänen HK and Laitala-Leinonen T. Osteoclast lineage and function. Arch Biochem Biophys 2008;473:132-138.

Walsh WR, Sherman P, Howlett CR, Sonnabend DH and Ehrlich MG. Fracture healing in a rat osteopenia

104model. Clin Orthop Relat Res 1997:218-227.

Wang L, McMahan CA, Banu J, Okafor MC and Kalu DN. Rodent model for investigating the effects of estrogen on bone and muscle relationship during growth. Calcif Tissue Int 2003;72:151-155.

Wang XD, Masilamani NS, Mabrey JD, Alder ME and Agrawal CM. Changes in the fracture toughness of bone may not be reflected in its mineral density, porosity, and tensile properties. Bone 1998;23:67-72.

Waterston RH, Lindblad-Toh K, Birney E, Rogers J, Abril JF, Agarwal P, Agarwala R, Ainscough R, Alexandersson M, An P, Antonarakis SE, Attwood J, Baertsch R, Bailey J, Barlow K, Beck S, Berry E, Birren B, Bloom T, Bork P, Botcherby M, Bray N, Brent MR, Brown DG, Brown SD, Bult C, Burton J, Butler J, Campbell RD, Carninci P, Cawley S, Chiaromonte F, Chinwalla AT, Church DM, Clamp M, Clee C, Collins FS, Cook LL, Copley RR, Coulson A, Couronne O, Cuff J, Curwen V, Cutts T, Daly M, David R, Davies J, Delehaunty KD, Deri J, Dermitzakis ET, Dewey C, Dickens NJ, Diekhans M, Dodge S, Dubchak I, Dunn DM, Eddy SR, Elnitski L, Emes RD, Eswara P, Eyras E, Felsenfeld A, Fewell GA, Flicek P, Foley K, Frankel WN, Fulton LA, Fulton RS, Furey TS, Gage D, Gibbs RA, Glusman G, Gnerre S, Goldman N, Goodstadt L, Grafham D, Graves TA, Green ED, Gregory S, Guigo R, Guyer M, Hardison RC, Haussler D, Hayashizaki Y, Hillier LW, Hinrichs A, Hlavina W, Holzer T, Hsu F, Hua A, Hubbard T, Hunt A, Jackson I, Jaffe DB, Johnson LS, Jones M, Jones TA, Joy A, Kamal M, Karlsson EK, Karolchik D, Kasprzyk A, Kawai J, Keibler E, Kells C, Kent WJ, Kirby A, Kolbe DL, Korf I, Kucherlapati RS, Kulbokas EJ, Kulp D, Landers T, Leger JP, Leonard S, Letunic I, Levine R, Li J, Li M, Lloyd C, Lucas S, Ma B, Maglott DR, Mardis ER, Matthews L, Mauceli E, Mayer JH, McCarthy M, McCombie WR, McLaren S, McLay K, McPherson JD, Meldrim J, Meredith B, Mesirov JP, Miller W, Miner TL, Mongin E, Montgomery KT, Morgan M, Mott R, Mullikin JC, Muzny DM, Nash WE, Nelson JO, Nhan MN, Nicol R, Ning Z, Nusbaum C, O'Connor MJ, Okazaki Y, Oliver K, Overton-Larty E, Pachter L, Parra G, Pepin KH, Peterson J, Pevzner P, Plumb R, Pohl CS, Poliakov A, Ponce TC, Ponting CP, Potter S, Quail M, Reymond A, Roe BA, Roskin KM, Rubin EM, Rust AG, Santos R, Sapojnikov V, Schultz B, Schultz J, Schwartz MS, Schwartz S, Scott C, Seaman S, Searle S, Sharpe T, Sheridan A, Shownkeen R, Sims S, Singer JB, Slater G, Smit A, Smith DR, Spencer B, Stabenau A, Stange-Thomann N, Sugnet C, Suyama M, Tesler G, Thompson J, Torrents D, Trevaskis E, Tromp J, Ucla C, Ureta-Vidal A, Vinson JP, Von Niederhausern AC, Wade CM, Wall M, Weber RJ, Weiss RB, Wendl MC, West AP, Wetterstrand K, Wheeler R, Whelan S, Wierzbowski J, Willey D, Williams S, Wilson RK, Winter E, Worley KC, Wyman D, Yang S, Yang SP, Zdobnov EM, Zody MC and Lander ES. Initial sequencing and comparative analysis of the mouse genome. Nature 2002;420:520-562.

Wehrli FW. Structural and functional assessment of trabecular and cortical bone by micro magnetic resonance imaging. J Magn Reson Imaging 2007;25:390-409.

Weinreb M, Rodan GA and Thompson DD. Osteopenia in the immobilized rat hind limb is associated with increased bone resorption and decreased bone formation. Bone 1989;10:187-194.

Welten DC, Kemper HCG, Post GB, van Mechelen W, Twisk J, Lips P and Teule GJ. Weight-bearing activity during youth is a more important factor for peak bone mass than calcium intake. J Bone Miner Res 1994;9:1089-1096.

Wu J-J, Woods PE and Eyre DR. Identification of cross-linking 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 1992;267:23007-23014.

Wuthier RE. A review of the primary mechanism of endochondral calcification with special emphasis on the role of cells, mitochondria and matrix vesicles. Clin Orthop Relat Res 1982:219-242.

105Yamada M, Ito M, Hayashi K and Nakamura T. Calcaneus as a site for assessment of bone mineral density. AJR 1993;161:621-627.

Yang R-S, Wang S-S, Lin H-J, Liu T-K, Hang Y-S and Tsai K-S. Differential effects of bone mineral content and bone area on vertebral strength in a swine model. Calcif Tissue Int 1998;63:86-90.

Yeh JK, Aloia JF, Chen MM, Tierney JM and Sprintz S. Influence of exercise on cancellous bone of the aged female rat. J Bone Miner Res 1993;8:1117-1125.

You L, Temiyasathit S, Lee P, Kim CH, Tummala P, Yao W, Kingery W, Malone AM, Kwon RY and Jacobs CR. Osteocytes as mechanosensors in the inhibition of bone resorption due to mechanical loading. Bone 2008;42:172-179.

Young HD and Freedman RA. University physics. Pearson, Addison and Wesley, San Francisco, 2004.

Ørtoft G, Mosekilde L, Hasling C and Mosekilde L. Estimation of vertebral body strength by dual photon absorptiometry in elderly individuals: comparison between measurement of total vertebral and vertebral body bone mineral. Bone 1993;14:667-673.

106

107

LIST OF ORIGINAL PUBLICATIONS

Kuopio University Publications D. Medical Sciences D 427. Saarelainen, Soili. Immune Response to Lipocalin Allergens: IgE and T-cell Cross-Reactivity. 2008. 127 p. Acad. Diss. D 428. Mager, Ursula. The role of ghrelin in obesity and insulin resistance. 2008. 123 p. Acad. Diss. D 429. Loisa, Pekka. Anti-inflammatory response in severe sepsis and septic shock. 2008. 108 p. Acad. Diss. D 430. Joukainen, Antti. New bioabsorbable implants for the fixation of metaphyseal bone : an experimental and clinical study. 2008. 98 p. Acad. Diss. D 431. Nykänen, Irma. Sepelvaltimotaudin prevention kehitys Suomessa vuosina 1996-2005. 2008. 158 p. Acad. Diss. D 432. Savonen, Kai. Heart rate response to exercise in the prediction of mortality and myocardial infatction: a prospective population study in men. 2008. 165 p. Acad. Diss. D 433. Komulainen, Pirjo. The association of vascular and neuroprotective status indicators with cognitive functioning: population-based studies. 2008. Acad. Diss. D 434. Hassinen, Maija. Predictors and consequences of the metabolic syndrome: population-based studies in aging men and women. 2008. Acad. Diss. D 435. Saltevo, Juha. Low-grade inflammation and adiponectin in the metabolic syndrome. 2008. 109 p. Acad. Diss. D 436. Ervasti, Mari. Evaluation of Iron Status Using Methods Based on the Features of Red Blood Cells and Reticulocytes. 2008. 104 p. Acad. Diss. D 437. Muukka, Eija. Luomun tie päiväkotiin: luomuruokailun toteutettavuus ja ravitsemuksellinen merkitys päiväkotilapsille. 2008. 168 p. Acad. Diss. D 438. Sörensen, Lars. Work ability and health-related quality of life in middle-aged men: the role of physical activity and fitness. 2008. 83 p. Acad. Diss. D 439. Maaranen, Päivi. Dissociation in the finnish general population. 2008. 97 p. Acad. Diss. D 440. Hyvönen, Juha. Suomen psykiatrinen hoitojärjestelmä 1990-luvulla historian jatkumon näkökulmasta. 2008. 279 p. Acad. Diss. D 441. Mäkinen, Heidi. Disease activity and remission in rheumatoid arthritis: comparison of available disease activity measures and development of a novel disease sctivity indes: the mean overall index for rheumatoid arthritis (MOI-RA). 2008. 129 p. Acad. Diss. D 442. Kousa, Anne. The regional association of the hardness in well waters and the incidence of acute myocardial infarction in rural Finland. 2008. 92 p. Acad. Diss.