UWA Research Publication

©2013 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. This is the accepted version of the following article: He, B., Wu, J.P., Chen, H.H., Kirk, T.B. & Xu, J. (2013). Elastin Fibers Display a Versatile Microfibril Network in Articular Cartilage Depending on the Mechanical Microenvironments. JOURNAL OF ORTHOPAEDIC RESEARCH, 31(9), 1345-1353, which has been published in final form at http://dx.doi.org/10.1002/jor.22384 This version was made available in the UWA Research Repository on the 1st of September 2104, in compliance with the publisher’s policies on archiving in institutional repositories.

Use of the article is subject to copyright law.

1

Elastin Fibres Display a Versatile Microfibril Network in Kangaroo Articular 1

Cartilage Depending on the Mechanical Microenvironments 2

Bo He, 1 Jian Ping Wu,

2 Hong Hui Chen,

3 Thomas Brett Kirk,

2 Jiake Xu,

1 3

Running title: Elastin fibres in Kangaroo Articular Cartilage 4

1School of Pathology and Laboratory Medicine, University of Western Australia, Western 5

Australia, Australia 6

2Department of Mechanical Engineering, Curtin University, Western Australia, Australia 7

3Guangzhou Institute of Traumatic Surgery, Guangzhou Red Cross Hospital, Medical College, 8

Jinan University, Guangzhou, 510220, China 9

10

* Address correspondence and reprint requests to: Jiake Xu11

School of Pathology and Laboratory Medicine, University of Western Australia, QE II Medical 12

centre, Nedlands, Perth, Australia 6009, Tel: 618 9346 2739, Fax: 618 9346 2891, E-mail: 13

jiake.xu@uwa.edu.au 14

15

2

ABSTRACT 16

Elastin fibres are major extracellular matrix macromolecules that are critical in maintaining 17

the elasticity and resilience of tissues such as blood vessels, lungs and skins. However, the 18

role of elastin in articular cartilage is poorly defined. The present study investigated the 19

organisation of elastin fibre in articular cartilage, its relationship to collagen fibres and the 20

architecture of elastin fibres from different mechanical environments by using a kangaroo 21

model. Five morphologies of elastin fibres were identified: straight fibre, straight fibre with 22

branches, branching fibres directly associated with chondrocyte, wave fibre and fine elastin. 23

The architecture of the elastin network varied significantly with cartilage depth. In the most 24

superficial layer of tibial plateau articular cartilage, dense elastin fibres formed a distinctive 25

cobweb-like meshwork which was parallel to the cartilage surface. In the superficial zone, 26

elastin fibres were well organized in a preferred orientation which was parallel to collagen 27

fibres. In the deep zone, no detectable elastin fibre was found. Moreover, differences in the 28

organization of elastin fibres were also observed between articular cartilage from the tibial 29

plateau, femoral condyle and distal humerus. This study unravels the detailed microarchitecture 30

of elastin fibres which display a well-organized three-dimensional versatile network in articular 31

cartilage. Our findings imply that elastin fibres may play a crucial role in maintaining the 32

integrity, elasticity and the mechanical properties of articular cartilage, and that the local 33

mechanical environment affects the architectural development of elastin fibres. 34

Keywords: articular cartilage; collagen fibres; elastin fibres, mechanical environment 35

36

Elastin fibres are macromolecules comprising an amorphous core of extensively cross-linked 37

3

elastin, surrounded by fibrillin-rich microfibrils.1-3

They are abundant in the extracellular 38

matrix (ECM) of many tissues such as connective tissue proper, elastin cartilage and fetal tissue, 39

as well as in organs such as skin, lungs, arteries, veins. The prevailing mechanical functions of 40

elastin fibres are thought to endow the tissues with the critical properties of elasticity and 41

resilience.4, 5

Increasing studies have shown that elastin fibres also play a critical role in 42

regulating arterial development via controlling proliferation of vascular smooth muscle and 43

stabilizing arterial structure;4 mediating disease states of supravalvular aortic stenosis (SVAS)

6, 44

7 and Williams syndrome;

8 and enhancing the mechanical integrity of the lumbar annulus 45

fibrosus.9 46

The role of elastin in articular cartilage is poorly defined as early histochemistry studies 47

suggested that little elastin was present in articular cartilage.5, 10-12

In articular cartilage, water, 48

collagen matrix and proteoglycans (PGs) have been recognised to be the most influential 49

elements in maintaining the mechanical functions of articular cartilage.13-16

A recent study 50

using nonlinear optical microscopy has shown elastin-like fibres in bovine articular cartilage,17

51

and further immunohistochemistry studies confirm the identity of elastin fibres in bovine17, 18

52

and equine19

articular cartilage. Although the fibres are reported to occupy the superficial zone 53

of articular cartilage, the arrangement and function of elastin fibres in the articular cartilage, 54

especially of different joints with distinct mechanical environments, remain to be elucidated. 55

By using multiphoton microscopy and sulforhodamine B (SRB) which stains elastin 56

fibres,20

this study aims to examine the elastin fibre network within kangaroo articular cartilage 57

and their relationship with the mechanical function of articular cartilage. We directly confirmed 58

that the two-photon fluorescence (TPF) signal was of elastin fibre and they had a close 59

4

relationship with the collagen matrix and chondrocytes. We also found that the architecture and 60

content of the elastin fibre network varied significantly with cartilage depth and with the 61

mechanical environments of articular cartilage. 62

METHODS 63

Sampling 64

Six forelimb and six hindlimb joints were obtained from six male kangaroos aged at 65

approximately 5 years. Articular cartilage from the tibial plateau, femoral condyle and distal 66

humerus was inspected for the absence of macroscopic degeneration. Cylindrical cartilage 67

samples (3 plugs from each site) were harvested using 5 mm diameter punches from the medial 68

tibial plateau (meniscus covered area), the weight-bearing areas of the medial femoral condyle 69

and medial distal humerus (squares in Fig. 1). The cartilage plugs were sliced into samples 70

corresponding to the superficial and deep zone. 71

Articular Cartilage Staining 72

For imaging elastin, articular cartilage sample was stained with SRB as previously described. 21

73

Briefly, the samples were stained in 1 mg/ml SRB solution for 1 minute. After a thorough wash 74

in PBS (PBS, pH 7.2), the cartilage samples were mounted between a coverslip and a glass 75

slide. 76

Imaging 77

Images of the elastin and collagen were acquired by a multiphoton confocal laser scanning 78

microscope (Leica TCS SP2 acousto-optical beam splitter) (AOBS) equipped with several light 79

sources, including a diode-pumped solid-state (DPSS) laser at 561 nm and the multiphoton 80

laser used in this study. The multiphoton laser was generated by a Spectra Physics Mai Tai 81

5

sapphire system and could be tuned from 710 to 990 nm. As a comparison to previous reported 82

elastin-like fibres from TPF signal,17, 19

an unstained cartilage sample was excited by a laser of 83

890 nm. The TPF signal was collected in the epi-direction at 500 to 550 nm emission 84

wavelength and the second harmonic generation (SHG) signal from collagen was collected in 85

the forward direction at 445 nm. For SRB stained samples, the DPSS laser at 561 nm was used 86

and the multiphoton laser at 890 nm. The SHG signal from collagen was acquired at 445 nm 87

and the SRB fluorescent signal from elastin was acquired at 565-590 nm emission wavelength. 88

All images were acquired as a series of stacks at an imaging step size of 0.5 μm. 89

Fibre Orientation Analysis 90

Digital image analysis software ImageJ (NIH, Maryland, USA) was used to conduct the image 91

analysis. The orientation of the elastin fibres and the collagen fibres was analysed using 92

OrientationJ (an ImageJ-plug-in) which was validated to study the collagen orientation in a 93

previous study.22

The hue-saturation-brightness (HSB) colour coded images, which used 94

colours to show the orientation of an object, were generated from the images acquired by 95

microscopy. The relative orientation of the elastin fibres and collagen fibres was plotted and 96

their predominant orientation and coherency value were calculated. The image stacks of the 97

elastin and collagen fibres acquired at the same region were merged and reconstructed into 98

three-dimensional (3D) images to study their spatial relationship. 99

RESULTS 100

Verification of the Elastin Fibre in Articular Cartilage 101

To confirm the presence of elastin fibre in articular cartilage, unstained articular cartilage of 102

kangaroo tibial plateau was firstly imaged using a multiphoton confocal laser scanning 103

6

microscope and was later stained for SRB excitation. From an unstained surface of articular 104

cartilage, two signals were collected: SHG indicating the collagen fibre network and TPF 105

representing an additional fluorescent fibre network which did not overlap with the collagen 106

fibres (Fig. 2A and B). For the SRB stained articular cartilage sample, three signals, TPF, SHG 107

and SRB were collected (Fig. 2C-E). The elastin fibre network shown in the SRB fluorescent 108

image (Fig. 2E) was similar to that in the TPF image (Fig. 2C) and the montage of the left half 109

part of TPF image and the right half part of SRB image showed a perfect match between the 110

TPF image and SRB image in terms of fibre extension (Fig. 2F). A merged image of TPF and 111

SHG (Fig. 2G) and a merged image of SRB and SHG (Fig. 2H) further confirmed the identity of 112

elastin fibres. The comparison of TPF and SRB images indicated a higher resolution and less 113

digital noise in SRB stained fluorescent images. 114

Morphological Characteristics of Elastin Fibres in Articular Cartilage 115

As visualized by SRB fluorescence, five morphologically different elastin fibres were found in 116

tibial plateau articular cartilage of kangaroo: straight fibre with branches in the ECM (Fig. 3A 117

and C), branching fibres associated with chondrocyte (squared in Fig. 3D, Fig. 3E), straight 118

fibre in the ECM (Fig. 3F), wave fibre in the ECM (Fig. 3B) and fine elastin on the cell surface 119

and in the pericellular matrix (Fig. 3G-J). Elastin fibres with branches were most abundant and 120

formed a complex three dimensional network by cross-linking extensively in the same plane 121

and branching into different layers (Fig. 3A and C). Straight elastin fibres represented the 122

second abundant type and they were highly oriented and parallel to each other (Fig. 3F). Fine 123

elastin was found in the pericellular matrix (arrow in Fig. 3G-J) and the surface of chondrocytes 124

(arrow head in Fig. 3H) where no individual fibre was resolvable (Fig. 3G-J). A few wave fibres 125

7

representing a relax status were also found in the ECM of kangaroo articular cartilage (Fig. 3B). 126

Branching fibres directly associated with chondrocyte represented the least common type of 127

elastin fibres in the articular cartilage of kangaroo tibial plateau (square in Fig. 3D, Fig. 3E). As 128

shown, a relative bigger elastin fibre stem (arrowhead in Fig. 3E) was directly associated with a 129

chondrocyte (square in Fig. 3E), and smaller branches (arrow in Fig. 3E) extended from the 130

stem to various directions which formed a complex three- dimensional root-like structure. 131

Variation of Elastic Network with Depth 132

In the most superficial layer of tibial plateau articular cartilage, from the articular surface down 133

to around 3μm depth, the elastin fibres ran parallel to the articular surface and were organized in 134

a cobweb-like structure with extensive cross-linking to one another (Fig. 4A-C). Beneath the 135

most superficial layer, the elastin fibres in the superficial zone showed a preferred orientation 136

and ran parallel to the long axis of the chondrocytes (Fig. 4D-F). In the deep zone, no individual 137

elastin fibres could be resolved but dense and fine elastin was found in the pericellular matrix 138

and on chondrocyte surfaces (arrow in Fig. 4G-I). 139

The Structural Relationship between the Elastin and Collagen Fibres 140

The confocal fluorescent images (Fig. 5A and B) and the corresponding HSB-colour-coded 141

images (Fig. 5 D and E) indicated that the elastin fibres were randomly distributed but the 142

collagen fibres were highly oriented in the most superficial layer of tibial plateau articular 143

cartilage. Quantitative study showed that the elastin fibres were distributed more widely in all 144

directions (Fig. 5C) with a lower coherency of 17% and a predominant degree of -71° (Fig. 5H). 145

In contrast, the orientation of the collagen fibres in the same region was more isotropic with 146

higher coherency of 51% and a dominant degree of 25° (Fig. 5F and H). The merged image (Fig. 147

8

5G) and 3 dimensional reconstruction image (Fig. 7A) further confirmed the coarse distribution 148

of elastin fibres but highly oriented collagen fibres in the most superficial layer. 149

In the superficial zone, the elastin fibres (Fig. 6A and B) mostly ran parallel to the collagen 150

fibres (Fig. 6D and E). Quantitative study showed that the elastin fibres and the collagen matrix 151

had similar distribution ranges, close predominant degrees and coherency values (Fig. 6C, F 152

and H). The merged image (Fig. 6G) and three dimensional reconstruction image (Fig. 7B) also 153

indicated that both the elastin fibres and the collagen fibres were highly oriented in a similar 154

direction. 155

Variation of Elastin Fibres with Joints 156

The variation between joints in terms of morphology and structure of the elastin fibre networks 157

was most significant in the most superficial layer. In the most superficial layer of articular 158

cartilage, the elastin fibres were large, coarse and heavily cross-linked without a preferred 159

orientation in tibial plateau (Fig. 8A); were also coarse but showed a predominant orientation in 160

femoral condyle (Fig. 8B); and were finer and shorter in distal humerus (Fig. 8C). 161

Compared to the most superficial layer, less elastin fibres with more clear orientation were 162

found in the superficial zone of all the joints studied (Fig. 8D-F). However, the elastin fibre 163

variation between joints was not obvious except the seemingly larger fibres found in the tibial 164

plateau articular cartilage (Fig. 8D). 165

166

DISCUSSION 167

The present study shows that elastin exists in the ECM of kangaroo articular cartilage. The 168

elastin fibres in the superficial zone are integrated as a three-dimensional network which varies 169

9

significantly with depth of articular cartilage and different joint of kangaroo. Given the 170

significant volume and complexity of the elastin network, it is suggested that elastin fibres 171

could have both biochemical and biomechanical roles. 172

The elastin fibre system has a close relationship with other components of tissues. Early 173

investigations have found an association between ealstin fibres and PGs in dermis,23

cultured 174

smooth muscle cell24

and skin.25

Bridging molecules were also identified to anchor elastin 175

fibres to collagens (e.g. MP78/70, the α1 chain of type VIII collagen)26, 27

and cell surface (e.g. 176

Fibulin-5).28, 29

We revealed that the elastin fibres in articular cartilage were closely related to 177

collagen (Fig 5, 6, and 7) and that some elastin fibres were anchored to chondrocytes (Fig. 2D 178

and E). Additionally, the cross-linking of massive elastin fibres themselves in the same plane 179

and branching into different depths (as shown in Fig. 3A-E and Fig. 4A-F) contribute a three 180

dimensional elastic network for the articular cartilage surface. Thus, along with other ECM 181

components, elastin fibres could also play a crucial role in structural support to the articular 182

cartilage. 183

The varied architecture of elastin with cartilage depth could have significant biomechanical 184

implications. Firstly, fine and dense elastin found around the chondrocytes (Fig. 2D-E, G-J) 185

forms a niche and may provide a crucial microenvironment where chondrocytes suffer only a 186

mild stretching despite the vigorous forces applied to the articular cartilage. The cobweb-like 187

elastin fibre network found in the most superficial layer could render the articular cartilage 188

surface more resistant to stretches in different directions, as previous studies have shown that 189

tensile strain predominates near the surface with hydrostatic compressive forces dominating 190

below the surface layers.30, 31

Moreover, the elastin fibres in the superficial zone were found to 191

10

be parallel to the adjacent collagen matrix and interdispersed in the collagen matrix (Fig. 6, Fig. 192

7B). Whether the elasticity and resilience properties of elastin fibres could enable a protective 193

mechanism to collagen under harsh shear stresses is unclear from this study, but this finding 194

merits further investigation to reveal the detailed relationship between elastin fibres and 195

collagen matrix in articular cartilage under stress. 196

Previous studies in other tissues have showed that mechanical stimuli affect the 197

development of elastin fibres. Cyclical mechanical stretches upregulate gene expression of 198

elastin in human lamina cribrosa cells in vitro;32

mechanical stress could induce the deposition 199

of fibrillin-1 in vitro;33

and pressure simulating orthodontic force could upregulate the 200

tropoelastin gene.34

The current study revealed variations in elastin fibre network between tibial 201

plateau, femoral condyle and distal humerus articular cartilage (Fig. 8). This variation with 202

different types of joints implies a potential relationship with the local mechanical environments. 203

Further studies, however, are essential in order to learn whether mechanical stimuli also affect 204

the development of the elastin fibre network in articular cartilage. 205

Marfan syndrome in humans is caused by mutations in the FBN1 gene which encodes 206

fibrillin-1 protein that is essential for the proper formation and biogenesis of elastin fibres.35, 36

207

Marfan syndrome develops signs of abnormal joint flexibility, and is linked to early 208

osteoarthritis.37, 38

Our findings of elastin fibres in articular cartilage might provide novel 209

insights into the importance of elastin fibers in skeletal anomalies, such as osteoarthritis. 210

In conclusion, the present work assessed the microstructural characteristics of elastin fibres 211

in kangaroo articular cartilage. A well organized three dimensional elastin network was 212

revealed in the ECM of articular cartilage surface as well as surrounding the chondrocyte 213

11

surface and in the pericellular matrix. The architecture of elastin fibres may provide novel 214

insights into the role of the elastin network in articular cartilage and cartilage degeneration. 215

Changes of elastin fibres in diseased articular cartilage, such as osteoarthritic cartilage, will be 216

an important subject of further investigation. 217

218

Acknowledgements 219

The authors acknowledge the facilities, scientific and technical assistance of the Australian 220

Microscopy & Microanalysis Research Facility at the Centre for Microscopy, Characterisation 221

& Analysis, The University of Western Australia, a facility funded by the University, and the 222

State and Commonwealth Governments. This study was also supported by a grant from the 223

National Natural Science Foundation of China (NSFC) (No. 81228013) to Jiake Xu and an 224

award of Australia NHMRC fellowship (ID404179 JP Wu). The authors thank Professor Robert 225

Cook for his critical edition of this manuscript. 226

227

References 228

1. Debelle L, Tamburro AM. 1999. Elastin: molecular description and function. The International Journal of 229

Biochemistry & Cell Biology 31:261-272. 230

2. Hesse I. 1987. The occurrence of elastic system fibres in the matrix of normal articular cartilage. Cell and 231

Tissue Research 248:589-593. 232

3. Montes GS. 1996. Structural biology of the fibres of the collagenous and elastic systems. Cell Biol Int 233

20:1527. 234

4. Li DY, Brooke B, Davis EC, et al. 1998. Elastin is an essential determinant of arterial morphogenesis. Nature 235

12

393:276280. 236

5. Kielty CM, Sherratt MJ, Shuttleworth CA. 2002. Elastic fibres. J Cell Sci 115:28172828. 237

6. Curran ME, Atkinson DL, Ewart AK, et al. 1993. The elastin gene is disrupted by a translocation associated 238

with supravalvular aortic stenosis. Cell 73:159168. 239

7. Metcalfe K, Rucka AK, Smoot L, et al. 2000. Elastin: mutational spectrum in supravalvular aortic stenosis. 240

Eur J Hum Genet 8:955963. 241

8. Ewart AK, Morris CA, Atkinson D, et al. 1993. Hemizygosity at the elastin locus in a developmental disorder, 242

Williams syndrome. Nat Genet 5:1116. 243

9. Smith LJ, Byers S, Costi JJ, Fazzalari NL. 2008. Elastic fibers enhance the mechanical integrity of the human 244

lumbar anulus fibrosus in the radial direction. Ann Biomed Eng 36:214223. 245

10. Ushiki T. 2002. Collagen fibers, reticular fibers and elastic fibers. A comprehensive understanding from a 246

morphological viewpoint. Arch Histol Cytol 65:109126. 247

11. Cotta-Pereira G, Del-Caro LM, Montes GS. 1984. Distribution of elastic system fibers in hyaline and fibrous 248

cartilages of the rat. Acta Anat (Basel) 119:8085. 249

12. Naumann A, Dennis JE, Awadallah A, et al. 2002. Immunochemical and mechanical characterization of 250

cartilage subtypes in rabbit. J Histochem Cytochem 50:1049-1058. 251

13. Mow VC, Holmes MH, Lai WM. 1984. Fluid transport and mechanical properties of articular cartilage: a 252

review. J Biomech 17:377394. 253

14. Eyre D. 2002. Articular cartilage and changes in Arthritis: Collagen of articular cartilage. Arthritis Res 254

4:3035. 255

15. Shoulders MD, Raines RT. 2009. Collagen structure and stability. Annu Rev Biochem 78:929-958. 256

16. Poole AR, Nelson F, Hollander A, et al. 1995. Collagen II turnover in joint diseases. Acta Orthop Scand 257

13

Suppl 266:88-91. 258

17. Yeh AT, Hammer-Wilson MJ, Van Sickle DC, et al. 2005. Nonlinear optical microscopy of articular cartilage. 259

Osteoarthritis and cartilage / OARS, Osteoarthritis Research Society 13:345352. 260

18. Yu J, Urban JP. 2010. The elastic network of articular cartilage: an immunohistochemical study of elastin 261

fibres and microfibrils. Journal of anatomy 216:533541. 262

19. Mansfield J, Yu J, Attenburrow D, et al. 2009. The elastin network: its relationship with collagen and cells in 263

articular cartilage as visualized by multiphoton microscopy. Journal of anatomy 215:682691. 264

20. Ricard C, Vial J-C, Douady J, Sanden Bvd. 2007. In vivo imaging of elastic fibers using sulforhodamine B. 265

Journal of Biomedical Optics 12:064017. 266

21. He B, Wu JP, Chim SM, et al. 2013. Microstructural analysis of collagen and elastin fibres in the kangaroo 267

articular cartilage reveals a structural divergence depending on its local mechanical environment. 268

Osteoarthritis and Cartilage 21:237-245. 269

22. Rezakhaniha R, Agianniotis A, Schrauwen J, et al. 2011. Experimental investigation of collagen waviness 270

and orientation in the arterial adventitia using confocal laser scanning microscopy. Biomechanics and 271

Modeling in Mechanobiology:113. 272

23. Baccarani-Contri M, Vincenzi D, Cicchetti F, et al. 1990. Immunocytochemical localization of proteoglycans 273

within normal elastin fibers. Eur J Cell Biol 53:305-312. 274

24. Kielty CM, Whittaker SP, Shuttleworth CA. 1996. Fibrillin: evidence that chondroitin sulphate 275

proteoglycans are components of microfibrils and associate with newly synthesised monomers. FEBS Lett 276

386:169-173. 277

25. Zimmermann DR, Dours-Zimmermann MT, Schubert M, Bruckner-Tuderman L. 1994. Versican is expressed 278

in the proliferating zone in the epidermis and in association with the elastic network of the dermis. J Cell 279

14

Biol 124:817-825. 280

26. Gibson MA, Kumaratilake JS, Cleary EG. 1997. Immunohistochemical and ultrastructural localization of 281

MP78/70 (betaig-h3) in extracellular matrix of developing and mature bovine tissues. J Histochem 282

Cytochem 45:1683-1696. 283

27. Sawada H, Konomi H. 1991. The alpha 1 chain of type VIII collagen is associated with many but not all 284

microfibrils of elastic fiber system. Cell Struct Funct 16:455-466. 285

28. Yanagisawa H, Davis EC, Starcher BC, et al. 2002. Fibulin-5 is an elastin-binding protein essential for elastic 286

fibre development in vivo. Nature 415:168-171. 287

29. Nakamura T, Lozano PR, Ikeda Y, et al. 2002. Fibulin-5/DANCE is essential for elastogenesis in vivo. Nature 288

415:171-175. 289

30. Vanwanseele B, Lucchinetti E, Stussi E. 2002. The effects of immobilization on the characteristics of 290

articular cartilage: current concepts and future directions. Osteoarthritis and cartilage / OARS, 291

Osteoarthritis Research Society 10:408-419. 292

31. Carter DR, Wong M. 2003. Modelling cartilage mechanobiology. Philos Trans R Soc Lond B Biol Sci 293

358:1461-1471. 294

32. Kirwan RP, Fenerty CH, Crean J, et al. 2005. Influence of cyclical mechanical strain on extracellular matrix 295

gene expression in human lamina cribrosa cells in vitro. Mol Vis 11:798-810. 296

33. Lorena D, Darby IA, Reinhardt DP, et al. 2004. Fibrillin-1 expression in normal and fibrotic rat liver and in 297

cultured hepatic fibroblastic cells: modulation by mechanical stress and role in cell adhesion. Lab Invest 298

84:203-212. 299

34. Redlich M, Asher Roos H, Reichenberg E, et al. 2004. Expression of tropoelastin in human periodontal 300

ligament fibroblasts after simulation of orthodontic force. Arch Oral Biol 49:119-124. 301

15

35. Dietz HC, Loeys B, Carta L, Ramirez F. 2005. Recent progress towards a molecular understanding of Marfan 302

syndrome. Am J Med Genet C Semin Med Genet 139C:4-9. 303

36. Kainulainen K, Karttunen L, Puhakka L, et al. 1994. Mutations in the fibrillin gene responsible for dominant 304

ectopia lentis and neonatal Marfan syndrome. Nat Genet 6:64-69. 305

37. van de Laar IM, Oldenburg RA, Pals G, et al. 2011. Mutations in SMAD3 cause a syndromic form of aortic 306

aneurysms and dissections with early-onset osteoarthritis. Nat Genet 43:121-126. 307

38. Van de Velde S, Fillman R, Yandow S. 2006. Protrusio acetabuli in Marfan syndrome. History, diagnosis, 308

and treatment. J Bone Joint Surg Am 88:639-646. 309

310

Figure legends: 311

Fig. 1. A photograph of the tibial plateau, femoral condyle and distal humerus of a kangaroo. 312

Images were obtained from the meniscus covered articular cartilage of medial tibial plateau 313

(square area in A), the medial femoral condyle (square area in B) and the medial distal humerus 314

(square area in C). 315

Fig. 2. Verification of elastin fibres in the tibial plateau articular cartilage of kangaroo. 316

Distinctive fibre networks in TPF image (A) and collagen matrix in SHG image (B) were found 317

on the cartilage surface of an unstained sample. In the SRB stained articular cartilage sample, 318

fibres found in the TPF image (C) were identical to elastin fibres in SRB image (E); collagen 319

matrix is imaged by its SHG signal (D). A montage image (F) and merged images (G and H) 320

confirmed the fibres in TPF were elastin fibres (Green = TPF and Red = SHG in G; Green = 321

SRB and Red = SHG in H). 322

Fig. 3. Representative morphology of elastin fibres stained with SRB in the tibial plateau 323

16

articular cartilage of kangaroo: straight elastin fibres with branches (A and C); branching elastin 324

fibres associated with chondrocyte (square in D; and E,); straight elastin fibres (F); wave elastin 325

fibres (B); and fine elastin (G–J, arrow indicated fine elastin in the pericellular matrix, arrow 326

head indicated fine elastin on the chondrocyte surface). 327

Fig. 4. Variation of elastin fibre with depth in the tibial plateau articular cartilage of kangaroo. 328

In the most superficial layer, elastin fibres were extensively cross-linked without a predominant 329

orientation (A-C). Underneath the most superficial layer, elastin fibres were generally oriented 330

in one direction in the superficial zone (D-F). In the deep zone, fine elastin was presented on the 331

chondrocyte surface (indicated as arrow head in H) and in the pericellular matrix (indicated as 332

arrow in G–I) 333

Fig. 5. Orientation analysis showing the organization of elastin fibres was different from that of 334

collagen matrix in the most superficial layer of tibial plateau articular cartilage. 335

Fig. 6. Orientation analysis showing that elastin fibreswere generally parallel to collagen 336

matrix in the superficial zone of tibial plateau articular cartilage. 337

Fig. 7. Three-dimensional reconstruction of cartilage surface showing the relationship between 338

elastin fibres and collagen fibres in the most superficial layer (A) and superficial zone (B). 339

Green = elastin fibres; Red = collagen. 340

Fig. 8. Organization of elastin fibres in articular cartilage differed between from the tibial 341

plateau (A and B), femoral condyle (C and D) and distal humerus (E and F). (stain: SRB) 342

343

17

Fig 1. 344

345

346

347

348

18

349

Fig 2. 350

351

352

353

19

354

Fig 3. 355

356

357

358

359

20

Fig 4. 360

361

362

21

Fig 5. 363

364

365

22

Fig 6. 366

367

368

23

Fig 7. 369

370

371

372

373

374

Fig 8. 375

24

376