research.vu.nl dissertation.pdf · VRIJE UNIVERSITEIT. KU LEUVEN. Biomechanical changes in...

Biomechanical changes in articulation of the jaw joint due to aging

Fereshteh Mirahmadiفرشته میراحمدی


Fereshteh MirahmadiOctober 2018

The studies described in this thesis were carried out at the section Oral Cell Biology and Functional Anatomy of the Academic Centre for Dentistry Amsterdam (ACTA), University of Amsterdam and VU University Amsterdam, MOVE Research Institute Amsterdam, The Netherlands, and the department of Rehabilitation Sciences and the Department of Mechanical Engineering, KU Leuven, with funding by the European Commission through Move-Age, an Erasmus Mundus Joint Doctorate programme (2011-0015).

Cover and Layout by Fereshteh Mirahmadi

Printed by lpskamp printing

ISBN 978-94-028-1160-5

Copyright © by Fereshteh Mirahmadi, Amsterdam 2018. All right reserved

No part of this book may be reproduced, stored in retrievable system, or transmitted in any form or by any means without prior written permission of the author.

VRIJE UNIVERSITEIT

KU LEUVEN


ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad Doctor aan de Vrije Universiteit Amsterdam,op gezag van de rector magnificus prof.dr. V. Subramaniam,

en de graad Doctor in de Biomedische Wetenschappen aan de KU Leuvenop gezag van de rector prof.dr. L. Sels

in het openbaar te verdedigen ten overstaan van de promotiecommissievan de Faculteit der Tandheelkunde van de Vrije Universiteit Amsterdam,

en de Faculteit Bewegings- en Revalidatiewetenschappen van KU Leuven,

op maandag 22 oktober 2018 om 13.45 uur

in de aula van de Vrije Universiteit Amsterdam,

De Boelelaan 1105

door

Fereshteh Mirahmadi

geboren te Najafabad, Iran

promotoren: prof.dr. V. Everts prof.dr. S. Verschuerencopromotoren: dr.ir. J.H. Koolstra prof.dr. G.H. van Lenthe

Dit proefschrift is tot stand gekomen op basis van een daartoe tussen de Vrije Universiteit en de KU Leuven, België, overeengekomen samenwerkingsverband ter regeling van een gezamenlijk promotie als bedoeld in het Promotiereglement Vrije Universiteit, hetgeen mede tot uiting wordt gebracht door de weergave van beide universiteiten op deze titelpagina.

به پدر و مادر عزیزم...

To my parents …

Contents

Chapter 1 11General Introduction

Chapter 2 23Ex vivo thickness measurement of cartilage covering the temporomandibular joint

Chapter 3 37Mechanical stiffness of TMJ condylar cartilage increases after artificial aging by ribose

Chapter 4 57Diffusion of charged and uncharged contrast agents in equine mandibular condylar cartilage is not affected by an increased level of sugar-induced collagen crosslinking

Chapter 5 73Aging does not change the compressive stiffness of mandibular condylar cartilage in horses

Chapter 6 91General Discussion

Appendices 103Summary

Nederlandse samenvatting

Acknowledgment

List of publications

About the author

Abbreviations

AGE Advanced glycation end productANOVA Analyses of variancesBaSO4 Barium sulfateCECT Contrast-enhanced computed tomography CS Chondroitin sulfateCT Computed tomographyDMMB Dimethyl methylene blueDW Dry weight E Ins Instantaneous modulusE St Steady state modulusECM Extracellular matrixEDTA Ethylendiamintetraacetic acidFCD Fixed charge densityGAG GlycosaminoglycanHCl Hydrochloric acid HPLC High performance liquid chromatography Hyp Hydroxy prolineIFP Interstitial fluid pressurizationKS Keratan sulfateLSD Least significant difference MRI Magnetic resonance imaging MW Molecular weight OA OsteoarthritisPBS Phosphate buffer salinePen PentosidinePG ProteoglycanPI Protease inhibitorPicro Picrosirius red PLM Polarized light microscopy R Pearson correlation coefficientSafO/FG SafraninO/Fast green SD Standard deviationTMJ Temporomandibular jointVOI Volume of interest WHO World health organization

Chapter 1

General Introduction

Chapter 1

12

Aging and cartilageThe life expectancy among older adults has shown a noticeable increase during the last decades. “For the first time in human history, people aged 65 and over will outnumber children under age 5 in 2050”, as reported by WHO [1]. The general biological characteristic of aging is defined as an intrinsic and progressive decline in the physiological integrity of the tissue, which happens during the adult period of life leading to deteriorative changes and impaired function [2]. As a consequence of the aging world, the world population will increasingly face age-related complications for instance osteoarthritis (OA) and osteoporosis.

OA is describes as progressive and degenerative disease among older adults in which articular cartilage loss is one of the prominent features [3]. Aging does not necessarily lead to deterioration of cartilage but it has been accepted as one of the major risk factors which make joints susceptible to damage, repair failure and OA [4]. OA is rare in people younger than 45 years, even in those having other main risk factors such as obesity and joint injury [5-7]. A survey of radiological OA in the Netherlands showed a sharp increase in OA of different joints in people older than 50 years. For instance, more than 50% of people between 60-70 suffer from OA in their distal interphalangeal joints [8].

Articular joints consist of an articular cartilage covering the bone surfaces which is surrounded by a synovium. The articular cartilage in such joints provides an almost friction-free surface with unique characteristics for energy absorption and distribution during joint movement. In OA, cartilage degeneration happens alongside the changes in other joint compartments [9]. Such conditions may cause pain, activity limitation, and dependency in daily activities which lead to an impaired quality of life [10]. Global burden of disease studies of 2010 quantified OA as the 11th cause of disability in the world [11].

In humans, the temporomandibular joint (TMJ) is considered to be load-bearing during masticatory function like other articular joints [12]. OA can potentially affect all articular joints in the body, most prevalently knee, hip, and hand [3] possessing hyaline cartilage in which collagen type II is the main type of collagen. However, OA also affects the temporomandibular joint (TMJ) which is fibrocartilage with both collagen type I and II. Among patients with temporomandibular disorders, 11% has TMJ-OA [13]. TMJ-OA causes swelling, pain, and limited mandibular movement [14]. TMJ-OA has a pathobiology similar to OA in the other joints [15].

Cartilage (either fibro- or hyaline) has a simple structure compared with other tissues (single cell type without blood vessels and nerves). However, there is remarkable little repair of damaged cartilage which is thought to be mainly due to the lack of blood supply [16]. Due to the limited repair capacity of cartilage, deterioration and damage can accumulate during aging. There are several theories behind the aging effect on cartilage

General introduction

1

13

leading to degeneration, e.g. the accumulation of microtrauma to cartilage over time due to injuries or overloading, age-related changes in tissue matrix, and stiffening of the cartilage due to an increase in crosslink level in the tissue matrix [17, 18]. Tissue characteristics of cartilage affected by aging can be divided into biochemical and biophysical categories. The biophysical properties of the cartilage such as stuffiness, diffusion, and friction largely depend on its biochemical composition i.e., extracellular matrix (ECM) components. During aging, changes happening in biomechanical properties of cartilage affect its functionality and biophysical properties. These age-related changes in biochemical and biophysical aspects of cartilage will be addressed in the following sections.

Fibrocartilage versus hyaline cartilage TMJ condylar cartilage is characterized as a fibrocartilage and a secondary cartilage without intrinsic growth potential, while hyaline cartilage of other joints, such as knee and hip, can be characterized as primary cartilage. This means that the growth and adaptation of condylar cartilage occur as a response to external forces and applied pressures. The growth of condylar cartilage depends on differentiation of mesenchymal stem cells rather than mitosis of cartilage progenitor cells, which is characteristic for hyaline cartilage [19]. TMJ condylar cartilage has both an articulating function and a growth function in the craniofacial complex [20]. Similar to hyaline cartilage, condylar cartilage of the TMJ consists of several layers from the articulating surface to the subchondral bone. However, it has a distinct morphology and microstructure: in contrast to hyaline cartilage, collagen type I can be found throughout the TMJ condylar cartilage. In the fibrous superficial layer only type I collagen is present, while proteoglycans (PGs) are absent. Collagen type II can be detected in the deeper layers of TMJ cartilage [20, 21]. The orientation of the fibers is parallel to the surface at the superficial layer. Since they have a distinct alignment, this creates an anisotropic response under tensile conditions [22]. The amount of PG is lower in the TMJ than in hyaline cartilage [23]. It is worth noting that large defects in hyaline cartilage are replaced with fibrocartilage with dominant collagen type I, which is the characteristic of mandibular condylar cartilage [21]. The repaired hyaline cartilage has less collagen type II, more collagen type I, and reduced mechanical properties compared to the primary hyaline cartilage [24].

Biochemical composition and agingArticular cartilage consists of two main structural constituents in its ECM, namely collagen and proteoglycans with their associated glycosaminoglycans (GAGs), which together with water are responsible for the functionality of the cartilage.

Collagens

Collagen is the major structural protein in the ECM, which constitutes more than 60 %

Chapter 1

14

of dry weight in articular cartilage [25]. Collagen type II is the principal type of collagen in hyaline cartilage, while collagen type I is found throughout the TMJ [15, 26]; collagen fibers are responsible for the general tissue shape and their main function is to maintain tissue integrity and stability under tensile loading [27, 28]. Cartilaginous collagen fibers have an extremely slow turnover rate; therefore, non-enzymatic crosslinks, which are created through glycation of proteins after maturation, accumulate in the tissue. The final product of non-enzymatic crosslinking is called advanced glycation end products (AGEs) [29, 30]. Pentosidine is a well-characterized and reliable measure of AGEs, of which the highest amount has been reported in articular cartilage [30]. While the enzymatic crosslinking during development and maturation enables the collagen fibers to perform reversible elastic behavior under loading and unloading without degradation, non-enzymatic glycation during aging leads to stiffening of the cartilaginous ECM with reduced flexibility. Tissue stiffening effects can also happen at an accelerated speed for example under diabetic conditions due to the exposure of the tissues to the high concentration of sugar [28]. The stiffening effect of glycation has been shown in hyaline cartilage of different animal models [31, 32] as well as human [30]. Yet, there is lack of studies about such effect in the TMJ condylar cartilage.

Proteoglycans and associated glycosaminoglycans

Proteoglycans (PGs) and their associated glycosaminoglycans (GAGs), the second major ECM constituent, are the most important non-collagenous components of articular cartilage matrix; they are linear polysaccharides, which are classified into different subgroups on the basis of their subunits (e.g., chondroitin sulfate and keratan sulfate).

GAGs covalently link to a core protein to create proteoglycan aggregates. PGs contribute to the compressive stiffness of cartilage due to the fixed-charge density of negatively charged GAGs [33]. This provides a gel-like hydrophilic environment with a water content of up to 80% [27]. Although the water content of both hyaline and fibrocartilage reduces by advancing age [27], the total amount of PG does not remarkably change during aging [34]. It is, therefore, likely that the composition of the PGs changes with aging [33, 35-37]. Consequently, the compressive stiffness of the cartilage depends not only on GAG content [36, 38], but also on PG composition and its alterations in hydrodynamic size of GAGs, sulfation pattern, and ratios of different GAGs [33, 36, 37]. For instance, age-related reduction of aggregate size and the ratio of chondroitin sulfate to keratan sulfate (which lead to reduced sulfation) in human articular cartilage resulted in a weaker response under compression loading [35]. Although the TMJ condyle has not been investigated, it is likely that a similar alteration in compressive responses of TMJ condylar cartilage can also be seen since a comparable alteration of PGs has been observed for TMJ [34]. In addition, as mentioned above, PGs have negatively charged molecules which attract a high amount of water and produce a swelling pressure in


1

15

cartilage. This swelling pressure is controlled by the collagen network and plays an important role in tissue support under biomechanical loading [15].

Biophysical propertiesIn addition to the above-mentioned structural role of PGs, they also have an important physiological role, i.e., influencing the metabolic activity due to controlling the transport of solute within the cartilage via their highly hydrated nature and negative electrostatic charges [36]. The flow of water through the cartilage and across the articular surface helps to transport and distribute nutrients, in addition to providing lubrication. It is expected that these parameters of TMJ fibrocartilage changes as a consequence of aging. Understanding such alterations might improve the knowledge of the mechanism underlying the pathology associated with aging.

Diffusion and changes with aging

Like other articular cartilages, the fibrocartilaginous TMJ condyle is avascular, which implies that the transport of metabolites within the matrix depends on diffusion [15, 39]. Changes in joints due to aging or to a degenerative disease may potentially be caused by an alteration in diffusion. The diffusion of solutes in cartilage strongly depends on PGs and water content, which dictate the steric hindrance and the porosity of the matrix [40]. Lee et al. have shown that net charge of PGs decreases due to changes in the composition and aggregation size of PGs during aging [35]. Consequently, these changes might result in the reduction of the effective porosity of cartilage matrix. The diffusion in a matrix is also sensitive to the available fluid to flow [41] which has shown a decline during aging as mentioned above [27]. In addition to these parameters, diffusion also changes concomitantly with alterations in tissue integrity, collagen structure, collagen crosslink level, and compressive stiffness of the cartilage superficial layer [32, 40, 42, 43]. However, these studies did not assess the effects of natural aging on diffusion. Although the diffusive properties of cartilage have obtained a lot of attention during the past years in the light of the development of enhanced contrast computed tomography (CT) and MRI [42, 44], little is known about the effect of aging-associated changes on that. Furthermore, to the best of our knowledge, no study has been performed on the effect of aging on the diffusion of TMJ cartilage.

Mechanical properties

Like other diarthrodial joints, the fibrocartilaginous TMJ condyle is loaded and unloaded during movement; the movement of TMJ is not limited to rotation [45]. The rotational and translational movement in TMJ causes it to experience shear, tension, and compression [22]. Articular cartilage is a visco-elastic tissue; its viscous properties are derived from interstitial fluid, which flows away from the loaded region. This flow is governed by permeability. It has been shown that a stiffer cartilage is less permeable.

Chapter 1

16

The elastic properties are derived from the solid matrix of the cartilage [46]. The fibrous structure of the TMJ causes its anisotropic mechanical responses. For instance, TMJ condylar cartilage has been shown 2.4 fold stronger in the anteroposterior direction than in mediolateral direction during tension [22]. This coincides with the dominant anteroposterior collagen fiber orientation in its superficial layer. Topographically different compressive responses have also been reported across the cartilage regions and zones for the TMJ condyle [47, 48], which are thought to be due to zonal and regional variation in its composition and structure. In other words, alterations in GAG and collagen amount or in the thickness of the cartilage and in the orientation of the collagens can lead to different responses to loading.

An age-associated increase in the number of collagen crosslinks has been shown to positively correlate with the stiffness of hyaline cartilage [18, 49]. The higher the amount of crosslinks, the stiffer the matrix. Julkunen et al. have also shown that age-related changes in the orientation of collagen fibrils significantly influenced the mechanical responses in the rabbit knee joint [32]. It has been shown that hyaline cartilage becomes stiffer and more brittle with aging, reducing its capability to handle an overload. Although no data are available on age-related changes in the mechanical response of TMJ condylar cartilage, a similar trend can be expected.

Motivation and overall hypothesisAging is a predisposing factor for the development of OA. It has already been shown that age-related changes that occur in hyaline cartilage made it more susceptible to the onset or development of OA [48]. Such changes have been investigated extensively in hyaline cartilage. There are several studies indicating gradual structural and biochemical changes occurring in the hyaline extracellular matrix (ECM) that affect its biophysical and mechanical performance [27]. However, there is a lack of studies about such effects in the TMJ condylar cartilage. The TMJ is interesting specifically because its structure, mechanical stiffness, and the type of cartilage differ from hyaline cartilage. Understanding the normal age-related changes in this type of cartilage will provide insight into the possible similarities and differences with hyaline cartilage. Consequently, it might reveal similarities and differences in tissue responses under pathological conditions. Finally, it might result in modifications of the clinical assessment of patients suffering from TMJ-OA, not only regarding its diagnosis but also regarding its possible prevention, management, and prognosis. Therefore, the present thesis aimed at understanding the aging of the mandibular condylar cartilage and its consequences for biophysical properties.

The purpose was to test the hypothesis that age-related changes of mandibular condylar cartilage in the temporomandibular joint result in cartilage stiffening. This stiffening is considered to be induced during aging by an increased level


1

17

of collagen crosslinking. Moreover, a higher amount of crosslinks is assumed to affect diffusion of metabolites.

The objective, therefore, is to relate biophysical properties of cartilage (thickness, stiffness, diffusion) to age-related biochemical changes. In order to understand the biological background of changes in biophysical properties of aging, an increased matrix cross-linking was induced by ribose and used as a model that mimics aging.

Thesis outlineThe thickness of cartilage is an important parameter; a parameter that positively correlates with the stiffness of the TMJ condylar cartilage [50]. In Chapter 2, we developed a non-destructive method to measure the thickness of TMJ condylar cartilage. We used a micro-CT method for a non-destructive and fast thickness measurement of intact cartilage. This method was compared with two other methods, i.e., histology and needle penetration. We used this micro-CT method in the following chapters.

To improve our understanding of the biological background of normal aging on the biophysical properties of TMJ condylar cartilage, we introduced in Chapter 3 an aging-like effect of collagen crosslinking. We investigated the correlation between biomechanical properties of TMJ cartilage and the number of collagen crosslinks (measured with pentosidine). The crosslinks were induced by application of ribose in various concentrations.

In Chapter 4, we used the aging model of collagen crosslinking, as tested in chapter 3, in young equine samples. We assessed the diffusion as well as the stiffness of artificially aged samples and their correlation with aging-induced effects.

Chapter 5 deals with the effect of natural aging on biochemical, compositional, and biophysical properties of equine samples of different ages. To investigate the effect of the nature of contrast agent on the diffusion with aging, we also examined the diffusion of a negatively charged and an uncharged contrast agent into the TMJ condylar cartilage.

Chapter 6 provides a general discussion of the thesis and draws the final conclusions and suggests future possibilities for additional research.

Chapter 1

18

References1. He, W., D. Goodkind, and P. Kowal, An Aging World: 2015. 2016, U.S. Census Bureau, U.S. Government Publishing Office: Washington, DC. p. P95/16-1.2. Flatt, T., A New Definition of Aging? Frontiers in Genetics, 2012. 3: p. 148.3. Shane Anderson, A. and R.F. Loeser, Why is osteoarthritis an age-related disease? Best Pract Res Clin Rheumatol, 2010. 24(1): p. 15-26.4. Arden, N. and M.C. Nevitt, Osteoarthritis: epidemiology. Best Pract Res Clin Rheumatol, 2006. 20(1): p. 3-25.5. Lotz, M. and R.F. Loeser, Effects of aging on articular cartilage homeostasis. Bone, 2012. 51(2): p. 241-248.6. Oliveria, S.A., et al., Incidence of symptomatic hand, hip, and knee osteoarthritis among patients in a health maintenance organization. Arthritis Rheum, 1995. 38(8): p. 1134-41.7. Zhang, Y. and J.M. Jordan, Epidemiology of osteoarthritis. Clin Geriatr Med, 2010. 26(3): p. 355-69.8. van Saase, J.L., et al., Epidemiology of osteoarthritis: Zoetermeer survey. Comparison of radiological osteoarthritis in a Dutch population with that in 10 other populations. Annals of the Rheumatic Diseases, 1989. 48(4): p. 271-280.9. Loeser, R.F., et al., Osteoarthritis: a disease of the joint as an organ. Arthritis Rheum, 2012. 64(6): p. 1697-707.10. Palazzo, C., et al., Risk factors and burden of osteoarthritis. Annals of Physical and Rehabilitation Medicine, 2016. 59(3): p. 134-138.11. Vos, T., et al., Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: A systematic analysis for the Global Burden of Disease Study 2010. The Lancet, 2012. 380(9859): p. 2163-2196.12. Brehnan, K., et al., Direct measurement of loads at the temporomandibular joint in Macaca arctoides. J Dent Res, 1981. 60(10): p. 1820-4.13. Mejersjo, C., Therapeutic and prognostic considerations in TMJ osteoarthrosis: a literature review and a long-term study in 11 subjects. Cranio, 1987. 5(1): p. 69-78.14. Ferrazzo, K.L., L.B. Osório, and V.A. Ferrazzo, CT Images of a Severe TMJ Osteoarthritis and Differential Diagnosis with Other Joint Disorders. Case Reports in Dentistry, 2013. 2013: p. 5.15. Kuroda, S., et al., Biomechanical and biochemical characteristics of the mandibular condylar cartilage. Osteoarthritis Cartilage, 2009. 17(11): p. 1408-1415.16. Grande, D.A., et al., Articular Cartilage Repair: Where We Have Been, Where We Are Now, and Where We Are Headed. Cartilage, 2013. 4(4): p. 281-285.17. Aigner, T., et al., Aging theories of primary osteoarthritis: from epidemiology to molecular biology. Rejuvenation Res, 2004. 7(2): p. 134-45.18. Verzijl, N., et al., Crosslinking by advanced glycation end products increases the stiffness of the collagen network in human articular cartilage: a possible mechanism through which age is a risk factor for osteoarthritis. Arthritis nad Rheumtism, 2002. 46(1): p. 114-23.19. Copray, J.C., H.W. Jansen, and H.S. Duterloo, Growth and growth pressure of mandibular condylar and some primary cartilages of the rat in vitro. American journal of orthodontics and dentofacial orthopedics, 1986. 90(1): p. 19-28.20. Mizoguchi, I., N. Toriya, and Y. Nakao, Growth of the mandible and biological characteristics of the mandibular condylar cartilage. Japanese Dental Science Review, 2013. 49(4): p. 139-150.21. Athanasiou, K.A., et al., Tissue engineering of temporomandibular joint cartilage. Synthesis Lectures on Tissue Engineering. Vol. 1. 2009: Morgan & Claypool Publishers. 1-122.


1

19

22. Singh, M. and M.S. Detamore, Tensile Properties of the Mandibular Condylar Cartilage. Journal of Biomechanical Engineering, 2008. 130(1): p. 011009-011009.23. Delatte, M., et al., Primary and secondary cartilages of the neonatal rat: the femoral head and the mandibular condyle. Eur J Oral Sci, 2004. 112(2): p. 156-62.24. Kulmala, K.A.M., et al., Contrast-Enhanced Micro–Computed Tomography in Evaluation of Spontaneous Repair of Equine Cartilage. Cartilage, 2012. 3(3): p. 235-244.25. Sophia Fox, A.J., A. Bedi, and S.A. Rodeo, The Basic Science of Articular Cartilage: Structure, Composition, and Function. Sports Health, 2009. 1(6): p. 461-468.26. Wang, L. and M.S. Detamore, Tissue engineering the mandibular condyle. Tissue Eng, 2007. 13(8): p. 1955-71.27. Williams, G.M., S.M. Klisch, and R.L. Sah, Bioengineering cartilage growth, maturation, and form. Pediatr Res, 2008. 63(5): p. 527-534.28. Snedeker, J.G. and A. Gautieri, The role of collagen crosslinks in ageing and diabetes - the good, the bad, and the ugly. Muscles Ligaments Tendons J, 2014. 4(3): p. 303-308.29. Verzijl, N., et al., Effect of collagen turnover on the accumulation of advanced glycation end products. J Biol Chem, 2000. 275(50): p. 39027-31.30. Bank, R.A., et al., Ageing and zonal variation in post-translational modification of collagen in normal human articular cartilage. The age-related increase in non-enzymatic glycation affects biomechanical properties of cartilage. Biochem J, 1998. 330(1): p. 345-351.31. Moshtagh, P.R., et al., Effects of non-enzymatic glycation on the micro- and nano-mechanics of articular cartilage. Journal of the Mechanical Behavior of Biomedical Materials, 2017.32. Julkunen, P., et al., Biomechanical, biochemical and structural correlations in immature and mature rabbit articular cartilage. Osteoarthritis Cartilage, 2009. 17(12): p. 1628-38.33. Bayliss, M.T. and S.Y. Ali, Age-related changes in the composition and structure of human articular-cartilage proteoglycans. Biochem J, 1978. 176(3): p. 683-693.34. Platt, D., J.L. Bird, and M.T. Bayliss, Ageing of equine articular cartilage: structure and composition of aggrecan and decorin. Equine Vet J, 1998. 30(1): p. 43-52.35. Lee, H.-Y., et al., Age-related nanostructural and nanomechanical changes of individual human cartilage aggrecan monomers and their glycosaminoglycan side chains. J Struct Biol, 2013. 181(3): p. 264-273.36. Dudhia, J., Aggrecan, aging and assembly in articular cartilage. Cell Mol Life Sci, 2005. 62(19-20): p. 2241-56.37. Bayliss, M.T., et al., Sulfation of Chondroitin Sulfate in Human Articular Cartilage: The effect of age, topographical position, and zone of cartilage on tissue composition. J Biol Chem, 1999. 274(22): p. 15892-15900.38. Singh, M. and M.S. Detamore, Biomechanical properties of the mandibular condylar cartilage and their relevance to the TMJ disc. J Biomech, 2009. 42(4): p. 405-17.39. O’Hara, B.P., J.P. Urban, and A. Maroudas, Influence of cyclic loading on the nutrition of articular cartilage. Ann Rheum Dis, 1990. 49(7): p. 536-9.40. Kokkonen, H.T., et al., Detection of mechanical injury of articular cartilage using contrast enhanced computed tomography. Osteoarthritis Cartilage, 2011. 19(3): p. 295-301.41. Maroudas, A., et al., The permeability of articular cartilage. J Bone Joint Surg Br, 1968. 50-B(1): p. 166-177.42. Kokkonen, H.T., et al., Computed tomography detects changes in contrast agent diffusion after collagen cross-linking typical to natural aging of articular cartilage. Osteoarthritis Cartilage, 2011. 19(10): p. 1190-1198.

Chapter 2

20

43. Kulmala, K.A., et al., Diffusion of ionic and non-ionic contrast agents in articular cartilage with increased cross-linking--contribution of steric and electrostatic effects. Med Eng Phys, 2013. 35(10): p. 1415-20.44. Xie, L., et al., Nondestructive assessment of sGAG content and distribution in normal and degraded rat articular cartilage via EPIC-microCT. Osteoarthritis Cartilage, 2010. 18(1): p. 65-72.45. Koolstra, J.H., Dynamics of the Human Masticatory System. Critical Reviews in Oral Biology & Medicine, 2002. 13(4): p. 366-376.46. Barker, M.K. and B.B. Seedhom, The relationship of the compressive modulus of articular cartilage with its deformation response to cyclic loading: does cartilage optimize its modulus so as to minimize the strains arising in it due to the prevalent loading regime? Rheumatology, 2001. 40(3): p. 274-284.47. Lamela, M.J., et al., Dynamic compressive properties of articular cartilages in the porcine temporomandibular joint. Journal of the Mechanical Behavior of Biomedical Materials, 2013. 23(0): p. 62-70.48. Tanaka, E., et al., Dynamic compressive properties of the mandibular condylar cartilage. J Dent Res, 2006. 85(6): p. 571-5.49. Moriyama, H., et al., Effects of aging and exercise training on the histological and mechanical properties of articular structures in knee joints of male rat. Biogerontology, 2012. 13(4): p. 369-81.50. Singh, M. and M.S. Detamore, Stress relaxation behavior of mandibular condylar cartilage under high-strain compression. Journal of biomechanical engineering, 2009. 131(6): p. 0610081.

Chapter 2

Ex vivo thickness measurement of cartilage covering the temporomandibular joint

Fereshteh Mirahmadi, Jan Harm Koolstra, Frank Lobbezoo, G. Harry van Lenthe, Vincent Everts

Published in Journal of Biomechanics, 52, 165-168,2017

Chapter 2

24

Abstract Articular cartilage covers the temporomandibular joint (TMJ) and provides smooth and nearly frictionless articulation while distributing mechanical loads to the subchondral bone. The thickness of the cartilage is considered to be an indicator of the stage of development, maturation, aging, loading history, and disease. The aim of our study was to develop a method for ex vivo assessment of the thickness of the cartilage that covers the TMJ and to compare that with two other existing methods. Eight porcine TMJ condyles were used to measure cartilage thickness. Three different methods were employed: needle penetration, micro-computed tomography (micro-CT), and histology; the latter was considered the gold standard. Histology and micro-CT scanning results showed no significant differences between thicknesses throughout the condyle. Needle penetration produced significantly higher values than histology, in the lateral and anterior regions. All three methods showed the anterior region to be thinner than the other regions. We concluded that overestimated thickness by the needle penetration is caused by the penetration of the needle through the first layer of subchondral bone, in which mineralization is less than in deeper layers. Micro-CT scanning method was found to be a valid method to quantify the thickness of the cartilage, and has the advantage of being non-destructive.

Thickness measurement of TMJ cartilage

25

2

IntroductionArticular cartilage covers the articulating surfaces in joints. It provides smooth and nearly frictionless articulation while distributing mechanical loads to the subchondral bone [1]. The thickness of the cartilage layer is influenced by development, maturation, aging, and health status of the joints. Furthermore, it depends on gender, joint loading, and disease [2-6]. It has been shown that either moderate exercise or immobilization of the joint results in remarkable changes in the thickness of the cartilage layers [5]. Once the mutual position of the joint components alters, the thickness of cartilage layers could also significantly change [6]. Furthermore, when the thickness distribution changes, the mechanical loading also differs [7]. It is clinically important to track the cartilage thickness and its alterations, since this parameter might indicate disease progression [4]. Moreover, thickness is an essential parameter in the characterization of biomechanical properties of cartilage.

Various methods have been used in vivo and ex vivo to measure cartilage thickness in joints, including magnetic resonance imaging (MRI), computed tomography (CT), ultrasound, needle penetration, and histology. Each of these methods has advantages and disadvantages. For instance, MRI is a non-invasive method although it is relatively expensive and not always available [3]. In ultrasound, cartilage thickness is determined on the basis of the assumption that cartilage has a constant sound velocity [1, 8, 9] which could results in up to 33.6% error in measured thickness [10]. For CT, a contrast agent is usually applied that penetrates the cartilage, since the non-mineralized cartilage is translucent to X-rays [1, 11]. Needle penetration is a destructive method which relies on detecting force alterations when a probe equipped with a needle penetrates through the cartilage surface and reaches the bone [12-14]. Histology makes visualization possible of the structure of the cartilage and the underlying bone [6, 15]. The latter method is usually used to validate the results of the other methods; yet, it has the disadvantage of being destructive.

In the present study, needle penetration, CT scanning, and histology were used for measuring the cartilage thickness of Temporomandibular Joint (TMJ). TMJ has a unique cartilaginous structure in comparison to other articular joints. TMJ condylar cartilage has both fibrocartilaginous and hyaline-like characteristics and is, therefore, structurally different from most hyaline articular cartilages [16, 17]. The reliability of the various methods for the measurement of its cartilage thickness has not been established. Therefore, the aim of our study was to develop a method for ex vivo assessment of the thickness of TMJ condylar cartilage and to compare that with two other existing methods.

Chapter 2

26

Materials and methods

Sample preparation

The heads from 8 young pigs were obtained from a local slaughterhouse. The left and right TMJ condyles were dissected within 12 h after sacrifice and kept at -20˚C until the actual experiment. Before each test, the condyles were thawed being immersed in phosphate buffer saline (PBS, Gibco, USA) at room temperature for 1 h. Five regions were tested in this study: central, lateral, medial, posterior, and anterior.

Thickness measurement

After thawing, all specimens were subjected to needle penetration according to the protocol by Swann and Seedhom [14]. Afterwards, a micro-CT 40 (Scanco Medical AG, Bruttisellen, Switzerland) was used to obtain and analyze 3D reconstructions of segmented volumes of cartilage. Before scanning, the entire cartilage surface was covered with a thin layer of barium sulfate (BaSO4, Sigma-Aldrich, MW: 233.39). The BaSO4 suspension was prepared in agarose gel to prevent cartilage dehydration during scanning. These condyles were then imaged using isotropic 36-μm voxel size

Figure 1: The cartilage thickness steps for measurement method with micro-CT and histology. Distribution of the cartilage thickness after scanning and 3D reconstruction is presented as a pseudocolor-scaled image i.e. from blue to red shows increasing cartilage thickness. Histological staining shows boundaries between cartilage and bone i.e. red/orange color shows the presence of GAG and blue color shows collagen-rich parts which include superficial layer of cartilage on top of the stained section and subchondral bone on the bottom.


27

2

at 70 kVp and 114 μA. To make a 3D reconstruction of the samples, the specimen was segmented by choosing the upper and lower threshold. The thresholds were defined based on visual inspection. The average cartilage thickness was measured using micro-CT software for each defined region, i.e., central, lateral, medial, posterior, and anterior (Figure 1). Following micro-CT scanning, the different regions of interest were cut and cryosectioned at 10 μm along the sagittal plane. The sections were stained with SafraninO/Fast green and digital images were taken for measuring the thickness by using ImageJ software (Figure 1). For more details about materials and methods, see Appendix.

Statistical analysis

Analyses of Variance (one-way ANOVA) were used to compare region-wise thickness in each individual method using IBM SPSS Statistics23. The thickness value of each region obtained by micro-CT and needle penetration was compared separately with the corresponding thickness as assessed with histologic sections with t-test. Linear regression with a Pearson correlation analysis was also performed to look at the correlations between methods. The statistical significance level was set at 0.05.

ResultsThe results from three different methods were shown in Table 1. In all regions, the thickness measured with needle penetration was higher than the thickness assessed by histology. The differences were significant in the lateral (p = 0.019) and anterior (p = 0.031) regions. The average range of thickness measured in the central, lateral, medial, and posterior regions were between 2.0 to 2.8 mm. The highest value was measured with needle penetration in the lateral region, i.e., 2.8 mm (Table 1).

The thicknesses measured with micro-CT proved to be similar to those measured in the histological sections. No significant differences were noted (Figure 2). The results also showed that the anterior region was significantly thinner than all other regions. This was apparent using either method. The values were 0.63 with histology, 0.68 with CT-

Table 1: The average thickness values in mm ± standard deviations for three different methods.

*, ** p<0.05; ** shows difference from all others. Needle penetration method measured generally higher thicknesses in all regions.

Chapter 2

28

scanning, and 0.73 mm, with needle penetration. Analyses of correlation have shown that there was a good agreement between the thickness results from micro-CT and histology (R=0.808, p<0.001); this agreement was confirmed when differences between the thickness results from micro-CT and histology was not significant. There was also a significant positive correlation between the thickness obtained with needle penetration and histology (R=0.743, p<0.001).

DiscussionIn this study, we used three different methods to measure the thickness of the TMJ condylar cartilage, i.e. micro-CT, needle penetration, and histology. The results showed that the thickness of the cartilage varied with location along the porcine TMJ condyle. The variation of the thicknesses between different regions was generally the same, regardless of the measurement method, e.g. anterior was the thinnest region with all three methods. The observed differences in thickness among regions of the condyle are in line with data presented by Lu et al. [13]. They used the needle penetration method and found that the thinnest region in porcine TMJ condyle was the anterior region. However, Tanaka et al. [18, 19] reported different thickness and variation in defined regions e.g. the thickness measured from the anterior region in their study was larger than the thickness of the same defined region in our study (1.09 vs. 0.73 mm). Such a variation in thickness might be related to differences in animal breed, age, and weight. The cartilage thickness of the mandibular condyles as well as their dimensions in rats fed a soft diet were found significantly lower than the ones fed a hard diet [20]. The

Figure 2: Cartilage thicknesses assessed by needle penetration, micro-CT, and histology. Significant differences were seen in lateral and anterior regions in cartilage thicknesses obtained with needle penetration and histology shown by *, p<0.05; the cartilage thicknesses measured with micro-CT and histology were not significantly different. Anterior region was significantly thinner than all other regions in all three methods (shown with **, p<0.05).


29

2

variation in thickness and size has also been reported for rabbit articular cartilage during lifetime, especially before maturation [21].

Cartilage thickness is a crucial factor to determine the biomechanical behavior of cartilage [18, 22, 23]. Hence, it is important to have a reliable and feasible measurement method for cartilage thickness. Needle penetration has been used widely to measure the thickness of articular cartilage, especially in biomechanical assessments [18, 19, 23, 24]. This method is relatively simple and fast. However, the results depend on the speed of the needle. In preliminary tests we observed a plateau in the force-displacement curve before reaching a sharp increase of the force when a low speed was applied. We, therefore, concluded that a lower speed might not produce an appropriate curve to determine bone boundaries. In addition, the stage of maturation can also have an effect on the results of this method. It has been shown that during growth and maturation, several changes occur in the stiffness of cartilage and subchondral bone. Due to the maturation, the thin and immature subchondral bone is replaced by the thicker compact bone; therefore it is harder for a needle to penetrate [21, 25]. We found that in all regions the thicknesses measured with needle penetration were higher than the ones measured with micro-CT and histology. We hypothesized that in the young samples we used this could be due to a lower level of mineralization in bony layers near the cartilage than the ones in deeper layers. It was observed indeed in our micro-CT images that the amount of mineral in the layer close to the cartilage is less than that in the deeper zone of the subchondral bone.

Contrast-enhanced CT, which is used to visualize cartilage thickness as well as glycosaminoglycan (GAG) distribution, requires immersion of the specimens in a contrast agent solution. The diffusion process can take a long time to reach equilibrium, depending on the thickness of the cartilage [1, 26]. In contrast, the protocol for micro-CT scanning as used in the present study provides the entire sample thickness without a penetrating step, and is fairly quick and valid. Moreover, the obtained results are similar to the results obtained with the histological sections. Their significant correlation also demonstrated the agreement between the results of these two methods (R=0.808, p<0.001). Histology was considered the gold standard in quantifying cartilage thickness to validate other methods. Although tissue undergoes some shrinkage during fixation and histological processing [27-29], we made use of cryosections. Shrinkage of the tissue does hardly occur using this approach. Gamble et al. have shown a shrinkage of about 8% after histological preparation of cryosection slices from artery [30]. However, paraffin embedding has shown to create 15% shrinkage caused mostly by dehydration steps [29].

As shown in figure 1, the interface of cartilage and bone is not smooth and regular. This irregularity could lead to error in thickness measurement specifically if the thickness

Chapter 2

30

is measured with the needle penetration method [31]. For histology and micro-CT measurements this irregularity was taken into account as the average of several points was calculated for each region.

In conclusion, thickness measurements obtained with histology or micro-CT scanning proved to yield similar values for all parts of the condyle. However, significant differences were apparent between needle penetration and histology. Needle penetration appeared to overestimate the thickness. This is likely due to the penetration of the needle through the first layer of subchondral bone, in which mineralization is less than in deeper layers. Micro-CT scanning method was found to be a valid and feasible method to quantify the thickness of the cartilage, and has the advantage of being non-destructive.

AcknowledgmentThis research was funded by the European Commission through MOVE-AGE, an Erasmus Mundus Joint Doctorate programme (2011-0015). The authors would like to thank Albert van der Veen for technical help.


31

2

References 1. Aula, A.S., J.S. Jurvelin, and J. Toyras, Simultaneous computed tomography of articular cartilage and subchondral bone. Osteoarthritis Cartilage, 2009. 17(12): p. 1583-8.2. Carter, D.R., et al., The mechanobiology of articular cartilage development and degeneration. Clin Orthop Relat Res, 2004(427 Suppl): p. S69-77.3. Draper, C.E., et al., Is cartilage thickness different in young subjects with and without patellofemoral pain? Osteoarthritis Cartilage, 2006. 14(9): p. 931-7.4. Jones, G., et al., Sex and site differences in cartilage development: a possible explanation for variations in knee osteoarthritis in later life. Arthritis Rheum, 2000. 43(11): p. 2543-9.5. Kiviranta, I., et al., Moderate running exercise augments glycosaminoglycans and thickness of articular cartilage in the knee joint of young beagle dogs. J Orthop Res, 1988. 6(2): p. 188-95.6. Proff, P., et al., Histological and histomorphometric investigation of the condylar cartilage of juvenile pigs after anterior mandibular displacement. Annals of Anatomy - Anatomischer Anzeiger, 2007. 189(3): p. 269-275.7. Li, G., O. Lopez, and H. Rubash, Variability of a three-dimensional finite element model constructed using magnetic resonance images of a knee for joint contact stress analysis. Journal of Biomechanical Engineering 2001. 123(4): p. 341-6.8. Toyras, J., et al., Characterization of enzymatically induced degradation of articular cartilage using high frequency ultrasound. Phys Med Biol, 1999. 44(11): p. 2723-33.9. Toyras, J., et al., Estimation of the Young’s modulus of articular cartilage using an arthroscopic indentation instrument and ultrasonic measurement of tissue thickness. J Biomech, 2001. 34(2): p. 251-6.10. Yao, J.Q. and B.B. Seedhom, Ultrasonic measurement of the thickness of human articular cartilage in situ. Rheumatology, 1999. 38(12): p. 1269-1271.11. Kerckhofs, G., et al., Contrast-enhanced nanofocus computed tomography images the cartilage subtissue architecture in three dimensions. Eur Cell Mater, 2013. 25: p. 179-89.12. Kim, K.W., et al., Biomechanical tissue characterization of the superior joint space of the porcine temporomandibular joint. Annals of Biomedical Engineering, 2003. 31(8): p. 924-30.13. Lu, X.L., V.C. Mow, and X.E. Guo, Proteoglycans and mechanical behavior of condylar cartilage. Journal of Dental Research, 2009. 88(3): p. 244-8.14. Swann, A.C. and B.B. Seedhom, Improved techniques for measuring the indentation and thickness of articular cartilage. Proc Inst Mech Eng H, 1989. 203(3): p. 143-50.15. Li, Q., et al., Psychological stress alters microstructure of the mandibular condyle in rats. Physiology & Behavior, 2013. 110–111(0): p. 129-39.16. Wang, L., M. Lazebnik, and M.S. Detamore, Hyaline cartilage cells outperform mandibular condylar cartilage cells in a TMJ fibrocartilage tissue engineering application. Osteoarthritis and Cartilage, 2009. 17(3): p. 346-53.17. Herring, S.W., TMJ anatomy and animal models. Journal of musculoskeletal & neuronal interactions, 2003. 3(4): p. 391-394.18. Tanaka, E., et al., Stress relaxation behaviors of articular cartilages in porcine temporomandibular joint. Journal of Biomechanics, 2014. 47(7): p. 1582-7.19. Lamela, M.J., et al., Dynamic compressive properties of articular cartilages in the porcine temporomandibular joint. Journal of the Mechanical Behavior of Biomedical Materials, 2013. 23(0): p. 62-70.20. Bouvier, M. and W.L. Hylander, The effect of dietary consistency on gross and histologic morphology in the craniofacial region of young rats. American Journal of Anatomy, 1984. 170(1):

Chapter 2

32

p. 117-26.21. Julkunen, P., et al., Biomechanical, biochemical and structural correlations in immature and mature rabbit articular cartilage. Osteoarthritis Cartilage, 2009. 17(12): p. 1628-38.22. Tanaka, E., et al., The Effect of Removal of the Disc on the Friction in the Temporomandibular Joint. Journal of Oral and Maxillofacial Surgery, 2002. 64(8): p. 1221-4.23. Tanaka, E., et al., Dynamic compressive properties of the mandibular condylar cartilage. J Dent Res, 2006. 85(6): p. 571-5.24. Shepherd, D. and B. Seedhom, Thickness of human articular cartilage in joints of the lower limb. Annals of the Rheumatic Diseases, 1999. 58(1): p. 27-34.25. Lubsen, C.C., et al., Histomorphometry of age and sex changes in mandibular condyles of young human adults. Arch Oral Biol, 1987. 32(10): p. 729-33.26. Renders, G.A.P., et al., Contrast-enhanced microCT (EPIC-µCT) ex vivo applied to the mouse and human jaw joint. Dentomaxillofac Radiol, 2014. 43(2): p. 20130098.27. Boonstra, H., et al., Cervical tissue shrinkage by formaldehyde fixation, paraffin wax embedding, section cutting and mounting. Virchows Arch A Pathol Anat Histopathol, 1983. 402(2): p. 195-201.28. Gardella, D., et al., Differential tissue shrinkage and compression in the z-axis: implications for optical disector counting in vibratome-, plastic- and cryosections. J Neurosci Methods, 2003. 124(1): p. 45-59.29. Gardner, E.S., W.T. Sumner, and J.L. Cook, Predictable Tissue Shrinkage During Frozen Section Histopathologic Processing for Mohs Micrographic Surgery. Dermatologic Surgery, 2001. 27(9): p. 813-818.30. Gamble, G., et al., B-mode ultrasound images of the carotid artery wall: correlation of ultrasound with histological measurements. Atherosclerosis, 1993. 102(2): p. 163-73.31. Jurvelin, J.S., et al., Comparison of optical, needle probe and ultrasonic techniques for the measurement of articular cartilage thickness. Journal of Biomechanics, 1995. 28(2): p. 231-235.


33

2

Appendix A. Supporting information

Sample preparation

The heads from 18 pigs (weighting about 100kg and 6-7 months old) were obtained from a local slaughterhouse. The left and right TMJ condyles, including the articular discs, were dissected within 12 h after sacrifice. After the condyles had been carefully detached from the articular discs, they were immediately wrapped in gauze soaked in Phosphate Buffer Saline (PBS) containing a protease-inhibitor (PI) cocktail (2 mM Ethylenediaminetetraacetic acid (EDTA); 5 mM benzamidine; 10 mM N-ethyl-maleimide; 1 mM phenylmethylsulfonyl fluoride) and stored at -20˚C before the start of the actual experiments. Before each test, the condyles were thawed being immersed in PBS at room temperature for 1 h. Five regions were tested in this study: central, lateral, medial, posterior, and anterior.

Needle penetration

After thawing, all specimens were subjected to needle penetration. A testing apparatus (Instron, Norwood, MA) equipped with a sharp needle (0.6 mm diameter) was utilized for this method. A custom-made container was used to align each region of interest perpendicular to the needle by visual inspection [1]. Each region was placed at a distance of 0.5-1 mm from the needle tip. This moved downwards towards the cartilage at a constant rate of 60 mm/s [2]. Subsequently, the force-displacement data were recorded (Figure A1). When the force reached 30 N or the displacement reached 6 mm,

Figure A1: The cartilage thickness as measured by the needle penetration. Force-displacement curve produced by needle penetration, green and red lines show the slopes of two regions with distinctive hardness. Sharp increase in the slope shows that needle has reached the subchondral bone.

Chapter 2

34

the test was automatically terminated. Per site, one test was performed. The condyles were kept in PBS for the entire duration of the test.

Cartilage thickness was determined using the force-displacement curve. Sharp changes in the force slope were considered as the point in which needle reached the bone after going through the cartilage. The surface of the condyle was determined from the point a continuous increase in the recorded force was measured. Two areas of different slopes were selected manually by visual inspection. Linear regression was performed between force and displacement data. The thickness coinciding with the point where the two discriminated slopes crossed was used as an approximation of cartilage thickness [1].

Micro-CT scanning

A micro-CT 40 (Scanco Medical AG, Bruttisellen, Switzerland) was used to obtain and analyze 3D reconstructions of segmented volumes of cartilage. The condyles were placed in a sample holder, such that the sagittal plane was perpendicular to the longitudinal axis of the sample holder. Before scanning, a thin layer of Barium sulfate (BaSO4, Sigma-Aldrich, MW: 233.39) with a concentration of 20% w/v in 2% agarose was used to cover the entire cartilage surface. This was used as a contrast-enhancement medium to visualize the border of the cartilage. The agarose gel around the sample prevented drying of the sample during scan. The BaSO4 covered samples were imaged using 36-μm voxel size at 70 kVp and 114 μA. The whole scanning time was approximately 90 min depending on the actual size of the condyle. A rectangular volume of Interest (VOI) including the relevant cartilage surface and underlying bone was manually chosen. To make a 3D reconstruction of the samples, the specimen was segmented by choosing the upper and lower threshold. The thresholds were defined based on visual inspection. The locations of needle penetration spots were used to define measuring regions in 3D reconstruction. Then, the relevant scanned slices for each defined region, i.e., central, lateral, medial, posterior, and anterior were selected to measure the thickness of the cartilage by mans of micro-CT software. The perpendicular distance between bone and BaSO4 layer covering the cartilage was measured as cartilage thickness. At least sixteen measurements in total, four measurements from four selected slices per region, were made for each specimen and average cartilage thickness was reported.

Histology

Following micro-CT scanning, each condyle was divided into a medial, posterior, anterior, lateral and central region using a scalpel. The different parts were embedded in Tissue-Tek (Sakura Finetek, Leiden, Netherlands), frozen in -80˚C, and cryosectioned at 10 μm along the sagittal plane. The sections were mounted on saline-coated glass and stained with SafraninO/Fast green. Three sections per region were used to analyze cartilage thickness. Digital images were captured by light microscopy and thickness for


35

2

each region was measured with ImageJ software and defined as an average of up to 15 measurements.

References:1. Swann, A.C. and B.B. Seedhom, Improved techniques for measuring the indentation and thickness of articular cartilage. Proc Inst Mech Eng H, 1989. 203(3): p. 143-50.2. Shepherd, D. and B. Seedhom, Thickness of human articular cartilage in joints of the lower limb. Annals of the Rheumatic Diseases, 1999. 58(1): p. 27-34.

Chapter 3Mechanical stiffness of TMJ condylar cartilage

increases after artificial aging by ribose

Fereshteh Mirahmadi, Jan Harm Koolstra, Frank Lobbezoo, G. Harry van Lenthe, Samaneh Ghazanfari , Jessica Snabel, Reinout Stoop, Vincent

Everts

Published in Archives of Oral Biology, 87, 102-109, 2018.

Chapter 3

38

AbstractAging is accompanied by a series of changes in mature tissues that influence their properties and functions. Collagen, as one of the main extracellular components of cartilage, becomes highly crosslinked during aging. In this study, the aim was to examine whether a correlation exists between collagen crosslinking induced by artificial aging and mechanical properties of the temporomandibular joint (TMJ) condyle. To evaluate this hypothesis, collagen crosslinks were induced using ribose incubation. Porcine TMJ condyles were incubated for 7 days with different concentrations of ribose. The compressive modulus and stiffness ratio (incubated versus control) were determined after loading. Glycosaminoglycan and collagen content, and the number of crosslinks were analyzed. Tissue structure was visualized by microscopy using different staining methods. Concomitant with an increasing concentration of ribose, an increase of collagen crosslinks was found. The number of crosslinks increased almost 50 fold after incubation with the highest concentration of ribose. Simultaneously, the stiffness ratio of the samples showed a significant increase after incubation with the ribose. Pearson correlation analyses showed a significant positive correlation between the overall stiffness ratio and the crosslink level; the higher the number of crosslinks the higher the stiffness. The present model, in which ribose was used to mimic certain aspects of age-related changes, can be employed as an in vitro model to study age-related mechanical changes in the TMJ condyle.

Artificial aging effect on stiffness of TMJ cartilage

39

3

IntroductionCollagen is the main component of the extracellular matrix (ECM) of connective tissues like bone and cartilage. Its molecules have helical structures which pack together to form collagen fibrils in fiber-type collagens [1-3]. These fibrils progressively rearrange and crosslink at the inter-fibrillar level to provide structural integrity needed by a tissue for proper remodeling and load-bearing capacity [4]. Every change in molecular or structural organization of collagen caused by development, pathological conditions, or aging, has an impact on collagen’s contributions to tissue function [2, 4-6].

Aging is accompanied by a series of cellular, molecular, and structural changes in the mature tissues that ultimately influence their properties and functions [2]. Non-reversible collagen crosslinking is a prominent change in the ECM during aging. By altering the mechanical properties of collagen, it could result in an age-related degenerative effect on tissue maintenance [7-10]. Such age-related changes are thought to play a key role in the etiology of osteoarthritis (OA) which is characterized by progressive destruction of cartilage [11, 12]. Mechanical loading is considered a dominant factor for the onset or progression of cartilage damage. It has been shown that the age-related increase of collagen crosslinking increased tissue stiffness. This was considered to make the tissue more susceptible to mechanically-induced damage [13].

Collagen crosslinking often occurs due to enzymatic and/or non-enzymatic reactions. Enzymatic crosslinking mainly occurs during development and maturation. Non-enzymatic crosslinking reactions can take place due to aging-associated modifications and certain diseases [4]. These reactions, called glycation, happen in the presence of a reducing sugar and lead to the production of advanced glycation end-products (AGEs) in the tissue [2, 9, 14]. When AGEs are produced, they remain in the tissue until it is renewed. Cartilaginous collagen found in the joints has an extremely slow turnover rate and remains in the tissue after maturation without renewing [15]. As a result, long-term accumulation of AGEs in the joints happens at older age [11, 14, 16]. Such an accumulation of AGEs would also happen in young adults due to the presence of extra sugar in the body in diabetes [5, 17, 18]. Resembling the same condition, the excessive exposure of tissue to sugar has been used in vivo and ex vivo, as a method for the induction of artificial aging [16, 19-21]. Pentosidine, a crosslink between arginine and lysine residues in collagen molecules, is a well-characterized and reliable measure of AGEs [11, 22].

The effect of this artificial aging has been studied specifically on mechanical properties of ECM in different types of tissues including bone [2, 23], hyaline cartilage [11, 20, 24], and tendon [25, 26]. It has also been utilized to study aging effects as the main risk factor for OA [20, 21, 27]. There is, however, no study in the literature on the effect of such a crosslinking induction on cartilage of the temporomandibular joint (TMJ).

Chapter 3

40

In the present study, the aim was to examine whether a correlation exists between collagen crosslinking induced by artificial aging and mechanical properties of the TMJ condyle. Evaluation of TMJ cartilage is of particular interest because the TMJ condyle is covered with fibrocartilage rather than hyaline cartilage which cover other articular joints [28, 29]. It has a different cartilaginous structure and composition as compared to articular cartilage. Interestingly, in the regeneration of damaged hyaline cartilage, it is replaced by fibrocartilage containing collagen type I. This is generally considered a material of inferior quality at these locations, while in the TMJ it performs well under various loading conditions. The TMJ condyle of the porcine was selected on the basis of its similarities to the human jaw joint, especially in its mechanical characterization [30].


Sample preparation

The heads of young pigs (aged 6-8 months old, gender not known) were collected from a local slaughterhouse. The left and right TMJ condyles were dissected within 12 h after sacrifice, and stored at -20 ̊ C. The dissected condyles were inspected visually to exclude the ones with gross abnormality, such as vascularization and fibrillation of the surface. In total, 24 condyles were used for this study.

Crosslinking induction

The condyles were divided into four groups: three groups for incubation with different concentrations of ribose and one control group. For each incubation group, five condyles were used (all from the right side of the mandibles); for the control group, nine condyles (all from the left side of the mandibles). For crosslinking induction, intact condyles were incubated in phosphate buffered saline (PBS) containing 125, 250, or 500 mM ribose, a protease-inhibitor (PI) cocktail (2 mM Ethylenediaminetetraacetic acid; 5 mM benzamidine; 10 mM N-ethyl-maleimide; 1 mM phenylmethylsulfonyl fluoride [31]), and sodium azide 0.02% for 7days at 37˚C. These groups were labeled as R125, R250, and R500, respectively. The control samples were incubated in the same condition, but without ribose. PI was added to minimize tissue degradation and sodium azide to prevent bacterial growth during the incubation time. For the next steps of the study, five regions per condyle were examined, i.e., central, medial, lateral, posterior, and anterior.

Mechanical characterization

For the mechanical loading, a compression loading test was performed by using a custom-made instrument equipped with a 25 N load-cell [32]. A rigid cylindrical indenter with a diameter of 4 mm was used for loading. This apparatus produces cyclic displacements of an indenter and simultaneously measures the compressive reaction force. The average thickness of each region was measured prior to loading using micro-CT scanning as described in a previous study [33]. To locate each region of interest


41

3

perpendicular to the cylindrical indenter, the condyles were fixed in a container with adjustable tilting capability. The container was filled with PBS containing PI during the test. A tare load of 0.2 N was introduced to each region to maintain proper loading plate contact, followed by 5 min of relaxation time. Afterwards, 60 cycles of 1% strain at a frequency of 1 Hz were applied as preconditioning, followed by 5 min relaxation time. Thereafter, the condyle was cyclically loaded at a strain level of 5% for 20 cycles at 1 Hz. It has been shown that 10 cycles of loading would be enough for TMJ disc to reach a steady state [34]. Force-displacement curves were used to calculate stress-strain values using the initial thickness of each region and the cross-sectional area of the indenter. Instantaneous modulus (EIns) was calculated from the peak stress of the first cycle of loading. Steady state modulus (ESt) was similarly calculated from the average peak stresses of the last five cycles. Representative stress-strain curves from different groups are shown in Figure 1.

Biochemical analyses

To examine the effect of crosslinking induction on biochemical properties, a plug of 4 mm was punched out from each mechanically-loaded region of the condyles. All plugs were weighed and lyophilized overnight; the dry weight (DW) was measured afterwards to calculate water content.

To measure the amount of GAG, about 2 mg of the dried samples were incubated in papain digestion buffer at 60 ˚C overnight and used for colorimetric quantification after reaction with dimethyl methylene blue (DMMB) [35]. The GAG values were normalized

Figure 1: Representative stress curves of control and treated groups from the posterior region.

Chapter 3

42

to the tissue dry weight.

One portion of the digested sample was hydrolyzed with 6M HCl at 95 ˚C for 20 h, dried overnight, and used for the quantification of hydroxyproline (Hyp) and pentosidine (Pen) as measures of collagen content and of the number of collagen crosslinks, respectively. The amount of Hyp was measured following neutralization and a reaction with chloramine-T and dimethyl amino-benzaldehyde [36]. To determine the collagen content, it was assumed that Hyp comprises approximately 10% of the collagen weight [26].

For the determination of the Pen level, high-performance liquid chromatography (HPLC) was used as described in detail elsewhere [37]. The number of Pen crosslinks were determined and expressed as mole per mole collagen, assuming 300 Hyp residues per triple-helical of collagen molecule and a molecular weight of 300,000 g/mol [14, 38].

Microscopy

The plugs adjacent to those used for the biochemical analysis from all regions were fixed in 4% formalin and used for histology. Cryosections with the thickness of 10 μm were stained with SafraninO/Fast green and Picrosirius red. To eliminate the possible effect of sectioning and processing on the structure of collagen fibers, two pieces with a thickness of approximately 1 mm from the central region of one sample incubated with the highest concentration of ribose (R500) and another sample form the control group were stained with collagen type I and II antibodies (see Supplementary material for more detail). Confocal microscopy was used to make the images from thick slices.

Statistical analyses

To analyze the effect of ribose incubation and region on mechanical and biochemical properties, two-way analyses of variance (ANOVA) with LSD post-hoc test with the assumption of normal distribution of data were performed. Linear regression with Pearson correlation analysis was conducted to assess the correlation between the ribose concentration and the quantified parameters as well as the correlation between pentosidine and stiffness. All values are reported as mean ± standard deviation. The statistical tests were carried out using SPSS 23 for Windows (SPSS Inc., Chicago, IL, USA), a p<0.05 was considered statistically significant.

Results

Crosslinking induction

Collagen crosslinking increased with the amount of ribose added to the samples. The condyles incubated with a high concentration of ribose, i.e., R250 and R500 showed a significant increase in the pentosidine (Pen) level (Figure 2B); being almost 50-fold after incubation with R500. Linear regression analysis demonstrated a significant


43

3Figure 2: Pentosidine (Pen) amount of different regions and ribose concentrations. Regional (A) and overall (B) Pen amount for different ribose concentrations. Pen level increased significantly after ribose treatment in all region; # shows significant different between R500 and other groups with one-way ANOVA). There was also a significant positive correlation between overall Pen content and ribose concentration (Pearson correlation analyses). ** p<0.01, *** p<0.001 significantly different from control group with one-way ANOVA.

Figure 3: Instantaneous (A) and steady state (D) moduli (EIns and ESt)of different regions showed that the anterior region has the lowest elastic moduli of all regions (# p<0.01). General ratio of elastic moduli for incubated samples versus control (B) and (E) showed a significant positive association between increased ribose concentration and elastic modulus with Pearson correlation analyses. R500 and R250 groups had significantly higher general EIns ratios than the control and R125 groups (one-way ANOVA, * p<0.05). There were also significant positive correlations between Pen amount and both general ratios with Pearson correlation analyses (C and F).

Chapter 3

44

positive correlation between ribose concentration and Pen level. The increasing trend of Pen with ribose concentration was similar for all regions (Figure 2A).

Mechanical properties

When ribose concentration increased, there was an increasing trend in stiffness as defined by instantaneous and steady state moduli (EIns and ESt). This was apparent for all regions (Figure 3A, D). Statistical analyses revealed the absence of a statistically significant interaction between concentration and region on compressive moduli (p=0.954 for instantaneous and p= 0.708 for steady state moduli). The anterior regions showed the lowest stiffness for both incubated and control samples; posterior and central regions were the stiffest regions (Figure 3A, D). A general stiffness ratio was calculated by dividing the stiffness of the incubated group with the average stiffness of the control group. Linear regression revealed significant correlations between general stiffness ratios and the ribose concentration (Figure 3B, E). Linear regression also showed significant correlations between general stiffness ratios and Pen increase (R= 0.308 for general EIns ratio and R= 0.265 for general ESt ratio, both p<0.01; Figure 3C, F). An almost 50-fold increase in cartilage Pen levels increased the general EIns ratio by about 1.5-fold from the control to R500.


GAG content results demonstrated no differences among the groups with respect to regions (Figure4 A). However, when all regions for each condyle were grouped on the basis of the ribose incubation (Figure 4 B), GAG concentration negatively correlated with ribose. This reduction was significant for condyles incubated with the highest ribose concentration, i.e., from 8.6±1.0 µg/mgDW for the R500 to 7.2±0.4 µg/mgDW for the control group.

Figure 4: GAG amount of different regions. Regional (A) and overall (B) GAG content for different ribose concentrations. There was a significant negative correlation between overall GAG content and ribose concentration (Pearson correlation analysis). GAG content decreased significantly after incubation with R500 (one-way ANOVA, ** p<0.01).


45

3With respect to the amount of collagen no significant difference was found between the various regions (Figure 5A). The results also showed that less collagen was measured when the concentration of ribose increased (Figure 5B). The reduction of collagen was significant for the R500 group when compared to the control group.

Microscopy

Histology showed that the TMJ condylar cartilage was covered with a dense collagenous layer (Figure 6A-D). When the concentration of ribose increased, spaces appeared within the collagenous matrix (arrows in Figures 6 and 7). In the control group, collagen fibers in the superficial layer were aligned parallel to the surface (Figure 6E). However, this parallel orientation changed following the incubation with ribose. As it is shown in Figure 6 (panels E-H), collagen fibers were oriented in different directions when the concentration of ribose was increased, being most pronounced in the R500 group (Figure 6 H). This was confirmed with immunostaining, when thicker sections of the central region from the control group and the R500 group were stained with collagen type I and II antibodies (Figure 7A, B). In line with the Picrosirius red stained sections (see Figure 6) confocal microscopy revealed a disorganized orientation of the collagen fibers. Confocal microscopy images also confirmed that the disorganization in collagen fibers observed in histological sections was not due to tissue processing.

Figure 5: Collagen content of different regions. Regional (A) and overall (B) collagen content for different ribose concentrations. There was a significant negative correlation between overall collagen content and ribose concentration (Pearson correlation analysis). Samples incubated with R500 showed significantly lower collagen content (one-way ANOVA, *** p<0.001).

Chapter 3

46

Figure 7: Confocal microscopy of the central region from the control group (A), and the central region of the R500 group (B). Thick samples, that stained for collagen type I (green), II (red), and nuclei (blue), showed disorganization in the collagen fibers specifically in the superficial layer in the incubated sample in comparison to the control. Arrows show the spaces appeared within the matrix.

Figure 6: Micrographs of histologic sections showing a sagittal view of the central region of control and incubated samples. A-D show sections stained with SafraninO/Fast green. The superficial fibrous layer stained in blue shows the presence of collagen fibers in this layer. This area is shown at a higher magnification in the inserted figure (40X). The micrographs show that spaces within the matrix of the cartilage increased gradually when the concentration of ribose increased. E-H represent the collagenous structure of the sections under polarized light microscope specifically in the superficial layer. Sections were stained with Picrosirius red. As the concentration of ribose increased, spaces between the collagen fibers also increased. The arrangement of collagen fibers, which was parallel to surface in the control group, changed after incubation. Arrows show the spaces appeared within the matrix.


47

3

DiscussionThe present study investigated the aging-like effect of ribose-induced crosslinking on the crosslink level of collagen and the concomitant effect on biomechanical characteristics of the TMJ condylar cartilage. Ribose incubation significantly correlated with the number of collagen crosslinks, approximated with Pen, and the compressive stiffness. The level of Pen as a measure of AGEs showed a significant positive correlation with the increasing ribose concentrations. The rise in Pen was in line with those reported in previous studies [11, 14, 16, 21, 27]. Bank et al. measured about 50 times more Pen in human articular cartilage after its incubation with 600 mM ribose for 86 h; the increase in Pen level measured in our study was similar to the increase they obtained in a shorter time with a comparable ribose concentration. A possible explanation for this different level of Pen could be the difference in tissue types. We used TMJ condylar cartilage; a cartilage quite different from articular cartilage: fibrocartilage versus hyaline cartilage. In other words, it contains both collagen type I and type II In contrast to hyaline cartilage, collagen type I can be detected throughout the TMJ condylar cartilage [39]. Although both collagen type I and II are fibrillar collagens, Antipova and Orgel (2010) have shown that collagen type II has more possible interfibrillar crosslinking sites than collagen type I. Collagen type II has similar peptide chains in its triple helical structure with three lysine groups for interfibrillar crosslinking, whereas collagen type I has different peptide chains of which only two can provide potential crosslinking sites [40]. We speculate that these differences in collagen molecular structure may create the different results in the crosslinking with ribose. Another reason could be the difference in the tissue’s exposure to sugar. As Bank and coworkers have used small cartilage plugs rather than the intact joint that was used here, the tissue plugs were exposed to ribose solution from different directions, whereas the intact condyle was exposed to the solution only from the surface. The method used in the present study is more mimicking the in vivo situation than the method in which tissue plugs are used.

The results of stiffness measurement depicted that EIns and ESt were region–dependent; a finding in line with previous studies [41-43]. The stiffness in compressive loading is thought to be due to the presence of the negatively charged proteoglycans, which are able to trap water and thus contribute to the cartilage viscosity [44]. A recent study has also shown that the compressive properties of the tissue are determined by the Pen amount for both artificially induced and naturally occurring crosslinks [45]. Moreover, after ribose incubation, roughly 10% of the detected crosslinks belonged to proteoglycans [11]. These findings suggested that compressive properties of articular cartilage during maturation and aging were controlled not only by proteoglycan content but also by Pen. Our results confirm that there is also such a correlation between Pen and compressive stiffness in fibrocartilaginous TMJ condyle.

Chapter 3

48

During AGE formation, crosslinks are created particularly between lysine and arginine residues of collagen molecules [22]. Is has been assumed that collagen crosslinking make the whole collagen network stiffer; however, it has been recently indicated that it is not the collagen network but sliding capacity in the fiber at the fibril level is influenced. The fibril sliding in the fiber of tendon collagen was reduced remarkably after getting more crosslinks [46]. Fessel et al. also showed that the induction of more crosslinks does not influence the stiffness of the collagen fibers in tendon, but instead alters fiber-fiber and fibril-fibril sliding which consequently influences the mechanical response [47, 48]. The same explanation could be extrapolated to the results of TMJ condylar cartilage, in which the dominant collagen is type I, specifically in the superficial fibrous layer. This layer of the TMJ condylar cartilage plays an important role in its compressive and friction properties [41, 49].

The above-mentioned effects of crosslinking on fibrils are also in good agreement with our microscopic observations, which showed a dramatic change in the fibrous layer of condylar cartilage after incubation with ribose. It has been observed that inter-fibrillar crosslinks were created due to diabetes or following incubation with a sugar, and that those crosslinks caused closer packing of several fibrils [5, 7]. We therefore assumed that, due to packing of fibrils in some areas, more Pen can result in the shrinkage in some areas and the formation of spaces in neighboring areas (arrows in Figure 5 and 6).

It is known that in OA, the collagen network is damaged at early stages; this damage can further lead to proteoglycan loss. Therefore, structural changes of collagen related to the increase in Pen level as shown here may result in the onset or progression of OA as it has been shown by De Groot et al. [50]. In contrast, at the late stage of OA, the amount of AGEs was inversely correlated with cartilage damage, most probably due to an altered collagen and proteoglycan turnover rate in OA [38].

Surprisingly, when Pen increased, a reduction in the amount of GAG as an indicator of the level of proteoglycans was observed. However, this reduction in GAG content did not reflect the aging-associated reduction in GAG which has been previously shown [44]. Firstly, the water content which has a direct correlation with the amount of proteoglycans did not change (see Supplementary material). Secondly, it has been shown that after creation of crosslinks in cartilaginous matrix, about 10% of detected Pen in the tissue was derived from proteoglycans [11]. This means that proteoglycans were also involved in crosslinking reactions which could result in a lower digestibility. Therefore, we assume that the decrease in the amount of GAG was due to differences in the digestibility of crosslinked proteoglycans. The same reasoning as mentioned above for GAG reduction can applied to the observed reduction in collagen, because the molecular solubility of collagen decreases by an accumulation of AGE [4]. Kulmala et al. has also reported a reduction in the collagen content after incubation of the


49

3

articular cartilage with threose [51]. Crosslinking has also been shown to decrease the susceptibility of articular cartilage to enzymatic digestion [16, 52].

It has been demonstrated that compressive properties of the superficial layer in the TMJ condyle play an important role in interstitial fluid pressurization (IFP). IFP provides around 90% of tissue support under applied load [49]. AGE production reactions modify free amino groups of the collagen molecules and other proteins in the tissue matrix [22]. These modifications change the fixed charge density and its distribution within the matrix, thereby influencing the tertiary structure of collagen molecules, the interactions of collagen with other proteins and water, and the fluid pressure resulting from fixed charge density in the matrix [14]. Our results demonstrated a clear increase in compressive stiffness, concomitant with a rise in Pen level as well as structural changes in the fibrous layer of the condyles. The potential correlation between our finding in this study and the fixed charge density of proteoglycans, permeability, IFP, and loading distribution within the condylar cartilage provides more insights into the mechanism behind the aging-like effect of crosslinking on tissue function. .

Although our study provided insights into biomechanical and biochemical consequences of AGE formation in the TMJ condylar cartilage, it had some limitations that should be considered. First, ribose used in this study can induce only an effect of collagen crosslinking. Although this method has been used widely, there are several other parameters which can influence the mechanical responses in the cartilaginous matrix. Moreover, there are some contradictory results which noted that crosslinking had a different effect than aging on collagen structure. Belkacem et al. have found that ribose incubation increased collagen fibril diameter, while aged samples have shown an overall reduction in their collagen fibril diameter [10].

Second, the condyles used in the present study were from young pigs in maturation stage, which undergo enzymatic crosslinking. Even though the enzymatic and non-enzymatic reactions have different routes, it might have an influence on crosslinking of the collagen molecules. The obtained results were not compared with results from naturally aged samples. Such a comparison can give a better insight into the differences and the similarities of the aging model in this study on biomechanical responses.

Thirdly, the increase in the collagen crosslinking correlated significantly with moduli ratios in our study. However, the individual regions did not show a significant increase in modulus both for steady state and instantaneous. Since the moduli of the R500 samples were always higher than for the other groups, it can be the small number of samples used for the experiments which did not result in significant differences.

Finally, some assumptions were made in mechanical characterization which can influence the results, e.g., the average of the last five reaction force cycles was used to

Chapter 3

50

calculate the steady state modulus. In addition, the edge effect of indenter on measured stiffness was neglected assuming that the effect is the same in all experiments.

In conclusion, an ex vivo crosslinking method was used to stimulate AGE formation and to examine the effect of crosslinking on biomechanical properties of the TMJ condylar cartilage without other age-related alterations. Incubation with ribose correlated significantly with the amount of Pen and stiffness. The accumulation of AGEs not only influenced the compressive properties of the tissue but also the structure of the tissue itself. Although our approach does not mimic all aspects of aging, it can be considered as a practical model for simulating age-related (mechanical) changes in the TMJ condyle. Our findings are important since TMJ condyle has a unique structure among other load-bearing articular joints in which tissue matrix contains collagen type I throughout the cartilage.

Acknowledgment This research was funded by the European commission through move-age, an Erasmus Mundus Joint Doctorate programme (2011-0015). The authors would like to thank Sepanta Fazaeli and Kamran Nazmi for their technical help.


51

3

References1. Lodish, H., et al., Molecular cell biology. 4th ed. 2000, New York: W.H. Freeman and Co.2. Panwar, P., et al., Changes in Structural-Mechanical Properties and Degradability of Collagen during Aging-associated Modifications. The Journal of biological chemistry, 2015. 290(38): p. 23291-306.3. Ghazanfari, S., A. Khademhosseini, and T.H. Smit, Mechanisms of lamellar collagen formation in connective tissues. Biomaterials, 2016. 97: p. 74-84.4. Wilson, S.L., et al., A microscopic and macroscopic study of aging collagen on its molecular structure, mechanical properties, and cellular response. The FASEB Journal, 2014. 28(1): p. 14-25.5. Odetti, P., et al., Scanning force microscopy reveals structural alterations in diabetic rat collagen fibrils: role of protein glycation. Diabetes/Metabolism Research and Reviews, 2000. 16(2): p. 74-81.6. Scharf, B., et al., Age-related carbonylation of fibrocartilage structural proteins drives tissue degenerative modification. Chemistry & biology, 2013. 20(7): p. 922-34.7. Bai, P., et al., Glycation alters collagen fibril organization. Connective Tissue Research, 1992. 28(1-2): p. 1-12.8. Tang, S.Y., U. Zeenath, and D. Vashishth, Effects of non-enzymatic glycation on cancellous bone fragility. Bone, 2007. 40(4): p. 1144-51.9. Brüel, A. and H. Oxlund, Changes in biomechanical properties, composition of collagen and elastin, and advanced glycation endproducts of the rat aorta in relation to age. Atherosclerosis, 1996. 127(2): p. 155-165.10. Aït-Belkacem, D., et al., Microscopic structural study of collagen aging in isolated fibrils using polarized second harmonic generation. Journal of Biomedical Optics, 2012. 17(8): p. 0805061-0805063.11. Bank, R.A., et al., Ageing and zonal variation in post-translational modification of collagen in normal human articular cartilage. The age-related increase in non-enzymatic glycation affects biomechanical properties of cartilage. The Biochemical Journal, 1998. 330(1): p. 345-351.12. Basser, P.J., et al., Mechanical properties of the collagen network in human articular cartilage as measured by osmotic stress technique. Archive of Biochemistry and Biophysics, 1998. 351(2): p. 207-19.13. Bourne, J.W., J.M. Lippell, and P.A. Torzilli, Glycation Cross-Linking Induced Mechanical-Enzymatic Cleavage of Microscale Tendon Fibers. Matrix biology, 2014. 34: p. 179-184.14. DeGroot, J., et al., Age-related decrease in proteoglycan synthesis of human articular chondrocytes: the role of nonenzymatic glycation. Arthritis and Rheumatism, 1999. 42(5): p. 1003-9.15. Verzijl, N., et al., Effect of collagen turnover on the accumulation of advanced glycation end products. J Biol Chem, 2000. 275(50): p. 39027-31.16. DeGroot, J., et al., Age-related decrease in susceptibility of human articular cartilage to matrix metalloproteinase-mediated degradation: the role of advanced glycation end products. Arthritis and Rheumatism, 2001. 44(11): p. 2562-71.17. Illien-Junger, S., et al., Combined anti-inflammatory and anti-AGE drug treatments have a protective effect on intervertebral discs in mice with diabetes. PLoS One, 2013. 8(5): p. e64302.18. Kohn, R.R., A. Cerami, and V.M. Monnier, Collagen aging in vitro by nonenzymatic glycosylation and browning. Diabetes, 1984. 33(1): p. 57-9.19. DeGroot, J., et al., Accumulation of advanced glycation end products decreases collagen

Chapter 3

52

turnover by bovine chondrocytes. Experimental Cell Research, 2001. 266(2): p. 303-10.20. Verzijl, N., et al., Crosslinking by advanced glycation end products increases the stiffness of the collagen network in human articular cartilage: a possible mechanism through which age is a risk factor for osteoarthritis. Arthritis nad Rheumtism, 2002. 46(1): p. 114-23.21. Vos, P.A.J.M., et al., Elevation of cartilage AGEs does not accelerate initiation of canine experimental osteoarthritis upon mild surgical damage. Journal of Orthopaedic Research, 2012. 30(9): p. 1398-404.22. Sell, D.R. and V.M. Monnier, Structure elucidation of a senescence cross-link from human extracellular matrix. Implication of pentoses in the aging process. J Biol Chem, 1989. 264(36): p. 21597-602.23. Willems, N.M., et al., Higher number of pentosidine cross-links induced by ribose does not alter tissue stiffness of cancellous bone. Materials Science and Engineering: C, 2014. 42: p. 15-21.24. Chen, A.C., et al., Induction of advanced glycation end products and alterations of the tensile properties of articular cartilage. Arthritis and Rheumtism, 2002. 46(12): p. 3212-7.25. Reddy, G.K., Cross-linking in collagen by nonenzymatic glycation increases the matrix stiffness in rabbit achilles tendon. Experimental Diabesity Research, 2004. 5(2): p. 143-53.26. Reddy, G.K., L. Stehno-Bittel, and C.S. Enwemeka, Glycation-induced matrix stability in the rabbit achilles tendon. Archive of Biochemistry and Biophysics, 2002. 399(2): p. 174-80.27. Willett, T.L., et al., Enhanced levels of non-enzymatic glycation and pentosidine crosslinking in spontaneous osteoarthritis progression. Osteoarthritis and Cartilage, 2012. 20(7): p. 736-744.28. Koolstra, J.H., Dynamics of the Human Masticatory System. Critical Reviews in Oral Biology & Medicine, 2002. 13(4): p. 366-376.29. Athanasiou, K.A., et al., Tissue Engineering of Temporomandibular Joint Cartilage. Synthesis Lectures on Tissue Engineering. Vol. 1. 2009: Morgan & Claypool Publishers. 1-122.30. Herring, S.W., TMJ anatomy and animal models. Journal of musculoskeletal & neuronal interactions, 2003. 3(4): p. 391-394.31. Kim, K.W., et al., Biomechanical tissue characterization of the superior joint space of the porcine temporomandibular joint. Annals of Biomedical Engineering, 2003. 31(8): p. 924-30.32. Fazaeli, S., et al., The contribution of collagen fibers to the mechanical compressive properties of the temporomandibular joint disc. Osteoarthritis and Cartilage, 2016. 24(7): p. 1292-1301.33. Mirahmadi, F., et al., Ex vivo thickness measurement of cartilage covering the temporomandibular joint. Journal of Biomechanics, 2017. 52: p. 165-168.34. Tanaka, E., et al., Dynamic properties of bovine temporomandibular joint disks change with age. Journal of dental research, 2002. 81(9): p. 618-22.35. Ghazanfari, S., et al., In vivo Collagen Remodeling in the Vascular Wall of Decellularized Stented Tissue-Engineered Heart Valves. Tissue Eng Part A, 2015. 21(15-16): p. 2206-15.36. Paul, C.P.L., et al., Dynamic and Static Overloading Induce Early Degenerative Processes in Caprine Lumbar Intervertebral Discs. PLoS ONE, 2013. 8(4): p. e62411.37. Bank, R.A., et al., Amino acid analysis by reverse-phase high-performance liquid chromatography: improved derivatization and detection conditions with 9-fluorenylmethyl chloroformate. Analytical Biochemistry, 1996. 240(2): p. 167-76.38. Vos, P.A.J.M., et al., In end stage osteoarthritis, cartilage tissue pentosidine levels are inversely related to parameters of cartilage damage. Osteoarthritis and Cartilage, 2012. 20(3): p. 233-240.39. Almarza, A.J. and K.A. Athanasiou, Design characteristics for the tissue engineering of cartilaginous tissues. Annals of biomedical engineering, 2004. 32(1): p. 2-17.40. Antipova, O. and J.P.R.O. Orgel, In Situ D-periodic Molecular Structure of Type II Collagen.


53

3

Journal of Biological Chemistry, 2010. 285(10): p. 7087-7096.41. Lu, X.L., V.C. Mow, and X.E. Guo, Proteoglycans and mechanical behavior of condylar cartilage. Journal of Dental Research, 2009. 88(3): p. 244-8.42. Lamela, M.J., et al., Dynamic compressive properties of articular cartilages in the porcine temporomandibular joint. Journal of the Mechanical Behavior of Biomedical Materials, 2013. 23(0): p. 62-70.43. Tanaka, E., et al., Stress relaxation behaviors of articular cartilages in porcine temporomandibular joint. Journal of Biomechanics, 2014. 47(7): p. 1582-7.44. Kuroda, S., et al., Biomechanical and biochemical characteristics of the mandibular condylar cartilage. Osteoarthritis and Cartilage, 2009. 17(11): p. 1408-1415.45. Julkunen, P., et al., Biomechanical, biochemical and structural correlations in immature and mature rabbit articular cartilage. Osteoarthritis Cartilage, 2009. 17(12): p. 1628-38.46. Li, Y., et al., Advanced glycation end-products diminish tendon collagen fiber sliding. Matrix Biology, 2013. 32(3-4): p. 169-77.47. Fessel, G., et al., Advanced Glycation End-Products Reduce Collagen Molecular Sliding to Affect Collagen Fibril Damage Mechanisms but Not Stiffness. PLoS ONE, 2014. 9(11): p. e110948.48. Gautieri, A., et al., Advanced glycation end-products: Mechanics of aged collagen from molecule to tissue. Matrix Biol, 2016.49. Ruggiero, L., et al., Roles of the Fibrous Superficial Zone in the Mechanical Behavior of TMJ Condylar Cartilage. Annals of Biomedical Engineering, 2015. 43(11): p. 2652-2662.50. DeGroot, J., et al., Accumulation of advanced glycation end products as a molecular mechanism for aging as a risk factor in osteoarthritis. Arthritis and Rheumtism, 2004. 50(4): p. 1207-15.51. Kulmala, K.A., et al., Diffusion of ionic and non-ionic contrast agents in articular cartilage with increased cross-linking--contribution of steric and electrostatic effects. Medical engineering & physics, 2013. 35(10): p. 1415-20.52. Ghazanfari, S., et al., Modulation of collagen fiber orientation by strain-controlled enzymatic degradation. Acta Biomaterialia, 2016. 35: p. 118-26.

Chapter 3

54

Appendix A. Supporting information

Immuno-staining

To qualitatively assess the effect of ribose treatment, specifically on the structure of collagen fibers, two pieces with a thickness of approximately 1 mm from the central region of one sample treated with the highest concentration of ribose (R500) and another sample from the control group were cut and fixed with 4% formalin for 24 h. Afterwards, the specimens were washed in (phosphate buffer saline) PBS and incubated with blocking buffer (1% BSA, 20% goat serum, PBS) for 4 h to prevent non-specific staining. The samples were then incubated with a cocktail of primary antibodies containing rabbit anti-collagen I (1:1000, ab34710; Abcam, USA) and mouse anti-collagen II (1:10, CIIC1, Developmental Studies Hybridoma Bank, USA) for 24 h at room temperature. After three washing steps in PBS, samples were incubated overnight at room temperature with a cocktail of secondary antibodies containing Alexa 488 goat anti-rabbit (1:2000, Invitrogen, USA) and Alexa 555 goat anti-mouse (1:2000, Invitrogen, USA). Thereafter, the nuclei were stained with DAPI (Vector Laboratories, USA) for 1 h. Images were taken using an inverted confocal microscope (Leica SP8, Leica, Germany).

Water content

Water content, as calculated on the basis of wet and dry weight, showed similar values for all treated and control samples throughout the condyle, i.e., 75.5% (Figure A1). Neither region nor Pen level had a significant effect on water content.

Figure A1: Water content of different regions. Regional (A) and overall (B) water content for different ribose concentrations.

Chapter 4

Diffusion of charged and uncharged contrast agents in equine mandibular condylar cartilage

is not affected by an increased level of sugar-induced collagen crosslinking

Fereshteh Mirahmadi , Jan Harm Koolstra, Sepanta Fazaeli, Frank Lobbezoo, G. Harry van Lenthe, Jessica Snabel, Reinout Stoop, Vincent

Everts

Manuscript under review

Chapter 4

58

AbstractNutrition of articular cartilage relies mainly on diffusion of solutes through the interstitial fluid due to the lack of blood vessels. The diffusion is controlled by two factors: steric hindrance and electrostatic interactions between the solutes and the matrix components. Aging comes with changes in the cartilage structure and composition, which can influence the diffusion. In this study, we treated fibrocartilage of mandibular condyle with ribose to induce an aging-like effect by accumulating collagen crosslinks. The effect of steric hindrance or electrostatic forces on the diffusion was analyzed using either charged (Hexabrix) or uncharged (Visipaque) contrast agents. Osteochondral plugs from young equine mandibular condyles were treated with 500 mM ribose for 7 days. The effect of crosslinking on mechanical properties was then evaluated via dynamic compression loading. Thereafter, the samples were exposed to contrast agents and imaged using contrast-enhanced computed tomography (CECT) at 18 different time points up to 48 h to measure their diffusion. Normalized concentration of contrast agents in the cartilage and contrast agent diffusion flux, as well as the content of crosslink level (pentosidine), water, collagen, and glycosaminoglycan (GAG) were determined. Ribose treatment significantly increased the pentosidine level (from 0.01 to 7.6 mmol/mol collagen), which resulted in an increase in tissue stiffness (~1.5 fold). Interestingly, the normalized concentration and diffusion flux did not change after the induction of an increased level of pentosidine either for Hexabrix or Visipaque. The results of this study strongly suggest that sugar-induced collagen crosslinking in TMJ condylar cartilage does not affect the diffusion properties.

Artificial aging effect on diffusion in TMJ cartilage

59

4

IntroductionWith aging, the articular cartilage undergoes dynamic and gradual changes in its composition, structure, and biophysical properties. The cartilage changes from a bluish transparent appearance at younger age to a yellowish colored tissue at older age. The process of color change, called browning, is due to an accumulation of non-enzymatic crosslinks in proteins, mainly in collagens [1]. Pentosidine (Pen) is one the most studied crosslinks, creating the crosslinks between arginine and lysine residues [2]. Crosslinking influences cartilage in different ways [3-5]; for instance, it causes reduction in enzymatic cleavage of tissue matrix. It has been also shown that an increased crosslink level leads to stiffening of hyaline cartilage [1, 6, 7]. Such has also been demonstrated with the occurrence of diabetes. In order to mimic aging and the accompanying increased crosslinking, cartilage can be incubated with a sugar. This results in an increased cartilage stiffness resembling accelerated aging or a diabetic condition [4].

Crosslinking alters the tissue structure and matrix framework. These changes might reduce the effective porosity of the cartilage matrix which can, in turn, affect the diffusion of solutes [8]. Arginine and lysine are positively charged amino acids. The crosslink between these amino acids can govern collagen interactions through different mechanisms by influencing charge distribution and aggregating properties of collagen fibrils, and also by controlling collagen interactions with other matrix components [9, 10]. Collagen crosslinking also changes the packing distance of collagen fibers [11] and increases lateral packing of collagen fibrils [12, 13]. Since cartilage is avascular, diffusion of nutrients plays an essential role in its homeostasis. Understanding the changes in the diffusion of solutes through the cartilage during aging and the factors influencing diffusion is crucial for establishing treatment strategies after wear or damage. Furthermore, it may establish requirements for tissue engineering of this tissue.

The diffusion of solutes into cartilage is governed by steric hindrance and electrostatic effects [8, 14-16]. It has been shown that the diffusion of a negatively charged contrast agent (Hexabrix) into hyaline cartilage matrix decreased after an increase in crosslink content [14, 15]. An increase in crosslink content, however, did not change the diffusion of a neutral contrast agent (Visipaque). Contradictory to earlier studies, we found that in the fibrocartilage of the temporomandibular joint (TMJ), the diffusion of Visipaque significantly decreased due to normal aging, whereas the diffusion of Hexabrix did not (unpublished data). The decreased diffusion of Visipaque correlated with an elevated amount of collagen and Pen at early time points. In the present study, we aimed to investigate if the observed changes in TMJ cartilage could be explained by an increased sugar-induced collagen crosslink content. We hypothesized that an increased

Chapter 4

60

number of collagen crosslinks limits transport of solutes into the fibrocartilage, either through steric hindrance or through electrostatic interactions. To test our hypothesis, we induced collagen crosslinks by incubating young equine fibrocartilage of the TMJ with ribose. Moreover, we aimed to clarify the electrostatic or steric effects of such a collagen crosslinking on the diffusion. Thereby, we measured time-dependent diffusion of charged (Hexabrix) and uncharged (Visipaque) contrast agents in treated and control samples. Finally, we measured the mechanical properties of the cartilage to examine whether they were affected by crosslinking.


Sample preparation

Frozen equine heads from young animals (2.5, 3, and 4 years old) were obtained from the Faculty of Veterinary Medicine, University of Utrecht (gender not specified). The animals were sacrificed to serve as educational specimens in dissection lectures, hence, for a reason other than the present study. After thawing, the right and left mandibular condyles were dissected. Thus, six condyles were collected which were visually without any abnormalities. The thickness of the cartilage in the central region of condyles was measured using a method we described in a previous study [17]. Briefly, a thin layer of 2% agarose gel containing 20% barium sulfate was used to cover the condyle surface followed by scanning the condyle with a micro-CT 40 apparatus (Scanco Medical AG, Bruttisellen, Switzerland). The average thickness of cartilage in the central region was estimated using reconstructed images. Two groups were defined: treated and control. The samples of the treated group were incubated with 500mM ribose in phosphate buffer saline (PBS) in the presence of a protease inhibitor cocktail (PBS-PI; 2mM ethylenediaminetetraacetic acid; 5mM benzamidine; 10mM N-ethyl-maleimide; 1mM phenylmethylsulfonyl fluoride [18]), and sodium azide 0.02% for 7days at 37˚C. The inhibitor cocktail was added to minimize tissue degradation and sodium azide to prevent bacterial growth. The concentration of ribose was selected on the basis of a previous study [4]. The control group was incubated in the same solution without ribose. To perform mechanical loading and diffusion test, cylindrical plugs (diameter 6mm) containing the cartilage layer and the connected subchondral bone were drilled out from the central region of the condyle as shown in Figure 1A. The plugs were stored at -20˚C in PBS+PI prior to the other experimental steps. Rectangular pieces were cut for biochemical tests and histological staining.

Biophysical testing

To determine the stiffness, cylindrical plugs were fixed in a holder filled with PBS+PI. A tare load of 0.2N was introduced to the region of interest using a rigid cylindrical indenter (diameter 2mm). To create a uniform loading history in all samples, 60


61

4

compression cycles with 1% strain amplitude were applied at a frequency of 1 Hz. This strain was defined as a fraction of the local cartilage thickness. After 5 min relaxation time for equilibrating force, a strain level of 5% was applied for 20 cycles at 1Hz. Force-displacement data was recorded. Instantaneous modulus (EIns) was calculated from the peak force of the first cycle of loading. The steady state modulus (ESt) was calculated from the average peak force of the last five cycles [4].

After mechanical characterization, the cylindrical plugs were used for the diffusion tests. To provide an axial diffusion only from the cartilage surface, the lateral sides of the plugs were covered with cyanoacrylate (Histoacryl, Braun Surgical S.A., Rubi, Spain) and wrapped in parafilm (Figure 1B). Thereafter, the relevant contrast agent was added to the surface of the samples. Contrast-enhanced computed tomography (CECT) technique was used to track the time-dependent diffusion of contrast agents into the cartilage. The diffusion of Hexabrix (diluted 20:80 in PBS-PI; Hexabrix, 1269 g/mol, charge = -1, GE Healthcare, Netherlands) was first measured using a Bruker micro-CT (Bruker SkyScan 1272, SkyScan, Kontich, Belgium) with isotropic voxel size of 10μm with 66kV tube voltage and 166μA current at 18 time points up to 48h. After the diffusion test with Hexabrix, the plug was washed for 48h with PBS-PI at 4˚C with several changes of the solution to remove Hexabrix from the cartilage. Then, diffusion of Visipaque (Visipaque, 1550.191 g/mol, charge=0, GE Healthcare, Netherlands) was tracked with the same scanning parameters. Visipaque was used undiluted. To measure the diffusion of each solution by time, the gray index of the whole cartilage was averaged at each time point using micro-CT software (CTAn version1.16.4.1, SkyScan). A volume of interest was chosen from reconstructed images that include the whole thickness of the cartilage with the same software. Contrast agent concentration within the cartilage was normalized to the bath concentration with the assumption of a linear correlation between the gray index and the concentration of the contrast agents. The flux of solution into the cartilage was calculated with equation (1):

Figure 1: (A) Tissue preparation. The central region was divided in two groups: treated (with 500mM ribose) and control. The osteochondral plugs (Ø = 6mm) were drilled for the diffusion test, rectangular fragments were used for biochemical tests, and sagittal slices were collected for histology. (B) The osteochondral plug was covered with cyanoacrylate and parafilm to allow diffusion of contrast agent from the cartilage surface.

Chapter 4

62

J= - h.əC/ət (1)

where J is the flux, h is the cartilage thickness, t is time, and C is the bulk solution concentration within the cartilage.


Biochemical analyses were performed after drying the rectangular pieces according to [4]. In brief, samples were freeze-dried, digested with papain, and then analyzed for calculating the amount of GAG with dimethyl methylene blue (DMMB) assay. One portion of the digested solution was hydrolyzed with 6N HCl and a reaction of the hydrolyzed solution with chloramine-T and dimethyl amino-benzaldehyde used for Hydroxyproline (Hyp) assessment to estimate the amount of collagen [19]. The collagen content was determined assuming that Hyp comprises approximately 10% of the collagen weight [20]. To determine the amount of crosslink, Pentosidine (Pen) level was measured by means of high-performance liquid chromatography technique as described elsewhere [21].

Microscopy Sagittal sections of 7μm thickness were made from the rectangular pieces using a microtome and stained with SafraninO/Fast green (SafO/FG) and Picrosirius red (Picro). They were fixed in 4% formalin and decalcified with neutral EDTA for 6 weeks. The structure of the cartilage was visualized using light microscopy after SafO/FG staining and the spatial alignment of the collagen fibers was visualized with polarized light microscopy (PLM).


To assess the effect of ribose treatment on the biophysical parameters, independent t-tests were performed. The values were reported as mean ± standard deviation (mean±SD). SPSS 23 for Windows (SPSS Inc., Chicago, IL, USA) was used for statistical tests, considering p values less than 0.05 statistically significant.

Results

Biophysical properties

The calculated stiffness of the mandibular condyles of the samples is shown in Figure 2. After ribose treatment, the instantaneous modulus (EIns) was significantly higher in the ribose-treated group than in the control group. No significant difference was observed between steady state modulus (Est) values.

Using contrast-enhanced computed tomography (CECT), we measured the diffusion of Hexabrix and Visipaque. The normalized concentration of Visipaque at equilibrium was lower than that of Hexabrix at equilibrium. The final concentration of Hexabrix in both


63

4

Figure 2: Instantaneous (EIns) and steady state (ESt) Moduli of ribose-treated and control samples. The treated samples had significantly higher instantaneous modulus (EIns ) values than those in the control group (* p = 0.013).

Figure 3: Normalized concentration of (A) Hexabrix and (B) Visipaque over time. No differences were observed between the treated and control groups. Calculated flux for (C) Hexabrix and (D) Visipaque did not show significant differences between the control and treated groups.

Chapter 4

64

treated and control groups reached a similar concentration of about 67% at equilibrium (Figure 3A). No significant differences were found between the treated group and the control group after the diffusion of Visipaque had equilibrated; the control group reached 42% and the treated group 40% normalized concentration after 48h (Figure 3B). There were also no differences between the fluxes for both Hexabrix and Visipaque groups (Figure 3C, D).


The results of the biochemical analyses are summarized in table 1. We pooled the results of all ages, since samples were from a relatively young age and the data from the control samples was not significantly different. A similar water content of about 83% was found in both groups. No significant differences were found between groups in the amount of GAG and collagen. There was a slight increase in the amount of collagen after ribose-treatment from 64% to72%, but this was not significant. However, the number of sugar-induced collagen crosslinks as measured by the level of Pen, significantly increased from 0.01 in control samples to 7.36 mmol/mol collagen in the treated ones.

Microscopy

Histological staining of the samples is shown in Figure 4. As shown in sections stained with SafO/FG (Figure 4A), the ribose-treated tissues contained a more packed superficial layer of the cartilage in comparison to the control ones. The structure and the thickness of the fibrous superficial layer also varied among samples. Ribose treatment seemed to promote the interaction of the superficial layer with the underlying layer. Comparing the two groups, it became apparent that during sectioning the superficial layer of control samples frequently detached from the underlying layer (see arrows in Figure 4A upper row). This phenomenon was not seen with ribose-treated samples. Figure 4B represents the superficial layer of the samples. It seems that due to crosslinking, the crimp and waviness of the collagen fibers in the superficial layer slightly decreased.

Table 1: Biochemical values for the control and treated group (means±SD, n=6/group).

*** p<0.001


65

4

Figure 4: Histological slices from control and treated samples stained with (A) SafO/FG with 5x magnification (B) PLM microscopy images of Picro-stained slices with 20x magnification. Treated samples appeared to have a better attached superficial layer than the control ones; black arrows show the detaching of the superficial during sectioning in control samples (see in Figure 4A upper row). (black scale bars represent 400μm and white scale bar 50μm).

Chapter 4

66

DiscussionIn this study, we investigated the aging-like effect of sugar-induced collagen crosslinking on the diffusion of solutes into the fibrocartilaginous mandibular condylar cartilage of the horse. Diffusion of solutes into cartilage is likely to be controlled by steric hindrance and electrostatic interactions. These two modalities were analyzed using different compounds: a neutral and an anionic agent. We found that an increase in Pen level caused an increase in stiffness, while no difference was noted for the diffusion of both contrast agents into the cartilage.

In line with previous studies, the Hexabrix flux into the cartilage was remarkably slower than that of Visipaque [22]. Cartilage contains negatively charged glycosaminoglycans (GAGs), therefore electrostatic forces contribute significantly to the diffusion of charged contrast agents [23, 24]. Due to the presence of more GAGs in the hyaline cartilage than in, for instance, the meniscus, a cationic solute diffuses much faster in the former as a result of stronger electrostatic interactions [23]. Such an electrostatic effect of the GAGs in the diffusion of a charged contrast agent was confirmed after removal of GAG from the tissue. This resulted in an increase in the diffusion of negatively charged contrast agent and a decrease in the diffusion of positive charge contrast agent [24].

The mandibular condyle is covered with fibrocartilage in which collagen content is much higher than in hyaline cartilage [25]. Another type of fibrocartilage, the meniscus, also contains a high collagen content. It has been shown that a higher collagen content in the meniscus compared to hyaline cartilage created more steric hindrance and resulted in less diffusion of a negatively charge contrast agents at early time points. However, the presence of GAG in hyaline cartilage led to lower diffusion at equilibrium [26]. GAGs are the most important sources of an electrostatic charge in the cartilage. However, it is known that collagen’s amino acids are electrically charged as well [27]. Crosslinking in the presence of a sugar (glycation) changes the cartilage charge density. Glycation creates crosslinks between side chains of collagen polypeptides containing arginine and lysine molecules which carry a positive charge [9, 10]. Thereby, as shown by Kokkonen et al., after collagen crosslink induction by the sugar incubation, fixed charge density (FCD) of hyaline cartilage increased significantly [15]. As a consequence, the diffusion of a negatively charged contrast agent reduced [14, 15, 28]. Hence, it has been suggested that the contribution of the sugar-induced collagen crosslinks to diffusion is considered primarily due to a change in electrostatic interactions [14]. The findings presented in this study are partly in line with findings previously published. It has been shown that the diffusion of the neutral contrast agent (Visipaque), was not affected after collagen crosslinking [14]. Nevertheless, Leddey et al. showed that the diffusion of uncharged molecules can be affected by a change in the constituents in different layers of cartilage and their steric hindrance effects [29]. The histological images of this study showed that


67

4

the structure of the fibrocartilage of the mandibular condyle did not remarkably change after crosslinking (Figure 4A), although the Pen level increased significantly. Therefore, the induction of more crosslinks as measured with Pen did not effectively contribute to attenuation of the diffusion of Visipaque into fibrocartilage through steric hindrance.

The composition of collagen differs between different cartilage types. The TMJ condyle consists of fibrocartilage and contains high levels of collagen type I in addition to collagen type II with a ratio of about 4 to 1 [30]. The latter type of collagen is the main collagen type in hyaline cartilage [31]. These two collagen types are different in their peptides that form the triple helix. As a consequence, the overlapping and the spacing of their triple peptide chains are different. Moreover, their glycation pattern is also different [27]. It has been shown previously that a difference in tissue composition may have a remarkable effect on diffusion [23, 26, 29]. The presence of less collagen in the superficial layer of hyaline cartilage compared to the cartilage of the meniscus, caused faster diffusion of a negatively charged contrast agent in the former at early time points [26]. In contrast to hyaline cartilage, in which the diffusion of Hexabrix in ribose-treated samples remarkably decreased [14], we did not detect such a difference between Hexabrix diffusion in treated and control samples of TMJ condylar cartilage. This is most likely due to the different type of cartilage present in the latter. The variations in the collagen types might result in different crosslink effects eventually resulting in a different diffusion of a compound like Hexabrix. In addition, collagen type I in fibrocartilage is more prominently located in its superficial layer [31]. It has been shown that the composition of the superficial layer has an important effect on diffusion [26]. This may have contributed to the different results regarding Hexabrix diffusion into fibrocartilage.

The level of collagen crosslinks can also explain the contradictory results of the present study. A general understanding from similar studies suggests that the higher the collagen crosslink level, the larger the change in the FCD. For instance, Kokkonen et al. showed an increase in Pen level from 0.0014 to 18.2 mmol/mol concomitant with a remarkable change in the FCD of the tissue [15]. However, another study has shown an increase on Pen level from 0.91 to 4.72 mmol/mol without a significant increase in overall FCD [14]. It is probable that the increase of Pen level in our study (0.1 to 7.36 mmol/mol) did not cause a profound electrostatic effect for reducing Hexabrix diffusion.

One might argue that another factor influencing the results of this study is the different concentration of contrast agents we used. The possible effect of contrast agent concentration on the diffusion has been investigated by several investigators [22, 32]. They have shown that in low and high concentrations levels of Hexabrix and Visipaque, diffusion was independent of the solution concentration; hence, the concentration used in this study may not potentially influence the diffusion.

Chapter 4

68

The size of the diffusing molecules is another factor that can strongly contribute to diffusion [29, 32, 33]. However, this parameter was ruled out in our study by using two contrast agents with a similar molecular size.

Equine joints have been used widely as a model, specifically for orthopedic diseases, due to their comparable characteristics to human joints. The thickness of the hyaline cartilage and subchondral bone resembles closely the human knee joint [34, 35]. Although equine TMJ is more complex than that in human, the thickness of TMJ cartilage in mature horses (unpublished data) is also close to what has been measured in human [36]. The diffusion largely depends on the thickness of the cartilage and the thickness of subchondral bone [37]. Thus, equine mandibular condyle can provide an attractive model for the study of diffusing molecules in TMJ cartilage.

To conclude, the present results suggest that diffusion of contrast agents with a negative charge or without electric charge does not change after a considerable change in the level of Pen in fibrocartilage. To the best of our knowledge, this study is the first addressing the effect of artificial aging on the diffusive properties of fibrocartilage of the mandibular condyle. This study indicated that the overall effect of sugar-induced collagen crosslinks in fibrocartilage does not significantly contribute to diffusion; steric hindrance and electrostatic interactions as shown previously for hyaline cartilage were not found for the fibrocartilage in the equine TMJ. Further studies on naturally aged hyaline cartilage and the possible alteration of diffusion will provide more insight into different aging aspects of hyaline and fibrocartilage.

Acknowledgment This research was funded by the European commission through Move-age, an Erasmus Mundus Joint Doctorate programme (2011-0015). The authors would like to thank the Faculty of Veterinary Medicine, University of Utrecht for providing us with animal material for this study. The authors also thank Johannes Korfage for performing histology.


69

4

References 1. Bank, R.A., et al., Ageing and zonal variation in post-translational modification of collagen in normal human articular cartilage. The age-related increase in non-enzymatic glycation affects biomechanical properties of cartilage. The Biochemical Journal, 1998. 330(1): p. 345-351.2. Sell, D.R. and V.M. Monnier, Structure elucidation of a senescence cross-link from human extracellular matrix. Implication of pentoses in the aging process. The Journal of biological chemistry, 1989. 264(36): p. 21597-602.3. Bourne, J.W., J.M. Lippell, and P.A. Torzilli, Glycation Cross-Linking Induced Mechanical-Enzymatic Cleavage of Microscale Tendon Fibers. Matrix biology, 2014. 34: p. 179-184.4. Mirahmadi, F., et al., Mechanical stiffness of TMJ condylar cartilage increases after artificial aging by ribose. Archives of oral biology, 2018. 87: p. 102-109.5. Avery, N.C. and A.J. Bailey, The effects of the Maillard reaction on the physical properties and cell interactions of collagen. Pathologie Biologie, 2006. 54(7): p. 387-395.6. Verzijl, N., et al., Crosslinking by advanced glycation end products increases the stiffness of the collagen network in human articular cartilage: a possible mechanism through which age is a risk factor for osteoarthritis. Arthritis nad Rheumtism, 2002. 46(1): p. 114-23.7. Snedeker, J.G. and A. Gautieri, The role of collagen crosslinks in ageing and diabetes - the good, the bad, and the ugly. Muscles, Ligaments and Tendons Journal, 2014. 4(3): p. 303-308.8. Maroudas, A., et al., The Permeability of Articular Cartilage. Journal of Bone & Joint Surgery, British Volume, 1968. 50-B(1): p. 166-177.9. Hadley, J.C., K.M. Meek, and N.S. Malik, Glycation changes the charge distribution of type I collagen fibrils. Glycoconjugate Journal, 1998. 15(8): p. 835-840.10. Bailey, A.J., R.G. Paul, and L. Knott, Mechanisms of maturation and ageing of collagen. Mechanisms of Ageing and Development, 1998. 106(1–2): p. 1-56.11. Tanaka, S., et al., Glycation induces expansion of the molecular packing of collagen. Journal of Molecular Biology, 1988. 203(2): p. 495-505.12. Bai, P., et al., Glycation alters collagen fibril organization. Connective Tissue Research, 1992. 28(1-2): p. 1-12.13. Odetti, P., et al., Scanning force microscopy reveals structural alterations in diabetic rat collagen fibrils: role of protein glycation. Diabetes/Metabolism Research and Reviews, 2000. 16(2): p. 74-81.14. Kulmala, K.A., et al., Diffusion of ionic and non-ionic contrast agents in articular cartilage with increased cross-linking--contribution of steric and electrostatic effects. Medical engineering & physics, 2013. 35(10): p. 1415-20.15. Kokkonen, H.T., et al., Computed tomography detects changes in contrast agent diffusion after collagen cross-linking typical to natural aging of articular cartilage. Osteoarthritis and Cartilage, 2011. 19(10): p. 1190-1198.16. Muir, H., P. Bullough, and A. Maroudas, The distribution of collagen in human articular cartilage with some of its physiological implications. The Journal of bone and joint surgery. British volume, 1970. 52(3): p. 554-63.17. Mirahmadi, F., et al., Ex vivo thickness measurement of cartilage covering the temporomandibular joint. Journal of Biomechanics, 2017. 52: p. 165-168.18. Kim, K.W., et al., Biomechanical tissue characterization of the superior joint space of the porcine temporomandibular joint. Annals of Biomedical Engineering, 2003. 31(8): p. 924-30.19. Paul, C.P.L., et al., Dynamic and Static Overloading Induce Early Degenerative Processes in Caprine Lumbar Intervertebral Discs. PLoS ONE, 2013. 8(4): p. e62411.

Chapter 4

70

20. Reddy, G.K., L. Stehno-Bittel, and C.S. Enwemeka, Glycation-induced matrix stability in the rabbit achilles tendon. Archive of Biochemistry and Biophysics, 2002. 399(2): p. 174-80.21. Bank, R.A., et al., Amino Acid Analysis by Reverse-Phase High-Performance Liquid Chromatography: Improved Derivatization and Detection Conditions with 9-Fluorenylmethyl Chloroformate. Analytical Biochemistry, 1996. 240(2): p. 167-176.22. Pouran, B., et al., Isolated effects of external bath osmolality, solute concentration, and electrical charge on solute transport across articular cartilage. Medical Engineering & Physics, 2016. 38(12): p. 1399-1407.23. Honkanen, J.T.J., et al., Cationic Contrast Agent Diffusion Differs Between Cartilage and Meniscus. Annals of Biomedical Engineering, 2016. 44(10): p. 2913-2921.24. Cockman, M.D., et al., Quantitative imaging of proteoglycan in cartilage using a gadolinium probe and microCT. Osteoarthritis and Cartilage, 2006. 14(3): p. 210-214.25. Vonk, L.A., et al., Caprine articular, meniscus and intervertebral disc cartilage: An integral analysis of collagen network and chondrocytes. Matrix Biology, 2010. 29(3): p. 209-218.26. Honkanen, J.T.J., et al., Transport of Iodine Is Different in Cartilage and Meniscus. Annals of Biomedical Engineering, 2016. 44(7): p. 2114-2122.27. Antipova, O. and J.P.R.O. Orgel, In Situ D-periodic Molecular Structure of Type II Collagen. Journal of Biological Chemistry, 2010. 285(10): p. 7087-7096.28. Pouran, B., et al., Non-enzymatic cross-linking of collagen type II fibrils is tuned via osmolality switch. Journal of Orthopaedic Research, 2018: p. n/a-n/a.29. Leddy, H.A. and F. Guilak, Site-specific molecular diffusion in articular cartilage measured using fluorescence recovery after photobleaching. Annals of biomedical engineering, 2003. 31(7): p. 753-60.30. Wang, K.-H., et al., Histological and Immunohistochemical Analyses of Repair of the Disc in the Rabbit Temporomandibular Joint Using a Collagen Template. Materials, 2017. 10(8): p. 924.31. Athanasiou, K.A., et al., Tissue Engineering of Temporomandibular Joint Cartilage. Synthesis Lectures on Tissue Engineering. Vol. 1. 2009: Morgan & Claypool Publishers. 1-122.32. Silvast, T.S., et al., Bath Concentration of Anionic Contrast Agents Does Not Affect Their Diffusion and Distribution in Articular Cartilage In Vitro. Cartilage, 2013. 4(1): p. 42-51.33. Kokkonen, H.T., et al., Detection of mechanical injury of articular cartilage using contrast enhanced computed tomography. Osteoarthritis Cartilage, 2011. 19(3): p. 295-301.34. MacDonald, M.H., et al., Characterization of age- and location-associated variations in the composition of articular cartilage from the equine metacarpophalangeal joint. Journal of Equine Veterinary Science. 22(1): p. 25-32.35. McIlwraith, C.W., et al., Equine Models of Articular Cartilage Repair. Cartilage, 2011. 2(4): p. 317-326.36. Renders, G.A.P., et al., Contrast-enhanced microCT (EPIC-µCT) ex vivo applied to the mouse and human jaw joint. Dentomaxillofac Radiol, 2014. 43(2): p. 20130098.37. Pouran, B., et al., Solute transport at the interface of cartilage and subchondral bone plate: Effect of micro-architecture. Journal of Biomechanics, 2017. 52: p. 148-154.

Chapter 5

Aging does not change the compressive stiffness of mandibular condylar cartilage in horses

Fereshteh Mirahmadi , Jan Harm Koolstra, Sepanta Fazaeli, Frank Lobbezoo, G. Harry van Lenthe, Jessica Snabel, Reinout Stoop, Vahid

Arbabi, Harrie Weinans, Vincent Everts

Manuscript has been accepted for publication in “Osteoarthritis and Cartilage“; in press.

Chapter 5

74

AbstractAging can cause an increase in the stiffness of hyaline cartilage as a consequence of increased protein crosslinks. By induction of crosslinking, a reduction in the diffusion of solutions into the hyaline cartilage has been observed. However, there is lack of knowledge about the effects of aging on the biochemical and biophysical properties of the temporomandibular joint (TMJ) cartilage. Hence, the aim of this study was to examine the biophysical properties (i.e., thickness, stiffness, and diffusion) of the TMJ condylar cartilage of horses of different ages and their correlation with biochemical parameters. We measured the compressive stiffness of the condyles, after which the diffusion of two contrast agents into cartilage was measured using Contrast Enhanced Computed Tomography technique. Furthermore, the amount of the components of the extracellular matrix (water, collagen, GAG, pentosidine) was analyzed. Pentosidine, Collagen, and GAG showed positive correlations with age. Contrary to our expectations, the stiffness of the cartilage did not change with age (modulus remained around 0.7 MPa). The diffusion of the negatively charged contrast agent (Hexabrix) also did not alter. However, the diffusion of the uncharged contrast agent (Visipaque) decreased with aging. The flux was negatively correlated with the amount of collagen and crosslink level which increased by aging, implying the steric hindrance effect of aging on diffusion. In conclusion, our data demonstrated that aging was not necessarily reflected in the biophysical properties of TMJ condylar cartilage. The combination of the changes happening due to aging resulted in different diffusive properties, depending on the nature of the solution.

Aging effect on stiffness and diffusion in TMJ cartilage

75

5

IntroductionThe temporomandibular joint (TMJ) of the jaw is composed of the temporal fossa, mandibular condyle, and articular disc. It is a synovial, bilateral joint with a unique morphology and function. Unlike hyaline cartilage in other articular joints with collagen type II as the dominant collagen, condylar cartilage in TMJ is a fibrocartilaginous tissue which contains both collagen type I and type II. It has a dense superficial layer of fibrillar collagen type I. Condylar cartilage absorbs and distributes shear, tensile, and compressive forces during rotational and translational jaw movement. This cartilage is a secondary cartilage and the center of growth in the mandible [1, 2]. During aging, gradual structural and compositional changes occur in TMJ cartilaginous extracellular matrix (ECM) [3-5]. ECM changes have been reflected in biophysical and mechanical performance of hyaline cartilage [5, 6].

Similar to other articular cartilage, TMJ condylar cartilage consists of three main constituents, i.e., collagen, proteoglycans (PGs), and water which together create the functionality of the matrix. Each of these components changes due to aging. The bulk of the dry weight of the cartilage is collagen [7]. The prominent functional feature of collagen is maintaining tissue integrity and stability under tensile loading [8, 9]. Collagen fibers are oriented parallel to the cartilage surface at the superficial layer of TMJ condyle [10]. This layer controls the mechanical response of the tissue in hyaline and fibrocartilage [11, 12]. Cartilaginous collagen has an extremely slow turnover rate; as a consequence, non-enzymatic crosslinks between collagen molecules result in accumulation of advanced glycation end products (AGEs) in cartilage with increasing age [13, 14]. One of these AGEs, pentosidine, may serve as a well-characterized and reliable measure of AGEs, and has been reported abundantly present in articular cartilage. The accumulation of AGEs leads to stiffening of hyaline cartilage with aging [9, 14]. PGs, the second abundant component of cartilage, are considered to contribute to the compressive stiffness of cartilage due to their negatively charged glycosaminoglycan (GAG) chains [15, 16]. The resultant highly negative fixed-charge density (FCD) also provides a hydrophilic environment with a water content of up to 80% which reduces by increasing age [8]; however, the total amount of PG does not remarkably change during aging [3]. The changes happen in the composition, structure, and hydrodynamic size of PGs during aging [15, 17, 18]. Next to the structural role of PGs, they also contribute to nutrient and solute transport in cartilage due to their highly hydrated nature [18].

Articular cartilage is an avascular tissue in which the transport of metabolites depends mainly on diffusion [19, 20]. PGs and water content strongly contribute to the permeability and diffusion [20]. Yet, little is known about the effect of aging-associated changes on the diffusive properties of cartilage. This knowledge is considered also to provide valuable information for treatment and regeneration strategies. It has been

Chapter 5

76

shown that when the number of crosslinks increases, the hyaline cartilage not only becomes stiffer, but it becomes also less diffusive to anionic contrast agents [21, 22]. However, there are other compositional and structural changes in aging cartilage beside crosslinking that can influence diffusion. To the best of our knowledge, there is no study on the alteration of cartilage diffusion with natural aging in TMJ condylar cartilage. Therefore, the aim of this study was to investigate the effect of aging-associated changes on tissue diffusion and biomechanics, and their correlation with biochemical composition of mandibular condylar cartilage. We hypothesized that when the TMJ condylar cartilage becomes older, it gets more crosslinks, becomes stiffer and less diffusible. To evaluate this hypothesis, we measured the diffusion of a negatively charged (Hexabrix) and an uncharged contrast agent (Visipaque) into TMJ cartilage of horses in different ages, using contrast-enhanced computed tomography (CECT). Next, we determined the mechanical properties of the cartilage under compression. Finally, the biochemical and structural changes of the tissue matrix were assessed. The correlations between age and biophysical and biochemical properties of the tissue were then examined.


Sample preparation

Frozen equine heads from different ages were obtained from the faculty of veterinary medicine of the University of Utrecht, the Netherlands (gender not specified). The animals had been sacrificed to serve as educational specimens in dissection lectures, hence, for a reason other than the present study. Their use was according to the ethical standards of the University of Utrecht. The age of the animal, as estimated from their teeth, was between 2.5 and 18 years old. After thawing the whole head for 24 h, right and left condyles were dissected from the temporomandibular joints. Condyles used in this study were visually intact. The samples were divided into 3 groups on the basis of their ages as young (2.5, 3, and 4 years old), young adult (7, 8, and 10 years old), and middle-aged (12, 15, 17, and 18 years). In total, we had 6 condyles for the young as well as for the middle-aged group, and 8 condyles for the old-aged group. The thickness of the cartilage in the central region of condyles was measured using micro-CT scanning of the cartilage covered with 20% barium sulfate in agarose gel, as described in a previous study [23]. Knowing the thickness, strain-control cyclic compression was applied to the central region of condyles for calculating the cartilage stiffness. Afterward, cylindrical osteochondral plugs (diameter 6 mm) were drilled out for the diffusion tests while phosphate buffer saline (PBS) was continuously sprayed on the samples to prevent dehydration and damage. Sagittal tissue slices with a thickness of about 3-5 mm were cut from the adjacent area for histological staining using a band saw. Surrounding tissue of the cartilage plugs was collected for biochemical analysis as shown in Figure 1A. The


77

5plugs were stored at -20˚C in PBS enriched with a protease inhibitor cocktail (PBS-PI; 2mM Ethylenediaminetetraacetic acid (EDTA); 5mM benzamidine; 10mM N-ethyl-maleimide; 1mM phenylmethylsulfonyl fluoride [24]; and sodium azide 0.02%) prior to diffusion test. The inhibitor cocktail was added to minimize tissue degradation and sodium azide to prevent bacterial growth.

Mechanical testing

To determine the stiffness of the central region of the intact condyle, cyclic compression loading test was performed by using a custom-made instrument as described previously [25]. Briefly, a solid cylindrical indenter (diameter 4mm) applied strain-control cyclic displacements of the cartilage and simultaneously the compressive reaction force was recorded. A tare load of 0.2 N was introduced to the region of interest. Then, 60 cycles of 1% strain at a frequency of 1Hz were applied as preconditioning; after 5 min relaxation time, a strain level of 5% was applied for 20 cycles at 1Hz. Instantaneous modulus (EIns) was calculated from the peak force of the first cycle of loading. Under cyclic loading, a steady-state response is usually reached within 10 cycles [26]. Therefore, steady state modulus (ESt) was calculated from the peak force average of the last five loading cycles. As compressive stiffness is different topographically across the condyle [25], only the central region was tested from all the condyles.

Figure 1: Tissue preparation steps. (A) An osteochondral plug (Ø = 6mm) was drilled from the central region of the condyles for diffusion test, a sagittal slice was taken for histology, and the surrounding tissue was used for biochemical tests. (B) Sagittal view of a plug in μCT holder for diffusion test. (C) Pseudo-colored image of the contrast enhanced computed tomography of an osteochondral plug at zero hours and 48h. The attenuation of cartilage increased after diffusing of the contrast agent.

Chapter 5

78

Contrast enhanced computed tomography (CECT)

For diffusion measurements, each plug was thawed in PBS-PI at room temperature. The lateral sides of the plug were sealed using a thin layer of cyanoacrylate (Histoacryl, Braun Surgical S.A., Rubi, Spain), followed by wrapping in parafilm to allow diffusion only form the surface (Figure 1B). Diffusion of two contrast agent with different electric charge (negative and uncharged) was evaluated for each plug. Hexabrix solution containing ioxaglate (Hexabrix, 1269 g/mol, charge = -1, GE Healthcare, Netherlands) was prepared with a concentration of 0.08M in PBS-PI. After the diffusion test with Hexabrix, the plug was washed out for 48h with PBS-PI at 4˚C with several changes of solution. Thereafter, the plug was used for the diffusion of Visipaque, i.e., a neutral contrast agent. Visipaque solutions that contained iodixanol solutes (Visipaque, 1550.191 g/mol, charge=0, GE Healthcare, Netherlands) were used without any dilution with a concentration of 0.42M. Images were acquired using a micro-CT apparatus (Bruker SkyScan 1272, SkyScan, Kontich, Belgium) with an isotropic voxel size of 10μm with 66 kV tube voltage and 166 μA current at 18 time points. Average gray index for the whole cartilage thickness was calculated at each time point using micro-CT software (CTAn version1.16.4.1, SkyScan). A volume of interest from the reconstructed image was chosen with the same software as to include the whole thickness of the cartilage (Figure 1C). Gray values of cartilage at time zero were taken as zero concentration of the contrast agent. Contrast agent concentration within the cartilage was normalized to the bath concentration which was taken as 100%. The concentration gradient of the bulk at different time points was used to define the flux through the cartilage surface with the following formula:

J= - h.əC/ət

where J is the flux, h is the cartilage thickness, C is the bulk concentration within the cartilage, and t is time.


To examine the effect of aging on the tissue constituents, the tissue that surrounded the plugs was weighted and subsequently freeze-dried for 24h. Wet and dry weights were used for calculating water content. Dried tissue samples were digested with papain and the amount of GAG was measured with a colorimetric assay using dimethyl methylene blue. Hydroxyproline (Hyp) amount, for the assessment of collagen content, and pentosidine (Pen) amount, as a measure of the number of collagen crosslinks, were assessed after hydrolyzing the digested solution with 6M HCl. The amount of Hyp was measured following neutralization and a reaction with chloramine-T and dimethyl amino-benzaldehyde [27]. The collagen content was determined assuming that Hyp comprises approximately 10% of the collagen weight [28]. For the determination of the Pen level, high-performance liquid chromatography was used as described in detail elsewhere [29-31].


79

5

Microscopy

The sagittal slices of tissue from the central region were fixed in 4% formalin and decalcified with neutral EDTA for 6 weeks. Sections with a thickness of 10 μm were stained with SafraninO/Fast green (SafO/FG) and Picrosirius red (Picro). The structure of the cartilage was visualized using light microscopy after SafO/FG staining and collagen fibers spatial alignment was visualized using polarized light microscopy.


To analyze the effect of aging on a quantified parameter, One-way analyses of variance (ANOVA) with Fisher’s Least Significant Difference (LSD) post-hoc test were performed. Linear regression with Pearson correlation analysis was conducted to assess the correlation between the age, biophysical, and biochemical properties. Quantified parameters are reported as mean ± standard deviation (SD). The statistical tests were carried out using SPSS 23 for Windows (SPSS Inc., Chicago, IL, USA). P values less than 0.05 were considered statistically significant.

Results

Mechanical testing

The thickness of cartilage in the central region for each age (right and left taken together) and the average thickness of each age group are presented in Figure 2A, B. The thickness decreased from 0.8 mm for the youngest sample (2.5 years old) to 0.27 mm for the oldest sample (18 years old). Correlation analyses showed a significant negative correlation between age and the thickness of cartilage. When the average thickness of age groups was compared, the young group proved to be the thickest. The thickness of the young adult and middle-aged group was similar (Figure 2B).

Figure 2: Cartilage thickness of the central region of condyles in relation to age. (A) The thickness of the cartilage declined with aging. A significant negative correlation was found between age and thickness. (B) The thicknesses of cartilage in each age group were averaged. One-way ANOVA showed that the young group had significantly thicker cartilage than two other groups (***p<0.001).

Chapter 5

80

The compressive stiffness of the TMJ condylar cartilage in the central region was measured under cyclic compression. No significant difference was observed for both EIns and ESt, between the different age groups, although the older samples were slightly softer than those of the other age groups (Table 1). ESt was slightly lower than EIns in all age groups (0.5 MPa vs. 0.7 MPa). Consequently, no correlations were found between stiffness and age.

Contrast enhanced computed tomography (CECT)

After 48h of diffusion of Hexabrix, the final normalized concentration of the solution within the cartilage reached 62.1% for the young group, 61.6% for the Young adult group, and 63.3% for the middle-aged group (Figure 3A). No significant differences were found between age groups at different time points; however, the middle-aged group samples had a somewhat higher normalized concentration at all time points. Although the young group samples allowed a faster flux indicating faster accumulation of Hexabrix (Table 2), no significant differences were found between Hexabrix flux of the different groups (Figure 3B); In spite of that, the correlation analyses showed a significant negative correlation between Hexabrix flux at early time points with age. There was no significant correlation between the normalized concentration of Hexabrix and age (Table 3).

Table 2: Average fluxes at three different points in time after application of two different contrast agents, Hexabrix and Visipaque (mean±SD). At early time points, the flux of Visipaque was significantly higher in the young samples than in the other groups.

*** p<0.001

Table 1: Stiffness values, expressed as the instantaneous (EIns) and steady state (ESt) elastic modulus and their correlations with age (mean±SD). There were no significant changes in the stiffness of the samples with age.


81

5Figure 3: Change in the concentration and flux of contrast agents with time. A) Average normalized concentrations of Hexabrix diffusion. B) Average Hexabrix flux for the three age groups in time. There were no significant differences between age groups in normalized concentration and flux. C) Average normalized concentration of Visipaque. It was significantly higher in the young samples (n=6) than in the young adult (n=6) and the middle-aged group (n=8) after 6h (** p<0.01). D) Average Visipaque flux for the three age groups. In the young group, Visipaque flux was significantly higher than in the young adult and middle-aged group (*** p<0.001).

Table 3: R values of Pearson correlation analyses between age, biochemical contents, and contrast agent diffusion (Nor. Conc.: normalized concentration).

* p<0.05, ** p<0.01, *** p<0.001

Chapter 5

82

In contrast to Hexabrix, the normalized concentration of Visipaque was significantly higher in young samples than in the two other groups after 6h of diffusion. While the young group reached an average normalized concentration of 38%, this value reached about 23% in the young adult group and 25% in the middle-aged group. Both groups had a significantly lower normalized concentration than the young group after 6 h (Figure 3C). With respect to the Visipaque flux, the values calculated for the young group were significantly higher than those of the young adult and the middle-aged groups (Figure 3D). Flux values of both contrast agents at early time points (15min, 30 min, and 1h) are presented in Table 2. The Visipaque flux decreased concomitantly with a reduction in the water content and an increase in the collagen and Pen levels with aging (see the correlations in Table 3).


The results of the biochemical analyses of the different age groups are presented in Table 4. Water content in the young group was significantly higher than that in other groups. When the horse TMJ condyle gets older, the water content significantly decreases (R= -0.859, p<0.001, Pearson correlation). The amount of GAG increased slightly with advancing age, although this change was not significant. With respect to collagen content, the young group had a significantly lower content than that two other groups. The collagen content increased from 67% in the young samples to 94% in the middle-aged ones. Collagen crosslinks measured with Pen increased 15-fold from the young group to the middle-aged group. The increase in Pen was remarkably higher for the middle-aged group in comparison to the others. Pen amount correlated positively with the age of the samples (r=0.723, p<0.001).

Microscopy

As shown in the Figure 4A, the cartilage became thinner with increasing age. The amount and the distribution of the collagen and GAG in the SafO/FG stained sections demonstrated a series of age-related alterations. For instance, the young samples stained more intensely red, thus indicating a high level of GAG content; the young adult and middle-aged samples had less red staining and therefore likely containing less GAGs.

Table 4: Biochemical values of different age groups and R values of Pearson correlation analyses with age (mean±SD, n=6 for young and young adult and n=8 for middle-aged).

* p<0.05, ** p<0.01, *** p<0.001


83

5

Figure 4: (A) Histology of equine condylar cartilage of different age (staining with SafO/FG). In the young samples (upper row), a more uniform distribution of GAGs was apparent. There is an increasing amount of blue-stained matrix (primarily collagen and non-GAG components) with age. The intensity and distribution of GAG (red) varies considerably among the samples, with the highest amount found in the young samples. The structure becomes more compact with age; particularly in the superficial collagen layer. Black arrows indicate the detachment of the superficial layer during sample preparation. (black scale bar: 300 μm, labels in white boxes indicate the age of the horses). (B) Orientation and thickness of the collagen fibers in relation to age (staining with Picro staining). Polarized microscopy revealed orientation and thickness of collagen fibers in the superficial layer became more packed and oriented in parallel to the surface when the condyles get older (white scale bar 200μm).

Chapter 5

84

In the young samples, the distribution of the GAGs was more uniform, and throughout the cartilage except for the superficial zone. With aging, chondrocyte clusters increased within the cartilage, and loss of GAG staining in 1/3 upper part of cartilage was observed. During the preparation of the samples for histology, it was observed that the superficial layer of the cartilage easily detached from the underlying part in the samples of young ages, particularly in those of 2.5 years old. This can be observed in Figure 4A, indicated by black arrows. The structure and the thickness of the fibrous superficial layer also varied notably by age. The waviness of collagen fibers decreased by advancing age, while the fibers became thicker and more packed (Figure 4B).

DiscussionThis study investigated the effect of natural aging on biophysical properties of equine TMJ condylar cartilage and their correlation with their biochemical components. Our results showed that when the condylar cartilage became older, the diffusion of the neutral contrast agent reduced. The diffusion correlated positively with the water content and thickness of the samples. However, the stiffness of the TMJ condyle did not increase significantly, whereas the number of collagen crosslinks increased 15-fold.

Age-dependent alteration of the TMJ condylar cartilage included the change in its thickness; it decreased significantly with aging. Our results are in accordance with a reduced thickness observed in knee hyaline cartilage [6, 32] and also in the mandibular cartilage of human [33] and rat [34]. Paulsen et al. found a significant negative correlation between mandibular cartilage thickness and age in human [33]. Copray and Duterloo showed that the overall thickness of cartilage in rat TMJ condyle decreased by growing old while the thickness of the superficial layer increased [34]. Moriyama et al. suggested that the tidemark advancement that happened by aging lead to eventual thinning of cartilage [32]. An apparent decrease in the cellularity of the cartilage as seen in the histological staining can also result in the cartilage thickness reduction due to the lack of proliferation capacity of chondrocytes [18]. Our results proved that there was a decline in the diffusion of Visipaque with aging. Thereby, the lack of enough metabolite transport from cartilage surface might result in blood vessel invasion and subchondral bone advancement [35].

One of the most prominent age-associated changes in cartilage is the increase in the number of collagen crosslinks [9, 14]. As expected, we measured a significantly higher Pen level in the young adult and middle-aged group as compared to the samples of younger horses, but our finding in this study did not show any significant difference in the stiffness (neither in instantaneous nor steady state modulus) with aging (Table 1). We previously showed that incubation of porcine TMJ condyles with ribose as an aging-effect simulation increased Pen level and consequently the stiffness of the cartilage [25]. Most of the other studies on the effect of natural aging and the resultant Pen increase


85

5

also demonstrated that the stiffness of cartilage increased with aging, both in hyaline cartilage and in fibrocartilage of TMJ disc [6, 14, 26].

The origin of the seemingly contradictory findings could be related to several parameters that have changed with aging, e.g. Pen level, GAG, and the structure of the superficial layer. Firstly, in a previous study, we measured a 50 fold increase in Pen levels leading to a 1.5 fold increase in stiffness [25]. The crosslink level of naturally aged samples in this study was only 15-fold higher than the young samples. It is likely that the Pen increase in the older equine condyle of our study could not result in significant changes in the contribution of collagen network under compression. Secondly, it is well known that the compressive stiffness of the cartilage depends not only on GAG content [18, 36] but also on the amount of sulfation and ratios of different GAGs [15, 18, 37]. With advancing age, the amount of keratan sulfate (KS) increased and chondroitin sulfate (CS) decreased. Older aggrecan becomes weaker under compression as they have a lower proportion of CS/KS than younger ones [17]. The results of our biochemical analyses for GAG content showed no significant difference with increasing age, which is in line with previous studies on the equine knee joint [3] and human ankle joint [38]. However, our microscopy images clearly showed changes in the distribution and intensity of stained GAG within the matrix. Therefore, we can conclude from above-mentioned reasons that the changes in GAGs have been likely compensated with Pen increase with aging which resulted in the same mechanical response. Thirdly, the superficial layer contributes strongly to the tissue support under loading [12, 39]. However, the compressive properties of this layer of condylar cartilage are inferior to the hyaline-like part underneath [12, 36]. In addition, the thickness of the superficial layer and the degree of parallel orientation of collagen fibers have shown a positive correlation with the stiffness of the cartilage under indentation [39]. Considering our mechanical data, it is suggested that due to the presence of a thicker superficial layer as well as thicker and more homogeneously distributed GAG in young samples resulted in a similar stiffness as seen in older samples with thicker parallel collagen fibers.

The diffusive properties of cartilage depend on the size of the diffusing molecule and the biochemical and physical structure of the cartilage matrix [40]. Together, these contribute to the diffusion alterations through two main properties i.e., steric hindrance and/or electrostatic effects [21, 22, 41, 42]. In this study, Pen level and collagen content increased concomitant with a decreased water content while GAG content did not change with aging; moreover, Pen, Collagen, and GAG showed positive correlations with age (table 4). These changes are all in favor of reducing the diffusion. Visipaque showed a reduction in the flux and normalized concentration with aging as we expected. In contrast, Hexabrix diffusion did not change notably with aging although the flux was the highest for the young sample at early time points (Table 3). This means that steric hindrance was dominant early but not at equilibrium. Since Hexabrix has a similar

Chapter 5

86

molecular weight as Visipaque, the most important factor creating the difference between these two contrast agents was the electrostatic interactions [22]. Visipaque is a neutral contrast agent of which diffusion is mainly controlled by steric hindrance. Thereby, the young samples with higher water content and lower collagen and Pen level showed the highest flux and normalized concentration. On the other hand, electrostatic interactions within cartilage matrix are controlled through the GAGs molecules as well as Pen level [21, 22, 43]. Age-related changes in collagen crosslinking have been shown to lead to increased fix charge density while it reduces sulfation of GAGs and the total of their negative charge [15, 18, 37, 43]. Our microscopy images also confirmed that the overall contribution of GAG charge within the matrix varied due to remarkable change distribution of GAGs within the matrix. Although the GAG content showed a slight increase, the upper 2/3 of cartilage showed hardly any GAG.

We studied several aspects of aging in equine TMJ condylar cartilage. Yet, there is some limitation related to our study. First, we studied only one region of the condyle. The TMJ condyle experiences different loading pattern in each region. Second, we only applied the compression testing while the condyle also experiences shear and tensile in which tissue can differently respond which aging. And third, we did not measure the charge density and the changes in ECM sub-molecules. For instance, hyaluronan, which extensively change with aging [44], has shown an important role in diffusion [45]. Further investigation of distribution and content of different collagen types as well as GAG sub-chains with aging can reveal valuable information about the contribution of these components in the biophysical response of cartilage.

In conclusion, we measured a reduction in water content and in thickness of equine TMJ condylar cartilage with aging, whereas collagen content and Pen level increased. This combination of biochemical alterations resulted in a similar response under compressive loading and diffusion of Hexabrix into the cartilage in all ages. These findings demonstrated that the changes happening in aging mandibular condylar cartilage are not similar to that of hyaline cartilage. Our hypothesis of a reduction of diffusion with aging was confirmed using an uncharged contrast agent implying that the neutral nutrition exchange might be more affected with aging. For future work, investigating such effects in diffusion of different agents (negative, neutral, and positive) into pathological situations, and in hyaline cartilage by aging can provide relevant insight into clinically relevant topics.

Acknowledgment This research was funded by the European commission through move-age, an Erasmus Mundus Joint Doctorate programme (2011-0015). The authors would like to thank the Faculty of Veterinary Medicine, University of Utrecht for providing us with animal material for this study. The authors thank Johannes Korfage for performing histology.


87

5

References1. Herring SW. TMJ anatomy and animal models. Journal of musculoskeletal & neuronal interactions 2003; 3: 391-394.2. Zimmerman BK, Bonnevie ED, Park M, Zhou Y, Wang L, Burris DL, et al. Role of interstitial fluid pressurization in TMJ lubrication. Journal of dental research 2015; 94: 85-92.3. Platt D, Bird JL, Bayliss MT. Ageing of equine articular cartilage: structure and composition of aggrecan and decorin. Equine veterinary journal 1998; 30: 43-52.4. Luder HU. Age changes in the articular tissue of human mandibular condyles from adolescence to old age: a semiquantitative light microscopic study. The Anatomical record 1998; 251: 439-447.5. Panwar P, Lamour G, Mackenzie NC, Yang H, Ko F, Li H, et al. Changes in Structural-Mechanical Properties and Degradability of Collagen during Aging-associated Modifications. The Journal of biological chemistry 2015; 290: 23291-23306.6. Julkunen P, Harjula T, Iivarinen J, Marjanen J, Seppanen K, Narhi T, et al. Biomechanical, biochemical and structural correlations in immature and mature rabbit articular cartilage. Osteoarthritis Cartilage 2009; 17: 1628-1638.7. Athanasiou KA, Almarza AJ, Detamore MS, Kalpakci KN. Tissue Engineering of Temporomandibular Joint Cartilage. Volume 1, Morgan & Claypool Publishers 2009.8. Williams GM, Klisch SM, Sah RL. Bioengineering Cartilage Growth, Maturation, and Form. Pediatric Research 2008; 63: 527-534.9. Snedeker JG, Gautieri A. The role of collagen crosslinks in ageing and diabetes - the good, the bad, and the ugly. Muscles, Ligaments and Tendons Journal 2014; 4: 303-308.10. Wang L, Detamore MS. Tissue engineering the mandibular condyle. Tissue engineering 2007; 13: 1955-1971.11. Aula AS, Jurvelin JS, Toyras J. Simultaneous computed tomography of articular cartilage and subchondral bone. Osteoarthritis Cartilage 2009; 17: 1583-1588.12. Ruggiero L, Zimmerman BK, Park M, Han L, Wang L, Burris DL, et al. Roles of the Fibrous Superficial Zone in the Mechanical Behavior of TMJ Condylar Cartilage. Annals of Biomedical Engineering 2015; 43: 2652-2662.13. Verzijl N, DeGroot J, Thorpe SR, Bank RA, Shaw JN, Lyons TJ, et al. Effect of collagen turnover on the accumulation of advanced glycation end products. The Journal of biological chemistry 2000; 275: 39027-39031.14. Bank RA, Bayliss MT, Lafeber FP, Maroudas A, Tekoppele JM. Ageing and zonal variation in post-translational modification of collagen in normal human articular cartilage. The age-related increase in non-enzymatic glycation affects biomechanical properties of cartilage. The Biochemical Journal 1998; 330: 345-351.15. Bayliss MT, Ali SY. Age-related changes in the composition and structure of human articular-cartilage proteoglycans. Biochemical Journal 1978; 176: 683-693.16. Kuroda S, Tanimoto K, Izawa T, Fujihara S, Koolstra JH, Tanaka E. Biomechanical and biochemical characteristics of the mandibular condylar cartilage. Osteoarthritis and Cartilage 2009; 17: 1408-1415.17. Lee H-Y, Han L, Roughley PJ, Grodzinsky AJ, Ortiz C. Age-related nanostructural and nanomechanical changes of individual human cartilage aggrecan monomers and their glycosaminoglycan side chains. Journal of Structural Biology 2013; 181: 264-273.18. Dudhia J. Aggrecan, aging and assembly in articular cartilage. Cellular and molecular life sciences : CMLS 2005; 62: 2241-2256.

Chapter 5

88

19. O’Hara BP, Urban JP, Maroudas A. Influence of cyclic loading on the nutrition of articular cartilage. Annals of the rheumatic diseases 1990; 49: 536-539.20. Honkanen JTJ, Turunen MJ, Freedman JD, Saarakkala S, Grinstaff MW, Ylärinne JH, et al. Cationic Contrast Agent Diffusion Differs Between Cartilage and Meniscus. Annals of Biomedical Engineering 2016; 44: 2913-2921.21. Kokkonen HT, Mäkelä J, Kulmala KAM, Rieppo L, Jurvelin JS, Tiitu V, et al. Computed tomography detects changes in contrast agent diffusion after collagen cross-linking typical to natural aging of articular cartilage. Osteoarthritis and Cartilage 2011; 19: 1190-1198.22. Kulmala KA, Karjalainen HM, Kokkonen HT, Tiitu V, Kovanen V, Lammi MJ, et al. Diffusion of ionic and non-ionic contrast agents in articular cartilage with increased cross-linking--contribution of steric and electrostatic effects. Medical engineering & physics 2013; 35: 1415-1420.23. Mirahmadi F, Koolstra JH, Lobbezoo F, van Lenthe GH, Everts V. Ex vivo thickness measurement of cartilage covering the temporomandibular joint. Journal of Biomechanics 2017; 52: 165-168.24. Kim KW, Wong ME, Helfrick JF, Thomas JB, Athanasiou KA. Biomechanical tissue characterization of the superior joint space of the porcine temporomandibular joint. Annals of Biomedical Engineering 2003; 31: 924-930.25. Mirahmadi F, Koolstra JH, Lobbezoo F, van Lenthe GH, Ghazanfari S, Snabel J, et al. Mechanical stiffness of TMJ condylar cartilage increases after artificial aging by ribose. Archives of oral biology 2018; 87: 102-109.26. Tanaka E, Aoyama J, Tanaka M, Murata H, Hamada T, Tanne K. Dynamic properties of bovine temporomandibular joint disks change with age. Journal of dental research 2002; 81: 618-622.27. Paul CPL, Schoorl T, Zuiderbaan HA, Zandieh Doulabi B, van der Veen AJ, van de Ven PM, et al. Dynamic and Static Overloading Induce Early Degenerative Processes in Caprine Lumbar Intervertebral Discs. PLoS ONE 2013; 8: e62411.28. Reddy GK, Stehno-Bittel L, Enwemeka CS. Glycation-induced matrix stability in the rabbit achilles tendon. Archive of Biochemistry and Biophysics 2002; 399: 174-180.29. Bank RA, Jansen EJ, Beekman B, te Koppele JM. Amino Acid Analysis by Reverse-Phase High-Performance Liquid Chromatography: Improved Derivatization and Detection Conditions with 9-Fluorenylmethyl Chloroformate. Analytical Biochemistry 1996; 240: 167-176.30. DeGroot J, Verzijl N, Bank RA, Lafeber FP, Bijlsma JW, TeKoppele JM. Age-related decrease in proteoglycan synthesis of human articular chondrocytes: the role of nonenzymatic glycation. Arthritis and Rheumatism 1999; 42: 1003-1009.31. Vos PAJM, Mastbergen SC, Huisman AM, de Boer TN, DeGroot J, Polak AA, et al. In end stage osteoarthritis, cartilage tissue pentosidine levels are inversely related to parameters of cartilage damage. Osteoarthritis and Cartilage 2012; 20: 233-240.32. Moriyama H, Kanemura N, Brouns I, Pintelon I, Adriaensen D, Timmermans JP, et al. Effects of aging and exercise training on the histological and mechanical properties of articular structures in knee joints of male rat. Biogerontology 2012; 13: 369-381.33. Paulsen HU, Thomsen JS, Hougen HP, Mosekilde L. A histomorphometric and scanning electron microscopy study of human condylar cartilage and bone tissue changes in relation to age. Clinical orthodontics and research 1999; 2: 67-78.34. Copray JC, Duterloo HS. A comparative study on the growth of craniofacial cartilages in vitro. European journal of orthodontics 1986; 8: 157-166.35. Wang Y, Wei L, Zeng L, He D, Wei X. Nutrition and degeneration of articular cartilage. Knee Surgery, Sports Traumatology, Arthroscopy 2013; 21: 1751-1762.


89

5

36. Singh M, Detamore MS. Biomechanical properties of the mandibular condylar cartilage and their relevance to the TMJ disc. Journal of biomechanics 2009; 42: 405-417.37. Bayliss MT, Osborne D, Woodhouse S, Davidson C. Sulfation of Chondroitin Sulfate in Human Articular Cartilage: THE EFFECT OF AGE, TOPOGRAPHICAL POSITION, AND ZONE OF CARTILAGE ON TISSUE COMPOSITION. Journal of Biological Chemistry 1999; 274: 15892-15900.38. Aurich M, Poole AR, Reiner A, Mollenhauer C, Margulis A, Kuettner KE, et al. Matrix homeostasis in aging normal human ankle cartilage. Arthritis and rheumatism 2002; 46: 2903-2910.39. Korhonen RK, Wong M, Arokoski J, Lindgren R, Helminen HJ, Hunziker EB, et al. Importance of the superficial tissue layer for the indentation stiffness of articular cartilage. Medical Engineering & Physics 2002; 24: 99-108.40. Maroudas A. Transport of solutes through cartilage: permeability to large molecules. Journal of anatomy 1976; 122: 335.41. Maroudas A, Bullough P, Swanson SAV, Freeman MAR. THE PERMEABILITY OF ARTICULAR CARTILAGE. Journal of Bone & Joint Surgery, British Volume 1968; 50-B: 166-177.42. Muir H, Bullough P, Maroudas A. The distribution of collagen in human articular cartilage with some of its physiological implications. The Journal of bone and joint surgery. British volume 1970; 52: 554-563.43. Hadley JC, Meek KM, Malik NS. Glycation changes the charge distribution of type I collagen fibrils. Glycoconjugate Journal 1998; 15: 835-840.44. Luria A, Chu CR. Articular Cartilage Changes in Maturing Athletes: New Targets for Joint Rejuvenation. Sports Health 2014; 6: 18-30.45. Pluen A, Boucher Y, Ramanujan S, McKee TD, Gohongi T, di Tomaso E, et al. Role of tumor–host interactions in interstitial diffusion of macromolecules: Cranial vs. subcutaneous tumors. Proceedings of the National Academy of Sciences 2001; 98: 4628.

Chapter 6

General Discussion

Chapter 6

92

The temporomandibular joint (TMJ) in human and other mammals consists of two articulating surfaces that are located in the mandibular condyle and temporal bone. An articulating disc separates these two cartilaginous surfaces, thus creating a joint space [1]. Synovial fluid, filling the joint space, provides a medium by which transportation of nutrients to and waste products from articular surfaces is facilitated via diffusion [2]. Like in other joints, the cartilage covering the condyle is avascular which makes diffusion vital for metabolic activities in the matrix [3-6]. During joint functioning, the mandibular condyle moves along the disc and rotates to absorb and distribute forces. The fibrocartilage covering the mandibular condyle is subjected to different loading regimes i.e. compression, tension, and shear. The mode of loading in the TMJ condyle is primarily compression [7].

Compressive stiffness of cartilage depends on the tissue matrix composition namely, collagen, proteoglycans (PGs), and interstitial fluid [5]. Cartilage components and their characteristics change during aging. One of the most prominent changes in aging is the accumulation of non-enzymatic protein crosslinking called glycation. As cartilage has a very slow turnover rate, the accumulation of glycation products is higher in this tissue than other tissues [8, 9]. Previous studies showed that due to an increased level of crosslinking, hyaline cartilage becomes stiffer. This together with other compositional changes in the cartilage matrix due to aging could lead to a decreased capacity of cartilage, for instance, to compensate for loading during joint function [10].

The majority of research regarding cartilage relates to hyaline cartilage, which is abundantly present in the joints of the appendicular skeleton [8, 11-13]. However, there is a lack of knowledge of age-related effects on physical properties of mandibular condylar cartilage of the temporomandibular joint (TMJ). The fibrocartilaginous TMJ condyle differs from hyaline cartilage in numerous ways. The origin of the fibrochondrocytes differs as well as their differentiation and development, the composition of the matrix components, and finally, the load-bearing responses [14-17]. The prevalence of TMJ age-related degeneration, specifically TMJ osteoarthritis (OA), is remarkably low, despite the significant biomechanical loads [16, 18]. Therefore, the effects of the age-related changes in TMJ fibrocartilage might be different from that of hyaline cartilage. Furthermore, when hyaline cartilage becomes damaged, it is replaced by fibrocartilage, which in those cases is considered mechanically inferior. However, it is not always possible to investigate age-associated changes, since tissue from different age ranges is not easily accessible, especially not from human. It is important to make use of an alternative model to study aging effects. Therefore, the aim of this thesis was first, to investigate the biomechanical, diffusive, and biochemical changes in a model for TMJ condylar cartilage in which aging-like crosslinks were induced with ribose. The second aim was to improve our understanding of the effect of normal aging on biophysical (mechanical and diffusive) and biochemical characteristics of TMJ condylar cartilage.

General discussion

93

6

As an initial step for research of biomechanics of cartilage, it is essential to determine its thickness. The mechanical properties of cartilage have been shown to correlate with its thickness: The thicker the cartilage, the stiffer under loading [19, 20]. Ex vivo, needle penetration has been introduced for thickness determination of hyaline cartilage by Swann and Seedhom in 1989 [21]. They used an indentation technique by which the indenter was replaced with a sharp needle. The thickness was measured on the basis of two changes in the reaction force when the needle touched the cartilage surface and subsequently the subchondral bone. Although needle penetration is simple and repeatable, it only gives the thickness in one specific spot. In addition, the state of maturation also can affect the results. As our findings in chapter 2 showed, the border of subchondral bone next to the cartilage in young cartilage is not calcified as is the underlying bone. Therefore, overestimation of cartilage thickness can be expected. This was confirmed in our study. Recent developments in imaging enables us to visualize the soft tissue using contrast enhanced computed tomography (CECT). This method provides information about thickness throughout the cartilage surface [22]. However, this method requires diffusing of contrast agents into the cartilage. The method we developed in chapter 2 was on the basis of CECT in which cartilage surface was covered with barium sulfate as a contrast agent suspended in agarose gel. The contrast agent we used did not penetrate into the cartilage and provided a sharp contrast at the cartilage surface. This method was quick and its results were comparable to those realized by microscopy. In this way a 3D thickness distribution was obtained. Nevertheless, such a method is not feasible for in vivo applications. A recent study has shown a dual contrast CT method which has the potential to be used in vivo for imaging the 3D thickness of cartilage [23].

A previous study has shown that TMJ condylar cartilage varies in thickness in different regions [24]. This finding is in line with variations observed in cartilage thickness as we measured in chapters 2 and 3. As cartilage thickness differs throughout the cartilage surface, mechanical properties also change in different regions of hyaline and fibrocartilage [25, 26 and chapter 3]. However, thickness is not the only factor influencing the mechanical properties; cartilage matrix alterations also dictate the performance of tissue under compression. Different loading patterns after birth cause stiffness adaptation in different regions of cartilage matrix [27, 28]. During aging, the number of non-enzymatic crosslinks increases remarkably in cartilage which results in increased stiffness of hyaline cartilage. Similarly, sugar-induced crosslinking has been used widely in hyaline cartilage to accelerate this aging effect and to artificially induce aging [12, 13]. Hyaline cartilage has collagen type II as its dominant collagen type, whereas TMJ condylar cartilage contains collagen type I in addition to type II. Thus, it remains to be experimentally verified whether a stiffening effect can be induced after incubation of this tissue with a sugar. Therefore in chapter 3, we induced

Chapter 6

94

collagen crosslinks in TMJ condylar cartilage with different concentrations of ribose. The pentosidine level was used as an estimate of the number of crosslinks in cartilage [12]. As expected, a positive correlation was found between the concentration of ribose and the stiffness of the cartilage. The highest ribose concentration used in our study resulted in a 50 fold increase in pentosidine and 1.5 fold increase in the modulus ratio of treated compared to control samples. This increased level of crosslinks has been previously shown for hyaline cartilage of humans between 20 to 90 years old with a concomitant increased stiffness [8].

Cartilage is an avascular structure in which diffusion provides the nutrition as well as the loss of waste product out of the cartilage [3, 29]. Therefore, diffusion from the surface into the deeper layers of cartilage is vital for cartilage metabolism and homeostasis. Artificial aging with sugar incubation has been used also to investigate the aging-like effect of an increased crosslinking on diffusion of hyaline cartilage [30, 31]. It was shown that the diffusion of Hexabrix (a negatively charged contrast agent) reduced after artificial aging of hyaline cartilage, most likely due to electrostatic interactions. Artificial aging resulted in an increased fixed charge density of cartilage most likely due to the formation of covalent bonds between positively charged amino acids [32]. Steric hindrance, however, was not affected as, under similar conditions, no change was found in the diffusion of an uncharged contrast agent [31]. Yet, such an effect needed to be verified in TMJ condylar cartilage. Hence, in chapter 4, we treated young equine condyles with ribose to examine diffusion and mechanical properties in relation to elevated crosslink numbers. Two contrast agents were used for the diffusion test: a negatively charged and an uncharged one. A negatively charged contrast agent potentially shows the effect of crosslinking on diffusion through electrostatic interaction whereas the uncharged one shows such effects through steric hindrance. The crosslink number increased significantly concomitant with 1.5-fold increase in the stiffness of treated samples in comparison to the control ones. In agreement to Kulmala et al., we did not detect a change in the diffusion of an uncharged contrast agent after crosslinking [31]. Yet, in contrast with the findings presented by Kulmala and coworkers for hyaline cartilage we did not detect significant changes in diffusion of a negatively charged contrast agent. One possible reason for our contradictory result is the difference in composition of hyaline and fibrocartilage. It is known that collagen type I is present throughout the fibrocartilage of the TMJ condyle. This collagen has a different molecular composition [31, 33], which might change the effect of crosslinking fixed charge density in hyaline and fibrocartilage. In addition to this, fibrocartilage has an overall different fixed charge density and distribution due to a lower GAG content, and higher collagen content than hyaline cartilage [34, 35]. Overall, these findings suggest that collagen crosslinking did not contribute to the diffusion neither through electrostatic interactions, as shown for hyaline cartilage, nor through steric hindrance.

General discussion

95

6

Taken together, we concluded from chapters 3 and 4 that artificial aging in TMJ condylar cartilage did not create clear aging effects as it was shown to occur in hyaline cartilage. These findings made it important to investigate the normal aging effect in TMJ condylar cartilage with respect to mechanical and diffusive properties. Therefore, in chapter 5 we tested the following hypothesis: TMJ condylar cartilage becomes stiffer and less diffusive during aging. Indeed, we found around 15-fold increase in pentosidine level of fibrocartilage in horses from 2.5 to 18 years old which represented young to senior age. This was comparable with hyaline cartilage. Bramma et al. [36] showed a 10-fold increase in pentosidne in hyaline cartilage of aging horses (4-30 years old) while Bank et al. [8] showed 50-fold increase in human of 20-90 years old. The increase in crosslink level varies among species most probably due to differences in lifespan as well as the age range that were studied. Nevertheless, our findings in chapter 5 did not support our hypothesis; aging and consequent crosslink increase did not result in stiffening of TMJ condylar cartilage. Since the compressive stiffness of cartilage depends on GAG amount in addition to collagen content and crosslink level, it might be that the increased amount of collagen crosslinks was not enough to compensate the effect of GAG distribution. These findings demonstrated that a series of changes (e.g. GAG distribution) happening due to aging in mandibular cartilage might compensate other changes such as increased collagen crosslinking, which consequently results in unaffected biophysical properties. The diffusion of a negatively charged contrast agent also did not change. This is in contrast to what has been shown in artificial aging of hyaline cartilage [30, 31]. A possible explanation for this finding might be the remarkable changed distribution of GAGs in the tissue matrix during aging. GAGs are the main source of charge density of the matrix. GAGs were distributed homogeneously throughout the hyaline-like part of fibrocartilage in the condyle of the samples of young animals, whereas they were distributed sparsely within the matrix of samples of older age. However, the total amount of GAG did not change. Conversely, diffusion of an uncharged contrast agent reduced during normal aging. The two contrast agents used in “artificial aging” and “normal aging” studies were of a similar molecular weight; accordingly, we can conclude that the reduction in the diffusion of an uncharged contrast agent in normally aged cartilage could be mainly due to steric hindrance.

Comparing the results from “artificial” and “natural” aging suggested that an increased crosslink number in fibrocartilage might not be, in contrast to hyaline cartilage, a representative model of aging in TMJ condylar cartilage for mechanical testing and assessment of diffusion. As the stiffness did not change with aging, it appears that TMJ has adapted its architecture and composition to preserve joint function under loading as much as possible. In fact, TMJ has a unique structure and composition with respect to its origin, development, and growth. The unique bilayered joint surface with a very thick superficial fibrocartilage on top of a hyaline-like deeper layer appears to have

Chapter 6

96

impact on its adaptive change with aging. The presence of this superficial layer has a large contribution under loading [37]. From a clinical perspective, the development of an OA-like disease in this joint is remarkably lower than other joints despite the significant biomechanical loads. This is in sharp contrast to other loaded joints such as the knee, hip, and hand [38]. OA can potentially be initiated in cartilage due to its reduced capacity for loading absorption and distribution after stiffening. Therefore, it can be speculated that a relatively low occurrence of OA in TMJ during aging might be due to an absence of cartilage stiffening which has been reported for hyaline joints.

This knowledge appears to provide valuable information for treatment and regeneration strategies. It emphasizes that for regeneration of cartilage, we must take into account that we are dealing with aged tissue that, for instance, no longer has the ability to allow proper diffusion of molecules which might affect its integration with new regenerated tissue.

Limitations and future perspectivesThe major aim of the studies presented in this thesis was to examine whether the current model of hyaline aging (artificial aging with sugar treatment) for mechanical and diffusive changes can be utilized also for TMJ condylar cartilage studies. In order to do that, we started with porcine samples which were shown previously one of the best animal models for mechanical characterization of TMJ condyle. Porcine TMJ condyle has a similar size to that of human [1]. However, the porcine samples for these studies are often collected from the slaughterhouse. They are young and not completely mature. Therefore, for normal aging studies and comparing that with artificial aging we switched to horse as animal model. Horses are herbivore and the size of their TMJ is almost twice that of human. However, they are widely used in hyaline cartilage-related research [39, 40]. Different ages of equine samples are easily accessible, and the thickness of their cartilage [40, 41] and the frequency of TMJ loading is close to that of human [42, 43]. The age range of the horses used in chapter 5 covers skeletally mature animals from very young to senior ages (2.5-18).

Importantly, we did not analyze the changes happening in each biochemical components such as different GAG types as well as the composition and the ratio of different types of collagen. Such an analysis might reveal some clues into the underlying mechanism that made TMJ condyle different from hyaline cartilage in response to collagen crosslink increase in normal aging.

In addition, we only analyzed one region of TMJ condyle with respect to normal aging and comparing that with artificial aging. Therefore, the absence of clear differences in stiffness and diffusion might have been due to region-specific characteristics and loading history. Significant spatial variations in TMJ condylar cartilage mechanical properties

General discussion

97

6

have been reported [44]. It is also known that cartilage matrix changes in response to loading [25, 45]. Therefore, it would be interesting to investigate normal aging effect in other regions of TMJ condyle to figure out whether such an adaptive response as found in chapter 5 can also be detected in other regions.

Next to the studies performed in this thesis, it would be interesting to elucidate the effect of normal aging on the diffusive properties of hyaline cartilage of the same species for a side-by-side comparison with fibrocartilage.

Our findings of chapter 5 indicated that the superficial layer of the horse TMJ condyle changed remarkably with aging. The contribution of the superficial layer of cartilage has already been shown in hyaline cartilage of young samples. This layer affects stiffness [46] and interstitial fluid pressurization [37]. Thus, further investigation about the specific effect of superficial layer on mechanical and diffusive properties on TMJ during aging would yield important results.

It is generally believed that the transport of small molecules within cartilage matrix is through diffusion [3, 47], while large molecules’ transport depends mainly on loading which creates another form of transport called convection [4]. Accordingly, another interesting subject is to measure the changes in the transport of large molecules into TMJ cartilage of different ages under dynamic loading conditions.

ConclusionCollectively, in this thesis, we demonstrated that the mechanical properties of TMJ condylar cartilage considerably changes due to an increased pentosidine level, as induced by means of artificial aging. Yet, an increased pentosidine level due to natural aging did not cause changes in mechanical properties. Interestingly, we found different diffusion patterns in naturally-aged samples when the contribution of charge was studied; diffusion of an uncharged solution was considerably decreased, while diffusion of a negatively charged solution was not affected. If, however, this was compared with artificially induced aging, no effect on diffusion was found. The presented work in this thesis demonstrated biophysical differences between “natural aging” and “artificial aging” in TMJ fibrocartilage. Artificial aging only induces the crosslinking effect of natural aging; it appears that in natural aging a combination of other age-related changes such as GAG distribution and composition, collagen content and its distribution create an adaptation to an increased level of collagen crosslinking. The findings presented in this thesis with future research will provide valuable insights for an improved understanding of fibro- and hyaline cartilage difference and their adaptation potential during aging. The adaptive nature of fibrocartilage as shown in this thesis may help to develop new preventive strategies for improving hyaline cartilage adaptation during aging.

Chapter 6

98

References 1. Herring, S.W., TMJ anatomy and animal models. J Musculoskelet Neuronal Interact, 2003. 3(4): p. 391-394.2. Wadhwa, S. and S. Kapila, TMJ Disorders: Future Innovations in Diagnostics and Therapeutics. Journal of dental education, 2008. 72(8): p. 930-947.3. Maroudas, A., et al., The permeability of articular cartilage. J Bone Joint Surg Br, 1968. 50-B(1): p. 166-177.4. O’Hara, B.P., J.P. Urban, and A. Maroudas, Influence of cyclic loading on the nutrition of articular cartilage. Ann Rheum Dis, 1990. 49(7): p. 536-9.5. Kuroda, S., et al., Biomechanical and biochemical characteristics of the mandibular condylar cartilage. Osteoarthritis Cartilage, 2009. 17(11): p. 1408-1415.6. Jackson, A.R. and W.Y. Gu, Transport properties of cartilaginous tissues. Current rheumatology reviews, 2009. 5(1): p. 40.7. Herring, S.W. and Z.J. Liu, Loading of the temporomandibular joint: anatomical and in vivo evidence from the bones. Cells Tissues Organs, 2001. 169(3): p. 193-200.8. Bank, R.A., et al., Ageing and zonal variation in post-translational modification of collagen in normal human articular cartilage. The age-related increase in non-enzymatic glycation affects biomechanical properties of cartilage. Biochem J, 1998. 330(1): p. 345-351.9. Sell, D.R. and V.M. Monnier, Structure elucidation of a senescence cross-link from human extracellular matrix. Implication of pentoses in the aging process. The Journal of biological chemistry, 1989. 264(36): p. 21597-602.10. Aigner, T., et al., Aging theories of primary osteoarthritis: from epidemiology to molecular biology. Rejuvenation Res, 2004. 7(2): p. 134-45.11. Moriyama, H., et al., Effects of aging and exercise training on the histological and mechanical properties of articular structures in knee joints of male rat. Biogerontology, 2012. 13(4): p. 369-81.12. Verzijl, N., et al., Crosslinking by advanced glycation end products increases the stiffness of the collagen network in human articular cartilage: a possible mechanism through which age is a risk factor for osteoarthritis. Arthritis nad Rheumtism, 2002. 46(1): p. 114-23.13. Moshtagh, P.R., et al., Effects of non-enzymatic glycation on the micro- and nano-mechanics of articular cartilage. J Mech Behav Biomed Mater, 2018. 77: p. 551-556.14. Copray, J.C., H.W. Jansen, and H.S. Duterloo, Growth and growth pressure of mandibular condylar and some primary cartilages of the rat in vitro. American journal of orthodontics and dentofacial orthopedics, 1986. 90(1): p. 19-28.15. Mizoguchi, I., N. Toriya, and Y. Nakao, Growth of the mandible and biological characteristics of the mandibular condylar cartilage. Japanese Dental Science Review, 2013. 49(4): p. 139-150.16. Athanasiou, K.A., et al., Tissue engineering of temporomandibular joint cartilage. Synthesis Lectures on Tissue Engineering. Vol. 1. 2009: Morgan & Claypool Publishers. 1-122.17. Singh, M. and M.S. Detamore, Tensile Properties of the Mandibular Condylar Cartilage. Journal of Biomechanical Engineering, 2008. 130(1): p. 011009-011009.18. Mejersjo, C., Therapeutic and prognostic considerations in TMJ osteoarthrosis: a literature review and a long-term study in 11 subjects. Cranio, 1987. 5(1): p. 69-78.19. Tanaka, E., et al., Dynamic compressive properties of the mandibular condylar cartilage. J Dent Res, 2006. 85(6): p. 571-5.20. Singh, M. and M.S. Detamore, Stress relaxation behavior of mandibular condylar cartilage under high-strain compression. Journal of biomechanical engineering, 2009. 131(6): p. 0610081.

General discussion

99

6

21. Swann, A.C. and B.B. Seedhom, Improved techniques for measuring the indentation and thickness of articular cartilage. Proc Inst Mech Eng H, 1989. 203(3): p. 143-50.22. Kerckhofs, G., et al., Contrast-enhanced nanofocus computed tomography images the cartilage subtissue architecture in three dimensions. Eur Cell Mater, 2013. 25: p. 179-89.23. Saukko, A.E.A., et al., Dual Contrast CT Method Enables Diagnostics of Cartilage Injuries and Degeneration Using a Single CT Image. Annals of Biomedical Engineering, 2017.24. Lu, X.L., V.C. Mow, and X.E. Guo, Proteoglycans and mechanical behavior of condylar cartilage. Journal of Dental Research, 2009. 88(3): p. 244-8.25. Brommer, H., et al., Functional adaptation of articular cartilage from birth to maturity under the influence of loading: a biomechanical analysis. Equine Vet J, 2005. 37(2): p. 148-54.26. Mirahmadi, F., et al., Mechanical stiffness of TMJ condylar cartilage increases after artificial aging by ribose. Arch Oral Biol, 2018. 87: p. 102-109.27. Hu, K., et al., Regional structural and viscoelastic properties of fibrocartilage upon dynamic nanoindentation of the articular condyle. J Struct Biol, 2001. 136(1): p. 46-52.28. Patel, R.V. and J.J. Mao, Microstructural and elastic properties of the extracellular matrices of the superficial zone of neonatal articular cartilage by atomic force microscopy. Front Biosci, 2003. 8: p. a18-25.29. Yao, H. and W.Y. Gu, Convection and diffusion in charged hydrated soft tissues: a mixture theory approach. Biomechanics and modeling in mechanobiology, 2007. 6(1-2): p. 63-72.30. Kokkonen, H.T., et al., Computed tomography detects changes in contrast agent diffusion after collagen cross-linking typical to natural aging of articular cartilage. Osteoarthritis Cartilage, 2011. 19(10): p. 1190-1198.31. Kulmala, K.A., et al., Diffusion of ionic and non-ionic contrast agents in articular cartilage with increased cross-linking--contribution of steric and electrostatic effects. Med Eng Phys, 2013. 35(10): p. 1415-20.32. Hadley, J.C., K.M. Meek, and N.S. Malik, Glycation changes the charge distribution of type I collagen fibrils. Glycoconj J, 1998. 15(8): p. 835-840.33. Antipova, O. and J.P.R.O. Orgel, In Situ D-periodic Molecular Structure of Type II Collagen. Journal of Biological Chemistry, 2010. 285(10): p. 7087-7096.34. Honkanen, J.T.J., et al., Cationic Contrast Agent Diffusion Differs Between Cartilage and Meniscus. Ann Biomed Eng, 2016. 44(10): p. 2913-2921.35. Honkanen, J.T.J., et al., Transport of Iodine Is Different in Cartilage and Meniscus. Annals of Biomedical Engineering, 2016. 44(7): p. 2114-2122.36. Brama, P.A., et al., Influence of site and age on biochemical characteristics of the collagen network of equine articular cartilage. Am J Vet Res, 1999. 60(3): p. 341-5.37. Ruggiero, L., et al., Roles of the Fibrous Superficial Zone in the Mechanical Behavior of TMJ Condylar Cartilage. Ann Biomed Eng, 2015. 43(11): p. 2652-2662.38. van Saase, J.L., et al., Epidemiology of osteoarthritis: Zoetermeer survey. Comparison of radiological osteoarthritis in a Dutch population with that in 10 other populations. Annals of the Rheumatic Diseases, 1989. 48(4): p. 271-280.39. McIlwraith, C.W., et al., Equine Models of Articular Cartilage Repair. Cartilage, 2011. 2(4): p. 317-326.40. Frisbie, D.D., M.W. Cross, and C.W. McIlwraith, A comparative study of articular cartilage thickness in the stifle of animal species used in human pre-clinical studies compared to articular cartilage thickness in the human knee. Vet Comp Orthop Traumatol, 2006. 19(3): p. 142-6.41. Renders, G.A., et al., Contrast-enhanced microCT (EPIC-µCT) ex vivo applied to the mouse and

Chapter 6

100

human jaw joint. Dentomaxillofac Radiol, 2014. 43(2): p. 20130098.42. Bonin, S.J., et al., Comparison of mandibular motion in horses chewing hay and pellets. Equine Veterinary Journal, 2007. 39(3): p. 258-262.43. Lumpkins, S.B. and P.S. McFetridge, Regional variations in the viscoelastic compressive properties of the temporomandibular joint disc and implications toward tissue engineering. Journal of Biomedical Materials Research Part A, 2009. 90A(3): p. 784-791.44. Tanaka, E., et al., Stress relaxation behaviors of articular cartilages in porcine temporomandibular joint. Journal of Biomechanics, 2014. 47(7): p. 1582-7.45. Julkunen, P., et al., Biomechanical, biochemical and structural correlations in immature and mature rabbit articular cartilage. Osteoarthritis Cartilage, 2009. 17(12): p. 1628-38.46. Korhonen, R.K., et al., Importance of the superficial tissue layer for the indentation stiffness of articular cartilage. Medical Engineering & Physics, 2002. 24(2): p. 99-108.47. Maroudas, A., Biophysical chemistry of cartilaginous tissues with special reference to solute and fluid transport. Biorheology, 1975. 12(3-4): p. 233-48.

Appendices

SummaryNederlandse samenvatting

AcknowledgmentList of publications

About the author

Summary

104

Summary Articular cartilage has a unique structure, which provides a smooth and friction-free motion in joints. The biophysical properties of cartilage largely depend on its biochemical composition. During aging, the changes that occur in cartilage composition affect its functionality and consequently biophysical properties. Since life expectancy increases, age-related complications like cartilage degeneration and osteoarthritis occur more frequently. These conditions may cause pain, disabilities, and dependency on assistance during daily activities which lead to an impaired quality of life; furthermore, they lead to increased health-care costs. It is, therefore, essential to learn about age-associated changes that occur in the tissue and their effects on the biophysical properties of the tissue. Therefore, this thesis had two main aims. The first aim was to investigate the biomechanical, diffusive, and biochemical changes in temporomandibular joint (TMJ) condylar cartilage in which we mimicked aging by inducing collagen crosslinks. The second aim was to improve our understanding of the effect of normal aging on biophysical (mechanical and diffusive) and biochemical characteristics of TMJ condylar cartilage.

In order to perform biomechanical characterization, it was required to measure the thickness of the cartilage via a fast, reliable, and non-destructive method. Therefore, in the first experimental chapter of this thesis (Chapter 2), we developed an ex vivo thickness measurement method for condylar TMJ cartilage, using micro-computed tomography (micro-CT). A thickness distribution map obtained with this method showed that the anterior region of porcine condylar cartilage was the thinnest region of the mandibular condyle. The values obtained with micro-CT coincided with those obtained with histology. A third method, needle penetration, overestimated the thickness due to the penetration of the needle to the first layer of subchondral bone of young condyles. Thus, we used the micro-CT method for the samples analyzed in the next experimental chapters.

In Chapter 3, we induced an aging-like effect of increased collagen crosslinks in condylar cartilage of young pig samples with different concentrations of ribose. The amount of collagen crosslinks increased with ribose incubation; 50 times more crosslinks were found with the highest concentration of ribose. A positive correlation was found between the level of collagen crosslinks and compressive stiffness similar to those previously described for hyaline cartilage of other joints. The treatment also changed the orientation and packing of collagen fibers in the superficial layer of the cartilage due to interfibrillar bonds. This resulted in packing of collagen fibers next to open areas between fibers. Thus, this model, in which ribose was used to mimic certain aspects of age-related changes, might be employed as an in vitro model to study age-related mechanical changes in the TMJ condyle.

Summary

105

A

To elucidate the effect of crosslinking effect of natural aging on the diffusion, we induced an increased number of collagen crosslinks in a young equine cartilage samples in Chapter 4. The effect of an increased amount of collagen crosslinks on electrostatic interactions or steric hindrance was analyzed using a neutral and a negatively charged contrast agent. After ribose incubation of young equine TMJ samples, the crosslink level increased substantially; becoming even higher than what we observed in old equine samples (Chapter 5). As a consequence, the tissue stiffness increased 1.5 fold. Nevertheless, the diffusion changed neither for the neutral contrast agent nor for the negatively charged one. The results of this study strongly suggest that collagen crosslinking in TMJ condylar cartilage does not affect diffusion. However, it increases the stiffness for both equine and porcine (Chapter 3) TMJ fibrocartilage.

To investigate the effect of normal aging on TMJ condylar cartilage, we collected equine samples from animals of different ages. Biochemical and biophysical alterations of the mandibular condyle of different ages were assessed in Chapter 5. We showed that with aging, the diffusion of a neutral contrast agent decreased due to a decline in tissue water content, concomitant with an increase in collagen, crosslink, and collagen content. This confirmed the steric hindrance effect of aging-associated changes on the diffusion of neutral contrast agents. Surprisingly, diffusion of the negatively charged contrast agent was hardly affected. This could be explained by a remarkable change in the distribution of glycosmaninoglycans (GAGs) in the tissue matrix upon aging. It also might explain the unexpected results of tissue stiffness: stiffness did not change with aging, while the amount of collagen crosslink increased 15 fold. GAGs appear to contribute significantly to the mechanical response of cartilage; it seems that the change in GAG distribution reduced the effect of the increased amount of collagen crosslinks.

We conclude that the contribution of factors that change due to natural aging is not necessarily similar to artificially induced aging by increasing the amount of crosslinks. So it seems that mandibular condylar cartilage also in this respect differs from hyaline cartilage. Of the latter cartilage it was shown that a positive correlation exists between collagen crosslink number and stiffness during aging. Collectively, we improved our understanding of the effect of normal aging on biophysical (mechanical and diffusive) and biochemical characteristics of TMJ condylar cartilage.


106

Nederlandse samenvattingGewrichtskraakbeen heeft een unieke structuur die zorgt voor een soepele en wrijvingsvrije beweging van de gewrichten. De biofysische eigenschappen van kraakbeen hangen grotendeels af van de biochemische samenstelling. Tijdens het ouder worden beïnvloeden de veranderingen die optreden in de kraakbeensamenstelling de functionaliteit en zodoende ook de biofysische eigenschappen. Omdat de levensduur toeneemt, komen leeftijd gerelateerde complicaties zoals kraakbeendegeneratie en osteoartritis steeds vaker voor. Deze aandoeningen kunnen pijn, handicaps en hulp-afhankelijkheid veroorzaken bij dagelijkse activiteiten die vervolgens leiden tot een verminderde kwaliteit van leven. Bovendien leiden ze tot hogere kosten van de gezondheidszorg. Het is daarom essentieel om meer te weten over leeftijdsgebonden veranderingen die in het weefsel optreden en hun effecten op de biofysische eigenschappen van het weefsel. Daarom had dit proefschrift twee hoofddoelen. Het eerste doel was om de biomechanische, diffusie en biochemische veranderingen in het condylaire kraakbeen van het temporomandibulaire gewricht (TMG) te onderzoeken. Hierbij hebben we de veroudering nagebootst door collageen crosslinks te induceren. Het tweede doel was om inzicht te krijgen in het effect van normale veroudering op biofysica (mechanisch en diffusie) en biochemische kenmerken van TMG condylair kraakbeen.

Om biomechanische eigenschappen te onderzoeken was het nodig om de dikte van het kraakbeen te meten via een snelle, betrouwbare en niet-destructieve methode. Daarom is in het eerste experimentele hoofdstuk van dit proefschrift (Hoofdstuk 2) de ontwikkeling van een ex vivo werkwijze beschreven voor diktemeting van het TMG condylaire kraakbeen. Dit geschiedde met behulp van micro computed tomography (micro-CT). Met behulp van deze methode is een dikte-distributiekaart ontwikkeld die aantoonde dat het voorste gebied van varkens’ condylair kraakbeen het dunste gebied van de mandibulaire condylus is. De waarden verkregen met micro-CT kwamen overeen met waarden verkregen met behulp van histologie. Een derde methode, naaldpenetratie, overschatte de dikte als gevolg van de penetratie van de naald in de eerste laag subchondraal bot. Daarom hebben we de micro-CT-methode gebruikt voor de monsters die in de volgende experimentele hoofdstukken zijn geanalyseerd.

In Hoofdstuk 3 hebben we een verouderingsachtig effect geïnduceerd door het aantal collagene crosslinks te verhogen gebruikmakend van verschillende concentraties ribose. Onder deze omstandigheden werden 50 keer meer crosslinks gevonden met de hoogste concentratie van ribose. Een positieve correlatie werd gevonden tussen het niveau van collageen crosslinks en de kraakbeenstijfheid. Deze bevindingen zijn vergelijkbaar met die eerder zijn beschreven voor hyalien kraakbeen van andere gewrichten. De behandeling veranderde ook de oriëntatie en de organisatie van de collageenvezels


107

A

in de oppervlakkige laag van het kraakbeen als gevolg van interfibrillaire bindingen. Dit resulteerde in het samen klonteren van collageenvezels en open gebieden tussen de vezels. Dit model, waarin ribose werd gebruikt om bepaalde aspecten van leeftijd gerelateerde veranderingen na te bootsen, zou dus kunnen worden gebruikt als een in vitro-model om leeftijdsgebonden mechanische veranderingen in de TMJ-condyle te bestuderen.

Om het effect van crosslinking die optreedt bij natuurlijke veroudering op diffusie te verhelderen, induceerden we een verhoogd aantal crosslinks in collageen van kraakbeen afkomstig van jonge paarden (Hoofdstuk 4). Het effect van een verhoogde hoeveelheid collageen crosslinks op elektrostatische interacties en sterische hindering werd geanalyseerd met behulp van een neutraal en een negatief geladen contrastmiddel. Na ribose-incubatie van TMJ-kraakbeen van jonge paarden nam het aantal collagene crosslinks aanzienlijk toe; het aantal was zelfs hoger dan in kraakbeen van oude paarden (Hoofdstuk 5). Als gevolg hiervan nam de weefselstijfheid 1,5-voudig toe. Niettemin veranderde de diffusie niet voor het neutrale contrastmiddel noch voor het negatief geladen middel. De resultaten van dit onderzoek suggereren sterk dat collageen crosslinking in TMJ condylair kraakbeen de diffusie niet beïnvloedt. Echter, het verhoogt de stijfheid van het TMJ kraakbeen en dit geldt voor zowel paardachtigen als varkens (Hoofdstuk 3).

Om het effect van normale veroudering op TMJ condylair kraakbeen te onderzoeken, verzamelden we kraakbeen van paarden van verschillende leeftijden. Dit kraakbeen werd zowel biochemisch als biofysisch onderzocht in Hoofdstuk 5. We toonden aan dat met het ouder worden, de diffusie van een neutraal contrastmiddel daalt als gevolg van een daling van het watergehalte in het weefsel, gelijktijdig met een toename van het gehalte aan collageen, en aantal crosslinks. Dit bevestigde het sterische hinderingseffect van met verouderen geassocieerde veranderingen op de diffusie van neutrale contrastmiddelen. Verrassenderwijs werd de diffusie van het negatief geladen contrastmiddel nauwelijks beïnvloed. Dit kan worden verklaard door een opmerkelijke verandering in de verdeling van glycosaminoglycanen (GAG’s) in de weefselmatrix na veroudering. Het zou ook de onverwachte resultaten van weefselstijfheid kunnen verklaren: de stijfheid veranderde niet met het ouder worden, terwijl de hoeveelheid collagene crosslinks 15 keer toenam. GAG’s dragen aanzienlijk bij tot de mechanische respons van kraakbeen; het lijkt erop dat het effect van een opmerkelijke verandering in GAG-verdeling het effect van de toegenomen hoeveelheid collageen-crosslinks heeft verminderd.

We concluderen dat de bijdrage van factoren die veranderen als gevolg van natuurlijke veroudering niet noodzakelijkerwijs gelijk is aan kunstmatig geïnduceerde veroudering door de hoeveelheid crosslinks te verhogen. Het lijkt er dus op dat mandibulair condylair


108

kraakbeen in dit opzicht verschilt van hyalien kraakbeen. Van het laatste kraakbeen werd aangetoond dat er een positieve correlatie bestaat tussen het aantal collageen crosslinks en de stijfheid tijdens veroudering. Concluderend kan gesteld worden dat de studies beschreven in dit proefschrift een belangrijke bijdrage leveren aan onze kennis over het effect van de normale veroudering op biofysische (mechanisch en diffusie) en biochemische kenmerken van het TMG condylair kraakbeen.

Acknowledgement

109

A

AcknowledgmentThank you God for all your blessings to me, for the strength you give me each day, and for all the people around me who make life more meaningful.

After such an amazing and intensive journey that I began four years ago, today is the day! Writing my deepest thanks note is the finishing touch of this journey. I wouldn’t stand here without the support of so many people who helped me in different ways to reach this point.

Prof.dr. Vincent Everts, dear Vincent, I would like to express my sincere gratitude to you for your continuous guidance, for your patience, motivation, and immense knowledge. Your guidance helped me during all the time of research and writing of this thesis. Your super positive, energetic, and encouraging attitude made me always feel enthusiastic about every little thing I was doing. I am so grateful and lucky for having you as my promoter. Whenever I sent you my manuscripts, in the evening or even after midnight, your feedback was already in my inbox the next morning! I learned a lot from you how to extend my knowledge and to enjoy exploring unexpected findings. Thanks for all the skype meetings from Bangkok and even from your home. Thanks for everything.

Dr.ir. Jan Harm Koolstra, dear Jan Harm, I would like to express my deepest sense of gratitude for your support and encouragement throughout these four years. From the first skype interview we had, I felt your kindness. Whenever I knocked on the door of your office, you were always there to help me in both scientific and technical problems. I really appreciate your positive attitude and your guidance. Besides that, I won’t forget the first days when I was not familiar with a bike-dependent lifestyle in Amsterdam; you helped me to fix my bike flat tire and broken light. Thanks for everything.

Prof.dr. Frank Lobbezoo, dear Frank, many thank you for your continuous support, encouragement, and positive attitude during these years. Your invaluable thoughts and ideas helped me to improve my research a lot. I was so lucky having you in my promotion committee. After every meeting we had, you used to mention your positive thought about the order of my reports and presenting which made me feel happy and confident. I am so happy for having you as the chair of my promotion after you decided to quit the promotion committee to help me with administrative issues we had. I won’t forget this generous help in the last step of my PhD.

Prof.dr. Harry van Lenthe, dear Harry, thank you for your guidance, support, and inspiration. I really appreciate your engineering point of view and your insightful comments, also for the constructive criticism which incented me to widen my research from various perspectives. Thank you for your support and arrangement during my research in the Biomechanics section in Leuven.

Acknowledgement

110

Prof.dr. Sabine Verschueren, dear Sabine, special thanks for helping me to come up with the administrative issues I had about final registration at KU Leuven as a joint PhD student. I really appreciate your valuable support as well as your comments and input to my thesis.

Besides my promotion committee, I would like to thank the reading committee of my thesis: prof. Luyten, dr. Vanrenterghem, prof. Bank, prof. Becking, and prof. van Dieen, for their valuable time devoted reading my thesis and for their comments and suggestions. I gratefully acknowledge the funding sources that made my Ph.D. work possible. Special thanks to European Commission for granting a scholarship to me through MOVE-AGE, an Erasmus Mundus Joint Doctorate programme. Also, special thanks to Sjoukje for being so supportive and caring.

I would like to thank all my colleagues and friends at the Department of Oral Cell Biology and Functional Anatomy at ACTA for the pleasure to work with them. Geerling, thank you for teaching me the dissection of poor mice in the first weeks of my research. Thank you for your positive attitude and encouragement. Hans and Dirk Jan, thank you for teaching me histology techniques. Special thanks to Hans for helping me with the staining of a large number of sections for my last experiments. I really appreciate your patience with respect to my requests. Leo, thank you for helping me working with micro-CT and analyzing the images. Jolanda and Cor, thank you for always being there to help me in the lab. I really appreciate your supportive and helpful attitude that makes the lab a pleasant place to work. Behrouz, many thanks for always being helpful in everything. You have a golden heart for helping everybody. I really enjoyed all the BBQ you arranged at your house every year. Many thanks also go to Marion for helping me with microscopy problems whenever I needed help. Astrid, Sue, Cees, Jenneke, Ineke, Ton, Teun, and Clara, thank you for your advice during Monday meetings and also for nice chats during lunches and drinks.

I would also like to extend my gratitude to Kamran for his great technical support and generous help whenever I had a problem, especially in biochemical analyses. Thank you for your advice, and recommendations not only for the experiments but also for my questions regarding my stay in the Netherlands.

I would like to express my gratitude to Jessica and Rainout for their help and advice I received through collaboration with TNO group. Many thanks go to Prof. Weinans, dear Harrie, and Vahid for their guidance during our collaboration for my last experiment. I also would like to thank the animal facility group in the UPC for their help in the dissection of pig heads as well as the Faculty of Veterinary Medicine, the University of Utrecht for providing the horse samples. I also would like to thank Albert van der Veen for his technical help.

Acknowledgement

111

A

Special appreciation to people from the Biomechanics section in the Mechanical Engineering Department, KU Leuven. Karen, Ivo, Stefan, Rita, and Sepideh, thank you for your kind help and support during my stay in Leuven.

Rachel, my first Dutch friend, thank you for being so nice and friendly to me. I will never forget my first days in Amsterdam when you helped me with buying a bike, showing me the city, and sharing your experience living here. Wish you all the best in the rest of your PhD in Switzerland.

My Chinese roomies at the 12th floor, Xingnan and Fei, thank you for sharing your PhD life with me. Fei, we shared the lab in the first weeks you had arrived. I enjoyed all the chats we had about our culture and our hobbies. You are so determined and hardworking and I wish you all the best in finishing your PhD.

Sepanta, thanks for all your support and help from the beginning of my PhD even before I arrived in the Netherlands. I remember the first email I sent you to ask some questions about living in the Netherlands that you replied very quickly with long precise and detailed information. I would never forget all the chats we had during dissections and experiments together, and during dinner at the office, late in the evenings. Thank you for your great help when I was moving to my new apartment. Thank you for your helpful spirit and for your sense of humor which made doing exhausting experiments much easier. Whenever I see or hear something about “Aghuye hamsade”, I will definitely remember you! I wish you all the best in your new job.

Carolyn, thank you for being my best Dutch friend. Thank you for all warm and friendly chats, supports, and helps. I was very lucky having you as one of my roomies when I came to the 11th floor. You are such a warm and caring person that made you my best Dutch friend soon. I won’t forget all the lunches and dinners we had together. I am always impressed by your motivations and your ambitions. I will definitely miss the Champagne tea and coffee you made for us in the office when I start my next career. Thank you for the very beautiful and elegant gift for my birthday, and thank you for arranging very interesting activities with other friends together. Special thanks that you accept to be my paranymph. I am so happy having you as my paranymph. I am looking forward to sharing more joyful moments with you in future and I wish you the best in the rest of your PhD.

Many thanks also go to my roomies at the 11th floor. Bea, thank you for bringing energy and fun to our office. We had a lot in common about our projects and their difficulties; the simplest one was pig’s head dissection! I always enjoy talking with you about everything. I wish you all the best with finishing your PhD and passing BIG exam. Angela, thank you for all the nice chats we had about our families, for nice lunches we had together, and for all your advice which always helped me. I wish you and your family all the best

Acknowledgement

112

and success in finishing your PhD. Tijmen, thank you for being always supportive and helping with software questions. Thank you for the nice gift for my birthday. I wish you all the best for the rest of your PhD. Sophie, you always bring a lot of energy to the office. I wish you and your cute daughter the best. Jianfeng, thank you for the tasty noodle you made for my birthday and for all the cookies you brought from China for us. I wish you success with the rest of your PhD. Sara, thank you for being always positive and cheerful. I enjoyed our chats during lunches and drinks. Thanks for your nice gift for my birthday. Yixuan, having such a nice Chinese girl next to my desk was a great pleasure for me. Thank you for being always positive and optimistic. The note and the little angel statue that you left on my desk after a long experiment will remain in my memory. I wish you all the best in your life. Mahshid, thank you for all the chats during lunch and coffee time we had together. I do not forget the dinner we had together with Fatima at Delft. Yvon, thank you for being always nice and cheerful. I am so happy sharing the last months of my PhD with you. Thank you for generous sharing of your Christmas gift with me.

Yasaman, we shared a lot when you came to ACTA and before that, when we worked at Pasteur Institute. We have a lot of amazing memories in common and I am looking forward to having more and more in the future.

Haniyeh, thank you for being a loyal friend for more than 10 years. We shared a lot of joyful moments in Iran, Amsterdam, and Leuven. Thank you for all the good times we spent together. Staying in Leuven was not pleasant without you and your lovely mother ‘Giti khanoom’. I wish you and Farshid the best in moving to Australia and I am looking forward to visiting you there.

Samaneh, I remember the first day I came to ACTA I visited you discussing scientific issues with Sepanta. You are not only a very creative and hardworking scientist but also a very nice and lovely friend. I am so proud of having such an amazing friend. Thank you for the dinner we had together, for your advice on my personal life besides your scientific suggestions. Special thanks to you and your lovely sister Sara for helping me when I was moving to my new apartment. I am always impressed with your determined spirit and your motivation. You deserve more than what you have now and I wish you all the best in your scientific and personal life.

Rozita, thank you for good time we had together, for nice chats and BBQ we had together with your husband Masoud. I am looking forward to having more fun together.

I also would like to take the opportunity to thank all previous and present colleagues and friends in ACTA for all lunches, drinks, and days out. A big thank you to Thijs, Silvia, Cindy, Ivana, Nero, Francis, Patrick, Dagma, Nawal, Ceylin, Lin, Yi, Ying, Morvarid, Alireza, Hessam, Janak, Haroon, Ana, Rosaline, Alejandra, Sabrine, and Mattijs. To mention and thank everyone here, I should have an acknowledgment chapter as long as this thesis.

Acknowledgement

113

A

Therefore I would like to thank, as well as express my apologies to everyone whose name I did not mention.

A true friend enlightens entire life. My best and closest friends ever, Neda joonam and Parisa joonam, thank you for being ‘brightest’ light of my life. I can say that this journey started because of you. Maybe, I would never think of coming to the Netherlands if you were not here. I cannot believe that our friendships became 13 years old! We spent all of these years together with lots of joys, ups and downs. Thank you both for all we had together; our memories are countless and I am so happy and proud of having such exceptional and amazing friends like you. I am the luckiest to have my best friends close to me in the Netherlands. Neda joonam, special thanks that you accept to be my paranymph. I am so happy having you as my paranymph.

Parisa and Amin, my coolest neighbors and kind friends, thank you for always being so nice and kind to me. You made it easier for me to stand the dim and cloudy days away from my family. Thank you for all the amazing travels and biking trips, evening walking, volleyball and badminton playing, table games, Yalda night celebration, and countless cheerful gathering we had together during the past 3 years. I enjoyed each moment we had together and I’m looking forward to having more fun in future.

I would also like to thank all my Iranian friends in the Netherlands which made me feel home: Mina and Hamid, Bita and Ali, Negar, Mahnaz, Nazila, Vida and Kamal, Hamed, Erfan, Haleh and Babak, Samira and Mostafa, Elham and Seyed Esmaeal, Marzi and Hodjat, Fatemeh and Amir, Fatemeh and Amin, Fatemeh and Hamed, Zahra, Niloofar, Masoud, Hoda, Mojtaba, Mehran, Keyone, Hadi, Hosein, Ali, and Mohamadreza. Thanks for all the gathering and lots of fun we had together.

Many thanks to my friends all over the world: Melika jan and Ehsan, En’am, Zahra, Sajedeh, Maryam, Bharak, Mehdi, Mahtab, Hosein, Fatemeh, and Nooshin; thank you for your continuous support and love.

I would like to express my sincere gratitude to my husband’s family: Khale Maryam jan, dr. Rahimi, Maryam, and Ali; thank you for your love, support, constant encouragement, and positive energy.

My especial thanks go to my parents and siblings. Maman and Baba, I am never able to count all you did for me. I would not be here without your unconditional love, support, and guidance. You always encouraged me to keep going on even after my failures. You learned me that there are no failures but there is something new to learn. I consider myself so lucky to have you and I love you so much. Having a big family is really amazing and as the youngest girl in the family, I always received endless love and support throughout my life. Alireza, Maryam, Mali, Zohreh, Abbas, Fatemeh, and Farhad, I’m not able to thank you for your exceptional support and love during my life. I love you so

Acknowledgement

114

much. Fatemeh, I cannot imagine myself without you; having a twin sister is a special gift from God, and I’m blessed with having such a nice gift. After whole life sharing our room together, it took me long tough time to get used to living in a room without you. Thank you for your continued companion and pure love, Ghol jan.

مامان و بابای عزیزم، مهر و عشق و محبت شام همیشه همراهم بوده و هست. با هیچ جمله ای منیتونم از زحامت شام

قدردانی کنم. طی کردن این مسیر بدون عشق و راهنامیی و حامیتهای بی دریغ شام ممکن نبود. همه ی چیزهایی که

دارم بخاطر وجود شامست. دستتون رو میبوسم و امیدوارم که منو بخاطر رنج دوری که متحمل شدید ببخشید

Last but not the least, Yaser janam, my beloved, you brought your kindness and calmness in my life. You step in my life and my life got more colorful days. You always know how to make me happy even when I am the most upset one in the world. Without your love and faithful support, I could not finish this journey. I am so happy, lucky, and proud of having you by my side in every moment of my life. I love you so much.

Bedankt سپاس

Thank you

Fereshteh

October 2018

List of pulications

115

A

List of publicationsMirahmadi, F., Koolstra, J. H., Lobbezoo, F., van Lenthe, G. H., & Everts, V. (2017). Ex vivo thickness measurement of cartilage covering the temporomandibular joint. Journal of biomechanics, 52, 165-168.

Mirahmadi, F., Koolstra, J. H., Lobbezoo, F., van Lenthe, G. H., Ghazanfari, S., Snabel, J., Stoop, R., Everts, V. (2018). Mechanical stiffness of TMJ condylar cartilage increases after artificial aging by ribose. Archives of oral biology, 87, 102-109.

Mirahmadi, F., Koolstra, J. H., Lobbezoo, F., van Lenthe, G. H., Ghazanfari, S., Snabel, J., Stoop, R., Arbabi, V., Weinans, H., Everts, V. Aging does not change the compressive stiffness of mandibular condylar cartilage in horses. Accepted for publication in Osteoarthritis and Cartilage; in press.

Mirahmadi, F., Tafazzoli-Shadpour, M., Shokrgozar, M. A., & Bonakdar, S. 2013. Enhanced mechanical properties of thermosensitive chitosan hydrogel by silk fibers for cartilage tissue engineering. Materials Science and Engineering: C, 33(8), 4786-4794.

Solouk, A., Cousins, B.G., Mirahmadi, F., Mirzadeh, H., Nadoushan, MRJ., et al. 2015. Biomimetic modified clinical-grade POSS-PCU nanocomposite polymer for bypass graft applications: A preliminary assessment of endothelial cell adhesion and haemocompatibility. Materials Science and Engineering: C 46 (30), 400-408.

Jazayeri, M., Shokrgozar, MA., Haghighipour, N., Bolouri, B., Mirahmadi, F., Farokhi, M.; 2017. Effects of Electromagnetic Stimulation on Gene Expression of Mesenchymal Stem Cells and Repair of Bone Lesions, Cell Journal, V 19(1), 34-44

Nematollahi Z, Tafazzoli-Shadpour M, Zamanian A, Seyedsalehi A, Mohammad-Behgam Sh, Ghorbani F, Mirahmadi, F.; 2017. Fabrication of Chitosan Silk-based Tracheal Scaffold Using Freeze-Casting Method, Iranian Biomedical Journal. V 21(4), 228-239

Solouk, A., Cousins, B.G., Mirahmadi, F., Mirzadeh, H., Nadoushan, M.R.J., Shokrgozar, M.A., Seifalian, A.M., 2015. Biomimetic modified clinical-grade POSS-PCU nanocomposite polymer for bypass graft applications: A preliminary assessment of endothelial cell adhesion and haemocompatibility. Materials Science and Engineering: C 46, 400-408.

Jazayeri, M., Shokrgozar, M.A., Haghighipour, N., Mahdian, R., Farrokhi, M., Bonakdar, S., Mirahmadi, F., Abbariki, T.N., 2015. Evaluation of Mechanical and Chemical Stimulations on Osteocalcin and Runx2 Expression in Mesenchymal Stem Cells. MCB: Molecular & Cellular Biomechanics 12, 197-213.

Mottaghitalab, F., Farokhi, M., Zaminy, A., Kokabi, M., Soleimani, M., Mirahmadi, F., Shokrgozar, M.A., Sadeghizadeh, M., 2013. A biosynthetic nerve guide conduit based on silk/SWNT/fibronectin nanocomposite for peripheral nerve regeneration. PloS one 8,

List of pulications

116

e74417.

Mirahmadi, F., Koolstra, J. H., Lobbezoo, F., van Lenthe, G. H., Ghazanfari, S., Snabel, J., Stoop, R., Everts, V. Diffusion of charged and uncharged contrast agents in equine mandibular condylar cartilage is not affected by an increased level of sugar-induced collagen crosslinking. Manuscript under review.

About the author

117

A

About the authorFereshteh Mirahmadi was born in Najafabad, Iran, on 9th of February, 1986. She started her bachelor in 2005 in Biomedical Engineering-Biomaterials, at Amirkabir University of Technology (Tehran Polytechnic), Iran. As a top student of her bachelor, she got admission to choose a second field of study in 2007. Accordingly, she started her second bachelor in Polymer Engineering. In 2009, she received her first bachelor degree in Biomedical Engineering. Thereafter, she got accepted for the masters graduate program of the Biomedical Engineering Faculty. In 2011, she obtained both her Master in Biomedical Engineering-Tissue Engineering and her second Bachelor in Polymer Engineering. During her master program, she did a one-year research internship at the tissue engineering lab at the Pasteur Institute of Iran, where she worked as a research assistant after her graduation from 2012 until 2014. In October 2014, she moved to the Netherlands and started her PhD, granted by the European Commission through MOVE-AGE, an Erasmus Mundus Joint Doctorate programme. For her PhD, she worked at the Academic Centre for Dentistry Amsterdam (ACTA), Vrije Universiteit Amsterdam, and at the Biomechanics section of the Department of Mechanical Engineering, Katholieke Universiteit (KU) Leuven, Belgium. During her PhD, she collaborated with the TNO (Leiden), and with the Orthopedic Department of the Utrecht University Medical Center (UMC). The results of her PhD research are presented in this thesis.

research.vu.nl dissertation.pdf · VRIJE UNIVERSITEIT. KU LEUVEN. Biomechanical changes in...

Documents

Transcript of research.vu.nl dissertation.pdf · VRIJE UNIVERSITEIT. KU LEUVEN. Biomechanical changes in...