A THERAPEUTIC DOSE OF ADENOSINE TRIPHOSPHATE (ATP) FOR CARTILAGE TISSUE ENGINEERING

Transcription

1 A THERAPEUTIC DOSE OF ADENOSINE TRIPHOSPHATE (ATP) FOR CARTILAGE TISSUE ENGINEERING by Jenna Felice Usprech A thesis submitted to the Department of Mechanical and Materials Engineering In conformity with the requirements for the degree of Master of Applied Science Queen s University Kingston, Ontario, Canada (September, 2010) Copyright Jenna Felice Usprech, 2010

2 Abstract Tissue engineering holds great promise for developing functional tissue to repair damaged articular cartilage. However, cartilaginous tissues formed in vitro typically possess poor mechanical properties in comparison to native tissue due to the inability of chondrocytes to accumulate adequate amounts of extracellular matrix (ECM). While mechanical stimuli can enhance the properties of engineered cartilage, it may be more efficient to harness the underlying mechanotransduction pathways responsible. Substantial evidence suggests that the mechanotransduction signaling cascade is initiated by rapid adenosine 5 -triphosphate (ATP) release from chondrocytes (purinergic receptor pathway). Thus, the purpose of this study was to investigate the effects of exogenous ATP supplementation on the biochemical and mechanical properties of tissue engineered cartilage. Primary bovine articular chondrocytes, seeded on Millipore TM filters, were grown in the presence of 0, 62.5 or 250 µm ATP for a period of four weeks. Both anabolic and catabolic effects were examined and a therapeutic dose range of ATP was determined. ATP stimulation ( µm) enhanced ECM synthesis by 23-43% and long-term collagen accumulation by 16-26%, in a dose-dependent manner; however, long-term proteoglycan accumulation decreased as a result of 250 µm ATP. In addition, ATP supplementation significantly improved the mechanical properties of the developed ii

3 tissues (5- to 6.5-fold increase in tissue stiffness). Interestingly, high doses of ATP (250 µm) also elicited a 2-fold increase in MMP-13 gene expression, 39% increase in MMP- 13 activity, and 54% more extracellular inorganic pyrophosphate (eppi) an ATP degradation product. Dose-dependent increases in MMP-13 activity suggested that catabolic effects were occurring alongside anabolic effects, which initiated the investigation of a therapeutic dose of ATP. Doses of µm and 125 µm ATP were added to cartilaginous tissues and investigated in terms of MMP-13 activity and ECM synthesis. Tissues supplemented with µm ATP exhibited a balance between anabolic and catabolic effects. By harnessing the purinergic receptor pathway, anabolic effects of mechanical stimuli were achieved in the absence of externally applied forces. Understanding how the catabolic effects of ATP are manifested would be valuable in order to further maximize the therapeutic potential of ATP stimulation. iii

4 Acknowledgements I d like to thank everyone who has supported me academically, personally and financially during my pursuit of a masters degree. I feel grateful to know that all of the knowledge and skill I ve gained cannot possibly be transferred into the pages of this document, but will remain with me for years to come. Thank you NSERC for recognizing my potential at the final hour and funding my research. Dr. Stephen Waldman, thank you for your excellent supervision and dedication throughout this thesis project and for believing in my abilities when I doubted them myself. You are definitely the master of a pep talk. Thank you to Dr.Yat Tse, Dr. Lauren Flynn, Richard Casselman and Jake Kaupp for your invaluable help with PCR Thank you, Jackie Fan, for taking me on that inaugural trip to the slaughterhouse and for transferring your knowledge of mechanical testing to me. Thank you summer/ 4 th year students Kathleen Martin, Gavin Chu and Kate Colizza for helping this project take on a mind of its own. iv

5 Thank you, Jill Brenner. Your presence in the lab has kept my spirits up, even in the most frustrating times. Thank you to my friends, roommates, co-workers and lab-mates especially to those of you who hold multiple titles. You have all made my time at Queen s, and in Kingston, unforgettable. James Hayami, I don t know how I could have gone through this process without your continued love, support and advice. You make me a much stronger person than I ever thought I could be. Lastly, thank you to my family. They would support me in anything and for that I am truly lucky. v

6 Table of Contents Abstract... ii Acknowledgements... iv Chapter 1 Introduction Articular Cartilage Damage and Degradation Strategies for Articular Cartilage Repair Cartilage Tissue Engineering Mechanical Stimulation and Mechanotransduction Case for Exogenous ATP Supplementation Research Outline... 6 Chapter 2 Literature Review Structure and Function of Articular Cartilage Extracellular Matrix Water Collagens Proteoglycans Chondrocytes Zonal Organization Cartilage Biomechanics Biphasic Nature Viscoelasticity Mechanical Stimulation Anabolic versus Catabolic Responses Effect of Loading Mode Compression Tension Shear Hydrostatic Pressure Mechanotransduction vi

7 Integrins Ion Channels Autocrine/Paracrine Signaling Purinergic Receptor Pathway Regulation of ATP Signaling ATP release ATP binding ATP degradation Effects of Exogenous ATP on Cartilage and Chondrocytes Matrix Turnover Inflammation Inorganic Pyrophosphate Homeostasis Mineralization Chapter 3 Materials and Methods Cartilage Tissue Culture Cell Isolation High Density 3D Culture Adenosine Triphosphate (ATP) Supplementation Extracellular Matrix (ECM) Synthesis P2 Receptor Inhibition Tissue Assessment Tissue Harvest Tissue Wet/Dry Weights Tissue Digestion Extracellular Matrix (ECM) Accumulation DNA Assay GAG Assay Hydroxyproline Assay Thickness vii

8 3.4.6 Mechanical Properties Histology and Immunohistochemistry Transmission Electron Microscopy (TEM) Gene Expression Genes of Interest Primer Design and Optimization RNA Isolation Reverse Transcription End-point Polymerase Chain Reaction (PCR) Evaluation of PCR Products MMP-13 Protein Activity Protein Extraction Fluorescent Resonance Energy Transfer (FRET) assay Measurement of Inorganic Pyrophosphate (PPi) Quantification of Inflammatory Mediator Release Prostaglandin E 2 (PGE 2 ) Assay Nitric Oxide (NO - ) Assay Optimization of ATP Dose Statistical Analyses Chapter 4 Results Effect of exogenous ATP on extracellular matrix (ECM) synthesis Confirmation of purinergic receptor signaling Long-term effects of ATP supplementation on tissue formation and properties Catabolic matrix turnover gene expression and enzyme activity Inorganic pyrophosphate homeostasis Release of inflammatory mediators Optimization of ATP dose Chapter 5 Discussion Purinergic receptor mediated synthetic effects of exogenous ATP viii

9 5.2 Long-term effects of ATP supplementation on tissue formation and properties Long-term effects of ATP supplementation on catabolism, inflammation and mineralization The balance of anabolic and catabolic responses and the optimization of ATP dose A possible mechanism for exogenous ATP action on the development of cartilaginous tissues Chapter 6 Conclusions and Recommendations Conclusions Overall perspective Anabolic versus catabolic effects A therapeutic dose range of ATP for cartilaginous tissues Recommendations References Appendix A: Transmission Electron Microscopy Appendix B: Tissue Thickness Measurements ix

10 List of Figures Figure 2-1: Side view and cross-section of a knee joint... 9 Figure 2-2: Representation of a proteoglycan aggregate Figure 2-3: Cross section of articular cartilage representing the zonal organization Figure 2-4: Representation of the in vivo mechanical environment of articular cartilage under joint loading and motion Figure 2-5: Multiple pathways through which cells detect mechanical deformation Figure 2-6: A schematic representation of the regulation of ATP signaling Figure 4-1: Effect of exogenous ATP on collagen and proteoglycan synthesis Figure 4-2: Effect of exogenous ATP (250 µm) on collagen and proteoglycan (PG) synthesis in the presence or absence of Reactive Blue 2 (100 µm) Figure 4-3: Effect of long term (4 week) ATP supplementation on collagen and proteoglycan accumulation Figure 4-4: Histological assessment of the developed cartilaginous tissues after 4 weeks of ATP supplementation Figure 4-5: Immunohistochemical assessment of collagen types II and I in the developed cartilaginous tissues after 4 weeks of ATP supplementation Figure 4-6: Semi-quantitative end-point PCR results for the expression of catabolic matrix turnover genes Figure 4-7: Effect of long term (4 week) ATP supplementation on MMP-13 activity Figure 4-8: Effect of long-term ATP stimulation on inorganic pyrophosphate levels in the conditioned culture media surrounding the developed cartilaginous tissues Figure 4-9: Effect of long-term ATP stimulation on inorganic pyrophosphate levels in the developed cartilaginous tissues Figure 4-10: Effect of long term ATP supplementation on the release of inflammatory mediators Figure 4-11: ECM synthesis results for the optimization of ATP dose Figure 4-12: MMP-13 activity optimization x

11 Figure 4-13: Changes in the anabolic and catabolic effects of exogenous ATP on engineered cartilage constructs xi

12 List of Tables Table 3-1: Values used to interpolate for correction factor K Table 3-2: Oligonucleotide primers used for semi-quantitative end-point PCR Table 3-3: Thermal cycler program for PCR Table 4-1: Effect of exogenous ATP on cartilaginous tissue formation and properties.. 58 xii

13 Chapter 1 Introduction 1.1 Articular Cartilage Damage and Degradation Articular cartilage covers the ends of articulating bones in order to provide a near frictionless surface for joint motion while acting as a shock absorber. Despite being able to withstand large, repetitive mechanical loads, cartilage has a very limited capacity for self-repair primarily due to a complete lack of blood supply. Injuries to cartilage typically occur from a single excessive load or from prolonged joint overloading [1]. Such injuries can result in the development of osteoarthritis (OA) the degradation of the extracellular matrix (ECM) and shutdown of various cartilage-protective mechanisms [2, 3]. Additionally, old age and a poor diet can affect cartilage biochemistry and contribute to cartilage degeneration and OA [4]. OA is characterized by changes in chondrocyte shape, an increase in tissue degrading enzymes and osteophytes (bony nodules) as well as increased cell death [5]. OA affects approximately 3 million Canadians, who experience pain and limited mobility as a result of the disease [6]. 1.2 Strategies for Articular Cartilage Repair Repairing cartilage therapeutically is also challenging due to the lack of appropriate donor tissue, development of cartilage that has inferior mechanical properties and propensity for re-injury. Strategies that currently exist to repair cartilage include: (i) cartilage debridement and lavage, (ii) induced chondrogenesis, (iii) tissue grafting and 1

14 (iv) autologous chondrocyte implantation (ACI). Lavage and debridement are minor corrective treatments aimed at reducing the pain associated with cartilage injury. Lavage is the irrigation of a joint with sodium chloride [7] and debridement is the process of shaving away damaged or diseased regions of cartilage [8]. Despite being successful in the short-term, debridement creates wounds that lead to apoptosis (programmed cell death) and tissue necrosis which impair the properties of cartilage in the long-term [9]. Inducing chondrogenesis involves penetrating the subchondral bone to initiate a wound healing response which will fill in the defected area [10]. Chondrogenesis is induced via abrasion arthroplasty, drilling and microfracture [11]. Consequently, the regenerated tissue is more fibrous, has characteristics of scar tissue, and lacks the unique mechanical properties of articular cartilage [7]. Tissue grafting or osteochondral transfer, on the other hand, involves fitting in osteochondral grafts (one or several cylindrical plugs of cartilage attached to bone) into the defect site [10]. The grafts are obtained from either the patient s own cartilage (autograft), in non-load-bearing regions, or from cadavers (allografts). Both autografts and allografts have advantages and disadvantages associated with their transplantation. An autograft does not promote an immunogenic response since it comes from the patient, in contrast to an allograft, which can be rejected by the patient s body and/or transmit disease [12, 13]. However, since autografts are taken from healthy cartilage, damage is inevitably inflicted at the donor site. Donor site cartilage is also prone to degeneration or damage from overloading in the transplant site, since it comes from a non-load-bearing region [7]. Allogenic grafts can be taken from load bearing regions of cartilage, which would have similar properties to that of the transplant 2

15 site. As long as the allogenic tissue is transplanted fresh, these properties are better maintained as cryopreserved tissue leads to the death of chondrocytes, contributing to cartilage deterioration and loss of mechanical integrity [14]. Transplanted tissue grafts, regardless of their origin, are subject to damage once they are implanted. This is due to a lack of mechanical support from the surrounding tissue, zones of death created by drilling, and harmful compressions sustained during surgical hammering [7]. ACI involves directly implanting cultured chondrocytes, obtained from a biopsy of the patient s cartilage, into the defect site [13]. A periosteal flap (connective tissue that surrounds long bones) is sutured over the site to hold the chondrocytes in place [13]. ACI is not a suitable method of treatment for OA and is recommended for medium-sized defects [11]. Even with the numerous methods of repair, there is currently no treatment that effectively restores the full functionality of articular cartilage. 1.3 Cartilage Tissue Engineering Tissue engineering is a strategy that has the potential to solve issues related to cartilage degeneration and repair. It involves fabricating living tissue replacements for implantation into the body [12]. Even though there are many areas of focus within the field, tissue engineering is based upon three fundamental components: cells, scaffolds and signals [12]. Mesenchymal stem cells show promise as a multipotent cell source that can differentiate into chondrocytes provided that the specific differentiation factors are present [15]. However, mesenchymal stem cells are almost exclusively found in bone marrow in low quantities, making them more difficult to obtain [1, 16]. Chondrocytes, 3

16 however, are a more plentiful cell source, but they have a limited proliferative potential [13]. Scaffolds are of equal importance to the cell source, since they provide the structural framework which immobilizes cells in a 3D environment in order to maintain the differentiated phenotype of chondrocytes [16]. Preservation of chondrocyte shape is also important for ECM synthesis and accumulation, which is why 3D scaffolds develop tissue that more closely resembles native cartilage tissue in comparison to chondrocytes grown in 2D monolayer [13, 17, 18]. Good scaffold cell seeding, indicated by uniformly distributed cells, is also required for the formation of functional cartilage [13]. Scaffolds can be made from both synthetic (poly (glycolic acid), poly (lactic acid)) and natural (collagen, hyaluronate, fibrin, alginate, chitosan) polymers that are biocompatible [19]. Signaling factors traditionally include cytokines, growth factors, and other regulatory molecules that modulate cellular metabolism [20]. Since cartilage depends on mechanical forces for the development and maintenance of functional tissue, physical forces are also critical signaling factors. Tissue engineering, like many attempts at repairing cartilage, is not without its share of problems. Typically, tissue engineered constructs exhibit inferior mechanical properties in comparison to native articular cartilage tissue due to the insufficient accumulation of matrix components [21-24]. Issues that also need to be addressed include implant fixation, integration of the native and implanted tissues and uniformity in mechanical properties between the transplanted and host tissues [16]. 4

17 1.4 Mechanical Stimulation and Mechanotransduction Chondrocytes require an appropriate mechanical environment in order to produce cartilage with the ability to withstand the typical forces within a joint [2]. Applying mechanical loads in vitro has been shown to improve the quality of developed cartilage through enhancements in ECM synthesis and cellular proliferation [22, 24-28]. Forms of mechanical stimulation that have been applied to cartilage include: compression, tension, shear, vibration, fluid flow and hydrostatic pressure. Regardless of the type, mechanical stimulation triggers mechanotransduction cascades which link the application of external stimuli to the production of chemical signals inside the cell. These chemical signals can then elicit changes in cellular metabolism and biochemistry. A molecule that appears to be integral to one chondrocyte mechanotransduction pathway is adenosine triphosphate (ATP) (termed the purinergic receptor pathway). ATP signaling has been implicated in mechanotransduction pathways in many different cell types including epithelial cells, astrocytes, osteoblasts and chondrocytes [29]. In articular chondrocytes, ATP is the first known molecule to be released as a result of mechanical stimulation [30]. ATP can then bind to receptors on the chondrocyte membrane to trigger anabolic or catabolic effects Case for Exogenous ATP Supplementation Despite the benefits of mechanical stimulation, it can be difficult to do so in a highly repeatable fashion, due to limitations by the size and shape of tissue constructs. Anatomical shapes would pose a particularly difficult challenge for the effective transmission of the different types of forces that are often applied in bench-scale 5

18 investigations [31-33]. Since ATP is a critical part of a mechanotransduction cascade, supplementing ATP to the engineered tissue as it grows could be a promising alternative to the traditional contact-based methods of mechanical stimulation. However, exogenous ATP can have a variety of undesirable effects on chondrocytes which include matrix turnover, release of inflammatory mediators, dysregulation of inorganic pyrophosphate levels and the initiation of matrix mineralization. The effect of ATP on tissue development appears to be dose-related, with lower doses yielding more anabolic responses and higher doses yielding catabolic responses. The extent of this doseresponse relationship, however, is unknown. 1.5 Research Outline The purpose of this study was to harness the anabolic effects of the mechanotransduction cascade, in the absence of externally applied forces, by supplementing engineered cartilage cultures with ATP. This study included a detailed investigation on both the positive and negative effects of ATP on cartilage tissue development in vitro with a focused attention on a therapeutic dose range for ATP. The objectives of the thesis are to: 1) determine doses of ATP that elicit an anabolic tissue response, 2) determine doses of ATP that elicit a (negative) catabolic, inflammatory and/or mineralization response and 3) determine an optimal, therapeutic dose of ATP for engineered cartilage cultures. 6

19 In order to achieve the first objective, engineered cartilage cultures were grown in the presence of high and low doses of ATP and were assessed in terms of biochemical and mechanical properties. Histology and immunohistochemistry were performed in order to visualize tissue components and mineralization. The following experiments were performed in order to achieve the second objective: expression of matrix turnover genes, catabolic enzyme activity assays (MMP-13) and assays for the quantification of inflammatory mediator release and inorganic pyrophosphate an ATP degradation product. Lastly, the final objective was achieved by growing the engineered cartilage cultures with a finer range of ATP concentrations in order to identify a therapeutic dose range. 7

20 Chapter 2 Literature Review 2.1 Structure and Function of Articular Cartilage Articular cartilage allows diarthroidal joints to function optimally by providing sufficient lubrication for joint motion and cushioning from loading (Figure 2-1). Synovial fluid bathes the surface of the joint, supplying the necessary lubrication while also acting as a nutrient source [34]. Nutrient exchange is restricted to the diffusion of small molecules from the surrounding synovial fluid or underlying bone because cartilage is avascular [35]. These properties severely limit the ability of cartilage to repair itself, which poses a significant problem when cartilage is injured. The material properties of articular cartilage have yet to be matched by synthetic materials or replicated in the laboratory due to the unique structure of articular cartilage [36]. Cartilage is arranged in distinct zones consisting of a small fraction of cells, called chondrocytes, surrounded by a hydrated network of structural proteins that make up the extracellular matrix (ECM). 8

21 Figure 2-1: Side view and cross-section of a knee joint [37] Extracellular Matrix For the most part, the ECM of cartilage consists of water, collagens and proteoglycans. Other non-collagenous structural proteins, lipids, phospholipids and glycoproteins are present in the ECM, but to a much lesser degree [34, 38]. The ECM exhibits both structural and informational functions, providing the tissue with mechanical support in addition to an avenue for cell signal communication [1] Water Water accounts for 75-80% of the total wet weight of articular cartilage and accordingly contributes to a variety of functions [39, 40]. Such functions include tissue hydration, lubrication, nutrient transport and the ability to withstand mechanical loading [34, 41]. 9

22 The high degree of tissue hydration is maintained through a balance between the swelling pressure exerted by proteoglycans and the constraining forces developed within the collagen network [34]. The interaction between collagens, proteoglycans and water create molecular-sized pores that contribute to a very low permeability for the ECM [34]. Therefore, a significant amount of pressure is needed to force water to flow through the ECM; creating a large frictional resistance against the flow as a result [34]. Frictional resistance and pressurization of water are two ways in which cartilage is able to support high compressive loads [34]. The fluid pressurization is also responsible for the excellent lubrication at the joint surface, which can be predicted by weeping or squeeze film lubrication theories [40, 41]. Dissolved ions such as sodium, calcium, chloride and potassium can be transported to chondrocytes [34]. These dissolved ions also create an osmotic pressure within the tissue, further supporting loads once they reach equilibrium [42]. It is estimated that the tissue s osmotic pressure can support up to 13-22% of the total applied load to the tissue [42] Collagens Collagens of articular cartilage give the tissue its tensile and shear properties and immobilize proteoglycans within the ECM [34]. Although collagen fibres alone do not offer significant resistance to compression due to a high slenderness ratio (length/diameter), their interaction with proteoglycans is critical in maintaining the physical properties of articular cartilage [43]. Collagen types II, VI, IX, X and XI are found in healthy articular cartilage to varying degrees [13]. Collagen type II is the most prevalent, accounting for 90-95% of the collagen in the ECM [13]. Type X collagen is 10

23 found near the calcified region of cartilage and is thought to be a marker of chondrocyte hypertrophy, differentiation and mineralization [34, 43]. Collagen consists of three polypeptide chains, cross-linked covalently, that orient in a triple helical fashion [43]. The primary amino acids that form collagen chains are glycine, proline and hydroxyproline, which account for 33, 25 and 10% of the total residues, respectively [34]. Glycine is present in every third position on the amino acid chain and acts to pull the three chains toward the interior axis of the triple helical structure [34]. The individual chains naturally structure themselves in a left-handed configuration due to their proline content, but are wound around a common axis in a right-handed configuration to resist tensile forces in the triple helix [34]. The presence of hydroxyproline stabilizes the collagen helix because it allows for the formation of intramolecular hydrogen bonds along the length of the molecule [34]. Cross-links that occur between multiple collagen chains result in the formation of an ECM that is stable and strong [34]. Since the collagen network is more stable, collagen turnover does not occur as often as proteoglycan turnover [34]. In normal cartilage collagen degradation occurs at a slow rate, however, there is evidence for accelerated breakdown of collagen in cartilage undergoing repair and remodeling or degeneration [34] Proteoglycans Proteoglycans consist of a protein core with covalently bound glycosaminoglycan (GAG) chains [34]. GAG chains are unbranched polysaccharides that contain a negatively charged sulfate and/or carboxyl group [34]. In a proteoglycan molecule, GAG chains repel each other and other anions, while attracting cations and facilitating interactions 11

24 with water that cause swelling [13]. This is referred to as the Donnon osmotic pressure [34]. Chondroitin sulfate, keratin sulfate and dermatan sulfate are the three main GAGs that are found in articular cartilage [13]. Hyaluronan is another GAG, although it is not sulphated or covalently bound to a protein core [34]. The vast majority of proteoglycans (80-90%) found in articular cartilage are known as aggrecans, which bind to hyaluronan via link proteins to form large proteoglycan aggregates [43] (Figure 2-2). Figure 2-2: Representation of a proteoglycan aggregate. Multiple aggrecans bind to a hyaluronate backbone to form a large macromolecule complex [44]. Within the ECM, the size, structural rigidity and molecular conformation of the charged proteoglycans trapped inside a collagen network influences the mechanical behaviour of cartilage [34]. The complex organization and size of proteoglycans promote networking 12

25 between proteoglycans and collagen, enhancing the stiffness of the ECM [34]. Yet, collagen fibers can slide through the proteoglycan gel because this networking does not include covalent bonds [34]. Control mechanisms for proteoglycan synthesis are very sensitive to biochemical, mechanical and physical stimuli [34]. Proteoglycan turnover in healthy articular cartilage is an ongoing process that occurs at an accelerated rate during cartilage degeneration [34] Chondrocytes Although sparsely populated in articular cartilage, at a modest 10% of the total tissue volume, chondrocytes perform vital functions in directing the production and maintenance of the ECM [34] Chondrocytes possess a surprisingly high metabolic activity for maintaining the ECM under largely anaerobic conditions [45]. In order to perform these functions, chondrocytes are capable of responding to a wide variety of environmental stimuli (chemical and mechanical). Such stimuli include soluble mediators such as growth factors, interleukins and pharmacological agents as well as mechanical loads and hydrostatic pressure changes [34]. Chondrocytes are generated by mesenchymal stem cell differentiation through a process known as chondrogenesis [36]. This process is tightly regulated by growth factors and both cell-cell and cell-matrix interactions [36]. The rounded morphology of chondrocytes is both supported by the 3D ECM and is a critical part of their ability to synthesize matrix components [39]. Cultured chondrocytes in 2D monolayer tend to produce an undesirable ECM that consists of a majority of collagen type I and also tend to differentiate into fibroblast-like cells, losing their rounded shape in favour of a more elongated morphology [18, 46]. There is also 13

26 evidence to suggest that mechanical loads can alter the shape of chondrocytes by reorienting matrix components, which has been linked to changes in the shape and volume of the chondrocyte nucleus [38, 39]. Additionally, as they age, chondrocytes start to lose their ability to effectively respond to soluble mediators which can contribute to apoptosis and osteoarthritis [36] Zonal Organization Cartilage is organized into three distinct zones (Figure 2-3): the superficial zone, intermediate/transitional zone and deep zone [47]. In the superficial zone, collagens as well as the long axis of chondrocytes are arranged parallel to the surface in the direction of load articulation [34, 48]. This region makes up 10-20% of the total thickness of cartilage containing the highest percentage of water, and the lowest percentage of proteoglycans [47]. These properties contribute to a very low coefficient of friction (estimated to be < 0.01) [49]. The superficial zone also contains a multifunctional proteoglycan known as the superficial zone protein (SZP) which has lubricating, growthpromoting, cytoprotective and matrix binding properties which contribute significantly to the functional properties of articular cartilage [50]. In the middle/transitional zone, there is no clear organization of collagen; however, collagen fibril diameter is increased relative to the surface [34]. This zone encompasses 40-60% of the total thickness and is characterized by more spherically shaped chondrocytes [47]. The deep zone, comprising up to 30% of the total thickness, contains the highest percentage of proteoglycans and correspondingly, the lowest percentage of water [47]. Collagen fibrils are large in diameter in this zone and are oriented perpendicular to the surface. Chondrocytes in the 14

27 deep zone are spherical and are arranged in columns [34]. The tidemark (Figure 2-3) separates cartilage from the subchondral bone [47]. Figure 2-3: Cross section of articular cartilage representing the zonal organization [47]. The area around chondrocytes is also divided into separate regions according to the chondrocyte s proximity to the ECM. These regions are classified as the pericellular, territorial and inter-territorial matrices and they differ in their content and in the collagen fibril diameter and organization [34]. The pericellular region surrounds each chondrocyte with proteoglycans and other non-collagenous matrix components [34]. The territorial region surrounds the pericellular region and is made up of thin collagen fibrils [34]. The interterritorial region fills the space between territorial regions, accounting for the majority of the ECM and the bulk of the material properties for the tissue [34]. Collagen 15

28 fibrils in the interterritorial region are of larger diameters and are usually oriented parallel to each other [34]. Proteoglycan content is also highest in the interterritorial region and the chondrocyte content is low [34]. 2.2 Cartilage Biomechanics The compressive modulus of human cartilage ranges from MPa [51, 52], and the tensile modulus ranges from MPa [53]. Bovine articular cartilage has a similar compressive indentation modulus that ranges from MPa [54, 55]. Mature articular cartilage has unique mechanical properties that are best understood when the tissue is described as a biphasic, viscoelastic material [34] Biphasic Nature Articular cartilage is considered to be a biphasic material consisting of a solid phase and a liquid phase. The biphasic theory of articular cartilage was first described by Mow et al.., and provides a theoretical framework for the interrelationship between the composition, structure and function of articular cartilage [56]. This biphasic nature involves three major forces that contribute to balancing externally applied loads [43]. The forces include: (i) the stress developed within the solid phase, (ii) the pressures developed within the fluid phase and (iii) the frictional drag due to fluid flow through the solid phase [43]. Interactions among these three forces give rise to a variety of viscoelastic effects exhibited in articular cartilage [43]. 16

29 2.2.2 Viscoelasticity In addition to being biphasic, articular cartilage is viscoelastic; exhibiting time dependent behaviour when subjected to a constant load or deformation [34]. The two characteristic responses of a viscoelastic material are creep and stress relaxation. Under constant loading, creep describes the initial rapid deformation that slowly diminishes over time until equilibrium is reached [34, 43]. Stress relaxation indicates a response where, under constant deformation, stress is initially high followed by decreased stress over time until equilibrium is reached [34, 43, 57]. Both flow-dependent and flow-independent viscoelastic mechanisms are implicated in the deformational behaviour of articular cartilage [34, 43, 57]. Physical entanglements, mechanical friction and chemical and electrostatic interactions between collagens and proteoglycans of the solid phase give rise to flow-independent properties [57]. Fluid-flow induced drag is a major contributor to flow-dependent properties of articular cartilage [57]. 2.3 Mechanical Stimulation Mechanical loading is directly involved in the development of functional articular cartilage and its eventual deterioration due to trauma or OA. Cartilage is subjected to repetitive mechanical loads and varying stresses that can range from 2 MPa, during activities like climbing stairs, to 20 MPa during jumping [20, 58]. As chondrocytes develop ECM, mechanical stimuli can augment cellular metabolism resulting in a tissue that is capable of withstanding such large and repetitive loads [20]. Anabolic effects of mechanical loading are distinguished from catabolic effects by mechanotransduction pathways that dictate growth or decay. 17

30 2.3.1 Anabolic versus Catabolic Responses Anabolic responses to mechanical loading include increases in ECM synthesis, chondrocyte proliferation and expression of matrix degradation inhibitors. Catabolic responses are characterized by increased expression and activity of matrix degrading enzymes as well as decreases in ECM synthesis and chondrocyte proliferation. Both anabolic and catabolic responses are normal in the development of cartilaginous tissue. Healthy tissue maintains a balance between anabolic and catabolic effects, whereas an imbalance is observed during growth (favouring anabolism) or degradation (favouring catabolism). The mode and magnitude of mechanical loading plays a pivotal role in determining whether the chondrocyte response will be anabolic or catabolic Effect of Loading Mode The effect of mechanical loading on articular cartilage depends on the type, magnitude, duration and frequency of mechanical stimulation that is applied. Compression, tension, shear, hydrostatic pressure and fluid flow are all types of mechanical stimulation that are exerted on cartilage in vivo, with individual zones of cartilage experiencing different mechanical loading environments [40] (Figure 2-4). These types of mechanical stimuli have been employed in bench scale experiments on excised cartilage as well as in vitro tissue engineered constructs, yielding somewhat different results depending on the type of loading. 18

31 Figure 2-4: Representation of the in vivo mechanical environment of articular cartilage under joint loading and motion [40] Compression Dynamic compression is the most widely used form of mechanical stimulation that has been applied to articular cartilage, likely due to the fact that cartilage is routinely exposed to dynamic compressive loads during daily activities. For example, dynamic compression, applied to cartilage explants for 4-48 hours at 3-7% strain and Hz, has an anabolic effect on ECM synthesis [26, 59, 60]. Similarly, cellular proliferation and proteoglycan synthesis was also stimulated by dynamic compression applied to chondrocytes seeded in agarose for 2-48 hours at 5-15% strain and Hz [28, 61]. Generally, high strain amplitudes contribute to proteoglycan synthesis and lower amplitudes favour collagen synthesis [25, 58]. Intermittent applications of dynamic compression elicited greater ECM synthesis and proliferation in chondrocytes seeded on 19

32 calcium polyphosphate substrates, or in foams and gels, in comparison to a continuous application of dynamic compressive forces [25, 62-66]. Overall, the effect of dynamic compression on cartilaginous tissues developed in vitro contributed to increased tissue thickness, strength and modulus [25, 62-64]. Static compression, on the other hand, has been shown to inhibit biosynthesis and proliferation of chondrocytes regardless of how they were cultured [21, 67-69]. This inhibition increased with the magnitude of static compression applied [68-70]. Cartilage explants experience similar consequences of static compression, with proteoglycan and collagen synthesis being inhibited [59, 71] Tension The effects of dynamically applied tension on 3D chondrocyte-seeded scaffolds are influenced by the duration and amplitude of loading. Short applications (3 hours) of dynamic tension up-regulated collagen I, II and X gene expression, as well as the expression of inhibitors of matrix degradation (TIMP-2) and markers for hypertrophy and bone development (Cbfa1) [71]. Longer applications of tensile strain (24-48 hours) did not affect ECM synthesis at low amplitudes (5%) [72], but increased tissue cellularity [73] and inhibited ECM synthesis at higher amplitudes (10 20%) [72, 73]. In contrast, intermittent applications of dynamic tension for long durations (12 hours) had a beneficial effect on 3D constructs by promoting proteoglycan synthesis and accumulation [73]. Dynamic tension, applied to 2D chondrocyte monolayers, demonstrate similar duration and amplitude dependent loading effects. Dynamic tensile strain applied to monolayers for short durations (20 minutes 3 hours) up-regulated the expression of ECM genes (aggrecan and type II collagen) as opposed to longer durations (12 hours) of 20

33 tensile strain which down-regulated ECM gene expression [74, 75]. Tension that is applied statically to chondrocyte cultures has not been widely investigated; however, one study found an increase in ECM synthesis with intermittent applications of static tension [76] Shear There are two forms of shear stress that have been applied to cartilage cultures in vitro: mechanically-induced shear, achieved by lateral movement of clamped surfaces, and fluid-induced shear. Mechanically-induced dynamic shear, applied to cartilage explants for 24 hours at 1-3% shear strain and Hz resulted in an increase in ECM synthesis, where collagen exceeded proteoglycan synthesis two-fold [24]. Mechanical shear, applied intermittently to chondrocytes seeded on calcium polyphosphate substrates for 400 cycles at 2-12% shear strain and 1 Hz, showed an increase in matrix synthesis at low amplitudes and a suppression of matrix synthesis at higher amplitudes [77]. Intermittent applications of dynamic shear over a 4 week period also allowed for the development of thicker cartilage with improved mechanical properties [77]. Fluid-flow is thought to contribute to ECM synthesis by increasing the rate of transport of nutrients and growth factors, while imparting fluid-induced shear forces [20]. In order to capitalize on the effects of flow-induced convective transport and shear, spinner flask, rotating wall, and perfusion culture systems have been utilized in cartilage tissue engineering [20]. In general, laminar flow fields increase the mechanical and biochemical properties of bioreactor-cultivated tissue constructs while turbulent flow promotes the formation of fibrous outer layers for tissue that leads to inferior mechanical properties [22]. Fluid- 21

34 induced shear increased proteoglycan synthesis in chondrocytes grown in 2D monolayer [78]. Similarly, fluid-induced shear can also improve the structural integrity of chondrocyte-seeded hyaluronate fibre scaffolds [79] Hydrostatic Pressure When cartilage tissue is compressed during daily physiologic loading, hydrostatic loads can reach 7-10 MPa [80]. Hydrostatic pressure applied to bovine articular chondrocytes and explants in vitro has been shown to promote or inhibit chondrocyte biosynthesis depending on the stimulus [81]. Chondrocytes subjected to cyclic hydrostatic pressures of 5 MPa for long durations (20 hrs) at frequencies of 0.25 and 0.5 Hz experienced an increase in proteoglycan synthesis in contrast to shorter durations (1.5 hrs) at lower frequencies ( and 0.05), which had an inhibitory effect [81]. Alternatively, the explants experienced significant proteoglycan synthesis from the shorter duration stimulus implicating the ECM in regulating the response [81]. In pressures ranging from 5-10 MPa, both aggrecan and collagen gene expression and synthesis were upregulated [82-84]. Intermittent applications of hydrostatic pressure were shown to stimulate greater proteoglycan synthesis [78] Mechanotransduction Mechanotransduction pathways connect external mechanical forces to internal cellular events, involving a myriad of complex signal-receptor interactions that are poorly understood. Mechanical signals are sensed by mechanoreceptors on the chondrocyte surface which include integrins and ion channels and can be mediated by 22

35 autocrine/paracrine signaling [34, 40] (Figure 2-5). These processes are often isolated from one another in cartilage literature, contradicting what is becoming increasingly apparent; that the transduction of a mechanical signal does not involve a single, signaling pathway resulting in a biological response [40, 85]. Chondrocytes respond to a number of input signals which travel from the cell membrane to the cell nucleus in addition to interacting with other signaling pathways. Figure 2-5: Multiple pathways through which cells detect mechanical deformation [86] Integrins Integrins are a class of receptors that are involved in the attachment of the cytoskeleton (a protein scaffold within the cytoplasm of cells) to its extracellular matrix [87]. Integrinmediated adhesions are intrinsically mechanosensitive and transmit forces to chondrocytes as the ECM is deformed [88]. They exist in a dimeric form with each 23

36 subunit, α and β, spanning the plasma membrane [89]. Chondrocytes express at least 16 α and 8 β subunits, to form over 20 specific receptor combinations [90]. Which integrin subunits are expressed can change with the local microenvironment, mechanical stimulation and because of osteoarthritis [89, 91]. For example, in chondrocytes, the α5β1 integrin is responsible for the attachment of chondrocytes to fibronectin and plays a role in proliferation and adhesion [40]. Another important integrin, α2β1, preferentially binds to collagen type II [40]. Stimulation of both α5β1 and α2β1 in chondrocytes activates the mitogen-activated protein kinase pathway (MAPK) one of the most widely studied signal transduction pathway responsible for changes in gene expression associated with proliferation, differentiation and inflammation [40]. The MAPK pathway is also activated by ligation of β3 integrins as a result of P2Y (ATP receptor) binding [92] Ion Channels Mechanical stimuli also influence the bioelectricity of the cell, which is regulated by a collection of membrane transporters and voltage-gated, ligand-gated, stretch-activated and inactivated ion channels [93]. Channels that are activated by membrane stretch as well as via ligands of second messenger systems are considered to be mechanosensitive [94]. Once stimulated, stretch-activated ion channels (SACs) cause a rapid rise in intracellular calcium. Calcium can then either act as a second messenger in G protein coupled receptor cascades, or as an activator of potassium channels, which are involved in membrane hyperpolarization following loading [20]. There is also strong evidence to 24

37 suggest that membrane structures beyond the channel are critical in signal conditioning of mechanosensitive ion channels [93] Autocrine/Paracrine Signaling Mechanical stimulation can activate an ion channel or a kinase pathway in outside-in and inside-out signaling systems [86]. Autocrine signals originate from and stimulate the same cell, whereas paracrine signals stimulate neighbouring cells. Since chondrocytes are isolated from one another by their own pericellular matrix as well as the surrounding ECM, soluble mediators are critical in maintaining communication between cells to coordinate whole tissue responses. In response to mechanical stimulation, chondrocytes have been known to release nucleotides (e.g. ATP), cytokines (e.g. IL-4), growth factors (e.g. TGFβ) and inflammatory mediators (e.g. NO) [40, 86]. Of particular note, ATP is the first known molecule to be released as a result of mechanical stimulation [30]. 2.4 Purinergic Receptor Pathway ATP acts as both an autocrine and paracrine signal, binding to purinoceptors on chondrocyte membranes which trigger intracellular calcium mobilization and cell signaling cascades [95, 116]. This signaling pathway is called the purinergic receptor pathway Regulation of ATP Signaling The regulation of ATP signaling can be separated into three categories: (i) ATP release, (ii) ATP binding and (iii) ATP degradation. Anabolic or catabolic responses are an indication of an imbalance within these regulatory mechanisms (Figure 2-6). 25

38 Figure 2-6: A schematic representation of the regulation of ATP signaling ATP release ATP is released from cells via stretch-activated channels, hemichannels and exocytosis of matrix vesicles [29]. The putative mechanism for ATP release as a result of mechanical stimulation in chondrocytes is through connexin 43 hemichannels [29]. Connexins are membrane proteins that form hemichannels, which are the precursors for gap junctions (Figure 2-5) [29]. Gap junctions are critical in maintaining channels for cellular communication. When connexin hemichannels are blocked with the specific inhibitors (e.g. flufenamic acid), the increase in ATP release following mechanical stimulation is abolished [29]. 26

39 ATP binding Once released from the cell, ATP and its associated breakdown products can bind to membrane-bound P1 and P2 purinoceptors [95]. Adenosine binds to P1 receptors, while P2 purinoceptors favour highly phosphorylated nucleotides like ATP, ADP and UTP that bind according to receptor subtype [96]. P2X receptors are ligand-gated, while P2Y receptors are G protein-coupled, which incite a second messenger cascade [96]. When P2 receptors are blocked using the non-specific P2 inhibitors suramin or reactive blue 2, no response is seen with respect to mechanical loading, implicating these receptors in mechanotransduction [35]. More specifically, the P2Y 2 receptor has been implicated in ATP binding for chondrocytes [95]. After ATP binds to the P2Y 2 receptor, an increase in intracellular calcium from the extracellular space and from internal stores corresponds to an increase in protein synthesis [95]. Ligation of β3 integrins has also been linked with ATP binding to purinoceptors [92]. Activation of P2X receptors may also facilitate extracellular calcium entry via sodium influx, triggering membrane depolarization and activating voltage operated calcium channels [29] ATP degradation Another mechanism by which ATP signaling is regulated is via nucleotide hydrolysis in the extracellular space. Metabolism of extracellular ATP can occur by ecto-enzymes that are present on the cell membrane or by soluble enzymes that have been co-released by chondrocytes [97]. Enzymes that are present in articular cartilage which degrade ATP include ecto-nucleotide pyrophosphatase /phosphodiesterases (E-NPPs), alkaline phosphatases, tissue transglutaminases and cartilage intermediate layer proteins (CILPs) 27

40 [97]. The E-NPP family is comprised of three enzymes: ENPP1, ENPP2 and ENPP3. In articular chondrocytes, ENPP1 and 3 are expressed, with NPP1 being expressed in the cell membrane and in matrix vesicles and NPP3 being expressed in the pre-golgi [98]. E-NPPs hydrolyze ATP to AMP and inorganic pyrophosphate (PPi). Tissue nonspecific alkaline phosphatases (TNAP) are localized to the plasma membrane of chondrocytes, and sequentially dephosphorylate all nucleotides [99]. Tissue transglutaminases, which are further categorized as tissue transglutaminase type 2 and Factor XIIIA in cartilage, as well as CILPs are located in the cartilage ECM [100, 101]. Although the ATPases present on articular chondrocytes and in cartilage have been characterized, their contribution in regulating mechanotransduction pathways has not been widely studied. One interesting study involving tendon cells, demonstrated that enzymatic activity increased significantly with cyclic loading [3]. Due to their observations, Tzusaki et al. concluded that mechanical load-activation of ATP/ATPase is complex and that elimination of ATP may be as important as ATP release and action [3] Effects of Exogenous ATP on Cartilage and Chondrocytes The concentration of intracellular ATP in eukaryotic cells is on the order of 1 mm, with normal extracellular levels being very low in comparison [102]. Changes in extracellular ATP concentration through supplementation with exogenous ATP can elicit numerous effects on cartilage and chondrocytes. Exogenous ATP can lead to matrix turnover, inflammation, imbalances in PPi levels and mineralization. These effects are not mutually exclusive and seem to be dependent on the dose administered, although the exact dose effect remains unknown. 28

41 Matrix Turnover Chondrocytes incite matrix turn-over as a way of maintaining homeostasis in metabolism and tissue properties, but it can also be a signal of cartilage degeneration and lead to the development of OA [95]. Croucher et al. investigated the effect of extracellular nucleotides (ATP and uridine 5-triphosphate UTP) on the synthesis of ECM macromolecules in bovine chondrocyte pellet cultures [103]. A single dose of extracellular nucleotides (500 µm ATP or UTP) was shown to result in greater accumulation of collagen and proteoglycans determined 21 days after exposure [103]. Repeated additions of ATP or UTP over the 21 day period, however, corresponded to lower levels of accumulated collagen and proteoglycans. A possible explanation for this decrease in collagen and proteoglycan accumulation could be that matrix turnover is occurring at an accelerated rate. Matrix metalloproteinases (MMPs), a disintegrin and metalloproteinase with thrombospondin motif (ADAMTSs), and tissue inhibitor of metalloproteinases (TIMPs) are the enzymes and inhibitors that are responsible for matrix turnover [104]. MMPs generally degrade collagens, ADAMTSs degrade aggrecan and TIMPs inhibit degradation. Collagenases (MMP-1, -8 and -13) are the only enzymes that can cleave the triple-helical portion of collagen molecules [34]. Collagenase 3, also known as MMP-13, has a high specificity towards cleaving collagen type II [34]. Gelatinases (MMP-2 and - 9) cleave denatured alpha chains generated by previous collagenase activity [34] while stromelysins (MMP-3, -10 and -11) act on the non helical domain of type II and IV 29

42 collagen [34]. Aggrecanases1 and 2 (ADAMTS-4 and -5) are the primary proteases responsible for aggrecan cleavage [105]. ADAMTS-1 is one of the most highly expressed ADAMTSs in normal cartilage and has been linked to the coordination of inflammatory responses [106, 107]. In contrast to the catabolic nature of MMPs and ADAMTSs, TIMPs inhibit degradation. TIMP-1 and -2 are potent inhibitors of the MMPs and TIMP-3 is a potent inhibitor of both aggrecanases [105, 108]. The exact role of TIMP-4 is unknown; however it is upregulated in OA and can inhibit ADAMTS-4 as well as some MMPs [109] Inflammation Prostaglandin E 2 (PGE 2 ) has anti-proliferative and pro-apoptotic effects which can stimulate the catabolism of chondrocytes [110]. In addition to acting as an intracellular messenger, nitric oxide (NO) is also implicated in chondrocyte apoptosis, which can lead to the formation of calcium-containing crystals [111, 112]. As a result of mechanical stress, chondrocytes are known to release inflammatory mediators like PGE 2 and NO [113, 114]. However, contrasting results have also been reported, linking mechanical loading to the downregulation of NO and PGE 2 production [111, 115]. Both upregulation and down-regulation of NO and PGE 2 production/release is involved in the purinergic receptor signaling pathway. Extracellular ATP has been shown to increase the amount of PGE 2 released from human, rabbit and bovine chondrocytes seeded in monolayer and bovine articular cartilage explants in a dose-dependent fashion [102, 110, 116, 117]. Doses of 500 µm ATP stimulate the release of PGE 2 and NO from bovine articular chondrocytes seeded in monolayer [117]. Alternatively, inhibiting the 30

43 purinergic receptor pathway leads to the up-regulation of NO release in bovine articular chondrocytes seeded in agarose, implicating the purinergic receptor pathway in the suppression of NO release [118] Inorganic Pyrophosphate Homeostasis Inorganic pyrophosphate (PPi) is composed of two molecules of inorganic phosphate (Pi) joined by a hydrolysable high energy bond [119]. PPi is a byproduct of multiple intracellular reactions, yet it cannot diffuse across the cell membrane [120]. In addition to the generation of PPi from the degradation of ATP by ectoenzymes ( ), PPi is known to be transported from the cytosol into the extracellular space by the one-way transmembrane channel, ANK [119, 121]. The level of extracellular PPi (eppi) plays a critical role in either accelerating or inhibiting the formation of pathological mineral [121]. Excess eppi in articular cartilage has been shown to contribute to calcium pyrophosphate dihydrate (CPPD) crystal deposition, which can elicit deleterious effects on chondrocytes [119]. Conversely, low levels of eppi coupled with a greater presence of Pi, promotes hydroxyapatite or hydroxyapatite-like crystal formation [119]. Regulating eppi levels within such a narrow range is complex, but it is vital in maintaining chondrocyte homeostasis and preventing mineralization [120] Mineralization Pathological mineralization of articular cartilage is often associated with the catabolic and inflammatory effects of ATP discussed previously. It is controversial as to whether hydroxyapatite and CPPD crystals are intimately involved in the progression and 31

44 pathogenesis of OA [122, 123]; however, it is clear that there is an increased presence of both types of mineral in end stage OA [123]. The diseased state of chondrocalcinosis is characterized by CPPD crystal deposition in articular cartilage and is often associated with joint pain and inflammation [124]. Both CPPD and hydroxyapatite crystals induce MMP production and inflammation [125, 126]. ATP has been detected in the synovial fluid of patients with OA, rheumatoid arthritis and chondrocalcinosis; implicating a possible involvement in these diseases [102, 127]. Adding exogenous ATP to articular chondrocyte cultures has been widely used as a mineralization model, inducing CPPD deposition at doses >1 mm [ ]. Articular cartilage vesicles (ACVs) have the capacity to produce CPPD and hydroxyapatite crystals in vitro and have been linked to the production of these crystals in vivo as well [128, 131]. Ryan et al., also found increased 45 Ca retention (a signifier of mineralization) in articular cartilage explants as a result of low doses of exogenous ATP supplementation (100 µm) [124]. Jubeck et al. found that type II collagen suppressed the ability of ATP to stimulate mineralization [132]. A combination of type II and type I collagen, however, increased the effect that ATP had on stimulating mineralization with the amount of type I collagen affecting ACV mineralization in a dose-dependent manner [132]. 32

45 Chapter 3 Materials and Methods 3.1 Cartilage Tissue Culture Cell Isolation Articular cartilage was dissected from calf (12-18 months old) metacarpal-phalangeal joints provided by Brian Quinn s Meats (Yarker, ON). Dissections were performed aseptically in a UV sterilized laminar flow hood on the same day that the calves were sacrificed. Cartilage sections were then digested in a Petri dish containing 0.5% protease (w/v) (Sigma-Aldrich, Oakville, ON) in 25 ml of Ham s F12 media (HyClone, Logan, UT, USA) supplemented with 25 mm HEPES (4-(2-hydroxyethyl)piperazine-1- ethanesulfonic acid) (Sigma-Aldrich) and an antibiotic solution containing: 100 Units/mL penicillin, 100 µg/ml streptomycin and 0.25 µg/ml amphotericin B (Sigma-Aldrich) [133] in an incubator at 37 C with 95% relative humidity and 5% CO 2 for 1.5 hours. Following protease digestion, cartilage sections were washed three times with Ham s F12 media and was replaced with 20 ml of 0.15% collagenase A (w/v) (Roche Diagnostics Canada, Laval, QC) in Ham s F12 media. After 18 hours of incubation with collagenase, the cartilage digest was passed through a 200 mesh filter (Sigma-Aldrich) in order to remove undigested tissue and bone fragments. Isolated chondrocytes were centrifuged at 600 RCF for 7 minutes to obtain a cell pellet, which was subsequently washed three times 33

46 with Ham s F12 media. Cell viability was assessed using the Trypan Blue (Sigma- Aldrich) dye exclusion assay [134] High Density 3D Culture Isolated chondrocytes were seeded in high density, 3D culture on Millicell TM filter units (Millipore, Billerica, MA, USA) which promote the development of cartilaginous tissue [135]. Filter units were prepared for chondrocyte seeding by placing them into 24-well culture plates and coating them with 100 µl of 0.5 mg/ml type II collagen (from chicken sterna) (Sigma-Aldrich) dissolved in 0.1 N glacial acetic acid. After being left to dry overnight, the filters were UV sterilized for 30 minutes and washed twice with Ham s F12 media to remove residual acid. The prepared filters were then seeded with 35,000 viable cells/ mm 2 and maintained in 1 ml of Ham s F12 media supplemented with 5% fetal bovine serum (FBS) (v/v) (Sigma-Aldrich). The culture media was changed every 2-3 days for the duration of the culture period typically for 4 weeks. On the fifth day of culture, the percentage of FBS in the culture media was increased to 20%. Culture media was further supplemented with 100 µg/ml ascorbic acid starting on the seventh day of culture Adenosine Triphosphate (ATP) Supplementation Cartilage constructs were supplemented with different doses of adenosine 5 triphosphate (ATP) (Sigma-Aldrich) throughout the culture period, starting on the third day of culture. A stock concentration of 50 mm ATP (dissolved in culture media) was prepared fresh at each media change. The stock ATP solution was sterile filtered through a 0.2 µm 34

47 acrodisc syringe filter (Pall Life Sciences, East Hills, NY, USA). The sterile ATP stock solution was diluted with fresh culture media to obtain a secondary stock concentration of 12.5 mm. A volume of 5 µl of each stock ATP solution was mixed into filter wells receiving either a low, 62.5 µm dose or a high, 250 µm dose of ATP. Control wells, receiving no ATP, were supplemented with 5 µl sterile culture media. 3.2 Extracellular Matrix (ECM) Synthesis ECM synthesis was measured to quantify the anabolic response of the cartilage tissue to ATP supplementation. Incorporation of [ 35 S] sulphate and [ 3 H] proline correlates with GAG [136] and collagen synthesis [137], respectively. At the 4 week culture point 5 µci of each isotope, [ 35 S] sulfate and [ 3 H] proline (PerkinElmer, Waltham, MA, USA), were added to the cartilage cultures. After incubation for 24 hours, the cultures were harvested ( 3.4.1) and digested ( 3.4.2). Radioisotope incorporation was determined by measuring the radioactivity of the tissue digest using a β-liquid scintillation counter (Beckman Coulter, Mississauga, ON). Synthesis of proteoglycans and collagen were normalized to the DNA content of the tissue ( ). 3.3 P2 Receptor Inhibition In order to verify that the effect of exogenous ATP on chondrocyte metabolism was mediated by the purinergic receptor pathway, P2 receptors were inhibited. The nonspecific P2 antagonist Reactive Blue 2 was used to inhibit both P2X and P2Y receptors, and has a greater affinity for P2Y receptor inhibition [138, 139]. After 4 weeks of culture, a single dose of Reactive Blue 2 (100 µm) was administered prior to ATP 35

48 supplementation ( 3.1.3) and incubation with radioisotopes ( 3.2). ECM synthesis was assessed as outlined previously ( 3.2). 3.4 Tissue Assessment Tissue Harvest Cartilage cultures were harvested by cutting out cartilaginous tissues (attached to filter membranes) from Millicell TM filter inserts using a No. 11 scalpel blade. Samples were washed three times in phosphate buffered saline (PBS), ph 7.4, blotted on weigh paper and placed in 1.5 ml microcentrifuge tubes Tissue Wet/Dry Weights Tissue wet and dry weights were measured using a model P-114 balance (Denver Instrument, Denver, CO, USA). Tissue wet weight was determined by weighing the microcentrifuge tube containing the blotted tissue sample. The empty tube weight and average wet filter weight was subtracted from the total measurement. After wet weight measurement, each microcentrifuge tube was covered with parafilm, holes were poked in the top and the samples were lyophilized (-45 C under vacuum) overnight. After lyophilization, the parafilm was quickly removed, tubes were closed and the tube was weighed once more. Dry tissue weight was determined by subtracting the dry weight of filter membranes from the total measurement. The average dry and wet mass of three bare filters was 1.3 and 1.6 mg, respectively. 36

49 Water content of the tissue was determined by using Equation 3.1: (3.1) Tissue Digestion Cartilage constructs were digested in papain (40 µg/ml) (Sigma-Aldrich) in 20 mm ammonium acetate, 1mM ethyldiaminetetraaceetic acid (EDTA) and 2 mm dithiothreitol (DTT) (Sigma-Aldrich) for 72 hours at 65 C [62]. Tissue digests were stored at -20 C until further analysis Extracellular Matrix (ECM) Accumulation The total amount of DNA, proteoglycans and collagen that accumulated in the developed cartilage tissues was measured from the tissue digests ( 3.4.3) using established biochemical assays DNA Assay DNA content was measured fluorometrically using the Hoescht (Sigma-Aldrich) dye binding assay as described previously by Kim et al. [140]. Aliquots of tissue digest were diluted with PBS, ph 7.4, and standard curves were prepared using calf thymus DNA (Sigma-Aldrich) directly prior to analysis. Samples and standards were added in triplicate to black 96 well plates (VWR International, Mississauga, ON) that were kept on ice until fluorescence measurement at an excitation wavelength of 350 nm and emission wavelength of 450 nm. 37

50 GAG Assay Proteoglycan content in the developed tissue was estimated by measuring the sulphated GAG content using the dimethylmethylene blue dye (Polysciences, Warrington, PA, USA) binding assay, as first described by Farndale et al. [141] and later improved by Goldberg et al. [142]. Directly prior to performing the assay, aliquots of tissue digest were diluted with 1% bovine serum albumin (w/v) (Sigma-Aldrich) in PBS, ph 7.4. Standard curves were generated using bovine cartilage chondroitin sulphate A (Sigma- Aldrich) in order to correlate absorbance with GAG content. Samples and standards were added in triplicate to 96 well plates (VWR International) followed by absorbance measurement at 525 nm Hydroxyproline Assay Collagen content in the developed tissue was estimated by measuring the hydroxyproline content, which is assumed to account for 10% of the total collagen mass [143]. Hydroxyproline content was measured using the chloramine T/Ehrlich s reagent assay described by Woessner et al. [144]. Aliquots of the tissue digest were hydrolyzed in 6 N HCl at 110 C for 18 hours. The samples were then neutralized with an equal volume of 5.7 N NaOH and further diluted with distilled water. Standard curves were generated using L-hydroxyproline (Sigma-Aldrich) in order to correlate absorbance with hydroxyproline content. Standards and samples of hydrolyzate were added in triplicate to 96 well plates (VWR International) and reacted with chloramine T and Ehrlich s reagent (Sigma-Aldrich) followed by absorbance measurement at 560 nm. 38

51 3.4.5 Thickness Separate cultures were established for the determination of tissue thickness and mechanical properties. After 4 weeks of culture, each tissue filter unit was placed on a stainless steel stage with a flat surface and the tissue thickness was measured using the needle probe method described by Hoch et al. [145]. Briefly, a 25 Ga. needle (Becton Dickinson, Franklin Lakes, NJ, USA) was attached to a 1 kg load cell on a MACH-1 mechanical tester (Biomomentum Inc., Laval, QC, Canada) and displaced into the tissue at a rate of 5 μm/s. The MACH-1 mechanical tester (Biomomentum Inc.) has a displacement resolution of 0.5 µm and a force resolution of 0.2 mn. Resistive compressive forces and needle displacement were recorded at a frequency of 10 Hz using the MACH-1 motion data acquisition software (Biomomentum Inc.). Abrupt changes in force signified needle contact with the tissue and then the underlying support, and this difference was interpreted as tissue thickness [146]. This process was repeated at another random location in the tissue and the average value between the two measurements was documented. All measurements were conducted in culture media at 37 C Mechanical Properties Directly following tissue thickness measurements ( 3.4.5), the compressive stiffness of the tissues were determined by uniaxial, compression indentation tests [147] using a plane-ended indenter (2 mm diameter) attached to the MACH-1 mechanical tester (Biomomentum Inc.). The diameter of the indentation probe was selected so that the ratio of the indenter radius-to-tissue thickness was within acceptable limits for this technique (0.2-8) [147]. Each tissue filter unit was placed on a porous stainless steel stage 39

52 and was surrounded by culture media at 37 C. Contact between the indenter and the tissue was achieved by preloading the tissue to 5 mn, which was defined as the zerostrain state. Compressive step strains of 2% total thickness ( 3.4.5) were then applied incrementally to the tissue until a maximum of 20% strain was reached. Each step was held until there was a change in force of less than 2 mn/min (which was chosen to represent equilibrium). The apparent equilibrium indentation modulus was determined at 20% strain using the expression derived by Hayes et al. for indentation testing of cartilage (Equation 3.2) [147]: (3.2) where P is equilibrium load, v is Poisson s ratio, a is indenter radius, w is displacement, and K is a correction factor based on v and the aspect ratio of a to tissue thickness h. Equation 3.2 was derived from a model which defines articular cartilage and the underlying subchondral bone as an infinite elastic layer bonded to a rigid half-space. The K value for each tissue culture sample was interpolated from values provided by Hayes et al. (Table 3-1) [147]. The Poisson s ratio for bovine articular cartilage was determined to be 0.4 by Buckwalter et al., which was used to determine tissue modulus [34]. 40

53 Table 3-1: Values used to interpolate for correction factor K. These values were provided by Hayes et al. for indentation of cartilage by a plane-ended cylindrical indenter and a Poisson s ratio of 0.4 [147]. a/h K Histology and Immunohistochemistry Selected cultures from each experimental group (not subjected to mechanical testing) were harvested for histological and immunohistochemical assessment. Immediately after harvest, tissue samples attached to the filter membranes were rinsed three times in PBS, ph 7.4, fixed in 4% paraformaldehyde (v/v) (Sigma-Aldrich) overnight at 4 C, and embedded in paraffin. Undecalcified, thin sections (5 µm) were cut and bound to glass slides by Mr. John DaCosta in the Department of Pathology and Laboratory Medicine, Queen s University, Kingston, ON. For histological staining, tissue sections were immersed in toluene to remove the paraffin, followed by immersion in a series of graded 41

54 alcohol and water solutions to rehydrate the tissue sample. Slides were stained with Gill s hemotoxylin-eosin (H&E), a general connective tissue stain [148], toluidine blue (TB) which stains sulphated proteoglycans blue [149], and von Kossa (VK) which stains calcium phosphate black [150]. After staining, the tissues were dehydrated by immersion in a series of graded alcohol, mounted with PRO-TEXX mounting medium (Fisher Scientific, Pittsburgh, PA, USA) and examined by light microscopy. Immunohistochemistry was used to detect collagen type I and II in the developed cartilaginous tissues. Following paraffin embedding and sectioning, tissue sections were briefly dehydrated in alcohol and re-hydrated in water. The sections were then enzymatically pretreated to facilitate primary antibody binding. Sections, that were stained for type II collagen, were first treated to 0.25% trypsin (w/v) (Sigma-Aldrich) in Tris-buffered saline (50 mm Tris with 150 mm NaCl), ph 7.6, for 30 minutes at room temperature [151]. All sections were then treated with 0.5 units/ml chondroitinase ABC (Sigma-Aldrich) in Tris-acetate buffer (40 mm Tris acetate with 1 mm EDTA), ph 8.5 for 1 hour at 37 C, and subsequently with 2.5% hyaluronidase (w/v) (Sigma-Aldrich) in PBS, ph 7.4 for 30 minutes at 37 C [151]. Afterwards, sections were immersed in 0.3% H 2 O 2 (v/v) for 30 minutes to quench endogenous peroxidase activity [152]. In order to reduce non-specific protein binding, type I and II collagen sections were blocked with 20% goat and horse serum (Vector Laboratories Inc., Burlingame, CA, USA), respectively, in PBS, ph 7.4, with 0.1% Triton X-100 (v/v) for 30 minutes at room temperature. Type I collagen sections were incubated with rabbit polyclonal type I 42

55 primary antibody (T40113R: Biodesign International, Saco, ME, USA) at a 1:200 dilution in 10% goat serum (v/v) in PBS, ph 7.4, with 0.1 % Triton X-100 (v/v) overnight at 4 C. Type II collagen sections were incubated with mouse monoclonal type II primary antibody (II-II6B3: Developmental Studies Hybridoma Bank, University of Iowa, IA, USA) at a 1:10 dilution in 10% horse serum (v/v) in PBS, ph 7.4, with 0.1 % Triton X-100 (v/v) overnight at 4 C. Following primary antibody incubation, sections were rinsed in PBS, ph 7.4, and incubated with biotinylated secondary antibody using the Vectastain Elite ABC kit (Vector Laboratories) with 10 % goat serum (v/v) in PBS with 0.1% Triton X-100 (v/v) for type I collagen, or 10% horse serum (v/v) in PBS with 0.1% Triton X-100 (v/v) for type II collagen. According to the ABC kit instructions (Vector Laboratories), immunodetection was performed using 3, 3 -diaminobenzidine tablets (Sigma-Aldrich) for colour development. Sections were counterstained with 1 % methyl green (w/v) (Sigma-Aldrich) for two minutes. Finally, sections were cleared and dehydrated by immersion in water and a series of graded alcohol, mounted with PRO- TEXX mounting medium (Fisher Scientific) and examined by light microscopy. Nonspecific staining was assessed by omitting the primary antibody from each type of section. This protocol was adapted from Chevrier et al. [151] and performed by Ms. Aasma Khan in the Department of Chemical Engineering, Queen s University, Kingston, ON Transmission Electron Microscopy (TEM) Separate cultures from each experimental condition were fixed in a 2% glutaraldehyde (v/v) in 0.1 M sodium cacodylate buffer, ph 7.3 while still in filter inserts. The samples 43

56 were left to sit in the fixative for 1 hr at room temperature before being transferred to 4 C overnight. The samples were then harvested ( 3.4.1) and sent to Mr. Doug Holmyard at Mount Sinai Hospital, Toronto, ON for further processing followed by embedding, sectioning and TEM. After fixation, the samples were rinsed with 0.1 M sodium cacodylate buffer with 0.2 M sucrose, ph 7.3 and then post-fixed with 1% osmium tetroxide (Electron Microscopy Sciences) in 0.1 M sodium cacodylate buffer, ph 7.3 for 1.5 hours. The samples were rinsed again in 0.1 M sodium cacodylate buffer with 0.2 M sucrose, ph 7.3, dehydrated in a series of graded ethanol solutions and infiltrated with a 1:1 propylene oxide:quetol-spurr resin mixture (700 mm Quetol 651 resin, 2.5 M nonenyl succinic anhydride (NSA), 770 mm vinylcyclohexene dioxide (ERL), 550 mm diglycidil ether of polypropylene glycol (DER 736) and 130 mm benzyl dimethylamine (BDMA)). After infiltration with resin, the samples were placed in embedding molds and polymerized in an oven at 65 C overnight. TEM was performed using a FEI Tecnai 20 transmission electron microscope (FEI, Hillsboro, OR, USA). All reagents were purchased from Electron Microscopy Sciences, Hatfield, PA, USA. 3.5 Gene Expression Genes of Interest Gene expression of MMPs, ADAMTSs and TIMPs were examined in the ATPsupplemented cultures in order to assess matrix turnover. The following genes were chosen based on their primary involvement in degrading, or inhibiting degradation, of collagen and proteoglycan in bovine articular cartilage: MMP 1, 3, 13, ADAMTS 1, 4, 5 and TIMP S ribosomal RNA was used as a housekeeping gene since its 44

57 expression remains unchanged in chondrocytes under similar culture conditions irrespective of mechanical stimulation [153, 154] Primer Design and Optimization Primers were designed using published sequences specific to the Bos Taurus genome (basic local alignment search tool - BLAST) and Primer 3 software [155]. Primers were designed to be between 100 and 250 base pairs, with a Guanine-Cytosine (GC) content between 45 and 60% and an optimal melting temperature (Tm) of 60 C (Table 3-2). In order to detect genomic contamination, primer sets were selected to span intron-exon boundaries. The program Amplify 3 was also used to pre-screen primers for potential dimers or unwanted products [156]. Primers were purchased from Integrated DNA Technologies Inc. (Coralville, IA, USA). All primers were optimized by determining the appropriate conditions for amplification and semi-quantization (Table 3-2). Such conditions include optimizing for cycle number, annealing temperature and reagent composition (MgSO 4 concentration). In cases where unwanted products or dimers were present, the primer was re-designed RNA Isolation Tissue constructs were snap frozen in liquid nitrogen after 4 weeks of culture and pulverized using a mortar and pestle. Following homogenization of each sample with Tri-Reagent (Sigma-Aldrich), RNA was extracted according to the acid-guanidiniumphenol-choloroform method [157]. RNA concentration and purity was assessed by using a NanoDrop Spectrophotometer (ND1000; Nanodrop Products, Wilmington, DE, USA) 45

58 and by resolving RNA on 1% formaldehyde-agarose gels. All samples generally had a 260/280 value that ranged from and a densitometry ratio of 1:2 (18S:28S) Reverse Transcription First strand cdna was synthesized from 1 µg of total RNA using oligo (dt)15 (Coretec DNA Service Laboratories Inc., Kingston, ON, Canada) and the Multiscribe Reverse Transcription Kit (Applied Biosystems Inc., Foster City, CA, USA) which included random primers. Ribolock RNase inhibitor was also added (40 units; Fermentas International Inc., Burlington, ON, Canada). 46

59 Name Table 3-2: Oligonucleotide primers used for semi-quantitative end-point PCR. Accession Number Primer Sequence # bases Product Size # cycles Annealing Temp (deg C) [MgSO 4 ] MMP-1 NM_ Fwd: 5'-TCCCTTGGACTTGCTCATTC-3' 20/ bps mm Rev: 5'-ACTGGCTGAGTGGGATTTTG-3' MMP-3 XM_ Fwd: 5'-CTTGTCCTTCGATGCAGTCA-3' 20/ bps mm Rev: 5'-CTGATGGCCCAGAACTGATT-3' MMP-13 NM_ Fwd: 5'-CCCAGGAGCACTCATGTTTC-3' 20/ bps mm Rev: 5'-GGCGTTTTGGGATGTTTAGA-3' ADAMTS-1 NM_ Fwd: 5'-AGCCCTGGTCTCCCTGTAGT-3' 20/ bps mm Rev: 5'-AGATCGGAGGGGAGCTGTAT-3' ADAMTS-4 NM_ Fwd: 5'-AGCAGTCTGGCTCCTTCAAA-3' 20/ bps mm Rev: 5'-ATTCACCGTTGAGGGCATAG-3' ADAMTS-5 XM_ Fwd: 5'-GACGTGTGCAAACTGACCTG-3' 20/20 94 bps mm Rev: 5'-TGTATGGCCTGCACTCTGTC-3' TIMP-1 NM_ Fwd: 5'-GATGTTCAAAGGGTTCAGTGC-3' 21/ bps mm Rev: 5'-GCAGAACTCATGCTGTTCCA-3' TIMP-2 NM_ Fwd: 5'-TACCAGATGGGCTGTGAGTG-3' 20/ bps mm Rev: 5'-GTCCAGAAACTCCTGCTTGG-3' TIMP-3 NM_ Fwd: 5'-ATCGATATCACCTGGGCTGT-3' 20/ bps mm Rev: 5'-TGCAAGCGTAGTGTTTGGAC-3' TIMP-4 NM_ Fwd: 5'-GCCTTTTGATTCCTCCCTCT-3' 20/ bps mm Rev: 5'-TTCTCCCAGGGCTCAATGTA-3' 18S DQ Fwd: 5'-ATGGCCGTTCTTAGTTGGTG-3' 20/ bps mm Rev: 5'-ATGGCCGTTCTTAGTTGGTG-3' 47

60 3.5.5 End-point Polymerase Chain Reaction (PCR) End-point PCR was performed on cdna samples using a High purity Taq DNA polymerase kit (UBI Life Sciences Inc., Saskatoon, SK, Canada) and a Bio-Rad C1000 thermal cycler (Bio-Rad, Hercules, CA, USA. Diluted cdna (~2 ng of input RNA) accounted for 2 µl of the 20 µl reaction master mix. The reagent concentrations were as followed, unless otherwise stated (Table 3-2): 2.5 units Taq, 1X Taq reaction buffer, 2 mm MgSO 4, 200 µm dntp, 1X Q solution (QIAGEN Inc., Valencia, CA, USA) and 0.5 µm for each primer. Following the mixture of reagents, samples were briefly centrifuged and placed in the thermal cycler (program outlined in Table 3-3). Minus reverse transcription and no template controls were included in every PCR run. Table 3-3: Thermal cycler program for PCR. Step Temperature Duration 1 Initialization 95 C 2 min 2 Denaturation 95 C 20 s 3 Annealing 55 C 20 s 4 Extension /Elongation 72 C 20 s *Repeat steps 2-4 for optimized cycle number (Table 3-2) 5 Final Elongation 72 C 5 min 6 Final Hold 4 C Indefinitely Evaluation of PCR Products Samples of PCR product mixed with orange 6X loading dye (4:1; Fermentas) were pipetted into 2.5% agarose (w/v) TBE gels containing 0.2 µg/ml ethidium bromide. An ultra low DNA ladder was also included in the gel (Fermentas). All gels ran for 1 hour 48

61 15 minutes at 50 V in a wide mini-sub cell GT system (Bio-Rad, Hercules, CA, USA). Gels were visualized under UV light (G:Box Chemi HR16, Syngene, Cambridge, UK) and semi-quantitative gene expression levels were determined by densitometry (GeneTools, Syngene, Cambridge, UK). All expression levels were normalized to the housekeeping gene, 18S ribosomal RNA, which was included in all gels. 3.6 MMP-13 Protein Activity Protein Extraction Intracellular protein was extracted from tissue digests according to a protocol from Nielsen et al. [158]. After tissue harvest ( 3.4.1), samples were snap frozen in liquid nitrogen and pulverized in a pointed bottom microcentrifuge tube. The sample was then homogenized twice (30 s each time) at high speed (Power Gen 125; Fisher Scientific) in extraction buffer (50 mm Tris-HCl ph 7.4, 0.1 M NaCl and 0.1% Triton X-100 (v/v)) supplemented with one Complete Mini EDTA-free protease inhibitor cocktail tablet (Roche Diagnostics Canada) per 10 ml of buffer. The samples were then centrifuged at 1700 g for 10 minutes at 4 C. Afterwards, the supernatants were harvested and further centrifuged at 10,000 g for 10 minutes at 4 C. The supernatants were stored at -20 C until further analysis Fluorescent Resonance Energy Transfer (FRET) assay In order to assess the amount of active MMP-13 in the tissue, a FRET based assay outlined by Knauper et al. was used [159]. The Mca MMP-13 FRET fluorogenic 49

62 substrate (MCA-Pro-Cha-Gly-Nva-His-Ala-Dpa-NH2) (AnaSpec Inc., Freemont, CA, USA) consists of a fluorophore coupled to a quencher which fluoresces when cleaved by MMP-13. Although the FRET peptide was designed to be MMP-13 specific, both MMP- 8 and MMP-9 have also been shown to cleave the FRET substrate (AnaSpec Inc.). Mca MMP FRET Peptide Fluorescence Standards (MCA-pro-leu-OH) (AnaSpec Inc.) were reconstituted to a concentration of 100 µm in 20% DMSO (v/v) and serially diluted to create a standard curve. Samples and standards were then added to wells of a black 96 well plate in triplicate. An aliquot of 100 µm Mca MMP-13 FRET substrate in 20% DMSO (v/v) was mixed with assay buffer (1:4; 200 mm NaCl, 50 mm Tris Base, CaCl 2, 20µM ZnSO 4, and 0.05% BRIJ 35 solution (w/v), ph 7.5) and added to the wells in the dark. The plate was covered and incubated at 37 o C with 95% humidity for 18 hours. After incubation, fluorescence was measured at 325 nm excitation and 393 nm emission. 3.7 Measurement of Inorganic Pyrophosphate (PPi) The amount of PPi generated intracellularly and in the surrounding tissue was measured from tissue digests ( 3.4.3) using a protocol, modified from Justesen et al., that makes use of a coupled enzymatic reaction resulting in the formation of NADPH from PPi [160]. The triethanolamine acetate buffer mixture (TM) was injected into all wells before samples or standards were added. The TM consisted of 70 mm triethanolamine, 20 mm magnesium acetate, 20 mm potassium acetate, 2 mm EDTA, 2 mm DTT and 245 µl of 17.4 N glacial acetic acid, which were combined and adjusted to a ph of 7.6. A standard curve was prepared using sodium pyrophosphate (NaPPi) in papain digestion buffer 50

63 ( 3.4.3) immediately prior to addition of the enzyme mixture (EM) in the dark. The EM contained 30% TM (v/v), 3 mm nicotinamide adenine dinucleotide phosphate (NADP + ), 2 mm uridine diphosphoglucose (UDPG), 30 µm glucose 1,6-diphosphate, 7.5 U UDPG pyrophosphorylase from beef liver, 5 U phosphoglucomutase from rabbit muscle, and 4 U glucose-6-phosphate dehydrogenase from yeast. Directly following EM addition, the plate was dark adapted for 5 minutes and fluorescence was measured at 340 nm excitation and 460 nm emission at 37 C. Extracellular PPi was also measured from the conditioned culture media surrounding 4 week old tissue samples using the same methods outlined above, with culture media occupying the standard curve instead of papain digestion buffer. All reagents for the PPi assay were purchased from Sigma-Aldrich. 3.8 Quantification of Inflammatory Mediator Release The release of inflammatory mediators, prostaglandin E 2 (PGE 2 ) and nitric oxide (NO), was assessed from conditioned culture media that was collected at the end of the culture period Prostaglandin E 2 (PGE 2 ) Assay The concentration of PGE 2 in the conditioned media was determined using a commercially available enzyme immunoassay (EIA) kit (Cayman Chemical Co., Ann Arbor, MI, USA) according to the manufacturer s instructions. The EIA, which has a minimum detection limit of 15 pg /ml, is based on the competition between PGE 2 and a PGE 2 -acetylcholinesterase (AChE) conjugate (PGE 2 tracer) for a limited amount of PGE 2 monoclonal antibody. Since the PGE 2 tracer concentration is held constant while the 51

64 PGE 2 concentration varies, the amount of PGE 2 tracer that binds to the antibody is inversely proportional to the concentration of PGE 2 in the well. The conditioned media samples and PGE 2 EIA standards (diluted with culture media) were added in triplicate to the wells, which were subsequently mixed with PGE 2 tracer. Wells that addressed non specific binding (PGE 2 tracer added in the absence of standard or sample) and maximum binding (PGE 2 tracer and antibody added in the absence of standard or sample) were also included in triplicate at this point. The plate was then incubated at 4 C for 18 hours to allow the PGE 2 -antibody complex to bind to a goat polyclonal anti-mouse IgG that came pre-attached to the well plate. Unbound reagents were washed away with Wash Buffer, and Ellman s reagent, which contained the AChE substrate, was added to all wells. At this point, a small amount of PGE 2 tracer was added to three wells to assess total enzymatic activity. The plate was shaken and allowed to develop in the dark for 60 minutes prior to absorbance measurement at 412 nm Nitric Oxide (NO - ) Assay NO - is difficult to measure directly because of its instability, however, NO - readily reacts to produce the stable end product, nitrate (NO - 3 ). The concentration of nitrate in the conditioned media was detected by the Griess reaction according to the protocol described by Van De Loo et al. [161]. The Griess reagent was composed of 0.2% N-(1- napthyl) ethylenediamine dihydrochloride (w/v) (Sigma-Aldrich) diluted 1:1 with 2.0% sulfanilamide (w/v) (Sigma-Aldrich) in 3N HCl. Briefly, equal amounts of sodium 52

65 nitrate (NaNO 3 ) standards and conditioned media samples were mixed, in triplicate, with Griess reagent in a flat bottom 96 well plate and absorbance was read at 540 nm. 3.9 Optimization of ATP Dose A range of doses were administered to the cartilage cultures in order to determine an optimal dose of ATP that would maximize the anabolic and minimize the catabolic responses that were observed ( 4.1, 4.2 and 4.3). According to the protocol outlined previously ( 3.1.3), the following doses of ATP were added to the cultures: 0, 31.25, 62.5, 125 and 250 µm. After 4 weeks, half of the cultures were assessed for ECM synthesis ( 3.2) and half of the cultures were assessed for MMP-13 activity ( 3.6) Statistical Analyses In order to reduce variability in the samples, articular chondrocytes were obtained from a pooled source of at least two dissected legs for all experiments that were performed. Each experimental group had a sample size of at least 4, but it typically ranged from 6-8. Most experiments were repeated at least twice using chondrocytes from different animals, and the combined data were analyzed. Exceptions include samples prepared for visual assessment (histology, immunohistochemistry, TEM) and gene expression. All results were expressed as the mean ± standard error of the mean (SEM). Collected data were analyzed statistically using a one-way analysis of variance (ANOVA) and the Fisher s least significant difference (LSD) post-hoc test (SPSS version 16, SPSS Inc., Chicago, IL, USA) to determine the effect of ATP dose. The data were checked prior to performing statistical tests for both normality and equal-variance. Significance was 53

66 associated with p-values less than Trends were noted with p-values between 0.05 and

67 Chapter 4 Results 4.1 Effect of exogenous ATP on extracellular matrix (ECM) synthesis Cartilage cultures were supplemented with 0, 62.5 or 250 µm of ATP throughout a 4 week culture period. Synthesis of collagen and proteoglycans were determined over a 24 hour period after the final exposure to ATP, at the end of the culture period, by radioisotope incorporation ( 3.2). ECM synthesis was significantly enhanced by ATP supplementation (p < 0.02) (Figure 4-1). Collagen synthesis (measured by incorporation of [ 3 H] proline) was significantly increased 46 ± 11% over control in the 250 µm ATP dose (p < 0.01). However, the 23 ± 11% increase with respect to the 62.5 µm dose (p < 0.09) and the difference in [ 3 H] proline incorporation between the two doses of ATP (p < 0.08) reflected trends in the data. Proteoglycan synthesis (measured by incorporation of [ 35 S] sulfate) was increased 18 ± 10 % and 37 ± 12% over control in the 62.5 µm and 250 µm ATP doses, respectively. Similar to the [ 3 H] proline data, the 62.5 µm dose of ATP elicited an increasing trend from control (p < 0.09) while the 250 µm dose elicited a significant increase from control (p < 0.01). Additionally, there were no significant differences in proteoglycan synthesis between the two doses of ATP (62.5 and 250 µm) (p > 0.1). 55

68 Figure 4-1: Effect of exogenous ATP on collagen and proteoglycan synthesis. Data are presented as the mean ± SEM normalized to control, n = 7-15 samples per group. * Denotes a significant difference from control (p < 0.01) ** Denotes a trend from control (p < 0.09) Confirmation of purinergic receptor signaling In order to confirm that the changes in ECM synthesis were due to stimulation of the purinergic signaling pathway, synthesis experiments were also conducted in the presence or absence of 250 µm ATP and 100 µm of the P2 receptor inhibitor, Reactive Blue 2 (RB2).In the presence of RB2, the stimulation effect of exogenous ATP on collagen and proteoglycan synthesis was abolished (p < 0.01) (Figure 4-2). 56

69 Figure 4-2: Effect of exogenous ATP (250 µm) on collagen and proteoglycan (PG) synthesis in the presence or absence of Reactive Blue 2 (100 µm). Data are presented as the mean ± SEM normalized to control, n = 7-8 samples per group. * Denotes a significant difference between the ATP stimulated and no-atp control group (p < 0.05); ** Denotes a significant difference between the ATP stimulated inhibitor and control groups (p < 0.01). 4.2 Long-term effects of ATP supplementation on tissue formation and properties The long-term effects of ATP supplementation on cartilaginous tissue formation and properties were determined after tissue-engineered cartilage constructs were cultured in the presence of exogenous ATP for a period of 4 weeks. Separate sets of cultures were grown to determine the effect on ECM accumulation, mechanical properties, histological appearance, immunohistochemistry staining for collagen types (I and II) and transmission electron microscopy (TEM) for the presence of mineralization. 57

70 As a result of long-term ATP supplementation, the wet weight of the developed tissue constructs did not differ from the control cultures regardless of ATP dose supplied (Table 4-1). The wet weight was, however, significantly different between the 62.5 and 250 µm dose of ATP (p < 0.05). Similarly, the dry weight of the developed cartilaginous tissues was also unchanged by ATP supplementation (Table 4-1). However, the water content was significantly lower in the 250 µm dose compared to the low dose of ATP (62.5 µm) or control cultures (p < 0.05) (Table 4-1). Table 4-1: Effect of exogenous ATP on cartilaginous tissue formation and properties. Control 62.5 M ATP 250 M ATP (n=6-8) (n=6-8) (n=6-8) Wet Weight [mg] 8.4 ± ± 0.4** 7.8 ± 0.4** Dry Weight [mg] 1.7 ± ± ± 0.2 Water Content [%] 80 ± 1 81 ± 1 77 ± 1 * DNA [ g] 8.6 ± ± ± 0.2 GAG/Dry Weight [ g/mg] 37 ± 2 35 ± 3 24 ± 1 * GAG/DNA [ g/ g] 7.0 ± ± ± 0.4 * Collagen/Dry Weight [ g/mg] 42 ± 9 * 60 ± 3 66 ± 5 Collagen/DNA [ g/ g] 9.5 ± ± ± 0.1 * Collagen-to-GAG Ratio 1.1 ± ± ± 0.5 * Thickness [ m] 230 ± 5 * 354 ± ± 8 Indentation Modulus [kpa] 54 ± 13 * 352 ± ± 64 Data presented as mean ± SEM Indentation modulus determined at 20% compressive strain * Significantly different from all other groups (p < 0.05) **Significant difference between groups 58

71 The DNA content of the developed cartilaginous tissues did not change significantly with ATP supplementation (Table 4-1). However, there were observed changes in ECM accumulation with ATP supplementation. The amount of collagen that accumulated in the developed ECM significantly increased by 16 ± 3 % and 26 ± 3% in the 62.5 µm and 250 µm conditions, respectively (p < 0.05) (Figure 4-3). The accumulated collagen also accounted for a significantly larger portion of the dry weight in the ATP stimulated cultures (independent of dose) (p < 0.05) (Table 4-1). The amount of accumulated collagen was also significantly increased in relation to the total DNA content of the developed tissues, but only with respect to the high dose (250 µm) of ATP (p < 0.05) (Table 4-1). Proteoglycan accumulation was unchanged under the low dose (62.5 µm) of ATP, but under the high dose (250 µm) of ATP there was a significant decrease in proteoglycan accumulation (p < 0.05) (Figure 4-3). In contrast to collagen accumulation, the accumulated proteoglycans accounted for a significantly smaller portion of the dry weight in the 250 µm ATP stimulated cultures (p < 0.05) (Table 4-1). The amount of accumulated proteoglycans also decreased in relation to the total DNA content of the developed cartilaginous tissues supplemented with 250 µm ATP (p < 0.05) (Table 4-1). 59

72 Figure 4-3: Effect of long term (4 week) ATP supplementation on collagen and proteoglycan accumulation. Data are presented as the mean ± SEM, n = 6 samples per group. * Denotes a significant difference between all other groups (p < 0.05) The ATP-supplemented cultures yielded cartilaginous tissue that was significantly thicker than the unstimulated controls, independent of the dose supplied (p < 0.05) (Table 4-1). Tissue thickness increased by 54 ± 16% and 50 ± 4% for the 62.5 and 250 µm doses, respectively. Stimulation by ATP also resulted in significantly improved mechanical properties, with a 5- to 6.5-fold increase in indentation modulus for cultures grown in the presence of ATP, irrespective of the dose (p < 0.01) (Table 4-1). Histological assessment of the developed tissues demonstrated that all of the cultures stained positive for sulphated proteoglycans (as assessed by toluidine blue (TB) staining) and displayed a balanced distribution of connective tissue proteins (as assessed by 60

73 hematoxylin and eosin (H&E) staining) (Figure 4-4). There was also no sign of mineralization (as assessed by von Kossa (VK) staining) in any of the developed cultures (with or with ATP supplementation) (Figure 4-4). Transmission electron microscopy (TEM) analysis was also conducted to detect the presence of mineral deposits in the vicinity of the cells which also showed no evidence of mineralization (Appendix A). Immunohistological assessment established that the primary form of collagen in the cartilaginous tissue constructs was type II collagen, with minor amounts of type I collagen and no observable effect as a result of ATP supplementation (Figure 4-5). Figure 4-4: Histological assessment of the developed cartilaginous tissues after 4 weeks of ATP supplementation. Tissues were stained with H&E, TB and VK. Original magnification of 100X and the scale bar represents 100 µm. Representative images are shown. 61

74 Figure 4-5: Immunohistochemical assessment of collagen types II and I in the developed cartilaginous tissues after 4 weeks of ATP supplementation. Original magnification of 100X and the scale bar represents 100 µm. Representative images are shown. 4.3 Catabolic matrix turnover gene expression and enzyme activity The expression of 10 catabolic matrix turnover genes ( 3.5) was determined after long term ATP supplementation using RT-PCR. The only gene that was differentially expressed was MMP-13 which exhibited an over 2- fold significant increase in gene expression in the cultures supplemented with 250 µm of ATP compared to both the low dose (62.5 µm) and control (Figure 4-6). ADAMTS-4 also exhibited a slight increasing trend in gene expression (20 ± 11%) for the high dose of ATP (250 µm) (p < 0.06) (Figure 4-6). 62

75 Figure 4-6: Semi-quantitative end-point PCR results for the expression of catabolic matrix turnover genes. (A) Representative expression of PCR products analyzed by electrophoresis on 2.5% gels, (B) Relative expression of PCR products normalized to 18S ribosomal RNA (housekeeping gene) by densitometry. Data are presented as the mean ± SEM normalized to control, n = 4 samples per group. * Denotes a significant difference between all other groups (p < 0.05) **Denotes a trend between the high dose of ATP (250 µm) and other groups 63

76 Since MMP-13 was the only gene that displayed a marked increase in expression with ATP supplementation, MMP-13 activity was examined in the cartilage cultures using a FRET based enzymatic assay ( 3.6). There was a dose-dependent effect of ATP supplementation on MMP-13 activity in the developed cartilaginous tissues (p < 0.05) (Figure 4-7). MMP-13 activity was increased by 19 ± 3% and 39 ± 7% with 62.5 and 250 µm doses of ATP, respectively (Figure 4-7). Figure 4-7: Effect of long term (4 week) ATP supplementation on MMP-13 activity. Data are presented as the mean ± SEM, n = 8 samples per group. * Denotes a significant difference between all other groups (p < 0.05) 64

77 4.4 Inorganic pyrophosphate homeostasis Levels of inorganic pyrophosphate (PPi) were measured in the tissue and the media surrounding tissue constructs after long-term culture (4 weeks) in order to elucidate the effects of ATP stimulation on PPi homeostasis. Media PPi was measured ( 3.7) in the conditioned culture media and normalized to the DNA content of the tissue. Tissue PPi was measured from the papain digests of cartilaginous tissues and normalized to DNA content. In terms of media PPi, there was a significant 54 ± 11% increase from all other doses with respect to the high dose of ATP (250 µm) (p < 0.01) (Figure 4-8). Media PPi was not significantly different from control with the 62.5 µm dose of ATP (Figure 4-8). 65

78 Figure 4-8: Effect of long-term ATP stimulation on inorganic pyrophosphate levels in the conditioned culture media surrounding the developed cartilaginous tissues. Data are presented as the mean ± SEM normalized to control, n = 14 samples per group. * Denotes a significant difference between all other groups (p < 0.05) Alternatively, tissue PPi did not appear to be affected by ATP stimulation exhibiting no significant difference among the ATP supplementation conditions (Figure 4-9). The baseline (control) level of PPi in the media and in the tissue varied with each experiment, but was higher in the media on average (~0.5 µg PPi/µg DNA, or 10 µm PPi without DNA normalization, in the media vs. ~0.17 µg PPi/µg DNA in the tissue). 66

79 Figure 4-9: Effect of long-term ATP stimulation on inorganic pyrophosphate levels in the developed cartilaginous tissues. Data are presented as the mean ± SEM normalized to control, n = 16 samples per group. 4.5 Release of inflammatory mediators Inflammatory mediators, PGE 2 and NO, were measured from the conditioned culture media over a 4 week period ( 3.8). There was no significant effect of ATP on the release of PGE 2 or NO. Both PGE 2 and NO were detected in the conditioned media at levels ~ 81 pg/ml and 20 µm, respectively which decreased by ~8 % and 6-fold, respectively over the culture period for all conditions (Figure 4-10). 67

80 Figure 4-10: Effect of long term ATP supplementation on the release of inflammatory mediators. Inflammatory mediators, PGE 2 and NO, were measured from the conditioned culture media after 4 weeks. Data are presented as the mean ± SEM, n = 6 samples per group. 4.6 Optimization of ATP dose In order to determine a therapeutic dose of ATP, a range of doses (0, 31.25, 62.5, 125 and 250 µm ATP) were supplemented to cartilage cultures over a 4-week period. Afterwards, cultures were either assessed in terms of collagen and proteoglycan synthesis by radioisotope incorporation ( 3.2) and MMP-13 activity ( 3.6) using a FRET based enzymatic assay. 68

81 In terms of collagen synthesis, [ 3 H] proline incorporation appeared to be slightly inhibited under the µm dose of ATP and promoted with the 62.5 µm dose of ATP, although these results reflected trends in the data (p < 0.09) (Figure 4-11). The 125 µm dose of ATP did not seem to elicit increases in [3H] proline incorporation in accordance with the dose effect observed with 62.5 µm ATP since it was not significantly different from control or the 62.5 µm dose (Figure 4-11). On the other hand, 250 µm ATP induced significantly more [3H] incorporation than control (p < 0.01), and exhibited an increased trend from the 62.5 µm dose (p < 0.08) (Figure 4-11). Analogous to [ 3 H] proline incorporation, [ 35 S] sulfate incorporation was only significantly increased from control in the 250 µm dose (p < 0.02) (Figure 4-11). Although not significantly different from control, the 62.5 µm and 125 µm doses exhibited similar increasing trends in [ 35 S] sulfate incorporation that did not seem to depend on dose (p < 0.08) (Figure 4-11). The µm dose was not significantly different from control, but was significantly different from all other doses (p < 0.01) (Figure 4-11). 69

82 Figure 4-11: ECM synthesis results for the optimization of ATP dose. Data are presented as the mean ± SEM normalized to control, n = 7-15 samples per group. * Denotes a significant difference from control (p < 0.05) ** Denotes a trend from control (p < 0.09) Comparable to the [ 3 H] proline and [ 35 S] incorporation data, MMP-13 activity was affected by the dose of ATP, however, MMP-13 activity did not always increase as the dose increased (Figure 4-12). Rather, it appeared that MMP-13 activity increased in a stepwise fashion as a result of 62.5 µm and 250 µm ATP. There was no significant difference in MMP-13 activity between the µm dose of ATP and control, whereas 62.5 µm ATP induced a significant (p < 0.01), 20 ± 4% increase in MMP-13 activity in relation to the unstimulated controls. While the 125 µm dose of ATP displayed an 70

83 increasing trend from control for MMP-13 activity of 13 ± 3 % (p < 0.08), it was not significantly different from 62.5 µm ATP. Stimulation with 250 µm ATP elicited a significant, 41 ± 6% increase with respect to all other doses of ATP (p < 0.01). Figure 4-12: MMP-13 activity optimization. Data are presented as the mean ± SEM normalized to control, n = 8 samples per group. Denotes a significant difference from all other doses (p < 0.01) * Denotes a significant difference from control (p < 0.05); ** Denotes a trend from control (p < 0.08) In order to better visualize the changes in the balance between the anabolic and catabolic responses of the developed tissues, the ratio between ECM synthesis and MMP-13 activity was plotted (Figure 4-13). Between 62.5 and 250 µm ATP, the ratio between ECM synthesis (collagen and proteoglycan) and MMP-13 activity of the developed 71

84 cartilaginous tissues appeared to be greater than or equal to 1, except in the case of 125 µm ATP on [ 3 H] proline incorporation/mmp-13 activity (Figure 4-13). Figure 4-13: Changes in the anabolic and catabolic effects of exogenous ATP on engineered cartilage constructs. Data is presented as the mean [ 3 H] proline or [ 35 S] sulfate incorporation (CPM/µg DNA) with respect to MMP-13 protein activity (µg substrate hydrolyzed / µg DNA), normalized to control. 72

85 Chapter 5 Discussion This study demonstrates that the anabolic effects of mechanical stimuli can be achieved in the absence of externally applied forces through the supplementation of exogenous ATP. ATP plays a pivotal role in chondrocyte extracellular communication, acting as a signaling molecule in the purinergic mechanotransduction cascade [30]. Chondrocytes release ATP as a result of mechanical loading where it binds to the P2 receptor and stimulates ECM gene expression and matrix synthesis [95, 97]. In accordance with other published studies, the effects of exogenous ATP supplementation on the development of engineered cartilaginous tissue constructs in vitro was shown to be both anabolic and catabolic in nature [103, 162]. These anabolic and catabolic responses depended on the dose applied, and also occurred in concert with one another. This thesis was separated into three objectives. The first objective determined the dose range of ATP that would elicit an anabolic tissue response. The second objective was centred on the determination of the dose range of ATP that would elicit a (negative) catabolic, inflammatory and/or mineralization response(s). The third and final objective was to determine an optimal, therapeutic dose of ATP for engineered cartilage cultures. Each objective was accomplished and the results are discussed below. 73

86 5.1 Purinergic receptor mediated synthetic effects of exogenous ATP Exogenous ATP had an anabolic effect on both collagen and proteoglycan synthesis at 250 µm, with a trend towards anabolism at the 62.5 µm dose. There was also no apparent difference between collagen and proteoglycan synthesis trends. The anabolic effects of ATP supplementation have only previously been observed in bovine chondrocytes cultured in high density pellets in a study performed by Croucher et al. [103]. However, the anabolic effect observed in this study was in response to a single 500 µm dose of ATP [103]. This dose was excluded from the present study since Croucher et al. observed inhibition of ECM accumulation with repeated doses of ATP at 500 µm and since both inflammatory and catabolic responses have been associated with doses of this magnitude [103, 117]. The effects of exogenous ATP on ECM synthesis in this study appear to be stimulated through the purinergic receptor pathway, as the response was abolished in the presence of the well-known P2 receptor antagonist, Reactive Blue 2. This observation is supported by Pingguan-Murphy et al. who used P2 receptor antagonists (Reactive Blue 2 or suramin) to prevent ATP-mediated increases in calcium mobilization as a result of cyclic compression [95]. However, the P2 antagonist may also be acting on other unknown pathways since overall ECM synthesis in the absence of ATP was also inhibited. Unfortunately, there is currently no specific inhibitor of the P2Y 2 receptor, a receptor that has been linked with ATP binding and subsequent mechanotransduction pathways [95]. It has been estimated that the level of ATP in the extracellular space immediately surrounding chondrocytes as a result of mechanical 74

87 stimulation is on the order of 10 µm [30]. Perhaps, the doses used in the present study better represent the doses of ATP that chondrocytes are exposed to from mechanical loading rather than a dose of ATP that might represent cellular stress and subsequent catabolism (>500 µm). 5.2 Long-term effects of ATP supplementation on tissue formation and properties Since ATP supplementation of 62.5 and 250 µm promoted increases in ECM synthesis mediated through the purinergic receptor pathway, the long-term changes in biochemical, mechanical and physical properties of cartilaginous tissues cultured in response to ATP stimulation were then assessed (over 4 weeks of culture). The water content of native articular cartilage is approximately 75-80% [39, 40], which is consistent with the water content of all of the developed cartilaginous tissues, regardless of ATP supplementation (ranging from 77 81%). Despite all of the developed tissues having water contents within the range of native cartilage, long-term supplementation with 250 µm ATP resulting in a slightly, but significantly lower tissue water content (77%). Lower tissue water contents are consistent with lower levels of proteoglycans (which interact strongly with water) [13] and higher levels of collagen (which reduce hydration) [163] both of which were observed for the high dose of ATP (250 µm). In fact, ATP supplementation had a significant effect on collagen accumulation in a dosedependent manner, with increases ranging between 16-26%. Proteoglycan accumulation was not significantly different from control under the 62.5 µm ATP dose, 75

88 but it was ~27 % lower in response to 250 µm ATP. The accumulated ECM macromolecules (collagen and proteoglycans) in the tissue constructs (regardless of ATP supplementation) did not approach that of native cartilage, which makes up approximately 700 µg/mg and 200 µg/mg of the tissue s dry weight, respectively [23]. The amount of collagen per dry weight of tissue increased significantly from control with ATP supplementation to ~10 % of the level in native cartilage. Regardless, the increases in collagen observed in this study provide a step towards improving tissue properties of in vitro cartilaginous tissues which are often deficient in collagen [22, 23, 164]. The amount of proteoglycans per dry weight of tissue decreased significantly with 250 µm ATP to ~10% that of native cartilage, but accounted for ~20 % in the control and low ATP dose (62.5 µm). Even though they accounted for a fraction of native articular cartilage ECM components, the ratio of collagen-to-glycosaminoglycans in the developed tissues supplemented with ATP appeared to approach the ratio typical for native tissue (2:1 to 3:1 overall) [23, 165] in a dose-dependent manner (significant increase for the 250 µm condition, and an increasing trend for 62.5 µm). These increases in ECM accumulation as a result of ATP supplementation are likely due to an increased synthetic response from chondrocytes opposed to increased tissue cellularity as the DNA content remained relatively constant among the developed tissues, irrespective of ATP dose. ATP supplementation was also shown to considerably increase cartilaginous tissue thickness by 50 55%, independent of dose. The indentation modulus of the developed tissues was also substantially improved (5- to 6.5-fold increase) with ATP stimulation 76

89 compared to the unstimulated controls. Not only were the mechanical properties significantly enhanced, but they appear to approach the indentation modulus of native bovine articular cartilage tissue, which ranges from kpa [54, 55]. The improvements in mechanical properties were likely due to increased collagenproteoglycan interactions, which are known to increase tissue stiffness [34, 77]. Moderate changes in ECM accumulation, like those observed in this study as a result of exogenous ATP, can dramatically influence the thickness and mechanical properties of cartilaginous tissues. Cartilaginous tissues subjected to intermittent applications of shear or compression to chondrocytes cultured in a rotating wall bioreactor displayed relatively small increases in ECM accumulation in comparison to large increases that were observed for equilibrium modulus and thickness [25, 62, 77, 164]. A more uniform deposition of ECM as a result of ATP could also contribute to the increased mechanical properties since ATP is a soluble mediator that can potentially exert its affects on all chondrocytes resident in the cartilage constructs. Waldman et al. cultured bovine articular chondrocytes in high density on calcium polyphosphate substrates (a culture method similar to the one utilized in this study) and found increases in equilibrium modulus for compression and shear stimulated tissues that were 4 and 5.6 fold that of unstimulated cultures, respectively [62]. The increases in equilibrium modulus observed with ATP supplementation in this study (by 5- to 6.5-fold) are on par with that of shear stimulated tissues and surpass that of compression stimulated tissues observed by Waldman et al. [62]. These results are promising in promoting ATP as an alternative way 77

90 of exerting the beneficial effects of mechanical stimulation, without having to mechanically stimulate the tissue constructs. Histology and immunohistochemistry of the developed tissues showed that there were no obvious differences in tissue formation as a result of ATP supplementation, aside from a trend for increased tissue thickness that was observed previously in this study. Sulphated proteoglycans and type II collagen were detected in the ECM of the developed tissues with little presence of type I collagen staining; all of which are characteristics of normal articular cartilage formation [13]. 5.3 Long-term effects of ATP supplementation on catabolism, inflammation and mineralization Although the positive (anabolic) effects of ATP on cartilaginous tissue development have been outlined thus far, ATP is also known to elicit a variety of responses that can hinder normal tissue formation and function. Such responses to exogenous ATP that have been reported include: catabolism, release of inflammatory mediators, dysregulation of inorganic pyrophosphate (PPi) and mineralization [117, 119, 128, 129, 162]. Matrix turnover is a process that can be triggered by ATP supplementation ( ) [162]. As part of the limited, natural repair response of articular chondrocytes, matrix turnover involves the break down of collagen and proteoglycans by matrix degrading 78

91 enzymes in order to generate new ECM components. Increased matrix turnover, however, can lead to the degeneration of articular cartilage. In particular, increased MMP-13 gene expression (which degrades the main type of collagen (type II) in articular cartilage) is implicated in OA [166]. The synthetic response of ATP on the ECM of developed cartilaginous tissues translated to greater collagen synthesis in a dosedependent manner; however, overall long-term increases in collagen accumulation were lower in comparison (23-43% increased synthesis vs % increased accumulation). Additionally, the short-term increase in proteoglycan synthesis did not appear to correspond to the long-term decrease in proteoglycan accumulation in response to high doses of exogenous ATP (250 µm). The overall lover levels of ECM accumulation, compared to ECM synthesis, could be attributed to increased matrix turnover. For this reason, the expression of matrix turnover genes was investigated in response to long term exposure to ATP. Ten matrix turnover genes were investigated: MMP 1, 3, 13, ADAMTS 1, 4, 5 and TIMP 1 4. The only gene that was differentially expressed with ATP stimulation was MMP-13, which increased over 2-fold in response to 250 µm of ATP. Although not statistically significant, 250 µm ATP also elicited an increasing trend in ADAMTS-4 gene expression. MMP-13 (collagenase-3) primarily degrades collagen type II, but like other MMPs it can also degrade aggrecan at a different site than the aggrecanases (the Asn 341 Phe 342 bond) [167]. ADAMTS-4 (aggrecanase-1) is one of the primary genes associated with the degradation of aggrecan [105]. These increases in gene expression in response to 250 µm ATP appeared to correlate well with the decrease 79

92 in proteoglycan accumulation observed in this study at the same dose. Increased MMP- 13 and ADAMTS-4 gene expression could also potentially account for the lack of ECM accumulation that was observed upon repeated doses of 500 µm ATP in bovine chondrocyte pellet cultures [103], provided that increases in gene expression continue with the concentration of ATP. It is likely that the increased MMP-13 gene expression is responsible for the lower levels of ECM accumulation in comparison to the short-term increases in ECM synthesis in response to exogenous ATP. To confirm this assertion, MMP-13 protein activity was also explored. Similar to the observed changes in gene expression, exogenous ATP increased MMP-13 protein activity in a dose-dependent manner implicating MMP-13 as a prime candidate for eliciting matrix turnover in response to ATP stimulation. Although inflammatory mediator release has been connected to ATP signaling in previous studies, it could be a secondary effect of increased matrix turn-over at higher doses (> 500 µm) rather than a precursor for catabolism [117] ( ). This would explain how the release of inflammatory mediators, PGE 2 and NO, was not affected by ATP supplementation ( 4.5). Consequently, inflammation was ruled out as a cause of catabolism in the developed tissues for the range of doses of ATP supplied in this study. 80

93 Inorganic pyrophosphate homeostasis is critical to the maintenance of healthy articular cartilage as excess extracellular PPi (eppi) contributes to pathological mineralization ( ) [121]. Development of calcium pyrophosphate dihydrate (CPPD) crystal deposition disease is characterized by an increased presence of PPi in the extracellular space due to increased ATP degradation and/or by increased transport of PPi out of the cell via the ANK clearance receptor [119, 121]. Extracellular PPi excess has also been shown previously to induce MMP-13 gene expression in monolayers of bovine articular chondrocytes [168]. PPi, measured from the conditioned culture media after 4 weeks of nucleotide exposure, was significantly increased in response to ATP stimulation ( 4.4). This is consistent with increased extracellular PPi (eppi) detected from conditioned culture media surrounding both porcine explants and chondrocytes cultured in monolayer as a result of P2 receptor signaling [169]. There is a very narrow range for the physiological eppi concentration in normal articular cartilage with synovial fluid in the knee containing 10 ± 0.5 µm PPi (determined over 50 individuals) [170]. The average basal level of eppi detected in the conditioned media in the present investigation was similar to these reported values at around 10 µm. Since ATP supplementation induced increased MMP-13 gene expression and activity along with the increased presence of eppi in the present study, excess eppi could be contributing to the catabolic effects observed under high doses of ATP supplementation. PPi levels from the developed tissue digests were also investigated in order to gain a better understanding of PPi homeostasis. No significant differences were observed in the amount of PPi present in the tissue 81

94 digests, which indicates that the excess PPi was somehow being utilized. This concept is discussed in greater detail in 5.5. Exogenous ATP has been previously associated with cartilage mineralization ( ). CPPD crystal deposition is primarily noted as a result of ATP supplementation [ ]. Hydroxyapatite mineralization is also pathological in articular cartilage as it is associated with chondrocyte differentiation and bone development [34]. Therefore, the potential effect of ATP on mineralization was assessed in the developed cartilaginous tissues. There was no apparent histological evidence of hydroxyapatite mineralization from the von Kossa stain. Even though hydroxyapatite mineralization was not identified by von Kossa staining, this does not rule out the potential for CPPD crystal deposition since pyrophosphate is detectable by this technique [171]. Transmission electron microscopy (TEM) was used to confirm the presence of both CPPD and hydroxyapatite mineralization in the developed tissues. Again, no evidence of mineralization was observed. However, the detection of CPPD and hydroxyapatite crystals is often fraught with difficulty [123, 172]. Crystals can be small and sparse, requiring more sophisticated methods of analysis [172]. It might also be possible that amorphous forms of calcium pyrophosphate (as there are more than 30 types) caused by excess PPi are eliciting catabolic responses from chondrocytes [173]. Most studies that positively identified CPPD crystals in articular cartilage as a result of exogenous ATP supplementation utilized significantly greater doses than was used in the present study (> 1mM ATP) 82

95 [128, 130, 174]. However, one study noted an increase in 45 Ca retention by articular cartilage explants under 100 µm of ATP (a dose within the range used in the present study) and eluded to the presence of CPPD despite not being able to characterize electron dense deposits due to sublimation and small size of the crystals [124]. 5.4 The balance of anabolic and catabolic responses and the optimization of ATP dose From the results discussed thus far it appeared that a therapeutic dose of ATP to optimally stimulate engineered cartilage constructs would exist between 62.5 µm and 250 µm ATP, where the anabolic effects would potentially outweigh the catabolic effects. In order to find an optimal dose of ATP to supply to developing cartilaginous tissues in vitro, doses that flanked 62.5 µm were investigated (31.25 and 125 µm). MMP-13 activity was chosen as a measure of catabolism and ECM synthesis was chosen as a measure of anabolism since both techniques displayed clear, dose-dependent effects on the developed cartilaginous tissues. There are, however, limitations in this approach as MMP-13 activity likely does not reflect all of the catabolic processes at work. Similarly, ECM synthesis also only reflects the anabolic response over a 24 hour period after ATP exposure. For this reason, conclusions were derived by comparing patterns of MMP-13 activity and ECM synthesis with the long-term tissue formation and properties data observed previously ( 4.2). 83

96 The ratio between ECM synthesis and MMP-13 activity as a result of µm ATP is on par with unstimulated controls ( 4.6; Figure 4-13), yet improved mechanical properties were observed at 62.5 and 250 µm. Although the long-term effects of ATP were not investigated at 125 µm, the mechanical properties as a result of 62.5 and 250 µm did not significantly differ from each other, leading to the assumption that 125 µm ATP would elicit similar increases in mechanical properties. Therefore, instead of discovering a therapeutic dose of ATP, a therapeutic dose range of ATP between 62.5 and 125 µm has emerged for developing cartilaginous tissues. Since 250 µm ATP exhibits MMP-13 gene expression and activity that is considerably higher than all other doses, which contributed to lower overall proteoglycan accumulation, it was excluded from the therapeutic dose range. Despite the many beneficial effects of ATP that have been noted, increases in MMP-13 activity seem to be inevitable as a result of ATP supplementation. Understanding the mechanism for exogenous ATP action on the development of cartilaginous tissues is important in order to further minimize the catabolic effects in order to develop more functional tissue. 84

97 5.5 A possible mechanism for exogenous ATP action on the development of cartilaginous tissues After ATP binding to the P2 receptor, anabolic and catabolic events were observed to occur in tandem. However, it seems unlikely that these effects were transduced through the same intracellular pathway due to a lack of evidence supporting a direct catabolic pathway as a result of ATP binding, as well as the noted involvement of PPi in inducing MMP-13 gene expression [168]. Although PPi seems like the probable candidate for eliciting catabolism, its method of action is unclear as there is no direct route of entry for PPi into the cell [119]. ANK is the only channel associated with PPi transport, functioning solely as a clearance receptor to expel accumulated intracellular PPi [119, 121]. Alternatively, PPi mineralization may be a possible mechanism for the observed catabolic response as a result of ATP supplementation. CPPD has been shown to spontaneously form at relatively low concentrations of PPi (>10 µm) in the presence of calcium (and influenced by ph and the concentration of sodium and magnesium) [175]. These CPPD crystals can then bind to toll-like receptors (specifically TLR2) on the plasma membrane and elicit MMP-13 gene expression [168, 176]. However, as CPPD crystals were not detected by TEM analysis of the ATP supplemented tissues it may be possible that the CPPD crystals were endocytosed and subsequently degraded. Both CPPD and hydroxyapatite crystals can be endocytosed by chondrocytes [177]. Additionally, small hydroxyapatite crystals are more likely to be endocytosed than larger crystals (17 µm vs. 46 or 105 µm) and have also been shown to preferentially induce 85

98 metalloproteinase (MMP) activity in fibroblasts [178]. Analysis of the synovial fluid is a common clinical practice for identifying chondrocalcinosis (pathological presence of CPPD crystals in the joint space) [125, 179]. CPPD and hydroxyapatite crystals have also been detected in the synovial fluid of osteoarthritic patients without eliciting an inflammatory response [179] a result which was also demonstrated in this study. There is also evidence that links increased eppi with changes in the chondrocyte phenotype [180]. Chondrocyte hypertrophy is characterized by enlarged chondrocytes that express higher levels of MMP-13, collagen X and the bone marker, cbfa1 [34]. Hypertrophic chondrocytes are found in the deep zone of cartilage directly above the region where calcification occurs [34]. Chondrocyte hypertrophy is associated with dysregulated matrix synthesis, increased PPi generation [181] and heightened production of matrix vesicles (which are equipped for mineralization) [180, 182]. Chondrocyte hypertrophy could occur as a result of CPPD crystal deposition and the induction of MMPs, as CPPD crystal deposits have been shown to surround hypertrophic chondrocytes [180]. A better understanding of the mechanism of action for ATP and its degradation products is necessary in order to take advantage of the anabolic effects while keeping catabolic effects at bay. Despite the lack of direct identification, CPPD crystals are most likely responsible for the catabolic effects noted in the present study, which could also result in chondrocyte hypertrophy. The involvement of chondrocyte hypertrophy and CPPD crystal formation need to be investigated in more detail before 86

99 implementing ATP supplementation as a therapeutic method of developing cartilaginous tissues. 87

100 Chapter 6 Conclusions and Recommendations 6.1 Conclusions Overall perspective For the first time, exogenous ATP has been utilized to improve the biochemical and mechanical properties of cartilaginous tissues developed in vitro for the role it plays in chondrocyte mechanotransduction. Previous studies have established that exogenous ATP can elicit various effects on chondrocytes and their turnover of extracellular matrix (ECM) components [102, 103, 129, 162]. Some of these effects were echoed in this thesis, as ATP was also shown to cause catabolic effects, demonstrated by increases in MMP-13 gene expression and activity, along with disruptions in the homeostasis of inorganic pyrophosphate (PPi) an ATP degradation product. Other known effects of ATP were not observed, such as hydroxyapatite and calcium pyrophosphate dihydrate (CPPD) mineral formation and inflammation (characterized by PGE 2 and NO release). The dose of ATP seems to be a significant factor in determining the extent of purinergic receptor-mediated anabolic effects, in addition to catabolic effects which appear to be a result of excess extracellular PPi. Except, rather than having a single, optimal dose of ATP, a range of doses of ATP were shown to exert similar, therapeutic effects. In general, it seems that maintaining a balance of ATP degradation products is essential to the production and preservation of healthy cartilage. What this exact balance is has yet to 88

101 be elucidated, but it is clear that ATP supplementation of cartilaginous tissues grown in vitro is beneficial towards developing more functional tissue constructs. The first two objectives of this thesis were to determine doses of exogenous ATP that elicited anabolic and catabolic effects. Implicit in these objectives was the hypothesis that there would be an onset of catabolic effects distinct from the onset of anabolic effects. What was observed was contrary to this hypothesis in that anabolic effects (ECM synthesis and accumulation) occurred in conjunction with catabolic effects (MMP-13 activity). This was illustrated most clearly in the results obtained from completion of the third objective: the determination of an optimal, therapeutic dose of ATP to add to cartilaginous cultures Anabolic versus catabolic effects From this investigation, it has been demonstrated that exogenous ATP between µm can promote ECM synthesis in a dose-dependent manner that is mediated by the P2 receptor signaling cascade. These positive effects manifest themselves in the long-term accumulation of collagen in developed tissues cultured in the presence of exogenous ATP as well, but to a lesser degree. On the other hand, proteoglycan accumulation was not significantly affected by ATP stimulation. Matrix turnover provided a probable explanation for these discrepancies between collagen and proteoglycan synthesis in the short-term, and accumulation in the long-term. Not only was increased MMP-13 gene expression and protein activity characteristic under the high (250 µm) dose of ATP, but 89

102 MMP-13 activity increased as a result of ATP stimulation ( µm). ATP supplemented cartilaginous tissues were also significantly thicker and possessed improved mechanical properties A therapeutic dose range of ATP for cartilaginous tissues Knowledge of the anabolic and catabolic effects of exogenous ATP can be utilized to create more functional tissue constructs. The therapeutic dose range of ATP was determined based on trends observed in two areas ECM synthesis and MMP-13 activity. It was determined that a range of doses between µm elicited more overall anabolic effects, which were seen to contribute to improved mechanical properties (6.5-fold increase with 62.5 µm ATP). 6.2 Recommendations The ultimate goal of tissue engineering is to re-create human tissue for eventual implantation into a patient. Investigation of a therapeutic dose range of ATP for the development of human cartilaginous tissue in vitro would be beneficial, as bovine articular chondrocytes could exhibit different responses compared to articular chondrocytes from different species. Human tissue is substantially thicker than native bovine articular cartilage and exhibits slightly different mechanical properties as a result ( 2.2) [183]. 90

103 In the present study, the indentation modulus of the developed cartilaginous tissues was assessed by unconfined indentation tests. Alternate testing methods, such as confined compression testing, should also be employed for future mechanical testing of the developed cartilaginous tissues as indentation testing of articular cartilage relies on several assumptions. On a similar note, the indentation modulus determined in this study was remarkably affected by ATP supplementation. Explanations for this substantial change in modulus were discussed previously; however, more work is needed to determine why such small changes in matrix accumulation could be associated with large changes in mechanical properties. Although the results of exogenous ATP supplementation on developing cartilaginous tissues are promising, further information pertaining to the mechanism of action of ATP and its degradation products should be determined before this therapeutic strategy is implemented. Extracellular PPi (eppi) has been linked to the catabolic effects of ATP demonstrated in this study. Similarly, chondrocyte hypertrophy has also been linked to increases in eppi. Although eppi excess was initiated by ATP degradation, changes in the chondrocyte phenotype could be occurring as a result. Gene expression of collagen X and cbfa1 (markers of chondrocyte hypertrophy) could confirm whether or not 91

104 chondrocyte hypertrophy is occurring and lead to a better understanding of ATPmediated effects. An interesting extension of this study relates to the management of extracellular inorganic pyrophosphate (eppi) levels in conjunction with ATP supplementation. Metal ion complexes have recently been shown to be ideal binding sites for PPi in aqueous solutions [184]. Although these metal-ppi complexes are currently used for the detection of PPi, removing excess eppi through the formation of a metal-ppi complex could be a possible means to managing eppi levels. One such complex includes Zn 2+ and bis(2- pyridylmethyl)amine (DPA), which can bind PPi approximately 1000 times more tightly than inorganic phosphate (Pi) in conditions of physiological ph [184, 185]. However, the effects of Zn 2+ and bis(2-pyridylmethyl)amine (DPA) on chondrocytes is presently unknown. Finding a complex that specifically binds to PPi and does not adversely affect cells may be extremely difficult since removal of the metal complex would not occur easily. Degrading excess PPi could be a potential method to manage eppi levels caused by high doses of exogenous ATP; however, all of the enzymes that degrade PPi also degrade ATP ( ). Alternatively, inhibiting mineralization may be a preferred method to manage the effects of excess PPi, by preventing the catabolic effects induced by CPPD crystals. Phosphocitrate, a naturally occurring inhibitor of mineralization has been used in many in vitro studies to inhibit CPPD and BCP crystal formation [186]. Phosphocitrate acts by restricting the nucleation, growth and aggregation of calcium salts 92

105 [186] and has also been shown to block crystal induced MMP synthesis [187]. Using phosphocitrate in conjunction with ATP could be an even more effective way of developing more functional cartilaginous tissue in vitro as the catabolic effects of ATP induced by eppi and calcium crystal formation could be inhibited. 93

106 References [1] B. Palsson and S. N. Bhatia, "Tissue Engineering. 2004, Upper Sadle River. [2] R. J. Wilkins, "Chondrocyte regulation by mechanical load," Biorheology, vol. 37, pp , [3] M. Tsuzaki, D. Bynum, L. Almekinders, J. Faber and A. J. Banes, "Mechanical loading stimulates ecto-atpase activity in human tendon cells," J. Cell. Biochem., vol. 96, pp , [4] T. Aigner and L. McKenna, "Molecular pathology and pathobiology of osteoarthritic cartilage," Cellular and Molecular Life Sciences, vol. 59, pp. 5-18, [5] L. J. Sandell and T. Aigner, "Articular cartilage and changes in arthritis. An introduction: cell biology of osteoarthritis," Arthritis Res., vol. 3, pp , [6] The Arthritis Society, "Osteoarthritis," 2010 (06/15). [7] E. Hunziker, "Articular cartilage repair: basic science and clinical progress. A review of the current status and prospects," Osteoarthritis and Cartilage, vol. 10, pp , [8] A. C. McLaren, C. P. Blokker, P. J. Fowler, J. N. Roth and M. G. Rock, "Arthroscopic debridement of the knee for osteoarthrosis," Can. J. Surg., vol. 34, pp , [9] S. R. Tew, A. P. L. Kwan, A. Hann, B. M. Thomson and C. W. Archer, "The reactions of articular cartilage to experimental wounding: role of apoptosis," Arthritis & Rheumatism, vol. 43, pp , [10] A. P. Newman, "Articular cartilage repair," Am. J. Sports Med., vol. 26, pp , [11] J. E. Gilbert, "Current treatment options for the restoration of articular cartilage," Am. J. Knee Surg., vol. 11, pp , [12] E. Bell, Ed., Tissue Engineering in Perspective. New York: Academic Press,

107 [13] J. S. Temenoff and A. G. Mikos, "Review: tissue engineering for regeneration of articular cartilage," Biomaterials, vol. 21, pp , [14] W. W. Tomford, D. S. Springfield and H. J. Mankin, "Fresh and frozen articular cartilage allografts," Orthopedics, vol. 15, pp , [15] F. Djouad, B. Delorme, M. Maurice, C. Bony, F. Apparailly, P. Louis-Plence, F. Canovas, P. Charbord, D. Noël and C. Jorgensen, "Microenvironmental changes during differentiation of mesenchymal stem cells towards chondrocytes," Arthritis Research and Therapy, vol. 9, pp , [16] M. V. Risbud and M. Sittinger, "Tissue engineering: advances in in vitro cartilage generation," Trends Biotechnol., vol. 20, pp , [17] F. Guilak and V. C. Mow, "The mechanical environment of the chondrocyte: a biphasic finite element model of cell-matrix interactions in articular cartilage," J. Biomech., vol. 33, pp , [18] N. C. Zanetti and M. Solursh, "Effect of cell shape on cartilage differentiation," Cell Shape: Determinants, Regulation, and Regulatory Role, pp , [19] P. M. Van der Kraan, P. Buma, T. Van Kuppevelt and W. B. Van den Berg, "Interaction of chondrocytes, extracellular matrix and growth factors: relevance for articular cartilage tissue engineering," Osteoarthritis and Cartilage, vol. 10, pp , [20] C. Lee, S. Grad, M. Wimmer and M. Alini, "The influence of mechanical stimuli on articular cartilage tissue engineering," Topics in Tissue Engineering, vol. 2, [21] M. D. Buschmann, Y. A. Gluzband, A. J. Grodzinsky and E. B. Hunziker, "Mechanical compression modulates matrix biosynthesis in chondrocyte/agarose culture," J. Cell. Sci., vol. 108, pp , [22] G. Vunjak-Novakovic, I. Martin, B. Obradovic, S. Treppo, A. J. Grodzinsky, R. Langer and L. E. Freed, "Bioreactor cultivation conditions modulate the composition and mechanical properties of tissue-engineered cartilage," Journal of Orthopaedic Research, vol. 17, pp , [23] S. D. Waldman, M. D. Grynpas, R. M. Pilliar and R. A. Kandel, "Characterization of cartilagenous tissue formed on calcium polyphosphate substrates in vitro," J. Biomed. Mater. Res., vol. 62, pp ,

108 [24] M. Jin, E. H. Frank, T. M. Quinn, E. B. Hunziker and A. J. Grodzinsky, "Tissue Shear Deformation Stimulates Proteoglycan and Protein Biosynthesis in Bovine Cartilage Explants," Arch. Biochem. Biophys., vol. 395, pp , [25] S. D. Waldman, C. G. Spiteri, M. D. Grynpas, R. M. Pilliar and R. A. Kandel, "Long-term intermittent compressive stimulation improves the composition and mechanical properties of tissue-engineered cartilage," Tissue Eng., vol. 10, pp , [26] Y. J. Kim, R. L. Y. Sah, A. J. Grodzinsky, A. H. K. Plaas and J. D. Sandy, "Mechanical regulation of cartilage biosynthetic behavior: physical stimuli," Arch. Biochem. Biophys., vol. 311, pp. 1-12, [27] E. J. Vanderploeg, C. G. Wilson and M. E. Levenston, "Articular chondrocytes derived from distinct tissue zones differentially respond to in vitro oscillatory tensile loading," Osteoarthritis and Cartilage, vol. 16, pp , [28] D. A. Lee, T. Noguchib, S. P. Freanc, P. Leesc and D. L. Badera, "The influence of mechanical loading on isolated chondrocytes seeded in agarose constructs," Biorheology, vol. 37, pp , [29] M. Knight, S. McGlashan, M. Garcia, C. Jensen and C. Poole, "Articular chondrocytes express connexin 43 hemichannels and P2 receptors a putative mechanoreceptor complex involving the primary cilium?" J. Anat., vol. 214, pp , [30] R. D. Graff, E. R. Lazarowski, A. J. Banes and G. M. Lee, "ATP release by mechanically loaded porcine chondrons in pellet culture," Arthritis Rheum., vol. 43, pp , [31] C. T. Hung, E. G. Lima, R. L. Mauck, E. Taki, M. A. LeRoux, H. H. Lu, R. G. Stark, X. E. Guo and G. A. Ateshian, "Anatomically shaped osteochondral constructs for articular cartilage repair," J. Biomech., vol. 36, pp , [32] E. H. Han, W. C. Bae, N. D. Hsieh-Bonassera, V. W. Wong, B. L. Schumacher, S. Görtz, K. Masuda, W. D. Bugbee and R. L. Sah, "Shaped, stratified, scaffold-free grafts for articular cartilage defects," Clinical Orthopaedics and Related Research, vol. 466, pp , [33] J. J. Ballyns, J. P. Gleghorn, V. Niebrzydowski, J. J. Rawlinson, H. G. Potter, S. A. Maher, T. M. Wright and L. J. Bonassar, "Image-guided tissue engineering of 96

109 anatomically shaped implants via MRI and micro-ct using injection molding," Tissue Engineering Part A, vol. 14, pp , [34] J. Buckwalter, T. Einhorn and S. Simon, "American Academy of Orthopaedic Surgeons," Orthopaedic Basic Science Biology and Biomechanics of the Musculoskeletal System. Rosemont, IL: American Academy of Orthopaedic Surgeons, [35] S. J. Millward-Sadler, M. O. Wright, P. W. Flatman and D. M. Salter, "ATP in the mechanotransduction pathway of normal human chondrocytes," Biorheology, vol. 41, pp , [36] D. Nesic, R. Whiteside, M. Brittberg, D. Wendt, I. Martin and P. Mainil-Varlet, "Cartilage tissue engineering for degenerative joint disease," Adv. Drug Deliv. Rev., vol. 58, pp , [37] F. H. Netter, Atlas of Human Anatomy. East Hanover: Novartis, [38] A. Clark, L. Barclay, J. Matyas and W. Herzog, "In situ chondrocyte deformation with physiological compression of the feline patellofemoral joint," J. Biomech., vol. 36, pp , [39] F. Guilak, "Compression-induced changes in the shape and volume of the chondrocyte nucleus," J. Biomech., vol. 28, pp , [40] M. Wong and D. R. Carter, "Articular cartilage functional histomorphology and mechanobiology: a research perspective," Bone, vol. 33, pp. 1-13, [41] V. Mow and G. Ateshian, "In Lubrication and Wear of Diarthrodial Joints: Basic Orthopaedic Biomechanics; Mow, VC; Hayes, WC, Eds," [42] D. D. Sun, X. E. Guo, M. Likhitpanichkul, W. M. Lai and V. C. Mow, "The influence of the fixed negative charges on mechanical and electrical behaviors of articular cartilage under unconfined compression," J. Biomech. Eng., vol. 126, pp. 6-16, [43] N. P. Cohen, R. J. Foster and V. C. Mow, "Composition and dynamics of articular cartilage: structure, function and maintaining healthy state," J Orthop Sports Phys Ther, vol. 28, pp , [44] C. Summer, "Proteoglycans," 2010 (06/15). 97

110 [45] F. Guilak, R. L. Sah and L. A. Setton, "Physical regulation of cartilage metabolism," in Basic Orthopaedic Biomechanics, V. C. Mow and W. C. Hayes, Eds. Philadelphia: Lippincott-Raven, [46] K. von der Mark, V. Gauss, H. von der Mark and P. Müller, "Relationship between cell shape and type of collagen synthesised as chondrocytes lose their cartilage phenotype in culture," [47] M. Ulrich-Vinther, M. D. Maloney, E. M. Schwarz, R. Rosier and R. J. O'Keefe, "Articular cartilage biology," J. Am. Acad. Orthop. Surg., vol. 11, pp , [48] I. C. Clarke, "Articular cartilage: a review and scanning electron microscope study: 1. The interterritorial fibrillar architecture," Journal of Bone & Joint Surgery, British Volume, vol. 53, pp , [49] V. C. Mow and C. C. B. Wang, "Some bioengineering considerations for tissue engineering of articular cartilage," Clin. Orthop., vol. 367, pp. S204, [50] C. R. Flannery, C. E. Hughes, B. L. Schumacher, D. Tudor, M. B. Aydelotte, K. E. Kuettner and B. Caterson, "Articular cartilage superficial zone protein (SZP) is homologous to megakaryocyte stimulating factor precursor and is a multifunctional proteoglycan with potential growth-promoting, cytoprotective, and lubricating properties in cartilage metabolism," Biochem. Biophys. Res. Commun., vol. 254, pp , [51] A. J. Almarza and K. A. Athanasiou, "Design characteristics for the tissue engineering of cartilaginous tissues," Ann. Biomed. Eng., vol. 32, pp. 2-17, [52] D. C. Fithian, M. A. Kelly and V. C. Mow, "Material properties and structurefunction relationships in the menisci," Clin. Orthop., vol. 252, pp , [53] S. Akizuki, V. C. Mow, F. Müller, J. C. Pita, D. S. Howell and D. H. Manicourt, "Tensile properties of human knee joint cartilage: I. Influence of ionic conditions, weight bearing, and fibrillation on the tensile modulus," Journal of Orthopaedic Research, vol. 4, pp , [54] R. K. Korhonen, M. S. Laasanen, J. Töyräs, J. Rieppo, J. Hirvonen, H. J. Helminen and J. S. Jurvelin, "Comparison of the equilibrium response of articular cartilage in unconfined compression, confined compression and indentation," J. Biomech., vol. 35, pp ,

111 [55] N. K. Simha, H. Jin, M. L. Hall, S. Chiravarambath and J. L. Lewis, "Effect of indenter size on elastic modulus of cartilage measured by indentation," J. Biomech. Eng., vol. 129, pp. 767, [56] V. C. Mow, S. C. Kuei, W. M. Lai and C. G. Armstrong, "Biphasic creep and stress relaxation of articular cartilage in compression: theory and experiments," J. Biomech. Eng., vol. 102, pp , [57] L. A. Setton, W. Zhu and V. C. Mow, "The biphasic poroviscoelastic behavior of articular cartilage: Role of the surface zone in governing the compressive behavior," J. Biomech., vol. 26, pp , [58] A. J. Grodzinsky, M. E. Levenston, M. Jin and E. H. Frank, "Cartilage Tissue Remodeling in Response to Mechanical Forces," Annu. Rev. Biomed. Eng., vol. 2, pp , [59] T. M. Quinn, A. J. Grodzinsky, M. D. Buschmann, Y. J. Kim and E. B. Hunziker, "Mechanical compression alters proteoglycan deposition and matrix deformation around individual cells in cartilage explants," J. Cell. Sci., vol. 111, pp , [60] L. J. Bonassar, A. J. Grodzinsky, E. H. Frank, S. G. Davila, N. R. Bhaktav and S. B. Trippel, Journal of Orthopaedic Research, vol. 19, pp , [61] S. H. Elder, S. A. Goldstein, J. H. Kimura, L. J. Soslowsky and D. M. Spengler, "Chondrocyte differentiation is modulated by frequency and duration of cyclic compressive loading," Ann. Biomed. Eng., vol. 29, pp , [62] S. D. Waldman, C. G. Spiteri, M. D. Grynpas, R. M. Pilliar, J. Hong and R. A. Kandel, "Effect of Biomechanical Conditioning on Cartilaginous Tissue Formation in Vitro," J. Bone Joint Surg. Am., vol. 85, pp , [63] R. L. Mauck, M. A. Soltz, C. C. Wang, D. D. Wong, P. H. Chao, W. B. Valhmu, C. T. Hung and G. A. Ateshian, "Functional tissue engineering of articular cartilage through dynamic loading of chondrocyte-seeded agarose gels," J. Biomech. Eng., vol. 122, pp , [64] J. D. Kisiday, M. Jin, M. A. DiMicco, B. Kurz and A. J. Grodzinsky, "Effects of dynamic compressive loading on chondrocyte biosynthesis in self-assembling peptide scaffolds," J. Biomech., vol. 37, pp ,

112 [65] O. Demarteau, D. Wendt, A. Braccini, M. Jakob, D. Schäfer, M. Heberer and I. Martin, "Dynamic compression of cartilage constructs engineered from expanded human articular chondrocytes," Biochem. Biophys. Res. Commun., vol. 310, pp , [66] O. Schmidt, J. Mizrahi, J. Elisseeff and D. Seliktar, "Immobilized fibrinogen in PEG hydrogels does not improve chondrocyte-mediated matrix deposition in response to mechanical stimulation," Biotechnol. Bioeng., vol. 95, pp , [67] A. C. Chen and R. L. Sah, "Effect of static compression on proteoglycan biosynthesis by chondrocytes transplanted to articular cartilage in vitro," Journal of Orthopaedic Research, vol. 16, pp , [68] L. J. Bonassar, A. J. Grodzinsky, A. Srinivasan, S. G. Davila and S. B. Trippel, "Mechanical and physicochemical regulation of the action of insulin-like growth factor-i on articular cartilage," Arch. Biochem. Biophys., vol. 379, pp , [69] C. R. Lee, A. J. Grodzinsky and M. Spector, "Biosynthetic response of passaged chondrocytes in a type II collagen scaffold to mechanical compression," Journal of Biomedical Materials Research Part A, vol. 64, pp , [70] P. A. Torzilli, R. Grigiene, C. Huang, S. M. Friedman, S. B. Doty, A. L. Boskey and G. Lust, "Characterization of cartilage metabolic response to static and dynamic stress using a mechanical explant test system," J. Biomech., vol. 30, pp. 1-9, [71] M. Wong, P. Wuethrich, M. D. Buschmann, P. Eggli and E. Hunziker, "Chondrocyte biosynthesis correlates with local tissue strain in statically compressed adult articular cartilage," Journal of Orthopaedic Research, vol. 15, pp , [72] J. T. Connelly, E. J. Vanderploeg and M. E. Levenston, "The influence of cyclic tension amplitude on chondrocyte matrix synthesis: experimental and finite element analyses," Biorheology, vol. 41, pp , [73] E. J. Vanderploeg, S. M. Imler and K. R. Brodkin, "Oscillatory tension differentially modulates matrix metabolism and cytoskeletal organization in chondrocytes and fibrochondrocytes," J. Biomech., vol. 37, pp , [74] J. Huang, L. R. Ballou and K. A. Hasty, "Cyclic equibiaxial tensile strain induces both anabolic and catabolic responses in articular chondrocytes," Gene, vol. 404, pp , [75] S. Millward-Sadler, M. Wright, L. Davies, G. Nuki and D. Salter, "Mechanotransduction via integrins and interleukin-4 results in altered aggrecan and 100

113 matrix metalloproteinase 3 gene expression in normal, but not osteoarthritic, human articular chondrocytes," Arthritis & Rheumatism, vol. 43, pp , [76] J. C. Y. Fan and S. D. Waldman, "The Effect of Intermittent Static Biaxial Tensile Strains on Tissue Engineered Cartilage," Ann. Biomed. Eng., vol. 38, pp , [77] S. D. Waldman, C. G. Spiteri, M. D. Grynpas, R. M. Pilliar and R. A. Kandel, "Long-term intermittent shear deformation improves the quality of cartilaginous tissue formed in vitro," J. Orthop. Res., vol. 21, pp , [78] R. L. Smith, M. C. D. Trindade, T. Ikenoue, M. Mohtai, P. Das, D. R. Carter, S. B. Goodman and D. J. Schurman, "Effects of shear stress on articular chondrocyte metabolism," Biorheology, vol. 37, pp , [79] M. T. Raimondi, F. Boschetti, L. Falcone, G. B. Fiore, A. Remuzzi, E. Marinoni, M. Marazzi and R. Pietrabissa, "Mechanobiology of engineered cartilage cultured under a quantified fluid-dynamic environment," Biomechanics and Modeling in Mechanobiology, vol. 1, pp , [80] N. M. Bachrach, V. C. Mow and F. Guilak, "Incompressibility of the solid matrix of articular cartilage under high hydrostatic pressures," J. Biomech., vol. 31, pp , [81] J. J. Parkkinen, J. Ikonen, M. J. Lammi, J. Laakkonen, M. Tammi and H. J. Helminen, "Effects of cyclic hydrostatic pressure on proteoglycan synthesis in cultured chondrocytes and articular cartilage explants," Arch. Biochem. Biophys., vol. 300, pp , [82] R. L. Smith, S. F. Rusk, B. E. Ellison, P. Wessells, K. Tsuchiya, D. R. Carter, W. E. Caler, L. J. Sandell and D. J. Schurman, "In vitro stimulation of articular chondrocyte mrna and extracellular matrix synthesis by hydrostatic pressure," Journal of Orthopaedic Research, vol. 14, pp , [83] R. L. Smith, M. C. D. Trindade, T. Ikenoue, M. Mohtai, P. Das, D. R. Carter, S. B. Goodman and D. J. Schurman, "Effects of shear stress on articular chondrocyte metabolism," Biorheology, vol. 37, pp , [84] T. Toyoda, B. B. Seedhom, J. Q. Yao, J. Kirkham, S. Brookes and W. A. Bonass, "Hydrostatic pressure modulates proteoglycan metabolism in chondrocytes seeded in agarose," Arthritis & Rheumatism, vol. 48, pp ,

114 [85] R. B. Lee, R. J. Wilkins, S. Razaq and J. P. G. Urban, "The effect of mechanical stress on cartilage energy metabolism," Biorheology, vol. 39, pp , [86] A. J. Banes, G. Lee, R. Graff, C. Otey, J. Archambault, M. Tsuzaki, M. Elfervig and J. Qi, "Mechanical forces and signaling in connective tissue cells: cellular mechanisms of detection, transduction, and responses to mechanical deformation." Current Opinion in Orthopaedics, vol. 12, pp , [87] G. G. Ernstrom and M. Chalfie, "Genetics of sensory mechanotransduction," [88] A. Katsumi, A. W. Orr, E. Tzima and M. A. Schwartz, "Integrins in mechanotransduction," J. Biol. Chem., vol. 279, pp , [89] S. J. Kim, E. J. Kim, Y. Н. Kim, S. B. Hahn and J. W. Lee, "The Modulation of Integrin Expression by the Extracellular Matrix in Articular Chondrocytes," Yonsei Med. J., vol. 44, pp , [90] S. J. Millward-Sadler and D. M. Salter, "Integrin-dependent signal cascades in chondrocyte mechanotransduction," Ann. Biomed. Eng., vol. 32, pp , [91] G. Lapadula, F. Iannone, C. Zuccaro, V. Grattagliano, M. Covelli, V. Patella, G. L. Bianco and V. Pipitone, "Integrin expression on chondrocytes: correlations with the degree of cartilage damage in human osteoarthritis," Clin. Exp. Rheumatol., vol. 15, pp , [92] J. C. Kudirka, N. Panupinthu, M. A. Tesseyman, S. J. Dixon and S. M. Bernier, "P2Y nucleotide receptor signaling through MAPK/ERK is regulated by extracellular matrix: Involvement of β3 integrins," J. Cell. Physiol., vol. 213, pp , [93] F. Sachs and C. Morris, "Mechanosensitive ion channels in nonspecialized cells," Reviews of Physiology Biochemistry and Pharmacology, Volume 132, pp. 1-77, [94] V. S. Markin and T. Y. Tsong, "Reversible mechanosensitive ion pumping as a part of mechanoelectrical transduction," Biophys. J., vol. 59, pp , [95] B. Pingguan-Murphy, M. El-Azzeh, D. L. Bader and M. M. Knight, "Cyclic compression of chondrocytes modulates a purinergic calcium signalling pathway in a strain rate- and frequency-dependent manner," J. Cell. Physiol., vol. 209, pp ,

115 [96] M. Koolpe, D. Pearson and H. P. Benton, "Expression of both P 1 and P 2 purine receptor genes by human articular chondrocytes and profile of ligand-mediated prostaglandin E 2 release," Arthritis & Rheumatism, vol. 42, pp , [97] R. D. Graff, M. Picher and G. M. Lee, "Extracellular nucleotides, cartilage stress, and calcium crystal formation," Curr. Opin. Rheumatol., vol. 15, pp , [98] K. Johnson, S. Vaingankar, Y. Chen, A. Moffa, M. B. Goldring, K. Sano, P. Jin- Hua, A. Sali, J. Goding and R. Terkeltaub, "Differential mechanisms of inorganic pyrophosphate production by plasma cell membrane glycoprotein-1 and B10 in chondrocytes," Arthritis & Rheumatism, vol. 42, pp , [99] J. A. Rees and S. Y. Ali, "Ultrastructural localisation of alkaline phosphatase activity in osteoarthritic human articular cartilage," Ann. Rheum. Dis., vol. 47, pp , [100] B. T. Summey Jr, R. D. Graff, T. S. Lai, C. S. Greenberg and G. M. Lee, "Tissue transglutaminase localization and activity regulation in the extracellular matrix of articular cartilage," J. Orthop. Res., vol. 20, pp , [101] A. Cardenal, I. Masuda, A. L. Haas and D. J. McCarty, "Specificity of a porcine 127-kd nucleotide pyrophosphohydrolase for articular tissues," Arthritis & Rheumatism, vol. 39, pp , [102] C. J. Brown, A. M. Caswell, S. Rahman, R. G. G. Russell and D. J. Buttle, "Proteoglycan breakdown from bovine nasal cartilage is increased, and from articular cartilage is decreased, by extracellular ATP," BBA-Molecular Basis of Disease, vol. 1362, pp , [103] L. J. Croucher, A. Crawford, P. V. Hatton, R. G. G. Russell and D. J. Buttle, "Extracellular ATP and UTP stimulate cartilage proteoglycan and collagen accumulation in bovine articular chondrocyte pellet cultures," BBA-Molecular Basis of Disease, vol. 1502, pp , [104] M. B. Goldring, "Update on the biology of the chondrocyte and new approaches to treating cartilage diseases," Best Practice & Research Clinical Rheumatology, vol. 20, pp , [105] E. C. Arner, "Aggrecanase-mediated cartilage degradation," Current Opinion in Pharmacology, vol. 2, pp ,

116 [106] G. C. Jones and G. P. Riley, "ADAMTS proteinases: a multi-domain, multifunctional family with roles in extracellular matrix turnover and arthritis," Arthritis Res Ther, vol. 7, pp , [107] K. Kuno, N. Kanada, E. Nakashima, F. Fujiki, F. Ichimura and K. Matsushima, "Molecular cloning of a gene encoding a new type of metalloproteinase-disintegrin family protein with thrombospondin motifs as an inflammation associated gene," J. Biol. Chem., vol. 272, pp , [108] M. Kashiwagi, M. Tortorella, H. Nagase and K. Brew, "TIMP-3 is a potent inhibitor of aggrecanase 1 (ADAM-TS4) and aggrecanase 2 (ADAM-TS5)," J. Biol. Chem., vol. 276, pp , [109] W. Huang, W. Q. Li, F. Dehnade and M. Zafarullah, "Tissue inhibitor of metalloproteinases-4 (TIMP-4) gene expression is increased in human osteoarthritic femoral head cartilage," J. Cell. Biochem., vol. 85, pp , [110] F. Berenbaum, L. Humbert, G. Bereziat and S. Thirion, "Concomitant Recruitment of ERK1/2 and p38 MAPK Signalling Pathway Is Required for Activation of Cytoplasmic Phospholipase A2via ATP in Articular Chondrocytes," J. Biol. Chem., vol. 278, pp , [111] T. T. Chowdhury and M. M. Knight, "Purinergic pathway suppresses the release of.no and stimulates proteoglycan synthesis in chondrocyte/agarose constructs subjected to dynamic compression," J. Cell. Physiol., vol. 209, pp , [112] S. Hashimoto, R. L. Ochs, F. Rosen, J. Quach, G. McCabe, J. Solan, J. E. Seegmiller, R. Terkeltaub and M. Lotz, "Chondrocyte-derived apoptotic bodies and calcification of articular cartilage," Proc. Natl. Acad. Sci. U. S. A., vol. 95, pp , [113] M. Gosset, F. Berenbaum, A. Levy, A. Pigenet, S. Thirion, S. Cavadias and C. Jacques, "Mechanical stress and prostaglandin E2 synthesis in cartilage," Biorheology, vol. 45, pp , [114] B. Fermor, J. B. Weinberg, D. S. Pisetsky, M. A. Misukonis, A. J. Banes and F. Guilak, "The effects of static and intermittent compression on nitric oxide production in articular cartilage explants," Journal of Orthopaedic Research, vol. 19, pp ,

117 [115] D. A. Lee, S. P. Frean, P. Lees and D. L. Bader, "Dynamic Mechanical Compression Influences Nitric Oxide Production by Articular Chondrocytes Seeded in Agarose," Biochem. Biophys. Res. Commun., vol. 251, pp , [116] A. M. Caswell, W. S. Leong and R. G. G. Russell, "Evidence for the presence of P2-purinoceptors at the surface of human articular chondrocytes in monolayer culture," Biochimica Et Biophysica Acta (BBA)-General Subjects, vol. 1074, pp , [117] K. Varani, M. De Mattei, F. Vincenzi, A. Tosi, S. Gessi, S. Merighi, A. Pellati, F. Masieri, A. Ongaro and P. A. Borea, "Pharmacological characterization of P2X1 and P2X3 purinergic receptors in bovine chondrocytes," Osteoarthritis and Cartilage, vol. 16, pp , [118] T. T. Chowdhury and M. M. Knight, "Purinergic pathway suppresses the release of. NO and stimulates proteoglycan synthesis in chondrocyte/agarose constructs subjected to dynamic compression," J. Cell. Physiol., vol. 209, pp , [119] R. A. Terkeltaub, "Inorganic pyrophosphate generation and disposition in pathophysiology," American Journal of Physiology- Cell Physiology, vol. 281, pp. C1, [120] L. M. Ryan and A. K. Rosenthal, "Metabolism of extracellular pyrophosphate," Curr. Opin. Rheumatol., vol. 15, pp , [121] J. C. Costello, A. K. Rosenthal, I. V. Kurup, I. Masuda, M. Medhora and L. M. Ryan, "Parallel regulation of extracellular ATP and inorganic pyrophosphate: roles of growth factors, transduction modulators, and ANK," Connect. Tissue Res., pp. 1-8, [122] G. M. McCarthy and H. S. Cheung, "Point: Hydroxyapatite crystal deposition is intimately involved in the pathogenesis and progression of human osteoarthritis," Curr. Rheumatol. Rep., vol. 11, pp , [123] K. P. H. Pritzker, "Counterpoint: Hydroxyapatite crystal deposition is not intimately involved in the pathogenesis and progression of human osteoarthritis," Curr. Rheumatol. Rep., vol. 11, pp , [124] L. M. Ryan, I. V. Kurup, B. A. Derfus and V. M. Kushnaryov, "ATP-induced chondrocalcinosis," Arthritis & Rheumatism, vol. 35, pp , [125] R. A. Terkeltaub, "Pathogenesis and treatment of crystal-induced inflammation," in Arthritis and Allied Conditions: A Textbook of Rheumatology, 15th ed., W. J. Koopman 105

118 and L. W. Moreland, Eds. Philadelphia: Lippincott Williams and Wilkins, 2005, pp [126] E. S. Molloy and G. M. McCarthy, "Calcium crystal deposition diseases: update on pathogenesis and manifestations," Rheumatic Disease Clinics of North America, vol. 32, pp , [127] L. M. Ryan, J. W. Rachow and D. J. McCarty, "Synovial Fluid ATP: A Potential Substrate for the Production of Inorganic Pyrophosphate," J. Rheumatol., vol. 18, pp , [128] B. A. Derfus, J. W. Rachow, N. S. Mandel, A. L. Boskey, M. Buday, V. M. Kushnaryov and L. M. Ryan, "Articular cartilage vesicles generate calcium pyrophosphate dihydrate-like crystals in vitro," Arthritis & Rheumatism, vol. 35, pp , [129] R. Kandel, M. Hurtig and M. Grynpas, "Characterization of the Mineral in Calcified Articular Cartilagenous Tissue Formed in Vitro," Tissue Eng., vol. 5, pp , [130] A. K. Rosenthal, C. M. Gohr, M. Uzuki and I. Masuda, "Osteopontin promotes pathologic mineralization in articular cartilage," Matrix Biology, vol. 26, pp , 3, [131] S. Y. Ali, "Apatite-type crystal deposition in arthritic cartilage," Scan. Electron Microsc., vol. (Pt 4), pp , [132] B. Jubeck, C. Gohr, M. Fahey, E. Muth, M. Matthews, E. Mattson, C. Hirschmugl and A. K. Rosenthal, "Promotion of articular cartilage matrix vesicle mineralization by type I collagen," Arthritis & Rheumatism, vol. 58, pp , [133] K. E. Kuettner, B. U. Pauli, G. Gall, V. A. Memoli and R. K. Schenk, "Synthesis of cartilage matrix by mammalian chondrocytes in vitro. I. Isolation, culture characteristics, and morphology." J. Cell Biol., vol. 93, pp , [134] B. Mishell and S. Shiigi, "Selected Methods in Cellular Immunology," 1980, pp 486. [135] J. Boyle, B. Luan, T. F. Cruz and R. A. Kandel, "Characterization of proteoglycan accumulation during formation of cartilagenous tissue in vitro*," Osteoarthritis and Cartilage, vol. 3, pp ,

119 [136] H. Boström and B. Månsson, "On the enzymatic exchange of the sulfate group of chondroitinsulfuric acid in slices of cartilage," J. Biol. Chem., vol. 196, pp , [137] B. Peterkofsky and R. Diegelmann, "Use of a mixture of proteinase-free collagenases for the specific assay of radioactive collagen in the presence of other proteins," Biochemistry (N. Y. ), vol. 10, pp , [138] G. Burnstock and J. J. Warland, "P2-purinoceptors of two subtypes in the rabbit mesenteric artery: reactive blue 2 selectively inhibits responses mediated via the P2y-but not the P2x-purinoceptor." Br. J. Pharmacol., vol. 90, pp , [139] C. Kennedy, "P1- and P2-purinoceptor subtypes--an update," Arch. Int. Pharmacodyn. Ther., vol. 303, pp , [140] Y. Kim, R. L. Y. Sah, J. H. Doong and A. J. Grodzinsky, "Fluorometric assay of DNA in cartilage explants using Hoechst 33258," Anal. Biochem., vol. 174, pp , 10, [141] R. W. Farndale, D. J. Buttle and A. J. Barrett, "Improved quantitation and discrimination of sulphated glycosaminoglycans by use of dimethylmethylene blue," Biochimica Et Biophysica Acta (BBA) - General Subjects, vol. 883, pp , [142] R. L. Goldberg and L. M. Kolibas, "An Improved Method for Determining Proteoglycans Synthesized by Chondrocytes in Culture," Connect. Tissue Res., vol. 24, pp , [143] D. Heinegard, M. Bayliss and P. Lorenzo, "Biochemistry and metabolism of normal and osteoarthritic cartilage," in Osteoarthritis, K. D. Brandt, M. Doherty and L. S. Lohmander, Eds. New York: Oxford University Press, 1998, pp [144] J. F. Woessner, "The determination of hydroxyproline in tissue and protein samples containing small proportions of this imino acid," Arch. Biochem. Biophys., vol. 93, pp , 5, [145] D. H. Hoch, A. J. Grodzinsky, T. J. Koob, M. L. Albert and D. R. Eyre, "Early changes in material properties of rabbit articular cartilage after meniscectomy," Journal of Orthopaedic Research, vol. 1, pp. 4-12, [146] C. Y. Fan, "Effects of Intermittent Static and Dynamic Tension on Articular Chondrocytes in High Density Culture,"

120 [147] W. C. Hayes, L. M. Keer, G. Herrmann and L. F. Mockros, "A mathematical analysis for indentation tests of articular cartilage," J. Biomech., vol. 5, pp , [148] G. W. Gill, J. K. Frost and K. A. Miller, "A new formula for a half-oxidized hematoxylin solution that neither overstains nor requires differentiation," Acta Cytol., vol. 18, pp , [149] G. Geyer and W. Linss, "Toluidine blue staining of cartilage proteoglycan subunits," Acta Histochem., vol. 61, pp , [150] J. Von Kossa, "Ueber die im Organismus künstlich erzeugbaren Verkalkungen," Beitr. Pathol. Anat., vol. 29, pp , [151] A. Chevrier, E. Rossomacha, M. D. Buschmann and C. D. Hoemann, "Optimization of histoprocessing methods to detect glycosaminoglycan, collagen type II, and collagen type I in decalcified rabbit osteochondral sections," Journal of Histotechnology, vol. 28, pp , [152] E. Weir, T. G. Pretlow, A. Pitts and E. E. Williams, "A more sensitive and specific histochemical peroxidase stain for the localization of cellular antigen by the enzymeantibody conjugate method," Journal of Histochemistry and Cytochemistry, vol. 22, pp , [153] M. A. Wimmer, S. Grad, T. Kaup, M. Hänni, E. Schneider, S. Gogolewski and M. Alini, "Tribology approach to the engineering and study of articular cartilage," Tissue Eng., vol. 10, pp , [154] J. A. Kaupp and S. D. Waldman, "Mechanical vibrations increase the proliferation of articular chondrocytes in high-density culture," Proc. Inst. Mech. Eng. [H], vol. 222, pp , [155] S. Rozen and H. Skaletsky, "Primer3 on the WWW for general users and for biologist programmers," in Bioinformatics Methods and Protocols: Methods in Molecular Biology, S. Krawertz and S. Misener, Eds. Totowa: Humana Press, 2000, pp [156] B. Engels, "Amplify 3," vol. 3.1, [157] P. Chomczynski, "A reagent for the single-step simultaneous isolation of RNA, DNA and proteins from cell and tissue samples," BioTechniques, vol. 15, pp ,

121 [158] R. H. Nielsen, R. Stoop, D. J. Leeming, M. Stolina, P. Qvist, C. Christiansen and M. A. Karsdal, "Evaluation of cartilage damage by measuring collagen degradation products in joint extracts in a traumatic model of osteoarthritis," Biomarkers, vol. 13, pp , [159] V. Knäuper, H. Will, C. López-Otin, B. Smith, S. J. Atkinson, H. Stanton, R. M. Hembry and G. Murphy, "Cellular mechanisms for human procollagenase-3 (MMP-13) activation," J. Biol. Chem., vol. 271, pp , [160] J. Justesen and N. O. Kieldgaard, "Spectrophotometric pyrophosphate assay of 2, 5 -ollgoadenylate synthetase," Anal. Biochem., vol. 207, pp , [161] F. A. J. Van de Loo, O. J. Arntz and W. B. Van den Berg, "Effect of Interleukin 1 and Leukaemia Inhibitory Factor on Chondrocyte Metabolism in Articular Cartilage from Normal and Interleukin-6-Deficient Mice: Role of Nitric Oxide and Il-6 in the Suppression of Proteoglycan Synthesis," Cytokine, vol. 9, pp , [162] W. S. Leong, R. G. G. Russell and A. M. Caswell, "Stimulation of cartilage resorption by extracellular ATP acting at P2-purinoceptors," Biochimica Et Biophysica Acta (BBA)-General Subjects, vol. 1201, pp , [163] J. J. Reynolds, "The effect of ascorbic acid on the growth of chick bone rudiments in chemically defined medium," Exp. Cell Res., vol. 42, pp , [164] I. Martin, B. Obradovic, S. Treppo, A. J. Grodzinsky, R. Langer, L. E. Freed and G. Vunjak-Novakovic, "Modulation of the mechanical properties of tissue engineered cartilage," Biorheology, vol. 37, pp , [165] H. J. Mankin, V. C. Mow, J. A. Buckwalter, J. P. Iannotti and A. Ratcliffe, "Form and function of articular cartilage," in Orthopaedic Basic Science, S. R. Simon, Ed. Columbus: American Academy of Orthopaedic Surgeons, 1994, pp [166] H. Takaishi, T. Kimura, S. Dalal, Y. Okada and J. D'Armiento, "Joint diseases and matrix metalloproteinases: a role for MMP-13," Curr. Pharm. Biotechnol., vol. 9, pp , [167] A. J. Fosang, P. J. Neame, K. Last, T. E. Hardingham, G. Murphy and J. A. Hamilton, "The interglobular domain of cartilage aggrecan is cleaved by PUMP, gelatinases, and cathepsin B." J. Biol. Chem., vol. 267, pp ,

122 [168] K. Johnson and R. Terkeltaub, "Upregulated ank expression in osteoarthritis can promote both chondrocyte MMP-13 expression and calcification via chondrocyte extracellular PPi excess," Osteoarthritis and Cartilage, vol. 12, pp , [169] A. K. Rosenthal, D. Hempel, I. V. Kurup, I. Masuda and L. M. Ryan, "Purine Receptors Modulate Chondrocyte Extracellular Inorganic Pyrophosphate Production," Osteoarthritis and Cartilage, [170] M. Doherty, A. Chuck, D. Hosking and E. Hamilton, "Inorganic pyrophosphate in metabolic diseases predisposing to calcium pyrophosphate dihydrate crystal deposition," Arthritis Rheum., vol. 34, pp , [171] L. F. Bonewald, S. E. Harris, J. Rosser, M. R. Dallas, S. L. Dallas, N. P. Camacho, B. Boyan and A. Boskey, "von Kossa staining alone is not sufficient to confirm that mineralization in vitro represents bone formation," Calcif. Tissue Int., vol. 72, pp , [172] A. K. Rosenthal, E. Mattson, C. M. Gohr and C. J. Hirschmugl, "Characterization of articular calcium-containing crystals by synchrotron FTIR," Osteoarthritis and Cartilage, vol. 16, pp , 11, [173] K. P. H. Pritzker, "Calcium pyrophosphate crystal formation and dissolution," in Calcium Phosphates in Biological and Industrial Systems, Z. Amjad, Ed. Massachusetts: Kluwer Academic Publishers, 1998, pp [174] R. Kandel, M. Hurtig and M. Grynpas, "Characterization of the Mineral in Calcified Articular Cartilagenous Tissue Formed in Vitro," Tissue Eng., vol. 5, pp , [175] P. T. Cheng, K. P. H. Pritzker, M. E. Adams, S. C. Nyburg and S. A. Omar, "Calcium pyrophosphate crystal formation in aqueous solutions," J. Rheumatol., vol. 7, pp , [176] R. Liu-Bryan, K. Pritzker, G. S. Firestein and R. Terkeltaub, "TLR2 signaling in chondrocytes drives calcium pyrophosphate dihydrate and monosodium urate crystalinduced nitric oxide generation," The Journal of Immunology, vol. 174, pp , [177] H. S. Cheung, "Calcium crystal effects on the cells of the joint: implications for pathogenesis of disease," Curr. Opin. Rheumatol., vol. 12, pp , [178] H. S. Cheung, T. R. Devine and W. Hubbard, "Calcium phosphate particle 110

123 induction of metalloproteinase and mitogenesis: effect of particle sizes*," Osteoarthritis and Cartilage, vol. 5, pp , [179] P. A. Gibilisco, H. R. Schumacher Jr, J. L. Hollander and K. A. Soper, "Synovial fluid crystals in osteoarthritis," Arthritis & Rheumatism, vol. 28, pp , [180] R. Terkeltaub, "Diseases associated with articular deposition of calcium pyrophosphate dihydrate and basic calcium phosphate crystals," Kelley's Textbook of Rheumatology.Philadelphia: Elsevier Saunders, pp , [181] A. K. Rosenthal and L. A. Henry, "Thyroid hormones induce features of the hypetrophic phenotype and stimulate correlates of CPPD crystal formation in articular chondrocytes," Journal of Rheumatology, vol. 26, pp , [182] T. Kirsch, H. D. Nah, I. M. Shapiro and M. Pacifici, "Regulated production of mineralization-competent matrix vesicles in hypertrophic chondrocytes," J. Cell Biol., vol. 137, pp , [183] E. B. Hunziker, T. M. Quinn and H. J. Häuselmann, "Quantitative structural organization of normal adult human articular cartilage," Osteoarthritis and Cartilage, vol. 10, pp , [184] S. K. Kim, D. H. Lee, J. I. Hong and J. Yoon, "Chemosensors for pyrophosphate," Acc. Chem. Res., vol. 42, pp , [185] H. N. Lee, Z. Xu, S. K. Kim, K. M. K. Swamy, Y. Kim, S. J. Kim and J. Yoon, "Pyrophosphate-selective fluorescent chemosensor at physiological ph: Formation of a unique excimer upon addition of pyrophosphate," J. Am. Chem. Soc., vol. 129, pp , [186] H. S. Cheung, I. V. Kurup, J. D. Sallis and L. M. Ryan, "Inhibition of calcium pyrophosphate dihydrate crystal formation in articular cartilage vesicles and cartilage by phosphocitrate," J. Biol. Chem., vol. 271, pp , [187] H. Cheung S., J. Sallis D. and J. Struve A., "Specific inhibition of basic calcium phosphate and calcium pyrophosphate crystal-induction of metalloproteinase synthesis by phosphocitrate," BBA-Molecular Basis of Disease, vol. 1315, pp ,

124 Appendix A: Transmission Electron Microscopy Transmission electron microscopy (TEM) was performed by Mr. Doug Holmyard at Mount Sinai Hospital, Toronto, ON, on separate cartilaginous cultures that had been supplemented with 0, 62.5 and 250 µm and grown for 4 weeks. All samples appeared to be healthy as noted by Mr. Holmyard (which was also indicated by the clearly defined chondrocytes). 112

125 Transmission electron microscopy (TEM) image of cartilaginous tissues grown for 4 weeks in (A) the absence of ATP, (B) the presence of 62.5 µm ATP and (C) the presence of 250 µm ATP. Magnification is 5000x. Representative images are shown. The top left corner of each image represents the porous Millipore filter that the cartilaginous tissues were grown on. 113