Osteoarthritis is a highly prevalent, debilitating disease characterized by the wear and degradation of articular cartilage. While many surgical interventions exist, few are consistently effective and those that are effective are not necessarily suitable for all patients. The objective of this dissertation is to improve patient care through the development of a new surgical technique and through basic science studies which seek to better understand articular cartilage wear initiation. Four studies, which address this objective are summarized below. Osteochondral allograft transplantation provides a safe and effective treatment option for large cartilage defects, but its use is limited partly due to the difficulty of matching articular surface curvature between donor and recipient. We hypothesize that bendable osteochondral allografts may provide better curvature matching for patella transplants in the patellofemoral joint. The finite element study presented in Chapter 2 investigates patellofemoral joint congruence for unbent and bendable osteochondral allografts, at various flexion angles. Finite element models were created for 12 femur-patella osteochondral allograft pairings. Two grooves were cut into the bony substrate of each allograft, allowing the articular layer to bend. Patellofemoral joints with either unbent (OCA) or permanently bent (BOCA) allografts were articulated from 40 to 70 degrees flexion and contact area was calculated. OCAs and BOCAs were then shifted 6 mm distally toward the tibia (S-OCA, S-BOCA) to investigate the influence of proximal-distal alignment on congruence. On average, no significant difference in contact area was found between native patellofemoral joints and either OCAs or BOCAs (p > 0.25), indicating that both types of allografts restored native congruence. This result provides biomechanical support in favor of an emerging surgical procedure. S-BOCAs resulted in a significant increase in contact area relative to the remaining groups (p < 0.02). The fact that bendable osteochondral allografts produced equally good results implies that these bendable allografts may prove useful in future surgical procedures, with the possibility of transplanting them with a small distal shift. Surgeons who are reluctant to use osteochondral allografts for resurfacing patellae based on curvature matching capabilities may be more amenable to adopting bendable osteochondral allografts. The recent development of bendable osteochondral allografts provides the potential for improved osteoarthritis treatment for joints whose current treatment is unsatisfactory. One such joint is the carpometacarpal joint in the thumb. While the current standard of care for carpometacarpal osteoarthritis, ligament reconstruction and tendon interposition, can reduce pain in the joint, it does not restore full joint function and mobility. A proposed alternative includes using an osteochondral allograft harvested from the femoral trochlea in a donor knee, machining grooves in the bone to allow the allograft to bend, and replacing the trapezium with this bent osteochondral allograft [1,2]. Chapter 3 of this dissertation discusses adjustments to the original design of the bendable allograft and the design of a custom surgical tool to perform the proposed surgery. Specification changes of the allograft included an overall size reduction in order to better fit within the carpometacarpal joint, minimum bone thickness requirements to avoid bone cracking during the surgical procedure, and a reduction from three grooves to two grooves, which provided sufficient bending yet avoided fracture of the allograft. The surgical tool was designed to be a custom forceps device, whose primary features included (1) jaws with an angled face to match the angle of allograft bending and (2) insertion holes for the Kirschner wire and compression screws used to anchor the allograft in the bent position. These customizations allow the tool to be used to bend the allograft, fix it in the bent configuration, and place the allograft in its proper position in the hand during anchoring of the bent allograft to the native trapezium. The final two studies presented in this dissertation focus on furthering our current understanding of wear and structure-function relationships of articular cartilage. We hypothesize that cartilage wears due to fatigue failure in reciprocating compression instead of reciprocating friction. Chapter 4 compares reciprocating sliding of immature bovine articular cartilage against glass in two testing configurations: (1) a stationary contact area configuration (SCA), which results in static compression, interstitial fluid depressurization and increasing friction coefficient during reciprocating sliding, and (2) a migrating contact area configuration (MCA), which maintains fluid pressurization and low friction while producing reciprocating compressive loading during reciprocating sliding. Contact stress, sliding duration, and sliding distance were controlled to be similar between test groups. SCA tests exhibited an average friction coefficient of μ=0.084±0.032, while MCA tests exhibited a lower average friction coefficient of μ=0.020±0.008 (p<10^(-4)). Despite the lower friction, MCA cartilage samples exhibited clear surface damage with a significantly greater average surface deviation from a fitted plane after wear testing (R_q=0.125±0.095 mm) than cartilage samples slid in a SCA configuration (R_q=0.044±0.017 mm, p=0.002), which showed minimal signs of wear. Polarized light microscopy confirmed that delamination damage occurred between the superficial and middle zones of the articular cartilage in MCA samples. The greatest wear was observed in the group with lowest friction coefficient, subjected to cyclical instead of static compression, implying that friction is not the primary driver of cartilage wear. Delamination between superficial and middle zones imply the main mode of wear is fatigue failure under cyclical compression, not fatigue or abrasion due to reciprocating frictional sliding. The final study of this dissertation, presented in Chapter 5, investigates the importance of collagen fibril distribution in articular cartilage computational models. Finite element models were created to approximate a bovine humeral head and replicate previous experimental loading conditions [3]. Five different finite element analyses were run, each using a different fibril distribution model. Three of the models used two, four, or eight discrete fibril bundles, while two models used continuous fibril distributions with either isotropic or depth-dependent ellipsoidal distributions. Two primary findings arose from this investigation. The first was the discovery that as the fibril distribution became more isotropic, the strain throughout the tissue decreased, even though the contact area between the articular surface and rigid platen remained relatively equal across distribution models. This suggests that computational models which approximate the collagen fibrils with an isotropic distribution may be underestimating the strain through the depth of the tissue. The second primary finding was that in the discrete distribution model with two fibril bundles, which followed the classically described Benninghoff structure [4], the greatest magnitude of shear strain during compressive loading was observed in the middle zone. However, the highest magnitude of shear strain observed in the isotropic fibril distribution model occurred in the deep zone near the subchondral surface. The observed results suggest that the type of fibril distribution used to model collagen in articular cartilage plays a role in depth-dependent strain magnitude and strain distribution.

Damage to articular cartilage, whether through degeneration (i.e. osteoarthritis) or acute injury, is particularly debilitating due to the tissue's limited regenerative capacity. These impairments are common: nearly 27 million Americans suffer from osteoarthritis and 36% of athletes suffer from traumatic cartilage defects. Allografts are the preferred treatment for large cartilage defects, but demand for these tissues outweighs their supply. To generate additional replacement tissues, tissue engineering strategies have been studied as a cell-based alternative therapy. Our laboratory has had great success repeatedly achieving native or near-native mechanical and biochemical properties in grafts engineered from chondrocyte-seeded agarose hydrogels. The most common iteration of this technique yields a disk of ~4 mm diameter and ~2.3 mm thickness. However, much work is still needed to increase the potential for clinical translation of this product. Our laboratory operates under the premise that in vivo success is predicated on replicating native graft properties in vitro. Compared to these engineered grafts, native grafts are larger in size. They consist of cartilage, which has properties varying in a depth-specific manner, anchored to a porous subchondral bone base. They are able to be stored between harvest and transplantation. This dissertation presents strategies to address a subset of the remaining challenges of reproducing these aspects in engineered grafts. First, graft macrostructure was addressed by incorporating a porous base to generate biomimetic osteochondral grafts. Previous studies have shown advantages to using porous metals as the bony base. Likewise, we confirmed that osteochondral constructs can be cultured to robust tissue properties using porous titanium bases. The titanium manufacturing method, selective laser melting, offers precise control, allowing for tailoring of base shape and pore geometry for optimal cartilage growth and osteointegration. In addition to viability studies, we investigated the influence of the porous base on the measured mechanical properties of the construct's gel region. Through measurements and correlation analysis, we linked a decrease in measured mechanical properties to pore area. We promote characterization of such parameters as an important consideration for the generation of functional grafts using any porous base. Next, we investigated a high intensity focused ultrasound (HIFU) denaturation of gel-incorporated albumin as a strategy for inducing local tissue property changes in constructs during in vitro growth. HIFU is a low cost, non-contact, non-invasive ultrasound modality that is used clinically and in the laboratory for such applications as ablation of uterine fibroids and soft tissue tumors. Denaturing such proteins has been shown to increase local stiffness. We displayed the ability incorporate albumin into tissue engineering relevant hydrogels, alter transport properties, and visualize treatment with its denaturation. HIFU treatment is aided by the porous metal base, allowing for augmented heating. Though heating cartilage is used in the clinic, it is associated with cell death. We investigated this effect, finding that the associated loss of viability remains localized to the treatment zone over time. This promotes the option of balancing desired changes in tissue properties against the concomitant cell viability loss. In order to match clinically utilized allografts, engineered constructs must be scaled up in size. This process is limited by diffusional transport of nutrients and other chemical factors, leading to preferential extracellular matrix deposition in the construct periphery. Many methods are being investigated for overcoming this limitation in fixed-size constructs. In this chapter, we investigated a novel strategy in which small constructs are cultured for future assembly. This modular assembly offers coverage of variable sized defects with more consistent growth with more uniform distribution of biochemical constituents than large constructs cultured on their own. Physiologic failure testing showed that integration of these tissues may be strengthened by increased subunit strength or assembled culture. It is expected that bioadhesive caulking and/or the incorporation of osteochondral bases would further increase integration of the assembled large graft. Finally, we sought to provide a preservation/storage protocol for engineered cartilage constructs. Such a technique is critical for clinical translation, providing the engineered graft with a “shelf-life.” We adopted and evaluated the Missouri Osteochondral Allograft Preservation System (MOPS), which had been shown to maintain cell viability in native grafts for at least 63 days at room temperature without serum or growth factors. Within the current clinical of 28 days, MOPS maintained chondrocyte viability and 75% of the pre-preservation Young's modulus without significant decline in biochemical content, however it did not extend the clinical window as it had with native grafts. Refrigeration with MOPS did not show any benefit at day 28, but proved better with longer preservation times. These results are the first evaluating engineered cartilage storage. Further optimization is necessary to extend storage tissue property maintenance in storage. Overall, this dissertation presents four strategies for increasing the translation potential of engineered articular cartilage grafts by better matching the clinically utilized native allograft system. Combining these techniques, one could ideally engineer small, interlocking ostechondral constructs with HIFU modified interface properties, which could be stored from maturity to implantation. Future optimization is required to better understand and utilize these methods to engineer fully functional, clinically relevant grafts.

L’arthrose et l’arthrite rhumatoïde sont deux pathologies articulaires caractérisées par la dégradation du cartilage articulaire, subséquente à la production de divers médiateurs inflammatoires. Le traitement de ces pathologies se limite généralement à soulager le patient des épisodes douloureux et inflammatoires et à améliorer sa qualité de vie. Dans le cas de l’arthrose, peu de traitements permettent d’enrayer significativement l’évolution de la dégradation du cartilage et donc de la maladie. Par contre, l’arthrite rhumatoïde peut être efficacement ralentie grâce à l’administration de certaines molécules. Néanmoins, ces traitements n’ont généralement montré qu’une efficacité à court-terme, requérant une administration fréquente. L’objectif de ce travail repose donc sur l’élaboration de nouvelles options thérapeutiques permettant de réduire la fréquence d’administration ainsi que les effets indésirables des traitements actuels. La délivrance de molécules en intra-articulaire associée à une libération prolongée offre l’avantage d’exposer les sites directement impliqués dans l’évolution de la maladie à une ou plusieurs molécules efficaces contre l’inflammation et la douleur, et aidant à la régénération du cartilage, durant plusieurs semaines, voire des mois.<p>Des microparticules de PLGA chargées en clonidine ou en bétaméthasone ont donc été optimisées afin d’obtenir des efficacités d’encapsulation appréciables (clonidine HCl :EE ≈ 20% ;dipropionate de bétaméthasone :EE ≈ 70%), une taille adaptée à l’administration intra-articulaire (12 – 38 µm) et une libération de la molécule s’échelonnant sur 5 à 8 semaines. La libération prolongée de la clonidine implique des mécanismes de diffusion de la molécule ainsi que de dégradation/érosion du polymère. Au vu de l’absence de réaction inflammatoire, les microparticules développées sont correctement tolérées par les chondrocytes, synoviocytes, PBMC et neutrophiles, principales cellules impliquées dans les mécanismes inflammatoires de l’arthrose et de l’arthrite rhumatoïde. L’évaluation de l’efficacité anti-inflammatoire des microparticules vides et chargées en clonidine ou en bétaméthasone via l’étude de l’expression et de la sécrétion de différents médiateurs de l’inflammation a permis d’aboutir à plusieurs conclusions :(i) les microparticules vides sont associées à un effet anti-inflammatoire, (ii) les microparticules chargées en clonidine n’ont pas montré d’activité anti-inflammatoire propre pouvant être attribuée à la clonidine, et (iii) les microparticules de bétaméthasone ont confirmé l’effet anti-inflammatoire de la bétaméthasone. Enfin, l’étude de la toxicité des principes actifs et microparticules vides ou chargées a montré une toxicité significative de la clonidine sur les synoviocytes. Néanmoins, l’encapsulation des principes actifs dans les microparticules de PLGA a permis d’éliminer cette toxicité, protégeant donc efficacement les cellules articulaires.<p>Les microparticules développées permettent alors d’envisager l’encapsulation d’autres molécules anti-inflammatoires ou une combinaison de molécules ayant des effets complémentaires (anti-inflammatoire et antidouleur). L’utilisation de la clonidine dans ces indications devra être réévaluée en étudiant de façon approfondie son efficacité dans la douleur. / Both osteoarthritis and rheumatoid arthritis are articular diseases characterized by the degeneration of the joint cartilage, resulting from the production of various inflammatory mediators. The current treatment of these diseases is restricted to alleviate the painful and inflammatory episodes of the patients and to improve its quality of life. In osteoarthritic patients, few treatments allow to significantly stop the evolution of the degradation of the cartilage and, consequently, the disease. In rheumatoid arthritis, the evolution can be slowed down following the administration of some drugs. Nevertheless, these treatments are often associated to a short-term efficacy. The objective of this work is to develop new therapeutic options that allow to reduce the frequency of administration and the side effects of the current treatments. The intraarticular delivery combined to controlled-release presents the advantage to expose the sites directly involved in the evolution of the disease to one or more molecules effective to relieve the pain, inflammation and to help the regeneration of the cartilage.<p>Clonidine or betamethasone-loaded PLGA microparticles were optimized to reach suitable encapsulation efficiencies (clonidine HCl: EE ≈ 20%; betamethasone dipropionate: EE ≈ 70%), an appropriate size for an intraarticular delivery (12 – 38 µm) and a controlled-release of the molecule over 5 to 8 weeks. The release of clonidine implies mechanisms of diffusion and degradation/erosion of the polymer. Given the absence of an inflammatory reaction, the developed microparticles were properly tolerated by the chondrocytes, synoviocytes, PMBC and neutrophils, which are the main cells involved in the inflammatory reaction of osteoarthritis and rheumatoid arthritis. The assessment of the anti-inflammatory efficacy of the drug-free and drug-loaded microparticles through the evaluation of the expression and the secretion of various inflammatory mediators allowed to draw several conclusions: (i) drug-free microparticles were associated to an anti-inflammatory effect, (ii) clonidine-loaded microparticles did not show any anti-inflammatory activity that could be attributed to clonidine, and (iii) betamethasone- loaded microparticles confirmed the anti-inflammatory effect of betamethasone. Finally, the evaluation of the toxicity of the drugs and microparticles showed a significant toxicity of clonidine against synoviocytes. Nevertheless, the encapsulation of the drugs in PLGA microparticles induced the suppression of this toxicity, protecting in this way the articular cells. <p>Entrapping other anti-inflammatory molecules or a combination of molecules with complementary effects (anti-inflammatory and anti-nociceptive drugs) in the PLGA microparticles developed has to be considered. Moreover, the use of clonidine in these indications has to be reassessed by a thorough study of its anti-nociceptive potential.<p><p> / Doctorat en Sciences biomédicales et pharmaceutiques / info:eu-repo/semantics/nonPublished