Strategies for engineering cartilage with improved content and organization

Abstract

Cartilage degeneration, from injury or osteoarthritis, is an important problem in current orthopaedic practice. Articular cartilage is unable to repair itself, resulting in a permanent defect and the formation of mechanically inferior fibrocartilage. Tissue engineering is a promising approach for the treatment of cartilage injuries, as it may eventually allow for production of engineered tissue indistinguishable from native cartilage. An important advantage of using tissue engineered material is that you implant a healthy, living tissue. Therefore it is more likely to integrate with the surrounding cartilage tissue. Currently, it is possible to engineer cartilage constructs with native proteoglycan content. However, current tissue-engineered cartilage is not suitable for implantation, because of its insufficient mechanical properties. Two major contributors to this poor mechanical quality are explored in this thesis. First, only 15-35% of the native collagen content is reached in tissue-engineered cartilage. Second, native extracellular matrix (ECM) organization on macro and micro scale is not reproduced. This thesis aims to improve the mechanical quality of tissue-engineered cartilage by exploring approaches to enhance both collagen content and ECM organization. Since our studies are to a great extent dependent on mechano-responsiveness of chondrocytes, we had to establish an appropriate culture model, which then could be used to transmit mechanical forces to the chondrocytes. A well-characterized and widely used model system involves the culturing of chondrocytes in agarose. We demonstrated that loading applied on 3% agarose constructs was sensed and transduced by the embedded chondrocytes. We found that RGD-dependent integrins were involved in mediating compression-induced alterations in ECM gene expression and protein production, and that this effect was dependent on the loading frequency applied. We observed in our and other studies that ECM is deposited mainly in a dense layer close to the chondrocytes. This inhomogeneity is believed to negatively affect mechanical properties of the engineered tissue. The second aim was to improve ECM distribution at the micro scale in chondrocyte-seeded agarose constructs. We demonstrated that distribution of ECM components was more uniform throughout the constructs when these were cultured with no or low agarose concentration, and when cultured in presence of growth factor TGF-s3. Zonal characteristics in matrix content and distribution have shown to be essential for the mechanical functioning of cartilage tissue, but are not reproduced in tissue-engineered cartilage. Therefore, our next aim was to create depth-dependent zonal variations in engineered cartilage constructs. We explored the hypothesis that depth-dependent mechanical cues, induced by a new, dedicated loading method called ‘sliding indentation’, would stimulate ECM synthesis depth-dependently. Numerical evaluation of this sliding indentation loading regime has shown that it can induce depth-varying strain fields in chondrocyte-seeded agarose constructs. It was confirmed that sliding indentation results in a depth-dependent response by the embedded chondrocytes, which was strongest in the regions that received highest strains. Another shortcoming of current tissue-engineered cartilage is its low collagen fraction. Since the mechanical function of collagen in articular cartilage is to resist tension, we postulated that in order to stimulate collagen formation we need to apply tension to the engineered cartilage constructs. Dynamic tension was applied by the aforementioned sliding indentation loading regime. In two separate studies, we demonstrated that application of dynamic tension induced by sliding indentation stimulated collagen type II production both in periosteum-derived cartilage and in chondrocyte-seeded agarose constructs. In conclusion, it has been shown that application of sliding indentation leads to increased collagen fractions and depth-dependent ECM distribution in tissue-engineered cartilage. The latter is major asset of using sliding indentation over alternative, regular loading protocols for tissue engineering, such as unconfined compression. Furthermore, it was demonstrated that ECM distribution at the micro scale appears more homogeneous in constructs cultured with no or low concentration agarose and in presence of TGF-s3. These findings may be used to define the optimal culture conditions for tissue engineering of cartilage with native collagen content, depth-dependent matrix organization, and sufficient mechanical properties, which are of pivotal importance for the engineering of mechanically stable, functional tissue engineered cartilage.