Modeling collagen remodeling in tissue engineered cardiovascular tissues

Commonly, heart valve replacements consist of non-living materials lacking the ability to grow, repair and remodel. Tissue engineering (TE) offers a promising alternative to these replacement strategies since it can overcome its disadvantages. The technique aims to create an autologous living tissue with the potential to grow and adapt in response to changing functional demands. The concept is based on seeding autologous cells on a biodegradable material and delivering the mechanical and biomechanical stimuli in a bioreactor system to stimulate tissue development. In TE of cardiovascular tissues, it is crucial to obtain a tissue equipped with appropriate mechanical properties to withstand the hemodynamic loads. The mechanical properties are defined by a well organized network of collagen fibers. Therefore, mechanical conditioning should be optimized to create TE cardiovascular tissues with a properly organized collagen fiber network. Collagen orientation in engineered tissues is related to mechanical stimuli yet, the mechanism itself is not fully understood. It is known that in cardiovascular tissues, collagen is predominately produced by fibroblasts and fibroblast like cells (myofibroblasts). Collagen fibrils align with cells in many native tissues. Several studies suggest a guiding role of the cell cytoskeleton in directing collagen synthesis. Furthermore, the co-alignment of collagen fibers and cells may also be the result of the contractile forces generated by the cells a-actin fibers. Cells also react to the resistance they sense by reorganizing their cytoskeleton. For example, they orient differently when submitted to static or dynamic conditioning. In the conditioning protocols, the cellular response to mechanical stimuli, the development of contractile forces and the collagen synthesis direction play a crucial role. It is essential to investigate these processes to improve protocols and optimize TE mechanical properties. Yet, these processes are highly coupled and may only be unraveled with the assistance of mathematical models. In this thesis, focus was given to the study of the collagen architecture remodeling in cardiovascular tissues. First, focus was given to native tissues. A structure-based model (Driessen et al., 2008) was applied to assess and evaluate the the mechanical properties of pairs of aortic and pulmonary valves (Chapter 2). Finite element analyses were performed to simulate the mechanical response of both leaflets to a transvalvular aortic pressure load. Furthermore, remodeling laws were applied to assess the change in properties of the pulmonary valve leaflets in this position. The result from biaxial tensile tests were used to determine the model parameters. When the results from both valves were compared it was observed that the PV presented to be more extensible and less anisotropic than the AV. When under the aortic valve environment, the stresses in the PV leaflet were also higher and the coaptation area was smaller than in the AV leaflets. Furthermore, our study showed that the PV leaflets appeared remodel by increasing its thickness and rotating its fibers towards the circumferential direction. Yet, this remodeling did not result in properties that are completely identical to the AV leaflet. Although Driessen et al. model succeeded in predicting the typical collagen fiber architecture found in the native leaflet, it shows to be unsuitable to describe the collagen remodeling in tissue engineered tissue developed under static loading conditions since under these conditions no external force is applied. During static culture, experiments show that these tissues gradually compact due to contractile stresses developed by cells. These results suggest that collagen alignment under static loading conditions is a result of tissue compaction. Therefore, compaction was incorporated in the Driessen et al. employing the volumetric growth theory (Chapter 3). Using this extended model, the distribution of the collagen architecture of TE vessels developed under static loading conditions could be successfully described. However, the underlying mechanism for tissue compaction was not incorporated in the model, but assumed a priori. Therefore, the mechanisms by which the cells remodel the collagen architecture were discussed in Chapter 4. A new hypothesis was formulated and collagen orientation was linked to contractile stresses that develop in the a-actin fibers of the cell. For this purpose, two models were therefore integrated: the first one describing the mechanical behavior of collagen fibers and the second one the synthesis and degradation of a-actin stress fibers in the cell and the active, contractile forces that also develop in the cells. It was assumed that the collagen direction and content were equal to the a-actin stress fiber direction and activation level. A feasibility study was performed and the influence of the different parameters were evaluated (Chapter 5). The framework was then applied to study the collagen remodeling of tissue engineered constructs developed under static loading conditions (Chapter 6). The model successfully described the experimental results of TE strips developed under static loading conditions. The model also successfully predicted the non intuitive collagen orientation in TE small diameter vessels.