부산대학교 도서관

Valvular heart disease is a major cause of mortality worldwide. Mitral valve regurgitation represents the second major valvular disorder in the western world. Current strategies for mitral valve reconstruction are imperfect. The aim of this study was to investigate the tissue engineering of mitral valve leaflets for mitral valve leaflets reconstruction. The approach taken was to utilise decellularised porcine pericardium seeded with porcine mesenchymal stem cells (pMSCs) and to mechanically condition the cell seeded constructs using biaxial strain in a bespoke bioreactor. The biomechanical and biological properties of porcine mitral valve leaflets and native and decellularised porcine pericardium were studied for comparative purposes. The porcine pericardium was decellularised using a propriety method based upon low concentration sodium dodecyl sulphate (SDS) and proteinase inhibitors. Histological characterisation showed the four and three layered structure of leaflets and fresh/decellularised pericardium, respectively. Histological analysis of decellularised pericardium did not reveal any remaining cells. Moreover, the histoarchitecture of collagen and elastic fibres appeared to be well preserved. Biochemical analysis showed that the mitral valve leaflets and the fresh pericardium had hydroxyproline and glycosaminoglycan (GAG) contents similar to those reported in the literature. Decellularised pericardium had higher hydroxyproline content than that of fresh pericardium but lower GAG content. The levels of deoxyribonucleic acid (DNA) in fresh and decellularised pericardium were determined and this showed that there was a 99% reduction in the DNA content following decellularisation. In vitro biocompatibility studies showed that the decellularised pericardium was not cytotoxic to porcine skin fibroblasts and pMSCs. Biomechanical properties were determined using low strain rate uniaxial loading to failure. Both fresh and decellularised pericardium demonstrated rather isotropic behaviours, possessing similar mechanical properties along the two orthogonal directions studied. Anterior leaflet specimens cut circumferentially were stiffer than those cut radially and the posterior leaflets. This anisotropic mechanical behaviour of the anterior leaflet was related to the main orientation of the collagen fibres along the circumferential direction. Conversely, the posterior leaflets were more isotropic. The optimum seeding density for culturing decellularised pericardial samples was 1×105 cells·cm-2 for pMSCs and the ideal time for culturing prior to loading the reseeded scaffold in the bioreactor stations was 3 days. With regards to fibroblasts and porcine smooth muscle cells (pSMCs), the optimum seeding density was 2×105 cells·cm-2 at 1 week culture for fibroblasts, whereas pSMCs had the tendency of forming agglomerates on the surface of tissue rather than penetrating throughout the thickness of the scaffold, thus, they were not considered ideal for remodelling the histoarchitecture of tissue during dynamic experiments. The actuator of the bioreactor was calibrated and the general set up of the system was completed. Suitable conditions to apply during static culture in the bioreactor, in terms of oxygen and pH control, in order to maintain the ECM integrity and cell viability in the bioreactor, were initially determined using fresh pericardium, however this proved problematic due to the variability in the tissue that was available. One day static experiment performed with the bioreactor by using decellularised pericardium reseeded with pMSCs showed that there was no difference in terms of viability when keeping the tissue statically in the tissue holders and in the bioreactor stations. Thus, tissue holders could be used as positive static controls during dynamic experiments. One day dynamic experiments (10% strain) performed with the bioreactor by using decellularised pericardium reseeded with pMSCs showed that cells were viable both when kept dynamically in the bioreactor and statically in the tissue holders. Cells started aligning along specific directions when kept dynamically in the bioreactor. Overall porcine mitral valve leaflets and porcine fresh/decellularised pericardium shared similar histoarchitectures, but had different biochemical composition and biomechanics. Decellularised pericardium was shown to be an optimum material for cell repopulation, delivering the necessary biological and biomechanical cues to seeded cells. The bioreactor was optimised for both static and dynamic culture and is now ready for further investigation at longer time points. This research provided the basics into the optimal strategy for mechanostimulation of cell-seeded pericardial scaffolds in vitro in order to generate heart-valve like tissue.