Tissue engineered heart valves have the potential to meet the growing need for valve replacement for pediatric patients, adolescents, and active adult patients with congenital defects, rheumatic fever, and valve disease, respectively. Currently engineered valve replacements are not durable post-implantation often due to cell-mediated tissue retraction and subsequent regurgitation. This limitation is largely due to the lack of knowledge required to inform the effective design of engineered tissues, and lack of platforms to efficiently investigate the integrative effects of multiple microenvironmental cues particularly including heart valve-relevant 3D mechanical stimulation on cell functions. Specifically, the optimization of mechanical loading protocols and methods for monitoring of tissue functional properties remain as primary challenges and priority for heart valve tissue engineering.

In this thesis, a platform was presented that enabled 3D dynamic mechanical stretch of arrayed cell-laden hydrogel constructs and simultaneous continuous stiffness measurement of the hydrogel constructs in situ with integrated strain sensors. Cell-seeded polyethylene glycol norbornene (PEG-NB) hydrogels were bound to deformable membranes via a thiol-ene reaction with off-stoichiometry thiol-ene based polydimethylsiloxane as the membrane material. Dynamic 3D mechanical stimulation of human mesenchymal stromal cells (MSCs)-seeded PEG-NB significantly promoted myofibroblast differentiation. As captured by the on-chip strain sensors, significant evolution in the stiffness of cell-hydrogel constructs during culture due to cell-mediated remodeling was also confirmed. In addition, a proof-of-concept study to screen and identify the optimal 3D mechanical stimulation combination for promoting collagen expression by MSCs was demonstrated, using factorial experimental design and regression parametric modeling. For the first time, a significant mechanical stimulation interaction effect was revealed that dominantly determined collagen production by cells.

The precisely controllable biomaterial and biochemical properties of the PEG-NB system also make this platform readily amenable to include other relevant microenvironmental cues for comprehensive combinatorial screening studies, and has the potential to identify strategies to predictably control the construction of functional engineered tissues in vitro. This platform with on-chip strain sensing also represents a promising approach to address the limitation of end-point analysis, enables exploration of bioprocess control strategies, and would provide insights otherwise not available into the engineered tissue development.