The anterior cruciate ligament (ACL) is the most frequently injured ligament of the knee, with upwards of 100,000 ACL reconstructions performed annually. Current grafting techniques are limited by insufficient integration with subchondral bone and donor site morbidity issues related to graft harvest, potentially resulting in revision surgery and long-term joint pain. Therefore, significant demand exists for alternative grafting solutions that do not require additional surgery and can regenerate the native ACL-tobone interface to promote biological fixation of the implanted ACL graft. To address this need, the ideal system must be able to withstand the functional demands of the native tissue by demonstrating physiologically equivalent mechanical properties, be comprised of compositionally varying phases in order to recapitulate the inherent heterogeneity of the native ligament to bone transition and be biodegradable such that it is gradually replaced by the regenerated tissue following implantation. It is hypothesized that a biomimetic, multi-phased scaffold comprised of optimized bone, interface and ligament regions coupled with controlled chemical and/or mechanical stimulation in vitro will guide phasespecific differentiation of mesenchymal stem cells (MSC) and result in a biologically integrated boneligament-bone complex in vivo. Mesenchymal stem cells are particularly attractive for this application as they can be routinely harvested from bone marrow, have been shown to respond to chemical, mechanical and structural cues, and are capable of differentiating towards the primary cell types (fibroblasts, osteoblasts and chondrocytes) found within ligament, bone and the ligament-to-bone interface.

To this end, a nanofiber-based synthetic graft was designed with compositionally-varying phases to regenerate ligament, bone and interface tissues. The ligament phase was optimized in terms of nanofiber alignment, composition, mechanical stimulation and chemical stimulation. It was demonstrated that an aligned nanofiber substrate coupled with controlled mechanical stimulation was necessary to differentiate MSCs towards a fibroblastic phenotype. The bone phase was optimized in terms of ceramic content and it was shown that a threshold of mineral incorporation into nanofibers was necessary to differentiate MSCs towards an osteogenic phenotype. Lastly, a mechanoactive nanofiber collar was designed to induce interface formation. It was demonstrated that compressive stimulation applied via nanofiber collar contraction induced chondrogenic differentiation of MSCs. Subsequently, the three phases were incorporated to form a synthetic graft, for which graft architecture and cell seeding density were optimized. The resulting graft was cultured in vitro under the optimized parameters, demonstrating the formation of distinct and structurally continuous regions of bone, interface and ligament tissue. The graft was implanted in vivo where it was shown to be suitable for ACL reconstruction as it maintained knee stability and promoted ligament regeneration.

In summary, this thesis focuses on the design of a biomimetic, nanofiber-based, integrated boneligament-bone construct, and elucidates chemical, mechanical and scaffold design-related parameters that can guide MSC differentiation towards desired tissue types. The impact of these studies extends beyond ligament reconstruction as they yield valuable scaffold design criteria, establish scaffold and culturing-related parameters to induce stem cell differentiation and can readily be applied to the formation of interfaces between soft-to-hard tissues as well as other complex tissues.