Large bone defects pose a significant clinical challenge, affecting large numbers of patients at high costs. The current clinical standard for treating these defects is implantation of bone grafts. While autograft bone is the gold standard for graft material, there is generally an insufficient amount available for treating large bone defects. Devitalized allograft bone from cadavers is more readily available;​ however this material displays limited integration with host bone resulting in as many as 1/3 of these grafts failing within 2-3 years after implantation. Bone tissue engineering strategies aim to replace bone grafting procedures with treatment by a combination of a structural scaffold, biochemical cues, and / or cells capable of enhancing healing. Cellular therapies may be of particular importance when treating large bone defects because many patients lack an adequate endogenous supply of osteogenic cells or osteoprogenitor cells.

The goal of this thesis was to quantitatively compare stem-cell based strategies for treating large bone defects. First, we developed a challenging large bone defect model in immunocompromised rats for use as a reproducible test bed to quantitatively compare human stem cell-based therapies, and then we evaluated the abilities of adult and fetal stem cells to enhance defect healing when delivered on porous polymer scaffolds. Our results showed that stem cell-seeded porous polymer scaffold therapy enhanced defect healing compared to treatment with acellular scaffolds alone in the absence of added osteogenic signals, but was insufficient to fully regenerate limb function. Second, we sought to label stem cells with an in vivo tracking agent, the quantum dot, to determine biodistribution of delivered cells during the bone healing process. We showed that while quantum dots effectively label human stem cells in vitro and have negligible effects on cell viability and osteogenic differentiation in vitro, their use as a long term in vivo tracking agent was inconclusive due to uptake by host macrophages. Post mortem immunohistochemistry analysis confirmed that at least a small population of human cells remained at defect sites four weeks post implantation. Finally, we treated defects with both in vitro and in vivo osteogenic gene therapy approaches, using scaffolds coated with an adeno-associated viral (AAV) vector to encode the gene for the osteogenic signal bone morphogenetic protein 2 (BMP2) in human stem cells prior to implantation or in host defect cells after scaffold implantation. Effective BMP2 gene transfer to stem cells and induction of osteogenic differentiation was first verified in vitro. However, treatment of segmental defects with scaffolds containing BMP2-transduced stem cells (in vitro gene therapy) produced less robust healing than the in vivo gene therapy approach with scaffolds delivering the BMP2 gene to host cells.

In conclusion, this work has produced a challenging and reproducible model of large bone defects that can be used to gain new insights into the cell-mediated defect repair process through quantitative comparison of human stem cell-based bone tissue engineering therapies. This work has confirmed the therapeutic benefit of stem cellseeded construct delivery over acellular construct delivery for enhancement of defect healing in the absence of added osteogenic stimuli and suggested the therapeutic potential of fetal amniotic-fluid derived stem cells as an alternative to adult marrow-derived stem cells for treatment of large bone defects. This work has refuted the ability of the fluorescent quantum dot to serve as an effective long term in vivo cell tracking agent, which will impact the choice of cell tracking agents used in future studies of cell-mediated tissue repair therapies. Finally, this work is the first to present proof of concept results of a true off-the-shelf, donor bone graft-free orthotopic large bone defect repair therapy in which pre-sized thermostable porous polymer scaffolds lyophilized with scAAV2.5-BMP2 could be frozen at length until needed for clinical implantation in large bone defect sites.