With the long-term goal of elucidating how mechanical stimuli can direct skeletal tissue differentiation, this dissertation employed a combination of experimental and computational approaches to quantify the local mechanical environment and the corresponding cellular responses in multiple scenarios of bone healing. These scenarios included distraction osteogenesis, a clinical procedure that uses applied tension to promote formation of new bone tissue within a bone defect;​ and mechanically generated pseudoarthrosis, a procedure in which an applied bending motion is used to promote formation of cartilage within the defect. Use of these two procedures in conjunction with each other allowed a richer investigation of the effects of mechanical cues on formation of both bone and cartilage than would have been possible with either alone. The local mechanical environment was quantified principally by an experimental technique, digital image correlation, that allowed measurement of heterogeneous strain fields within and surrounding the defect site. In the case of the pseudoarthrosis model, finite element analysis was also employed, although with the caveat that the tissue material properties that were used as input were not measured experimentally but were instead given assumed values. The techniques used to quantify the cellular responses included in situ hybridization, immunohistochemistry and histology. A final method, nanoindentation, was used in this dissertation as a first step towards addressing the limitation of the finite element analyses that this dissertation and many other prior studies have use to estimate the local mechanical environment.

Results from the distraction osteogenesis model revealed that the heterogeneity in local strain fields was perhaps both a result and a cause of the existence of distinct zones of tissue phenotypes within the healing callus. In the pseudoarthrosis model in which cartilage formation was mechanically induced, substantial spatial variations in the mRNA expression of cartilage-related genes were found in the defect region, and the patterns of expressed cartilage genes were in agreement with the predicted distribution of candidate stimuli (e.g. octahedral shear strain, fluid flow and tensile strain). Lastly, a custom-designed nanoindentation protocol was developed for quantification of tissue-level mechanical property of regenerating callus tissues.

Taken together, these findings illustrate how variations in local mechanical cues can influence tissue formation and the molecular aspects of bone healing during skeletal repair. The experimental techniques developed in this project can serve as a powerful tool for investigation of a variety of different loading scenarios to study the mechanobiology of bone healing in vivo.