Osteoporosis is a chronic disease characterized by decreasing bone mineral density and integrity, leading to an increased risk of bone fracture. In the United States, more than half of the population over 50 years old is affected by low bone mass or osteoporosis. The progression of age-related osteoporosis has been shown to correlate with reduced bone marrow mesenchymal stem cell (MSC) differentiation to osteoblasts and increased MSC differentiation to adipocytes. Current pharmacological therapies act during the bone remodeling process by either enabling the osteoblast through an anabolic process or by an antiresorption action when inhibiting the osteoclast. Shown herein, synthesized PPARδ agonists exhibit the ability to control MSC differentiation in vitro by encouraging osteogenesis. This action offers a possible novel route for treating osteoporosis while simultaneously presenting a novel use for peroxisome proliferatoractivated receptor δ (PPARδ) agonists, a receptor with currently no marketed selective drug and few selective agonists.

The lack of known PPARδ agonists limited our ability to determine what potential PPARδ has for directing osteogenesis and possibly treating osteoporosis and other bonerelated diseases. Therefore, we first synthesized additional PPARδ agonists as probes to explore the structure-activity relationships (SAR) of the receptor with regard to inducing osteogenesis in MSCs. The PPARδ agonists synthesized mimic the scaffold of the experimental PPARδ agonist, GW0742, developed by GlaxoSmithKline. The analogs were designed to conserve the pharmacologically important carboxylic acid along with the overall scaffold design containing a central heterocycle and terminal phenyl ring. Literature SAR indicated that the phenoxy acetic acid moiety used to display the carboxylic acid in GW0742, offered limited benefit over an E-cinnamic acid. Therefore, the latter was substituted in all analogs because it was more synthetically accessible and could help differentiate our synthesized analogs from those described in the literature.

Four structural locations on the scaffold were identified as points of interest for improving SAR knowledge and PPARδ activity. The first two areas explored were the central heteroatom and the terminal phenyl ring. Synthesized probes were tested using an in vitro MCS differentiation assay with compounds being scored based on their ability to direct MSC differentiation towards the osteogenic or adipogenic lineages. Commercially available PPAR standards were helpful in these experiments for understanding how different PPAR agonism affects MSC differentiation, and in-turn for understanding the PPAR activity of the synthesized compounds. Standard P PARδ agonists resulted in osteogenesis whereas PPARγ agonism resulted in adipogenesis. The results from this initial series of analogs indicated that a central N-methyl and para-trifluoromethyl were preferred for osteogenic activity and were therefore carried forward in the structural design of almost all further synthesized analogs.

The last two structural areas explored for SAR were the central heterocycle and cinnamic acid phenyl ring. Three different heterocycles were designed including two different thiazoles and a triazole. Alkyl substitutions were appended to the phenyl ring of the cinnamic acid moiety based on literature SAR indicating the importance of at least a methyl substitution for providing enhanced activity with the phenoxy moiety as present in GW0742. The results seemed to confirm the value of the methyl group and showed that lengthening the alkyl chain further was not beneficial. In total, 28 analogs were synthesized and tested for their ability to prompt in vitro MSC differentiation.

The second major part of this dissertation was to generate additional biological data on the synthesized compounds. Seven compounds were selected from the initial family of synthesized probes for testing in a biochemical assay involving transfecting COS-7 cells with the specific PPAR gene of interest, along with a luciferase reporter gene for quantifying the PPAR activation. The results showed a general agreement with the qualitative results of the MSC differentiation assay, with only small variations between the two data sets. Lastly, six compounds were selected for a study utilizing ovariectomized mice as an in vivo model for osteoporosis. In general, the results were very encouraging with the synthesized compounds improving bone density and trabecular thickness. Two compounds were especially promising and presented an osteogenic profile comparable to that of the PPARδ agonist standard, GW0742.