Thesis (Ph.D)--Chemistry and Biochemistry, Georgia Institute of Technology, 2009. / Committee Chair: Doyle, Donald; Committee Member: Barry, Bridgette; Committee Member: Bommarius, Andreas; Committee Member: Ledoux, Joe; Committee Member: Matsumura, Ichiro; Committee Member: Oyelere, Adegboyega. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Nuclear receptors (NRs) are modular ligand-activated transcription factors that control a broad range of physiological processes by regulating the expression of essential genes involved in cell physiology, differentiation, and metabolism. These receptors are implicated in a number of diseases and due to their profound role in development and disease progression and their modularity, much emphasis is being put forth into nuclear receptor based drug discovery and engineering these receptors to bind novel small molecules Chemical Complementation (CC) is a yeast three-hybrid genetic selection system that was developed to aid in the discovery of these engineered receptors by linking the survival of a yeast cell to a small molecules ability to activate the receptor. Due to several advantages, to include faster growth times and higher transformation efficiencies, we have attempted to extend chemical complementation from yeast to E. coli. The bacterial chemical complementation system (BCC) was designed, based on a bacterial two hybrid system, to parallel yeast CC system. However, bacterial chemical complementation did not produce ligand dependent activation due to heterologous protein expression. In a second project designed to further NR based protein engineering and drug discovery, CC was used to evaluate a library of charge reversal variants rationally designed to gain a better understanding of nuclear receptor function and structure and to produce orthogonal ligand receptor pairs. A library of retinoic acid receptor (RARα) variants were developed based on five residues in the binding pocket known to stabilize the natural negatively charged ligand, all-trans retinoic acid (atRA). We altered the binding selectivity of the receptor to bind positively charged retinoid ligands. We were able to engineer two triple variants capable of activating with the positively charged retinoid but not the natural atRA ligand, however they do not activate as well as RARα wild-type does with atRA. In a third project we characterized covalently linked tamoxifen and histone deacetylase inhibitor based dual inhibiting compounds as breast cancer therapeutics. Several dual inhibiting compounds were found to decrease the proliferation of ER positive breast cancer cells better than tamoxifen alone, the HDACi alone, or noncovalently linked HDACi and tamoxifen.