The goal of this research is to develop an experimental evolution system in which hormone receptors evolve affinity for new ligands. The human estrogen receptor alpha (ERalpha and two major breast cancer drugs that are ERalpha antagonists or response modulators will be used as a test case. This system will allow the structure-function relationships that determine interactions between estrogen receptors and their ligands to be evaluated using the exponential efficiency of natural selection -- a major advance over the currently used methods of directed mutagenesis and library screening, which require the painstaking production and evaluation of mutant receptors for new functions. This system will also address fundamental questions about the dynamics and mechanistic basis of molecular evolution and applied questions about disease progression, because it will recapitulate in the laboratory the processes by which tumor cells evolve resistance to drugs and, over evolutionary time, receptor proteins in normal cells evolve new functions. He system will consist of an engineered yeast strain in which growth rate depends upon the expression of genes controlled by estrogen response elements, and expression of these genes depends in turn on the activation of a recombinant human ERalpha by ligands added to the culture medium. When cultured in the laboratory, yeast with variant ERs that are better activated by the ligand will be generated by mutation; in the presence of a growth-limiting dose of a novel ligand, these variants will increase in frequency due to natural selection. Over many generations, ERs that use the novel ligand as a high-affinity agonist will evolve. The gene sequences of evolving receptors will be obtained at frequent intervals, their functions characterized using reporter gene assays, their three-dimensional structures modeled, and the dynamics of their evolution evaluated. Repeated over numerous experimental and control replicates, this system will allow the rigorous testing of hypotheses about ER structure-function relationships and the dynamics of receptor-ligand coevolution. This system can ultimately be extended to study the coevolution of virtually any receptor-ligand interaction, including those between any hormone, growth factor, or neurotransmitter and its receptor, as well as those between host and pathogen proteins.

R21GM070792

2004-05-01

2007-04-30