We propose to determine by X-ray crystallography the three-dimensional structures of two membrane-bound enzymes from human placenta: cytochrome P450 aromatase (P450arom) and estrone/dehydroepiandrosterone sulfatase (ES). P450arom and ES, along with 17beta-hydroxysteroid dehydrogenase type 1, constitute the three-enzyme system responsible for biosynthesis of active estrogens. High levels of estrogen have been linked to the initiation and proliferation of breast tumors. Current therapeutic agents, such as Tamoxifen and its analogs, are estrogen-receptor antagonists that have limited efficacy and produce undesirable side effects. The enzymes involved in estrogen biosynthesis provide alternative and attractive anti-breast cancer drug targets, and the ultimate objective of this project is to design such drugs. The strategic lowering of active estrogen levels at or near breast tumors, by simultaneous inhibition of two or all three members of this enzyme system, could turn out to be the most pragmatic way to prevent proliferation of breast-tumor cells. Elucidation of the P450arom and ES crystal structures will lead to better understanding of the molecular mechanisms of action of these enzymes and the rational design of transition-state analogs as inhibitors, thereby minimizing or eliminating undesirable side effects. We have already grown diffraction-quality crystals of ES and small single crystals of P450arom. We will continue to use both conventional and automated methods to screen systematically a large number of crystallization conditions for P450arom. We have prepared several antibody fragments that bind P450arom with high affinity and specificity. We will use these fragments as tools for crystallizing P450arom in the form of soluble complexes, enhancing the overall chances of obtaining good crystals of this difficult-to-crystallize membrane protein. The latter approach provides an opportunity to investigate the molecular basis of immune-specific interactions between P450arom and the complementarity-determining regions of the antibody, and this information could be exploited for the design of activity-suppressing pharmacophores or immune-responsive peptides with vaccine implications.
