Many enzymes that synthesize steroid hormones are strongly bound to membranes. This membrane anchor is important for activity but complicates biochemical and biophysical studies. P450c17 (17- hydroxylase/17, 20-1yase) occupies a critical position in steroidogenesis, regulating the flux of precursors down pathways to mineralocorticoids, glucocorticoids, and sex steroids. The 17, 20-1yase activity of P450c17 is stimulated 10-fold by cytochrome b5 (b5), but the mechanism of this stimulation is not known. The action of b5 is part of the physiology of adrenarche and gonadal development, and this b5-P450c17 interaction looms as an ideal target for pharmacological inhibition of sex steroid synthesis. Abnormally high 17, 20-1yase activity underlies common androgen excess states including polycystic ovary syndrome, which affects 5-10% of reproductive-aged women, and 21-hydroxylase deficiency, which affects 1/14,000 live births. Detailed knowledge of the fine structure of these proteins and identification of the residues responsible for the b5-P450c17 interaction would advance our understanding of the complex mechanisms that regulate 17, 20-1yase activity and will ultimately facilitate rational design of highly specific, non-steroidal inhibitors of sex steroid biosynthesis, for treating androgen excess and breast and prostate cancer. We have developed a system of fusion proteins that link human P450c17 with cytochrome b5 to bury the hydrophobic surface of P450c17 in a protein-protein interaction rather than in the membrane. Several initial versions of these fusion proteins are active when expressed in yeast or HEK-293 cells. We will build a series of constructs with portions of the membrane-spanning regions deleted, with the goal of preserving at least partial activity while improving solubility and optimizing expression. We will determine if the fusion proteins are active in HEK-293 cells and yeast. Our approach allows subcloning, using purification tags, into vectors that permit high-level expression in E coil or Sf9 cells for purification. As a second goal, we will use computer modeling and site-directed mutagenesis to identify the residues on b5 that stimulate the 17, 20- lyase activity of P450c17. The work performed in this pilot and feasibility study will thus identify fusion proteins that are suitable for crystallization and structure determination studies and will ascertain which region(s) of b5 stimulate high 17, 20-lyase activity of P450c17. In addition, this approach might be generally applicable to structural studies of membrane-bound proteins.
