Medroxyprogesterone acetate and dexamethasone are competitive inhibitors of different human steroidogenic enzymes

Medroxyprogesterone acetate (MPA), a widely used progestin, can suppress the hypothalamic-pituitary-gonadal axis but can also directly inhibit gonadal steroidogenesis; the success of MPA as a treatment for gonadotropin-independent sexual precocity derives from its direct action on steroidogenic tissues. Dexamethasone, a widely used glucocorticoid, can suppress the hypothalamic-pituitary-adrenal axis, but its potential effect directly on the adrenal is unclear. Previous reports suggested that these two drugs may act on the initial steps in the rodent steroidogenic pathway; therefore, we investigated their abilities to inhibit the first three human enzymes in steroidogenesis: the cholesterol side-chain cleavage enzyme (P450scc), the 17alpha-hydroxylase/17,20-lyase (P450c17), and type II 3beta-hydroxysteroid dehydrogenase/isomerase (3betaHSDII). We found no effect of either drug on P450scc in intact human choriocarcinoma JEG-3 cells. Using microsomes from yeast expressing human P450c17 or microsomes from human adrenals, we found that dexamethasone inhibited P450c17 with a Ki of 87 micromol/L, which is about 1000 times higher than typical therapeutic concentrations, but that MPA has no detectable action on P450c17. Using microsomes from yeast expressing human 3betaHSDII, we found that this enzyme has indistinguishable apparent Km values of 5.2-5.5 micromol/L and similar maximum velocities of 0.34-0.56 pmol steroid/min x microg microsomal protein for the three principal endogenous substrates, pregnenolone, 17-hydroxypregnenolone, and dehydroepiandrosterone. In this system, MPA inhibited 3betaHSDII with a Ki of 3.0 micromol/L, which is near concentrations achieved by high therapeutic doses of 5-20 mg MPA/kg x day. These data establish the mechanism of action of MPA as an inhibitor of human steroidogenesis, and are in contrast with the results of earlier studies indicating that MPA inhibited both P450c17 and 3betaHSD in rat Leydig cells. These studies establish the "humanized yeast" system as a model for studying the actions of drugs on human steroidogenic enzymes and suggest that 3betaHSDII may be an appropriate target for pharmacological interventions in human disorders characterized by androgen excess or sex steroid dependency.