Petrow V, Padilla G M, McPhail A T, Bruchovsky N, Schneider S L
Department of Cell Biology, Duke University, Durham, NC 27710.
J Steroid Biochem. 1989 Mar;32(3):399-407. doi: 10.1016/0022-4731(89)90213-6.
To aid in the design of new inhibitors of steroidal 5 alpha-reductase for treatment of prostate cancer, we have studied the topography of the 5 alpha-reductase active site (5 alpha-R) and of the related androgen (RA) and progesterone (RP) receptors in the region complementary to C.6 of progesterone. To this end we have determined the total structures of 17 alpha-acetoxy-6-methylene-4-pregnene-3,20-dione (VII; R = H) and of 17 beta-hydroxy-6,6-ethylene-4-androsten-3-one (VIa) by X-ray crystal structure analysis and, using these data, have developed Newman projections of the 6 alpha-Me, 6 beta-Me, 6-methylene and 6,6-ethylene derivatives of progesterone. From them we have developed a Newman projection of a composite model formed from steroids (V), (VI), (VIIIa) and (VIIIb). This is shown in Fig. 4 and illustrates the relative conformations of these substituents around C.6. From there we proceeded to receptor-binding studies. Our results led to the conclusion that androgen receptor, (RA), takes up preferred but different conformations when bound to testosterone (T) and to 17 beta-hydroxy-5 alpha-androstan-3-one (5 alpha-dihydrotestosterone, DHT), respectively, and that the resulting steroid-receptor complexes bind preferentially to different chromatin acceptor sites. We have therefore used the convention RT and RDHT in place of RA as appropriate. Working on the assumption that binding affinities reflect spatial contours, we have developed comparative silhouettes for the 5 alpha-R, RP and RDHT protein binding sites complementary to C.6 of the steroidal ligand. These data show that the 5 alpha-reductase active site is characterized by a hydrophobic pocket which specifically accommodates a 6-methylenic moiety and partially accommodates a 6 beta-methyl group. RDHT, in contrast, shows much less specificity and largely accommodates all the above substituents. Progesterone receptor differs in failing to accommodate 6,6-ethylene and 6 beta-methyl, with minimal accommodation of 6-methylene. It possesses a hydrophobic pocket skewed towards the alpha-face of the steroid, thereby allowing optimal binding of the 6 alpha-methyl substituent to the receptor. 6-Methylene-4-pregnene-3,20-dione (V) fails to bind significantly to androgen and progesterone receptors thereby supporting the postulate that its antiprostatic activity stems primarily from 5 alpha-reductase inhibition.
为辅助设计用于治疗前列腺癌的新型甾体5α-还原酶抑制剂，我们研究了5α-还原酶活性位点（5α-R）以及相关雄激素（RA）和孕激素（RP）受体在与孕酮C.6互补区域的拓扑结构。为此，我们通过X射线晶体结构分析确定了17α-乙酰氧基-6-亚甲基-4-孕烯-3,20-二酮（VII；R = H）和17β-羟基-6,6-亚乙基-4-雄烯-3-酮（VIa）的完整结构，并利用这些数据绘制了孕酮的6α-甲基、6β-甲基、6-亚甲基和6,6-亚乙基衍生物的纽曼投影式。基于此，我们构建了由甾体（V）、（VI）、（VIIIa）和（VIIIb）组成的复合模型的纽曼投影式。如图4所示，该图展示了这些取代基在C.6周围的相对构象。在此基础上，我们开展了受体结合研究。我们的研究结果表明，雄激素受体（RA）在分别与睾酮（T）和17β-羟基-5α-雄甾烷-3-酮（5α-双氢睾酮，DHT）结合时会呈现出不同但偏好的构象，且由此形成的甾体-受体复合物会优先结合到不同的染色质受体位点。因此，我们在适当的时候使用RT和RDHT来代替RA。基于结合亲和力反映空间轮廓这一假设，我们绘制了与甾体配体C.6互补的5α-R、RP和RDHT蛋白结合位点的比较轮廓图。这些数据表明，5α-还原酶活性位点的特征是一个疏水口袋，它能特异性容纳一个6-亚甲基部分，并部分容纳一个6β-甲基基团。相比之下，RDHT的特异性要低得多，它能容纳上述所有取代基。孕激素受体的不同之处在于它不能容纳6,6-亚乙基和6β-甲基，对6-亚甲基的容纳也极少。它有一个偏向甾体α面的疏水口袋，从而使6α-甲基取代基能与受体实现最佳结合。6-亚甲基-4-孕烯-3,20-二酮（V）与雄激素和孕激素受体的结合不显著，这支持了其抗前列腺活性主要源于对5α-还原酶抑制作用的假设。
