Discovery, chemistry, and reproductive pharmacology of asoprisnil and related 11beta-benzaldoxime substituted selective progesterone receptor modulators (SPRMs).

Asoprisnil (J 867; benzaldehyde, 4-[(11beta, 17beta)-17-methoxy-17-(methoxymethyl)-3-oxoestra-4, 9-dien-11beta-yl]-, 1-oxime) is the prototype of a novel class 11beta-benzaldoxime-substituted selective progesterone receptor modulators (SPRMs) and the first-in-class SPRM to reach an advanced stage of clinical development for the treatment of uterine fibroids and endometriosis. This compound was selected in a drug discovery program aimed to identify progesterone receptor (PR) ligands with predominant agonist but also some antagonist activities. The screening program included a range of receptor binding studies and a hierarchy of in vivo tests. A series of 11beta-benzaldoxime-substituted steroidal compounds exhibiting mixed PR agonist/antagonist effects were synthesized and characterized. For inclusion in this class of compounds, two methods of synthesis were developed and optimized. The 11beta-benzaldoxime-substituted SPRMs showed high PR binding affinities, reduced glucocorticoid receptor affinities compared with the antiprogestin mifepristone, marginal androgen receptor binding affinities, and no binding to estrogen receptors. Animal tests in guinea pigs (luteolysis inhibition assay) and rabbits (McPhail test) constituted the secondary screening tests. A mosaic of progesterone agonist and antagonist effects were found in various models. The most agonistic compounds were selected for further evaluation in animal models with respect to labor induction and endometrial effects. Unlike progesterone antagonists, asoprisnil and related compounds showed marginal effects on labor and parturition in guinea pigs. Proof-of-concept studies in nonhuman primates revealed endometrial antiproliferative effects of selected compounds, including asoprisnil and J 1042, in the presence of amenorrhea and follicular phase estradiol concentrations. Asoprisnil was selected for further clinical development. It shows promising results in the treatment of uterine leiomyomata and endometriosis.

The significance of the 20-carbonyl group of progesterone in steroid receptor binding: a molecular dynamics and structure-based ligand design study.

Polar functional groups in the A- and D-ring (positions 3 and 17beta or 20) are common to all natural and synthetic steroid hormones. It was assumed that these pharmacophoric groups are involved in strong hydrogen bonding interactions with the respective steroid receptors. High resolution X-ray structures of the estrogen and androgen receptors have confirmed these assumptions. Also site-directed mutagenesis studies of the human progesterone receptor (hPR) suggest an important role for Cys891 in the recognition of the progesterone 20-carbonyl group. Surprisingly, the crystal structure of the hPR ligand binding domain (LBD) in complex with progesterone suggests that the carbonyl oxygen in position 20 (O20) is not involved in hydrogen bond contacts. To investigate these surprising and contradicting results further, we performed a molecular dynamics simulation of the hPR-progesterone complex in an aqueous environment. The simulation revealed hPR-Cys891 as the sole but weak hydrogen bonding partner of progesterone in the D-ring. In contrast to the site-directed mutagenesis data a major role of hPR-Cys891 in progesterone recognition could not be confirmed. Isolated hydrogen bond acceptors, such as the prosterone O20 group, in a relatively lipophilic environment of the receptor led to a decrease in affinity of the ligand. Based on this consideration and the structure of the PR, we designed compounds lacking such an acceptor function. If the X-ray structure and the calculations were right, these compounds should bind with comparable or higher affinity versus that of progesterone. E-17-Halomethylene steroids were synthesized and pharmacologically characterized in vitro and in vivo. Although the compounds are unable to form hydrogen bonds with the hPR in the D-ring region, they bind with superior affinity and exert stronger in vivo progestational effects than progesterone itself. Our investigations have confirmed the results of the X-ray structure and disproved the old pharmacophore model for progestogenic activity, comprising two essential polar functional groups on both ends of the steroid core. The 20-carbonyl group of progesterone is likely to play a role beyond PR-binding, e.g. in the context of other functions via the androgen and mineralocorticoid receptors and as a site of metabolic inactivation.