New lead compounds in the search for selective modulators of glucocorticoid activities

BURTON, GERARDO

Buenos Aires

Workshop; Strategies for the development of new drugs; 2008

Facultad de Farmacia y Bioquímica, UBA

Antiglucocorticoids (antiGC) that act as antagonists at the GR level may be used to block or modulate the undesirable effects of GC excess  (from endogenous or exogenous origin). The search for an antiGC resulted in RU-486 in the early 80’s, also a potent antiprogestin and abortifacient. Subsequent developments were mostly focused in derivatives with antiprogestin activity and free of antiGC activity. Although the need for a pure antiGC persisted, work in this direction generally used RU-486, with its bulky aromatic substituent at C-11, as a lead. Further on, in view of the variety of physiological processes in which GCs are involved, selective antiGCs that can block only some of these processes (eventually with tissue specificity) would be highly desirable. Our approach started from the observation that the introduction of a 6,19-epoxy bridge in the progesterone molecule led to a bent structure at the A/B ring junction devoid of MC and GC activity. Introduction of a 21-hydroxyl in the latter molecule resulted in a GR antagonist (21OH-619OP) still with no affinity por MR and PR. 1 The 3D structure of this molecule resembles that adopted by RU-486 in the complex with the ligand binding domain (LBD) of GR.2Antagonistic activity was evidenced by partial blocking of dexamethasone induction of TAT and thymocyte apoptosis. Taking this compound as a lead, we replaced the oxygen bridge by a sulfur bridge to give a less bent, more flexible molecule. 21-hydroxy-6,19-Epithioprogesterone exhibited improved antiapoptotic activity on thymocytes but was less effective blocking TAT induction. This selectivity was improved further by oxidation to the sulfone. The sulfoxide only had agonistic activity. None of these compounds showed antiprogestin activity and some of them are potential dissociated GCs or antiGCs. A similar overall molecular shape but less lipophylic and with higher metabolic stability was attained by introduction of a methylene bridge (6,19-methanoprogesterone). More recently we synthesized a highly bent steroid with a direct bond between C-6 and C-19 (6,19-cycloprogesterone and its 21-hydroxy derivative). This structure showed affinity for GR but not for the other nuclear receptors. Molecular dynamics simulations of 21OH-619OP  - GR-LBD complex have shown that the ligand does not interact with helix 12 of the GR, resulting in a passive antagonism which is selective for the GR. The antagonistic activity would result from a change in receptor conformation affecting the dimerization region.3 Finally introduction of a hemisuccinate at C-21 of 21OH-619OP changes the activity profile rendering agonistic properties. At the molecular level, a molecular dynamics simulation shows an interaction of the hemisuccinate terminal carboxylate with Tyr-735 of helix 11 that returns the dimerization region to its “active” conformation.