Constrained corticotropin-releasing factor antagonists with i-(i + 3) Glu-Lys bridges

Hypothesis driven and systematic structure-activity relationship (SAR) investigations have resulted in the development of effective central nervous system (CNS) antagonists of corticotropin (ACTH)-releasing factor (CRF) such as alpha-helical CRF(9-41) and analogues of our assay standard [DPhe12,Nle21,38]hCRF(12-41). On the other hand, equally potent CRF antagonists that block the hypothalamic/pituitary/adrenal (HPA) axis had not been described until recently. Predictive methods, physicochemical measurements (nuclear magnetic resonance spectrometry and circular dichroism spectroscopy), and SAR studies suggest that CRF and its family members (urotensins and sauvagine) assume an alpha-helical conformation when interacting with CRF receptors. To further test this hypothesis, we have systematically scanned the hCRF(9-41) or hCRF(12-41) sequences with an i-(i + 3) bridge consisting of the Glu-Xaa-Xbb-Lys scaffold which we and others had shown could maintain or enhance alpha-helical structure. From this series we have identified seven analogues that are either equipotent to, or 3 times more potent than, the assay standard; in addition, as presented earlier cyclo(30-33)[DPhe12,-Nle21,38,Glu30, Lys33]hCRF(12-41) (astressin) is 32 times more potent than the assay standard in blocking ACTH secretion in vitro (rat pituitary cell culture assay). In vivo, astressin is also significantly more potent than earlier antagonists at reducing hypophysial ACTH secretion in intact stressed or adrenalectomized rats. Since the corresponding linear analogues that were tested are significantly less potent, our interpretation of the increased potency of the cyclic analogues is that the introduction of the side chain to side chain bridging element (Glu30-Lys33, and to a lesser extent that of Glu14-Lys17, Glu20-Lys23, Glu23-Lys26, Glu26-Lys29, Glu28-Lys31, Glu29-Lys32, and Glu33-Lys36) induces and stabilizes in the receptor environment a putative alpha-helical bioactive conformation of the fragment that is not otherwise heavily represented. The effect of the introduction of two favored substitutions [(cyclo(20-23) and cyclo(30-33)] yielded 37 with a potency 8 times that of the assay standard but actually 12 times less than expected if the effect of the two cycles had been multiplicative. These results suggest that the pituitary CRF receptor can discriminate between slightly different identifiable conformations, dramatically illustrating the role that secondary and tertiary structures play in modulating biological signaling through specific protein-ligand interactions.