Corticotropin releasing factor (CRF) is a 41-peptide amide which stimulates the release of ACTH (Vale et al. Science 1981, 213, 1394). CRF has been postulated to assume an alpha-helical conformation upon binding to its pituitary receptor (Hernandez et al. J. Med. Chem. 1993, 36, 2860). We have exploited this hypothesis in the design of a limited series of cyclic analogues and have taken into consideration the effects of side-chain deletion (Alanine scan, Kornreich et al. J. Med. Chem. 1992, 35, 1870) as well as of changes in chirality (Rivier et al. J. Med. Chem. 1993, 36, 2851), with the rationale that side chains necessary for binding could also be replaced by side-chain bridges. In particular, we have used computer modeling to predict likely side chain bridging opportunities and evaluated the effects of such replacements by correlating biological results with those derived from CD spectroscopy. We have synthesized 38 monocyclic peptide amides, competitive antagonists of human/rat CRF, using solid-phase methodology on MBHA resin. After purification by preparative RP-HPLC, the peptides were analyzed by RP-HPLC and capillary zone electrophoresis and characterized by mass spectroscopy and amino acid analysis. CRF antagonists were tested for their ability to interfere with CRF-induced release of ACTH by rat anterior pituitary cells. In most cases, one of the bridge heads was located at a position where substitution by a D-residue was tolerated (i.e., positions 12 and 20). It has become clear that careful optimization of bridge length and chirality is critical. This is best exemplified by the fact that out of the 38 analogues that were synthesized and tested, only two, [cyclo(20-23)[DPhe12,Glu20,Lys23, Nle21,38]h/rCRF12-41 and cyclo(20-23)[DPhe12,Glu20,Orn23,Nle21,38] h/rCRF12-41], were found to be more potent (3 and 2 times, respectively) than [DPhe12,Nle21,38]h/rCRF12-41, the parent compound. Six analogues belonging to two different families were found to be half as potent as the standard, 18 had 2-20% of the potency of the standard, and the others were significantly less potent. CD results of all analogues in 50% TFE (a concentration of TFE that induced nearly maximum helicity of [DPhe12,Nle21,38]h/rCRF12-41) suggest that while helicity may be an important factor for CRF analogue recognition, little correlation is found between percent helicity as determined by spectral deconvolution and biological activity in vitro.

alpha-Conotoxin MII, isolated from Conus magus, is a potent peptidic toxin which specifically targets the mammalian neuronal nicotinic acetylcholine receptor, alpha3beta2 subtype. The three-dimensional structure of alpha-conotoxin MII in aqueous solution has been determined by two-dimensional 1H NMR spectroscopy. NOE-derived distances, refined by an iterative relaxation matrix approach, as well as dihedral and chirality restraints were used in high-temperature biphasic simulated annealing calculations. Fourteen minimum energy structures out of 50 subjected to the SA simulations were chosen for evaluation; these 14 structures have a final RMS deviation of 0.76 +/- 0.31 and 1.35 +/- 0.34 A for the backbone and heavy atoms, respectively. The overall structure is unusually well-defined due to a large helical component around the two disulfide bridges. The principal backbone folding motif may be common to a subclass of alpha-conotoxins. There are two distinct surfaces on the molecule almost at right angles to one another. One entirely consists of the hydrophobic residues Gly1, Cys2, Cys3, Leu15, and Cys16. The second comprises the hydrophilic residues Glu11, His12, Ser13, and Asn14. These surfaces on the ligand could be essential for the subtype-specific recognition of the receptor.

In three earlier publications (Miranda et al. J. Med. Chem. 1994, 37, 1450-1459; 1997, 40, 3651-3658; Gulyas et al. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 10575-10579) we have hypothesized that covalent constraints such as side-chain-to-side-chain lactam rings would stabilize an alpha-helical conformation shown to be important for the recognition and binding of the CRF C-terminus 30 residues, to CRF receptors. These studies led to the discovery of useful CRF antagonists such as alpha-helical CRF (alpha-hel-CRF) and Astressin both in vitro and in vivo. To test the hypothesis that such lactam rings may also be modulating activation of the receptor when introduced at the N-terminus of CRF, we studied the influence of the successive introduction from residues 4 to 14 of a cyclo(i, i+3)[Lysi-Glu(i+3)] and a cyclo(i,i+3)[Glui-Lys(i+3)] bridge on the in vitro potency of the agonist [Ac-Pro4,dPhe12,Nle21,38]hCRF(4-41) and related compounds. We have also introduced the favored cyclo(Glu30-Lys33) substitution found to be remarkable in several families of antagonists (such as Astressin) and in a number of CRF agonists and investigated the role of residues 4-8 on receptor activation using successive deletions. Earlier studies had shown that in both oCRF and alpha-helical CRF, deletion of residues 1-6, 1-7, and 1-8 led to gradual loss of intrinsic activity (IA) (from 50% IA to <10% IA) resulting in alpha-hel-CRF being a potent competitive antagonist. We show that acetylation of the N-terminus of these fragments generally increases potency by a factor of 2-3 with no influence on IA. While cyclo(30-33)[Ac-Leu8,dPhe12,Nle21, Glu30,Lys33,Nle38]hCRF(8-41) (30) is the shortest reported analogue of CRF to be equipotent to CRF (70% IA), the corresponding linear analogue (31) is 120 times less potent (59% IA). Addition of one amino acid at the N-terminus ¿cyclo(30-33)[Ac-Ser7,dPhe12,Nle21, Glu30,Lys33,Nle38]hCRF(7-41) (28)¿ results in a 5-fold increase in agonist potency and full intrinsic activity (113%). The most favored modifications were also introduced in other members of the CRF family including sauvagine (Sau), urotensin (Utn), urocortin (Ucn), and alpha-hel-CRF. Parallel and consistent results were obtained suggesting that the lactam cyclization at residues 29-32 and 30-33 (for the members of the CRF family with 40 and 41 amino acid residues, respectively) will induce (in the shortened agonists) a structural constraint (alpha-helix) that stabilizes a bioactive conformation similar to that shown in the Astressin family of CRF antagonists and that residue 8 (leucine or isoleucine) bears the sole responsibility for activation of the receptor since deletion of that residue leads to potent antagonists (Gulyas et al. Proc. Natl. Acad.Sci. U.S.A. 1995, 92, 10575-10579).