Conformational restraints revealing bioactive beta-bend structures for halpha CGRP8-37 at the CGRP2 receptor of the rat prostatic vas deferens

CGRP physiology, pharmacology, and therapeutic targets: migraine and beyond

Calcitonin gene-related peptide (CGRP) is a neuropeptide with diverse physiological functions. Its two isoforms (α and β) are widely expressed throughout the body in sensory neurons as well as in other cell types, such as motor neurons and neuroendocrine cells. CGRP acts via at least two G protein-coupled receptors that form unusual complexes with receptor activity-modifying proteins. These are the CGRP receptor and the AMY1 receptor; in rodents, additional receptors come into play. Although CGRP is known to produce many effects, the precise molecular identity of the receptor(s) that mediates CGRP effects is seldom clear. Despite the many enigmas still in CGRP biology, therapeutics that target the CGRP axis to treat or prevent migraine are a bench-to-bedside success story. This review provides a contextual background on the regulation and sites of CGRP expression and CGRP receptor pharmacology. The physiological actions of CGRP in the nervous system are discussed, along with updates on CGRP actions in the cardiovascular, pulmonary, gastrointestinal, immune, hematopoietic, and reproductive systems and metabolic effects of CGRP in muscle and adipose tissues. We cover how CGRP in these systems is associated with disease states, most notably migraine. In this context, we discuss how CGRP actions in both the peripheral and central nervous systems provide a basis for therapeutic targeting of CGRP in migraine. Finally, we highlight potentially fertile ground for the development of additional therapeutics and combinatorial strategies that could be designed to modulate CGRP signaling for migraine and other diseases.

The Current State of Peptide Drug Discovery: Back to the Future?

Over the past decade, peptide drug discovery has experienced a revival of interest and scientific momentum, as the pharmaceutical industry has come to appreciate the role that peptide therapeutics can play in addressing unmet medical needs and how this class of compounds can be an excellent complement or even preferable alternative to small molecule and biological therapeutics. In this Perspective, we give a concise description of the recent progress in peptide drug discovery in a holistic manner, highlighting enabling technological advances affecting nearly every aspect of this field: from lead discovery, to synthesis and optimization, to peptide drug delivery. An emphasis is placed on describing research efforts to overcome the inherent weaknesses of peptide drugs, in particular their poor pharmacokinetic properties, and how these efforts have been critical to the discovery, design, and subsequent development of novel therapeutics.

Three-dimensional structure and orientation of rat islet amyloid polypeptide protein in a membrane environment by solution NMR spectroscopy

Islet amyloid polypeptide (IAPP or amylin) is a 37-residue peptide hormone associated with glucose metabolism that is cosecreted with insulin by beta-cells in the pancreas. Since human IAPP is a highly amyloidogenic peptide, it has been suggested that the formation of IAPP amyloid fibers is responsible for the death of beta-cells during the early stages of type II diabetes. It has been hypothesized that transient membrane-bound alpha-helical structures of human IAPP are precursors to the formation of these amyloid deposits. On the other hand, rat IAPP forms transient alpha-helical structures but does not progress further to form amyloid fibrils. To understand the nature of this intermediate state and the difference in toxicity between the rat and human versions of IAPP, we have solved the high-resolution structure of rat IAPP in the membrane-mimicking detergent micelles composed of dodecylphosphocholine. The structure is characterized by a helical region spanning the residues A5 to S23 and a disordered C-terminus. A distortion in the helix is seen at R18 and S19 that may be involved in receptor binding. Paramagnetic quenching NMR experiments indicate that rat IAPP is bound on the surface of the micelle, in agreement with other nontoxic forms of IAPP. A comparison to the detergent-bound structures of other IAPP variants indicates that the N-terminal region may play a crucial role in the self-association and toxicity of IAPP by controlling access to the putative dimerization interface on the hydrophobic face of the amphipathic helix.

Clinical and preclinical rationale for CGRP-receptor antagonists in the treatment of migraine

Calcitonin gene-related peptide (CGRP) is linked to migraine and other primary headache disorders. It is found in every location described in migraine genesis and processing, including meninges, trigeminal ganglion, trigeminocervical complex, brainstem nuclei, and cortex. It is released in animal models following stimulation of the CNS similar to that seen in migraine, and triptans inhibit this release. Injection of CGRP into migraineurs results in delayed headache similar to migraine. Elevation of CGRP occurs during migraine, resolving following migraine-specific treatment. Finally, and most importantly, CGRP receptor antagonists terminate migraine with efficacy similar to triptans. Both intravenous olcegepant (BIBN 4096 BS) and oral telcagepant (MK-0974) have been effective, safe, and well tolerated in phase I and II studies. Telcagepant is currently in phase III trials, and preliminary results are favorable. The potential for a migraine-specific medication without vasoconstrictive or vascular side effects is enormous. CGRP receptor blockade may also have applications in other pathologic and pain syndromes.

Calcitonin gene-related peptide analogues with aza and indolizidinone amino acid residues reveal conformational requirements for antagonist activity at the human calcitonin gene-related peptide 1 receptor

Calcitonin gene-related peptide antagonists have potential for the treatment and prevention of disease states such as non-insulin-dependent diabetes mellitus, migraine headache, pain, and inflammation. To gain insight into the spatial requirements for CGRP antagonism, three strategies were employed to restrict the conformation of the potent undecapeptide antagonist, [D31,P34,F35]CGRP27-37. First, aza-amino acid scanning was performed, and ten aza-peptide analogues were synthesized and examined for biological activity. Second, (3S,6S,9S)-2-oxo-3-amino-indolizidin-2-one amino acid (I2aa) and (2S,6S,8S)-9-oxo-8-amino-indolizidin-9-one amino acid (I9aa) both were introduced at positions 31-32, 32-33, 33-34, and 34-35, regions of the backbone expected to adopt turns. Finally, the conformation of the backbone and side-chain of the C-terminal residue, Phe35-Ala36-Phe37-NH2, was explored employing (2S,4R,6R,8S)-9-oxo-8-amino-4-phenyl-indolizidin-9-one amino acid (4-Ph-I9aa) as a constrained phenylalanine mimic. The structure-activity relationships exhibited by our 26 analogues illustrate conformational requirements important for designing CGRP antagonists and highlight the importance of beta-turns centered at Gly33-Pro34 for potency.

Identification of the key residue of calcitonin gene related peptide (CGRP) 27-37 to obtain antagonists with picomolar affinity at the CGRP receptor

Calcitonin gene related peptide (CGRP) plays an important role in the CNS and in the cardiovascular system. To identify high-affinity antagonists in competitive binding studies, we identified a novel radioactive tracer, [(3)H-propionyl-K(24)]-halphaCGRP 8-37, which was labeled in solution by a recently developed strategy using photolabile protecting groups at reactive side chains. This tracer was shown to be as potent as commercially available (125)I-tracers for the determination of agonists and to have increased sensitivity for antagonists. We applied it to investigate the predicted turn structures centered at Pro(29) and Pro(34). The substitution at positions 29 and 34 by turn-inducing amino acid mimetica showed that these turns are highly diverse. At position 29, a hydrophobic residue is preferred that constricts the secondary structure, whereas position 34 is required to stabilize the conformation of the backbone. All high-affinity analogues showed antagonistic properties with potency similar to CGRP 8-37.

Design, synthesis, and application of azabicyclo[X.Y.0]alkanone amino acids as constrained dipeptide surrogates and peptide mimics

Azabicyclo[X.Y.0]alkanone amino acids are challenging synthetic targets and useful tools for studying structure-activity relationships of native peptide ligands. They have been employed to increase potency and stability in conformationally rigid enzyme inhibitors and receptor ligands. Since last reviewed in 1997, activity in their synthesis and application has increased significantly and access is now available to a wider diversity of these peptide mimics. This review focuses on recent syntheses of these heterocyclic amino acids and their application in the investigation of biologically active peptides and peptide mimics.

Modifications to the N-terminus but not the C-terminus of calcitonin gene-related peptide(8-37) produce antagonists with increased affinity

Seventeen novel analogues of human calcitonin gene-related peptide(8-37) (hCGRP(8-37)) were synthesized by solid-phase methods and purified to apparent homogeneity by semipreparative cation exchange and/or reversed-phase high-performance liquid chromatography. The C-terminal Phe was replaced by Gly, cyclohexylalanine (Cha), Tyr, all four isomers of beta-methylphenylalanine (beta-MePhe), and l- and d-tetrahydroisoquinoline carboxylic acid (Tic), resulting in analogues 3-11. For the synthesis of the beta-MePhe-containing analogues 6-9, crystallization was used to separate a mixture of all four isomers of beta-MePhe into the erythro pair of enantiomers (2S,3S, 2R,3R) and the threo pair of enantiomers (2S,3R, 2R,3S), which were then converted to Fmoc derivatives and used in two separate syntheses. Two diastereomeric peptides were obtained from each synthesis and were separated by RP-HPLC to yield enantiomerically pure 6-9. Substitution of Tyr for Phe caused no change in binding affinity at CGRP receptors. All other substitutions for Phe resulted in substantial reductions in binding affinity. Indeed, no binding was observed for analogues 7, 9, and 11, all of which contained a d-amino acid residue in the C-terminal position, and the binding affinities of the remaining analogues were >10-fold lower than that of h-alpha-CGRP(8-37). These data suggest that a conformationally flexible phenyl ring in the C-terminal position of h-alpha-CGRP(8-37) is preferred for high-affinity binding to CGRP receptors. Acetylation, benzoylation, and benzylation of the N-termini of h-alpha-CGRP(8-37) and h-beta-CGRP(8-37) produced analogues 12-14 and 16-18, respectively. A byproduct was isolated by RP-HPLC from the resin-cleaved crude product of each benzylated analogue, which was characterized as the dibenzylated derivative of h-alpha-CGRP(8-37) and h-beta-CGRP(8-37) (analogues 15 and 19, respectively). Amino acid analysis and (1)H NMR showed that the second benzyl group was located on the C4 carbon of the imidazole ring of His(10). Radioligand binding experiments showed that derivatizing the N-termini substantially increased binding affinities at CGRP receptors. The benzoylated and dibenzylated derivatives had the highest affinities, which were approximately 50-fold greater than those of h-alpha-CGRP(8-37). Functional experiments confirmed that the N-terminally derivatized analogues of h-alpha-CGRP(8-37) are antagonists that are more potent than h-alpha-CGRP(8-37). In conclusion, these studies underscore the importance of Phe(37) of h-alpha-CGRP(8-37) for binding to CGRP receptors and have identified the N-terminus and His(10) as two positions that can be used for the design of antagonists with increased affinity for CGRP receptors.

The role of the 8-18 helix of CGRP8-37 in mediating high affinity binding to CGRP receptors; coulombic and steric interactions

1. The role of individual residues in the 8-18 helix of CGRP(8-37) in promoting high-affinity binding to CGRP(1) receptors expressed on rat L6 and human SK-N-MC cells has been examined. The relative potencies of various derivatives were estimated from their ability to inhibit the human alphaCGRP-mediated increase in cyclic AMP production and the binding of [(125)I]-human alphaCGRP. 2. Arg(11) and Arg(18) were replaced by serines to give [Ser(11,18)]CGRP(8-37). These bound with pKi values <6 to SK-N-MC cells and had apparent pA(2) values of 5.81+/-0.04 and 5.31+/-0.11 on SK-N-MC and L6 cells. CGRP(8-37) had a pKi of 8.22 on SK-N-MC cells and pK(b) values on the above cell lines of 8.95+/-0.04 and 8.76+/-0.04. 3. The arginines were replaced with glutamic acid residues. [Glu(11)]CGRP(8-37) had a pK(b) of 7.14+/-0.14 on SK-N-MC cells (pKi=7.05+/-0.05) and 6.99+/-0.08 on L6 cells. [Glu(18)]CGRP(8-37) had a pK(b) of 7.10+/-0.0.08 on SK-N-MC cells (pKi=6.91+/-0.23) and 7.12+/-0.09 on L6 cells. 4. Leu(12), Leu(15) and Leu(16) were replaced by benzoyl-phenylalanine (bpa) residues. On SK-N-MC cells, the apparent pA(2) values of [bpa(12)]-, [bpa(15)]- and [bpa(16)]CGRP(8-37) were respectively 7.43+/-0.23, 8.34+/-0.11 and 5.66+/-0.16 (pKi values of 7.14+/-0.17, 7.66+/-0.21 and <6): on L6 cells they were 7.96+/-0.36, 8.28+/-0.21 and 6.09+/-0.04 (all n=3). 5. It is concluded that the Arg(11) and Arg(18) are involved in specific electrostatic interactions with other residues, either on the CGRP(1) receptors or elsewhere on CGRP(8-37). Leu(16) is in a conformationally restricted site when CGRP(8-37) binds to CGRP(1) receptors, unlike Leu(12) and Leu(15).

Probing opioid receptor-ligand interactions by employment of indolizidin-9-one amino acid as a constrained Gly(2)-Gly(3) surrogate in a leucine-enkephalin mimic

The relationship between the conformation and biological activity of Leu-enkephalin was studied using (2S,6R,8S)-9-oxo-8-N-(Boc)amino-1-azabicyclo[4.3.0]nonane-2-carboxylic acid [(2S,6R,8S)-1, I(9)AA] as a constrained Gly(2)-Gly(3) dipeptide surrogate. [I(9)AA](2,3)-Leu-enkephalin 12 was assembled using solid-phase peptide synthesis on Merrifield resin with TBTU as the coupling reagent. The in vitro assays indicated that [I(9)AA](2,3)-Leu-enkephalin 12 exhibited affinities for the mu- and delta-opioid receptors that were three orders of magnitude lower than that of Leu-enkephalin, as well as partial agonist character for both receptors. In in vivo assays for spinal analgesia, the indolizidinone analog 12 showed significantly enhanced duration of action, indicating an increased metabolic stability. Conformational analysis was performed using NMR and CD spectroscopy. The amide temperature coefficients and 3J(NH-CalphaH) coupling constants for 12 could not support a hydrogen-bonded beta-turn structure; however, its CD spectrum indicated a turn conformation. Incorporation of indolizidinone amino acid 1 into Leu-enkephalin thus provided additional support for the importance of a turn conformation for the biological activity of the native peptide.

CGRP(2) receptor in the internal anal sphincter of the rat: implications for CGRP receptor classification

The CGRP receptor mediating relaxation of the rat internal anal sphincter (IAS) has been characterized using CGRP analogues, homologues, the antagonist CGRP(8 - 37) and its analogues. In isolated IAS strips, the spontaneously developed tone was concentration-dependently relaxed by halpha CGRP, hbeta CGRP and rat beta CGRP (pEC(50) 8.1+/-0.2, 8.3+/-0.1 and 8.4+/-0.2, respectively; 100% maximum response). Vasoactive intestinal polypeptide (VIP) was around 7 fold more potent than halpha CGRP (pEC(50) 9.0+/-0.1; 100% maximum relaxation). [Cys(ACM(2.7))] halpha CGRP and salmon calcitonin were inactive (up to 10(-5) M). Halpha CGRP(8 - 37) (10(-5) M) antagonized responses to halpha CGRP (apparent pK(B) 5.7+/-0.3) and rat beta CGRP (apparent pK(B) 5.8+/-0.2), but not to VIP. Hbeta CGRP(8 - 37) (10(-5) M) was an antagonist against halpha CGRP (apparent pK(B) 6.1+/-0.1). Halpha CGRP(8 - 37) analogues (10(-5) M), with substitutions at the N-terminus by either glycine(8) or des-NH(2) valine(8) or proline(8), antagonized halpha CGRP responses with similar affinities (apparent pK(B) 5.8+/-0.1, 5.8+/-0.1 and 5.5+/-0.1, respectively). Peptidase inhibitors (amastatin, bestatin, captopril, phosphoramidon and thiorphan, 10(-6) M each) did not increase the agonist potency of either halpha CGRP or [Cys(ACM(2,7))] halpha CGRP, or the antagonist affinity of halpha CGRP(8 - 37) against halpha CGRP or rat beta CGRP. These data demonstrate for the first time a CGRP receptor in the rat IAS for which halpha CGRP (8 - 37) and its analogues have an affinity that is consistent with a CGRP(2) receptor. However, there is a marked species difference as the antagonist has a 100 fold lower affinity in the rat than in the same tissue of the opossum (Chakder & Rattan, 1991).

Bioactive beta-bend structures for the antagonist halpha CGRP(8 - 37) at the CGRP(1) receptor of the rat pulmonary artery

The aim of this study was to determine beta-bend structures and the role of the N- and C-terminus in the antagonist halpha CGRP(8 - 37) at the rat pulmonary artery CGRP receptor mediating halpha CGRP relaxation. Halpha CGRP(8 - 37) Pro(16) (10(-6) M), with a bend-biasing residue (proline) at position 16, did not antagonize halpha CGRP responses, while a structure-conserving amino acid (alanine(16)) at the same position retained antagonist activity (apparent pK(B) 6.6+/-0.1; 10(-6) M). Halpha CGRP(8 - 37) Pro(19) (10(-6) M), with proline at position 19 was an antagonist (apparent pK(B) 6.9+/-0.1). Incorporation of a beta-bend forcing residue, BTD (beta-turn dipeptide), at positions 19 and 20 in halpha CGRP(8 - 37) (10(-6) M) antagonized halpha CGRP responses (apparent pK(B) 7.2+/-0.2); and BTD at positions 19,20 and 33,34 within halpha CGRP(8 - 37) was a competitive antagonist (pA(2) 7.2; Schild plot slope 1.0+/-0.1). Halpha CGRP(8 - 37) analogues, substituted at the N-terminus by either glycine(8) or des-NH(2) valine(8) or proline(8) were all antagonists (apparent pK(B) 6.9+/-0.1; (10(-6) M), 7.0+/-0.1 (10(-6) M), and pA(2) 7.0 (slope 1.0+/-0.2), respectively); while replacements by proline(8) together with glutamic acid(10,14) in halpha CGRP(8 - 37) (10(-6) M) or alanine amide(37) at the C-terminus of halpha CGRP(8 - 37) (10(-5) M) were both inactive compounds. In conclusion, possible bioactive structures of halpha CGRP(8 - 37) include two beta-bends (at 18 - 21 and 32 - 35), which were mimicked by BTD incorporation. Within halpha CGRP(8 - 37), the N-terminus is not essential for antagonism while the C-terminus may interact directly with CGRP(1) receptors in the rat pulmonary artery.