Combinatorial library screening offers a rapid process for identifying potential therapies to toxins. Hinge peptide libraries, which rely on conformational diversity rather than traditional molecular diversity, reduce the need for huge numbers of syntheses and screening steps and greatly expedite the discovery process of active molecules. Hinge peptide libraries having the structures: Acetyl-X1–X2–hinge–X3–X4–NH2 (capped) and X1–hinge–X2–X3 (uncapped), where X1 through X4 are near-equimolar mixtures of twelve L-amino acids and hinge = 4-aminobutyric acid, were screened for inhibitory activity in bioassays for botulinum neurotoxins A and B (BoNT/A, BoNT/B) and saxitoxin. The zinc protease activity of the reduced light chains of BoNT/A and /B was assayed by measuring the cleavage of synthetic substrates. Saxitoxin activity was measured by the restoration of the viability of neuroblastoma cells treated with ouabain and veratridine. Deconvolution of libraries was accomplished by fixing one position at a time beginning with the C-terminus. Primary library subsets in which position 4 was fixed showed moderate levels of inhibition for BoNT/A. Secondary library subsets showed stronger inhibition in the bioassays. In each of the bioassays, inhibitory potency was stronger when the second position to be fixed was on the opposite side of the hinge, rather than on the same side with respect to the C-terminus, suggesting that the hinge facilitates the interaction of side chains. Inhibitors for all three of the toxins studied were discovered within library subsets, although not necessarily in primary subsets. These studies demonstrate that (1) the best strategy for deconvoluting hinge peptide libraries is by fixing residues alternately on each side of the hinge moiety, and (2) it is essential to investigate secondary subsets even when primary subsets are inactive. The present findings support the concept that the increased flexibility imposed by the inclusion of a central hinge residue in small peptides increases the opportunity for side chain interactions, providing a distinct advantage for hinge peptide libraries over conventional peptide libraries. Hinge peptide libraries are a rich source of novel ligands for modulation of biomechanisms. The library subsets uncovered in this study may possess peptides that will lead to effective therapies to neurotoxin poisoning.

Botulinum neurotoxin serotype A (BoNT/A) is a proteolytic enzyme that induces muscle paralysis. It is a cause of food poisoning, a potential bioterrorist threat and, in low doses an emerging pharmaceutical product. No effective treatment is currently available for BoNT intoxication. Previously we developed a BoNT/A light chain enzyme assay using a peptide substrate based on the SNAP-25 protein target, with HPLC separation and UV detection of assay products, and applied the method to screen combinatorial peptide libraries for inhibitory activity to BoNT/A. We now report on development of a capillary electrophoresis laser-induced fluorescence (CE-LIF) method for measuring BoNT/A activity. The enzyme assay products were labeled with CBQCA dye followed by CE separation on a bare fused silica column in a HEPES-based buffer and LIF detection. All assay products were separated in CE within 8 min compared to incomplete separation of assay products within 1h by HPLC. The labeled products showed linear dependence of intensity versus concentration, and quantitative mole-fraction assignments. We used the CE-LIF method to screen combinatorial peptide libraries for potential modulating effects on BoNT/A peptidase activity. With some of the libraries, peptides co-migrated with assay products and interfered with quantitation. In such cases, interference was reduced by substituting sodium dodecyl sulfate (SDS) for Tween-20 in the running buffer. Separation in the capillaries then occurred by micellar electrokinetic chromatography (MEKC). The CE-LIF method is quick and lends itself to high-throughput or microfluidic formats.

The development of drugs that selectively block angiotensin receptors has resulted largely from a process of trialand-error medicinal chemistry on an early lead. The new generation of angiotensin antagonists or angiotensin mimetics are essentially devoid of side effects and are set to replace the angiotensin converting enzyme inhibitors for the treatment of cardiovascular diseases.

Cyclic amide-linked angiotension II (ANGII) analogues have been synthesized by novel strategies, in an attempt to test the ring clustering and the charge relay bioactive conformation recently suggested. These analogues were synthesized by connecting side chain amino and carboxyl groups at positions 1 and 8, 2 and 8, 3 and 8, and 3 and 5, N-terminal amino and C-terminal carboxyl groups at positions 1 and 8, 2 and 8, and 4 and 8, and side chain amino to C-terminal carboxyl group at positions 1 and 8. All these analogues were biologically inactive, except for cyclic [Sar1, Asp3, Lys5]ANGII (analogue 10) which had high contractile activity in the rat uterus assay (30% of ANGII) and [Lys1, Tyr(Me)4, Glu8]ANGII (analogue 7) which had weak antagonist activity (PA2 approximately 6). Precyclic linear peptides synthesized using 2-chlorotrityl chloride resin and N alpha-Fmoc-amino acids with suitable side chain protection were obtained in high yield and purity and were readily cyclized with benzotriazol-1-yloxytris(dimethylamino)-phosphonium hexafluorophosphate as coupling reagent. Molecular modeling suggests that the ring structure of the potent analogue can be accommodated in the charge relay conformation proposed for ANGII.

Analogues of [Sar1]angiotensin II, Sarilesin (type I antagonist), and Sarmesin (type II antagonist) with L-azetidine-2-carboxylic acid (Aze) and L-pipecolic acid (Pip) at position 7 have been prepared by the solid-phase method, purified by reverse-phase HPLC, and bioassayed in the rat uterus. Analogues of the superagonist [Sar1]ANGII with Aze or Pip at position 7 and sarcosine (Sar) or aminoisobutyric acid (Aib) at position 1 had high intrinsic activity in the rat isolated uterus assay (34-184%). Analogues of Sarilesin ([Sar1,Ile8]ANGII) with Aze or Pip at position 7 and Sar or Aib at position 1 retained high antagonist activity (pA2 = 7.1-8.3). Analogues of Sarmesin ([Sar1,Tyr-(OMe)4]ANGII) with Aze and Pip at position 7 had pA2 Values of 7.4 and 6.5, respectively. [Aze7]-ANGII and [Pip7] ANGII had low activities (1 2 % and 1 %, respectively), and deletion of Sar at position 1 of Sarmesin analogues abolished binding (or affinity) as judged from pA2 ValUes. Nuclear Overhauser effect (NOE) spectroscopy studies of [Sar1,Aze7]ANGII in DMSO-d6 have indicated a clustering of the three aromatic rings (Tyr, His, Phe) and proximity of Sar C(alpha) and Arg C(delta) protons to the Tyr/Phe ring protons. These data emphasize that replacement of Pro with the lower and higher homologs Aze and Pip does not greatly alter the structural requirements necessary for expression of agonist or antagonist activity, when sarcosine occupies position 1, but not when Asp occupies position 1, suggesting that there is an intimate relationship between the N-terminal and penultimate residues of the molecule in the biologically active conformation of the molecule.