Porcine leukocytes contained three homologous peptides, PG‐1, 2 and 3, that manifested potent microbicidal activity against Escherichia coli, Listeria monocytogenes and Candida albicans in vitro. The peptides (‘protegrins’) were composed of 16 (PG‐2) or 18 amino acid residues, and, like tachyplesins (broad‐spectrum antibiotic peptides of horseshoe crab hemocytes), they contained two intramolecular cystine disulfide bonds. Considerably smaller than defensins, protegrins nevertheless showed substantial homology to them, especially to the ‘corticostatic’ rabbit defensin, NP‐3a. The relatively simple structure of protegrins should provide useful prototypes for constructing congeners with selectively enhanced host defense activities.

We purified three homologous antimicrobial peptides ('gallinacins') from chicken leukocytes, examined their antimicrobial activity in vitro, and established their primary sequences by a combination of gas phase microsequencing and on-line IX-ESI-MS analysis of endo-and exoprotease peptide digests. The peptides contained 3&39 amino acid residues, were relatively cationic due to their numerous lysine and arginine residues, and each contained 3 intramolecular cystine disulfide bonds. Gallinacins showed primary sequence homology to the recently delineated p-defensin family, heretofore found only in the respiratory epithelial cells and neutrophils of cattle, suggesting that /3-defensins originated at least 250 million years ago, before avian and mammalian lineages diverged. The 9 invariant residues (6 cysteines, 2 glycines and 1 proline) common to avian gallinacins and bovine p-defensins are likely to constitute the essential primary structural motif of this ancient family of host-defense peptides.

No abstract

Two novel 21-residue antimicrobial peptides, arenicin-1 and arenicin-2, exhibiting activity against Gram-positive and Gram-negative bacteria and fungi, were purified from coelomocytes of marine polychaeta Arenicola marina (lugworm) by preparative gel electrophoresis and RP-HPLC. Molecular masses (2758.3 and 2772.3 Da) and complete amino acid sequences (RWCVYAYVRVRGVLVRYRRCW and RWCVY-AYVRIRGVLVRYRRCW) 1 were determined for each isoform. Each arenicin has one disulfide bond (Cys3-Cys20). The total RNA was isolated from the lugworm coelomocytes, RT-PCR and cloning were performed, and cDNA was sequenced. A 202-residue preproarenicin contains a putative signal peptide (25 amino acids) and a long prodomain. Arenicins have no structure similarity to any previously identified antimicrobial peptides.

No abstract