Antimicrobial, cytotoxic, and insulin-releasing activities of the amphibian host-defense peptide ocellatin-3N and its L-lysine-substituted analogs

The host-defense peptide ocellatin-3N (GIFDVLKNLAKGVITSLAS.NH2), first isolated from the Caribbean frog Leptodactylus nesiotus, inhibited growth of clinically relevant Gram-positive and Gram-negative bacteria as well as a strain of the major emerging yeast pathogen Candida parapsilosis. Increasing cationicity while maintaining amphipathicity by the substitution Asp(4)-> Lys increased potency against the microorganisms by between 4- and 16-fold (MIC <= 3 mu M) compared with the naturally occurring peptide. The substitution Ala(18)-> Lys and the double substitution Asp(4)-> Lys and Ala(18)-> Lys had less effects on potency. The [D4K] analog also showed 2.5- to 4-fold greater cytotoxic potency against non-small-cell lung adenocarcinoma A549 cells, breast adenocarcinoma MDA-MB-231 cells, and colorectal adenocarcinoma HT-29 cells (LC50 values in the range of 12-20 mu M) compared with ocellatin-3N but was less hemolytic to mouse erythrocytes. However, the peptide showed no selectivity for tumor-derived cells [LC50 = 20 mu M for human umbilical vein endothelial cells (HUVECs)]. Ocellatin-3N and [D4K]ocellatin-3N stimulated the release of insulin from BRIN-BD11 clonal beta-cells at concentrations >= 1 nM, and [A18K]ocellatin-3N, at concentrations >= 0.1 nM. No peptide stimulated the release of lactate dehydrogenase at concentrations up to 3 mu M, indicating that plasma membrane integrity had been preserved. The three peptides produced an increase in intracellular [Ca2+] in BRIN-BD11 cells when incubated at a concentration of 1 mu M. In view of its high insulinotropic potency and relatively low hemolytic activity, the [A18K] ocellatin analog may represent a template for the design of agents with therapeutic potential for the treatment of patients with type 2 diabetes.