No antimicrobial peptide has been identified in cephalopods to date. Annotation of transcriptomes or genomes using basic local alignment Search Tool failed to yield any from sequence identities. Therefore, we searched for antimicrobial sequences in the cuttlefish (Sepia officinalis) database by in silico analysis of a transcriptomic database. Using an original approach based on the analysis of cysteine-free antimicrobial peptides selected from our Antimicrobial Peptide Database (APD3), the online prediction tool of the Collection of Anti-Microbial Peptides (CAMPR3), and a homemade software program, we identified potential antibacterial sequences. Nine peptides less than 25 amino acids long were synthesized. The hydrophobic content of all nine of them ranged from 30 to 70%, and they could form alpha-helices. Three peptides possessed similarities with piscidins, one with BMAP-27, and five were totally new. Their antibacterial activity was evaluated on eight bacteria including the aquatic pathogens Vibrio alginolyticus, Aeromonas salmonicida, or human pathogens such as Salmonella typhimurium, Listeria monocytogenes, or Staphylococcus aureus. Despite the prediction of an antimicrobial potential for eight of the peptides, only two-GR21 and KT19-inhibited more than one bacterial strain with minimal inhibitory concentrations below 25 µM. Some sequences like VA20 and FK19 were hemolytic, while GR21 induced less than 10% of hemolysis on human blood cells at a concentration of 200 µM. GR21 was the only peptide derived from a precursor with a signal peptide, suggesting a real role in cuttlefish immune defense.
Keywords: Antimicrobial peptides; Design; Predictive tools; Sepia officinalis; Transcriptome.