In silico optimization of analogs derived pro-adrenomedullin peptide to evaluate antimicrobial potential

Diverse computational approaches have been widely used to assist in designing antimicrobial peptides with enhanced activities. This tactic has also been used to address the need for new treatment alternatives to combat resistant bacterial infections. Herein, we have designed eight variants from a natural peptide, pro-adrenomedullin N-terminal 20 peptide (PAMP), using an in silico pattern insertion approach, the Joker algorithm. All the variants show an α-helical conformation, but with differences in the helix percentages according to circular dichroism (CD) results. We found that the C-terminal portion of PAMP may be relevant for its antimicrobial activities, as revealed by the molecular dynamics, CD, and antibacterial results. The analogs showed variable antibacterial potential, but most were not cytotoxic. Nevertheless, PAMP2 exhibited the most potent activities against human and animal-isolated bacteria, showing cytotoxicity only at a substantially higher concentration than its minimal inhibitory concentration (MIC). Our results suggest that the enhanced activity in the profile of PAMP2 may be related to their particular physicochemical properties, along with the adoption of an amphipathic α-helical arrangement with the conserved C-terminus portion. Finally, the peptides designed in this study can constitute scaffolds for the design of improved sequences.