Computer-aided design of mastoparan-like peptides enables the generation of non-toxic variants with extended antibacterial properties.

Diverse peptides have been evaluated for their activity against pathogenic microorganisms. Here, five mastoparan variants were designed based on mastoparan-L, among which two (R1 and R4) were selected for in-depth analysis. Mastoparan-L (parent/control), R1 and R4 inhibited susceptible/resistant bacteria at concentrations ranging from 2 to 32 μM, whereas only R1 and R4 eradicated P. aeruginosa biofilms at 16 μM. Moreover, the toxic effects of mastoparan-L toward mammalian cells were drastically reduced in both variants. In skin infections, R1 at 64 μM was the most effective variant, reducing P. aeruginosa bacterial counts 1,000-times at day four post-infection. Structurally, all the peptides showed varying levels of helicity and structural stability in aqueous and membrane-like conditions, which may affect the different bioactivities observed here. By computationally modifying the physicochemical properties of R1 and R4 we reduced the cytotoxicity and optimized the therapeutic potential of these mastoparan-like peptides both in vitro and in vivo.