MD simulations and multivariate studies for modeling the antileishmanial activity of peptides
MetadataShow full item record
Leishmaniasis, a protozoan-caused disease, requires alternative treatments with minimized side-effects and less prone to resistance development. Antimicrobial peptides represent a possible choice to be developed. We report on the prospection of structural parameters of 23 helical antimicrobial and leishmanicidal peptides as a tool formodeling and predicting the activity of new peptides. This investigation is based on molecular dynamic simulations (MD) in mimetic membrane environment, as most of these peptides share the feature of interacting with phospholipid bilayers. To overcome the lack of experimental data on peptides' structures, we started simulations from designed 100% -helices. This procedure was validated through comparisons with NMR data and the determination of the structure of Decoralin-amide. From physicochemical features and MD results, descriptors were raised and statistically related to the minimum inhibitory concentration against Leishmania by the multivariate data analysis technique. This statistical procedure confirmed five descriptors combined by different loadings in five principal components. The leishmanicidal activity depends on peptides' charge, backbone solvation, volume, and solvent-accessible surface area. The generated model possesses good predictability (q(2)=0.715, r(2)=0.898) and is indicative for the most and the least active peptides. This is a novel theoretical path for structure-activity studies combining computational methods that identify and prioritize the promising peptide candidates.