The growing problem of resistance of bacteria and other microbes to current antibiotic drugs directs the search for new antimicrobial substances. Among possible substitutes for traditional antibiotic drugs are antimicrobial peptides (AMPs), which show promising levels of antimicrobial activity but are often toxic to humans. We plan to design new potent AMPs that are nontoxic to human cells by studying a set of related peptides. Our current goal is to determine the sequence and structural properties of specific peptides that are responsible for their antimicrobial activity and toxicity through the use of molecular dynamics (MD) simulations. Such simulations provide molecular level detail about the interactions of the peptides with membrane mimics.

Thus far we have carried out MD simulations of four related ß-hairpin antimicrobial peptides in zwitterionic dodecylphosphocholine (DPC) micelles and anionic sodium dodecylsulfate (SDS) micelles. These systems are considered to be models of mammalian and bacterial membrane interfaces, respectively. Our peptides of interest are based on the sequence of Protegrin-1 (PG-1), a potent AMP isolated from porcine leukocytes that is harmfully toxic to humans. Simulations of PG-1 [RGGRL CYCRR RFCVC VGR] have yielded insights into the sequence elements responsible for activity against bacterial species, notably Leu-5, and those elements responsible for toxicity, in particular Phe-12 and Val-14. Further simulations of peptides with mutations on the C-terminus of PG-1 have supported this hypothesis, as mutations decreasing the hydrophobicity of the C-terminus reduce toxicity, but retain the activity of the peptide. Currently we are working with our collaborators at UCLA Medical School to design and test new sequences based on these results.

This work was supported by a grant from NIH (GM 070989). Computational support from the Minnesota Supercomputing Institute is gratefully acknowledged. This work was also partially supported by National Computational Science Alliance.