Computational design of heterochiral peptides against a helical target

摘要

Polypeptides incorporating D-amino acids occasionally occur in nature and are an important class of pharmaceutical molecules. With the use of heterochiral Monte Carlo (HCMC), a method inspired by the de novo design of proteins, we develop peptide scaffolds for interacting with a molecular target, a left-handed alpha-helix. The HCMC approach concurrently seeks to optimize a peptide sequence, its internal conformation, and its docked conformation with a target surface. Several major classes of interactions are observed: (1) homochiral interactions between two alpha(L) helices, (2) heterochiral interactions between an alpha(L) and an alpha(R) helix, and (3) heterochiral interactions between the alpha(L) target and novel nonhelical structures. We explore the application of HCMC to simulating the preferential enantioselectivity of heterochiral complexes. Implications for biomimetic design in molecular recognition are discussed.