Computational design and optimization of novel d-peptide TNFα inhibitors.

Abstract

Compared to small molecule drugs, peptide therapeutics provides greater efficacy, selectivity, and safety. The intrinsic disadvantages of peptides are their sensitivity to proteases. To overcome this, we have developed a general computational strategy for denovo design of protein binding helical d-peptides. A d-helical fragment library was established and used in generating flexible d-helical conformations, which were then used to generate suitable sequences with the required structural and binding properties. Using this strategy, we successfully denovo designed d-helical peptides that bind to tumor necrosis factor-α (TNFα), inhibit TNFα-TNFR1 binding, reduce TNFα activity in cellular assays, and are stable against protease digestion. Our strategy of helical d-peptide design is generally applicable for discovering d-peptide modulators against protein-protein interactions.