Rational design of viral inhibitors: Application to SARS

开发能够有效靶向和预防病毒感染的合理方法是一项重要任务

The development of rational approaches that can effectively target and prevent viral infection is a strategic objective of the WRCE. Although rational design efforts have met with occasional success, a key shortcoming is the difficulty associated with the dynamic nature of protein conformation. Namely, the viral protein targets (usually envelope proteins) do not behave as the static structures that are used to depict them. Instead, these proteins are dynamic and experience conformational fluctuations. This poses the obvious difficulty associated with designing a ligand for a structurally heterogeneous target-the ligand must be compatible with either one low energy conformation of the protein or multiple higher energy conformations in order to result in a high enough binding affinity. Here, this problem is addressed with a unique computational approach, called COREX_Design, that has been developed over the past decade, and which models proteins and peptides as ensembles of conformational states. Recent studies on the design of inhibitors to the human prion protein (PrP), and the coronavirus agent of severe acute respiratory syndrome (SARS) (SCoV), have provided proof-of-principle that COREX_Design is able to; 1) identify "thermodynamically compatible" potential binding site(s), 2) design a conformationally constrained, disulfide cross-linked cyclic peptide ligand that is structurally compatible with this site, and 3) optimize the sequence to maximize the conformational compatibility between the protein and the peptide. Using this tool, we were able to successfully design potent inhibitors of amyloid formation by PrP, and we have collected strong preliminary results for antiviral activity against SCoV, one of the targets of this proposal. The goal of this project is to demonstrate that COREX_Design can be applied as a general strategy to the development of antiviral agents. Although this new design tool can in principle be applied to any system where structural information is known about the target, it is applied here to domain 3 of the envelope protein of dengue 2 virus (DN2V) and the spike (S) protein of SCoV. During the period of funding, we will demonstrate the efficacy of the designed peptides in cell based assays, and over the course of the grant we will test lead compounds in animal studies. RELEVANCE (See instructions): NIAID Category A, B and C viruses are responsible for millions of infections each year. Few antiviral therapeutics are available to treat these infections. The approach described here leverages a unique computational modeling strategy into the development of antiviral agents. The success of this approach could offer new avenues for combating infections and thus saving lives