Antibiotics- Protein/Peptide Interfaces

While significant success has been achieved in the field of rational protein design, the computational design of protein/peptide interfaces is a tantalizing problem, which remains largely unsolved. The ability to reliably and repeatably design high-affinity interactions between proteins and peptides has been a long sought goal of computational structural biology, which would allow the manipulation of key biological processes and enable vital therapeutic applications.

One such therapeutic application that our group has been actively investigating is the potential to computationally design proteins to function as antibiotic and antimicrobial agents. A specific microbial system we have been scrutinizing to help achieve that goal is the interaction between the essential gram positive bacterial cell wall D-Ala-D-ala peptide and the natural-product glycopeptide antibiotic vancomycin. Here we have an example of a protein-like molecule interacting with a non-standard peptide ligand to therapeutic effect. We hope to use this natural system to inform our attempts to design a protein capable of replicating vancomycin’s antibiotic mode of action without being vulnerable to the same mechanisms of evolved bacterial resistance.

To accomplish this ambitious goal we will draw on established, successful Rosetta computational techniques in the areas of de novo, enzyme and protein/protein interface design. The ultimate goal of these studies and our larger exploration of protein/peptide interfaces is the creation and validation of a computational method using Rosetta for the rational, repeatable and broadly applicable design of high-affinity protein/peptide interfaces and a broader understanding of the chemical and physical basis for protein/peptide interactions.

Alumni Project Members: Andrew Morin, Joel Harp, Joseph Crivelli, Nicole Shen