From Concepts to Inhibitors: A Blueprint for Targeting Protein-Protein Interactions

Protein-protein interactions are no longer considered undruggable because of the conceptual and technical advances that allow inhibitors to be generated using rational design principles and high-throughput screening methods. Here we review the concepts and approaches that have underpinned the progress in this field. We begin by assessing what makes a protein surface more tractable than others with a focus on the recent success in targeting Ras, which has long served as a poster child of a therapeutically important yet undruggable target. We discuss computational approaches to dissect protein surfaces to design macrocycles and miniprotein ligands. Traditional drug discovery has benefitted from leveraging natural products but this benefit has not extended to the design of ligands for protein surfaces because few natural products have been characterized as inhibitors of protein complexes. However, nature does provide a template in the form of binding epitopes of partner proteins. We review design of protein structure mimics that enable rational design of inhibitors through multiple weak contacts. Lastly, we focus on contemporary screening methods that are being merged with constrained peptides to offer unprecedented side chain diversity on conformationally defined scaffolds. We will focus on the concepts underlying advancements in the field rather than the application of these concepts and technologies that have led to inhibitors of specific interactions.