Knowledge-Based Modeling of G-Protein Coupled Receptors and their Binding Partners

Abstract

G-protein coupled receptors (GPCRs) represent the largest family of membrane proteins and the most heavily targeted classes of proteins for therapeutic intervention. Relatively little is known about the structure of these proteins. At present, there are experimental structures available for only about 14% of pharmacologically relevant GPCRs. This knowledge gap between known GPCR structures and the reliance on GPCR-based therapeutics in the clinical setting underlies the need for novel tools to predict the structure of these proteins. To this end, I developed tools specific to the prediction of GPCR structures using the macromolecular modeling suite Rosetta. Special attention is given to the membrane constraints and structural conservation of these often highly sequence-diverse proteins. Additionally, I present methods for the docking of either small molecule ligands, peptide ligand, or protein ligands to understand the pharmacological basis of signaling at GPCRs. Lastly, I combine the novel methods with experimental data to predict binding interactions of important peptide hormones. Taken together, these tools will best be used in a drug discovery pipeline for the identification of novel GPCR structures coupled to in silico screening of drug compounds.