Computational design of allosteric pathways reprograms ligand-selective GPCR signaling

Abstract

G-protein-coupled receptors (GPCRs) constitute the largest family of signaling receptors and drug targets. However, understanding how variations in receptor sequence, ligand chemical structure, and binding impact signaling functions remains a challenge, hindering drug discovery. Here, we developed a computational protein structure and dynamics approach to infer and design GPCR responses to multiple ligands. We created 32 dopamine D1 and D2 receptor variants with widely reprogrammed agonist-induced signal transductions. Subtle natural and designed receptor sequence variations, predicted to alter specific structural and dynamic mechanisms of ligand responses, profoundly impacted ligand potency and efficacy in agreement with our calculations. Our study provides a rational blueprint for computing the effect of sequence polymorphisms on ligand-selective protein signaling and paves the way for advancements in pharmacogenomics, drug selectivity, and the design of signaling receptors from first principles.