GPCR-MAPS: high-resolution functional and allosteric mapping of G protein-coupled receptor activation and bias

Abstract G protein-coupled receptors (GPCRs) are the largest family of human receptors and drug targets, but how the signaling properties of receptors are encoded in their sequence is incompletely understood. Most GPCR drugs bind a conserved orthosteric pocket, which can result in non-specificity, toxicity and clinical failure. Precision modalities including biased signaling and allosteric modulation could lead to GPCR therapeutics with improved efficacy and safety profiles. The rational design of precision therapies is, however, limited by the lack of high-resolution functional annotation of GPCR structures. Here we present a fast and general approach to rapidly build complete, high-resolution, multi-modal functional and allosteric maps of receptors. The approach, GPCR-MAPS, quantifies direct recruitment of proteins to a receptor and deconvolves the substantial effects of mutations on receptor expression. Applying GPCR-MAPS to the β2 adrenergic receptor generates >150,000 phenotypic measurements, including the full activation functions for all possible amino acid substitutions (>7,500 unique variants) in a single experiment. The multi-modal maps provide numerous mechanistic insights and reveal a modular receptor architecture, with a core activation network surrounded by residues controlling quantitative parameters and bias. The maps also identify multiple allosteric surface pockets, including pockets bound by serendipitously discovered allosteric ligands and a novel pocket with no known ligands. The application of this approach across the superfamily of GPCRs will provide comparative maps of receptor mechanisms and a functional framework for the rational design of precision GPCR therapeutics.