Probing GPCR allosteric activation and biased signaling with rigidity propagation and geometric Monte-Carlo simulations.

Abstract

GPCR diverse signaling pathways are driven by complex pharmacology arising from a functional conformational ensemble and allosteric modifications rarely captured by structural methods. Our allostery algorithms, which are based on mathematical rigidity theory, provide a mechanical interpretation of allosteric signaling and are designed to predict if perturbation of rigidity at one site of the protein can propagate across a network and in turn cause a change in rigidity at a second distant site, resulting in allosteric transmission. We model the GPCR as a constraint network (graph) consisting of nodes (atoms) and edges (i.e., constraints such as covalent bonds, electrostatic interactions, hydrogen bonds, and hydrophobic contacts). Starting with a set of known GPCR structures we can detect how ligands at either orthosteric site or other regions trigger rigidity changes which propagate to functionally important regions (such as G-protein interface or GDP pocket) in turn cause a rearrangement and alteration of the shape of the cytoplasmic binding region. In collaborations with NMR experimentalists, 19FNMR and computational allostery predictions was used to delineate allostery and distinct key functional activation and precoupled intermediate states in A2AR complexed with heterotrimeric G-protein. NMR showed that binding of G-protein stabilizes a precoupled activation intermediate state and two distinct active states which facilitate nucleotide exchange by full or partial agonists. Rigidity theory allostery algorithms showed Gbg is critical in facilitating allostery within the ternary complex. Recent work investigates the selectivity and efficacy in A2AR in the presence of different G protein subtypes using simulations, allostery predictions and NMR.