Decrypting the languages of allostery in membrane-bound K-Ras4B using four complementary in silico approaches

Abstract

Proteins are evolved molecular machines whose diverse biochemical functions can be dynamically regulated by allostery, through which even faraway protein residues can conformationally communicate. Allostery can express itself in different ways, akin to different "languages": allosteric control pathways predominating in an unperturbed protein are superseded by others as soon as a perturbation arises--e.g., a mutation--that alters its function (pathologically or not). Accurately modeling these often-unintuitive phenomena could therefore help explain functional changes in a specific protein whenever they are unclear. Unbiased molecular dynamics (MD) simulations are a possibility; however, since allostery can operate at longer timescales than those accessible by MD, simulations require integration with a reliable method able to, e.g., detect regions of incipient allosteric change or likely perturbation pathways. Several methods exist but are typically applied singularly: we argue their joint application, in a "multilingual" approach, could significantly enrich this picture. To prove this, we perform unbiased MD simulations ([~]100 {micro}s) of the widely studied, allosterically active oncotarget K-Ras4B, solvated and embedded in a phospholipid membrane, proceeding to decrypt its allostery using four showcase "languages": Distance Fluctuation analysis and the Shortest Path Map capture allosteric communication hotspots at equilibrium; Anisotropic Thermal Diffusion and Dynamical Non-Equilibrium MD simulations assess them once the GTP that oncogenically activates K-Ras4B is, respectively, either superheated or hydrolyzed. "Languages" provide a uniquely articulate, mutually coherent, experimentally consistent picture of allostery in K-Ras4B. At equilibrium, pathways stretch from the membrane-embedded hypervariable region all the way to the active site, touching known flexible allosteric "switches" and proposed pockets. Upon GTP cleavage/perturbation, known to inactivate K-Ras4B, allosteric signals most reverberate on switches and interfaces that recruit effector proteins. Our work highlights the benefits of integrating different allostery detection methods with unbiased MD.