Dynamical nonequilibrium molecular dynamics simulations reveal allosteric networks, signal transduction mechanisms, and sites associated with drug resistance in biomolecular systems

The dynamical approach to nonequilibrium molecular dynamics (D-NEMD), conceptualised by Ciccotti et al. in the 1970s, has seen resurgence in recent years. In the biomolecular simulation field, the technique provides novel utility in the study of signal propagation and allosteric effects in biological macromolecules. Through comparison of equilibrium MD simulations and perturbed nonequilibrium simulations, the D-NEMD approach provides clear maps of the time-dependent structural response of proteins to a perturbation, and straightforward assessment of the statistical significance of the responses. D-NEMD has recently been shown to complement various equilibrium-based allosteric analysis techniques, such as shortest path maps and distance fluctuation analyses. Here, we review recent applications of D-NEMD to biomolecular systems. D-NEMD simulations identify allosteric ‘hotspots’ in the oncotarget K-Ras4B; an allosteric binding site, and sites associated with drug resistance in the SARS-CoV-2 main protease. In the SARS-CoV-2 spike, D-NEMD simulations showed the fatty acid binding site connects to distant, functionally relevant sites, and have probed the effects of pH changes. D-NEMD identified a general mechanism of signal propagation in nicotinic acetylcholine receptors. In class A β-lactamases, reveal the communication networks between allosteric and active sites, and pinpoint sites which, when mutated, alter antibiotic resistance spectrum of activity.