Covalent Docking and Molecular Dynamics Simulations Reveal the Resistance-Granting Mutations A237R/K in TEM Beta-Lactamase

Abstract

The rate of modern drug discovery using experimental screening methods still lags far behind the rate at which pathogens mutate, highlighting the dire need for accurate and fast predictive simulations of macromolecular evolution. Particularly threatening, multidrug-resistant bacteria evade our defenses by expressing a series of resistance factors, the most famous of which is the 29 kilodaltons antibiotic cleaver, TEM beta-lactamase. Rising to this challenge, we applied biophysical simulations and predictions to identify two new resistance-conferring mutations in TEM and their respective modes of action. Using homology modeling and traditional all-atom molecular dynamics simulations, we predicted the structures of 50 TEM mutants, then deployed covalent docking to estimate their binding affinities against 85 beta-lactam-containing drugs to identify potential substrate specificity shifts. Finally, we used binding pose metadynamics and normal mode analyses to further scan the most promising results (validated against an empirical fitness dataset). This heuristic identified putative mechanisms for the previously-unreported, resistance-conferring mutations A237R and A237K. We also uncovered concerted structural rearrangements that could explain the specificity shifts observed in the clinically-entrenched mutations R164S/G/D. In addition to reaffirming the power of using simulations as molecular microscopes, our results could prove useful for guiding the rational design of next-generation beta-lactam antibiotics and bring the community closer to retaking the lead against the recurrent threat of multidrug-resistant pathogens.