Computational Study of the Interaction between Antimicrobial Compounds and Efflux Systems of Gram-Negative Bacteria

Abstract

Nowadays, multidrug resistance in Gram-negative bacteria is a major issue for public health. The efflux pumps of the resistance nodulation division (RND) family contribute largely to this phenomenon. A well-known example is represented by the AcrAB-TolC efflux system of E. coli, in which the inner membrane translocase AcrB is the main responsible for the uptake and extrusion of different substrates. Due to the limited availability of crystallographic structures of AcrB-substrate complexes, computational methods represent an alternative approach to elucidate the nature of interactions between diverse antimicrobial compounds and the efflux protein. Here, two antibiotics known for their different affinity for the efflux pump in P. aeruginosa, namely meropenem and imipenem (respectively, strongly and poorly affected by MexB, the homologous of AcrB in P. aeruginosa), were chosen for this study. A similar behaviour for the two antibiotics has not been yet reported in the literature for the AcrB of E. coli.In this study we first performed a preliminary docking investigation using ensemble of conformations extracted from molecular dynamics simulations of both ligands and AcrB structures. The top poses of ligands were selected as initial conformations for molecular dynamics simulations. Results from MD simulations evidence a larger propensity of meropenem than imipenem to bind to the putative affinity sites of AcrB, in analogy to what observed for MexB. Moreover, a detailed analysis of the molecular interactions of the two compounds with AcrB allowed identifying the key residues involved in the extrusion/retention processes. Our results provide information that could help in designing new antibiotics less likely to be extruded by RND efflux pumps, as well as inhibitors more effective in blocking the whole efflux process.