MECHANISTIC INSIGHTS INTO MULTIDRUG EFFLUX PUMP ACTIVITIES AND THEIR INHIBITION

Abstract

The rise of multidrug-resistant (MDR) Gram-negative pathogens poses a critical threat to global public health. The Resistance-Nodulation-Division (RND) family efflux pumps are central to this resistance, which actively expel a broad range of antibiotics, detergents, and host-derived compounds from bacterial cells. Despite their clinical importance, the detailed molecular mechanisms by which RND pumps coordinate substrate recognition, energy transduction, and channel gating remain incomplete, particularly regarding how environmental cues and inhibitors dynamically regulate these processes. This dissertation dissects the molecular mechanisms and functional regulation of two archetypal RND systems: the AcrAB—TolC efflux pump of Escherichia coli and the MexEF—OprN pump of Pseudomonas aeruginosa. Through a multidisciplinary approach integrating site-directed mutagenesis, in vivo efflux assays, disulfide trapping, inhibitor studies, and collaborative hydrogen-deuterium exchange mass spectrometry (HDX-MS) conducted by Eamonn Reading’s group at King’s College London, molecular docking carried out by Dr. Paolo Ruggerone's group at the University of Cagliari, and molecular dynamics (MD) simulations performed by James C. Gumbart’s group at Georgia Institute of Technology, we provide new insights into how conformational dynamics, environmental conditions, and small-molecule inhibitors converge to control efflux pump activity and antibiotic resistance. Chapter 3 elucidates the role of the membrane fusion protein (MFP), AcrA, in the AcrAB—TolC efflux system. Consistent with prior structural studies, our mutagenesis and covalent locking studies reveal that interdomain flexibility within AcrA is essential for coupling substrate-induced conformational changes in AcrB to TolC opening. Targeted cysteine substitutions at eight hinge residues demonstrated that AcrA tolerates single-point mutations without losing efflux function. However, thiol-reactive locking of key hinge sites, L50, T205, and N232, abolished efflux activity, pinpointing the importance of these regions. Environmental modulation experiments further demonstrated that AcrAB—TolC is functional at a wide range of pH (from pH 5.0 to pH 8.0) but its activity is pH-sensitive, with optimal efflux between pH 6.0 and 7.0, and sharply reduced activity at extreme acidic or alkaline pH. Through mutational analysis, H285 of AcrA emerged as a protonation-sensitive protein residue: substitution to alanine abolished efflux under acidic conditions but preserved activity at neutral pH, suggesting that protonation of H285 destabilizes the interaction network essential for pump activation. Collaborative HDX-MS conducted by Eamonn Reading’s group at King’s College London and MD simulations performed by James C. Gumbart’s group at Georgia Institute of Technology provided complementary insights, revealing that protonation of H285 and the absence of Mg2+ increased AcrA flexibility, particularly in the membrane-proximal and αβ barrel domains. Collectively, these findings support a model in which AcrA acts as a pH-and cation-sensitive MFP, whose interdomain flexibility is finely tuned to efflux competency in response to environmental cues. Chapter 4 builds on these findings to investigate how the coupling between AcrA and TolC governs the opening of the channel, TolC. Through systematic cysteine mutagenesis and in vivo disulfide trapping, we mapped the interface between the α-helical hairpin of AcrA and the periplasmic coiled coil of the TolC domain. Single cysteine substitutions at key tip residues, AcrA Q136, Y137, L132 and TolC Q139, Q142, R143, preserved efflux activity under tested conditions, suggesting that the interface tolerates some static contacts. However, enforced covalent trapping of the AcrAQ136C—TolCQ142C pair abolished efflux activity, indicating that structural flexibility at this interface is essential for function. Inhibitor studies demonstrated that the small molecule NSC-60339 selectively disrupted disulfide bond formation at specific AcrA–TolC sites, suggesting that it affects efflux by modulating conformational dynamics rather than physically occluding the interface. In vitro reconstitution experiments with disulfide-stabilized AcrA—AcrB complexes further support that productive TolC engagement requires pre-alignment and is not driven by spontaneous oxidation alone. These results refine the mechanistic model of efflux activation, proposing that the AcrA—TolC interface is modulated by flexible domain alignment. In conclusion, we propose that NSC-60339 stabilizes AcrA in a conformation incompatible with the TolC opening, thus turning off the efflux function of the tripartite assembly. Chapter 5 shifts our focus to MexEF—OprN efflux pump of P. aeruginosa, an RND system whose substrate specificity remains poorly understood despite the important role of MexEF—OprN in the resistance of fluoroquinolones, trimethoprim, and chloramphenicol. Through targeted mutagenesis of four nonconserved MexF residues, D132, P136, G626, and S729, we mapped the structural determinants of substrate recognition and transport. Mutations in P136 and S729 enhanced or impaired the efflux of multiple substrates, while mutations at D132 and G626 produced substrate-specific effects. Efflux accumulation assays and MIC profiling revealed that substrate-specific phenotypes are distributed throughout the MexF structure, with fluoroquinolones and trimethoprim sharing overlapping but distinct efflux pathways. Competitive inhibition studies using Hoechst 33342 further demonstrated that chemically distinct substrates compete for shared binding determinants. These findings align with collaborative ensemble docking and cluster analysis carried out by Dr. Paolo Ruggerone's group at the University of Cagliari, which showed that certain mutations reshape ligand trajectories and alter contact frequencies within the translocation pathway. Together, we propose that MexF substrate specificity arises from an integrated network of adaptable checkpoints distributed along the access and deep pockets, modulating efflux in a substrate-selective manner. Chapter 6 builds upon this mechanistic understanding to explore pharmacological targeting of MexF. By screening a small-molecule library, Dr. Zgurskaya's lab identified SLU-1642, a 2-aminobenzothiazole derivative, as a selective inhibitor of MexF-mediated efflux, and our study provides direct mechanistic details to establish SLU-1642 as a selective inhibitor of MexF. Hoechst 33342 accumulation assays demonstrated that SLU-1642 selectively inhibited MexF without affecting MexB or efflux-deficient strains, distinguishing it from broad-spectrum efflux inhibitors. SLU-1642 potentiated the activity of trimethoprim and doxycycline in MexF-overproducing strains, but had limited impact on other antibiotics, suggesting substrate-selective inhibition. Structure-activity relationship (SAR) analysis of SLU-1642 analogs revealed that modifications to the core scaffold or side-chain flexibility dramatically altered MexF inhibition, underscoring the narrow chemical window required for activity. Mechanistically, single-point mutations at MexF residues, especially S729, variably impaired SLU-1642 function. The S729W mutation abolished SLU-1642 activity across all substrates, confirming that S729 is a critical residue for substrate translocation and inhibitor engagement. The four chapters of this dissertation build a unified mechanistic framework to understand how RND efflux pumps integrate environmental cues, conformational flexibility, and pharmacological inhibition to maintain their functions. In the AcrAB—TolC system, AcrA acts as a dynamic bridge between AcrB and TolC, whose flexibility is modulated by pH, Mg2+, and small-molecule inhibitors, NSC-60339, to control efflux function. In MexEF—OprN, substrate specificity and inhibition arise from a distributed network of adaptable checkpoints within MexF, with key residues such as S729 coordinating substrate binding and transport. Across both systems, conformational trapping by selective inhibitors—NSC-60339 for AcrA and SLU-1642 for MexF—demonstrates that efflux inhibition can be achieved by stabilizing nonproductive conformations that prevent channel opening and substrate extrusion. These findings have broad implications for the development of antimicrobial drugs. They suggest that targeting the dynamics of efflux pumps may provide a new avenue for their inhibition and future overcoming multidrug resistance. By stabilizing specific conformational states of key efflux components, it may be possible to selectively disable efflux without broadly inhibiting homologous transporters, reducing the likelihood of toxicity and resistance development. In addition, our work highlights the critical role of environmental modulation in efflux pump regulation, suggesting that therapeutic efficacy may vary between infection niches with different pH and ion concentrations. In conclusion, this dissertation advances the mechanistic understanding of multidrug efflux systems in Gram-negative bacteria, demonstrating that flexibility of certain regions, environmental sensing, and conformational inhibition converge to regulate efflux competency. By integrating the experimental dissection of AcrAB-TolC and MexEF-OprN with collaborative structural and computational studies, this work establishes a basis for rational design of next-generation efflux pump inhibitors (EPIs) that target bacterial resistance mechanisms.