Machine Learning-Guided Discovery of AcrB and MexB Efflux Pump Inhibitors
Abhishek Bera1, Rakesh Kumar Roy1, Pritish Joshi1
1Department of Chemistry & Chemical Biology, Indian Institute of Technology (ISM) Dhanbad, Dhanbad 826004, India.
Abstract:
Multidrug efflux pump is one of the reasons behind the antimicrobial inactivity related to infection caused by Gram-negative pathogens. The inner membrane resistance-nodulation-cell division transporter proteins, AcrB and MexB, in association with outer membrane proteins, TolC and OprM, are responsible for the extrusion of a broad range of substrates, followed by recognizing them. Although various inhibitors were proposed to stop the efflux activity of the transporter protein, none of them had been approved clinically. Our study aims to identify potent inhibitor-like molecules employing supervised classification models trained upon the molecular descriptors of previously known inhibitors. Based on the intrinsic minimum inhibitory concentration (MIC) values of the reported inhibitors, they were classified into highly potent and less potent categories. A total of 10 different classification models were built using various molecular descriptors; among them, support vector machine, Random Forest, AdaBoost, and LightGBM models appeared to deliver promising results with >80% accuracy. These top four models were implemented on a library of 5043 to obtain 8 hit molecules after the multistep filtering process. To assess their activity toward AcrB and MexB, several molecular dynamics simulations of their ligand-bound structures were performed. We also calculated the binding free-energy values and analyzed other structural properties. Mol.3488 of the unknown molecules showed higher binding affinities for both AcrB and MexB. Also, the presence of "pyridopyrimidone" and "benzothiazole" moieties in the molecules and "V"-shaped orientation of ligands inside the deep binding pocket increase the binding affinity, thereby higher inhibitory properties.
Insights
Researchers identified novel drug efflux pump inhibitors for Gram-negative infections. Machine learning models screened a library, yielding 8 hit molecules, with Mol.3488 showing high binding affinity to AcrB and MexB.
Area of Science:
- Medicinal Chemistry
- Computational Chemistry
- Microbiology
Background:
- Multidrug efflux pumps, like AcrB and MexB, confer antimicrobial resistance in Gram-negative pathogens.
- Current inhibitors targeting these pumps lack clinical approval, necessitating new therapeutic strategies.
Purpose of the Study:
- To identify novel, potent inhibitor-like molecules targeting Gram-negative multidrug efflux pumps.
- To leverage supervised classification models for virtual screening of potential inhibitors.
Main Methods:
- Developed supervised classification models (SVM, Random Forest, AdaBoost, LightGBM) using molecular descriptors of known inhibitors.
- Screened a library of 5043 compounds, filtering to identify 8 hit molecules.
- Performed molecular dynamics simulations and binding free-energy calculations to assess inhibitor activity against AcrB and MexB.
Main Results:
- Four classification models achieved >80% accuracy in predicting inhibitor potency.
- Mol.3488 exhibited high binding affinity to both AcrB and MexB efflux pumps.
- Specific molecular moieties (pyridopyrimidone, benzothiazole) and ligand orientation enhanced binding affinity.
Conclusions:
- Machine learning-based virtual screening is effective for discovering efflux pump inhibitors.
- Mol.3488 represents a promising lead compound for combating Gram-negative bacterial infections.
- Understanding structure-activity relationships can guide the design of more potent inhibitors.
Related Concept Videos
01:06Carrier-Mediated Transport
Active transport involves two types of membrane-spanning transporters: uptake and efflux. Uptake transporters are expressed in the small...
01:26Drug Discovery: Overview