Tuberculosis is an infectious air-borne disease and one of the leading causes of death globally among all infectious diseases. There is an urgent need to develop antitubercular drugs that would be highly efficient and less toxic than the presently available marketed drugs. Mycobacterium membrane protein large 3 (MmpL3) is an emerging drug target in tuberculosis with various classes of molecules that have been known to inhibit it. In this study, a dataset of indole-2-carboxamides showing antitubercular activity by inhibiting MmpL3 was utilized. Initially, a chimera-based homology model was developed and docking was performed with the filtered dataset to analyse the interactions. Thereafter, molecular dynamics simulations were run with representative molecules to gain a better insight on the binding patterns. To attain a more quantitative correlation, an atom-based 3D QSAR model was developed which complemented the results from the previous models. A library of novel indole-2-carboxamides was then generated using core hopping-based ligand enumeration and upon screening on our workflow model it predicted three molecules as potent antitubercular compounds. This work not only helps to gain new insights on the interactions at the MmpL3 binding site but also provides novel indole-2-carboxamides having the potential to become antitubercular drugs in future.
Keywords: MmpL3; atom-based 3D QSAR; homology modelling; molecular docking; molecular dynamics; tuberculosis.