Discovery of potent anti-tuberculosis agents targeting aminoacyl-tRNA synthetases.

Background. The most serious challenge in the treatment of tuberculosis is the multidrug resistance of Mycobacterium tuberculosis to available antibiotics. This provides the stimulus for the search of novel molecular targets and approaches to overcome resistance emergence. Nowadays, M. tuberculosis leucyl-tRNA synthetase (LeuRS) is a clinically validated molecular target for antituberculosis drug discovery. The aim of our study was to develop small-molecular inhibitors of M. tuberculosis LeuRS with antibacterial activity. Methods. Molecular docking, pharmacophore screening and machine learning approaches were used for in silico design of M. tuberculosis LeuRS inhibitors. Molecular docking was performed with DOCK 4.0 software. Ligand-oriented pharmacophore models were built using web-server PharmaGist. Pharmacophore screening was conducted using PharmDeveloper program. The artificial neural networks were built using module net from R 3.6.1. The inhibitory activity of compounds toward recombinant aminoacyl-tRNA synthetase was investigated in aminoacylation assays. The antibacterial activity of compounds was studied toward pathogenic strain M. tuberculosis H37Rv in microdilution assay. Results. Using molecular docking approach we have found inhibitors of M. tuberculosis LeuRS among N-Benzylidene-N’-thiazol2-yl-hydrazines with IC50 values in the range from 2.27 mM to 19.3 mM. The most promising compound among studied derivatives inhibits growth of pathogenic bacteria M. tuberculosis H37Rv with IC50 = 10 mM. Then, as a strategy to overcome resistance emergence in M. tuberculosis we used the idea of multitarget inhibition, since the probability of mutations simultaneously in both enzymes is negligible. As molecular targets we have selected M. tuberculosis LeuRS and methionyl-tRNA synthetase (MetRS). Using molecular docking approach we identified dual-targeted inhibitors among N-benzylidene-N’-thiazol2-yl-hydrazines. The most promising compound among studied derivatives inhibits growth of M. tuberculosis H37Rv with IC50 = 1.56 mM. Applying a ligand-based pharmacophore approach we found five novel dual-targeted hit compounds from different chemical classes. The most active compound inhibits LeuRS and MetRS with IC50 values of 13 mM and 13.8 mM, correspondingly. Using the single-hidden-layer neural network models we identified dual-targeted inhibitors among 2-(quinolin2-ylsulfanyl)-acetamides. The most active compound among studied derivatives inhibits MetRS and LeuRS with IC50 values of 33 mM and 23.9 mM, respectively. Conclusions. Using in silico approaches we identified several chemical classes of inhibitors targeting M. tuberculosis LeuRS and MetRS with IC50 values in micromolar concentration range. It was found that the most promising compounds which possess antibacterial activity belong to N-benzylidene-N’-thiazol-2-yl-hydrazine derivatives. Therefore, the compounds in this class can be valuable for further biological research and optimization. [ABSTRACT FROM AUTHOR]