Identification of potent indolizine derivatives against Mycobacterial tuberculosis: In vitro anti-TB properties, in silico target validation, molecular docking and dynamics studies

In the current study, two sets of compounds: (E)-1-(2-(4-substitutedphenyl)-2-oxoethyl)-4-((hydroxyimino)methyl)pyridinium derivatives (3a-3e); and (E)-3-(substitutedbenzoyl)-7-((hydroxyimino)methyl)-2-substitutedindolizine-1-carboxylate derivatives (5a-5j), were synthesized and biologically evaluated against two strains of Mycobacterial tuberculosis (ATCC 25177) and multi-drug resistant (MDR) strains. Further, they were also tested in vitro against the mycobacterial InhA enzyme. The in vitro results showed excellent inhibitory activities against both MTB strains and compounds 5a-5j were found to be more potent, and their MIC values ranged from 5 to 16 μg/mL and 16-64 μg/mL against the M. tuberculosis (ATCC 25177) and MDR-TB strains, respectively. Compound 5h with phenyl and 4-fluorobenzoyl groups attached to the 2- and 3-position of the indolizine core was found to be the most active against both strains with MIC values of 5 μg/mL and 16 μg/mL, respectively. On the other hand, the two sets of compounds showed weak to moderate inhibition of InhA enzyme activity that ranged from 5 to 17 % and 10-52 %, respectively, with compound 5f containing 4-fluoro benzoyl group attached to the 3-position of the indolizine core being the most active (52 % inhibition of InhA). Unfortunately, there was no clear correlation between the InhA inhibitory activity and MIC values of the tested compounds, indicating the probability that they might have different modes of action other than InhA inhibition. Therefore, a computational investigation was conducted by employing molecular docking to identify their putative drug target(s) and, consequently, understand their mechanism of action. A panel of 20 essential mycobacterial enzymes was investigated, of which β-ketoacyl acyl carrier protein synthase I (KasA) and pyridoxal-5'-phosphate (PLP)-dependent aminotransferase (BioA) enzymes were revealed as putative targets for compounds 3a-3e and 5a-5j, respectively. Moreover, in silico ADMET predictions showed adequate properties for these compounds, making them promising leads worthy of further optimization.