Discovery of inhibitors against mycobacterium branched-chain amino acid aminotransferases through in silico screening and experimental evaluation.

Tuberculosis (TB) is one of the most dangerous infectious diseases and is caused by Mycobacterium bovis (Mb) and Mycobacterium tuberculosis (Mt). Branched-chain amino acid aminotransferases (BCATs) were reported to be the key enzyme for methionine synthesis in Mycobacterium. Blocking the methionine synthesis in Mycobacterium can inhibit the growth of Mycobacterium. Therefore, in silico screening of inhibitors can be a good way to develop a potential drug for treating TB. A pyridoxal 5'-phosphate (PLP)-form of Mycobacterium bovis branched-chain amino acid aminotransferases (MbBCAT), an active form of MbBCAT, was constructed manually for docking approximately 150 000 compounds and the free energy was calculated in Autodock Vina. The 10 compounds which had the highest affinity to MbBCAT were further evaluated for their inhibitory effects against MbBCAT. Within the selected compounds, compound 4 (ZINC12359007) was found to be the best inhibitor against MbBCAT with the inhibitory constant K i of 0·45 μmol l -1 and IC 50 of 2·37 μmol l -1 . Our work provides potential candidates to develop effective drugs to prevent TB since the well-known structural information would be beneficial in the structure-based modification and design.