Characterisation of Potential Antimicrobial Targets for Tuberculosis. 2. Branched-Chain Amino Acid Aminotransferase and Methionine Regeneration in Mycobacterium Tuberculosis

Tuberculosis remains an important problem for the Canadian Forces in many of its overseas deployments. With the spread of drug-resistant strains of Mycohactenum tubercuThsis, there is an increased need to characterise novel drug targets in the organism. The final step of methionine recycling from methylthioadenosine has been examined in M tuberculosis, and has been found to be catalysed by a branched-chain amino acid aminotransferase. The enzyme was found to be a member of the aminotransferase lila subfamily, and closely related to the corresponding aminotransferase in Bacillus subtilis, but not to that found in B. anthracis or B. cereus (Berger et al., Journal of Bacteriology, 185, p. 2418-2431, 2003). The amino donor preference for the formation of methionine from ketomethiobutyrate was isoleucine, leucine, valine, glutamate, and phenylalanine. The en%me catalysed branched-chain amino acid and ketomethiobutyrate transamination with a Km of 1.77 - 7.44 mM and a Vmax of 2.17 - 5.70 %mol/min/mg protein, and transamination of ketoglutarate with a Km of 5.79 - 6.95 mM and a Vmax of 11.82 - 14.35 %mol/min/mg protein. Aminooxy compounds were examined as potential enzyme inhibitors, with O-benzylhydroxylamine, o-t- butylhydroxylamine, carboxymethoxylamine, and 0-allylhydroxylamine yielding mixed-type inhibition with Ki values of 8.20 - 21.61 %M. These same compounds were examined as antimycobacterial agents in a M inarinum model and were found to completely prevent cell growth. 0-allylhydroxylamine was the most effective growth inhibitor with an MIC of 78 tjM and an 1C50 of 8.49 %M.