Oxidative stress increases susceptibility of Mycobacterium tuberculosis to isoniazid

Abstract

Isoniazid is a first-line antibiotic used in the treatment of infections caused by Mycobacterium tuberculosis. Isoniazid is a prodrug requiring oxidative activation by the catalase-peroxidase hemoprotein, KatG. Resistance to isoniazid can be obtained by point mutations in the katG gene, with one of the most common being a threonine-for-serine substitution at position 315 (S315T). The S315T mutation is found in more than 50% of isoniazid-resistant clinical isolates and results in an approximately 200-fold increase in the MIC of isoniazid compared to that for M. tuberculosis H37Rv. In the present study we investigated the hypothesis that superoxide plays a role in KatG-mediated isoniazid activation. Plumbagin and clofazimine, compounds capable of generating superoxide anion, resulted in a lower MIC of isoniazid for M. tuberculosis H37Rv and a strain carrying the S315T mutation. These agents did not cause as great of an increase in isoniazid susceptibility in the mutant strain when the susceptibilities were assessed by using the inhibitory concentration that causes a 50% decrease in growth. These results provide evidence that superoxide can play a role in isoniazid activation. Since clofazimine alone has antitubercular activity, the observation of synergism between clofazimine and isoniazid raises the interesting possibility of using both drugs in combination to treat M. tuberculosis infections.

FIG. 1.

INH susceptibilities of M. tuberculosis (Mtb) strains H37Rv (A) and TBC3 (B) using BACTEC radiometric analysis. The control reaction represents that in a vial containing a starter culture diluted 1:100 relative to the inoculum in the vials containing INH. The asterisks indicate the MICs of INH for each strain, as defined in Materials and Methods. GI data were obtained on a daily basis following inoculation of BACTEC vials, but data for only selective days are shown for clarity.

FIG. 2.

Susceptibilities of M. tuberculosis (Mtb) H37Rv (A) and TBC3 (B) to plumbagin and clofazimine.

FIG. 3.

Effects of plumbagin and clofazimine on INH susceptibility. The percent GI is plotted versus the log[INH] (p[INH]), expressed in units of nanograms per milliliter, alone (top) or in the presence of plumbagin (middle) or clofazimine (bottom), for M. tuberculosis H37Rv (A) and TBC3 (B). Weakly inhibitory concentration of clofazimine (0.1 μg/ml) and plumbagin (20 μM in panels A, 10 μM in panels B) were included in each vial. IC₅₀s were calculated for each graph as the concentration of INH resulting in a 50% decrease in the GI value relative to that for the control vials, as described in Materials and Methods. The vertical dotted line in each set of panels represents the IC₅₀ of INH alone (0.015 μg/ml in panels A, 2.1 μg/ml in panels B). (C) Reduction of IC₅₀ of INH for H37Rv and TBC3 by plumbagin and clofazimine. The percent GI is expressed as the average value ± the standard error for three experiments.

FIG. 3.

Effects of plumbagin and clofazimine on INH susceptibility. The percent GI is plotted versus the log[INH] (p[INH]), expressed in units of nanograms per milliliter, alone (top) or in the presence of plumbagin (middle) or clofazimine (bottom), for M. tuberculosis H37Rv (A) and TBC3 (B). Weakly inhibitory concentration of clofazimine (0.1 μg/ml) and plumbagin (20 μM in panels A, 10 μM in panels B) were included in each vial. IC₅₀s were calculated for each graph as the concentration of INH resulting in a 50% decrease in the GI value relative to that for the control vials, as described in Materials and Methods. The vertical dotted line in each set of panels represents the IC₅₀ of INH alone (0.015 μg/ml in panels A, 2.1 μg/ml in panels B). (C) Reduction of IC₅₀ of INH for H37Rv and TBC3 by plumbagin and clofazimine. The percent GI is expressed as the average value ± the standard error for three experiments.

FIG. 4.

The INH shunt, a proposed mechanism for INH oxidation by ferric superoxo-KatG.