Nitric Oxide Generated from Isoniazid Activation by KatG: Source of Nitric Oxide and Activity against Mycobacterium tuberculosis

Abstract

Isonicotinic acid hydrazide (INH) is a frontline antituberculosis agent. Once taken up by Mycobacterium tuberculosis, INH requires activation by the catalase-peroxidase KatG, converting INH from its prodrug form into a range of bactericidal reactive species. Here we used ¹⁵N-labeled INH together with electron paramagnetic resonance spin trapping techniques to demonstrate that nitric oxide (NȮ) is generated from oxidation at the hydrazide nitrogens during the activation of INH by M. tuberculosis KatG. We also observed that a specific scavenger of NȮ provided protection against the antimycobacterial activity of INH in bacterial culture. No significant increases in mycobacterial protein nitration were detected, suggesting that NȮ and not peroxynitrite, a nitrating metabolite of NO^·, is involved in antimycobacterial action. In conclusion, INH-derived NO^· has biological activity, which directly contributes to the antimycobacterial action of INH.

Mycobacterium tuberculosis infections cause 2 million deaths every year and latently persist in over 1 billion individuals worldwide (33). Isoniazid (isonicotinic acid hydrazide [INH]) is a frontline antituberculosis agent that is prescribed daily in millions of doses worldwide. While its mechanism(s) of action and cellular targets continue to be uncovered (16, 20), its full range of effects on mycobacterial cells still remains to be resolved (14, 27). The tubercle bacillus is exceptionally sensitive to INH (7, 36), a prodrug, which is peroxidatively activated by the M. tuberculosis catalase-peroxidase KatG to produce damaging species within the bacteria.

In keeping with the critical role of INH activation via KatG, the main site for mutations associated with resistance to INH is the katG gene (37). Several INH-derived intermediates generated during INH activation, such as isonicotinic acyl NADH (20) and mycobacterial targets including enzymes from the mycobacterial type II fatty acid synthase system (1, 16), have been identified. Other hypotheses regarding INH activation have focused upon INH-derived free radicals as important antimycobacterial intermediates (12, 24, 31). Despite this progress, the exact mechanism(s) of INH action that underlies its exceptional and specific potency against M. tuberculosis remains to be fully delineated, as multiple targets and pathways have been considered (7, 14, 17, 26).

NȮ and other reactive nitrogen species have previously been demonstrated to have appreciable activity against M. tuberculosis (5, 15, 35). The immune response-derived NȮ is considered to contribute to defenses against mycobacterial infection (22). In this study, we were prompted by reports indicating tyrosine nitration during oxidation of INH (29), NȮ formation from hydroxyurea in vivo (11), and NȮ formation during horseradish peroxidase-catalyzed oxidation of hydroxyurea in vitro (10) to examine whether NȮ is generated during INH activation by KatG. Based on partial molecular similarities between INH and hydroxyurea, we hypothesized that pathways might exist for NȮ production from INH during its oxidation catalyzed by KatG.

Here we describe the previously unappreciated generation of nitric oxide during activation of INH from its prodrug form into reactive intermediates by the M. tuberculosis catalase-peroxidase KatG. We also present in vivo analysis indicating that NȮ production during INH activation may potentially contribute to the antimycobacterial action of INH.

MATERIALS AND METHODS

In vitro spin trapping of INH-derived NO.

A sensitive and NȮ-specific electron paramagnetic resonance (EPR) spin trapping technique was used (13). A 10 mM Fe^II (N-methyl-d-glucamine dithiocarbamate)₂ complex was incubated with 0.471 mg of purified M. tuberculosis H37Rv KatG (30) ml⁻¹ with 35 mM INH and 10 mM H₂O₂ in 10 mM phosphate buffer, pH 7, at 37°C for 5 min. [¹⁵N₂, ¹⁵N₃]INH was synthesized by using the method of Todorovic et al. (28), and purity was confirmed by thin-layer chromatography and ¹⁵N nuclear magnetic resonance. N-methyl-d-glucamine dithiocarbamate was synthesized by the method of Shinobu et al. (23). EPR spectrometry of these incubations was performed by using a Bruker Elexsys series spectrometer operating at X-band frequencies, at 25°C with samples held in 20-μl capillaries. Recombinant KatG was expressed in Escherichia coli and purified by chromatography on DEAE Sepharose CL-6B, Sephacryl S300-HR, and MonoQ HR5/5 columns as previously described (32).

Treatment of bacterial cultures with INH and NȮ scavenger.

Exponentially growing cultures in aerobic roller bottles containing M. tuberculosis var. bovis BCG were exposed to 2 mM 2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (CPTIO; a specific NȮ scavenger) (Alexis Inc., Carlsbad, Calif.) and/or 3.7 μM INH for 7 days at 37°C. CPTIO alone had a minor effect, reducing viability by from 100% ± 6.3% to 68% ± 15.5% (number of cultures, 3; P < 0.05). CPTIO showed no interactions with INH as tested by EPR: the reaction of INH with the nitroxide functional group of CPTIO would cause a loss of EPR signal, and a reaction with the nitrone functional group of CPTIO would cause an alteration in the hyperfine coupling pattern. Bacterial cultures were serially diluted and plated on 7H11 plates for CFU determination.

Nitrotyrosine levels upon exposure to INH.

Exponentially growing cultures of M. tuberculosis var. bovis BCG were treated overnight with 73 μM INH with or without 1 mM plumbagin (Sigma, St. Louis, Mo.) at 37°C. Cell extracts were obtained by bead beating with 0.1-mm zirconia beads (two 30-s cycles) in a Mini-Beadbeater (Biospec Products Inc., Bartsville, Okla.). Cell extracts were assayed for protein by the Pierce BCA kit (Pierce, Rockford, Ill.) and for nitrotyrosine by using the Hycult Biotechnology Hbt nitrotyrosine enzyme-linked immunosorbent assay (ELISA) kit (Uden, The Netherlands).

RESULTS AND DISCUSSION

NȮ production from KatG activation of INH.

We tested whether NȮ was produced during INH activation in an in vitro system consisting of purified components. One advantage of using purified and defined components is the elimination of complications associated with complex mixtures that may contain NADH and other chemical species known to complex with INH in the presence of NAD-binding proteins (20). NȮ production was detected during KatG-mediated oxidation of INH in a reaction mixture comprised of purified and previously characterized M. tuberculosis H37Rv KatG (31), INH, and H₂O₂ (32) (Fig. 1a, spectrum i). This mixture, consisting of purified components, was identical to the previously published INH activation system (31), with the single modification of substituting the previously used catalase-insensitive t-butyl hydroperoxide (31) with H₂O₂ at concentrations bypassing the intrinsic catalase activity of KatG. The identification of NȮ was based on the ¹⁴N hyperfine coupling (1.25 mT) and g value (2.04) and by comparison with an authentic NȮ standard (not shown).

FIG. 1.

Nitric oxide is generated from oxidation hydrazide nitrogen atoms during INH activation by KatG. (a) EPR spectra of NȮ derived from KatG activation of INH. NȮ was spin trapped with 10 mM Fe^II (N-methyl-d-glucamine dithiocarbamate)₂ complex after incubation of 0.471 mg of KatG ml⁻¹ with 35 mM INH and 10 mM H₂O₂ in 10 mM phosphate buffer, pH 7, at 37°C for 5 min. Spectrum i is all [¹⁴N]INH; spectrum ii is hydrazide-labeled [¹⁵N₂, ¹⁵N₃]INH. The EPR spectrometer settings were as follows: microwave power, 10 mW; modulation, 0.2 mT at 100 kHz; x-axis resolution, 1,024 points; conversion time, 82 ms; time constant, 164 ms; sweep, 8 mT; number of scans, 45. (b) Dependence of NȮ production from INH on KatG and H₂O₂. Conditions are the same as in panel a except that the concentration of INH is 10 mM. Data are the results of four experiments.

Conclusive confirmation that NȮ detected during oxidation with KatG and H₂O₂ is derived from INH was obtained by synthesis of INH containing the hydrazide group that was the doubly ¹⁵N-labeled [¹⁵N₂, ¹⁵N₃]INH. Using ¹⁵N-labeled INH in the KatG activation system resulted in a characteristic transformation of the ¹⁴NȮ triplet spectrum (Fig. 1a) to a ¹⁵NȮ doublet EPR spectrum (Fig. 1a, spectrum ii), with hyperfine coupling of 1.78 mT, resulting from the different nuclear spins and gyromagnetic ratios of ¹⁴N and ¹⁵N. The generation of NȮ in this in vitro system was consistently reproducible and showed absolute requirements for all components of the system: enzyme (KatG), substrate (INH), and oxidant (H₂O₂) (Fig. 1b). No signals were observed in the absence of the spin trap Fe(N-methyl-d-glucamine dithiocarbamate)₂.

Biological activity of NȮ generated during INH-derived NȮ activation.

We next tested whether exposure of M. tuberculosis to NȮ generated during the KatG oxidation of INH is potentially toxic to the bacteria. As a control, cultures of M. tuberculosis var. bovis BCG were treated with an NȮ donor,2,2-(hydroxynitrosohydrazono)-bis-ethanamine (DETANONOate). Treatment with 1 mM DETA NONOate resulted in a 70% decrease in bacterial survival (Fig. 2a), in accord with the known potent action of NȮ against M. tuberculosis (5, 15, 35). Next, M. tuberculosis was treated with INH in the presence or absence of the NȮ scavenger CPTIO. CPTIO was chosen as it is a widely used and specific scavenger of NȮ (19). Although hemoprotein NȮ scavengers such as myoglobin and hemoglobin are effective, they were not used to avoid complications due to the addition of iron to the medium and also because they are capable of oxidatively activating INH themselves to form radicals (25). The addition of CPTIO to INH-treated cultures inhibited the killing of aerobically growing M. tuberculosis var. bovis BCG (Fig. 2b). CPTIO increased bacterial survival, relative to INH treatment in the absence of CPTIO, by 30-fold from 0.53 to 17.5% (Fig. 2b). A direct reaction between INH and CPTIO was excluded (assayed by EPR of the nitroxide CPTIO), confirming that the action of CPTIO was mediated by the scavenging of INH-derived NȮ and not by a fortuitous sequestering or reaction with INH.

FIG. 2.

Antimycobacterial action of INH-derived nitric oxide. (a) Survival data for M. tuberculosis var. bovis BCG upon exposure to 1 mM DETA-NONOate. Exponentially growing cultures in aerobic roller bottle culture of M. tuberculosis var. bovis BCG were exposed to 1 mM DETA-NONOate for 7 days at 37°C. Data are normalized to the respective day 0 values and represent triplicate experiments (P = 0.026). (b) The NȮ scavenger CPTIO protects against the antimycobacterial action of INH. Exponentially growing cultures in aerobic roller bottle cultures of M. tuberculosis var. bovis BCG were exposed to 2 mM CPTIO and/or 3.7 μM INH for 7 days at 37°C. CPTIO alone had no effect on viability (data not shown). Cultures were serially diluted and plated on 7H11 plates for CFU determination. Values are normalized to CFU values of untreated controls at day 0 and represent triplicate experiments (P = 0.0009). The actual survival rate in culture with CPTIO was 17.5% compared to day 0 controls. (c) Nitrotyrosine levels upon exposure to INH. Exponentially growing cultures of M. tuberculosis var. bovis BCG (as in panel a) were treated overnight with 73 μM INH with or without 1 mM plumbagin at 37°C. Cell extracts (obtained by bead beating with 0.1-mm ZrSi beads) were assayed for nitrotyrosine by using the Hycult Biotechnology Hbt nitrotyrosine ELISA. Data represent the means of three replicate cultures. No levels of nitrotyrosine were significantly different from others (P < 0.05).

Analysis of biological indications for peroxynitrite formation.

NȮ is often thought to exert antimicrobial activity through its reaction with superoxide (O₂^·−) to form peroxynitrite, a reactive species capable of oxidizing and nitrating biomolecules (21). We examined the potential evidence for peroxynitrite-mediated killing by determining the levels of nitrotyrosine, a widely used marker for peroxynitrite formation and damage (2), in the proteins extracted from INH-treated M. tuberculosis var. bovis BCG. However, ELISAs of nitrotyrosine in protein extracts from INH-treated M. tuberculosis (Fig. 2c) showed no increase above background levels. There was no increase in nitrotyrosine even in the presence of plumbagin, a known redox-cycling agent stimulating O₂^·− production (4), which is known to potentiate INH action (3) and is thus expected to increase peroxynitrite formation (which is formed by reaction of O₂^·− and NȮ). Western blot analysis (data not shown) also showed no increases in nitrotyrosine levels above the background in M. tuberculosis var. bovis BCG protein extracts upon INH treatment.

Conclusions

The studies presented here with ¹⁵N-labeled INH unequivocally demonstrate that NȮ is generated from the INH hydrazide group during activation by M. tuberculosis KatG. A potential pathway of NȮ formation is shown in Fig. 3, based upon the known importance of the hydrogen at the hydrazide N₂ (i.e., a lack of substitution by alkyl or other groups) (8) and the requirement for oxygen for maximal INH activity (34). However, the reaction pathway after the N-O bond formation from the reaction of the nitrogen-centered radical with oxygen (9) is as yet unclear. Our bacterial survival assays suggest that the NȮ generated in vivo during oxidation of INH may have appreciable activity. Although the NȮ scavenger CPTIO caused substantial protection of mycobacteria against INH action, it is evident from the experimental data that it did not fully protect M. tuberculosis var. bovis BCG against INH. This is consistent with the presence of other known antimycobacterial products of INH, such as isonicotinic acyl-NADH (20). Alternatively, the incomplete protection by CPTIO can be explained as resulting from any of the following: (i) incomplete scavenging of NȮ, (ii) antimycobacterial activity of one of the products of NȮ scavenging by CPTIO, (iii) the activity of additional INH-derived radical species (31), and/or (iv) the slight effects of CPTIO on growth itself. It should be acknowledged that it is unlikely that NȮ is the only product of INH contributing to its overall antituberculosis action, as other studies have shown that INH-NAD adducts are another class of major toxic products responsible for the antimycobacterial action of INH (1, 20). Nevertheless, the unambiguous generation of NȮ from INH and its detectable effects on mycobacteria suggest the possibility of enhancing this property of INH as a potential strategy for generating new antituberculosis drugs. Such ideas already have preliminary validation, as it has recently been shown that the addition of NȮ-releasing groups to ciprofloxacin greatly increases its activity against M. tuberculosis (6).

FIG. 3.

Postulated pathway of NȮ production. Oxidation of INH (1) at N₂ (18, 25) to form the hydrazyl radical (2) is followed by oxygen addition to the hydrazyl radical (causing N-O bond formation) (3), followed by as yet undelineated fragmentation or elimination.