Tubercle and Lung Disease (1999) 79(4), 191–198 © 1999 Harcourt Publishers Ltd Article no. tuld.1998.0203
Toxicity of nitrogen oxides and related oxidants on mycobacteria: M. tuberculosis is resistant to peroxynitrite anion K. Yu,* C. Mitchell,*,† Y. Xing,* R. S. Magliozzo,‡ B. R. Bloom,† J. Chan* *Departments of Medicine and Microbiology & Immunology, † Albert Einstein College of Medicine, Bronx, NY ‡ Department of Chemistry, Brooklyn College, CUNY, Brooklyn NY, USA
Summary Objective: To test the toxicity of reactive nitrogen intermediates (RNI), including authentic nitric oxide (NO), nitrogen dioxide (NO2), and peroxynitrite anion (ONOO–), a potent oxidant derived from NO and superoxide anion, on various mycobacterial strains including M. tuberculosis. Design: Relatively avirulent mycobacteria including M. smegmatis and BCG, as well as the pathogenic M. Bovis Ravenel and M. tuberculosis Erdman and the clinical isolate M160 (also known as the C strain) were tested for their susceptibility to the toxic effects of NO, NO2, and ONOO–. Deaerated, NO-saturated solutions as well as an anaerobic in vitro system in which mycobacteria can be exposed to desired concentrations of authentic NO or NO2, were employed in these studies. An in vitro ONOO– killing assay was used to examine the adverse effects of this NO-derived oxidant on the various strains of mycobacteria. Results: Both NO and NO2 exhibit antimycobacterial activity, with the former being more potent. Results obtained using ONOO– killing assay revealed that while avirulent mycobacteria including BCG and M. smegmatis are susceptible to this NO-derived oxidant, the virulent Erdman strain of M. tuberculosis and M. bovis, as well as the clinical tuberculous isolate M160, are remarkably resistant. Conclusion: These results suggest that the interactions between RNI and various species of mycobactiera could be highly specific. And since activated macrophages produce peroxynitrite, the significance of the ONOO– resistance of M. tuberculosis strains in relation to intracellular survival deserves further investigation. © 1999 Harcourt Publishers Ltd INTRODUCTION Activated murine macrophages generate a variety of oxygen- and nitrogen-derived radicals that possess antimicrobial function.1–7 The enzyme complex of NADPH oxidase produces reactive oxygen intermediates (ROI) including hydrogen peroxide (H2O2), superoxide anion (O2–), and hydroxyl radicals (OH·) by the oxidative burst via sequential reduction of molecular oxygen.1–4 Macrophage inducible nitric oxide synthase (NOS2) generates nitric oxide (NO), the primary reactive nitrogen inter-
Correspondence to: John Chan, Departments of Medicine and Microbiology & Immunology, Montefiore Medical Center, Albert Einstein College of Medicine, 111 East 21th Street, Bronx, New York 10467; USA. Tel.: 718-9207247/2715; Fax: 718-652-0536; E-mail:
[email protected] Received: 9 October 1998; Revised: 6 November 1998; Accepted: 11 November 1998
mediate (RNI), from which other toxic nitrogen oxides, such as nitrogen dioxide (NO2), derive.5–7 These oxygen and nitrogen radical-based cytotoxic pathways of macrophages can be triggered by a variety of biological agents8 including interferon-gamma (IFN-γ) and tumor necrosis factor-alpha (TNF-α). We and others have established that RNI are effective antimycobacterial agents.9–15 Results obtained from experimental murine tuberculosis models using NOS inhibitors,12,14,15 as well as mice with disruption of the IFN-γ,11 p55 TNF-α receptor13 and NOS2 genes14 have provided compelling evidence that RNI play a critical role in host defense against murine tuberculosis, both during the acute and persistent/latent phase of infection.14,15 Despite their established role in resistance to M. tuberculosis, the biochemical and molecular basis for the antimycobacterial function of RNI is not well defined. Recently, the potent oxidant peroxynitrite anion (ONOO–), a derivative of NO and O2–, has been shown to be bacterio191
192
Yu, Mitchell, Xing et al.
cidal.16,17 The effect of ONOO– on Mycobacterium spp. is, however, unknown. This study is designed to examine the susceptibility of mycobacteria to various RNI, specifically, NO, NO2, and ONOO–. MATERIALS AND METHODS Chemicals and reagents All reagents were purchased from Sigma Chemical (St. Louis) unless otherwize indicated. Authentic NO and oxygen free nitrogen (containing < 5 p.p.m. O2) were obtained from County Welding (White Plains, New York). Bacteria M. smegmatis mc2155, BCG (Pasteur), the virulent M. bovis strain Ravenel (ATCC 35720), M. tuberculosis strains Erdman (provided by Dr Frank Collins, FDA) and M160, the most prevalent New York City clinical isolate (known also as the C strain),18 were cultured in Middlebrook 7H9 broth (Difco Laboratories, Detroit, MI) supplemented with oleic acid-albumin-dextrose complex (OADC) (Difco Laboratories). Passaging of M. smegmatis, BCG, and the Erdman strain of M. tuberculosis has been described previously.19 M. bovis Ravenel and the clinical isolate M160 were cultured as described for M. tuberculosis Erdman. E. coli Y1090 was cultured as described.20 Organisms in the log phase of growth were used in all experiments. Studies involving virulent mycobacterial strains were performed inside a Class II biohazard laminar flow hood in a biosafety level 3 facility. Peroxynitrite anion Peroxynitrite anion was either purchased (Alexis Biochemicals, San Diego, CA) or synthesized using a quenched-flow reactor built from an acrylic block as described.21–13 Briefly, 0.6M sodium nitrite (NaNO2) was allowed to react with 0.7M H2O2/0.6M hydrochloric acid (HCl) in the quench-flow reactor, and the product quenched with 1.5M sodium hydroxide (NaOH). All solutions were freshly prepared and ice-cold. Peristaltic pumps were used to deliver the three solutions at a flow rate of 26 ml/min. The product, collected on ice in aliquots, was stored frozen at -80°C. Excess H2O2 was removed by a manganese oxide column before use.22,23 The concentration of ONOO– was measured by determining absorbance at 302 nm in 1M NaOH as described.22,23 Potential background absorbance due to chemical species other than ONOO– was determined by measuring A302 of an equal volume of peroxynitrite anion that had been allowed to degrade in 100 mM potassium phosphate (pH 7.4). This potent oxidant disintegrates rapidly Tubercle and Lung Disease (1999) 79(4), 191–198
with a half-life ~1 s in phosphate buffer (pH 7.4) at 37°C.16,22 Treatment of BCG with NO in deaerated phosphatebuffered saline (PBS; pH 7.4) Nitric oxide solution was prepared by saturating deaerated PBS with authentic NO as described.24,25 O2-free N2 was bubbled through PBS in serum cap-sealed tubes for deaeration. The deaerated PBS was then saturated with authentic NO. Aliquots of log-phase BCG were placed in serum cap-sealed microfuge tubes. The bacilli were pelleted by centrifugation. Microfuge tubes containing bacterial pellets were then purged with O2-free N2. Bacteria (106–107 CFU) were resuspended in 1 ml of NOsaturated, deaerated PBS introduced through the serum cap using a gas tight syringe. Control groups were treated with deaerated PBS. At appropriate intervals after treatment (0 h, 6 h, 12 h, and 24 h), cells were pelleted and resuspended in 7H9 broth supplemented with OADC after removal of the supernatant using a gas-tight syringe. The antimycobacterial effect of deaerated, NO-saturated PBS was assessed by metabolic labeling using [3H]uracil and by direct quantitation of viable CFU as previously described.10 Treatment of mycobacteria with NO and NO2: The Calypso We have developed a system (referred to as Calypso) by which M. tuberculosis can be exposed to desired amounts of authentic NO or NO2. The system consists of an airtight, stainless-steel chamber connected to a pressure sensor (sensitive to 0.1 Torr), a supply of NO gas and O2free nitrogen (< 5 p.p.m. O2), and a vacuum pump (10–4 Torr). This system is able to hold a pressure of 1 Torr for > 20 h. Mycobacteria in the log phase of growth were plated onto a nylon membrane (Fisher Scientific, Springfield, NJ) and then placed aseptically inside the presterilized air-tight chamber. The air inside the chamber was evacuated via the vacuum pump, and refilled with O2-free N2. This gas exchange was repeated twice before the desired amount of NO, cleared of any contaminating NO2 by percolating through an air-tight container of potassium hydroxide pellets, is allowed to enter the chamber pre-filled with an appropriate volume of N2. The desired concentration of NO2 was achieved by injecting an appropriate volume of air into the chamber containing a predetermined amount of NO mixed with N2, via a port equipped with a 0.2 micron filter. A built-in fan at the bottom of the chamber assures even mixing of gases. At the end of an appropriate NO or NO2 treatment period, the nylon membrane was removed after all NO/NO2 had been flushed out by N2, and placed directly onto a 7H10 © 1999 Harcourt Publishers Ltd
Toxicity of nitrogen oxides and related oxidants on mycobacteria
agar plate to monitor growth. In control experiments, mycobacteria were similarly manipulated, except that during the treatment period, the bacilli were exposed only to N2. Using this system, the relative toxic effect of NO and NO2 on mycobacteria can be characterized. Treatment of mycobacteria with peroxynitrite anion: The ONOO– killing system The ONOO– killing model, designed to evaluate the antimycobacterial effects of ONOO–, is distinct from the Calypso system described above: the procedure involved in the peroxynitrite killing system is carried out in ambient air. Mycobacteria, in microfuge tubes, were subjected to treatment with various doses of ONOO– in potassium phosphate buffer (50 mM; pH 7.4) consecutively for 7–10 cycles. The duration of each treatment was 3 min. Prior to use, excess H2O2, one of the ingredients used in the synthesis of ONOO–, was removed by treatment with manganese dioxide as described.22,23 Between treatments, mycobacteria were pelleted by centrifugation, and then resuspended in fresh ONOO– solution. Tubes were placed in a 37°C heating block while not manipulated. Control groups underwent the same procedure except that they were exposed to ONOO– solution that had been allowed to disintegrate for 3 min. Viability of mycobacteria after treatment was assessed by plating serial dilutions of bacilli resuspended in saline on 7H10 plates. RESULTS The effects of NO-saturated PBS on BCG To test whether authentic NO has the ability to kill mycobacteria, BCG was treated with NO-saturated, deaerated PBS. Assuming saturation, the concentration of NO in the deaerated PBS used is approximately 1 mM.24–26 The results of these experiments indicate that (Table 1) this nitrogen radical possesses antimycobacterial activity. Compared to controls exposed to deaerated buffer alone, a 6-h treatment with NO-saturated PBS inhibited BCG incorporation of [3H]uracil by almost 100% (Table 1). Assessment of the viability of the treated bacilli revealed Table 1 Effects of NO-saturated, deaerated PBS on BCG Treatment CPM ± SD§ time –NO† +NO‡ 6h 12h 24h
67425 ± 30783 1556 ± 505 104920 ± 35559 1003 ± 320 43617 ± 17827 1198 ± 766
†
% Suppression of [3H] uracil % incorporation Viability 97.7 99.0 97.3
<1 <1 <1
Treatment with deaerated PBC alone; ‡Treatment with NOsaturated, deaerated PBS; §Mean of triplicates ± SD (data shown are representive of two experiments).
© 1999 Harcourt Publishers Ltd
193
that 6-h exposure to PBS containing 1 mM of NO resulted in ~100% killing (Table 1). Since the stringency for anaerobicity in this system may not be high enough to entirely exclude O2, the possibility exists that other RNI such as NO2 could have been formed as a result of oxidation of NO, and the mycobacteriocidal effect observed may not be entirely attributable to NO. In order to test the antimycobacterial effect of NO more rigorously, we developed the anaerobic Calypso system, by which mycobacteria can be exposed to desired concentrations of authentic NO for various time intervals. Toxicity of NO and NO2 on M. smegmatis, BCG, and the virulent Erdman strain of M. tuberculosis The toxicity of authentic NO against mycobateria was evaluated stringently using the anaerobic Calypso system. Because of its rapid growth rate and its relatively avirulent nature compared to M. tuberculosis, M. smegmatis (strain mc2155) was used as target organisms to standardize experimental conditions. M. smegmatis (100–300 colony forming units [CFU]), plated on an 85 mm nylon membrane, were exposed to various doses of authentic NO in the gas-tight chamber. The doses of NO used are 0, 10, 25, 50, and 100% (v/v in O2-free N2). With a treatment time of 5–15 min, authentic NO is toxic to M. smegmatis in this in vitro system (Fig. 1). Under similar experimental parameters, both BCG and M. tuberculosis Erdman are also susceptible to the mycobacteriocidal effect of authentic NO, with an LD50 of about 25%–60% (Fig. 1). Compared to NO, NO2, another nitrogen-based radical, is significantly more toxic to M. smegmatis, BCG, and M. tuberculosis, with an LD50 of 1% to 1.5%. These results, together with those demonstrating that deaerated PBS saturated with NO exhibits antimycobacterial activity, indicate that NO and NO2 are both potential effector molecules capable of killing the tubercle bacillus. Susceptibility of Mycobacterium Spp. to ONOO– Peroxynitrite anion, a highly reactive oxidant formed by O2– and NO, has recently been shown effectively to kill E. coli in an in vitro system.16,17 Since activated macrophages, which constitute a major cellular defense component against mycobacteria, have the ability to generate substantial amounts of ONOO–,27 studies were initiated to test the antimycobacterial activity of this potent oxidant. Preliminary experiments were performed to establish optimal conditions using E. coli Y1090. Optimized parameters were then utilized in studies in which mycobacteria were the target organisms. Results obtained from this ONOO– killing model, which is distinct from the Calypso system, revealed that peroxynitrite effectively kills the avirulent M. smegmatis mc2155 and BCG (as assessed by Tubercle and Lung Disease (1999) 79(4), 191–198
194
Yu, Mitchell, Xing et al.
Fig. 1 Toxicity of NO and NO2 against mycobacteria. One hundred to 300 viable CFU of M. smegmatis (top), BCG (middle), or the virulent Erdman strain of M. tuberculosis (bottom) were plated onto nylon membranes and treated with various doses of authentic NO (open circle) or NO2 (closed circle) for 15 min in the Calypso chamber. Data shown are representative of three experiments.
CFU quantitation) in a dose dependent manner, with an LD50 of ~0.5 mM (Fig. 2). In contrast, virulent M. tubercuiosis strain Erdman is markedly resistant to ONOO– (Fig. 2). To examine whether the ONOO–-resistant phenotype observed in M. tuberculosis Erdman represents a more general phenomenon among the virulent tubercle bacilli, M. bovis strain Ravenel and M. tuberculosis M160, the highly prevalent clinical isolate in New York City also known as the C-strain,18 were subjected to 7–10 cycles of Tubercle and Lung Disease (1999) 79(4), 191–198
Fig. 2 Toxicity of ONOO– against mycobacteria. Approximately 5 × 105 to 1 × 106 viable CFU of M. smegmatis (top), BCG (middle), and M. tuberculosis Erdman (bottom) were subjected to eight consecutive cycles of treatment with ONOO– (closed circle) at the various doses indicated. Controls (open circle) were treated similarly with ONOO– that had been allowed to degrade for 3 min in potassium phosphate buffer (50 mM; pH 7.4). The experiment was repeated three times with similar results.
© 1999 Harcourt Publishers Ltd
Toxicity of nitrogen oxides and related oxidants on mycobacteria
Fig. 3 M. tuberculosis strains and M. bovis, compared to the relatively avirulent M. smegmatis and BCG, are resistant to the toxic effects of ONOO–. The various mycobacterial strains were subjected to 7–10 consecutive cycles of treatment with 1 mM of ONOO– in 50 mM phosphate buffer (pH 7.4). Exposure of the tested mycobacterial strains with ONOO– that had been degraded for 3 min in the buffer did not affect viability (data not shown). The results presented are representative of two experiments. ò M. smegmatis; n BCG; l M. bovis; ô M160.
exposure to 1 mM of peroxynitrite anion. Remarkably, compared to BCG and M. smegmatis, both the clinical M. tuberculosis M160 and M. bovis Ravenel are highly resistant to peroxynitrite anion (Fig. 3). While there was a greater than 7-log killing of M. smegmatis after 8–10 treatments with 1 mM ONOO–, the same regimen decreased the viability of M160 and Ravenel by less than 1–1.5 log. This same ONOO– treatment protocol, when applied to the relatively avirulent BCG Pasteur, resulted in 2.5–3.5 log reduction in viability (Fig. 3). The effects of metal chelators and hydroxyl radical scavengers on the mycobacteriocidal activity of ONOO– Transition metals are known to catalyze the heterolytic cleavage of ONOO– to form nitronium ion (NO2+), particularly at physiological pH.23 The nitronium ion can potentially modify enzymatic function by virtue of its ability to react with phenolics, such as that found in tyrosine residues of proteins, to form nitrophenols, reviewed by Beckman and Koppenol.26 In addition, OH· can be generated from H2O2 by the iron-catalyzed Haber-WeissFenton reaction.28 This latter reaction could be a confounding factor particularly if the manganese dioxide treatment did not remove all H2O2 from the ONOO– preparation.22,23 Consequently, the chelators ethylenediamine tetraacetate (EDTA), diethylenetriamine penta© 1999 Harcourt Publishers Ltd
195
Fig. 4 Effects of metal chelators on the toxicity of ONOO– against mycobacteria. 1 × 106 viable CFU of M. smegmatis were subjected to six consecutive cycles of treatment with 1 mM of ONOO– (open bar) in 50 mM potassium buffer (pH 7.4) alone, or in the presence of 100 uM of ethylenediamine tetraacetate (EDTA), diethylenetriamine pentaacetic acid (DTPA), or desferrioxamine (desferal). Controls were treated with ONOO– that had been allowed to degrade for 3 min in the phosphate buffer (closed bar). Data presented are means + SD derived from three independent experiments. Metal chelators did not significantly alter the antimycobacterial activity of ONOO– except for desferral, which exhibited an enhancing effect. *P = 0.015 (Wilcoxon rank sum test).
acetic acid (DTPA), and desferrioxamine (desferal) were used to evaluate the role of transition metals in the antimycobacterial activity of ONOO– observed in our in vitro system. A dose of ONOO– equivalent to that of the LD50 was used in these studies. In keeping with the bacteriocidal nature of ONOO– in the E. coli system,16,17 the ability of this NO-derived oxidant to kill M. smegmatis was not diminished by the chelators tested at a concentration of 100 uM (Fig. 4). Interestingly, by mechanisms yet to be defined, the chelator desferrioxamine enhanced the toxicity of ONOO– against M. smegmatis, while EDTA and DTPA had no effect on the antimycobacterial activity of this NO-derived reactive species (Fig. 4). Finally, since hydroxyl radical has been implicated in various biochemical reactions of peroxynitrite anion,22,23,26 the role of this reactive oxygen species in the ONOO–-mediated killing of mycobacteria was examined using various OH·scavengers. Our results indicate that at a concentration of 50 mM, the three OH· scavengers tested (mannitol, benzoate, and DMSO) did not protect M. smegmatis mc2155 against the toxic effects of ONOO– (Fig. 5). Since degraded ONOO– did not display any toxic effects, these data, collectively, indicate that peroxynitrite anion is the agent responsible for killing the relatively avirulent mycobacteria in the in vitro killing assay described here. The corollary, therefore, is that the virulent M. bovis as well as M. tuberculosis Erdman and M160, are specifically resistant to the antimycobacterial effect of ONOO–. Tubercle and Lung Disease (1999) 79(4), 191–198
196
Yu, Mitchell, Xing et al.
Fig. 5 Effects of hydroxyl radical-scavengers on the toxicity of ONOO– against mycobacteria. 1 × 106 viable CFU of M. smegmatis were subjected to six consecutive cycles of treatment with 1 mM of ONOO– (open bar) in 50 mM potassium buffer (pH 7.4) alone, or in the presence of 50 mM of mannitol, benzoate or dimethyl sulfoxide (DMSO). Controls were treated with ONOO– that had been allowed to degrade for 3 min in the phosphate buffer (closed bar). Data presented are means ± SD derived from three independent experiments. The effect of mannitol, benzoate, and DMSO on the antimycobacterial activity of ONOO– did not reach statistical significance as analyzed by the Wilcoxon rank sum test (P ≥ 0.318).
DISCUSSION The significance of reactive nitrogen intermediates in host defense against M. tuberculosis has been well established in various murine models.11–15 Particularly compelling evidence for an antimycobacterial role of RNI has been provided by the observation that mice with disruption of NOS2 exhibit marked susceptibility to M. tuberculosis.14 Despite the apparent importance of reactive nitrogen oxides in the control of tuberculous infection, the specific NO-related species that mediate antimycobacterial effects remain unknown. This study, designed to evaluate the toxicity of specific RNI against mycobacteria, has provided evidence that the antimycobacterial effects of the various NO-derived molecules tested vary substantially. In addition, the degree of susceptibility to these RNI species spans a wide range among the various mycobacterial strains examined. Specifically, our results suggest that: (i) authentic nitric oxide and nitrogen dioxide are effective antimycobacterial agents, NO2 being far more potent than NO, and (ii) while the relatively avirulent BCG and non-pathogenic M. smegmatis are susceptible to the toxic effects of ONOO–, the virulent M. bovis and M. tuberculosis, including the laboratory Erdman strain and the clinical isolate M160, are remarkably resistant. Our finding that authentic NO and NO2 possess antimycobacterial activity suggests that these nitrogen oxides are potential effector molecules responsible for RNImediated protective effect against M. tuberculosis. Interestingly, it has been reported, in a system akin to the one Tubercle and Lung Disease (1999) 79(4), 191–198
employed in this study, that treatment with 1 mM aqueous NO in deaerated water had no effect on the viability of E. coli.17 This phenomenon of variability in susceptibility to NO delivered in deaerated aqueous media among different target organisms (in this case, E. coli and Mycobacterium spp.) has also been observed in a study29 using an anaerobic system similar to Calypso. Thus, Shank et al.29 reported that a 30-min treatment with 100% NO (1 atmosphere) resulted in 99.9%, 94%, 90%, 29% and 0% reduction in viability of Pseudomonas fluorescence, Staphylococcus aureus, Clostridium 3679, Lactobacillus K7B, and Streptococcus durans, respectively.29 Together, these results suggest that the interactions between RNI and microbes are likely to be highly specific. This specificity may explain, at least in part, the wide range of susceptibility to NO among diverse classes of bacteria.29 These in vitro data must be interpreted, however, bearing certain caveats in mind: (i) the method used in delivering NO-saturated, deaerated aqueous media to BCG in this study, as well as the one employed in the report in which E. coli was the target organism,17 may generate higher nitrogen oxides because of possible oxygen contamination; (ii) the use of authentic NO and NO2 does not entirely exclude the contribution of other nitrogen oxides to the antimycobacterial effects observed in the Calypso anaerobic system. For example, it is possible that N2O4, generated via dimerization of NO2,30 could have participated in the NO2-mediated antimycobacterial effects; (iii) in both Calypso and the anaerobic system described by Shank et al.29 the target bacteria were spread out either on a nylon membrane (this study) or a filter disc29 when exposed to gaseous nitrogen oxides. This ‘solid-phase’ experimental condition may have affected the tested bacilli in such a way that their physiological states are different than those of an invading microbe in the host; (iv) the apparently high LD50 of NO against M. tuberculosis (~25% v/v in O2-free N2) in the Calypso system raises issues concerning the physiological relevance of the antimycobacterial effects observed. It is noteworthy, however, that the amounts of NO afforded by Calypso that ultimately react with the bacilli on the nylon membrane in a 5–15 minute treatment period cannot be readily extrapolated (the same problem exists in the anaerobic system used by Shank et al. to test the antibacterial effects of NO).29 For example, since mycobacteria were only treated with NO for less than 15 min in this study, the effective antimycobacterial doses of NO may be considerably less, should target organisms be exposed to RNI for an extended period of time, as is likely to be the case for phagocytized bacilli, based on the kinetics of high-output NO production by activated macrophages.10 Nevertheless, the in vivo significance of the antimycobacterial effects of NO and NO2 observed in vitro remains to be established. As accumulating evidence suggests that RNI may © 1999 Harcourt Publishers Ltd
Toxicity of nitrogen oxides and related oxidants on mycobacteria
indeed play a role in host defense against M. tuberculosis in humans,31 understanding the interactions between the reactive nitrogen species and the tubercle bacillus may illuminate mechanisms involved in pathogenesis and host defense in tuberculosis, and hopefully, lead to the design of effective anti-tuberculous treatments. The stupendous array of NO reactions in biological systems4,5,32,33 predicts that the in vivo RMI-M. tuberculosis interations go beyond those mediated by NO and NO2 per se. For example, the interconvertible redox forms of nitrogen monoxide, inclucing nitrosonium cation (NO+), NO, and nitroxyl anion (NO–), each endowed with distinctive chemical characteristics, can all participate in a broad spectrum of biochemical processes. To our knowledge, the contribution of NO+ and NO– to the antimycobacterial effects of RNI has not been examined. Recently, the demonstration of the participation of ONOO– in various reactions in the biological system adds yet another dimension to the complex interactions between RNI and microbes. Peroxynitrite is a highly potent oxidant formed by the reaction of NO and O2– with a rate constant of 6.7 × 109 M–1 S–1.34 The chemistry of ONOO– in biological systems is just beginning to be unravelled. Nevertheless, existing evidence suggests the potential of ONOO– to react with a wide array of biologically relevant targets in a broad range of physiological and pathophysiological processes.26 Indeed, nitrotyrosine, a product of nitration of protein tyrosine residues by ONOO–, has been detected by immunocytochemistry in tissues obtained from humans with diseases such as atherosclerosis and acute respiratory distress syndrome, reviewed by Beckman and Koppenol.26 In the context of host defense, ONOO– has been shown to effectively kill E.coli.16,17 Peroxynitrite anion also exhibits cytotoxic effect toward Trypanosoma cruzi,35 probably by inactivating thiol-containing enzymes of the parasite,36 as well as toward Candida albicans.37 Interestingly, Leishmania major, an intracellular pathogen, is resistant to the toxic effects of authentic ONOO–.38 Together, these results suggest that the response of microbes to this NO-derived oxidant is not uniform. Relevant to diseases caused by M. tuberculosis is the capacity of activated rodent alveolar macrophages to generate ONOO–.27 It has been estimated that these phagocytes can generate ONOO– at a rate of 0.11 nmol/ 106 cells/min when stimulated.27 In the rat lung, since there are ~107 macrophages residing in the fluid (total volume: ~1 ul) lining the pulmonary epithelium, the average rate of ONOO– formation in this micro milieu can reach 1 mM/min if all these phagocytes are triggered to produce peroxynitrite anion.27 Higher local concentrations may be achievable when macrophages attack the tubercle bacillus at close range. Indeed, the computation by Beckman et al.16,23 serves as a reminder that because of © 1999 Harcourt Publishers Ltd
197
the short half-life of peroxynitrite anion, the dose added in bolus in vitro, such as that used in the current study, is likely to be matched by a much lower, physiologically relevant steady-state concentration sustained by continuous production of ONOO– by activated macrophages in vivo. Given that macrophages have the capacity to generate the NO-derived ONOO–, which can kill the relatively avirulent mycobacteria tested, the ability of M. tuberculosis species and M. bovis to resist killing may contribute be an evolutionary adaptation to enhance intracellular survival of tubercle bacilli. Recently, the ability of various strains of M. tuberculosis to resist toxicity mediated by acidified nitrite has been reported to correlate with virulence in guinea pigs.39 In addition, a novel M. tuberculosis gene noxR1 has been shown to confer resistance on heterologous hosts against the toxic effects of RNI in vitro.40 Finally, M. tuberculosis aphC (alkyl hydroperoxide reductase subunit C) protects human cells from RNIinduced necrosis and apoptosis.41 Together, these findings suggest that M. tuberculosis may have evolved mechanisms by which the antimicrobial activity of toxic nitrogen oxides can be evaded. The contribution of the in vitro ONOO–-resistant phenotype to the in vivo persistent and possibly latent states of the tubercle bacillus, as well as the mechanisms involved in mediating such resistance, remains to be determined. In this respect, we have recently isolated M. tuberculosis cosmid clones that can confer ONOO– resistance on the otherwise susceptible BCG. Genetic analysis of these putative M. tuberculosis ONOO– resistant determinants is currently underway.
ACKNOWLEDGEMENTS This work was supported in part by the New Jersey Foundation, the Heiser Foundation, and the Howard Hughes Medical Institute. We thank the machinists Tony Leggradro, Edward Schwartz, and Lester Arnholdt of the Albert Einstein College of Medicine Machine Shop and Henry Keller of County Welding (White Plains, New York) for the construction of Calypso and the acrylic quench-flow reactor.
REFERENCES 1. Chanock S J, el Benna J, Smith R M, Babior B M. The respiratory burst oxidase. J Biol Chem 1994; 269: 24519–24522. 2. Iyer G Y N, Islam M F, Quastel J H. Biochemical aspects of phagocytosis. Nature (London) 1961; 192: 535–541. 3. Nathan C F. Mechanisms of macrophage antimicrobial activity. Trans R Soc Trop Med Hyg 1983; 77: 620–630. 4. Wientjes F B, Segal A W. NADPH oxidase and the respiratory burst. Seminars in Cell Biol 1995; 6: 357–365. 5. Fang F C. Mechanisms of nitric oxide-related antimicrobial activity. J Clin Invest 1997; 99: 2818–2825. 6. MacMicking J, Xie Q, Nathan C. Nitric oxide and macrophage function. Annu Rev Immunol 1997; 15: 323–350. 7. Nathan C. Inducible nitric oxide synthase: What difference does it make? J Clin Invest 1997; 100: 2417–2423.
Tubercle and Lung Disease (1999) 79(4), 191–198
198
Yu, Mitchell, Xing et al.
8. Ding A H, Nathan C F, Stueher D J. Release of reactive nitrogen intermediates and reactive oxygen intermediates from mouse peritoneal macrophages. Comparison of activating cytokines and evidence for independent production. J Immunol 1988; 141: 2407–2412. 9. Denis M. Interferon-gamma-treated murine macrophages inhibit growth of tubercle bacilli via the generation of reactive nitrogen intermediates. Cellu Immunol 1991; 132: 150–157. 10. Chan J, Xing Y, Magliozzo R S, Bloom B R. Killing of virulent Mycobacterium tuberculosis by reactive nitrogen intermediates produced by activated murine macrophages. J Exp Med 1992; 175: 1111–1122. 11. Flynn J L, Chan J, Triebold K J, Dalton D K, Stewart T A, Bloom B R. An essential role for IFN-γ in resistance to Mycobacterium tuberculosis infection. J Exp Med 1993; 178: 2249–2254. 12. Chan J, Tanaka K, Carroll D, Flynn J, Bloom B R. Effects of nitric oxide synthase inhibitors on murine infection with M. tuberculosis. Infect Immun 1994; 63: 736–740. 13. Flynn J L, Goldstein M M, Chan J et al. Tumor necrosis factor is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity 1995; 2: 561–572. 14. MacMicking J, North R J, LaCourse R, Mudgett J S, Shah S K, Nathan C F. Identification of nitric oxide synthase as a protective locus against tuberculosis. Proc Natl Acad Sci USA 1997; 84: 5243–5248. 15. Flynn J, Scanga C A, Tanaka K, Chan J. Reactivation of latent murine tuberculosis by inhibition of inducible nitric oxide synthase. J Immunol 1998; 160: 1796–1803. 16. Zhu L, Gunn C, Beckman J S. Bactericidal activity of peroxynitrite. Arch Biochem Biophys 1992; 298: 452–457. 17. Brunelli L, Crow J P, Beckman J S. The comparative toxicity of nitric oxide and peroxynitrite to Escherichia coli. Arch Biochem Biophys 1995; 316: 327–334. 18. Friedman C R, Quinn G C, Kreiswirth B N et al. Widespread dissemination of a drug-susceptible strain of Mycobacterium tuberculosis. J Infect Dis 1997; 176: 478–484. 19. Riska P, Jacobs W R Jr, Bloom B R, McKitrick J, Chan J. Specific identification of Mycobacterium tuberculosis by the luciferase reporter mycobacteriophage: the use of β-nitro-α-acetylaminoβ-hydroxy-propiophenone. J Clinical Microbiol 1997; 35: 3225–3231. 20. Sambrook J, Fritsch E F, Maniatis T. Molecular Cloning 1989; Second Editions. Cold Spring Harbor Laboratory Press. 21. Reed J W, Ho H H, Jolly W L. Chemical syntheses with a quenched flow reactor. Hydroxytrihydroborate and peroxynitrite. J Amer Chem Soc 1973; 96: 1248–1249. 22. Beckman J S, Beckman T W, Chen J, Marshall P A, Freeman B A. Apparent hydroxyl radical production by peroxynitrite: Implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci USA 1990; 87: 1620–1624. 23. Beckman J S, Chen J, Ischiropoulos H, Crow J P. Oxidative chemistry of peroxynitrite. Methods in Enzymol 1994; 233: 229–240. 24. Ignarro L J, Byrns R E, Buga G M, Wood K S. Endothelium-
Tubercle and Lung Disease (1999) 79(4), 191–198
25.
26.
27.
28.
29. 30.
31.
32. 33. 34. 35.
36.
37.
38.
39.
40.
41.
derived relaxing factor from pulmonary artery and vein possesses pharmacologic and chemical properties identical to those of nitric oxide radical. Circ Res 1987; 61: 866–879. Stuehr D J, Nathan C F. Nitric oxide: A macrophage product responsible for cytostasis and respiratory inhibition in tumor target cells. J Exp Med 1989; 169: 1543–1555. Beckman J S, Koppenol W H. Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and the ugly. Am J Physiol 1996; 271: C1424–1437. Ischiropouls H, Zhu L, Beckman J S. Peroxynitrite formation from macrophage-derived nitric oxide. Arch Biochem Biophys 1992; 298: 446–451. Haber F, Weiss J. The catalytic decomposition of hydrogen peroxide by iron salts. Proc R Soc London Ser 1934; A 147: 332–351. Shank J L, Silliker J H, Harper R H. The effect of nitric oxide on bacteria. Appl Microbiol 1962; 10: 185–189. Cotton F A, Wilkinson G. In Advanced Inorganic Chemistry. A Comprehensive Text. John Wiley and Sons, Inc. 1980; p. 423–428. Nicholson S, Monecini-Almeida M da G, Lapa e Silva J R. et al. Inducible nitric oxide synthase in pulmonary alveolar macrophages from patients with tuberculosis. J Exp Med 1996; 183: 2293–2302. Stamler J S, Singel D J, Loscalzo J. Biochemistry of nitric oxide and its redox-activated forms. Science 1992; 258: 1898–1902. Stamler J S. Redox signaling: Nitrosylation and related target interactions of nitric oxide. Cell 1994; 78: 931–936. Huie R E, Padmaja S. The reaction rate of nitric oxide with superoxide. Free Radical Res Commun 1993; 18: 195–199. Denicola A, Rubbo H, Rodriguez D, Radi R. Peroxynitritemediated cytotoxicity to Trypanosoma cruzi. Arch Biochem Biophys 1993; 304: 279–286. Rubbo H, Denicola A, Radi R. Peroxynitrite inactivates thiolcontaining enzymes of Trypanosoma cruzi energetic metabolism and inhibits cell respiration. Arch Biochem Biophys 1994; 308: 96–102. Vasquez-Torres A, Jones-Carson J, Balish E. Peroxynitrite contributes to the candidacidal activity of nitric oxideproducing macrophages. Infect Immun 1996; 64: 3127–3133. Assreuy J, Cunha F Q, Epperlein M, Noronha-Dutra A, O’Donnell C A, Liew F Y, Moncada S. Production of nitric oxide and superoxide by activated macrophages and killing of Leishmania major. Eur J Immunol 1994; 24: 672–676. O’Brian L, Carmichael J, Lowrie D B, Andrew P W. Strains of Mycobacterium tuberculosis differ in susceptibility to reactive nitrogen intermediates in vitro. Infect Immun 1994; 62: 5187–5190. Ehrt S, Shiloh M U, Ruan J, Choi M, Gunzburg S, Nathan C, Xie Q-W, Riley L W. A novel antioxidant gene from Mycobacterium tuberculosis. J Exp Med 1997; 186: 1885–1896. Chen L, Xie Q-W, Nathan C. Alkyl hydroperoxide reductase subunit C (AhpC) protects bacterial and human cells against reactive nitrogen intermediates. Mol Cell 1998; 1: 795–805.
© 1999 Harcourt Publishers Ltd