Evidence for Mutagenesis by Nitric Oxide during Nitrate Metabolism in Escherichia coli
http://www.100md.com
《细菌学杂志》
Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, Georgia 30322
ABSTRACT
In Escherichia coli, nitrosative mutagenesis may occur during nitrate or nitrite respiration. The endogenous nitrosating agent N2O3 (dinitrogen trioxide, nitrous anhydride) may be formed either by the condensation of nitrous acid or by the autooxidation of nitric oxide, both of which are metabolic by-products. The purpose of this study was to determine which of these two agents is more responsible for endogenous nitrosative mutagenesis. An nfi (endonuclease V) mutant was grown anaerobically with nitrate or nitrite, conditions under which it has a high frequency of A:T-to-G:C transition mutations because of a defect in the repair of hypoxanthine (nitrosatively deaminated adenine) in DNA. These mutations could be greatly reduced by two means: (i) introduction of an nirB mutation, which affects the inducible cytoplasmic nitrite reductase, the major source of nitric oxide during nitrate or nitrite metabolism, or (ii) flushing the anaerobic culture with argon (which should purge it of nitric oxide) before it was exposed to air. The results suggest that nitrosative mutagenesis occurs during a shift from nitrate/nitrite-dependent respiration under hypoxic conditions to aerobic respiration, when accumulated nitric oxide reacts with oxygen to form endogenous nitrosating agents such as N2O3. In contrast, mutagenesis of nongrowing cells by nitrous acid was unaffected by an nirB mutation, suggesting that this mutagenesis is mediated by N2O3 that is formed directly by the condensation of nitrous acid.
INTRODUCTION
In the respiration of Escherichia coli and many other bacteria, nitrate and nitrite may become the predominant electron acceptors in the absence of oxygen (2, 5, 14). During nitrate or nitrite metabolism, mutagenic reactive nitrogen oxides, such as nitrous acid (HNO2) and nitric oxide (NO·), are formed. Thus, during hypoxia, bacteria must defend themselves against the toxic and mutagenic by-products of nitrate/nitrite respiration just as they have to defend themselves against reactive oxygen species during aerobic respiration. These agents are also encountered as ubiquitous environmental toxins, and NO· is generated as an antibacterial agent by mammalian macrophages (26, 29).
The combination of hypoxia and nitrate or nitrite induces the major nitrate and nitrite reductases of E. coli, and alternative anaerobic respiratory pathways (e.g., fumarate reductase) are turned off (14). Nitrate/nitrite metabolism generates a mutagenic by-product, the nitrosating agent N2O3 (dinitrogen trioxide, nitrous anhydride). N2O3 can arise either from the condensation of molecular HNO2 or from the autooxidation of NO· (Fig. 1). For the cell's self protection, nitrite levels are kept low by strong nitrite reductase activity and by nitrite efflux pumps, and reactive nitrogen species, if they are at all produced, are kept tightly bound to nitrite reductases during the six-electron reduction of nitrite to ammonia (14). Nevertheless, NO· is detectable in E. coli cell suspensions during nitrate respiration, and its production is dependent on nitrite reductase activity (8).
Nitrosating agents produce mutagenic lesions. When DNA is exposed in vitro to HNO2 or to NO·, the exocyclic amines of the nucleobases form unstable N-nitroso (-N-N O) derivatives that lead to deamination. Thus, adenine is deaminated to hypoxanthine, guanine is deaminated to xanthine, and cytosine is deaminated to uracil (23). The deaminated products pair with different bases than their parents do. Therefore, they almost always produce mutations during subsequent replication. Nitrosation of cellular secondary amines and amides produce alkylating agents that cause mutagenic lesions at many sites in DNA (22, 25, 29). Other DNA lesions (4) include interstrand and intrastrand cross-links, protein-DNA cross-links, and the formation of oxanine from guanine (24). With the aid of repair-deficient mutants, mutagenic deamination (21) and alkylation damage (22, 25) have been observed during nitrate and nitrite metabolism.
The present study focuses on the mutagenic nitrosative deamination of DNA bases. It seeks to answer two questions. The first question concerns the proximate source of mutagenic N2O3 during nitrate respiration in E. coli. Is it mainly HNO2 or NO· There are theoretical arguments for and against each possibility. Nitrite might be favored because it is an obligatory intermediate and a free metabolite, whereas NO· is a by-product. On the other hand, NO· might be favored because it is readily autooxidized to N2O3, whereas the formation of molecular HNO2 from nitrite is very poor at intracellular pH because of the low pK of HNO2. The second question concerns the source of O2 for the formation of N2O3 from NO·. Is it the residual air in the hypoxic cultures, or is it the air to which the cultures are exposed when they are subsequently plated to measure mutation frequencies
To answer these questions, the present study will make extensive use of an nfi mutant (15) of E. coli. It lacks endonuclease V, a DNA repair enzyme that recognizes hypoxanthine and xanthine in DNA. Although endonuclease V was discovered first in E. coli (12), it is now recognized as part of a superfamily of widely distributed proteins that are found in organisms as diverse as humans and bacteria (1). The enzyme cleaves the second phosphodiester bond 3' to a deaminated purine (16, 30), thereby initiating an excision repair of the mutagenic lesion. Compared to the wild type, nfi mutants have an elevated frequency of A:TG:C mutations induced by HNO2 (21) or by hypoxic growth in the presence of nitrate or nitrate (28). These properties are consistent with a defect in the repair of deaminated adenine (i.e., hypoxanthine) in DNA, which pairs with cytosine instead of the original thymine. In confirmation of this conclusion, there was no similar mutator effect for a base pair containing 6-methyladenine, which is resistant to nitrosative deamination (21). The unusual susceptibility of an nfi mutant to the mutagenic effects of nitrosating agents will be used in the present study to monitor endogenous nitrosation of DNA.
MATERIALS AND METHODS
Strains used. The E. coli K-12 strains used are shown in Table 1. Transductions with bacteriophage P1 dam rev6 were as previously described (15), except that selection for streptomycin resistance (a recessive trait) was performed after the transductants were grown for six generations in Luria-Bertani (LB) broth (19) containing 20 mM sodium citrate. nirB was transduced by selection for a linked rpsL (streptomycin resistance) marker and was scored by chemical detection of nitrite utilization, as previously described (6).
Media. LB-glycerol-fumarate medium for anaerobic growth was as previously described (28). Routine propagation of cultures was in LB broth. Streptomycin was used at a final concentration of 200 μg/ml, and tetracycline was used at 15 μg/ml.
Mutation frequencies. Measurements of lacZ and trpA reversion frequencies, anaerobic growth with sodium nitrate and sodium nitrite, and HNO2 mutagenesis were as previously described (21, 28).
Argon-purged cultures. Cultures containing 100 to 1,000 cells of the nfi mutant BW1506 were grown 20 h at 37°C in 18 ml of LB-glycerol-fumarate medium containing 100 mM NaNO3. The vessels were 22-ml screw-cap vials with polytetrafluoroethylene/silicone septa (Sigma-Aldrich, St. Louis, MO). The argon, from which traces of oxygen were removed (13), was bubbled through the culture for 5 min via a hypodermic needle inserted through the septum.
RESULTS
Metabolic source of N2O3. The endogenous production of nitrosating agents was monitored by measuring the mutator effect of an nfi mutation during anaerobic nitrate/nitrite metabolism. The strains used were derivatives of a lacZ mutant that can revert to the Lac+ phenotype only by an A:TG:C mutation (9). Anaerobic conditions were attained by growing the cultures in completely filled tubes, without a headspace. The medium contained glycerol as an electron donor. For the sake of the controls in which nitrate and nitrite were absent, fumarate was added to all of the cultures as an alternate electron acceptor. After growth to saturation, the cultures were plated on a lactose minimal medium to score the number of Lac+ revertants. To determine the relative importance of HNO2 or NO· as precursors of N2O3 during nitrate/nitrate metabolism, a nirB mutation was introduced into an nfi mutant. The nirB deletion affects the major, inducible nitrite reductase of E. coli, which is a cytoplasmic, NADH-dependent enzyme that produces about two-thirds of the detectable free NO· (8); the remainder is generated by a periplasmic cytochrome c nitrite reductase encoded by the nrf genes. During nitrate metabolism, the nirB mutant is expected to accumulate nitrite and HNO2 and to have a corresponding deficit in the production of NO· (Fig. 1). If nirB increases the mutator effect of nfi, then HNO2 is the cause of nitrosative mutagenesis under these growth conditions; if it decreases the mutator effect, then NO· is the mutagenic agent. The nfi mutant (Fig. 2) demonstrated a nitrate(nitrite)-dependent mutator phenotype, as was previously observed (28). However, the addition of the nirB mutation reduced its mutation frequency almost to the level of the nfi+ strain. The effect of nirB was similar for both nitrate- and nitrite-induced mutagenesis, suggesting that both agents produced DNA damage through a common pathway. The dependence of the mutator effect on nitrite reductase, a potential generator of NO·, implicates NO· rather than HNO2 as the proximate source of N2O3 during nitrate/nitrite metabolism.
An unusual finding within the raw data for Fig. 2, was that of a fairly symmetrical, nonlogarithmic distribution of the data. Within each set of samples, the arithmetic mean was within 5% of the median value, and there were no values differing by more than 2.5 standard deviations from the mean (see Discussion).
Source of O2 for N2O3 formation. The production of the nitrosating agent N2O3 from NO· requires O2, but the cultures were incubated anaerobically. There must have been a point in the experiments when both NO· and O2 were present. There are two possibilities for the source of the O2. The first is that the O2 that was dissolved in the medium at the start of the experiment may not have been fully consumed before significant amounts of NO· accumulated in the sealed tubes. The second possibility is that when the cells were plated, O2 in the air reacted with NO· that had accumulated in the anaerobic cultures. The question was resolved by flushing the cultures with O2-free argon either before or after their growth. The results (Table 2) indicated that removing the dissolved oxygen before growth had no significant effect on the mutation frequency. However, when the cultures were flushed with argon after growth, which would remove NO· before exposure to air, the mutation frequency was reduced to that seen without nitrate (as in Fig. 2). Therefore, nitrosative mutagenesis occurs when the bacteria are shifted from nitrate-dependent to oxygen-dependent respiration, during which accumulated NO· may be oxidized to N2O3.
Mechanism of mutagenic nitrosative deamination by HNO2. HNO2 mutagenesis may be effected by briefly exposing nongrowing E. coli to sodium nitrite at acid pH (31). It has been long assumed that the mechanism of HNO2-induced DNA deamination in intact E. coli is the same as that for phages and free DNA, i.e., it is produced by N2O3 that is derived directly from the condensation of HNO2. However, the results in the present study raise the possibility that NO· may be involved. Theoretically, the nitrite in the acid medium could be reduced enzymatically, yielding NO· as a by-product that is then autooxidized to N2O3. The required acid pH might function to convert the nitrite ions transiently into an uncharged form (molecular HNO2) that might traverse the bacterial membranes more easily.
To determine whether HNO2 mutagenesis is also mediated by NO·, a nirB mutation was again used. The nirB deletion was tested for its effect on the reversion of the trpA58 allele in the nfi mutant, BW1177. This reversion also occurs by an A:TG:C transition. A previous study (21) had found that the HNO2-induced reversion of trpA in this strain is nfi dependent (>99.5%) and occurs with a 16-fold-higher frequency than lacZ reversion in our lacZ nfi mutant. Thus, it should be a sensitive indicator of any effect produced by the nirB deletion. The results (Table 3) indicate that the nirB mutation has little, if any, effect on HNO2-induced mutagenesis, suggesting that most HNO2 mutagenesis is not mediated by NO·. There was an additional reason for choosing the trpA58 mutation indicator strain BW1177 for the present study, apart from its high frequency of mutability by HNO2. It has a different genetic background from our lacZ strains and, for unknown reasons that are now under investigation, it does not appear to be significantly mutagenized by nitrate or nitrite in anaerobic cultures (Table 3). However, these conditions of nitrate respiration and HNO2 treatment were previously shown to produce about an equal number of revertants in the lacZ indicator strain BW1506 (21, 28). This result further supports the conclusion that there are different mechanisms for the production of nitrosative DNA damage by HNO2 and by nitrite metabolism. Therefore, during HNO2 treatment, most of the N2O3 is formed directly from HNO2 rather than through its metabolism to NO·.
DISCUSSION
When mutations accumulate in a growing culture, the progeny of the earliest mutants will predominate. Variations in mutation frequency mainly reflect the different times when the first mutants appeared in each culture. The results usually demonstrate a skewed, logarithmic distribution rather than a symmetrical spread, and there are frequent jackpot values. Consequently, the arithmetic mean and the median values usually do not agree. However, in this and in previous work (28) with nitrate- and nitrite-induced mutagenesis, the results were evenly distributed, suggesting that the mutagenic events occurred at about the same time in each culture, which is consistent with the hypothesis that the lesions were produced when the saturated, NO·-containing cultures were exposed to air.
What is the source of the NO· Unlike the denitrifying bacteria, which reduce nitrite directly to NO·, E. coli appears to reduce nitrite directly to ammonia (2). Nevertheless, several studies (reviewed in reference 8) have documented the production of NO· during nitrate/nitrite metabolism in E. coli. With the aid of a recently available sensitive NO· electrode, it was found that the production of NO· from nitrite by extracts of E. coli could be eliminated by mutations in both the Nrf (periplasmic) and NirB (cytoplasmic) nitrite reductases (8). At the high level of nitrate used in the current study, the expression of Nrf should be repressed, whereas that of NirB should be fully induced (20, 27), thereby explaining why the nirB mutation was effective in preventing most of the nitrosative mutagenesis. However, in early studies with purified NirB, neither NO· nor any other partial reduction product of nitrite was detected (7); hydroxylamine was postulated to be a latent intermediate only because it can serve as a substrate for the enzyme (17). There are no known reactions in E. coli whereby the end product of NirB, ammonia, can be converted to NO·. Therefore, either NO· is a previously undetected latent intermediate that leaks from NirB in very small amounts or else it is the product of a reaction between a cytoplasmic constituent and an NirB-bound intermediate. With the current availability of more sensitive methods of detection, it may be worthwhile to reinvestigate the possibility that traces of NO· may be released from the purified enzyme during the NADH-dependent reduction of nitrite.
There are parallels between oxidative and nitrosative DNA damage. The reactive oxygen species and reactive nitrogen oxides are mostly by-products rather than intermediates of aerobic and anaerobic metabolism, respectively. The reactive oxygen species undergo electron chain reactions in which free radicals generate other free radicals (11). Similarly, N-nitroso derivatives undergo transnitrosation reactions (29). The deleterious nature of the passage from anaerobic nitrate-dependent to oxygen-dependent respiration, as demonstrated in this work, is reminiscent of reperfusion injury, in which reactive oxygen species are produced when mammalian tissues pass from a hypoxic to an aerobic state (3).
Although the present study has demonstrated mutagenesis occurring through a sudden shift in metabolic state, such a shift may not be a necessary prerequisite. Hypoxia rather than anaerobiosis is sufficient for the induction of nitrate and nitrite reductase activities (18). It is possible that when the cell is growing in the presence of nitrate under hypoxic conditions, nitrate and nitrite reductase activities may be sufficiently induced, and there may still be sufficient oxygen present, so that NO· and N2O3 are formed. Thus, nitrosative mutagenesis has been demonstrated in cultures grown to saturation in open flasks (28). There may be many natural environments, perhaps even the human intestine, where these conditions may exist.
ACKNOWLEDGMENTS
The capable technical assistance of Carrie L. Flood and Dawit Seyfe is gratefully acknowledged. I also thank Robert K. Poole and Stephen Spiro for their helpful discussions.
This study was supported by NIH grant ES11163.
Mailing address: Department of Pathology and Laboratory Medicine, Emory University, Whitehead Bldg., Rm. 141, 615 Michael St., Atlanta, GA 30322. Phone: (404) 712-2812. Fax: (404) 727-8538. E-mail: bweiss2@emory.edu.
REFERENCES
Aravind, L., D. R. Walker, and E. V. Koonin. 1999. Conserved domains in DNA repair proteins and evolution of repair systems. Nucleic Acids Res. 27:1223-1242.
Berks, B. C., S. J. Ferguson, J. W. B. Moir, and D. J. Richardson. 1995. Enzymes and associated electron transport systems that catalyse the respiratory reduction of nitrogen oxides and oxyanions. Biochim. Biophys. Acta 1232:97-173.
Bulkley, G. B. 1994. Reactive oxygen metabolites and reperfusion injury: aberrant triggering of reticuloendothelial function. Lancet 344:934-936.
Burney, S., J. L. Caulfield, J. C. Niles, J. S. Wishnok, and S. R. Tannenbaum. 1999. The chemistry of DNA damage from nitric oxide and peroxynitrite. Mutat. Res. 424:37-49.
Cole, J. 1996. Nitrate reduction to ammonia by enteric bacteria: redundancy, or a strategy for survival during oxygen starvation FEMS Microbiol. Lett. 136:1-11.
Cole, J. A., and F. B. Ward. 1973. Nitrite reductase-deficient mutants of Escherichia coli K-12. J. Gen. Microbiol. 76:21-29.
Coleman, K. J., A. Cornish-Bowden, and J. A. Cole. 1978. Purification and properties of nitrite reductase from Escherichia coli K-12. Biochem. J. 175:483-493.
Corker, H., and R. K. Poole. 2003. Nitric oxide formation by Escherichia coli. Dependence on nitrite reductase, the NO-sensing regulator Fnr, and flavohemoglobin Hmp. J. Biol. Chem. 278:31584-31592.
Cupples, C. G., and J. H. Miller. 1986. A set of lacZ mutations in Escherichia coli that allow rapid detection of each of the six base substitutions. Proc. Natl. Acad. Sci. USA 86:5345-5349.
Darwin, A., P. Tormay, L. Page, L. Griffiths, and J. Cole. 1993. Identification of the formate dehydrogenases and genetic determinants of formate-dependent nitrite reduction by Escherichia coli K-12. J. Gen. Microbiol. 139:1829-1840.
Fridovich, I. 1998. Oxygen toxicity: a radical explanation. J. Exp. Biol. 201:1203-1209.
Gates, F. T., and S. Linn. 1977. Endonuclease V of Escherichia coli. J. Biol. Chem. 252:1647-1653.
Gaudu, P., and B. Weiss. 1996. SoxR, a [2Fe-2S] transcription factor, is active only in its oxidized form. Proc. Natl. Acad. Sci. USA 93:10094-10098.
Gennis, R. B., and V. Stewart. 1996. Respiration, p. 217-261. In F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed. ASM Press, Washington, D.C.
Guo, G., and B. Weiss. 1998. Endonuclease V (nfi) mutant of Escherichia coli K-12. J. Bacteriol. 180:46-51.
He, B., H. Qing, and Y. W. Kow. 2000. Deoxyxanthosine in DNA is repaired by Escherichia coli endonuclease V. Mutat. Res. 459:109-114.
Jackson, R. H., J. A. Cole, and A. Cornish-Bowden. 1981. The steady-state kinetics of the NADH-dependent nitrite reductase from Escherichia coli K 12. Nitrite and hydroxylamine reduction. Biochem. J. 199:171-178.
Metheringham, R., and J. A. Cole. 1997. A reassessment of the genetic determinants, the effect of growth conditions and the availability of an electron donor on the nitrosating activity of Escherichia coli K-12. Microbiology 143:2647-2656.
Miller, J. H. 1992. A short course in bacterial genetics. A laboratory manual and handbook for Escherichia coli and related bacteria. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Page, L., L. Griffiths, and J. A. Cole. 1990. Different physiological roles of two independent pathways for nitrite reduction to ammonia by enteric bacteria. Arch. Microbiol. 154:349-354.
Schouten, K. A., and B. Weiss. 1999. Endonuclease V protects Escherichia coli against specific mutations caused by nitrous acid. Mutat. Res. 435:245-254.
Sedgwick, B. 1997. Nitrosated peptides and polyamines as endogenous mutagens in O6-alkylguanine-DNA alkyltransferase deficient cells. Carcinogenesis 18:1561-1567.
Shapiro, R., and S. H. Pohl. 1968. The reaction of ribonucleosides with nitrous acid. Side products and kinetics. Biochemistry 7:448-455.
Suzuki, T., H. Ide, M. Yamada, N. Endo, K. Kanaori, K. Tajima, T. Morii, and K. Makino. 2000. Formation of 2'-deoxyoxanosine from 2'-deoxyguanosine and nitrous acid: mechanism and intermediates. Nucleic Acids Res. 28:544-551.
Taverna, P., and B. Sedgwick. 1996. Generation of an endogenous DNA-methylating agent by nitrosation in Escherichia coli. J. Bacteriol. 178:5105-5111.
Victorin, K. 1994. Review of the genotoxicity of nitrogen oxides. Mutat. Res. 317:43-55.
Wang, H., and R. P. Gunsalus. 2000. The nrfA and nirB nitrite reductase operons in Escherichia coli are expressed differently in response to nitrate than to nitrite. J. Bacteriol. 182:5813-5822.
Weiss, B. 2001. Endonuclease V of Escherichia coli prevents mutations from nitrosative deamination during nitrate/nitrite respiration. Mutat. Res. 461:301-309.
Williams, D. L. H. 1988. Nitrosation. Cambridge University Press, Cambridge, United Kingdom.
Yao, M., Z. Hatahet, R. J. Melamede, and Y. W. Kow. 1994. Purification and characterization of a novel deoxyinosine-specific enzyme, deoxyinosine 3' endonuclease, from Escherichia coli. J. Biol. Chem. 269:16260-16268.
Zimmermann, F. 1977. Genetic effects of nitrous acid. Mutat. Res. 39:127-148.(Bernard Weiss)
ABSTRACT
In Escherichia coli, nitrosative mutagenesis may occur during nitrate or nitrite respiration. The endogenous nitrosating agent N2O3 (dinitrogen trioxide, nitrous anhydride) may be formed either by the condensation of nitrous acid or by the autooxidation of nitric oxide, both of which are metabolic by-products. The purpose of this study was to determine which of these two agents is more responsible for endogenous nitrosative mutagenesis. An nfi (endonuclease V) mutant was grown anaerobically with nitrate or nitrite, conditions under which it has a high frequency of A:T-to-G:C transition mutations because of a defect in the repair of hypoxanthine (nitrosatively deaminated adenine) in DNA. These mutations could be greatly reduced by two means: (i) introduction of an nirB mutation, which affects the inducible cytoplasmic nitrite reductase, the major source of nitric oxide during nitrate or nitrite metabolism, or (ii) flushing the anaerobic culture with argon (which should purge it of nitric oxide) before it was exposed to air. The results suggest that nitrosative mutagenesis occurs during a shift from nitrate/nitrite-dependent respiration under hypoxic conditions to aerobic respiration, when accumulated nitric oxide reacts with oxygen to form endogenous nitrosating agents such as N2O3. In contrast, mutagenesis of nongrowing cells by nitrous acid was unaffected by an nirB mutation, suggesting that this mutagenesis is mediated by N2O3 that is formed directly by the condensation of nitrous acid.
INTRODUCTION
In the respiration of Escherichia coli and many other bacteria, nitrate and nitrite may become the predominant electron acceptors in the absence of oxygen (2, 5, 14). During nitrate or nitrite metabolism, mutagenic reactive nitrogen oxides, such as nitrous acid (HNO2) and nitric oxide (NO·), are formed. Thus, during hypoxia, bacteria must defend themselves against the toxic and mutagenic by-products of nitrate/nitrite respiration just as they have to defend themselves against reactive oxygen species during aerobic respiration. These agents are also encountered as ubiquitous environmental toxins, and NO· is generated as an antibacterial agent by mammalian macrophages (26, 29).
The combination of hypoxia and nitrate or nitrite induces the major nitrate and nitrite reductases of E. coli, and alternative anaerobic respiratory pathways (e.g., fumarate reductase) are turned off (14). Nitrate/nitrite metabolism generates a mutagenic by-product, the nitrosating agent N2O3 (dinitrogen trioxide, nitrous anhydride). N2O3 can arise either from the condensation of molecular HNO2 or from the autooxidation of NO· (Fig. 1). For the cell's self protection, nitrite levels are kept low by strong nitrite reductase activity and by nitrite efflux pumps, and reactive nitrogen species, if they are at all produced, are kept tightly bound to nitrite reductases during the six-electron reduction of nitrite to ammonia (14). Nevertheless, NO· is detectable in E. coli cell suspensions during nitrate respiration, and its production is dependent on nitrite reductase activity (8).
Nitrosating agents produce mutagenic lesions. When DNA is exposed in vitro to HNO2 or to NO·, the exocyclic amines of the nucleobases form unstable N-nitroso (-N-N O) derivatives that lead to deamination. Thus, adenine is deaminated to hypoxanthine, guanine is deaminated to xanthine, and cytosine is deaminated to uracil (23). The deaminated products pair with different bases than their parents do. Therefore, they almost always produce mutations during subsequent replication. Nitrosation of cellular secondary amines and amides produce alkylating agents that cause mutagenic lesions at many sites in DNA (22, 25, 29). Other DNA lesions (4) include interstrand and intrastrand cross-links, protein-DNA cross-links, and the formation of oxanine from guanine (24). With the aid of repair-deficient mutants, mutagenic deamination (21) and alkylation damage (22, 25) have been observed during nitrate and nitrite metabolism.
The present study focuses on the mutagenic nitrosative deamination of DNA bases. It seeks to answer two questions. The first question concerns the proximate source of mutagenic N2O3 during nitrate respiration in E. coli. Is it mainly HNO2 or NO· There are theoretical arguments for and against each possibility. Nitrite might be favored because it is an obligatory intermediate and a free metabolite, whereas NO· is a by-product. On the other hand, NO· might be favored because it is readily autooxidized to N2O3, whereas the formation of molecular HNO2 from nitrite is very poor at intracellular pH because of the low pK of HNO2. The second question concerns the source of O2 for the formation of N2O3 from NO·. Is it the residual air in the hypoxic cultures, or is it the air to which the cultures are exposed when they are subsequently plated to measure mutation frequencies
To answer these questions, the present study will make extensive use of an nfi mutant (15) of E. coli. It lacks endonuclease V, a DNA repair enzyme that recognizes hypoxanthine and xanthine in DNA. Although endonuclease V was discovered first in E. coli (12), it is now recognized as part of a superfamily of widely distributed proteins that are found in organisms as diverse as humans and bacteria (1). The enzyme cleaves the second phosphodiester bond 3' to a deaminated purine (16, 30), thereby initiating an excision repair of the mutagenic lesion. Compared to the wild type, nfi mutants have an elevated frequency of A:TG:C mutations induced by HNO2 (21) or by hypoxic growth in the presence of nitrate or nitrate (28). These properties are consistent with a defect in the repair of deaminated adenine (i.e., hypoxanthine) in DNA, which pairs with cytosine instead of the original thymine. In confirmation of this conclusion, there was no similar mutator effect for a base pair containing 6-methyladenine, which is resistant to nitrosative deamination (21). The unusual susceptibility of an nfi mutant to the mutagenic effects of nitrosating agents will be used in the present study to monitor endogenous nitrosation of DNA.
MATERIALS AND METHODS
Strains used. The E. coli K-12 strains used are shown in Table 1. Transductions with bacteriophage P1 dam rev6 were as previously described (15), except that selection for streptomycin resistance (a recessive trait) was performed after the transductants were grown for six generations in Luria-Bertani (LB) broth (19) containing 20 mM sodium citrate. nirB was transduced by selection for a linked rpsL (streptomycin resistance) marker and was scored by chemical detection of nitrite utilization, as previously described (6).
Media. LB-glycerol-fumarate medium for anaerobic growth was as previously described (28). Routine propagation of cultures was in LB broth. Streptomycin was used at a final concentration of 200 μg/ml, and tetracycline was used at 15 μg/ml.
Mutation frequencies. Measurements of lacZ and trpA reversion frequencies, anaerobic growth with sodium nitrate and sodium nitrite, and HNO2 mutagenesis were as previously described (21, 28).
Argon-purged cultures. Cultures containing 100 to 1,000 cells of the nfi mutant BW1506 were grown 20 h at 37°C in 18 ml of LB-glycerol-fumarate medium containing 100 mM NaNO3. The vessels were 22-ml screw-cap vials with polytetrafluoroethylene/silicone septa (Sigma-Aldrich, St. Louis, MO). The argon, from which traces of oxygen were removed (13), was bubbled through the culture for 5 min via a hypodermic needle inserted through the septum.
RESULTS
Metabolic source of N2O3. The endogenous production of nitrosating agents was monitored by measuring the mutator effect of an nfi mutation during anaerobic nitrate/nitrite metabolism. The strains used were derivatives of a lacZ mutant that can revert to the Lac+ phenotype only by an A:TG:C mutation (9). Anaerobic conditions were attained by growing the cultures in completely filled tubes, without a headspace. The medium contained glycerol as an electron donor. For the sake of the controls in which nitrate and nitrite were absent, fumarate was added to all of the cultures as an alternate electron acceptor. After growth to saturation, the cultures were plated on a lactose minimal medium to score the number of Lac+ revertants. To determine the relative importance of HNO2 or NO· as precursors of N2O3 during nitrate/nitrate metabolism, a nirB mutation was introduced into an nfi mutant. The nirB deletion affects the major, inducible nitrite reductase of E. coli, which is a cytoplasmic, NADH-dependent enzyme that produces about two-thirds of the detectable free NO· (8); the remainder is generated by a periplasmic cytochrome c nitrite reductase encoded by the nrf genes. During nitrate metabolism, the nirB mutant is expected to accumulate nitrite and HNO2 and to have a corresponding deficit in the production of NO· (Fig. 1). If nirB increases the mutator effect of nfi, then HNO2 is the cause of nitrosative mutagenesis under these growth conditions; if it decreases the mutator effect, then NO· is the mutagenic agent. The nfi mutant (Fig. 2) demonstrated a nitrate(nitrite)-dependent mutator phenotype, as was previously observed (28). However, the addition of the nirB mutation reduced its mutation frequency almost to the level of the nfi+ strain. The effect of nirB was similar for both nitrate- and nitrite-induced mutagenesis, suggesting that both agents produced DNA damage through a common pathway. The dependence of the mutator effect on nitrite reductase, a potential generator of NO·, implicates NO· rather than HNO2 as the proximate source of N2O3 during nitrate/nitrite metabolism.
An unusual finding within the raw data for Fig. 2, was that of a fairly symmetrical, nonlogarithmic distribution of the data. Within each set of samples, the arithmetic mean was within 5% of the median value, and there were no values differing by more than 2.5 standard deviations from the mean (see Discussion).
Source of O2 for N2O3 formation. The production of the nitrosating agent N2O3 from NO· requires O2, but the cultures were incubated anaerobically. There must have been a point in the experiments when both NO· and O2 were present. There are two possibilities for the source of the O2. The first is that the O2 that was dissolved in the medium at the start of the experiment may not have been fully consumed before significant amounts of NO· accumulated in the sealed tubes. The second possibility is that when the cells were plated, O2 in the air reacted with NO· that had accumulated in the anaerobic cultures. The question was resolved by flushing the cultures with O2-free argon either before or after their growth. The results (Table 2) indicated that removing the dissolved oxygen before growth had no significant effect on the mutation frequency. However, when the cultures were flushed with argon after growth, which would remove NO· before exposure to air, the mutation frequency was reduced to that seen without nitrate (as in Fig. 2). Therefore, nitrosative mutagenesis occurs when the bacteria are shifted from nitrate-dependent to oxygen-dependent respiration, during which accumulated NO· may be oxidized to N2O3.
Mechanism of mutagenic nitrosative deamination by HNO2. HNO2 mutagenesis may be effected by briefly exposing nongrowing E. coli to sodium nitrite at acid pH (31). It has been long assumed that the mechanism of HNO2-induced DNA deamination in intact E. coli is the same as that for phages and free DNA, i.e., it is produced by N2O3 that is derived directly from the condensation of HNO2. However, the results in the present study raise the possibility that NO· may be involved. Theoretically, the nitrite in the acid medium could be reduced enzymatically, yielding NO· as a by-product that is then autooxidized to N2O3. The required acid pH might function to convert the nitrite ions transiently into an uncharged form (molecular HNO2) that might traverse the bacterial membranes more easily.
To determine whether HNO2 mutagenesis is also mediated by NO·, a nirB mutation was again used. The nirB deletion was tested for its effect on the reversion of the trpA58 allele in the nfi mutant, BW1177. This reversion also occurs by an A:TG:C transition. A previous study (21) had found that the HNO2-induced reversion of trpA in this strain is nfi dependent (>99.5%) and occurs with a 16-fold-higher frequency than lacZ reversion in our lacZ nfi mutant. Thus, it should be a sensitive indicator of any effect produced by the nirB deletion. The results (Table 3) indicate that the nirB mutation has little, if any, effect on HNO2-induced mutagenesis, suggesting that most HNO2 mutagenesis is not mediated by NO·. There was an additional reason for choosing the trpA58 mutation indicator strain BW1177 for the present study, apart from its high frequency of mutability by HNO2. It has a different genetic background from our lacZ strains and, for unknown reasons that are now under investigation, it does not appear to be significantly mutagenized by nitrate or nitrite in anaerobic cultures (Table 3). However, these conditions of nitrate respiration and HNO2 treatment were previously shown to produce about an equal number of revertants in the lacZ indicator strain BW1506 (21, 28). This result further supports the conclusion that there are different mechanisms for the production of nitrosative DNA damage by HNO2 and by nitrite metabolism. Therefore, during HNO2 treatment, most of the N2O3 is formed directly from HNO2 rather than through its metabolism to NO·.
DISCUSSION
When mutations accumulate in a growing culture, the progeny of the earliest mutants will predominate. Variations in mutation frequency mainly reflect the different times when the first mutants appeared in each culture. The results usually demonstrate a skewed, logarithmic distribution rather than a symmetrical spread, and there are frequent jackpot values. Consequently, the arithmetic mean and the median values usually do not agree. However, in this and in previous work (28) with nitrate- and nitrite-induced mutagenesis, the results were evenly distributed, suggesting that the mutagenic events occurred at about the same time in each culture, which is consistent with the hypothesis that the lesions were produced when the saturated, NO·-containing cultures were exposed to air.
What is the source of the NO· Unlike the denitrifying bacteria, which reduce nitrite directly to NO·, E. coli appears to reduce nitrite directly to ammonia (2). Nevertheless, several studies (reviewed in reference 8) have documented the production of NO· during nitrate/nitrite metabolism in E. coli. With the aid of a recently available sensitive NO· electrode, it was found that the production of NO· from nitrite by extracts of E. coli could be eliminated by mutations in both the Nrf (periplasmic) and NirB (cytoplasmic) nitrite reductases (8). At the high level of nitrate used in the current study, the expression of Nrf should be repressed, whereas that of NirB should be fully induced (20, 27), thereby explaining why the nirB mutation was effective in preventing most of the nitrosative mutagenesis. However, in early studies with purified NirB, neither NO· nor any other partial reduction product of nitrite was detected (7); hydroxylamine was postulated to be a latent intermediate only because it can serve as a substrate for the enzyme (17). There are no known reactions in E. coli whereby the end product of NirB, ammonia, can be converted to NO·. Therefore, either NO· is a previously undetected latent intermediate that leaks from NirB in very small amounts or else it is the product of a reaction between a cytoplasmic constituent and an NirB-bound intermediate. With the current availability of more sensitive methods of detection, it may be worthwhile to reinvestigate the possibility that traces of NO· may be released from the purified enzyme during the NADH-dependent reduction of nitrite.
There are parallels between oxidative and nitrosative DNA damage. The reactive oxygen species and reactive nitrogen oxides are mostly by-products rather than intermediates of aerobic and anaerobic metabolism, respectively. The reactive oxygen species undergo electron chain reactions in which free radicals generate other free radicals (11). Similarly, N-nitroso derivatives undergo transnitrosation reactions (29). The deleterious nature of the passage from anaerobic nitrate-dependent to oxygen-dependent respiration, as demonstrated in this work, is reminiscent of reperfusion injury, in which reactive oxygen species are produced when mammalian tissues pass from a hypoxic to an aerobic state (3).
Although the present study has demonstrated mutagenesis occurring through a sudden shift in metabolic state, such a shift may not be a necessary prerequisite. Hypoxia rather than anaerobiosis is sufficient for the induction of nitrate and nitrite reductase activities (18). It is possible that when the cell is growing in the presence of nitrate under hypoxic conditions, nitrate and nitrite reductase activities may be sufficiently induced, and there may still be sufficient oxygen present, so that NO· and N2O3 are formed. Thus, nitrosative mutagenesis has been demonstrated in cultures grown to saturation in open flasks (28). There may be many natural environments, perhaps even the human intestine, where these conditions may exist.
ACKNOWLEDGMENTS
The capable technical assistance of Carrie L. Flood and Dawit Seyfe is gratefully acknowledged. I also thank Robert K. Poole and Stephen Spiro for their helpful discussions.
This study was supported by NIH grant ES11163.
Mailing address: Department of Pathology and Laboratory Medicine, Emory University, Whitehead Bldg., Rm. 141, 615 Michael St., Atlanta, GA 30322. Phone: (404) 712-2812. Fax: (404) 727-8538. E-mail: bweiss2@emory.edu.
REFERENCES
Aravind, L., D. R. Walker, and E. V. Koonin. 1999. Conserved domains in DNA repair proteins and evolution of repair systems. Nucleic Acids Res. 27:1223-1242.
Berks, B. C., S. J. Ferguson, J. W. B. Moir, and D. J. Richardson. 1995. Enzymes and associated electron transport systems that catalyse the respiratory reduction of nitrogen oxides and oxyanions. Biochim. Biophys. Acta 1232:97-173.
Bulkley, G. B. 1994. Reactive oxygen metabolites and reperfusion injury: aberrant triggering of reticuloendothelial function. Lancet 344:934-936.
Burney, S., J. L. Caulfield, J. C. Niles, J. S. Wishnok, and S. R. Tannenbaum. 1999. The chemistry of DNA damage from nitric oxide and peroxynitrite. Mutat. Res. 424:37-49.
Cole, J. 1996. Nitrate reduction to ammonia by enteric bacteria: redundancy, or a strategy for survival during oxygen starvation FEMS Microbiol. Lett. 136:1-11.
Cole, J. A., and F. B. Ward. 1973. Nitrite reductase-deficient mutants of Escherichia coli K-12. J. Gen. Microbiol. 76:21-29.
Coleman, K. J., A. Cornish-Bowden, and J. A. Cole. 1978. Purification and properties of nitrite reductase from Escherichia coli K-12. Biochem. J. 175:483-493.
Corker, H., and R. K. Poole. 2003. Nitric oxide formation by Escherichia coli. Dependence on nitrite reductase, the NO-sensing regulator Fnr, and flavohemoglobin Hmp. J. Biol. Chem. 278:31584-31592.
Cupples, C. G., and J. H. Miller. 1986. A set of lacZ mutations in Escherichia coli that allow rapid detection of each of the six base substitutions. Proc. Natl. Acad. Sci. USA 86:5345-5349.
Darwin, A., P. Tormay, L. Page, L. Griffiths, and J. Cole. 1993. Identification of the formate dehydrogenases and genetic determinants of formate-dependent nitrite reduction by Escherichia coli K-12. J. Gen. Microbiol. 139:1829-1840.
Fridovich, I. 1998. Oxygen toxicity: a radical explanation. J. Exp. Biol. 201:1203-1209.
Gates, F. T., and S. Linn. 1977. Endonuclease V of Escherichia coli. J. Biol. Chem. 252:1647-1653.
Gaudu, P., and B. Weiss. 1996. SoxR, a [2Fe-2S] transcription factor, is active only in its oxidized form. Proc. Natl. Acad. Sci. USA 93:10094-10098.
Gennis, R. B., and V. Stewart. 1996. Respiration, p. 217-261. In F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed. ASM Press, Washington, D.C.
Guo, G., and B. Weiss. 1998. Endonuclease V (nfi) mutant of Escherichia coli K-12. J. Bacteriol. 180:46-51.
He, B., H. Qing, and Y. W. Kow. 2000. Deoxyxanthosine in DNA is repaired by Escherichia coli endonuclease V. Mutat. Res. 459:109-114.
Jackson, R. H., J. A. Cole, and A. Cornish-Bowden. 1981. The steady-state kinetics of the NADH-dependent nitrite reductase from Escherichia coli K 12. Nitrite and hydroxylamine reduction. Biochem. J. 199:171-178.
Metheringham, R., and J. A. Cole. 1997. A reassessment of the genetic determinants, the effect of growth conditions and the availability of an electron donor on the nitrosating activity of Escherichia coli K-12. Microbiology 143:2647-2656.
Miller, J. H. 1992. A short course in bacterial genetics. A laboratory manual and handbook for Escherichia coli and related bacteria. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Page, L., L. Griffiths, and J. A. Cole. 1990. Different physiological roles of two independent pathways for nitrite reduction to ammonia by enteric bacteria. Arch. Microbiol. 154:349-354.
Schouten, K. A., and B. Weiss. 1999. Endonuclease V protects Escherichia coli against specific mutations caused by nitrous acid. Mutat. Res. 435:245-254.
Sedgwick, B. 1997. Nitrosated peptides and polyamines as endogenous mutagens in O6-alkylguanine-DNA alkyltransferase deficient cells. Carcinogenesis 18:1561-1567.
Shapiro, R., and S. H. Pohl. 1968. The reaction of ribonucleosides with nitrous acid. Side products and kinetics. Biochemistry 7:448-455.
Suzuki, T., H. Ide, M. Yamada, N. Endo, K. Kanaori, K. Tajima, T. Morii, and K. Makino. 2000. Formation of 2'-deoxyoxanosine from 2'-deoxyguanosine and nitrous acid: mechanism and intermediates. Nucleic Acids Res. 28:544-551.
Taverna, P., and B. Sedgwick. 1996. Generation of an endogenous DNA-methylating agent by nitrosation in Escherichia coli. J. Bacteriol. 178:5105-5111.
Victorin, K. 1994. Review of the genotoxicity of nitrogen oxides. Mutat. Res. 317:43-55.
Wang, H., and R. P. Gunsalus. 2000. The nrfA and nirB nitrite reductase operons in Escherichia coli are expressed differently in response to nitrate than to nitrite. J. Bacteriol. 182:5813-5822.
Weiss, B. 2001. Endonuclease V of Escherichia coli prevents mutations from nitrosative deamination during nitrate/nitrite respiration. Mutat. Res. 461:301-309.
Williams, D. L. H. 1988. Nitrosation. Cambridge University Press, Cambridge, United Kingdom.
Yao, M., Z. Hatahet, R. J. Melamede, and Y. W. Kow. 1994. Purification and characterization of a novel deoxyinosine-specific enzyme, deoxyinosine 3' endonuclease, from Escherichia coli. J. Biol. Chem. 269:16260-16268.
Zimmermann, F. 1977. Genetic effects of nitrous acid. Mutat. Res. 39:127-148.(Bernard Weiss)