The influence of nitrate and nitrite on growth of Corynebacterium glutamicum under aerobic conditions in shake flasks was analysed. When dissolved oxygen became limiting at higher cell densities, nitrate was reduced almost stoichiometrically to nitrite by nitrate reductase (NarGHJI). The nitrite concentration also declined slowly, presumably as a result of several reactions including reduction to nitric oxide by a side-activity of nitrate reductase. The flavohaemoglobin gene hmp was most strongly upregulated (19-fold) in the presence of nitrite. Hmp is known to catalyse the oxygen-dependent oxidation of nitric oxide to nitrate and, in the absence of oxygen, with a much lower rate the reduction of nitric oxide to nitrous oxide. A Δhmp mutant showed strong growth defects under aerobic conditions in the presence of nitrate, nitrite and the NO-donating reagent sodium nitroprusside, but also under anaerobic nitrate-respiring conditions. Therefore, Hmp is likely to be responsible for nitric oxide conversion to either nitrate or nitrous oxide in C. glutamicum. The results suggest that a cyclic nitrate–nitrite conversion takes place in C. glutamicum under microaerobic conditions.
Corynebacterium glutamicum is a Gram-positive soil bacterium that has been used for the production of amino acids for decades, in particular l-glutamate and l-lysine, and has become a favourite model organism in industrial biotechnology (Eggeling & Bott, 2005; Burkovski, 2008; Yukawa & Inui, 2013). As suggested by the presence of narGHJI genes in the genome, it was recently confirmed in two studies that C. glutamicum can grow anaerobically by nitrate respiration, albeit poorly (Nishimura et al., 2007; Takeno et al., 2007). In this process, nitrate is reduced to nitrite by the membrane-bound multisubunit nitrate reductase NarGHJI, as shown by the fact that mutants with defective or absent narG, narH or narJ genes were unable to convert nitrate to nitrite. Nitrate reductase uses reduced menaquinone (MKH2) as electron donor. As two protons of MKH2 are released on the outer side of the membrane and two protons are consumed within the cell to form water, the reaction catalysed by nitrite reductase contributes to the generation of proton motive force and thus to energy conservation (Bott & Niebisch, 2003).
The nitrate reductase structural genes are located in the narKGHJI operon together with the gene for a putative nitrate/nitrite antiporter (NarK), which is involved, but not essential for nitrate respiration as shown by the phenotype of a ΔnarK mutant (Nishimura et al., 2007; Takeno et al., 2007). Regarding the regulation of the narKGHJI operon, the results of the two studies are to some extent contradictory. Takeno and co-workers reported that the ATCC 13032 wild type shows a specific nitrate reductase activity of 11–25 nmol min−1 mgprotein−1 even in cells cultured under atmospheric conditions, independent from the presence of nitrate (Takeno et al., 2007). No activity was found in mutants lacking narG or narJ. The authors concluded that the narGHJI operon is not regulated by either oxygen or nitrate levels. In contrast, Nishimura and co-workers reported that the expression of narKGHJI was strongly regulated by oxygen and nitrate. They found low mRNA levels in aerobically grown cells, irrespective of the presence of nitrate and a 10-fold higher level under anaerobic conditions in the absence of nitrate, which was further increased twofold to fourfold when nitrate was present. Therefore, they concluded that the narKGHJI genes are induced under anaerobiosis and additionally activated by nitrate (Nishimura et al., 2007). In a subsequent study, the transcriptional regulator, ArnR, was identified, which represses the narKGHJI operon under aerobic conditions (Nishimura et al., 2008).
It has been demonstrated that under nitrate-respiring conditions, nitrite accumulates in the medium under anaerobic and oxygen-limited conditions and is not consumed by C. glutamicum (Nishimura et al., 2007; Takeno et al., 2007). This behaviour is in agreement with the genome analysis, which suggested that C. glutamicum does not possess proteins with nitrite reductase activity (Bott & Niebisch, 2003). Nitrite is known to have antimicrobial properties and a possible reason could be its interaction with iron-sulphur clusters (Reddy et al., 1983). Moreover, under acidic conditions, nitrous acid can yield nitric oxide (NO) and other nitrogen oxides, which are also harmful to cells (Bowman et al., 2011). Acidic conditions can occur in C. glutamicum under oxygen-limited conditions, when the cells form organic acids, in particular l-lactate (Koch-Koerfges et al., 2012). The growth-inhibitory properties of nitrite and reactive nitrogen species formed from nitrite are likely reasons for the poor growth of C. glutamicum under anaerobic conditions by nitrate respiration.
In this study, we analysed the influence of nitrate and nitrite on aerobic growth and global gene expression of C. glutamicum as well as the role of the nitrate reductase in nitrate and nitrite metabolism. In the course of these studies, the flavohaemoprotein Hmp was identified and its role for coping with nitrite stress was analysed.
Materials and methods
Bacterial strains and culture conditions
A list of all bacterial strains and plasmids used in this study can be found in Table 1. Escherichia coli was cultivated at 37 °C and 170 r.p.m. in 5 mL LB medium (10 g L−1 bactotryptone, 5 g L−1 yeast extract, 5 g L−1 NaCl) supplemented with 50 μg mL−1 kanamycin when appropriate. C. glutamicum strains were precultivated in 5 mL brain heart infusion medium with 4% (w/v) glucose for 8 h at 30 °C and 170 r.p.m. A second preculture was performed in CGXII minimal medium (Keilhauer et al., 1993) containing 4% (w/v) glucose as carbon and energy source and the indicated concentrations of potassium nitrite, potassium nitrate or sodium nitroprusside (SNP) as supplement (aerobic cultivations). Additionally, the medium was supplemented with 30 mg L−1 3,4-dihydroxybenzoate as iron chelator. Precultures were incubated in 500-mL baffled shake flasks with 20 mL medium at 30 °C and 130 r.p.m. overnight. The second preculture was inoculated with 1 mL of the first preculture. The main cultures for aerobic growth were performed in 500-mL baffled shake flasks at 30 °C and 130 r.p.m. containing 50 mL of the same medium as the second preculture. Main cultures were inoculated to an optical density at 600 nm (OD600 nm) of 1 with cells washed twice with sterile 0.9% (w/v) sodium chloride solution. For anaerobic cultivation, 50 mL CGXII medium with 1.8% (w/v) glucose and 30 mM potassium nitrate as final electron acceptor was prepared in a 60-mL serum bottle closed air-tight with a butyl rubber stopper, followed by flushing the bottle with sterile nitrogen gas for 30 min. 300 μL of a sterile 60 g L−1l-cysteine solution was added with a syringe immediately before inoculation with cells washed twice with sodium chloride solution. Cells were grown at 30 °C with gentle stirring (50 r.p.m.). Sampling was carried out aseptically under constant flushing with nitrogen. When appropriate, 25 μg mL−1 kanamycin was added to the media.
Table 1. Strains, plasmids and oligonucleotides used in this study
KanR, pK19mobsacB derivative containing an overlap- extension PCR product covering the up- and downstream regions of hmp
KanR, pK19mobsacB derivative containing an overlap- extension PCR product covering the up- and downstream regions of narG
Amplification of hmp upstream region, XmaI restriction site
Amplification of hmp upstream region
Amplification of hmp downstream region
Amplification of hmp downstream region, XmaI restriction site
Control of hmp deletion by colony PCR
Control of hmp deletion by colony PCR
Amplification of narG upstream region, XbaI restriction site
Amplification of narG upstream region
Amplification of narG downstream region
Amplification of narG downstream region, EcoRI restriction site
Control of narG deletion by colony PCR
Control of narG deletion by colony PCR
Amplification of hmp for expression plasmid pEKEx2-hmp, includes PstI restriction site and ribosome binding site
Amplification of hmp for expression plasmid pEKEx2-hmp, BamHI restriction site
Determination of nitrite concentrations
Nitrite concentrations in the culture medium were determined by a Griess reaction as described previously (Green et al., 1982). One milliliter of appropriately diluted culture supernatant was mixed with 40 μL Griess reagent and incubated for 10 min at room temperature. Afterwards, the absorption at 540 nm was measured in a spectrophotometer against distilled water with Griess reagent as a blank. Nitrite concentrations were calculated from a calibration curve determined with 1 mL 0.4 to 5.2 μM nitrite solution each. The Griess reagent consists of 1 part 0.1% N-(1-naphthyl)ethylenediamine dihydrochloride in distilled water plus 1 part 1% sulphanilamide in 5% H3PO4. After mixing, the solution was used within 12 h and kept chilled.
Nitrite formation and consumption rates were determined similar to glucose uptake rates as described previously (Koch-Koerfges et al., 2012).
All enzymes used for restriction, dephosphorylation or ligation of DNA were obtained from Roche Diagnostics (Mannheim, Germany) or New England Biolabs (Frankfurt am Main, Germany) and used as described by the manufacturer. Plasmid DNA was isolated with the QIAprep Spin Miniprep kit from Qiagen (Hilden, Germany) according to the manufacturer's instructions.
Construction of Δhmp, ΔnarG and ΔnarGΔhmp deletion mutants
The in-frame deletion mutants, ΔnarG and Δhmp of C. glutamicum were constructed by the method described previously (Niebisch & Bott, 2001) using plasmids pK19mobsacB-Δhmp and pK19mobsacB-ΔnarG, respectively. For construction of these plasmids, the upstream region of hmp or narG was amplified by PCR using the oligonucleotides ΔXXX_1 and ΔXXX_2 and the downstream region using oligonucleotides ΔXXX_3 and ΔXXX_4 (Table 1). The two DNA fragments were fused by overlap-extension PCR using oligonucleotides ΔXXX_1 and ΔXXX_4. The resulting PCR product was cut with XmaI in the case of hmp, or XbaI and EcoRI in the case of narG, and cloned into pK19mobsacB cut with the same restriction enzymes. The correctness of the cloned fragment was confirmed by DNA sequence analysis. The first and second homologous recombination events were performed as described (Niebisch & Bott, 2001). By colony PCR using the oligonucleotides hmp_out_fw and hmp_out_rv (Table 1), 25 kanamycin-sensitive and sucrose-resistant clones were analysed, one of which contained the desired chromosomal hmp deletion, whereas 24 clones were found to be wild type. By colony PCR using the oligonucleotides narG_out_fw and narG_out_rv (Table 1), 22 kanamycin-sensitive and sucrose-resistant clones were analysed, 8 of which contained the desired chromosomal narG deletion, whereas 14 clones were found to be wild type. A double deletion mutant (ΔnarGΔhmp) was created by deletion of the hmp gene in the ΔnarG mutant, as described above. No growth defect compared to the wild type was observed for strains ΔnarG and ΔnarGΔhmp during aerobic cultivation in glucose minimal medium (data not shown).
Construction of the expression plasmid pEKEx2-hmp
The hmp gene was amplified from chromosomal DNA of C. glutamicum ATCC 13032 with the oligonucleotides PstI_RBS_hmp_f and BamHI_hmp_r. The former one introduced a PstI restriction site and a ribosomal binding site (TATACTGCAGAAGGAGATACCCCTTG ATCGGTTCCACCCA), the latter one a BamHI restriction site. The 1194 bp PCR product was cut with PstI and BamHI and cloned into the expression plasmid pEKEx2 (Eikmanns et al., 1991) cut with the same enzymes. The correctness of the cloned DNA fragment was confirmed by DNA sequencing.
Global gene expression analysis
Global gene expression analysis of C. glutamicum was performed by DNA microarray analysis as described previously (Frunzke et al., 2008). The wild type was cultivated aerobically in CGXII glucose minimal medium either in the presence or in the absence of 10 mM nitrite. Cells in the early exponential growth phase (OD600 nmc. 5) were harvested, and the total RNA was extracted as described (Möker et al., 2004). cDNA labelled with fluorescent dyes was obtained as described previously and hybridized to custom-made DNA microarrays (Operon, Cologne, Germany). Expression ratios were calculated after signals were filtered for signal-to-noise ratios of ≥ 3. For details on array coverage and experimental procedures, see Frunzke et al., 2008. Four biological replicates were performed.
Results and discussion
Influence of nitrate and nitrite on aerobic growth of C. glutamicum and role of nitrate reductase NarGHJI in nitrite formation and degradation
To probe the influence of nitrate and nitrite on the growth of C. glutamicum under aerobic conditions, the wild type was cultivated in CGXII minimal medium with 4% (w/v) glucose supplemented with different concentrations of potassium nitrate (5–500 mM) or potassium nitrite (3–100 mM). Both nitrate and nitrite decelerated growth in a concentration-dependent manner. The presence of 50, 100 or 500 mM nitrate reduced the growth rate from 0.41 ± 0.01 h−1 to 0.33 ± 0.01 h−1, 0.32 ± 0.01 h−1 and 0.29 ± 0.01 h−1, respectively, but did not influence the maximal OD600 nm of the cultures (data not shown). The presence of 10 and 25 mM nitrite reduced the growth rate from 0.41 h−1 to 0.29 ± 0.01 h−1 or 0.24 ± 0.01 h−1, respectively, while the presence of 100 mM nitrite completely inhibited growth (data not shown). These results demonstrate that, as expected, nitrite is much more toxic to C. glutamicum cells than nitrate. C. glutamicum was also cultivated with the addition of 5–500 mM potassium chloride. The resulting influence on the growth rate was much weaker than the one caused by nitrate, showing that the effects caused by nitrate were not simply due to osmotic stress.
In a next series of growth experiments, we analysed the fate of 100 mM nitrate and of 25 mM nitrite during aerobic cultivation in more detail. For the interpretation of the results, it is important to keep in mind that the oxygen conditions change significantly during the applied cultivation conditions: in the first 7–10 h, the dissolved oxygen concentration (DO) decreases from 100% air saturation to below 5%, where it stays until 40 h or later, when all carbon substrates have been oxidized, and then returns to full saturation (Koch-Koerfges et al., 2012). As shown in Fig. 1a and b, nitrate was converted almost stoichiometrically to nitrite when the culture became oxygen-limited after c. 7–10 h. The maximal rate of nitrite formation observed was 37.5 ± 1.5 nmol min−1 mgCDW−1. Some nitrite formation in the first hours of cultivation was presumably due to the fact that the cells used for inoculation contained nitrate reductase activity as they were harvested in the oxygen-limited phase. Nitrite formation could be completely abolished by the deletion of the narG gene (Fig. 1b). The ΔnarG mutant showed a slightly improved growth rate (0.36 ± 0.01 h−1) compared with the wild type (0.32 ± 0.01 h−1) in the presence of 100 mM nitrate, indicating that the growth defect caused by nitrate is due to nitrite or products formed from nitrite. Interestingly, a significant fraction (60 mM) of the 90 mM nitrite that had been formed after about 20 h disappeared within the next 100 h of cultivation (Fig. 1b) with a rate of 4.4 ± 1.9 nmol min−1 mgCDW−1. As the DO varies during cultivation as described above, nitrite conversion can involve both oxidative and reductive reactions.
In the wild-type cultures containing 25 mM nitrite, nitrite disappeared completely within 100–120 h (Fig. 1d). Remarkably, in the ΔnarG mutant, the rate of nitrite consumption within the first c. 10 h was comparable with that of the wild type leading to a net consumption of c. 1.5 mM nitrite, but was higher in the period when the culture became oxygen-limited (wt: 0.25 ± 0.18 nmol min−1 mgCDW−1 vs. ΔnarG: 0.98 ± 0.46 nmol min−1 mgCDW−1), leading to a complete nitrite disappearance within 60 h (Fig. 1d). As discussed below, this behaviour can be explained by the assumption of a cyclic nitrate/nitrite conversion cycle under microaerobic conditions, which is interrupted in the ΔnarG mutant. Like for nitrate, the ΔnarG mutant also showed a slightly improved growth rate (0.26 ± 0.01 h−1) compared to the wild type (0.24 ± 0.01 h−1) in the presence of 25 mM nitrite (Fig. 1c).
Influence of nitrite on global gene expression of C. glutamicum
To identify proteins involved in nitrite conversion, the influence of nitrite on global gene expression was studied using DNA microarrays. The mRNA levels of cells grown aerobically in the presence of 10 mM nitrite were compared with those of cells cultivated in the absence of nitrite in four biological replicates. The nitrite concentration was chosen to minimize growth defects. The mRNA was isolated from cells at an OD600 nm of c. 5, when oxygen was not yet limiting. As shown in Table 2 genes showed greater than or equal to twofold increased mRNA levels and four genes had greater than or equal to twofold decreased mRNA levels. The latter ones encode proteins of various functions, for example in l-lactate transport and oxidation (Stansen et al., 2005). A conclusive interpretation for the downregulation of these genes in the presence of nitrite cannot be provided.
Table 2. Genes with an altered expression level in the presence of 10 mM nitrite
mRNA ratio ± 10 mM nitrite
The wild type was cultivated aerobically in glucose minimal medium with or without 10 mM nitrite. RNA was isolated from cells in the exponential growth phase and used for DNA microarray analysis as described in 'Materials and methods'. Genes whose mean mRNA ratio calculated from four independent experiments was changed at least twofold and whose P-value was ≤ 0.05 are shown. In addition, the mRNA ratios of the genes encoding nitrate reductase, a putative nitrate/nitrite antiporter and the transcriptional regulator ArnR are also shown.
znuA1, ABC-type Mn2+/Zn2+ transport system, secreted lipoprotein
znuC1, ABC-type Mn2+/Zn2+ transport system, ATPase component
znuB1, ABC-type Mn2+/Zn2+ transport system, permease component
nadS, Cysteine sulphinate desulfinase/cysteine desulphurase or related enzyme
nadC, quinolinate phosphoribosyltransferase
nadA, quinolinate synthetase
ndnR, transcriptional repressor of NAD de novo biosynthesis genes (ndnR-nadA-nadC-nadS operon), NrtR family
Heavy metal binding transport protein
prpD2, 2-methylcitrate dehydratase
prpB2, 2-methylisocitrate lyase
prpC2, 2-methylcitrate synthase
ramB, transcriptional regulator involved in acetate metabolism
Extremely conserved hypothetical protein
copA, copper-exporting ATPase
arnR, transcriptional regulator of narKGHJI and hmp
The genes upregulated in the presence of 10 mM nitrite code for Znu, an ABC transporter for the import of Mn2+ or Zn2+ (Schröder et al., 2010), the iron storage protein ferritin (Wennerhold & Bott, 2006), several genes involved in NAD(P)H biosynthesis (Teramoto et al., 2010), two proteins involved in copper export (Schelder et al., 2011), genes involved in the methylcitrate cycle (Claes et al., 2002) and four proteins of unknown function. Improving the conditions for iron storage, copper export, and zinc import (Smith et al., 2009) probably serves as means to reduce oxidative and nitrosative stress. The expression of the narKGHJI operon and of the arnR gene was 1.3–1.7-fold upregulated, suggesting that nitrite has some influence on their transcription. Two additional transcriptional regulators showed an increased expression in the presence of 10 mM nitrite, the global regulator RamB (Gerstmeir et al., 2004; Auchter et al., 1967), and the regulator of NAD de novo biosynthesis genes NdnR (Teramoto et al., 2010). The increased expression of the ndnR-nadA-nadC-nadS operon indicates that the repressor NdnR is inactivated by nitrite or decreased NAD concentrations (Teramoto et al., 2010, 2012). A link of nitrite and NAD(P)H metabolism would be reasonable, as the most strongly upregulated (19-fold) gene hmp (cg3141) encodes an NAD(P)H-dependent flavohaemoprotein.
The C. glutamicum Hmp protein has 32% sequence identity to Hmp of E. coli (also known as flavoHb), which was shown to be involved in the detoxification of nitric oxide (Poole & Hughes, 2000). Under aerobic conditions, E. coli Hmp catalyses the reduction in NO to nitrate (NO + O2 + 0.5 NAD(P)H → + 0.5 NAD(P)+ + 0.5 H+). This reaction can proceed via two different chemical mechanisms, the ‘dioxygenation’ mechanism, in which O2 binds first to Hmp, and the ‘nitrosylation’ mechanism, in which NO binds first to Hmp. There is disagreement on which of the two mechanisms is the physiologically relevant one (Forrester & Foster, 2012; Hausladen & Stamler, 2012). Under anaerobic conditions, Hmp of several organisms was reported to detoxify NO by NAD(P)H-dependent reduction to N2O (2 NO + NAD(P)H + H+ → N2O + H2O + NAD(P)+), but the physiological relevance is discussed controversially, as the rate of the anaerobic reaction is only about 1% of the aerobic one (Kim et al., 1999; Mills et al., 2001; Gardner & Gardner, 2002; Forrester & Foster, 2012).
The hmp gene of C. glutamicum was previously shown to be repressed under aerobic conditions by ArnR (Nishimura et al., 2008), similar to the nar operon, and to be upregulated under anaerobic nitrate-respiring conditions (Nishimura et al., 2011). In a ΔarnR mutant grown under aerobic conditions in the absence of nitrate or nitrite, the mRNA level of the narKGHJI genes was increased threefold to sixfold compared with the wild type, whereas that of the hmp gene was increased 29-fold (Nishimura et al., 2008).
A stimulatory effect of nitrite on hmp expression was also observed in E. coli and Bacillus subtilis. In E. coli, 5 mM nitrite stimulated expression of a single-copy hmp-lacZ fusion 33-fold under anaerobic conditions, but only 1.6-fold under aerobic conditions (Poole et al., 1996). It was speculated that the actual signal molecule might be NO (Poole et al., 1996). In B. subtilis, expression of a hmp-lacZ fusion was also strongly stimulated by nitrite under aerobic and anaerobic conditions (Lacelle et al., 1996). However, in contrast to C. glutamicum (see below), a hmp-defective strain of B. subtilis showed the same growth behaviour as the wild type during anaerobic growth with nitrate (Lacelle et al., 1996).
Phenotype of a Δhmp mutant during aerobic growth in the presence of nitrate, nitrite and SNP
To evaluate the role of Hmp in the response of C. glutamicum towards nitrate, nitrite and nitrosative stress, a deletion mutant (Δhmp) was constructed. Whereas no differences in growth were observed during aerated cultivation in standard glucose minimal medium, a strong growth defect of the Δhmp mutant became apparent when the medium was supplemented with 100 mM nitrate (Fig. 1a) or 25 mM nitrite (Fig 1c). The mutant showed a prolonged lag phase as well as a decreased growth rate and reached only about half of the optical density as the wild type. In the presence of 100 mM nitrate, the kinetics of nitrite formation was similar for the Δhmp mutant and the wild type, but the Δhmp mutant converted nitrite much more slowly and only to about 50% of the extent of the wild type, showing that nitrite conversion by C. glutamicum involves Hmp, but also Hmp-independent reactions. The growth defects of the Δhmp mutant could be reversed by plasmid-based expression of hmp using pEKEx2-hmp (data not shown).
The phenotype of the Δhmp mutant gave rise to the question whether overexpression of hmp could improve the resistance of C. glutamicum towards nitrite. Therefore, pEKEx2-hmp was transferred into the wild type, and the recombinant strain was cultivated under aerobic conditions in glucose minimal medium containing different nitrite concentrations (5–20 mM) and different IPTG concentrations (0.05–0.1 mM) for induction. However, irrespective of the conditions used, hmp overexpression did not lead to a better growth in the presence of nitrite (data not shown), indicating that the native Hmp level in nitrite-stressed cells is already sufficiently high. The overproduction of Hmp was verified at the protein level. A 42-kDa protein band with an increased abundance in cells with plasmid pEKEx2-hmp cultivated in the presence of IPTG was identified by peptide mass fingerprinting as Hmp (data not shown).
As Hmp of E. coli was shown to detoxify nitric oxide and hmp deletion mutants of other organisms were hypersensitive towards NO-donating agents (Membrillo-Hernandez et al., 1999; Moore et al., 2004), we tested C. glutamicum Δhmp with respect to its sensitivity towards SNP, a substance reported to function as an NO donor. The presence of 20 mM SNP caused a decrease in the growth rate of the wild type from 0.41 to 0.29 ± 0.01 h−1, whereas the maximal optical density was not affected. In the case of the Δhmp mutant, the presence of 20 mM SNP led to a prolonged lag phase and the final optical density was decreased by 45% compared to the wild type grown under these conditions (Fig. 2). Thus, the Δhmp mutant was more sensitive towards SNP, indicating that Hmp is involved in the SNP-triggered NO stress response of C. glutamicum under aerobic conditions.
Phenotype of ΔnarGΔhmp double mutant during aerobic growth in the presence of nitrate and nitrite
To determine the phenotype of a strain lacking both nitrate reductase and Hmp, we constructed a ΔnarG Δhmp double mutant. In the presence of 100 mM nitrate, the double mutant showed the same growth rate as the wild type, but reached a lower maximal OD600 nm (Fig. 1a) and nitrate reduction to nitrite was abolished (Fig. 1b). In the presence of 25 mM nitrite, the double mutant reached a similar growth rate as the Δhmp mutant (Δhmp = 0.20 ± 0.01 h−1; ΔnarGΔhmp = 0.21 ± 0.01 h−1), but a much higher maximal OD600 nm (Fig. 1c). In the initial phase of cultivation, when oxygen was not limited, the double mutant showed no significant nitrite consumption, however, when oxygen became limiting after about 10 h, the kinetics of nitrite consumption of the double mutant was faster (0.76 ± 0.35 nmol min−1 mgCDW−1) than that of the Δhmp mutant (0.35 ± 0.22 nmol min−1 mgCDW−1) and the wild type (0.25 ± 0.18 nmol min−1 mgCDW−1), but slower than that of the ΔnarG mutant (0.98 ± 0.46 nmol min−1 mgCDW−1). The latter results again suggest that Hmp is involved in nitrite decomposition.
Phenotype of a Δhmp mutant during anaerobic growth by nitrate respiration
The hmp gene was shown to be strongly upregulated during anaerobic growth with nitrate (Nishimura et al., 2011). We therefore analysed the growth of the Δhmp mutant under these conditions. As shown in Fig. 3, the absence of Hmp caused a strong inhibition of anaerobic growth with nitrate; however, it did not prevent the reduction of nitrate to nitrite. Nitrate reduction by nongrowing cells was also reported in a previous study (Takeno et al., 2007). Under anaerobic conditions, the role of Hmp is presumably the reduction in NO formed in traces by nitrate reductase to N2O, thereby converting a very toxic compound to one which is much less toxic (Fig. 4b).
In this study, we analysed the influence of nitrate and nitrite on growth of C. glutamicum. As mentioned before, it is important to keep in mind that the DO changes during the applied aerobic cultivation conditions: after 7–10 h, the DO has fallen below 5%, where it stays until 40 h or later, when all carbon substrates have been oxidized, and then returns to full saturation (Koch-Koerfges et al., 2012). We assume that in the phase of microaerobic conditions, both reductive and oxidative reactions can occur in parallel and interpret our data under this premise. The fact that nitrate is converted almost stoichiometrically to nitrite in the ≤ 5% DO phase is in accord with the results of Takeno and co-workers (Takeno et al., 2007) and indicates that the repressor ArnR is inactivated under these conditions, thereby inducing expression of the nitrate reductase operon narKGHJI as well as of hmp. The fact that in our DNA microarray experiments a strong induction of hmp expression was observed in cells harvested when oxygen was above 5% DO can also be interpreted by the assumption that ArnR is activated before cells enter strong oxygen limitation. Alternatively, ArnR repression might be overcome by a yet unknown transcriptional activator or inactivation of ArnR cannot only be accomplished by oxygen limitation, but also by the presence of nitrate, nitrite or NO formed from nitrite. The inability of a ΔnarG mutant to reduce nitrate to nitrite is in agreement with the previous studies and shows that this reaction is exclusively catalysed by the nitrate reductase NarGHJI in C. glutamicum (Nishimura et al., 2007; Takeno et al., 2007).
The observation that nitrite formed from nitrate or added to the medium disappeared, leads to the question for the responsible reactions and enzymes. In the genome of C. glutamicum, none of the genes has been annotated to encode an enzyme involved in nitrite conversion. In particular, no genes encoding either an assimilatory or a dissimilatory nitrite reductase were identified (Bott & Niebisch, 2003). Our microarray data showed that the hmp gene is most strongly induced in the presence of nitrite and the phenotypes of the Δhmp and ΔnarGΔhmp mutants suggest that Hmp is involved in nitrite conversion. Hmp has been reported to have weak nitrite reductase activity resulting in NO formation, but the turnover rate is only 0.01–0.4% of the rate observed for NO oxidation to nitrate (Gardner & Gardner, 2002). A second candidate for nitrite conversion is the nitrate reductase NarGHJI. The homologous enzyme from E. coli and other enterobacteria was reported to have a weak nitrite reductase activity by which NO is formed. This activity is strongly inhibited by nitrate and the rate only about 0.03–0.1% of the nitrate reduction rate (Ji & Hollocher, 1988, 1989). Similar results were also obtained for the nitrate reductase of Salmonella typhimurium (Gilberthorpe & Poole, 2008). Also, plant nitrate reductases were shown to catalyse nitrite reduction to nitric oxide (Yamasaki & Sakihama, 2000). Although both NarGHJI and Hmp can be involved in nitrite consumption, alternative reactions must exist as the ΔnarGΔhmp double mutant was still able to consume nitrite (Fig. 1d). Acidification caused by the formation of l-lactate and acetate could contribute to nitrite decomposition, as mentioned in the introduction. Beyond that, nitrite could be degraded by the interaction with, for example, reactive oxygen species generated during aerobic growth such as H2O2, which could yield nitrate and water ( + H2O2 → + H2O) or with primary amines (Castellani & Niven, 1955; Klebanoff, 1993). The latter reactions most likely represent nonenzymatic reactions.
The reactions described above could form a nitrate–nitrite cycle under microaerobic conditions, in which nitrate is rapidly reduced by nitrate reductase to nitrite, which is subsequently slowly converted, either by reduction to NO or by re-oxidation to nitrate. NO is oxidized to nitrate by Hmp (Fig. 4). In a Δhmp mutant, NO will accumulate, explaining the growth defects of this mutant in the presence of nitrate and nitrite. In a ΔnarG mutant, this cycle would be interrupted, as nitrate reduction to nitrite is exclusively catalysed by NarGHJI, which can explain the increased nitrite consumption of the ΔnarG mutant shown in Fig. 1d. In summary, our results have shed light on the role of nitrate reductase and Hmp in nitrate and nitrite metabolism of C. glutamicum and shown the possibility of this species to cope with nitrite under microaerobic conditions despite the absence of a nitrite reductase.
Financial support by the Bundesministerium für Bildung und Forschung (BMBF) within the GenoMik-Transfer project ‘FlexFit’ (grant 0315589A) is gratefully acknowledged. We would like to thank Andrea Michel for advice on anaerobic cultivation of C. glutamicum.
L.P. and A.K.-K. contributed equally to this work.