The activity of the promoters involved in transcription of the genes (nirS, nirQ and norC) required for anaerobic reduction of nitrite and nitric oxide was investigated in NIR- and NOR-deficient mutants of Pseudomonas aeruginosa. The transcriptional activity of these three promoters was induced by nitrite in a wild-type strain and the activity was low in an nirS mutant. In norCBD and nirQOP mutants, which were expected to accumulate nitric oxide because of a lack of nitric oxide reductase activity, the norC and nirQ promoters showed significantly enhanced activity in promoting transcription relative to the parental strain, even at low nitrite concentrations. These results suggest that the nirQ and norC promoters are regulated by the concentration of endogenous nitric oxide rather than that of nitrite.
Pseudomonas aeruginosa reduces nitrate to dinitrogen by denitrification under anaerobic conditions. Among the enzymes involved in denitrification, nitrite reductase (NIR) catalyses the reduction of nitrite (NO−2) to nitric oxide (NO). NO is further reduced to nitrous oxide (N2O) by nitric oxide reductase (NOR). Expression of these two enzyme activities is thought to be coordinately regulated to avoid the accumulation of toxic NO. The NIR of P. aeruginosa is a cytochrome cd1, and NOR is a cytochrome bc complex which is structurally related to cytochrome oxidases functioning in aerobic respiration . The structural genes for NIR (nirS) and NOR (norCB) are clustered on the chromosome of P. aeruginosa. The gene encoding cytochrome c-551, an electron donor for NIR, and the genes required for biosynthesis of heme d1 are also clustered in this region [3, 4]. The nirQOP operon is located between nirS and norCB. nirQ encodes a putative ATP-binding protein and is reported to be necessary for post-translational activation of NOR in Pseudomonas stutzeri. Homologues of nirQ were found in the gene cluster encoding the ‘green-like’ ribulose-1,5-bisphosphate carboxylase/oxygenases (RubisCOs) in chemoautotrophic bacteria and designated as cbbQ[7, 8]. We have reported that the CbbQ of Pseudomonas hydrogenothermophila mediates post-translational activation of RubisCO . NirQ from P. aeruginosa also was effective in activating RubisCO . nirO and nirP (formerly ORF2 and ORF3, respectively) encode hydrophobic proteins . NirO is similar to subunit III of cytochrome oxidase [1, 5]. nirS and nirQ are oriented in a manner such that divergent transcription occurs. There is an FNR-binding motif in the intergenic promoter region between nirS and nirQ. The norCB genes also carry an FNR-binding motif in their promoter region . Transcription of the nirS, nirQ and norCB genes is under the control of a hierarchical dual regulatory system, ANR and DNR [11, 14]. ANR of P. aeruginosa is closely related to FNR, an oxygen-sensing regulator in Escherichia coli[12, 13]. ANR has 51% amino acid identity with FNR and has four cysteine residues required for sensing oxygen levels, suggesting that ANR is involved in a regulatory mechanism capable of responding to changes in oxygen concentration. Expression of DNR is under the control of ANR in P. aeruginosa and regulation of the denitrification genes by ANR occurs by an indirect mechanism by way of DNR . DNR also belongs to the CRP/FNR-family of transcriptional regulators, but it does not have the above-mentioned cysteine residues and it is phylogenetically distant from FNR, suggesting that the sensory signal for DNR is a factor other than oxygen [11, 14]. A candidate thought to be an effector molecule of DNR is nitrite or one of its metabolites because expression of DNR-dependent promoters requires nitrite [5, 15]. We also reported that DNR-dependent expression of a consensus FNR-dependent promoter was regulated by nitrite, whereas ANR-dependent expression was not . In this work, we investigated the activity of the DNR-dependent promoters in nir and nor mutant strains of P. aeruginosa to identify the role of nitrite and NO in regulation of denitrification gene expression.
2Materials and methods
2.1Strains, plasmids and growth conditions
P. aeruginosa PAO1 was used as a wild-type strain . RM488 , RM495 and RM450  are nirS, norCBD and nirQOP deficient mutants of P. aeruginosa, respectively. Escherichia coli JM109 was used as a host for gene manipulation . Plasmids pHA531, pHA532 and pHA533, which carry nirS::lacZ, nirQ::lacZ and norC::lacZ transcriptional fusions, respectively, were used to assay promoter activity . Bacterial strains were grown in LB medium under aerobic conditions . For assay of β-galactosidase, a synthetic medium was used . When necessary, sodium nitrite or sodium nitrate was added to the medium. Oxygen-limited and semi-aerobic culture conditions were described previously . The concentrations of antibiotics and growth conditions were also described previously .
2.2Recombinant DNA techniques and enzyme assays
Standard DNA techniques were employed as described by Sambrook et al. . Other gene manipulation techniques used were described previously [11, 14]. β-Galactosidase activity in stationary phase cultures grown under oxygen-limited conditions was measured by the standard method . NIR and NOR activity in cells grown under semi-aerobic conditions was measured as described previously .
2.3Construction of the norCBD mutant strain
Strain RM495 was constructed by homologous recombination from strain PAO1 with plasmid pHA495. pHA495 was constructed by insertion of an EcoRI–SphI fragment into pUC119 and replacement of an internal SacI–XhoI fragment (containing the 3′-terminus of norC, norB and the 5′-terminus of norD (formerly ORF6 )), with the tetracycline-resistance gene (tet) of pBR322 (Fig. 1). norD is transcribed together with norCB and is required for anaerobic growth by denitrification (unpublished data). A tetracycline-resistant and carbenicillin-sensitive colony was selected after transformation of strain PAO1 with pHA495 and this mutant was designated as strain RM495. The presence of the norCBD mutation was confirmed by Southern hybridisation analyses (data not shown).
3Results and discussion
3.1Effect of nitrite and NO on expression of the nir and nor genes
The transcriptional activity of the three promoters (nirS, nirQ and norC) (Fig. 1) in cells of the wild-type strain PAO1 grown in the presence of various concentrations of nitrite under oxygen-limited conditions was measured using lacZ transcriptional fusions (Fig. 2). The activity of each of the promoters increased depending on the concentration of nitrite, indicating that expression of these genes is regulated by nitrite or one of its metabolites. It seems reasonable that expression of NIR is regulated by its own substrate. It remains to be determined whether the expression of NOR is regulated by nitrite or by NO produced from nitrite in the periplasm.
In Pseudomonas stutzeri, the level of expression of NOR was found to be very low in the absence of NIR activity, suggesting that NOR expression may be regulated by endogenous NO levels . In the case of a NIR− mutant of Rhodobacter sphaeroides, addition of an NO generator, sodium nitroprusside, to the culture medium stimulated transcription of the nor genes, indicating the involvement of an NO-sensing regulatory system . We investigated the effect of endogenous NO by measuring the promoter activity in nirS and norCBD mutants of P. aeruginosa. In the nirS mutant strain RM488, in which NO was not produced from nitrite because of the lack of NIR activity, the transcriptional activities of the nirS, nirQ and norC promoters were about 1/3, 1/5 and 1/10 of those of strain PAO1, respectively (Fig. 2). These results suggest that either the expression of the nir and nor genes is regulated by endogenous NO levels, or an active NIR enzyme is required for full expression of these genes. We examined the effect of exogenous NO on fusion promoters in strains PAO1 and RM488 by adding NO gas to the headspace of the cultivation vials. The activities were similar to those when equivalent amount of nitrite was added to the medium (data not shown), probably because of the effect of nitrite which was produced by spontaneous oxidation of NO in the presence of oxygen.
The low nirS promoter activity in strain RM488 was an unexpected result. Probably, transcription of nirS is not regulated directly by nitrite. A similar phenomenon was also reported in the case of a NIR− mutant of R. sphaeroides. Since the norC promoter activity was significantly low in strain RM488, it seems that repression of the nirS promoter in this strain might be necessary for coordinate expression of the two enzymes. The sensory signal for induction of nirS and that for induction of nirQ and norC were shown to be different, because the nirS promoter and the nirQ and norC promoters showed different patterns of activity in the norCBD mutant strain RM495 (Fig. 2). The activity of the nirS promoter in strain RM495 was nearly equivalent to that in strain PAO1 in the presence of a low concentration of nitrite. In contrast, the nirQ and norC promoters in strain RM495 showed significantly strong activity in the presence of a low concentration of nitrite and the activity was not greater at high nitrite concentration.
The strong activity of the nirQ and norC promoters at low nitrite concentration could be explained by assuming that the promoters are regulated by endogenous NO levels. In strain RM495, the NO molecules generated from nitrite by the action of NIR are not further metabolised because of the lack of NOR. Probably, NO accumulates in the cells to a level above the threshold for activation of transcription even at low nitrite concentration, although NO could not be detected in the headspace of the incubation vials by gas chromatography (data not shown).
3.2Effect of the nirQOP mutation
The nirS promoter activity in the nirQOP mutant strain RM450 was almost the same as that in the norCBD mutant strain RM495 (Fig. 2A). Transcription directed by the nirQ and norC promoters in strain RM450 was strongly induced at the low nitrite concentration as in the case of strain RM495 (Fig. 2B,C). Thus, the nirQOP mutation resembles the norCBD mutation in terms of activation of the nir and nor genes. Analysis of enzyme activity showed that NOR was not active in strain RM450 (Table 1), in spite of the fact that the norC promoter was active in this strain (Fig. 2C). However, NIR was active in strain RM450 although the activity was low compared to that in PAO1. The difference in NIR activity between strains PAO1 and RM450 was a reflection of the difference in promoter activity (Fig. 2A), indicating that the nirQOP mutation had no effect on the translational and post-translational regulation of nirS. It seems likely that the effect of the nirQOP mutation is similar to that of the norCBD mutation because of the lack of NOR activity.
Table 1. Effect of the nirQOP mutation on NIR and NOR activity
In vivo activity in reduction ofa
aThe activity is expressed as nmol of N oxides reduced per minute per mg of protein with lactate used as the electron donor. The values are means of more than four independent experiments. Cells were grown in LB medium containing 20 mM sodium nitrate under semi-aerobic conditions.
We have reported that CbbQ, which is similar to NirQ, activates RubisCO post-translationally . NirQ of P. aeruginosa can also activate RubisCO of P. hydrogenothermophila. Probably, NirQ-type proteins are effective in mediating post-translational activation of certain enzymes and it seems likely that the physiological role of NirQ in P. aeruginosa is activation of NOR. Regulation of the nirQ promoter by NO rather than nitrite may occur because products of genes in the nirQ operon are required for activation of NOR, but not for activation of NIR.
The nirS and nirQ operons are divergently transcribed, suggesting that there is an interrelation between these two promoter activities. NIR and NOR enzyme activities should be coordinately expressed in order to avoid accumulation of toxic NO. Assuming that NirQ is a post-translational activator for NOR, well-balanced expression of the nirS and nirQ promoters may be advantageous for fine-tuning of the activities of NIR and NOR.