Regulation of Carbon and Nitrogen Utilization by CbrAB and NtrBC Two-Component Systems in Pseudomonas aeruginosa

Bacterial strains, media, and growth conditions.The bacterial strains and plasmids used in this study are listed in Table 1. Luria-Bertani (LB) and 2× tryptone-yeast extract media (33) were used for strain construction with the following supplements as indicated: ampicillin at 100 μg/ml, carbenicillin at 200 μg/ml, gentamicin at 50 μg/ml, tetracycline at 10 μg/ml for E. coli or 100 μg/ml for P. aeruginosa, and 5-bromo-4-chloro-3-indolyl-β-d-galactoside at 0.03% (wt/vol). Minimal medium P (MMP) supplemented with indicated carbon sources at 20 mM and nitrogen sources at 5 mM was used for the growth of P. aeruginosa (10).

Growth phenotype analysis.Phenotype microarray (PM; Biolog Co.) plates were used to analyze utilization of carbon, nitrogen, sulfate, or phosphate sources. Ammonium was provided as the sole nitrogen source when carbon sources were tested, and succinate was provided as the sole carbon source when nitrogen sources were tested. PM experiments were conducted according to the manufacturer's instruction. In short, fresh overnight cells from an agar plate were suspended in the inoculating fluid and the turbidity was adjusted to 85%. The cells were inoculated into PM plates and incubated at 37°C for 24 h for carbon plates and 48 h for nitrogen plates. Cell growth was reflected by the development of purple coloration as monitored and recorded by OmniLog PM Station and PM Kinetic (Biolog). Further information on the compounds tested with the PM kit can be found at the Biolog website.

Isolation of the suppressor mutants.Cells were grown in the LB medium overnight and harvested by centrifugation at 3,000 × g for 10 min. The cell pellet was washed twice, suspended, and spread onto MMP plates supplemented with 20 mM arginine or histidine as the sole source of carbon and nitrogen. The suppressor mutants that showed visible colony sizes after incubation at 37°C for 48 h were isolated.

Genomic library construction and screening.Genomic DNA of P. aeruginosa Arg18 was subjected to partial digestion with Sau3AI, and the DNA fragments of 4 to 20 kb were size fractionated by agarose gel electrophoresis and subjected to cloning into the dephosphorylated BamHI site of pUCP18 (34). After transformation into E. coli DH5α, the genomic library was further enriched by agarose gel electrophoresis to separate recombinant plasmids from the vector. The library was then introduced into PAO5100 by transformation (6), and the transformants were screened onto MMP plates supplemented with nutrients as indicated.

Construction of lacZ fusions.DNA fragments containing the glnA and hutU promoter regions were PCR amplified from the genomic DNA of PAO1 with the following synthetic oligonucleotide primers: 5′-ACGAT AAGCTT CAT GCG AGG GAC GCGAA-3′ and 5′-CTT GTA CGA CAT GCT GTC CTC CAG GAGG-3′ for glnA; 5′-GAC TAC AAGCTT CAG GCG CAG GGC TGG-3′ and 5′-CGG GGT GGT CAC GGC AGG CTC CTG GGC-3′ for hutU. The PCR products were purified from a 1% (wt/vol) agarose gel, digested by restriction endonuclease HindIII, and ligated to the HindIII and SmaI sites of the translational fusion vector pQF52 (26). The resulting plasmids, pSA500 and pSU500, contain the entire upstream intergenic sequence of glnA and hutU and the 5′ end of the coding sequence fused in-frame to the eighth codon of the lacZ gene in the vector. The nucleotide sequences of the resulting constructs were verified by nucleotide sequence determination.

Construction of mutant strains and plasmids.To make the ntrBC mutant, the DNA fragment containing the N-terminal part of ntrB was amplified by PCR from the genomic DNA of PAO5122 (Fig. 1) with a transposon-specific flanking primer containing synthetic oligonucleotides designed to generate the HindIII or BamHI restriction site (underlined): 5′-AGGG AAGCTT CGC GCT TGG TGA ACG GATGT-3′. The C-terminal DNA fragment of ntrC was PCR amplified from the PAO1 genomic DNA with the following two primers: 5′-TGGA AAGCTT GGT GCG CGA CGG CAA GTTC-3′ and 5′-GTC GGATCC ATC GAC GTT CAT CCC CAG CTC-3′. The PCR products were purified from a 1% (wt/vol) agarose gel, digested by appropriate restriction endonucleases, and assembled onto the EcoRI and BamHI sites of pRTP1-M (36). The Tet^r cassette of EZ::TN<TET-1> was gel purified after HindIII digestion of a transposon-inserted plasmid and ligated to the HindIII fusion site of the NtrB-NtrC hybrid as described above. For gene replacement, the resulting transposon insertion plasmid in E. coli SM10 was mobilized into the recipient strains by biparental mating (8). Following incubation at 37°C overnight, transconjugants were selected on LB plates supplemented with tetracycline (100 μg/ml) and streptomycin (500 μg/ml). The correct gene replacements were confirmed by PCR with the following two primers flanking the ntrBC locus: 5′-GGT CAG CGC GAT AGA GCC AAG CGA-3′ and 5′-GTT GGG TCG CTC ATG GGA GCA GTG GCG CTT-3′.

• Open in new tab
• Download powerpoint

FIG. 1.

glnA-ntrC loci of PAO strains and plasmids. The PA gene numbers are based on annotations of the Pseudomonas Genome Project (http://www.pseudomonas.com/ ). Drawings for genes and transposon cassettes (Tn) are symbolic and not scaled. Dashed lines denote regions of deletion. Thick arrows represent genes and open reading frames with orientations. Strain PAO5122 carries a tetracycline resistance Tn inserted in PA5122 that was used as the source of PCR to make the ntrBC deletion mutant PAO5125. Strain PAO5125 has an antibiotic resistance (Tet^r) cassette in place of the deleted ntrBC region. Plasmids pHIS, pHIS1, and pNTR18 carry an ntrB(Con) allele from Arg18 (Table 1); the asterisks depict the relative locations of the point mutations carried on the ntrB genes of these constructs. Only relevant restriction sites are indicated: B, BamHI; C, ClaI; K, KpnI; S, Sau3AI.

To analyze the nucleotide sequences of the histidine suppressor mutants, the ntrB and ntrC alleles were PCR amplified by Pfu polymerase from genomic DNA of the suppressor mutants and PAO5100 with the following synthetic oligonucleotides: for ntrB, 5′-GGT CAG CGC GAT AGA GCC AAG CGA-3′ and 5′-GAC GAT CCA GAC GGT CTC TGA TCG-3′; for ntrC, 5′-CCT GCC CCT GGA ACA AGG AGT GCA TTG ACC-3′ and 5′-GTT GGG TCG CTC ATG GGA GCA GTG GCG CTT-3′. The amplified fragments were purified from a 1% (wt/vol) agarose gel and ligated into the SmaI site of pUCP18 (34). To make the pCBR plasmid, the 7-kb EcoRI fragment covering the cbrAB operon was purified from a cosmid clone, which was identified from colony screening of a genomic library constructed in this laboratory, and subcloned into pUCP18 (34). The DNA sequences of the insert on the plasmid were confirmed by nucleotide sequencing.

Enzyme assays.GS was assayed as γ-glutamyltransferase activity in cetyltrimethyl ammonium bromide-treated cells (16). The GOGAT and anabolic GDH activities were assayed as described previously (11). β-Galactosidase activities were determined by the methods of Miller (22). Protein concentrations were determined by the method of Bradford (4) with bovine serum albumin as the standard.

Growth phenotype analysis of cbrAB and ntrBC mutants.The CbrAB two-component system was reported by Nishijyo et al. (23) to control the utilization of carbon/nitrogen sources in P. aeruginosa, and it was anticipated that the NtrBC system controls nitrogen utilization in this organism. To further understand the physiological function of CbrAB and NtrBC in carbon and nitrogen catabolism, we conducted extensive growth phenotype analyses by PMs with the wild-type strain PAO1, the cbrAB mutant PAO5100, the ntrBC mutant PAO5125, and the cbrAB ntrBC double mutant PAO5126.

In comparison to the growth pattern of PAO1, that of the ntrBC mutant PAO5125 showed an expected deficiency on the utilization of N-only compounds (e.g., nitrite, nitrate, and urea) and specific C/N compounds (e.g., uridine and cytidine) as the sole nitrogen source (see Table S1 in the supplemental material), as has been reported for E. coli and many other enteric bacteria (20, 29). In contrast, the cbrAB mutant was defective in carbon utilization of a variety of compounds (24 out of 49), including C-only compounds (e.g., sugars, acetate, pyruvate, and lactate) as well as C/N compounds (e.g., arginine, histidine, and proline) (see Table S1 in the supplemental material). While the cbr mutant lost the capability to utilize many C/N compounds as the sole carbon source, almost all of these compounds could still serve as the sole nitrogen source in this mutant as well as in the cbr ntr double mutant, with few exceptions (e.g., l-histidine and l-serine). These results clearly delineate the pivotal role of CbrAB and NtrBC in regulation of carbon and nitrogen utilization, respectively; however, it was also intriguing to note that a cbrAB ntrBC double mutant retained the capability to utilize most C/N compounds as the sole nitrogen source.

Suppressors of the ΔcbrAB mutant.The spontaneous suppressors of PAO5100 (ΔcbrAB) were isolated from plates that restored growth on arginine used as the sole source of carbon and nitrogen. Among over 100 suppressors obtained, 90 of these mutants were randomly picked for further growth phenotype analyses of histidine and glucose, which can be utilized by PAO1 but not PAO5100. It was found that these suppressors can be divided into two groups: 5 mutants showing a histidine-positive, glucose-negative phenotype, as represented by Arg18, and 85 mutants exhibiting a histidine-negative, glucose-positive phenotype, as represented by Arg34.

To further explore the differences in catabolic capability, Arg18 and Arg34 were subjected to more extensive analyses by PMs. For carbon source utilization, Arg34 exhibited a pattern almost identical to that of the wild-type strain PAO1, except showing no growth on histidine, ornithine, and hydroxy-l-proline (see Table S2 in the supplemental material). Arg18 regained the ability to utilize most of the N-containing compounds but meanwhile lost the capability to utilize tricarboxylic acid (TCA) cycle intermediates (e.g., succinate) and most carbon sources containing no nitrogen moiety (see Table S2 in the supplemental material). For nitrogen source analysis, Arg34 had the same pattern as the wild-type strain PAO1. Arg18 was found surprisingly to show no growth on ammonium but grew normally on nitrate, urea, and other compounds (see Table S2 in the supplemental material). Since ammonium was used as the supplement in carbon source analysis and succinate was added as the carbon supplement in nitrogen source analysis, these results suggest that Arg18 might be deficient in the uptake or assimilation of ammonium per se.

Arg18 possesses a mutation of nitrogen regulation in the ntrB gene.In order to uncover the suppressor mutation of Arg18, a genomic library of Arg18 was constructed and employed to conduct genetic complementation of PAO5100 on histidine utilization. By this approach, the recombinant plasmid pHIS was recovered from one clone. To substantiate the physiological function of pHIS, this plasmid was introduced into PAO5100, and the resulting recombinant strain of PAO5100 harboring pHIS exhibited the same growth phenotypes as Arg18.

The results of sequence analysis revealed that pHIS carries a 6.5-kb chromosomal fragment with five complete (PA5120 through ntrB) and two truncated (glnA and ntrC) open reading frames (Fig. 1). The glnA, ntrB, and ntrC genes encoding GS and the NtrBC sensor/response regulator of nitrogen regulation were annotated based on sequence analysis. PA5120 to PA5123, the four genes between glnA and ntrB, encoded hypothetical proteins with unknown functions. When the subclone pHIS1 (Fig. 1), carrying an intact ntrB and devoid of four upstream genes, was introduced into PAO5100, it exerted the same complementation effect as pHIS.

To define the nature of the mutation responsible for the Arg18 phenotype, the nucleotide sequence of the DNA fragment carried by pHIS was determined and subjected to sequence comparison to the finished genomic sequence of P. aeruginosa PAO1. Only a single T680G mutation, which led to the Asp227Ala substitution, was detected on the ntrB allele of pHIS. To confirm the authenticity of this identified mutation, the ntrB genes of PAO5100 and Arg18 were cloned after PCR and sequenced. Sequence analyses confirmed the same point mutation in ntrB of Arg18 as in pHIS. Furthermore, a complementation experiment demonstrated that only the plasmid containing the ntrB allele of Arg18 (pNTR18) (Fig. 1), but not the wild-type ntrB gene, was able to convert the growth phenotype of PAO5100 on carbon utilization to that of Arg18.

Analysis of four other histidine-positive suppressors of PAO5100.Like Arg18, four other suppressors of PAO5100 (Arg8, Arg9, Arg10, and Arg24) isolated from arginine plates that restored the ability to utilize histidine as the sole carbon and nitrogen source were subjected to genetic analysis. Based on the results of allele analysis of Arg18, we predicted that the suppressor genes of these four mutants might be either ntrB or ntrC. Accordingly, the ntrB and ntrC genes of these mutants were PCR amplified and cloned into pUCP18. The resulting constructs were subjected to DNA sequence analysis and complementation analysis. Each of these four mutants carried a point mutation in the ntrB gene which led to the following amino acid substitutions: L126W in Arg8, P228L in Arg9, and S229I in Arg10 and Arg24. Complementation of PAO5100 with these ntrB mutant clones also resulted in a growth phenotype identical to that of the suppressor mutants. These results were consistent to what was observed in Arg18 as described above.

Effect of CbrAB and NtrBC on histidine and arginine catabolism.To understand the molecular mechanism responsible for the growth phenotype of PAO5100 and Arg18, we selected histidine and arginine utilization as two model systems for further analysis. A hutU::lacZ translational fusion, pSU500, was constructed to investigate expression of the hut operon with histidine utilization in different strains (Table 2). The cells were grown in glutamate minimal medium in the presence and absence of histidine. In the wild-type strain PAO1, exogenous histidine induced the expression of the hutU promoter, and this induction effect was abolished in PAO5100 devoid of cbrAB. In Arg18, the hutU promoter exhibited a higher basal level of activity than in PAO1, and exogenous histidine could further induce this promoter over threefold higher than that of PAO1 (Table 2). A similar pattern of promoter activities was also observed when pST500, an aotJ::lacZ translational fusion (24), was used to investigate expression of the aot operon for arginine transport in the same set of strains in response to the presence and absence of exogenous arginine (Table 2).

Since NtrC and CbrB are enhancers of σ⁵⁴ promoters, we also conducted measurements of aotJ and hutU promoter fusions in the rpoN mutant. As shown in Table 2, these two promoters were no longer subjected to the substrate-specific induction, and only low basal levels of expression can be observed in the rpoN mutant. These results strongly support the hypothesis that the CbrAB and NtrBC systems participate in regulation of genes for uptake or catabolism of histidine and arginine and possibly other C/N sources, as revealed by PM analysis.

The Ntr system in Arg18 is constitutively active.The ntrB mutation in Arg18 was anticipated to exert an effect on nitrogen regulation. To characterize the nature of this effect, the specific activities of GDH, GOGAT, and GS, three major enzymes in ammonium assimilation under the control of Ntr, were measured in PAO1 (wild type), PAO5100 (ΔcbrAB), PAO5125 (ΔntrBC), and Arg18. The cells were grown in succinate minimal medium with ammonium or nitrate as the sole nitrogen source. For PAO5125 growing on nitrate and Arg18 on ammonium, limiting amounts of glutamine (0.4 mM) and glutamate (1 mM), respectively, were added to enable cell growth until the optical density at 600 nm reached 0.5.

In the wild-type strain PAO1, GDH and GOGAT of glutamate biosynthesis were induced by ammonium and repressed when nitrate was the nitrogen source (Table 3). As expected, the expression profile of GS in PAO1 was reciprocal to that of GDH and GOGAT, i.e., repression by ammonium and induction by nitrate. A similar expression pattern of these three enzymes was also observed in PAO5100, except that the level of GS in PAO5100 was about 50% of that in PAO1 when nitrate was provided as the nitrogen source (Table 3). These results suggested that CbrAB per se may not be essential for nitrogen regulation, consistent with the results of PM analyses.

In contrast, significant changes were observed when the Ntr system was disturbed. In PAO5125 devoid of ntrBC, both GDH and GOGAT activities were constitutively high regardless of the nitrogen sources while the levels of GS remained constitutively low (Table 3). These results strongly suggest that the Ntr system exerts a repression effect on expression of GDH and GOGAT and an activation effect on GS expression. In comparison, a completely reversed pattern of expression was observed in Arg18.

We also conducted experiments with pSA500, a glnA::lacZ fusion, to investigate the expression of the glnA gene at the transcriptional level in the same set of strains under the same growth conditions as those described above. As shown in Table 3, the resulting pattern of glnA promoter expression was basically identical to what was observed at the GS enzyme levels. Based on these results, we hypothesized that the ntrB allele of Arg18 makes the Ntr system constitutively active in this mutant.

An imbalanced act by cbrAB deletion and ntrBC activation in Arg18.As revealed by growth phenotype analyses, regaining the ability to utilize most of the N-containing carbon sources in Arg18 was compromised by losing the capability to utilize TCA cycle intermediates and most carbon-only sources when ammonium was the sole nitrogen source. To further dissect the possible contribution from Cbr deletion and Ntr activation in Arg18, PM analysis was conducted with Arg18 harboring pCBR, which contains and expresses functional CbrAB from the plasmid. The result showed that the pattern of carbon utilization by Arg18/pCBR was not different from that of PAO1, which strongly suggests that a constitutive Ntr system in Arg18/pCBR does not affect carbon utilization and that the unique growth phenotype of Arg18 might be the consequence of an extremely imbalanced act by Cbr deletion and Ntr activation.