A Functional Homolog of Escherichia coli NhaR in Vibrio cholerae

Abstract

Escherichia coli NhaR controls expression of a sodium/proton (Na⁺/H⁺) antiporter, NhaA. The Vibrio cholerae NhaR protein shows over 60% identity to those of Escherichia coli and Salmonella enteritidis. V. cholerae NhaR complements an E. coli nhaR mutant for growth in 100 mM LiCl–33 mM NaCl, pH 7.6, and enhances the Na⁺-dependent induction of an E. coli chromosomal nhaA::lacZ fusion. These findings indicate functional homology to E. coli NhaR. Two V. cholerae nhaR mutants were constructed by using kanamycin resistance cartridge insertion at different sites to disrupt the gene. Both mutants showed sensitivity to growth in 120 mM LiCl, pH 9.2, compared with the wild-type strain and could be complemented by the introduction of V. cholerae nhaR on a low-copy-number plasmid. An nhaR mutation had no detectable effect on the virulence of the V. cholerae strain in the infant mouse model, suggesting that the antiporter system involved is not required in vivo, at least in this animal model.

Vibrio cholerae is a human pathogen which also has the ability to survive in a wide range of conditions of pH and salinity in its aquatic environment. V. cholerae is most frequently isolated from environmental sites with salinities between 0.2 and 2.0% (4) and survives in vitro in 0.25 to 3.0% salt; the optimal salinity is considered to be 2.0% (12). The optimal pH for survival is between 7.0 and 9.0, depending on salinity (12), and the pH of seawater is between 7 and 8. In vivo, the bacteria must survive passage through the acidity of the stomach to colonize the small intestine. All organisms survive variation in pH and salinity by employing homeostatic mechanisms; however, such mechanisms have not been investigated in V. cholerae, despite their potential importance to survival in the environment and, possibly, in the infection process. In two other vibrio species, V. parahaemolyticus and V. alginolyticus, sodium/proton antiporter genes have been identified which are involved in Na⁺ extrusion and intracellular pH regulation (10, 13–15).

In Escherichia coli, adaptation to high salinity and alkaline pH is dependent upon the NhaA sodium/proton antiporter system (17). Three other antiporters are known in E. coli; NhaB is also an Na⁺/H⁺ antiporter (19), ChaA is involved in Ca²⁺ and Na⁺ extrusion (16), and a third, less specific antiporter exchanges K⁺, Li⁺, Na⁺, or Rb⁺ ions for H⁺ (1).

Expression of the E. coli NhaA antiporter is known to be induced by Na⁺ and Li⁺ ions (9) and involves the NhaR regulator (20). NhaR belongs to the LysR family of positive regulators, and the nhaR gene is located 59 nucleotides downstream of nhaA. A feature common to members of the LysR family is an N-terminal helix-turn-helix domain which binds DNA, and several members of this family of regulators, including NhaR, are involved in stress responses (7).

We describe here the identification of an NhaR homolog in V. cholerae, which suggests that this organism has at least one pH and ion homeostasis mechanism similar to that of E. coli. Such mechanisms are likely to be important for survival of V. cholerae both in the aquatic environment and in the human host.

Sequence and functional comparisons of V. cholerae, E. coli, and S. enteritidis nhaR genes.

We identified the V. cholerae nhaR gene by sequencing and homology search in the region upstream of the previously published hlyU gene (22). These genes are carried on a 2.7-kb PstI clone which was obtained from a V. cholerae O17 plasmid library (22). Figure 1 shows a ClustalW amino acid sequence alignment of the V. cholerae, Salmonella enteritidis, and E. coli NhaR proteins. The extent of identity among these proteins suggested that V. cholerae NhaR could substitute for E. coli NhaR in functional assays. nhaR mutant E. coli OR100 (20) was transformed with pPM3089, a clone carrying nhaR from V. cholerae O17 on a 1.5-kb PstI-MscI fragment in the vector pGB2 (Fig. 2). OR100 showed a reduced growth rate in nutrient broth (NB; 10 g each of Oxoid Lab Lemco and Oxoid Bacto Peptone per liter) containing 100 mM LiCl–33 mM NaCl at pH 7.6, while OR100(pPM3089) showed growth comparable to that of wild-type strain TA15 (Fig. 3A). The vector alone had no effect on growth (Fig. 3B). Plasmid pPM3492, derived from pPM3089 by deletion of a 490-bp HpaI fragment removing the N-terminal region of nhaR (Fig. 2), showed no complementation of OR100 (Fig. 3B). However, clone pPM3092, with the Km^r cartridge inserted into nhaR via the EcoRV site (amino acid position 83; Fig. 2), showed full complementation of OR100 in 100 mM LiCl–33 mM NaCl, pH 7.6 (Fig. 3B). This suggests that the C-terminal 60% of the V. cholerae NhaR protein is not required to promote expression of E. coli nhaA. The helix-turn-helix domain is localized at amino acids 20 to 40 of NhaR, and this domain, with as little as 40 amino acids downstream, may suffice for DNA binding and promotion of nhaA transcription.

FIG. 1.

ClustalW alignment of V. cholerae (V. ch), S. enteritidis (S. en), and E. coli (E. co) NhaR proteins. Asterisks indicate amino acid identity, and dots indicate amino acid similarity. Amino acids are numbered on the right, and the helix-turn-helix domain is underlined. The E. coli sequence data are from Mackie (11) and Rahav-Manor et al. (20), and the S. enteritidis sequence is from Pinner et al. (18).

FIG. 2.

Genetic organization in the vicinity of nhaR of V. cholerae. Relevant restriction sites are shown, and arrows indicate the direction of transcription. The helix-turn-helix region within nhaR is shown by the hatched box. The plasmid constructs used are outlined below the gene map.

FIG. 3.

Growth of E. coli OR100 and complemented strains in NB with 100 mM LiCl and 33 mM NaCl, pH 7.6. (A) Growth curves of TA15(pGB2) (□), OR100 (▴), and OR100(pPM3089) (▾). (B) Growth curves of OR100 (pPM3092) (•), OR100(pGB2) (×), and OR100(pPM3492) (⧫).

V. cholerae NhaR could also substitute for E. coli NhaR in the activation of an nhaA::lacZ chromosomal fusion in E. coli RK33Z (9). This strain enables β-galactosidase activity to be assayed as a measure of nhaA induction (9). Clone pPM3091 encodes nhaR from V. cholerae O17 on a 2.7-kb PstI fragment in vector pPM2182 (22) (Fig. 2). It was introduced into E. coli RK33Z, and β-galactosidase activity was determined as described by Karpel et al. (9). Introduction of pPM3091 into RK33Z increased the induction of lacZ in an Na⁺-dependent manner, showing the same effect as the E. coli nhaR clone (Table 1). This is consistent with the idea that V. cholerae NhaR activates the expression of E. coli nhaA and provides further evidence that V. cholerae NhaR is a functional homolog of E. coli NhaR.

TABLE 1.

V. cholerae NhaR increases Na⁺-dependent expression of an E. coli nhaA::lacZ protein fusion

Straina	β-Galactosidase activity (Miller units)b
	No NaCl	100 mM NaCl
RK33Z	59.7	317
RK33Z(pGM42T)	71	1,276
RK33Z(pPM3091)	119	1,978

^aStrain RK33Z (nhaA::lacZ) was transformed with pGM42T (E. coli nhaR [20]) or pPM3091 (V. cholerae nhaR). 

^bThe cells were induced with Na⁺ (100 mM) for 150 min, and β-galactosidase activity was determined as previously described (9). 

A V. cholerae nhaR mutant shows sensitivity to 120 mM LiCl, pH 9.2.

Confirmation of the role of V. cholerae NhaR was sought by the construction of chromosomal nhaR mutants. The first mutation interrupted the nhaR gene with a 1.2-kb Km^r cartridge at the EcoRV site (shown in Fig. 2). The nhaR::Km^r gene was cloned into vector pRK290 (5) and introduced into V. cholerae O17 by conjugation. Strains in which the nhaR::Km^r gene had recombined into the chromosome were selected by the introduction of incompatible plasmid pH1JI (2), selection for gentamicin resistance (pH1JI), and maintenance of selection for Km^r. Plasmid-free strains were then obtained after two further steps: firstly by conjugating pME305 (21) into strains to promote loss of incompatible plasmid pH1JI and secondly by selecting at 42°C for the loss of temperature-sensitive plasmid pME305. The replacement of nhaR with nhaR::Km^r was confirmed by Southern hybridization (data not shown), and the mutant strain was called V881. A second mutation in V. cholerae nhaR was introduced which removed the 490-bp HpaI fragment spanning the N terminus of the gene (Fig. 2), replacing it with the 1.2-kb Km^r cartridge. This mutation was recombined into the chromosome by using the pCVD442 suicide vector system as outlined by Butterton et al. (3). The mutant, designated V1242, was confirmed by Southern analysis (data not shown).

Growth of the nhaR mutants was comparable to that of wild-type O17 in NB over a wide range of NaCl concentrations and pH values, up to 4% (0.68 M) NaCl and pH 10, which was completely inhibitory to the growth of all strains. However, differences in growth in 120 mM LiCl, pH 9.2, were observed between nhaR mutants and O17 (typical doubling times of 57 and 37 min, respectively). At 150 mM LiCl, pH 9.5, the nhaR mutants showed no growth, while O17 reached mid-log density after overnight incubation. These results suggest that V. cholerae nhaR is involved in survival in high LiCl concentrations at high pH. Both mutations in nhaR (in V881 and V1242) had the same effect on the growth of V. cholerae in high LiCl concentrations at high pH. This is despite the previous finding that the Km^r insertion mutation in V881 did not affect the ability of the gene to complement the nhaR mutant E. coli OR100 (Fig. 3B). This suggests that the N-terminal 80 amino acids of V. cholerae NhaR can function in E. coli but is not sufficient to function in V. cholerae. Consistent with this, we found that pPM3092 (nhaR::Km^r), which complemented OR100, did not complement our V. cholerae nhaR mutants. The nhaR clone (pPM3089) was complementary, restoring the growth rate of V881 and V1242 to that of the O17 control in 120 mM LiCl, pH 9.2 (data not shown).

The nhaR gene lies immediately downstream of nhaA in E. coli (11). An IS1 element and the gene for ribosomal protein S20, rpsT, are downstream of, and divergent to, nhaR (11). Sequencing of the 420 bp upstream of nhaR in V. cholerae has not revealed any candidate nhaA open reading frame. Downstream of V. cholerae nhaR is the hlyU gene (a regulator of hemolysin and hcp expression [22, 23]), followed by a divergent open reading frame showing between 65 and 82% identity to rpsT genes from at least seven different bacterial species. The location of V. cholerae nhaA remains to be determined. Similarly, in both V. alginolyticus and V. parahaemolyticus, the nhaA and nhaR genes do not appear to be closely linked (10, 14).

The proximity of the nhaR and hlyU regulatory genes in V. cholerae may be significant. HlyU shows amino acid similarity to CadC of Bacillus firmus (8) and the CadC gene regulator from Staphylococcus aureus plasmid pI258 (6). B. firmus CadC has been shown to partially complement nhaA mutant E. coli NM81 (8). This complementation has been suggested to be nonspecific, due to the binding of Na⁺ by CadC, which is then transferred to membrane-bound antiporter systems. Since the mutations we introduced into nhaR could have been polar on hlyU expression, an hlyU::Km^r mutant, V876 (22), was also examined for growth in 120 mM LiCl, pH 9.2. This mutant showed no sensitivity to this medium compared to O17, so we conclude that HlyU has no role in antiporter activity under these growth conditions (data not shown). Therefore, the effects of the nhaR mutations observed cannot be due to polar effects on hlyU expression, a conclusion consistent with the complementation of nhaR mutants by the nhaR gene in pPM3089.

Infection by V. cholerae involves a significant change in environment, in particular, the gastric acid barrier presented by the human stomach, and the organism must have pH homeostasis mechanisms of importance to infection. Although gastric acid presents a low-pH assault, we tested the possibility that NhaR is part of a pH homeostasis mechanism with importance to the infection process. The 50% lethal dose of V881 in the infant mouse cholera model was 1.5 × 10⁴, compared with 2.9 × 10⁴ for the parent strain, O17 (determined as described in reference 22). This result suggests no role for NhaR during infection in this animal model. It is perhaps more likely that NhaR is important for survival of V. cholerae in the aquatic environment, where higher salinity and pH may be encountered.

The variation in growth conditions encountered by V. cholerae within the environment and upon entry into its human host must be countered by homeostatic mechanisms. We have reported the finding of an NhaR homolog in V. cholerae which is capable of regulating the E. coli antiporter, nhaA. The inability of V. cholerae nhaR mutants to grow in 150 mM LiCl, pH 9.5, which supported the growth of the parent strain, suggests that this gene regulates an antiporter responsible for adaptation to growth in high concentrations of LiCl at high pH. The relevance of LiCl to the growth environments of V. cholerae is unclear; however, it is possible that further examination may reveal other cations as substrates for this putative antiporter system. It is clear that alternative antiporter systems are likely to be identified in V. cholerae, and the discovery of such systems may reveal how this organism survives different conditions of salinity and pH.

Nucleotide sequence accession number.

The nucleotide sequence reported here has been assigned accession no. AJ002395 in the EMBL Nucleotide Sequence Database.