Vibrio cholerae OmpR Contributes to Virulence Repression and Fitness at Alkaline pH

Vibrio cholerae is a Gram-negative human pathogen and the causative agent of the life-threatening disease cholera. V. cholerae is a natural inhabitant of marine environments and enters humans through the consumption of contaminated food or water. The ability to transition between aquatic ecosystems and the human host is paramount to the pathogenic success of V. cholerae. The transition between these two disparate environments requires the expression of adaptive responses, and such responses are most often regulated by two-component regulatory systems such as the EnvZ/OmpR system, which responds to osmolarity and acidic pH in many Gram-negative bacteria.

ABSTRACT

Vibrio cholerae is a Gram-negative human pathogen and the causative agent of the life-threatening disease cholera. V. cholerae is a natural inhabitant of marine environments and enters humans through the consumption of contaminated food or water. The ability to transition between aquatic ecosystems and the human host is paramount to the pathogenic success of V. cholerae. The transition between these two disparate environments requires the expression of adaptive responses, and such responses are most often regulated by two-component regulatory systems such as the EnvZ/OmpR system, which responds to osmolarity and acidic pH in many Gram-negative bacteria. Previous work in our laboratory indicated that V. cholerae OmpR functioned as a virulence regulator through repression of the LysR-family transcriptional regulator aphB; however, the role of OmpR in V. cholerae biology outside virulence regulation remained unknown. In this work, we sought to further investigate the function of OmpR in V. cholerae biology by defining the OmpR regulon through RNA sequencing. This led to the discovery that V. cholerae ompR was induced at alkaline pH to repress genes involved in acid tolerance and virulence factor production. In addition, OmpR was required for V. cholerae fitness during growth under alkaline conditions. These findings indicate that V. cholerae OmpR has evolved the ability to respond to novel signals during pathogenesis, which may play a role in the regulation of adaptive responses to aid in the transition between the human gastrointestinal tract and the marine ecosystem.

INTRODUCTION

Vibrio cholerae is a Gram-negative facultative pathogen and the causative agent of the disease cholera. V. cholerae is a native member of the marine microbiota. Pathogenic strains of V. cholerae enter the human host from the environment through the consumption of V. cholerae-contaminated food or water. Following ingestion, V. cholerae traverses the gastrointestinal tract to ultimately colonize the surface of the distal small intestine (1). During this process, V. cholerae activates the expression of virulence factors. The most important virulence factors for disease progression are the enterotoxin cholera toxin (CT), which is responsible for the secretory diarrhea that is the hallmark of the disease cholera (2), and the toxin-coregulated pilus (TCP), which facilitates intestinal colonization (3). Expression of CT and TCP is tightly controlled by a hierarchical regulatory cascade known as the ToxR regulon (4). Activation of this regulon begins with two cytoplasmic transcription factors, AphA and AphB (5, 6). AphA and AphB together activate the transcription of tcpP. TcpP then acts along with ToxR to activate toxT expression. ToxT directly activates the expression of genes that encode proteins needed for the production of CT and TCP (4). Fine-tuning of the ToxR regulon is mediated in part by factors affecting the production and activity of AphA and AphB (7–9).

The complex life cycle of pathogenic V. cholerae, consisting of growth in aquatic ecosystems and the human gut, necessitates that V. cholerae adapt rapidly to sudden environmental changes, including changes in nutrient availability, osmolarity, temperature, and pH. There is an array of regulatory systems controlling these adaptive responses during V. cholerae infection (10–13). Early in infection, V. cholerae genes involved in the acid tolerance response (ATR) are induced as the cells pass through the acidic environment of the stomach (14). After exiting the stomach, V. cholerae adapts to the more alkaline environment in the small intestine by repressing ATR genes; a failure to do so results in an intestinal colonization defect (13). At the same time, V. cholerae activates the expression of genes required for colonization and disease development (i.e., the ToxR regulon). During the late stages of infection, before exiting the host in the diarrheal purge, V. cholerae represses virulence genes while inducing genes required for dissemination and transmission (9, 12, 15, 16). The genetic mechanisms and environmental cues that contribute to adaptive responses governing these late infection responses in V. cholerae in vivo remain largely unexplored.

Two-component regulatory systems (TCSs) play a critical role in regulating adaptive responses to environmental cues. TCSs are widespread in bacteria and transduce extracellular signals into transcriptional responses. The prototypical TCS consists of a membrane-bound sensor histidine kinase (HK) (17) and a cytoplasmic response regulator (RR). The HK contains a signal-sensing domain that monitors the extracellular environment for specific activating signals (17). Upon activation, the HK autophosphorylates and then transfers its phosphate group to a conserved aspartate residue on its cognate RR. Phosphorylation of the RR results in its activation, and it then functions to modulate transcriptional responses to the inducing cue (18). Most RRs function as transcription factors, but they can also effect adaptive responses by other mechanisms (19). One of the best studied TCSs is the EnvZ/OmpR system. EnvZ/OmpR is ubiquitous in Gram-negative bacteria and has been extensively studied for its role in regulating porin production in response to changes in environmental osmolarity and acidic pH (20–24). We recently reported that V. cholerae OmpR was a virulence repressor that regulated expression of the ToxR regulon by repression of aphB (7). Our work also showed that V. cholerae ompR did not respond to changes in osmolarity, the canonical ompR-inducing signal in other Gram-negative bacteria. Instead, V. cholerae ompR was induced by membrane stress-inducing agents that are found within the host gut. Collectively, these findings suggested that OmpR evolved to respond to niche-specific environmental signals in V. cholerae.

In this work, we further explored the function of OmpR in V. cholerae. Using transcriptional profiling and genetic approaches, we documented that OmpR functioned in the V. cholerae ATR by mechanisms mediated through its regulation of aphB. Further, we showed that ompR expression was induced at alkaline pH, which contributed to V. cholerae fitness at alkaline pH. Collectively, our results extend the function of OmpR in V. cholerae biology to include both virulence factor production and adaptation to alkaline pH, two phenotypes that are relevant to human infection.

RESULTS

Defining the V. cholerae OmpR regulon.

Recent work in our laboratory documented that V. cholerae OmpR did not appear to function in response to osmolarity but instead was a virulence repressor that was induced in response to membrane-intercalating agents (7). This suggested that the function of OmpR in V. cholerae diverges from what is observed in other gammaproteobacteria. To further explore the impact of OmpR on V. cholerae biology, we used high-throughput RNA sequencing (RNA-seq) to define the V. cholerae OmpR regulon during growth under virulence-inducing conditions (i.e., AKI conditions). In these experiments, we cultured a V. cholerae ΔompR mutant bearing either pBAD33 or pBAD33-ompR under virulence-inducing conditions, in AKI broth containing 0.05% arabinose inducer. Total RNA was then isolated at 4.5 h and used for RNA-seq, as described previously (25). The results from three independent experiments were then analyzed to identify differentially expressed genes, which were defined as genes exhibiting ≥2-fold changes in expression in the pBAD33-ompR samples, compared to the empty vector control (i.e., pBAD33), with P values of ≤0.05. This analysis resulted in the identification of 191 differentially expressed genes, including 67 genes that were upregulated and 124 that were repressed by ompR overexpression (see Table S1 in the supplemental material). Most of the differentially regulated genes were annotated as belonging to four functional groups (Table 1), i.e., pathogenesis (9.9%), transport and binding proteins (19.9%), metabolism (25.7%), and uncharacterized (31.4%).

TABLE 1.

Functional classes of OmpR-regulated genes in V. cholerae

Functional group	No. of genes
	Upregulated	Downregulated	Total
Transcription	1	0	1
Transport and binding proteins	21	17	38
Biosynthesis of cofactors, prosthetic groups, and carriers	2	3	5
Cellular processes	1	2	3
Regulatory function	2	4	6
Metabolism	17	32	49
Cell envelope	0	5	5
Protein fate	0	1	1
Amino acid biosynthesis	0	3	3
DNA replication, recombination, and repair	0	1	1
Pathogenesis	0	19	19
Conserved, hypothetical, and unknown	23	37	60
Total	67	124	191

The validity of our approach for identifying OmpR-regulated genes was confirmed by the finding that ompR overexpression resulted in the repression of pathogenesis genes, including most of the genes involved in CT and TCP production, a finding consistent with the recent report that OmpR was a V. cholerae virulence repressor (7). Further support for our approach was provided by the observation that 35% of the differentially expressed genes due to ompR overexpression were also identified as being differentially expressed in a V. cholerae resistance-nodulation-division (RND) efflux negative mutant (Table S1), in which ompR expression was constitutively upregulated (7, 26). Interestingly, overexpression of ompR did not affect the expression of its canonical sensor envZ. Because envZ is encoded in a bicistronic operon downstream of ompR, this finding suggested that OmpR may not autoregulate its own expression in V. cholerae, as it does in other bacterial species (27, 28). This finding was confirmed by transcriptional reporter data showing that ompR expression was similar in the wild-type (WT) strain and in its isogenic ΔompR mutant strain during growth under virulence-inducing conditions (Fig. S1).

The observation that about one-half (∼46%) of the differentially regulated genes fell into the metabolism and transport and binding functional groups suggested that a major role for OmpR is in metabolic and environmental adaptation (Table 1; also see Table S1). Significantly, most of the regulated metabolism genes were repressed by OmpR; this included repression of genes involved in central and intermediary metabolism and genes involved in the electron transport chain. Upregulated genes included genes for a fructose phosphotransferase system, sulfate metabolism genes, and anaerobic metabolism genes. This finding was similar to those observed for other Enterobacteriaceae, in which OmpR reprograms the cell transcriptome in response to inducing stimuli (29, 30).

OmpR has been linked to osmoregulation in the Enterobacteriaceae. In V. cholerae, osmoadaptation has been linked to the osmoregulator OscR and involves increased biofilm production, ectoine biosynthesis, and the import of compatible solutes through OmpW, OpuD, and PutP transporters (31–34). In the RNA-seq data set, OmpR repressed oscR (2.6-fold) and ompW (3.38-fold) but did not affect the expression of biofilm genes (i.e., vps genes) or ectoine biosynthesis genes (i.e., ectABC, opuD, and putP). These results, coupled with previous findings that V. cholerae ompR was dispensable for growth at high osmolarity, led us to conclude that V. cholerae OmpR does not function in osmoadaptation under the tested conditions.

V. cholerae OmpR functions in OMP gene regulation.

OmpR regulates the expression of outer membrane protein (OMP) genes in many Gram-negative bacteria (21–23, 35, 36). Consistent with this, the RNA-seq results suggested that V. cholerae OmpR regulated the expression of three OMP genes; ompW (VCA0867) and ompV (VC1318) were repressed 3.38- and 2.02-fold, respectively, and chiP (VC0972) was induced 2-fold. We validated this finding by quantifying the effects of ompR overexpression on ompW, ompV, and chiP plus the major V. cholerae OMP genes ompU and ompT, which were not differentially regulated in the RNA-seq data. To do this, V. cholerae ΔompR strains harboring either pBAD33 or pBAD33-ompR were cultured under AKI conditions for 5 h in the presence of 0.05% arabinose, and gene expression was quantified by quantitative reverse transcription (qRT)-PCR (Fig. 1). The results showed that OmpR repressed ompW while activating chiP. The expression of ompV was also reduced; however, the results did not reach statistical significance. Significantly, OmpR did not affect the expression of the two major V. cholerae OMP genes, ompT and ompU, consistent with our previous results (7). This contrasts with findings for other Enterobacteriaceae, in which OmpR differentially regulates the production of the major OMP genes ompC and ompF (28, 37, 38). Collectively, these results validated the OmpR transcriptomic data and indicated that V. cholerae OmpR retains conserved functions for the regulation of OMP genes but not for the regulation of ompT and ompU, which are under the control of ToxR and other effectors (39–42).

FIG 1.

V. cholerae OmpR modulates porin gene expression. V. cholerae ΔompR cells harboring either pBAD33 or pBAD33-ompR were cultured under AKI conditions with 0.05% arabinose for 5 h, at which time RNA was collected and used for qRT-PCR to quantify ompW, ompV, chiP, ompU, and ompT expression. The data indicate the average ± SD of a minimum of three independent experiments performed in triplicate. *, P < 0.01; **, P < 0.0001, relative to pBAD33, as determined using Dunnett’s multiple-comparison test. NS, not significant.

OmpR regulates ATR genes and is induced by alkaline stress.

OmpR has been studied for its role in activating ATR genes in response to low pH in a number of Gram-negative organisms (28, 36, 43–47). Interstingely, the V. cholerae ompR transcriptome indicated that OmpR repressed a number of ATR genes, including nhaP1, cadB, cadA, and clcA (13, 14, 48). The fact that cadC and clcA are positively regulated by AphB at low pH (8) and OmpR directly represses aphB transcription (7) led us to hypothesize that OmpR may function upstream of aphB to regulate genes involved in pH adaptation. To test this, we first sought to validate the RNA-seq results by quantifying by qRT-PCR the effects of OmpR on the expression of the AphB-dependent genes cadC, cadB, cadA, and clcA. Consistent with the RNA-seq data, the results showed that ompR overexpression repressed cadB, cadA, and clcA expression by ∼2-fold (Fig. 2A). Expression of cadC, which encodes the direct regulator of the cadBA operon, was also reduced but to a lesser extent, mirroring the RNA-seq results.

FIG 2.

OmpR represses ATR genes. (A) V. cholerae ΔompR cells harboring either pBAD33 or pBAD33-ompR were cultured under AKI conditions with 0.05% arabinose. RNA was collected at 5 h and used for qRT-PCR to quantify cadC, cadB, cadA, and clcA expression. The fold changes in gene expression in the pBAD33-ompR-harboring strains are relative to that in the pBAD33-harboring strains. The data are the average ± SD of a minimum of three independent experiments performed in triplicate. *, P < 0.05; **, P < 0.001; ***, P < 0.0001, relative to pBAD33, as determined by Dunnett’s multiple-comparison test. (B) WT V. cholerae cells harboring an ompR-lacZ reporter plasmid were cultured under AKI conditions for 4 h before NaOH was added to the indicated concentrations; pH values of the medium following NaOH addition, at the time of the experiment, are shown. β-Galactosidase activity was quantified 1 h after treatment. The data indicate the average ± SD of a minimum of three independent experiments performed in triplicate. *, P < 0.01; **, P < 0.0001, relative to 0, as determined by Dunnett’s multiple-comparison test. NS, not significant.

The aforementioned results suggested that V. cholerae OmpR regulated ATR genes. We next tested whether ompR was itself regulated in response to pH. We cultured WT cells harboring an ompR-lacZ reporter under AKI conditions for 4 h, and the culture medium pH was altered through the addition of either NaOH or HCl. We then quantified ompR-lacZ expression 60 min after exposure. The results showed that acidification of the medium with HCl addition did not alter ompR expression (data not shown). In contrast, there was an alkaline pH-dependent induction of ompR expression, suggesting that ompR was regulated in response to alkaline pH (Fig. 2B). The fact that OmpR repressed ATR genes, whose expression is likely to be detrimental during alkaline stress, and was induced under alkaline pH conditions suggested that OmpR may contribute to alkaline pH responses in V. cholerae.

OmpR represses aphB at alkaline pH.

The observation that ompR was induced by alkaline pH, combined with OmpR functioning as an aphB repressor (7), suggested that OmpR might regulate aphB expression during growth under alkaline pH conditions. To test this, we cultured WT and ΔompR strains bearing either an aphA-lacZ or aphB-lacZ reporter under AKI conditions for 4 h before adding 20 mM NaOH to raise the medium pH to 8.7; the control for these experiments received an equivalent amount of water. The cultures were incubated for an additional 1 h before quantification of aphA and aphB expression. The results showed that increased medium pH had little effect on aphA expression in either strain (Fig. 3A). In contrast, raising the pH with NaOH addition decreased aphB expression in the WT strain but not in the ΔompR mutant (Fig. 3B), indicating that OmpR repressed aphB transcription in response to alkaline pH.

FIG 3.

OmpR represses aphB and cadB expression at alkaline pH. (A to D) V. cholerae strains harboring aphA-lacZ (A), aphB-lacZ (B), cadC-lacZ (C), or cadB-lacZ (D) reporters were cultured under AKI conditions for 4 h, at which time 0 mM NaOH (black bars) or 20 mM NaOH (green bars) was added to the culture medium; the addition of 20 mM NaOH increased the medium pH to 8.7. The cultures were then incubated for an additional 1 h, after which β-galactosidase activity was quantified. The data are the average ± SD of at least three independent experiments, with each experiment being performed in triplicate. *, P < 0.01; **, P < 0.001, relative to 0 mM control for each strain, as determined by Student's t test. (E and F) E. coli strains harboring either pBAD33 (black bars) or pBAD33-ompR (blue bars) plus a cadC-lacZ (E) or cadB-lacZ (F) transcriptional reporter plasmid were cultured in LB with the indicated arabinose concentrations for 5 h, after which β-galactosidase activity was quantified. The data are the average ± SD of three independent experiments, with each experiment being performed in triplicate. MU, Miller units; NS, not significant.

To confirm that OmpR downregulated ATR genes at alkaline pH through aphB, we cultured WT, ΔompR, ΔaphA, and ΔaphB strains harboring either cadC-lacZ or cadB-lacZ reporters under AKI conditions and quantified gene expression in response to alkaline pH, as described above. As expected, both cadC and cadB were repressed at alkaline pH in the WT strain (Fig. 3C and D). The expression of cadC was also repressed at alkaline pH in the ΔompR mutant, suggesting that alkaline pH-dependent repression of cadC was OmpR independent. In contrast, the alkaline pH-dependent repression of cadB was lost in the ΔompR mutant, indicating that OmpR contributes to cadB repression at alkaline pH (Fig. 3D). Unexpectedly, the expression levels of both cadC and cadB were decreased under standard AKI conditions in the aphA mutant, relative to the WT strain, and their expression levels further decreased to a level like that in the WT strain at alkaline pH (Fig. 3C and D). This finding suggests that AphA may play a role in the basal expression of cadC and cadB at neutral pH but aphA likely does not contribute to their regulation at alkaline pH; additional studies will be required to assess the biological relevance of this observation. The expression levels of cadC and cadB were dramatically reduced in the aphB mutant (Fig. 3C and D), relative to the WT strain, during growth in standard AKI broth, confirming previous studies that showed that AphB positively regulated the Cad system (8). Expression of cadC was not further repressed at alkaline pH in the aphB mutant (Fig. 3C), whereas cadB expression was repressed at alkaline pH in the aphB mutant (Fig. 3D). This finding suggests the presence of an AphB- and OmpR-independent mechanism for cadB repression in response to alkaline pH. From these results, we concluded that the V. cholerae Cad system was repressed in response to alkaline pH and that OmpR contributed to aphB repression in response to alkaline pH. Our results further suggest that repression of the Cad system in response to alkaline pH is modulated by both AphB-dependent and AphB-independent mechanisms.

The expression of cadBA in Escherichia coli and Salmonella is directly repressed by OmpR (49). The findings described above indicated that V. cholerae OmpR repressed cadBA expression in response to alkaline pH in an AphB-dependent manner. However, the results described above did not exclude the possibility that OmpR also directly repressed the expression of cadC or cadBA. To address this possibility, we expressed V. cholerae ompR in E. coli strains bearing cadC-lacZ or cadBA-lacZ transcriptional reporters. The E. coli strains were cultured to mid-log phase in lysogeny broth (LB) containing a range of arabinose concentrations before cadC and cadBA expression levels were quantified (Fig. 3E and F). The results showed that overexpression of V. cholerae ompR did not affect the expression of either cadC or cadBA in E. coli, suggesting that OmpR indirectly regulated the Cad system in V. cholerae.

Alkaline pH represses V. cholerae virulence factor production in an OmpR-dependent manner.

Virulence factor production in V. cholerae is modulated in response to multiple environmental cues, including pH (8, 50, 51). Expression of the ToxR regulon in V. cholerae El Tor biotypes is optimal in AKI broth at pH 7.4 (52). For classic V. cholerae biotypes, elevated pH (∼pH 8.5) has been used as a virulence-noninducing condition in multiple studies (41, 50, 51, 53, 54). Because our results showed that ompR was induced at alkaline pH and OmpR functioned as a virulence repressor through aphB repression (7), we hypothesized that OmpR might contribute to virulence repression at alkaline pH. To test this, we cultured WT and ΔompR strains under AKI conditions in standard AKI broth at pH 7.4 (inducing conditions) and pH 8.7 (noninducing conditions). We then quantified CT production at 6 h and following overnight growth; levels were assessed at 6 h because AphB is thought to be most relevant during early induction of the ToxR regulon. When cultured in pH 7.4 AKI broth, the ΔompR mutant strain produced similar amounts of CT, compared to the WT strain, at both time points tested, confirming the previous report (7) (Fig. 4A and B). In contrast, growth of the WT strain in pH 8.7 AKI broth resulted in pronounced reductions of CT production at both 6 h and 24 h. The alkaline pH-dependent virulence attenuation at 6 h was OmpR dependent, as CT production in the ΔompR mutant strain was not affected by high pH (Fig. 4A). At 24 h, CT production was highly attenuated in the WT and ΔompR strains at pH 8.7, but the level of attenuation was less in the ΔompR background.

FIG 4.

OmpR represses virulence factor production at alkaline pH. V. cholerae WT and ΔompR cells were cultured under AKI conditions in standard AKI broth at pH 7.4 or at pH 8.7. Culture aliquots were collected at 6 h (A) and 24 h (B) to quantify CT production by GM1 ELISA or at 5 h (C) and 3 h (D) to quantify tcpA-lacZ (C) and aphB expression by qRT-PCR (D). The data are the average ± SD of three independent experiments. *, P < 0.001, relative to the WT strain, as determined by Student's t test. MU, Miller units.

CT production and TCP production are coordinately regulated by the ToxR regulon. Therefore, we further explored the impact of alkaline pH on virulence by quantifying the expression of tcpA and aphB under the same conditions as used above. The results showed that the expression of tcpA, which encodes the pilin subunit of the TCP, was repressed in an OmpR-dependent manner in response to alkaline pH (Fig. 4C). We also found that aphB expression was repressed in response to alkaline pH, by a mechanism that was partially dependent on OmpR (Fig. 4D). The fact that deletion of ompR did not fully restore aphB expression or CT production to WT levels (Fig. 4B) suggests that both OmpR-dependent and OmpR-independent mechanisms contribute to virulence repression in response to alkaline pH. Based on these results, we concluded that alkaline pH represses CT and TCP production. We further concluded that virulence repression is OmpR dependent at early time points and is mediated in part by OmpR-dependent repression of aphB. Our data suggest that OmpR contributes to alkaline pH-induced CT repression at later time points but OmpR-independent mechanisms also exist.

V. cholerae OmpR is required for fitness in alkaline pH.

The observations that OmpR repressed ATR genes and that ompR expression was induced at alkaline pH suggested that OmpR might contribute to fitness at alkaline pH. To test this, we determined the growth kinetics of WT, ΔompR, ΔaphA, and ΔaphB strains in LB at pH 7.4, at pH 8.7, and at pH 9.7. The results revealed that all four test strains had similar growth kinetics in LB at pH 7.4 (Fig. 5A) and pH 8.7 (data not shown). In contrast, the ΔompR mutant exhibited a pronounced growth defect, relative to the other strains, during growth at pH 9.7, as indicated by an increase in the lag phase (Fig. 5B). The growth attenuation of the ΔompR mutant at pH 9.7 was complemented by ectopic ompR expression (compare the black and green traces in Fig. 5D), whereas there was no difference in the growth of the two strains in LB at pH 7.4 (Fig. 5C). This finding confirms that the observed alkaline pH-dependent phenotype was due to the ompR mutation and not some nonspecific effect. We noted that the presence of pBAD33, independent of ompR, exhibited a nonspecific negative effect on the growth of all strains at alkaline pH, as shown by a decrease in the final culture density (Fig. 5D); the reasons for this are unknown. The ΔaphA and ΔaphB mutants exhibited little difference in growth kinetics, relative to the WT strain, at both pH values tested, which suggests that their contributions to growth under the test conditions were minimal (Fig. 5B).

FIG 5.

V. cholerae ΔompR cells show attenuated growth at alkaline pH. Overnight LB cultures of WT, ΔompR, ΔaphA, and ΔaphB V. cholerae strains were diluted 1:10,000 in LB (A and C) or LB supplemented with 35 mM NaOH (pH 9.7) (B and D). Arabinose (0.05%) was included in the growth medium for V. cholerae ΔompR strains harboring pBAD33, pBAD33-ompR, pBAD33-ompR D55E, or pBAD33-ompR D55A (C and D). The cultures were distributed into the wells of a 96-well microtiter plate in triplicate. The microtiter plate was then incubated at 37°C in a plate reader, with shaking, and cell growth was monitored by reading the OD at 630 nm every 30 min. The data are the average ± SD of at least three independent experiments.

The aforementioned experiments confirmed that V. cholerae OmpR contributed fitness during growth at alkaline pH. Previous studies in E. coli showed that OmpR phosphorylation was not required for OmpR-dependent responses to pH (45). To determine whether the phosphorylation status of OmpR played a role in V. cholerae responses to alkaline pH, we repeated the aforementioned complementation experiments with plasmids expressing mutant alleles of V. cholerae OmpR that mimicked phosphorylated OmpR (OmpR^D55E) and dephosphorylated OmpR (OmpR^D55A). Each of the mutant alleles was expressed in ΔompR cells during growth in LB at pH 7.4 and at pH 9.7, as described above. During growth at pH 7.4, the ompR^D55E allele resulted in a longer lag phase, relative to the ompR^D55A allele, and the final densities of both cultures (ompR^D55E and ompR^D55A) were decreased, relative to that of the WT strain (Fig. 5C). These findings suggest that the exclusive expression of either phosphomimic is detrimental for growth at neutral pH. The expression of all three ompR alleles (i.e., ompR, ompR^D55A, and ompR^D55E) complemented the ΔompR mutant for growth at alkaline pH (Fig. 5D). This finding suggests that, although OmpR is required for optimal fitness at alkaline pH, this phenotype is independent of the phosphorylation status of OmpR. Taken together, these results suggest that ompR contributes to V. cholerae fitness during growth under alkaline conditions and that this process is independent of the OmpR phosphorylation status.

DISCUSSION

The life cycle of pathogenic V. cholerae relies on transitioning between marine ecosystems and the human gastrointestinal tract. The ability of V. cholerae to cycle successfully between these two disparate environments is dependent on the activation of TCSs, which sense and respond to environmental cues to coordinate the expression of adaptive responses. In this study, we characterized the response of V. cholerae OmpR to alkaline pH. Based on our results, we propose the following model, in which V. cholerae OmpR functions at alkaline pH to coordinately repress genes involved in virulence and acid tolerance via AphB (Fig. 6). At alkaline pH, OmpR directly represses aphB transcription, resulting in downregulation of the ToxR virulence regulon to attenuate CT and TCP production. Repression of aphB also results in reduced expression of ATR genes (e.g., cadC, cadBA, and clcA), which likely contributes to V. cholerae fitness at alkaline pH. We speculate that this novel function of the V. cholerae EnvZ/OmpR TCS in regulating adaptive responses to alkaline pH may contribute to successful transitions between the human gastrointestinal tract and marine ecosystems.

FIG 6.

Model of V. cholerae OmpR in response to alkaline pH. Alkaline pH induces the expression of V. cholerae ompR, which then directly represses aphB transcription. Repression of aphB results in the downregulation of genes involved in both acid tolerance and virulence factor production, leading to enhanced fitness at alkaline pH and attenuated production of CT and TCP.

Bacterial regulatory networks evolve in response to selective pressures during growth in specific niches. This drives the evolution of regulatory networks to respond to novel signals and to alter the constitution of their target regulons (55, 56). This divergent evolution results in conserved TCSs fulfilling different physiological roles across bacterial species and genera (57), as has been documented with the EnvZ/OmpR TCS. EnvZ/OmpR is conserved among gammaproteobacteria, but multiple studies have shown divergence in OmpR-inducing signals and in the OmpR regulon among different species (28, 35, 58–60). This appears to be the case also in V. cholerae. The work presented here and in a previous report (7) indicates that the function of EnvZ/OmpR in V. cholerae has diverged from that in other well-studied enteric pathogens. We speculate that the unique lifestyle of V. cholerae has supplied the selective pressures for evolution of V. cholerae EnvZ/OmpR.

Our collective results suggest that the EnvZ/OmpR TCS fulfills novel physiological roles in V. cholerae. The observation that ompR was not induced by acidic pH or medium osmolarity but was instead induced by alkaline pH suggested that OmpR likely functioned in adaptation to alkaline pH. Support for this hypothesis was provided by the observation that a V. cholerae ompR mutant showed attenuated growth at alkaline pH. The discovery that OmpR repressed V. cholerae ATR genes (nhaP1, cadBA, and clcA) highlighted a potential mechanism for OmpR-dependent adaptation to alkaline pH. The V. cholerae ATR involves induction of AphB activity and the increased expression of genes that are collectively predicted to prevent acidification of the cytoplasm (i.e., cadBA, nhaP1, and clcA) (8, 13, 48). The Cad system contributes to the ATR by preventing cytoplasmic acidification. CadA is a lysine decarboxylase that consumes a proton during the decarboxylation reaction to produce cadaverine, which is then removed from the cell by the lysine-cadaverine antiporter CadB. ClcA and NhaP1 are H⁺/Cl⁻ and Na⁺/H⁺ antiporters, respectively, which are predicted to facilitate survival at low pH by expelling protons from the cytoplasm. Conversely, the repression of these genes at high pH would block intracellular proton depletion and likely contribute to adaptive responses to counteract alkalization of the cytoplasm (13). It is interesting to note that this contrasts with the function of EnvZ/OmpR in the Enterobacteriaceae. In E. coli and Salmonella, ompR is induced by high osmolarity and acidic pH and reprograms the cell transcriptome to enhance survival at low pH. Recent studies have shown that, during adaptation to low pH, E. coli and Salmonella acidify their cytoplasm by a mechanism facilitated by OmpR-dependent repression of the Cad system (45, 61, 62).

Recent studies have shown that alkaline pH-dependent repression of clcA is critical for infection in the infant mouse infection model (13). In infant mice, clcA is induced in the stomach but repressed in the more alkaline intestine. Expression of clcA in alkaline environments is detrimental, and failure to repress clcA in the intestine was associated with a 50-fold colonization defect, compared to the WT strain (13). The expression of clcA in response to low pH is mediated by AphB (63), but the molecular mechanism involved in clcA repression in the intestine is unknown. Studies indicate that rice water stool from cholera patients is alkaline (pH 7.5 to 8.5), making it interesting to speculate that OmpR may contribute to clcA repression in the intestine (64), but additional work will be required to verify this. We note that there are conflicting reports regarding the contribution of ompR to intestinal colonization, suggesting that there may be strain differences among V. cholerae isolates. A recent study reported that ompR was required for colonization (65), while three other studies suggested that ompR was dispensable for colonization (66–68).

Previous studies showed that alkaline pH represses ctxAB expression in classic biotypes of V. cholerae (5). Here, we show that this phenotype is conserved in El Tor biotypes and is in part due to OmpR-dependent repression of aphB at alkaline pH. Alkaline pH strongly repressed CT production at both 6 h and 24 h. Attenuated CT production at 6 h during growth at high pH was abrogated by deletion of ompR. However, ompR deletion only partially restored CT production at 24 h at alkaline pH. This finding indicates that OmpR is primarily responsible for alkaline pH-induced virulence repression at early time points during growth under AKI conditions and that other factors contribute to virulence repression at later time points. This observation is consistent with the ToxR regulon model, in which AphB, the target of OmpR repression, functions early during induction of the ToxR regulon. There are multiple OmpR-independent mechanisms that may also contribute to virulence repression in response to alkaline pH. For example, PepA, Crp, and TcpI have all been reported to contribute to virulence regulation in response to alkaline pH in classical biotypes (50, 51, 69). In El Tor strains, ToxR proteolysis has been shown to occur in response to alkaline pH (70).

We speculate that alkaline induction of ompR reflects an evolutionally adaptation of V. cholerae to its unique life cycle. Vibrio spp. are native to alkaline marine environments (17, 71) and have evolved efficient mechanisms to grow at high osmolarity and alkaline pH. This adaptation serves V. cholerae well in the human gut, where V. cholerae can rapidly multiply in the alkaline rice water stool before disseminating in the diarrheal purge to reenter aquatic ecosystems. The ability of V. cholerae to cycle between these two disparate environments requires adaptive transcriptional rewiring. Late in infection, in preparation for exiting the host and entering the aquatic ecosystem, V. cholerae represses virulence genes and induces genes important in transmission and dissemination (9, 12, 15, 16). The signals and regulatory mechanisms that control this transcriptional shift are poorly understood. Both seawater and the rice water stool of cholera patients are markedly alkaline (64, 72), and the high pH in the marine environment is an important signal that has been shown to induce adaptation though transcriptional regulation (73). Therefore, we postulate that the alkaline pH-induced expression of ompR may represent a genetic mechanism that functions during the late stages of infection to repress virulence gene expression and to adapt to alkaline pH in the lumen, thus preparing the bacteria for entrance back into the aquatic ecosystem (9, 12, 15, 16). This speculative model is supported by our results showing that alkaline pH induction of ompR results in the transcriptional silencing of ATR and virulence genes and that OmpR contributes to fitness in alkaline pH and activates the expression of the chitin-specific porin chiP (74, 75). The transition of pathogenic V. cholerae between the host gastrointestinal tract and aquatic ecosystems may have selected for the divergent evolution of V. cholerae OmpR to respond to alkaline pH, an adaptation that appears to be specifically suited to the V. cholerae life cycle.

MATERIALS AND METHODS

Bacterial strains and culture conditions.

The bacterial strains and plasmids used in this study are listed in Table S2 in the supplemental material. E. coli strain EC100Dpir+ was used for cloning. V. cholerae stain JB58 was used as the WT strain in all experiments. Bacterial strains were grown at 37°C in LB (56) or on LB agar. AKI growth conditions were used to induce V. cholerae virulence gene expression, as described previously (76, 77). Antibiotics were used at the following concentrations: streptomycin, 100 μg/ml; carbenicillin, 100 μg/ml; chloramphenicol, 20 μg/ml for E. coli and 1 μg/ml for V. cholerae.

Growth curve experiments.

Growth curves were generated in 96-well microtiter plates. Overnight LB cultures of WT and ΔompR strains were diluted 1:10,000 in fresh LB containing either 0 mM NaOH (pH 7.4), 20 mM NaOH (pH 8.7), or 35 mM NaOH (pH 9.7); 0.05% l-arabinose and 1 μg/ml chloramphenicol were added to the medium for the complementation studies. Two hundred microliters of the diluted cultures was then aliquoted into triplicate wells of a 96-well microtiter plate, the plates were incubated at 37°C in a BioTek ELX-808 microplate reader, with shaking, and growth was monitored every 30 min as the optical density (OD) at 630 nm.

Transcriptional reporter assays.

V. cholerae strains harboring the indicated lacZ reporters were cultured under AKI conditions or in LB. At the indicated times, aliquots were collected in triplicate and β-galactosidase activity was quantified as described previously (78). For experiments assessing transcriptional responses to alkaline pH, strains harboring the indicated lacZ transcriptional reporter fusions were cultured under AKI conditions for 4 h, at which time they were treated with 20 mM NaOH to raise the pH to ∼8.7 or treated with water as the control. Note that pH 8.7 was used for all of the transcriptional reporter analyses because it did not affect the growth of any of the analyzed strains under the conditions tested. The cultures were then incubated for an additional 1 h, with shaking, before culture aliquots were collected in triplicate and β-galactosidase production was assessed. All of the transcriptional reporter experiments were performed independently at least three times.

RNA-seq.

V. cholerae strain JB58, harboring either pBAD33-ompR (pTB11) or the empty vector pBAD33, was cultured under AKI conditions in the presence of 0.05% arabinose for 4.5 h, at which time total RNA was isolated using TRIzol, according to the manufacturer’s directions (Invitrogen), and further purified using a RNeasy kit with in-column DNase treatment (Qiagen). The resulting RNA was then processed and sequenced by the University of Pittsburgh Health Sciences Sequencing Core at Children’s Hospital of Pittsburgh. The methods for RNA processing, sequencing, and RNA-seq analysis have been described (25). The resulting FASTQ files from three independent experiments were mapped to the V. cholerae reference genome (GenBank accession numbers NC_002505 and NC_002506) using CLC Genomics Workbench (version 10.1; Qiagen) and default mapping parameters. Sample normalization and the identification of differentially expressed genes were accomplished using the differential expression for RNA-seq tool in CLC Genomics Workbench. Genes showing ≥2.0-fold differences in expression and false discovery rate P values of ≤0.05 were identified as differentially expressed genes.

Quantification of CT production.

CT production was determined by GM1 enzyme-linked immunosorbent assays (ELISAs), as described previously, using purified CT (Sigma) as a standard (79).

qRT-PCR.

V. cholerae strains harboring either pBAD33-ompR or pBAD33 empty vector were grown under AKI conditions with 0.05% arabinose for 5 h, at which time total RNA was isolated from the cultures using TRIzol (Invitrogen), according to the manufacturer’s instructions (80–84). cDNA was generated from the purified RNA using the Maxima first-strand cDNA synthesis kit (Thermo Fisher Scientific). The expression levels of specific genes were quantified by amplifying 25 ng of cDNA with 0.3 μM primers using the SYBR green PCR mix (Thermo Fisher Scientific) on a StepOnePlus real-time PCR system (Applied Biosystems). The relative expression level of genes in the mutant and WT cultures were calculated using the 2^−ΔΔCT method. The qRT-PCR primers used in these experiments are listed in Table S2. The results presented are the means ± standard deviations (SDs) from three biological replicates, with each biological replicate being generated from three technical replicates. DNA gyrase (gyrA) was used as the internal control.

Data availability.

The raw RNA-seq data files have been deposited at the National Center for Biotechnology Information Sequence Read Archive under accession number SRP109296.