ABSTRACT
Like many other pathogens, Vibrio cholerae, the causative agent of cholera, can modulate its gene expression to combat stresses encountered in both aquatic and host environments, including stress posed by reactive oxygen species (ROS). We previously reported that the virulence activator AphB in V. cholerae is involved in ROS resistance. In this study, we found that another key virulence regulator, ToxR, was important for V. cholerae resistance to hydrogen peroxide. Through a genome-wide transposon screen, we discovered that a deletion in mneA, which encodes a manganese exporter, restored ROS resistance of the toxR mutant. We then showed that ToxR did not affect mneA transcription but that the ToxR-regulated major porin OmpU was critical for ROS resistance. The addition of manganese in culture medium restored ROS resistance in both the toxR and ompU mutants. Furthermore, elemental analysis indicated that the intracellular concentration of manganese in both the toxR and ompU mutants was reduced. This may result in intracellular ROS accumulation in these mutants. Our data suggest that ToxR plays an important role in the resistance to reactive oxygen species through the regulation of manganese transport.
INTRODUCTION
The human pathogen Vibrio cholerae, a motile Gram-negative bacterium, is the causative agent of cholera (1, 2), which is still a major threat to public health in the developing world (3). V. cholerae survives in various environments by sensing and responding to environmental cues. Within a human host, V. cholerae senses signals such as changing oxygen tension and the presence of bile salts and bicarbonate, enabling the activation of a regulatory cascade leading to virulence gene expression (4–7). The master virulence regulator ToxT controls the expression of an array of virulence factors, including cholera toxin. The expression of toxT requires two transmembrane regulators, ToxR and TcpP (5); TcpP is in turn regulated by two unlinked regulators, AphA and AphB (Fig. 1A). V. cholerae also encounters oxidative stress during the later stages of infection (8, 9), as well as in the aquatic environment (10). In V. cholerae, the reactive oxygen species (ROS)-sensing activator OxyR is important for resistance to hydrogen peroxide (11), while OhrR, a regulator of the organic hydroperoxide resistance gene ohrA, regulates V. cholerae resistance to organic hydroperoxides (12). Quorum sensing systems (13) and the virulence regulator AphB also play important roles in the oxidative stress response (4, 12). In addition, V. cholerae may modulate mutation frequency as a temporal adaptive strategy to overcome oxidative stress and enhance infectivity (15).
The roles of virulence regulators in ROS resistance. (A) Virulence regulon in V. cholerae (adapted from references 5 and 40). (B) The importance of ToxR in H2O2 resistance. Overnight cultures of V. cholerae strains were inoculated into AKI medium containing appropriate antibiotics and incubated statically for 4 h at 37°C. When indicated, 0.1% arabinose was included in the medium to induce the pBAD promoter. Cultures were then diluted at 1:10 into saline with or without 300 μM hydrogen peroxide and further incubated for 1 h at 37°C with aeration. The number of CFU was determined by serial dilution and plating. The survival rate was determined by CFU counts of H2O2-treated cultures divided by those without H2O2. Data are the means and standard deviation (SD) of the results from three independent experiments. *, Student's t test, P < 0.05.
To deal with the ROS threat, bacteria produce inducible defensive enzymes and proteins. For example, in V. cholerae, similar to other bacteria, it has been previously demonstrated that ROS-induced catalases (KatG and KatB), peroxiredoxin (PrxA), organic hydroperoxide resistance protein (OhrA), and the DNA-binding protein DPS are important for V. cholerae ROS resistance (11, 12, 16). In Escherichia coli, H2O2 also induces the expression of the manganese importer MntH (17). Manganese has been shown to help protect different bacteria against oxidative stress (18, 19). The mechanisms of these protective effects of manganese are not very clear. It was proposed that manganese might mitigate oxidative stress by chemically scavenging superoxide and hydrogen peroxide (20). More recently, Anjem et al. demonstrated that manganese can substitute iron to metallate mononuclear enzymes that are normally iron loaded, avoiding the iron-derived Fenton reaction to prevent protein damage (21). MntH-like manganese import proteins have been found in a wide range of bacteria (22). Additionally, some marine bacteria, including some Vibrio spp., utilize a MntX transporter to take up manganese (23). However, no manganese import systems have been found in V. cholerae. Recently, Fisher et al. identified a novel inner membrane protein, MneA, that is involved in manganese export in V. cholerae (24). In this study, we examined the role virulence regulators played in V. cholerae ROS resistance and found that ΔtoxR mutants were more sensitive to hydroperoxide than the wild type. Further analyses showed that the ToxR-activated outer membrane porin OmpU may be involved in manganese import and manganese homeostasis, both of which are important for V. cholerae ROS resistance.
RESULTS AND DISCUSSION
The virulence activator ToxR is involved in ROS resistance.Previous studies show that in V. cholerae, the virulence regulator AphB plays an important role in the oxidative stress response (12), where the reduced form of AphB represses the ROS resistance gene ohrA, whereas oxidized AphB derepresses ohrA. An AphBC235S mutant that is not responsive to oxidation is thus sensitive to oxidative stress due to the constitutive inhibition of ohrA expression. To examine whether other virulence regulators along the virulence regulatory cascade (Fig. 1A) play a role in ROS resistance, wild-type and aphA, toxR, tcpP, and toxT mutant strains were grown in the virulence-inducing AKI medium (25) statically until mid-log phase. The cultures were then exposed to H2O2 for 1 h. Under such a treatment, approximately 10% of the wild type survived (Fig. 1B). While deletions in aphA, tcpP, and toxT showed little effect, a deletion in toxR severely impaired V. cholerae’s H2O2 resistance (Fig. 1B). The expression of toxR in trans restored the H2O2 resistance in toxR mutants (Fig. 1B). In addition to activating virulence genes, ToxR also directly and indirectly modulates the expression of a number of genes involved in other cellular processes, such as cellular transport and iron uptake, and those encoding outer membrane proteins (26, 27). Under the conditions tested, ROS resistance was not changed in the deletion mutant of toxT, whose expression is activated by ToxR and TcpP (28), suggesting that V. cholerae virulence genes may not be involved in ROS resistance.
Disruption of manganese export restores ROS resistance in toxR mutants.To investigate how ToxR regulates ROS resistance, we performed a genome-wide transposon screen. We introduced a mariner transposon, pNJ17, which contains an arabinose-inducible outward promoter (29) (Fig. 2A), into the toxR mutants and screened for mutants with restored H2O2 resistance. From approximately 20,000 independent transposon insertions, we identified two arabinose-dependent and one arabinose-independent H2O2-resistant mutant (Fig. 2B). Arbitrary PCR and DNA sequencing revealed that the ΔtoxR-Tn1 and ΔtoxR-Tn2 mutants had transposons inserted in the promoter regions of katG (VC1560) and oxyR (VC2636), respectively. As KatG catalase and the oxidative stress response regulator OxyR are critical for V. cholerae resistance to H2O2 (11, 30) and the growth phenotypes of these two mutants were arabinose dependent, it is likely that overexpression of KatG or OxyR leads to ROS resistance in toxR mutants. We measured katG-lux expression in wild-type and toxR mutant strains with or without H2O2, and we found that katG transcription was not regulated by ToxR (Fig. 2C). As katG expression is activated by OxyR (11), these data suggest that toxR mutant sensitivity to H2O2 is not directly linked to KatG and OxyR. The ΔtoxR-Tn3 mutant had the transposon inserted in the coding sequence (at bp 441) of the mneA gene, which encodes a manganese exporter (24). This mutant displayed arabinose-independent H2O2 resistance (Fig. 2B), suggesting that the deletion of mneA in the toxR mutant restored ROS resistance.
Genome-wide screens for ROS resistance-restored mutants in toxR mutants. (A) Genetic structure of the transposon used in the screens. The mariner transposon (Tn) contains an arabinose-inducible promoter and a kanamycin resistance cassette (Kanr) (29). (B) ROS resistance of selected mutants obtained from the screens. Overnight cultures of wild-type (WT), ΔtoxR mutant, and ΔtoxR::Tn mutant strains were inoculated into AKI medium with or without 0.1% arabinose and incubated statically for 4 h at 37°C. Cultures were treated with 300 μM hydrogen peroxide for 1 h at 37°C with aeration. The number of CFU was determined by serial dilution and plating. The survival rate was determined by CFU counts of H2O2-treated cultures divided by those without H2O2. Data are the means and standard deviation (SD) of the results from three independent experiments. (C) katG expression. WT and toxR mutants containing PkatG-luxCDABE plasmids were grown in AKI statically for 4 h at 37°C. When indicated, 50 μM H2O2 was added, and lux activity was measured after 2 h and normalized against the optical density at 600 nm (OD600). Data are the means and standard deviation (SD) of the results from three independent experiments. *, Student's t test, P < 0.05; ns, nonsignificant.
To further study the relationship between ToxR, MneA, and ROS resistance, we first confirmed the Tn screen results by constructing an in-frame deletion of mneA in the wild-type and toxR mutant backgrounds. Figure 3A shows that while mneA mutants did not affect ROS resistance, the mneA deletion in the toxR mutation background restored ROS resistance to the wild-type level. While overexpression of mneA in the wild type and in mneA mutants had little effect, the expression of mneA in trans in toxR mneA double mutants complemented the mneA phenotype (Fig. 3A), indicating that MneA may be the reason for toxR mutants directly or indirectly succumbing to H2O2 insults. In Fisher et al.’s study (24), MneA is a manganese exporter in classical V. cholerae strains, and an mneA deletion displayed a high-manganese growth defect. To test whether the MneA protein has a similar function in El Tor, we compared the growth of El Tor wild-type and mneA mutant strains in LB in the absence or presence of additional manganese. We found that at high Mn2+ concentrations, the mneA deletion in both the wild-type and toxR mutant strains resulted in growth defects (Fig. 3B), confirming the similar role of MneA in manganese export in the El Tor biotype. As manganese has been shown to help protect different bacteria against oxidative stress (18, 19) and the deletion of the MneA manganese exporter rescued ROS resistance in toxR mutants, we then tested whether the addition of manganese may increase ROS resistance in toxR mutants. We found that when 10 μM MnCl2 was added to the AKI medium, toxR mutants demonstrated a level of H2O2 resistance comparable to that of the wild type (Fig. 3C). However, this effect was not seen when 10 μM MgCl2 was added (Fig. 3C). These data suggest that ToxR may be involved in the regulation of cellular manganese homeostasis.
The effects of manganese on V. cholerae growth and ROS resistance. (A) MneA effects on ROS resistance of the ΔtoxR mutant. V. cholerae strains were grown in AKI for 4 h at 37°C. Ampicillin and 0.1% arabinose (ara), when indicated, were included for strains containing pBAD-mneA. Cultures were treated with 300 μM hydrogen peroxide for 1 h at 37°C with aeration. The number of CFU was determined by serial dilution and plating. The survival rate was determined by CFU counts of H2O2-treated cultures divided by those without H2O2. Data are the means and standard deviation (SD) of the results from three independent experiments. *, Student's t test, P < 0.05. (B) Growth of the ΔtoxR and ΔmneA mutants with or without manganese. V. cholerae strains were grown in LB without (left) or with (right) 10 μM MnCl2 at 37°C with aeration. The OD600 was measured. (C) Manganese effects on ROS resistance of the ΔtoxR mutant. V. cholerae strains were grown in AKI medium as described above in the absence or presence of 10 μM MnCl2 or MgCl2, and the ROS resistance of these strains was determined as described above.
The ToxR-activated OmpU porin is involved in manganese transport and ROS resistance.As the mneA toxR double-deletion strain displayed ROS resistance, we first tested whether ToxR regulates mneA transcription. We constructed an mneA promoter-luxCDABE transcriptional fusion reporter plasmid and introduced it into the wild-type and toxR mutant strains. We grew these strains under AKI conditions with or without additional MnCl2. We found that MnCl2 induced mneA expression (Fig. 4A), consistent with findings from the previous report (24). However, the mneA expression level in the toxR mutant was similar to that in the wild type. Moreover, overexpression of toxR in E. coli did not activate mneA expression (Fig. 4B). These data suggest that ToxR does not regulate the mneA promoter under the growth conditions tested.
The effects of manganese on mneA expression. (A) Wild-type V. cholerae and ΔtoxR mutants containing PmneA-luxCDABE reporter plasmids were grown in AKI medium with or without 50 μM MnCl2 at 37°C for 2 h. Luminescence was measured and normalized against the OD600. The results are the means and standard deviations of the results from three independent experiments. *, Student's t test, P < 0.05. (B) E. coli harboring PmneA-luxCDABE reporter plasmids with either pBAD24 vector or pBAD-toxRS was grown in LB containing 0.1% arabinose at 37°C for 2 h. Luminescence was measured and normalized against the OD600. The results are the means and standard deviations of the results from three independent experiments.
Among the genes directly regulated by ToxR, both ompU (activated by ToxR) and ompT (repressed by ToxR) encode outer membrane porins (31). It has been reported that in the Gram-negative bacterium Bradyrhizobium japonicum, an outer membrane protein acts as a channel to facilitate the translocation of Mn2+ (32). We therefore hypothesized that ToxR-regulated outer membrane proteins may contribute to manganese transport and thus to ROS resistance as well. To test this, we constructed in-frame deletions of ompU and ompT and tested the ROS resistance of these mutants. We found that ompU mutants, but not ompT mutants, were sensitive to H2O2 treatment, similar to toxR mutants (Fig. 5A). Overexpression of ompU in toxR mutants restored the ROS resistance (Fig. 5A), suggesting that ToxR-activated ompU expression is important for resistance to H2O2 insults. Furthermore, the addition of 10 μM MnCl2 in the medium restored ompU mutant ROS resistance (Fig. 5B), implying that OmpU may be involved in manganese transport. Moreover, the deletion of the manganese export gene mneA in an ompU mutant restored ROS resistance (Fig. 5A), further supporting the link between OmpU and manganese transport. To confirm this, we examined intracellular manganese concentrations of the wild type and different mutants using inductively coupled plasma-mass spectrometry (ICP-MS). Figure 5C shows that both toxR and ompU mutant cells contained less intracellular Mn2+ than the wild type. Deleting the Mn2+ exporter in toxR mutants or constitutively expressing ompU (and thereby bypassing the ToxR regulation) in toxR mutants increased the intracellular concentration of Mn2+ (Fig. 5C). These data suggest that OmpU may contribute to manganese uptake and is important for ROS resistance.
The role of OmpU in ROS resistance and Mn2+ transport. (A and B) V. cholerae strains were grown in AKI medium statically for 4 h at 37°C. Ampicillin and 0.1% arabinose were included in the medium to induce the pBAD promoter. When indicated (in panel B), 10 μM MnCl2 was added to the medium. H2O2 resistance was then determined, and the survival rate was calculated by the CFU counts of H2O2-treated cultures divided by those without H2O2. Data are the means and standard deviation (SD) of the results from three independent experiments. *, Student's t test, P < 0.05. (C) Intracellular manganese concentrations in V. cholerae strains. V. cholerae strains were grown in AKI medium statically for 4 h at 37°C. Ampicillin and 0.1% arabinose were included in the medium to induce the pBAD promoter. PBS-washed cells were digested with nitric acid overnight. The samples were then subjected to an inductively coupled plasma-mass spectrometry (ICP-MS; Agilent 7900) analysis. Data presented are the means and standard deviations from three analyses. *, Student's t test, P < 0.05.
To confirm our findings that ToxR-regulated manganese transport plays an important role in ROS resistance, we examined the intracellular accumulation of ROS in the aforementioned strains. We grew those strains in AKI medium with or without the addition of 30 μM MnCl2 and then exposed them to H2O2. Intracellular ROS was measured by staining the bacterial cells with the general ROS indicator 2′,7′-dichlorodihydrofluorescein diacetate (H2DCFDA). Figure 6A shows that without the addition of manganese, ROS accumulation was significantly higher in toxR and ompU mutants than in the wild type. Deleting mneA or overexpressing ompU in toxR mutants, in which intracellular Mn2+ was increased, decreased ROS accumulation in these cells (Fig. 6A). The addition of Mn2+ to the medium significantly reduced ROS accumulation in the toxR and ompU mutants (Fig. 6A). These results again suggest that cellular manganese homeostasis plays an important role in V. cholerae ROS resistance. It also suggests that OmpU is not the sole manganese transporter, as the addition of a high concentration of exogenous manganese rescued the ROS resistance in both toxR and ompU mutants.
The relationship between intracellular manganese and ROS accumulation. (A) Intracellular ROS accumulation. V. cholerae strains were grown in AKI medium statically at 37°C for 4 h. Arabinose (0.1%) was included for strains containing pBAD-ompU. When indicated, 30 μM MnCl2 was included in the culture. Samples were collected, washed with PBS, resuspended in PBS containing 100 μM H2O2, and incubated at 37°C statically for 1 h. PBS-washed samples were then incubated with 10 μM H2DCFDA for 0.5 h at 37°C. Each sample was washed, and the ROS level was measured using a FACSCalibur flow cytometer and determined with the use of mean fluorescent intensity (MFI). Data presented are the means and standard deviations from three experiments (10,000 events/experiment). *, 2-way analysis of variance (ANOVA), P < 0.05. AU, arbitrary units. (B) Working model. ToxR upregulated the expression of ompU. Manganese may be imported through OmpU directly or indirectly and exported through MneA to maintain the intracellular and extracellular balance, which is important for V. cholerae ROS resistance. IM, inner membrane; OM, outer membrane.
In this study, we surveyed whether any V. cholerae virulence regulators are involved in ROS resistance and found that a mutation in the key virulence regulatory gene toxR led to hypersensitivity toward H2O2 treatment. Based on the genetic and biochemical evidence, we propose that the ToxR-activated outer membrane protein OmpU is involved in the uptake of manganese, which is important for ROS resistance (Fig. 6B). Deletion of the manganese exporter mneA in a toxR deletion mutant increases the intracellular concentration of Mn2+ and thus restores ROS resistance. The link between virulence regulation and ROS resistance may reflect the need of V. cholerae to fight against ROS during colonization. Another key virulence regulator, AphB, has been shown to modulate ROS resistance as well (12). Of note, we do not know whether OmpU transports Mn2+ directly. It has been reported that Salmonella outer membrane proteins are involved in regulating membrane permeability to control the influx of ROS for defense against oxidative stress (33). Further investigation is needed to test whether V. cholerae ToxR-regulated outer membrane porins are important for the regulation of membrane permeability.
MATERIALS AND METHODS
Strains and growth conditions.V. cholerae E1 Tor C6706 (34) was used as the wild-type strain in this study. Bacteria were propagated at 37°C in LB medium with appropriate antibiotics and 1.5% agar when necessary. AKI medium (25) was used as the virulence-inducing condition. In-frame deletions of mneA, ompU, and ompT were constructed by the natural cotransformation and multiplex genome editing by natural transformation (MuGENT) approach (35, 36). The toxR, tcpP, aphA, and toxT mutants have been described previously (13). The transcriptional PmneA-lux reporter was constructed by cloning the promoter sequence of mneA into the pBBR-lux vector, which contains a promoterless luxCDABE reporter (37). The PkatG-lux reporter plasmid construct was previously described (11). A plasmid containing the pBAD inducible promoter for overexpressing ompU was constructed by cloning the PCR-amplified ompU coding regions into pBAD24 (38). The pBAD-toxRS construct was described previously (4). The primer sequences used for the study are available upon request.
H2O2 killing assays.Overnight cultures of V. cholerae strains were inoculated at a 1:1,000 dilution into fresh AKI medium and incubated statically for 4 h at 37°C. When indicated, 10 μM MnCl2 was included in the culture. Cultures were then diluted 1:10 into saline (0.8% NaCl) with or without 300 μM hydrogen peroxide and further incubated for 1 h at 37°C with aeration. The number of viable cells was determined by serial dilutions and plating on LB plates with the appropriate antibiotics. The survival rate was calculated by dividing the CFU of H2O2-treated samples by those of untreated samples.
Screening of transposon mutations in ΔtoxR mutants that restored ROS resistance.The ΔtoxR mutants were mutagenized by a mariner transposon (Tn) containing an outward arabinose-inducible promoter (29). The libraries containing over 2 × 104 independent transposon insertions were pooled, and H2O2 killing assays were performed. Arabinose (0.1%) was included during the library growth prior to the H2O2 treatment. The surviving mutants were collected, and a second round of H2O2 treatment was performed. After three rounds of enrichments, individual mutants were examined for ROS resistance. The transposon insertion was determined by arbitrary PCR and DNA sequencing (39).
Measuring mneA expression.Overnight cultures of V. cholerae or E. coli strains containing the PmneA-luxCDABE transcriptional fusion plasmids were inoculated at a 1:100 dilution into fresh LB medium with appropriate antibiotics in the absence or in the presence of 50 μM MnCl2. To induce toxRS in E. coli, 0.1% arabinose was included. After a 2-h incubation at 37°C with aeration, luminescence was measured, and the lux units were normalized for growth against the optical density at 600 nm.
ROS detection.Overnight cultures of V. cholerae strains were inoculated at a 1:1,000 dilution into fresh AKI medium and incubated statically for 4 h at 37°C. When indicated, 30 μM MnCl2 was included in the culture. Samples were collected, washed with phosphate-buffered saline (PBS), resuspended in PBS containing 100 μM H2O2, and incubated at 37°C statically for 1 h. Samples were then centrifuged, collected and washed twice with PBS, and incubated in 1 ml PBS containing 10 μM H2DCFDA ROS probe (Abcam, MA, USA) for 0.5 h at 37°C in the dark. Each sample was washed twice with cold PBS, and the ROS level was measured using a FACSCalibur flow cytometer (BD Immunocytometry Systems, NJ, USA) and determined with the use of mean fluorescent intensity (MFI). The resulting data were analyzed with the FlowJo 7.6.1 software (Ashland, OR, USA).
Inductively coupled plasma-mass spectrometry to determine cellular Mn concentration.Overnight cultures of V. cholerae strains were inoculated at a 1:100 dilution into AKI medium (with antibiotics and 0.1% arabinose when necessary) and incubated statically for 4 h at 37°C. The suspension was washed once with 0.8% (wt/vol) NaCl saline, and the pellets were resuspended in 1 M trace-metal-grade nitric acid (Fisher Scientific) and incubated at 30°C overnight. Cells were extracted with a diluted acid solution (2% [vol/vol] HNO3 and 0.5% [vol/vol] HCl) and then centrifuged to remove cellular debris. A 55Mn analysis was completed using an inductively coupled plasma-mass spectrometer (ICP-MS; Agilent 7900) using an online internal standard (103Rh) and He as a collision gas. Daily ICP-MS quality assurance steps included instrument tuning and determination of the detection limit and limit of quantification, which were batch dependent but generally ∼0.05 and 0.3 μg/liter, respectively.
ACKNOWLEDGMENTS
We thank Erica McKenzie for helping with ICP-MS analysis. We also thank Yuning Wang and Xiaolin Xin for technical help and Mark Goulian for helpful discussion.
This study was supported by the National Key Basic Research Program of China (grant 2013CB531604 to X.-H.G.), the Natural Science Foundation of Jiangsu Province (grant BK20191314 to H.W.), and the NIH/NIAID (grant R01AI120489 to J.Z.).
FOOTNOTES
- Received 17 December 2019.
- Accepted 18 December 2019.
- Accepted manuscript posted online 23 December 2019.
- Copyright © 2020 American Society for Microbiology.
