ZntA maintains zinc and cadmium homeostasis and promotes oxidative stress resistance and virulence in Vibrio parahaemolyticus

ABSTRACT

Although metals are essential for life, they are toxic to bacteria in excessive amounts. Therefore, the maintenance of metal homeostasis is critical for bacterial physiology and pathogenesis. Vibrio parahaemolyticus is a significant food-borne pathogen that mainly causes acute gastroenteritis in humans and acute hepatopancreatic necrosis disease in shrimp. Herein, we report that ZntA functions as a zinc (Zn) and cadmium (Cd) homeostasis mechanism and contributes to oxidative stress resistance and virulence in V. parahaemolyticus. zntA is remarkably induced by Zn, copper, cobalt, nickel (Ni), and Cd, while ZntA promotes V. parahaemolyticus growth under excess Zn/Ni and Cd conditions via maintaining Zn and Cd homeostasis, respectively. The growth of ΔzntA was inhibited under iron (Fe)-restricted conditions, and the inhibition was associated with Zn homeostasis disturbance. Ferrous iron supplementation improved the growth of ΔzntA under excess Zn, Ni or Cd conditions. The resistance of ΔzntA to H₂O₂-induced oxidative stress also decreased, and its virulence was attenuated in zebrafish models. Quantitative real-time PCR, mutagenesis, and β-galactosidase activity assays revealed that ZntR positively regulates zntA expression by binding to its promoter. Collectively, the ZntR-regulated ZntA is crucial for Zn and Cd homeostasis and contributes to oxidative stress resistance and virulence in V. parahaemolyticus.

KEYWORDS:

• Vibrio parahaemolyticus
• zinc homeostasis
• cadmium homeostasis
• ZntA
• oxidative stress resistance
• virulence

Introduction

Trace metals, such as iron (Fe), manganese (Mn), zinc (Zn), copper (Cu), cobalt (Co) and nickel (Ni), are essential nutrients for all living organisms. They are required by many enzymes to act as cofactors and thus stimulate their catalytic activity; additionally, they are structural components of various metalloproteins.^1,2 Restricting the availability of metals to invading pathogens is an important strategy developed by vertebrate hosts to control bacterial infections.³ To establish infection, bacteria have evolved diverse mechanisms to defend themselves against host-imposed metal restriction. For example, Yersinia pestis synthesizes yersiniabactin to overcome Fe and Zn restriction during host infection.⁴ Although metals are essential, they are toxic to bacteria when they accumulate in excessive amounts. A common cause of metal toxicity is the mismetallation of metalloproteins with nonpreferred metals.⁵ Furthermore, certain metals may produce reactive oxygen species via the Fenton reaction, thereby causing cell damage.⁶ Metal toxicity is another strategy exploited by hosts to control bacterial infections.⁷ Nonetheless, bacteria have developed countermeasures to prevent damage caused by metal overload, such as excess metal efflux or sequestration.⁵

Vibrio parahaemolyticus is a gram-negative, curved rod-shaped, halophilic bacterium widely distributed in temperate and tropical marine and coastal waters, and can be frequently isolated from seafood, including fish, shrimp, and shellfish.^8,9 It is a significant food-borne pathogen that mainly causes acute gastroenteritis, which is generally associated with the consumption of undercooked or raw contaminated seafood.¹⁰ In some cases, V. parahaemolyticus may cause septicemia in individuals with comorbidities, such as liver disease, or wound infections when open wounds are exposed to contaminated seawater.¹¹ An outbreak of V. parahaemolyticus-associated acute gastroenteritis was first reported in 1950 in Osaka, Japan, resulting in 272 cases and 20 deaths. Since then, this bacterium has become the leading cause of seafood-derived food poisoning worldwide.¹⁰ In the USA, V. parahaemolyticus causes approximately 50 000 infections per year.⁹ In Japan, this pathogen is responsible for 20%-30% of all food poisoning cases.¹⁰ V. parahaemolyticus ranks first in the prevalence of major pathogens in Chinese foods and becomes the leading cause of infectious diarrhea in China.^12,13 Moreover, it is a major causative agent of acute hepatopancreatic necrosis disease (AHPND) in shrimp, leading to a mortality rate of up to 100% within 30–35 days of stocking.^14,15 Since the first outbreak in China in 2009, AHPND outbreaks have been documented from various regions including Thailand, Mexico, South American, the USA, South Korea, and Japan.¹⁵ AHPND was estimated to cause collective losses over USD 43 billion across Asia (China, Thailand, Vietnam, Malaysia) and in Mexico.¹⁶

The presence of Fe in lysed erythrocytes contributes to V. parahaemolyticus virulence in mice.¹⁷ Swarming and type III secretion systems can be regulated by calcium (Ca) and Fe in V. parahaemolyticus.¹⁸ During Litopenaeus vannamei infection, Fe, Mn, Zn, and Cu concentrations in the serum and hepatopancreas of this host are dynamically changed, revealing that V. parahaemolyticus must respond to metal signals during the infection process.¹⁹ The mechanisms of metal homeostasis in V. parahaemolyticus have been preliminarily revealed. V. parahaemolyticus employs multiple strategies, including the use of iron-protein receptors, proteases, siderophores, and xenosiderophores, to acquire Fe during host colonization.²⁰ ZnuA, a horizontally acquired Zn transporter, is involved in V. parahaemolyticus virulence.²¹ Recently, we reported that the DmeRF system is required for Co homeostasis in V. parahaemolyticus, where DmeF functions as a Co efflux pump and DmeR acts as a repressor the dmeRF operon.²² Moreover, the Zur-regulated Zn-binding protein ZrgA plays a predominant role in Zn acquisition in V. parahaemolyticus.²³ RNA sequencing revealed that zntA (VP_RS04700) encoding a metal-transporting ATPase is the most highly upregulated gene of V. parahaemolyticus grown under Zn-replete conditions compared with that under Zn-deficient conditions.²³ However, the role of ZntA in the homeostasis of Zn and other metals remains unclear. Furthermore, little information is available regarding the effects of ZntA on the physiology and pathogenesis of V. parahaemolyticus.

In this study, we showed that ZntA, which is positively regulated by ZntR, is essential for Zn and cadmium (Cd) homeostasis in V. parahaemolyticus. ZntA also contributes to oxidative stress resistance and virulence in this organism.

Results

Zn, Cu, Co, Ni, and Cd significantly induce zntA expression in V. parahaemolyticus

In a previous study, zntA was found to be the most highly upregulated gene during V. parahaemolyticus growth under Zn-replete conditions.²³ ZntA has been identified as a Zn-transporting P-type ATPase.^24,25 Additionally, ZntA is involved in Cd homeostasis in Klebsiella pneumoniae.²⁶ In the present study, quantitative real-time PCR (qRT-PCR) analysis was performed to detect zntA expression following treatment with elevated concentrations of various metals, and thus explore the involvement of ZntA in V. parahaemolyticus response to Zn and other metals. The data showed that zntA expression was strongly induced (approximately 19-fold) upon exposure to 0.5 mM Zn (Figure 1). Additionally, zntA was upregulated by approximately 3-, 18-, 8-, and 19-fold following treatment with Cu, Co, Ni, and Cd, respectively (Figure 1). However, zntA expression was not induced by Fe(II) or Mn (Figure 1). These results suggest that ZntA may be involved in V. parahaemolyticus response to excess Zn, Cu, Co, Ni, and Cd.

Figure 1. V. parahaemolyticus upregulates zntA expression in response to various metals. The RIMD2210633 strain in the early exponential phase was treated with H₂O (control), 1 mM FeSO₄, 1 mM MnSO₄, 0.5 mM ZnSO₄, 1 mM CuSO₄, 0.5 mM CoSO₄, 1 mM NiSO₄, or 125 μM CdSO₄ for 15 min. Total RNA was isolated, and quantitative real-time PCR analysis was performed to determine zntA expression, which is expressed relative to the control sample. Results represent the means and standard deviations from three independent experiments. Significance was determined using one-way analysis of variance along with Bonferroni’s posttest. ***, p < .001.

ZntA is required for V. parahaemolyticus growth under excess Zn, Ni, or Cd conditions

The zntA deletion and complemented strains were constructed (Fig. S1A), and their growth in the medium supplemented with increased concentrations of metals was evaluated to further investigate the role of zntA in V. parahaemolyticus response to excessive metal. The wild-type (WT), ΔzntA, and CΔzntA strains exhibited identical growth in the absence of metal supplementation (Figure 2a). Nevertheless, Zn, Ni, and Cd elicited an inhibitory effect on the growth of ΔzntA compared with that of the WT and CΔzntA strains; the inhibition level was correlated with the concentrations of the three metals (Figure 2b–d). Additionally, the three strains displayed almost identical growth upon Fe(II), Mn, Cu, or Co supplementation (Fig. S2).

Figure 2. ZntA promotes V. parahaemolyticus growth under excess Zn, Ni, or Cd conditions. The strains were grown in TSB (A) supplemented with different concentrations of ZnSO₄ (B), NiSO₄ (C), or CdSO₄ (D), as indicated in each panel. Growth curves were drawn by measuring the OD₅₉₅. The graphs show the means and standard deviations from three wells in a representative experiment.

The WT, ΔzntA, and CΔzntA strains were treated with 100 μM Zn, 1.5 mM Ni, or 25 μM Cd for 6 h to further explain the effects of ZntA on V. parahaemolyticus resistance to Zn, Ni, and Cd toxicity. At 0, 3, and 6 h, aliquots were removed, serially diluted, and spotted onto Tryptic Soy Agar (TSA) plates. The three strains exhibited a similar ability to form colonies (Fig. S3). Therefore, Zn, Ni, and Cd could be bacteriostatic rather than bactericidal to the mutant.

Taken together, these results indicate that ZntA protects V. parahaemolyticus against Zn-, Ni-, and Cd-induced bacteriostatic effects.

ZntA protects V. parahaemolyticus against Zn/Ni and Cd toxicity via maintaining Zn and Cd homeostasis, respectively

The WT, ΔzntA, and CΔzntA strains grown in the presence of 25 μM Zn, 1 mM Ni, or 2.5 μM Cd were subjected to inductively coupled plasma optical emission spectroscopy (ICP-OES) analysis to better understand the mechanisms underlying ZntA-mediated resistance to Zn, Ni, and Cd toxicity. In the presence of 25 μM Zn, the intracellular Zn content accumulated in the mutant was significantly higher than that in the WT and CΔzntA strains (Figure 3a). Unexpectedly, in the presence of 1 mM Ni, the intracellular Ni content accumulated in ΔzntA was equivalent to that in the WT strain and slightly higher than that in CΔzntA (Figure 3b). In the presence of 2.5 μM Cd, ΔzntA accumulated significantly higher intracellular Cd content than the WT and CΔzntA strains (Figure 3c). Surprisingly, the Zn content that accumulated in ΔzntA was significantly higher than that in the WT and CΔzntA strains in the presence of 1 mM Ni (Figure 3d).

Figure 3. Quantified intracellular metal content in V. parahaemolyticus strains by ICP-OES. Intracellular Zn (A), Ni (B), and Cd (C) content of the strains grown in TSB supplemented with 25 μM ZnSO₄, 1 mM NiSO₄, and 2.5 μM CdSO₄, respectively. (D) Intracellular Zn content of the strains grown in TSB supplemented with 1 mM NiSO₄. Results represent the means and standard deviations from five biological samples. Significance was determined using one-way analysis of variance along with Bonferroni’s posttest. ns, no significant difference; **, p < .01; ***, p < .001.

The growth of the three strains in the presence of 1.5 mM Ni was further evaluated using Zn-deficient medium prepared with N,N,N′,N′-tetrakis (2-pyridylmethyl) ethylenediamine (TPEN), a Zn chelator, to verify whether ZntA-mediated resistance to Ni toxicity is correlated with Zn homeostasis. The three strains showed identical growth in the medium pretreated with DMSO (the solvent of TPEN) or 25 μM TPEN (Figure 4a, b). Supplementation with 1.5 mM Ni resulted in a substantial growth defect of ΔzntA in DMSO-pretreated medium (Figure 4c). Nevertheless, ΔzntA exhibited comparable growth after 1.5 mM Ni supplementation in the Zn-deficient medium (Figure 4d). Moreover, although 25 μM Zn or 1 mM Ni alone resulted in moderate growth inhibition of ΔzntA, their simultaneous supplementation led to severe growth defect of the mutant (Figure 4e).

Figure 4. Growth curves of V. parahaemolyticus strains in the presence of 1.5 mM Ni and/or TPEN. The strains were grown in TSB pretreated for 2 h with DMSO (A), 25 μM TPEN (B), DMSO-pretreated TSB supplemented with 1.5 mM NiSO₄ (C), TPEN (25 μM)-pretreated TSB supplemented with 1.5 mM NiSO₄ (D), or 25 μM ZnSO₄ plus with 1 mM NiSO₄ (E). Growth curves were drawn by measuring the OD₅₉₅. The graphs show the means and standard deviations from three wells in a representative experiment.

Collectively, these results revealed that ZntA contributes to V. parahaemolyticus resistance to Zn/Ni and Cd toxicity by maintaining Zn and Cd homeostasis, respectively.

ZntA is required for V. parahaemolyticus growth under Fe-restricted conditions

In Salmonella Typhimurium, zntA is required for fitness under Fe restriction.²⁷ Accordingly, we explored the role of zntA in V. parahaemolyticus response to Fe restriction. An Fe-restricted medium was prepared using the Fe chelator 2,2′-dipyridyl. qRT-PCR analysis revealed that zntA expression in V. parahaemolyticus grown under Fe-restricted conditions (Tryptic Soy Broth [TSB] pretreated with 150 μM 2,2ʹ-dipyridyl) was approximately 9-fold higher than that under control conditions (TSB pretreated with absolute ethanol; Figure 5a). In the medium supplemented with 100 μM 2,2ʹ-dipyridyl, the WT, ΔzntA, and CΔzntA strains exhibited comparable growth (Figure 5b). Nevertheless, in the medium supplemented with 125 or 150 μM 2,2ʹ-dipyridyl, the growth of ΔzntA was severely inhibited compared to that of the WT and CΔzntA strains (Figure 5c, d).

Figure 5. ZntA facilitates V. parahaemolyticus growth under Fe-restricted conditions by mediating Zn homeostasis. (A) Quantitative real-time PCR analysis of zntA expression in V. parahaemolyticus grown in TSB pretreated for 2 h with 150 μM 2,2ʹ-dipyridyl compared with absolute ethanol. Results represent the means and standard deviations from three independent experiments. Significance was determined using a two-tailed paired t-test. **, p < .01. (B-D) growth curves of the strains in TSB supplemented with different concentrations of 2,2ʹ-dipyridyl, as indicated in each panel (B-D). (E-F) growth curves of the strains in DMSO-pretreated TSB supplemented with 125 μM 2,2ʹ-dipyridyl (E), or TPEN (10 μM)-pretreated TSB supplemented with 125 μM 2,2ʹ-dipyridyl (F). Growth curves were drawn by measuring the OD₅₉₅. The graphs show the means and standard deviations from three wells in a representative experiment.

The growth of the three strains in Fe- and Zn-restricted media was evaluated to determine whether the growth defect of ΔzntA under Fe-restricted conditions could be attributed to Zn homeostasis disturbance. As expected, the growth of ΔzntA was severely impaired in the DMSO-pretreated medium supplemented with 125 μM 2,2ʹ-dipyridyl (Figure 5e). Conversely, the growth of ΔzntA was restored in TPEN (10 μM)-pretreated medium supplemented with 125 μM 2,2ʹ-dipyridyl (Figure 5f).

These results suggest that ZntA promotes V. parahaemolyticus growth under Fe-restricted conditions by maintaining Zn homeostasis.

Fe(II) supplementation could improve the growth of ΔzntA under excess Zn, Ni, or Cd conditions

A previous study showed that Fe supplementation partially restored the growth of Streptococcus pyogenes ΔmntE (a mutant of a Mn-specific efflux pump) at elevated Mn concentrations.²⁸ Therefore, we sought to explore whether the growth defect of V. parahaemolyticus ΔzntA under excess Zn, Ni, or Cd conditions could be alleviated by Fe(II) supplementation. The growth of ΔzntA was markedly inhibited in the presence of 100 μM Zn or 1.5 mM Ni (Figure 6a, c). Upon the addition of 0.25 mM Fe(II), the growth of ΔzntA was almost completely restored despite the presence of high Zn or Ni concentrations (Figures 6b, d). Fe(II) supplementation could also improve the growth of ΔzntA under excess Cd conditions, although the effect was not very profound (Figure 6e, f).

Figure 6. Growth defect of ΔzntA under excess Zn, Ni, or Cd conditions can be alleviated by Fe(II) supplementation. The strains were grown in TSB supplemented with 100 μM ZnSO₄ (A), 100 μM ZnSO₄ plus 0.25 mM FeSO₄ (B), 1.5 mM NiSO₄ (C), 1.5 mM NiSO₄ plus 0.25 mM FeSO₄ (D), 25 μM CdSO₄ (E), or 25 μM CdSO₄ plus 0.25 mM FeSO₄ (F). Trisodium citrate dihydrate (TCD) was also supplemented to the medium to alleviate Fe precipitation. Growth curves were drawn by measuring the OD₅₉₅. The graphs show the means and standard deviations from three wells in a representative experiment.

ZntA contributes to oxidative stress resistance and virulence in V. parahaemolyticus

In Agrobacterium tumefaciens, zntA inactivation confers hyper-resistance to H₂O₂.²⁹ In the present study, growth curve analysis was conducted to examine whether zntA deletion affects V. parahaemolyticus resistance to oxidative stress. The WT, ΔzntA, and CΔzntA strains exhibited comparable growth in the medium supplemented with 75 μM H₂O₂ alone (Figure 7a). In comparison, the growth of ΔzntA was completely inhibited in the medium supplemented with 75 μM H₂O₂ plus either 25 μM Zn or 1 mM Ni (Figure 7b, c). These results suggest that ZntA confers V. parahaemolyticus resistance to oxidative stress under excess Zn or Ni conditions.

Figure 7. ZntA is involved in oxidative stress resistance in V. parahaemolyticus under excess Zn or Ni conditions. The strains were grown in TSB supplemented with 75 μM H₂O₂ alone (A), 75 μM H₂O₂ plus 25 μM ZnSO₄ (B), or 75 μM H₂O₂ plus 1 mM NiSO₄ (C). Growth curves were drawn by measuring the OD₅₉₅. The graphs show the means and standard deviations from three wells in a representative experiment.

Zn homeostasis plays an important role in the virulence of various bacteria.³⁰ In the present study, the role of ZntA in V. parahaemolyticus virulence was assessed using zebrafish models. Initially, the survival curves of zebrafish infected intraperitoneally with the WT and ΔzntA strains were determined. All zebrafish infected with the WT strain (heat-killed) survived over the course of the experiment. The survival rates of zebrafish infected with ΔzntA were obviously higher than those infected with the WT stain, although the difference was not significant (p = 0.1028) (Figure 8a). Then, zebrafish were infected intraperitoneally with a 1:1 mixture of ΔzntA and the WT strain. At 24 h post-infection, the full intestines were collected, and the bacteria recovered from these samples were analyzed in terms of ΔzntA:WT. The competitive indexes (CIs) were significantly < 1 (Figure 8b), indicating that the ability of ΔzntA to colonize the intestine of zebrafish decreased compared with that of the WT strain. Finally, zebrafish were infected intramuscularly with a 1:1 mixture of ΔzntA and the WT strain. At 12 h post-infection, the muscle samples were collected, and the bacteria recovered from these samples were analyzed in terms of ΔzntA:WT. The CIs were significantly < 1 (Figure 8c), further indicating that the ability of ΔzntA to colonize and propagate in zebrafish decreased compared with that of the WT strain. Therefore, the virulence of ΔzntA was significantly attenuated in these models.

Figure 8. ZntA is involved in V. parahaemolyticus virulence in zebrafish models. (A) Survival curves of zebrafish infected intraperitoneally with the WT or ΔzntA strains. Zebrafish infected with the WT strain (heat-killed) served as the control. The data were analyzed using the log-rank test. (B) Competitive index (CI) of ΔzntA against the WT strain in the intestine of zebrafish. Eight zebrafish were injected intraperitoneally with a 1:1 mixture of ΔzntA and the WT strain. At 24 h post-infection, the full intestines were collected. Bacteria recovered from the samples were analyzed by colony PCR to determine the CI (the ΔzntA:WT ratio in each sample divided by that in the mixture). (C) CI of ΔzntA against the WT strain in the muscle of zebrafish. Eight zebrafish were injected intramuscularly with a 1:1 mixture of ΔzntA and the WT strain. At 12 h post-infection, the muscle samples were collected. Bacteria recovered from the samples were analyzed by colony PCR to determine the CI. Significance was determined using comparing the mean CI to 1 (equal competitiveness) via a two-tailed paired t-test. ***, p < .001.

zntA is positively regulated by ZntR in V. parahaemolyticus

Regulation of zntA by ZntR has been demonstrated in various bacteria.^29,31–37 However, their regulatory roles differ. For example, ZntR positively regulates zntA expression in Escherichia coli;^31–33 conversely, it silences zntA expression in Brucella abortus.³⁶ To assess the regulation of zntA expression by ZntR, we constructed the zntR deletion and complemented strains (Fig. S1B). qRT-PCR analysis revealed that zntA expression in ΔzntR was significantly lower than that in the WT and CΔzntA strains (Figure 9a), indicating that ZntR is a positive regulator of zntA in V. parahaemolyticus. A putative ZntR binding sequence, characterized by an inverted repeat, was recognized between the − 10 box and the start codon of zntA (Figure 9b). To determine whether ZntR regulates zntA by binding to this sequence, we replaced this sequence with an unrelated DNA sequence to generate P_zntA’, the mutant promoter of zntA (Fig. S4). We further performed β-galactosidase activity assays. The WT strain harboring P_zntA-pDM8 produced significantly higher β-galactosidase activities in the presence of Zn, Ni, and Cd than it did in control sample; additionally, β-galactosidase activities in the presence of Cu and Co were obviously higher than that in control sample, although the difference was not significant (Figure 9c). Nevertheless, ΔzntR harboring this plasmid produced baseline levels of β-galactosidase activity in the presence of Zn, Ni, and Cd (Figure 9d). Furthermore, the WT strain harboring P_zntA’-pDM8 produced baseline levels of β-galactosidase activity in the presence of Zn, Ni, and Cd (Figure 9d).

Figure 9. ZntR positively regulates zntA expression in V. parahaemolyticus. (A) Quantitative real-time PCR analysis of zntA expression in the WT, ΔzntR, and CΔzntR strains grown in TSB. Results represent the means and standard deviations from three independent experiments. Significance was determined using one-way analysis of variance along with Bonferroni’s posttest. ns, no significant difference; **, p < .01; ***, p < .001. (B) A putative ZntR binding site was recognized in the promoter region of zntA (P_zntA) by RegPrecise. (C) β-galactosidase activities of the WT strain harboring the P_zntA-pDM8 plasmid. The strain in the early exponential phase was treated with H₂O (control), 1 mM FeSO₄, 1 mM MnSO₄, 0.5 mM ZnSO₄, 1 mM CuSO₄, 0.5 mM CoSO₄, 1 mM NiSO₄, or 125 μM CdSO₄ for 15 min. Thereafter, the bacterial cells were harvested for measuring the β-galactosidase activities. (D) β-galactosidase activities of the WT and ΔzntR strains harboring either P_zntA-pDM8 or P_zntA’-pDM8. The strains in the early exponential phase were treated with H₂O (control), 0.5 mM ZnSO₄, 1 mM NiSO₄, or 125 μM CdSO₄ for 15 min. Thereafter, the bacterial cells were harvested for measuring the β-galactosidase activities. Results represent the means and standard deviations from four independent experiments. Significance between the treated sample and control was determined using one-way analysis of variance along with Bonferroni’s posttest. ns, no significant difference; *, p < .05; ***, p < .001.

These results suggested that ZntR positively regulates zntA by binding to its promoter.

Discussion

Although trace metals are essential elements for life, they are toxic to bacteria when present in excessive amounts.⁵ Moreover, vertebrate hosts exploit nutritional immunity (both metal restriction and metal toxicity) in response to bacterial infections.⁷ Consequently, bacteria have evolved multiple mechanisms to maintain metal homeostasis.⁵ Metals play a crucial role in the physiology and pathogenesis of V. parahaemolyticus.^17–19,21 Although the mechanisms of metal acquisition have been partly elucidated in V. parahaemolyticus,^20,21,23 limited information is available regarding the response of this bacterium to excess metals. Indeed, only recently we have characterized a metal-response system, DmeRF, which facilitates V. parahaemolyticus growth under high Co conditions.²² Zn is the second most abundant trace metal in living beings and is implicated in the host-pathogen interface.^38,39 By exploring the transcriptome changes during V. parahaemolyticus response to Zn repletion, we can focus on zntA which potentially functions in Zn homeostasis.²³ Herein, we showed that the ZntR-regulated ZntA protects V. parahaemolyticus against Zn/Ni- and Cd-induced bacteriostatic effects by maintaining Zn and Cd homeostasis, respectively. More importantly, ZntA contributes to oxidative stress resistance and virulence in V. parahaemolyticus.

Generally, bacteria utilize a number of metalloregulators to sense changes in metal concentrations and modulate the expression of specific genes in response to metal restriction and excess.⁵ Hence, zntA expression following exposure to elevated concentrations of various metals was detected using qRT-PCR analysis. Consistent with the RNA sequencing data,²³ qRT-PCR analysis revealed that zntA expression was significantly upregulated in response to excess Zn. Moreover, Cu, Co, Ni, and Cd significantly induced zntA expression, whereas Fe(II) and Mn did not. The significant induction of zntA expression by Zn, Cu, Co, Ni, and Cd indicated that ZntA could be involved in V. parahaemolyticus response to high concentrations of these metals. The subsequent growth evaluation showed that ΔzntA grew poorly under high Zn, Ni, and Cd conditions compared to the WT and CΔzntA strains; nonetheless, Cu and Co exhibited no obvious effects on the growth of ΔzntA. Similar observations have also been described in other studies.^22,40,41 In Sinorhizobium meliloti, although Cu, Co, and Ni strongly induce the expression of the dmeRF operon, ΔdmeF exhibits a growth defect only under high Co and Ni concentrations.⁴⁰ In V. parahaemolyticus, dmeF is significantly induced by Zn, Cu, and Co, whereas the growth of ΔdmeF is inhibited only under excess Co conditions.²² In Streptococcus suis, ΔpmtA grow poorly only in the presence of elevated concentrations of Fe(II) and Co, although pmtA expression is significantly upregulated in response to Fe(II), Co, and Ni.⁴¹ It is probable that excessive amounts of Zn, Cu, Co, Ni, or Cd can bind to ZntR and affect its activity, leading to the induction of zntA expression.

Following the observation of the growth defect of ΔzntA under excess Zn, Ni or Cd conditions, we aimed to explore the underlying mechanism using ICP-OES. As expected, when grown in the presence of Zn or Cd, the intracellular content of the corresponding metal accumulated in ΔzntA was higher than that in the WT and CΔzntA strains. Surprisingly, when cultured in the presence of Ni, ΔzntA accumulated higher levels of intracellular Zn, rather than Ni, compared with the WT and CΔzntA strains. Moreover, the high sensitivity of ΔzntA to excess Ni was markedly decreased in the Zn-deficient medium. As observed in S. Typhimurium,²⁷ V. parahaemolyticus zntA was significantly upregulated under Fe restriction, and the growth of ΔzntA was impaired in Fe-restricted medium. In the Zn-deficient medium, ΔzntA eliminated its sensitivity to Fe restriction. Based on these results, we speculated that excess Ni and Fe restriction disrupted Zn homeostasis in V. parahaemolyticus, while ZntA allowed this bacterium to cope with disrupted Zn homeostasis. Similarly, excess Zn perturbs Fe and Cu homeostasis in E. coli⁴²; the disruption of Zn homeostasis affects Mn and Fe homeostasis in K. pneumoniae.²⁶ The mismetalation of metalloproteins with nonpreferred metals is one of the mechanisms by which metals exert their toxicity to bacteria.⁵ Furthermore, Fe acquisition genes are significantly downregulated when V. parahaemolyticus is grown under Zn-replete conditions.²³ Therefore, we assumed that Fe(II) supplementation could mitigate the growth defect of ΔzntA under excess Zn, Ni, or Cd conditions. In line with this hypothesis, Fe(II) supplementation could improve the growth of ΔzntA in the presence of excess Zn, Ni, or Cd. Similar observations have been reported in other studies.^{28,41,43–45}

The correlation between Zn and the oxidative stress response has been well established.^29,46–49 ZosA, a Zn uptake protein, is important for Bacillus subtilis resistance to H₂O₂- and diamide-induced oxidative stress.⁴⁶ In Salmonella enterica, a mutant lacking the Zn importers ZnuABC and ZupT is hypersensitive to oxidative stress.⁴⁷ Interestingly, zntA inactivation in A. tumefaciens confers hyper-resistance to H₂O₂.²⁹ Unlike the finding in A. tumefaciens, our results showed that ZntA protected V. parahaemolyticus against H₂O₂-induced oxidative stress under excess Zn conditions. Consistent with our observations, previous studies have shown that PmtA, a metal transporter that accounts for hyper-resistance to Zn in a perR deletion mutant, contributes to resistance against H₂O₂ in S. pyogenes⁴⁸; the sensitivity of Streptococcus mutans zccE (encoding a Zn exporter) deletion mutant to H₂O₂ increases under excess Zn conditions.⁴⁹ Moreover, certain genes involved in the oxidative stress response, such as VP_RS10295 (encoding superoxide dismutase) and VP_RS16740 (encoding catalase), are significantly upregulated under Zn-replete conditions,²³ indicating that an interface exists between Zn homeostasis and oxidative stress.

The involvement of Zn homeostasis in virulence has been established in certain bacteria.³³ In S. pyogenes, the susceptibility of the czcD (encoding a zinc efflux pump) and gczA (encoding a regulator of czcD) deletion mutants to human neutrophil killing is increased and their virulence in mice is decreased.⁵⁰ Zn efflux systems also promote the virulence of Streptococcus agalactiae in mice.⁵¹ In Mycobacterium bovis, the Zn exporter CtpG enhances mycobacterial survival in THP-1 cells and mice.⁵² In S. Typhimurium, ΔzntA displays a significantly reduced ability to cause systemic infection in a competitive-infection experiment performed on mice.⁵³ Zebrafish models have been widely used for assessing the virulence of V. parahaemolyticus strains.^54–58 Hence, we evaluated the role of ZntA in V. parahaemolyticus virulence using zebrafish models. Similarly, we showed that the virulence of ΔzntA was attenuated compared to that of the WT strain during zebrafish infection. Considering that V. parahaemolyticus experiences dynamic changes in Zn concentrations during infection of L. vannamei,¹⁹ our result was not necessarily surprising. However, not all Zn exporters contribute to bacterial virulence. For example, a zntA zur double mutant of Y. pestis remains to be highly virulent in two mouse models.⁵⁹

RegPrecise recognizes a putative ZntR binding sequence in the promoter region of zntA, suggesting that ZntR potentially modulates zntA expression by binding to its promoter. In A. tumefaciens,²⁹ E. coli,^31–33 and Cupriavidus metallidurans,³⁷ ZntR positively regulates zntA expression. In Staphylococcus aureus³⁴ and B. abortus,³⁶ ZntR negatively regulates zntA expression. Using qRT-PCR, mutagenesis, and β-galactosidase activity assays, we demonstrated that ZntR positively regulates zntA expression in V. parahaemolyticus by binding to its promoter. The mechanism of ZntR regulation of zntA has been illustrated in E. coli: under Zn-deficient conditions, ZntR binds to the promoter of zntA, distorts the DNA, and represses its transcription; under Zn-replete conditions, Zn-bound ZntR introduces DNA conformation changes and enhances RNA polymerase binding to the promoter to initiate transcription.³² Based on our results, we speculate that in V. parahaemolyticus, ZntR regulates zntA expression through a similar mechanism. In a previous study, we showed that the Zur-regulated ZrgABCDE system plays a predominant role in Zn acquisition.²³ Together, V. parahaemolyticus utilizes the Zur-regulated ZrgABCDE and the ZntR-regulated ZntA to coordinate Zn import and export, respectively.

In summary, ZntA promotes V. parahaemolyticus growth under excess Zn/Ni and Cd conditions via maintaining Zn and Cd homeostasis, respectively. ZntA is also involved in V. parahaemolyticus growth under Fe restriction by maintaining Zn homeostasis. Importantly, ZntA contributes to oxidative stress resistance and virulence in V. parahaemolyticus. Finally, ZntR regulates zntA expression by binding to its promoter.

Materials and methods

Bacterial strains, plasmids, and growth conditions

Bacterial strains and plasmids are listed in Table S1. Primers are listed in Table S2. E. coli was grown in Luria-Bertani (LB) broth or on LB agar. V. parahaemolyticus strains were cultured at 37°C with either LB (for generation of mutant and complemented strains) or TSB/TSA (for other assays). When appropriate, chemicals were added at the following concentrations: carbenicillin, 50 μg/mL; chloramphenicol, 25 μg/mL; and isopropyl β-D-1-thiogalactopyranoside (IPTG), 1 mM.

Generation of mutant and complemented strains

Strains lacking zntA or zntR (ΔzntA and ΔzntR) were constructed via allelic exchange using pDM4, a suicide vector carrying the counterselectable sacB gene,⁶⁰ as previously described.²³ In brief, the left and right arms of the target gene were amplified from the V. parahaemolyticus RIMD2210633 genome and fused into a DNA fragment via overlap extension PCR. After enzymes digestion, the DNA fragment was cloned into pDM4. The resultant plasmid was transformed into E. coli S17–1 λpir and then introduced into RIMD2210633 via conjugation, as previously described.⁶¹ A single crossover recombination strain was selected from LB agar containing carbenicillin and chloramphenicol. This strain was further cultured in LB containing 15% sucrose for 12 h and streaked onto LB agar containing 15% sucrose, to select double crossover recombination mutants. The mutant was confirmed by PCR and DNA sequencing.

The pMMB207 vector⁶² was used to generate the complemented strains (CΔzntA and CΔzntR). Briefly, a DNA fragment containing the target gene and an additional ribosome-binding site was amplified from the RIMD2210633 genome, digested with the appropriate enzymes, and cloned into pMMB207. The resultant plasmid was transformed into E. coli S17–1 λpir and then introduced into the mutant via conjugation. LB agar containing carbenicillin and chloramphenicol was used to select the complementation strain, which was further verified by PCR.

qRT-PCR analysis

The V. parahaemolyticus RIMD2210633 strain was cultured in TSB until the early exponential phase (OD₆₀₀ of ~ 0.7). Eight aliquots (1 mL) were removed and treated with H₂O, 1 mM FeSO₄, 1 mM MnSO₄, 0.5 mM ZnSO₄, 1 mM CuSO₄, 0.5 mM CoSO₄, 1 mM NiSO₄, or 125 μM CdSO₄. After 15 min of treatment, the bacterial cells were harvested by centrifugation and subjected to total RNA isolation using an Eastep Super Total RNA Isolation Kit (Promega). In another experiment, the RIMD2210633 strain was grown in TSB until the early exponential phase. Two aliquots (1 mL) were removed and centrifuged. Thereafter, the bacterial pellets were resuspended in 1 mL TSB pretreated for 2 h with either absolute ethanol or 150 μM 2,2ʹ-dipyridyl. After 30 min of incubation, the bacterial cells were harvested for total RNA isolation. Additionally, the WT, ΔzntR, and CΔzntR strains were grown in TSB until the early exponential phase, and the bacterial cells were harvested for total RNA isolation.

After RNA integrity and concentrations were evaluated, cDNA was synthesized using HiScript II Q RT SuperMix for qPCR (+gDNA wiper; Vazyme). qRT-PCR was performed on a StepOnePlus Real-Time PCR System (Applied Biosystems) using ChamQ SYBR qPCR Master Mix (Vazyme). gyrB was used as an internal control, and relative gene expression was calculated using the 2^−ΔΔCT method.⁶³

Growth curve analysis

The WT, ΔzntA, and CΔzntA strains were grown in TSB until the mid-exponential phase (OD₆₀₀ of ~ 2) and diluted 1:100 in TSB supplemented with the indicated chemicals. The medium containing FeSO₄ was supplemented with 1 g/L trisodium citrate dihydrate (TCD) to alleviate Fe precipitation.⁴⁵ The cultures were incubated in 96-well plates (200 µL/well in triplicate) at 37°C with shaking at 120 rpm. Growth was monitored hourly by measuring the OD₅₉₅ using a microplate reader (Molecular Devices).

Spot dilution assays

The WT, ΔzntA, and CΔzntA strains were grown in TSB until mid-exponential phase. Two aliquots (1 mL) were removed and treated with 100 μM ZnSO₄, 1.5 mM NiSO₄, or 25 μM CdSO₄. At 0, 3, and 6 h, aliquots (100 μL) were removed and serially diluted 10-fold to 10⁻⁶. Thereafter, 5 μL of each dilution was spotted onto the TSA plates. The plates were incubated at 37°C for 8 h and then photographed.

ICP-OES

The WT, ΔzntA, and CΔzntA strains were grown in TSB until the mid-exponential phase and diluted 1:100 in TSB supplemented with 25 μM ZnSO₄, 1 mM NiSO₄, or 2.5 μM CdSO₄. After 6 h of incubation, the bacterial cells were harvested by centrifugation, washed thrice with PBS plus 250 mM EDTA, and washed thrice with PBS. The cells were dried at 110°C, weighed, digested in 66% nitric acid, and diluted to 2% nitric acid with metal-free H₂O. ICP-OES analysis was performed at Yangzhou University using PerkinElmer Optima 7300, following the manufacturer’s guidance, to determine the metal content, which was expressed as micrograms of metal per gram of cells (dry weight).

Zebrafish infections

The zebrafish experiment was approved by the Animal Welfare and Ethics Committees of Yangzhou University and performed in strict accordance with the guidelines of the Institutional Administrative Committee and Ethics Committee of Laboratory Animals.

The WT and ΔzntA strains were cultured in TSB at 30°C for 12 h, adjusted to 1 × 10⁹ CFU/mL in PBS. For analysis of survival curves, 45 zebrafish (3–4 months old) were randomly divided into three groups (15 zebrafish per group). The zebrafish in group 1 and 2 were anesthetized with 1% ethyl 3-aminobenzoate methane sulfonate (Sigma-Aldrich) and injected intraperitoneally with 20 μL of the WT and ΔzntA strains, respectively. The zebrafish in group 3 were anesthetized, injected intraperitoneally with 20 μL of the WT strain (heat-killed), and served as the control group. Zebrafish were monitored twice daily over seven days for survival rates. For the competitive-infection assays, the WT and ΔzntA strains (1 × 10⁹ CFU/mL) were mixed at a 1:1 ratio. Before mixing, the strain suspensions were serially diluted and plated on TSA plates to determine the actual ratio. Eight zebrafish were anesthetized and injected intraperitoneally with 10 μL of the mixture. The zebrafish were euthanized at 24 h post-infection, and the full intestines were collected, homogenized in PBS, serially diluted, and plated onto thiosulfate citrate bile-salts sucrose agar for the selective isolation of V. Parahaemolyticus. Another eight zebrafish were anesthetized and injected intramuscularly with 5 μL of the mixture. The zebrafish were euthanized at 12 h post-infection, and the muscle samples were collected, homogenized in PBS, serially diluted, and plated onto TSA plates. The ΔzntA:WT ratio was determined for each sample by PCR analysis of 96 colonies using the primer pair zntA-out-F/zntA-out-R, which yielded 484 and 2544 bp products for ΔzntA and WT, respectively. The CI was expressed as the ΔzntA:WT ratio in each sample divided by that in the mixture.

β-galactosidase activity assays

RegPrecise, a database for the exploration of transcription factor regulons,⁶⁴ recognizes a putative ZntR binding site in the promoter region of zntA. This site was replaced with an unrelated DNA sequence to obtain P_zntA’. DNA carrying P_zntA’ was synthesized by Tsingke Biotechnology Co., Ltd. (Beijing, China). The WT and mutant promoters of zntA were separately cloned into pDM8, a plasmid harboring the promoterless lacZ gene.⁶⁵ The resultant plasmids and empty pDM8 plasmid were separately transformed into S17–1 λpir and introduced into the WT and ΔzntR strains via conjugation to generate LacZ fusion strains.

β-galactosidase activity assays were performed as previously described.²² The LacZ fusion strains were cultured in TSB until the mid-exponential phase. Aliquots (1 mL) were removed and treated with the indicated metals. After 15 min of treatment, bacterial cells were harvested by centrifugation and resuspended in 1 mL PM buffer. The bacterial suspensions were measured in terms of OD₆₀₀. Thereafter, the bacterial suspension (200 µL), chloroform (30 µL), and 0.1% SDS (30 µL) were added to PM buffer (500 µL). Bacterial cells were lysed by vigorous vortexing. Next, o-nitrophenyl-β-galactopyranoside (4 mg/mL, 200 µL) was added to the mixture to initiate the reaction. Subsequently, 1 M Na₂CO₃ (400 µL) was added to stop the reaction when the mixture turned yellow. The supernatant was collected after centrifugation and measured in terms of OD₄₂₀. β-galactosidase activity was calculated as OD₄₂₀ ×1000 × min⁻¹ × mL⁻¹ × OD₆₀₀⁻¹ (Miller units).

Statistical analysis

Data were analyzed using GraphPad Prism 5 (San Diego, CA, USA). Gene expression was examined using a two-tailed paired t-test or one-way analysis of variance along with Bonferroni’s posttest. Metal content and β-galactosidase activity were assessed using one-way analysis of variance along with Bonferroni’s posttest. Survival curves of zebrafish were analyzed using the log-rank test. The competitive infection experiment was analyzed using a two-tailed paired t-test.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary material.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19490976.2024.2327377

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (No. 31802210), the 111 Project (D18007), Key projects from Shenzhen Science and Technology Innovation Commission (grant No. JCYJ20220818100616034), Shenzhen Medical Research Fund (B2302024), Changjiang Scholar Program, and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).