Functions of VPA1418 and VPA0305 Catalase Genes in Growth of Vibrio parahaemolyticus under Oxidative Stress

Abstract

The marine foodborne enteropathogen Vibrio parahaemolyticus has four putative catalase genes. The functions of two katE-homologous genes, katE1 (VPA1418) and katE2 (VPA0305), in the growth of this bacterium were examined using gene deletion mutants with or without complementary genes. The growth of the mutant strains in static or shaken cultures in a rich medium at 37°C or at low temperatures (12 and 4°C), with or without competition from Escherichia coli, did not differ from that of the parent strain. When 175 μM extrinsic H₂O₂ was added to the culture medium, bacterial growth of the ΔkatE1 strain was delayed and growth of the ΔkatE1 ΔkatE2 and ΔkatE1 ΔahpC1 double mutant strains was completely inhibited at 37°C for 8 h. The sensitivity of the ΔkatE1 strain to the inhibition of growth by H₂O₂ was higher at low incubation temperatures (12 and 22°C) than at 37°C. The determined gene expression of these catalase and ahpC genes revealed that katE1 was highly expressed in the wild-type strain at 22°C under H₂O₂ stress, while the katE2 and ahpC genes may play an alternate or compensatory role in the ΔkatE1 strain. This study demonstrated that katE1 encodes the chief functional catalase for detoxifying extrinsic H₂O₂ during logarithmic growth and that the function of these genes was influenced by incubation temperature.

INTRODUCTION

Various reactive oxygen species (ROS), such as superoxide anion (O₂⁻), hydrogen peroxide (H₂O₂), and hydroxyl radical (˙OH), are generated by intrinsic metabolic activity in bacteria or induced by environmental stresses (1–3). ROS are detrimental to cellular components, including proteins, DNA, and membrane lipids (4).

Most bacteria are equipped with various antioxidative enzymes for scavenging ROS. Superoxide dismutase (SOD) transforms superoxide anions into hydrogen peroxide, while catalase decomposes hydrogen peroxide into oxygen and water. Two families of catalases, HPI (KatG) and HPII (KatE), have been identified in Escherichia coli and some other enteric bacteria (5). HPI, which is the family of bifunctional catalases/peroxidases, is transcriptionally induced during logarithmic growth in response to low concentrations of hydrogen peroxide. This induction requires the positive activator OxyR, which directly senses oxidative stress. HPII, the family of monofunctional catalases, is not peroxide inducible and is transcribed at the transition from exponential growth to the stationary phase by the product of the rpoS gene, which is a critical factor in the survival of bacteria in the stationary phase or under other stresses (6, 7). OxyR also regulates the transcription of the alkyl hydroperoxide reductase subunit C (ahpC) gene, which encodes a 2-cysteine peroxiredoxin for detoxifying organic peroxides (8, 9).

Food processing commonly imposes stresses on foodborne pathogens, and these stresses may account for the formation of ROS. Campylobacter accumulates hydrogen peroxide under freeze-thaw treatment (10). Environmental stresses lower the level of cellular SOD and catalase in Vibrio parahaemolyticus, while increasing the susceptibility of this pathogen to oxidative stress (11, 12). We have previously demonstrated that the level of intracellular ROS is related to the survival of V. parahaemolyticus under a combination of cold, starvation, and low salinity (13). Therefore, the functions of antioxidative factors may be crucial to the persistence of these foodborne pathogens in the environment. Also, extracellular ROS may be generated by other bacteria or hosts of bacterial infection (14–16), and the presence of extracellular catalase has been demonstrated in Vibrio cholerae (17). The functions of antioxidative factors may enhance the virulence of infectious bacteria in human beings, establish natural symbionts in aquacultured animals (16), and enable the successful growth of bacteria in the presence of competitors.

V. parahaemolyticus is a halophilic Gram-negative bacterium that frequently causes foodborne gastroenteritis in Taiwan and some other Asian countries (18), and it has become a pathogen of global concern following the appearance of the first pandemic O3:K6 strain in 1996 (19). In a search of the genome sequence of the V. parahaemolyticus strain RIMD2210633 (20), two katE- and two katG-homologous genes were identified, namely, katE1 (VPA1418), katE2 (VPA0305), katG1 (VPA0768), and katG2 (VPA0453). Recently, four proteins exhibiting catalase or catalase/peroxidase activity were demonstrated using zymogram in V. parahaemolyticus, whereas two catalases are induced in the exponential/early stationary phase (21). Unfortunately, the identities of these proteins have not been determined (21), and the functions of specific catalase genes remain unclear. In addition to these catalase genes, an alkylhydroperoxide reductase subunit C gene (ahpC1) was also responsive to different peroxides (22, 23). To understand the role of specific catalase genes in the growth of V. parahaemolyticus under challenge with peroxides, low temperature, and the presence of a competitive bacterium, katE1 and katE2 deletion mutants with and without complementary genes were constructed and characterized.

MATERIALS AND METHODS

Bacterial strains and culture conditions.

V. parahaemolyticus strain KX-V231 (Kanagawa phenomenon positive, serotype O3:K6), isolated in Thailand from a clinical specimen, was used in this study (Table 1). It was stored frozen at −85°C in beads in Microbank cryovials (Pro-Lab Diagnostics, Austin, TX, USA). It was cultured at 37°C on tryptic soy agar (Becton-Dickinson Diagnostic Systems, Sparks, MD, USA) that was supplemented with 3% sodium chloride (TSA–3% NaCl) or in tryptic soy broth (TSB)–3% NaCl in a 5-ml tube which was shaken at 160 rpm. A 50-μl aliquot of the 16-h broth culture was inoculated into 5 ml of fresh TSB–3% NaCl and incubated at 37°C with shaking for 2 h for the cells to enter the exponential phase (around 10⁸ CFU/ml), and this culture was used as the inoculum in the following experiments. E. coli was cultured in Luria-Bertani (LB) broth (Becton-Dickinson) at 37°C and shaken at 160 rpm. Chloramphenicol (final concentration of 6 μg/ml) or chloramphenicol (20 μg/ml)-ampicillin (50 μg/ml) was added to the media as required for the cultivation of some of the V. parahaemolyticus or E. coli strains, respectively.

TABLE 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Characteristics	Source or reference
V. parahaemolyticus strains
KX-V231	Wild type, serotype O3:K6, KP+, clinical isolate	This study
ΔkatE1 mutant	KX-V231 ΔkatE1 (VPA1418)	This study
ΔkatE2 mutant	KX-V231 ΔkatE2 (VPA0305)	This study
ΔkatE1 ΔkatE2 mutant	KX-V231ΔkatE1ΔkatE2	This study
ΔkatE1/katE1 mutant	KX-V231 ΔkatE1/pSCB01-katE1	This study
ΔkatE2/katE2 mutant	KX-V231 ΔkatE2/pSCB01-katE2	This study
ΔahpC1 mutant	KX-V231 ΔahpC1 (VPA1683)	23
ΔkatE1 ΔahpC1 mutant	KX-V231ΔkatE1ΔahpC1	This study
KX-V231V	KX-V231 containing pSCB01	This study
E. coli strains
XL1-Blue	recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac [F′ proAB lacIqZΔM15 Tn10 (Tetr)]	Stratagene
SM10 λpir	thi thr leu tonA lacY supE recA::RP4-2-Tc::Mu λ pirR6K; Kmr	40
Plasmids
pGEM-T Easy	Cloning vector, Apr	Promega
pDS132	R6K ori, mobRP4, sacB, Cmr	41
pSCB01	Derived from pBR328 and pDS132, mobRP4, Apr, Cmr, Tcr	23
pSCB01-katE1	pSCB01 with katE1	This study
pSCB01-katE2	pSCB01 with katE2	This study

Construction of deletion mutants.

Mutants with deletions of the catalase genes (katE1 and katE2) were constructed following published methods (23, 24). For constructing the ΔkatE2 mutant strain, two DNA fragments were amplified by PCR with V. parahaemolyticus KX-V231 chromosomal DNA as the template, one with the primer pair VPA0305-1 and VPA0305-2 and the other with the primer pair VPA0305-3 and VPA0305-4 (Table 2). These two amplified fragments were then used as templates for a second PCR with primers VPA0305-1 and VPA0305-4, resulting in the construction of a fragment with a deletion in the VPA0305 gene. This fragment containing the deletion was purified and cloned into the pGEM-T Easy vector and transformed into E. coli XL1-Blue, following the protocol of the manufacturer (Promega Co., Madison, WI, USA). The inserted sequence was verified by sequencing. This fragment was then removed from the pGEM-T Easy vector by digestion using SacI and SphI and cloned into a suicide vector, pDS132, which contained the chloramphenicol resistance gene and the sacB gene, conferring sensitivity to sucrose. This plasmid (pDS132-katE2-deletion) was introduced into E. coli SM10 λpir, which was then mated with V. parahaemolyticus strain KX-V231. Thiosulfate-citrate-bile-sucrose (TCBS) agar that contained chloramphenicol was used to screen the V. parahaemolyticus cells containing the inserted plasmid. The V. parahaemolyticus clones were isolated and cultured in LB broth that was supplemented with 2% NaCl and chloramphenicol. DNA was extracted from these cultures, and the inserted sequence was detected by PCR using the VPA0305-1 and VPA0305-4 primers. The culture that contained the pDS132-katE2-deletion plasmid was incubated at 37°C for 3 h in LB broth that contained 2% NaCl and was then plated on an LB agar plate that contained 2% NaCl and 10% sucrose. The isolated colonies that were unable to grow on an LB agar plate that contained chloramphenicol were selected, and homologous recombination of the deleted fragment was verified by PCR (Table 2). Amplification of the katE2 gene with the primers VPA0305-0 and VPA0305-5 yielded amplicons of 3,275 bp and 1,733 bp in the wild-type strain and the ΔkatE2 strain, respectively. The mutated gene was also verified by nucleotide sequencing of the amplified fragments. Following the same procedures using different primers (Table 2), the ΔkatE1 strain was also constructed. The ΔkatE1 ΔkatE2 double mutant was prepared similarly by construction of the katE2 gene deletion in the ΔkatE1 strain, while the ΔkatE1 ΔahpC1 double mutant was prepared similarly by construction of the ahpC1 gene deletion in the ΔkatE1 strain.

TABLE 2.

Primers used in cloning experiments

Target	Primer	Sequence, 5′→′3
VPA0305	VPA0305-1	CGGCGTTGAAGTGGTGTTGG
	VPA0305-2	CCGTATTCTTTGTCTGCACGATTTTGCGCCTGTAGAGATGTG
	VPA0305-3	CACATCTCTACAGGCGCAAAATCGTGCAGACAAAGAATACGG
	VPA0305-4	GCGAACGTCTTCAAGTCGAG
	VPA0305-0	GGTCAGATTTATCCTTCGTC
	VPA0305-5	GTGATTGTGAATCTAGCTGC
	VPA0305-C1	CAGTGTAATCACTCTCGCCA
	VPA0305-C2	CAGAGCTGAGCAAGAATACG
VPA1418	VPA1418-1	CATTAAAGAGCCGAACTCGATGC
	VPA1418-2	TTGGTAAGCGTGGGTGACGTGGACATCTTGTAGGAGTTGAGGG
	VPA1418-3	CCCTCAACTCCTACAAGATGTCCACGTCACCCACGCTTACCAA
	VPA1418-4	CAGAACTTGCTGTGGAACTGG
	VPA1418-0	CAGGAGCCATGACTGAATACTTG
	VPA1418-5	GTTGGTAATGATAACGACGTACG
	VPA1418-C1	CATTAAAGAGCCGAACTCGATGC
	VPA1418-C2	TTATTTCGCTAAACCTAACGCCAG

Sequencing service was provided by Genomics BioSci & Tech, Inc., Taipei, using Sanger's method with an Applied Biosystems 3730 analyzer.

Construction of complementary strains.

The entire length of the katE2 gene was amplified by PCR with V. parahaemolyticus KX-V231 chromosomal DNA as the template using primer pair VPA0305-C1 and VPA0305-C2 with restriction enzyme linkers (SalI and SphI) (Table 2). The amplicon was digested with SalI and SphI and ligated to the shuttle vector pSCB01 which had been digested with the same enzymes (23). The plasmid, pSCB01-katE2, containing the entire length of the katE2 gene was propagated in E. coli SM10 λpir and conjugated to the corresponding ΔkatE2 strain to generate gene complementation, which was selected by chloramphenicol resistance (Table 1). The presence of the entire length of the katE2 gene in these strains was verified by PCR. Following the same procedures, the complementation of the katE1 gene in the ΔkatE1 strain was also constructed (Table 1).

Effects of peroxides on bacterial growth.

V. parahaemolyticus cultures in the exponential phase (200 μl) were dispensed into the wells of a microtiter plate, to which various concentrations of H₂O₂ (Santoku Chemical Industries, Tokyo, Japan), cumene hydroperoxide (cumene) (Alfa Aesar, Ward Hill, MA, USA), or tert-butyl hydroperoxide (t-BOOH) (Tokyo Kasei Chemicals, Tokyo, Japan) were added; the cultures were then incubated statically at 37°C or 22°C for 8 h or at 12°C for 56 h. Bacterial growth was determined by measuring the absorbance of the culture at 590 nm using an MRXII microplate reader (Dynex Technologies, Chantilly, VA, USA).

Low-temperature stress.

V. parahaemolyticus cultures in the exponential phase (200 μl) were dispensed into the wells of a microtiter plate and statically incubated at 12°C. Bacterial growth was determined by measuring the absorbance at 590 nm. In another experiment, the V. parahaemolyticus cultures in the exponential phase were 10-fold diluted in TSB–3% NaCl, and 100-ml volumes of these diluted cultures were incubated at 4°C. At intervals, the survivors were counted on TSA–3% NaCl agar.

Growth competition.

Wild-type and mutant V. parahaemolyticus strains were grown in a coculture with E. coli SM10 λpir harboring pDS132. The V. parahaemolyticus culture in the exponential phase was diluted 10-fold in fresh TSB–1% NaCl. E. coli was cultured in LB that contained chloramphenicol (20 μg/ml) until it reached the exponential phase. The V. parahaemolyticus and E. coli cultures were inoculated separately into TSB–1% NaCl or mixed in a 1:40 (vol/vol) ratio and then inoculated; they were subsequently incubated statically or with shaking at 160 rpm for 8 h. The V. parahaemolyticus and E. coli cells were counted on TSA–3% NaCl that was supplemented with 15 μg/ml ampicillin and on Luria-Bertani (LB) agar that was supplemented with 5 μg/ml chloramphenicol, respectively, following incubation at 37°C for 16 h. To count the bacteria with the complementary gene, LB agar was used, on which V. parahaemolyticus formed pale yellow, large colonies while E. coli formed white, small colonies.

RT-qPCR.

The expression of genes (Table 3) in the wild-type and ΔkatE1 strains was determined using real-time quantitative reverse transcription-PCR (RT-qPCR) (23). Briefly, bacterial strains were cultivated statically in TSB–3% NaCl at 22 or 37°C, and the cultures in exponential phase were challenged with 175 μM H₂O₂ for 1.5 h. Bacterial cells were harvested by centrifugation and broken using TRIzol reagent (Invitrogen, United Kingdom), and RNA samples were extracted using an RNApure kit (Genesis Biotech Inc., Taipei, Taiwan), following the manufacturer's instructions. RNA samples were treated with DNase I (TaKaRa Bio Inc., Shiga, Japan) and then reverse transcribed using SuperScript III first-strand synthesis SuperMix (Invitrogen, United Kingdom), following the instructions of the manufacturer. Primers (Table 3) were designed using the Primer Express Sequence Editor (http://www.fr33.net/seqedit.php) and Oligo Calculator (http://www.sciencelauncher.com/oligocalc.html), and 16S rRNA was used as the internal control. Real-time PCR was performed using the StepOnePlus real-time PCR system v.2.0 (Applied Biosystems) with a IQ² SYBR green fast qPCR system master mix and High ROX (DBU-008) and RT-PCR reagents. All the data were normalized with the 16S gene expression levels of the culture at each time point, and the normalized values for each gene were compared (Applied Biosystems). Expression of each target gene of the experimental group relative to the expression of the corresponding gene of the control was presented. Recombinant plasmids for the target genes were used as a calibration standard (Table 1) (23). The quality of the RNA samples and the quantification protocols that were used here was evaluated by previously described methods (23).

TABLE 3.

Primers used in RT-qPCR experiment

Designation	Sequence	Target	Amplicon size, bp
q16SrRNA-F	TCCCTAGCTGGTCTGAGA	16S rRNA gene	222
q16SrRNA-R	GGTGCTTCTTCTGTCGCT
VP0580-F	CGACAACCGTCTAGCTGA	ahpC2	202
VP0580-R	AGCAACACCTGCTTCTGG
VPA1683-F	CTACCCAGCAGACTTCAC	ahpC1	227
VPA1683-R	CTTCACGCATCACACCGA
VPA0305-F	AGAGTTGTGCACGCTCGT	VPA0305	228
VPA0305-R	CCCTACCAGATCCCAGTT
VPA1418-F	TACGACCGTTGCTGGTGA	VPA1418	235
VPA1418-R	TTCTGGCAGCGATGTCCA
VPA0453-F	TGCATGGCTCCATGACCA	VPA0453	257
VPA0453-R	CGCATGCCATGACATACG
VPA0768-F	GTGGTCATACCGTGGGTA	VPA0768	237
VPA0768-R	GGCTCTTCTTCAGTTCCC

Statistical analysis.

Triplicate experiments were performed, and the data of the bacterial growth experiments were obtained from triplicate determinations. The data were analyzed by performing one-way analysis of variance (ANOVA) or t test at a significance level of α = 0.05, using SPSS for Windows version 11.0 (SPSS Inc., Chicago, IL, USA).

RESULTS

Growth and survival of catalase gene mutants.

To evaluate the significance of these katE-homologous genes in the growth of V. parahaemolyticus under normal growth conditions, the growth of the single (ΔkatE1 and ΔkatE2) and double (ΔkatE1 ΔkatE2) catalase gene mutants, the gene-complemented (ΔkatE1/katE1, ΔkatE2/katE2) strains, and the wild-type strain (Table 1) in TSB–3% NaCl at 37°C under either shaking or static conditions was determined. Bacterial growth was promoted by shaking at 160 rpm, and the cells approached the late exponential phase after 4 h of incubation, when they reached a maximal absorbance of about 4 at 590 nm (see Fig. S1 in the supplemental material) and a cell density of about 10¹⁰ CFU/ml (data not shown). In the static culture, the growth of bacterial cells approached the stationary phase after 3 h of incubation, exhibiting a maximal absorbance of about 0.7 at 590 nm (see Fig. S1 in the supplemental material). No defective growth compared to the growth of the wild-type strain was observed for these mutants under these conditions. Nevertheless, the presence of the complementary katE2 gene in the ΔkatE2 strain slightly affected its growth, in particular enhancing the growth of the shaking culture after incubation for 6 to 8 h (see Fig. S1A in the supplemental material). This suggests that the shaking culture, in contrast to the static culture, may generate oxidative stress that activates the expression of the katE2 gene.

Incubating these cultures statically at 12°C slowed down the growth of the bacterial cells, and the cultures approached the late exponential phase after 25 h of incubation and reached a maximal absorbance of 0.5 after 55 h (data not shown).

The populations of culturable cells of the wild-type, ΔkatE1, ΔkatE1 ΔkatE2, and ΔkatE1/katE1 strains in TSB–3% NaCl were equal following static incubation at 4°C. A slow decline in the number of culturable cells was observed over time, and 10⁸ to 10⁹ CFU/ml remained culturable and about 0.5 × 10⁸ CFU/ml had been killed after 52 h of incubation (data not shown). The results revealed that deletion mutations of these katE-homologous genes did not influence the growth and survival of this pathogen in rich medium under growth-permitting (12 to 37°C) or refrigeration temperatures. These results also suggest the presence of an efficient compensatory mechanism in these catalase-deficient mutants under these conditions.

Growth of catalase gene mutants in coculture with E. coli.

Extracellular ROS are produced by some bacterial species, such as Enterococcus faecalis (25), while efflux of H₂O₂ also occurs in E. coli (26). Catalase-deficient cells have a growth disadvantage over catalase-proficient cells in a mixed culture (26). Thus, these catalase gene mutations may decrease the competition of V. parahaemolyticus in cocultures and influence its persistence in natural environment. In this study, the growth of the wild-type strain and different catalase mutants cocultured with E. coli was assayed. The TSB–1% NaCl medium provided rapid growth for both species in shaken culture (Fig. 1A). In the coculture, the initial density of E. coli was 10 times that of the V. parahaemolyticus strains. In the shaken single culture, both the V. parahaemolyticus strains and E. coli grew rapidly. In the coculture, V. parahaemolyticus strains, at a much lower initial density than E. coli, multiplied rapidly and became the dominant population after 2 to 3 h of incubation, after which the growth of E. coli was inhibited. The cell densities of the V. parahaemolyticus strains at 6 to 8 h of incubation were significantly lower in the coculture than in the single culture; nevertheless, deletion mutation of these catalase genes did not significantly affect their growth competition (Fig. 1).

FIG 1.

Effect of catalase gene mutation on growth of Vibrio parahaemolyticus in competition with Escherichia coli in shaken culture. V. parahaemolyticus wild-type and mutant strains and E. coli that harbored cloning vector pDS132 were cultured separately (control) or cocultured in TSB–1% NaCl at 37°C with at 160 rpm. (A) V. parahaemolyticus wild type; (B) ΔkatE1 mutant; (C) ΔkatE1 ΔkatE2 mutant; (D) ΔkatE1/katE1 mutant. ●, V. parahaemolyticus strain in coculture; ○, E. coli in coculture; ▼, V. parahaemolyticus strain in separate culture; △, E. coli in separate culture.

In the static culture, the population of E. coli remained at 10⁷ CFU/ml for 8 h of incubation when it was cultured separately or in the coculture. The V. parahaemolyticus strains, with a much lower initial density in the coculture, rapidly reached the maximal density of 10⁹ CFU/ml after 4 h of incubation. Deletion mutation of these catalase genes did not significantly affect its growth and competition under static culture (see Fig. S2 in the supplemental material).

Growth of catalase gene mutants in the presence of extrinsic H₂O₂.

The addition of 175 or 200 μM H₂O₂ to the TSB–3% NaCl medium significantly slowed the growth of the wild-type strain of V. parahaemolyticus at 37°C and delayed the reaching of the exponential and stationary phases (Fig. 2A). The concentrations of H₂O₂ used in this study were not lethal to V. parahaemolyticus, and it did not significantly decay during the incubation time (data not shown). When 175 μM H₂O₂ was applied to catalase mutant strains, the bacterial growth of the ΔkatE2 mutant was slightly delayed, that of the ΔkatE1 mutant was markedly delayed, and that of the ΔkatE1 ΔkatE2 and ΔkatE1 ΔahpC1 double mutants was completely inhibited (Fig. 2B). The growth of the ΔkatE1 strain, which was inhibited by H₂O₂, was restored by the complementary katE1 gene, while the growth of the ΔkatE2 strain in the presence of the complementary katE2 gene did not differ significantly from that of the wild-type strain that harbored the cloning vector (KX-V231V) (Fig. 2C). When the ΔkatE1/katE1, ΔkatE2/katE2, and KX-V231V strains were cultivated in medium containing chloramphenicol to maintain the plasmids in the cells, growth of the ΔkatE1/katE1 strain under extrinsic H₂O₂ was accelerated and it reached late exponential phase about 1 h earlier than the other two strains containing plasmids (Fig. 2C). The experimental results shown in Fig. 2 and in Fig. S1 in the supplemental material demonstrate that both katE1 and katE2 were functional, while katE1 was more important than katE2 as the protective gene in the exponential phase of V. parahaemolyticus against extrinsic H₂O₂, and the presence of complementary katE1 on a plasmid may provide sufficient protection against extrinsic H₂O₂ and the growth-inhibitory effect of chloramphenicol. These results also suggest that ahpC1 may be the H₂O₂ detoxifier in the absence of katE1 (Fig. 2B).

FIG 2.

Growth of Vibrio parahaemolyticus strains under challenge with extrinsic hydrogen peroxide in a static culture at 37°C. (A) Effect of concentration of H₂O₂ on growth of the wild-type strain (KX-V231). ●, 0 μM; ○, 175 μM; ▼, 200 μM. (B) Effect of 175 μM H₂O₂ on growth of different strains. ●, wild type; ○, ΔkatE1 mutant; ▼, ΔkatE2 mutant; △, ΔkatE1 ΔkatE2 double mutant. (C) Effect of 175 μM H₂O₂ on growth of wild-type and complemented strains. ●, wild type; ○, ΔkatE1 mutant with complementary katE1 gene; ▼, ΔkatE2 mutant with complementary katE2 gene; △, wild-type with cloning vector.

Growth of catalase gene mutants in the presence of extrinsic organic peroxides.

The addition of 60 or 90 μM cumene significantly slowed the growth of the wild-type strain of V. parahaemolyticus at 37°C (see Fig. S3A in the supplemental material). When 60 μM cumene was applied to the single and double catalase mutant strains at 37°C, their growth did not differ significantly from that of the wild-type strain (see Fig. S3B in the supplemental material).

Adding 100 or 130 μM t-BOOH to the wild-type culture slightly reduced the extent of bacterial growth at 37°C, and the bacteria reached a lower maximal absorbance than those in the control group without peroxide. Adding 130 μM t-BOOH did not cause the growth of these catalase mutant strains to differ significantly from that of the wild-type strain (data not shown). These experiments suggest that these katE-homogenous genes may not detoxify organic peroxides.

Effect of H₂O₂ on growth of catalase gene mutants at 22 and 12°C.

At 22°C, 175 μM H₂O₂ strongly inhibited the growth of the ΔkatE1, ΔahpC1, ΔkatE1 ΔkatE2, and ΔkatE1 ΔahpC1 mutants and had no significant effect on the growth of the ΔkatE2 mutant (Fig. 3A). The presence of the complementary katE1 gene restored the growth of the ΔkatE1 mutant that was inhibited by H₂O₂ (Fig. 3B).

FIG 3.

Growth of Vibrio parahaemolyticus strains under challenge with extrinsic 175 μM hydrogen peroxide in a static culture at 22°C. (A) Mutant strains. ●, wild type; ○, ΔkatE1 mutant; ▼, ΔkatE2 mutant; △, ΔkatE1 ΔkatE2 double mutant; ■, ΔahpC1 mutant; □, ΔkatE1 ΔahpC1 double mutant. (B) Complemented strains. ●, wild type with cloning vector; ○, ΔkatE1 mutant with complementary katE1 gene.

At 12°C, the growth of bacteria was slowed. The corresponding experiment was performed for 56 h. The presence of 70 μM H₂O₂ inhibited the growth of the ΔkatE1 ΔkatE2 double mutant only for a period of about 40 h, and the growth resumed thereafter. This concentration of H₂O₂ had no effect on the wild type or on the other mutant strains (Fig. 4A). A 100 μM concentration of H₂O₂ completely inhibited the growth of the ΔkatE1 ΔkatE2 mutant for a full 56 h (Fig. 4B). A 175 μM concentration of H₂O₂ completely inhibited the growth of the ΔkatE1 and ΔkatE1 ΔkatE2 mutants at 12°C but did not affect the growth of the wild-type strain or the ΔkatE2 mutant (Fig. 4C). These experiments showed that the susceptibility of the ΔkatE1 mutant to extrinsic H₂O₂ was sensitized at incubation temperatures lower than 37°C, and they suggest that the behavior of these genes in V. parahaemolyticus is influenced by the incubation temperature.

FIG 4.

Growth of Vibrio parahaemolyticus strains under challenge with different concentration of extrinsic hydrogen peroxide in a static culture at 12°C. (A) Seventy micromolar H₂O₂; (B) 100 μM H₂O₂; (C) 175 μM H₂O₂. ●, wild type; ○, ΔkatE1 mutant; ▼, ΔkatE2 mutant; △, ΔkatE1 ΔkatE2 double mutant.

Expression of catalase genes.

In order to study how these catalase genes are influenced by incubation temperature, expression of the catalase genes (katE1, katE2, katG1, and katG2) and the ahpC1 and ahpC2 genes in the exponential phase with and without the challenge of extrinsic H₂O₂ was determined by RT-qPCR. Under the stress of extrinsic H₂O₂, the expression of katE1, katE2, and ahpC1 in the wild-type strain was significantly higher at an incubation temperature of 22°C than at an incubation temperature of 37°C, whereas katE1 showed changes of 4.7- and 0.5-fold at 22°C and 37°C, respectively (Fig. 5A).

FIG 5.

Expression of antioxidative genes in wild-type and ΔkatE1 strains of Vibrio parahaemolyticus under H₂O₂ stress. (A) Expression of antioxidative genes in the wild-type strain incubated at 22 or 37°C under challenge with 175 μM extrinsic H₂O₂; (B) expression of different genes in the ΔkatE1 mutant incubated at 22 or 37°C without extrinsic H₂O₂ stress; (C) expression of different genes in the ΔkatE1 mutant incubated at 22 or 37°C under challenge with 175 μM extrinsic H₂O₂. Expression of genes in the exponential-phase culture with or without the H₂O₂ challenge was determined by RT-qPCR, the level of expression relative to that of the corresponding gene of the wild type at each point without H₂O₂ challenge was calculated, and values at 22 and 37°C were analyzed by a t test. * and **, significantly different values (P < 0.05 or P < 0.01, respectively).

When the ΔkatE1 strain was cultured under normal conditions without challenge by extrinsic H₂O₂, the expression of ahpC1 and ahpC2 was significantly higher at 37°C than at 22°C (Fig. 5B). Under the challenge of extrinsic H₂O₂, the expression of the ahpC1, ahpC2, and VPA0305 genes was significantly higher at 22°C than at 37°C (Fig. 5C), while no significant difference was observed between the expressions of the two katG-homologous genes (katG1 and katG2) (data not shown).

DISCUSSION

Vibrio species have one to four catalase genes. V. fischeri has a single katA gene, which is critical in forming symbionts in its squid host (16), whereas V. vulnificus and V. cholerae have two catalase genes that encode catalase and catalase/peroxidase (17). V. parahaemolyticus has four putative catalase genes, which may have different functions and regulatory characteristics than the catalase genes of E. coli and other bacterial species.

Among the four putative catalase genes in V. parahaemolyticus, katE1 was demonstrated here to be similar to the monofunctional peroxidase gene (katE1) of E. coli, and it is probably the chief functional catalase gene against extrinsic H₂O₂ in the exponential phase of growth of this bacterium (Fig. 2; see Fig. S3 in the supplemental material). The other two katG-homologous genes (katG1 and katG2) of V. parahaemolyticus did not exhibit a significant antioxidative role during logarithmic growth (27). The putative amino acid sequence of the KatE1 catalase exhibits high identities of 95.6% and 80.7% with those of KatE of V. alginolyticus (accession no. AGV18944) and KatA of V. fischeri (AF011784), respectively, and 29.6% identity with that of KatE of E. coli.

In different bacterial species, different catalase genes play the major role in detoxifying peroxides. In E. coli, KatG is the predominant peroxide scavenger in the exponential phase (28), and the katG gene of V. vulnificus has a similar protective function (29). In V. cholerae, both katG and katB (a katE-like gene) are protective against H₂O₂ (17). In Rhodobacter species, whether H₂O₂ induces the expression of katE or katG depends on the species (30).

Although katE1 is probably the chief functional catalase gene in the exponential phase, the deletion mutation of this gene did not harm the normal growth of these mutant strains (see Fig. S1 in the supplemental material), their survival at a refrigeration temperature, or their high competitiveness with E. coli (Fig. 1). The endogenous ROS that is generated by aerobic metabolism in these mutants (31) may be detoxified by other antioxidative factors (31). In V. parahaemolyticus, three superoxide dismutase genes (VP2118, VP2860, and VPA1514), four ahpC or ahpF factors (VPA1683, VP0580, VPA1684, and VPA1681), and two katG-homologous genes (katG1 and katG2) may compensate for the deletion of catalase genes in these mutants (ΔkatE1 and ΔkatE2 mutants) (32, 33). Catalases and AhpC scavenge endogenous H₂O₂ that is generated by aerobic metabolism (34, 35), whereas AhpC is the primary detoxifier in Bacillus abortus (33) and E. coli (36). The katE2 and ahpC genes may have alternate or compensatory roles in the ΔkatE1 mutant (Fig. 2 and 4).

Another feature of these catalase genes is the influence of incubation temperature. The sensitivity of the ΔkatE1 mutant to extrinsic H₂O₂ was increased as the incubation temperature was reduced below 37°C (Fig. 3 and 4). A similar effect of incubation temperature on the protective function of the ahpC genes of V. parahaemolyticus and its colony size has been demonstrated elsewhere (23). Low temperature also impairs the growth of the catalase mutant of Listeria monocytogenes (37). In the cited investigations, it was shown that more ROS may be produced as the temperature falls, increasing the need for a functional catalase. The accumulation of ROS may be attributed to the expression, stability, and activities of catalases and AhpCs. The critical function of katE1 under extrinsic H₂O₂ stress at 22°C was also supported here by the high expression of this gene in the parent strain (Fig. 5A) and much greater expression of the compensatory genes in the ΔkatE1 mutant at 22°C than at 37°C (Fig. 5C).

The expression of the aforementioned genes may be regulated by controlling the incubation temperature, as has been demonstrated in Yersinia pestis (38). The thermal regulation of the expression and function of these antioxidative factors may involve rpoS, oxyR, toxRS, and other regulatory factors. The OxyR (VP2752) regulon is known to regulate the expressions of catalase genes and ahpC genes, which exhibit compensatory patterns in several bacteria (32), whereas rpoS (VP2553) is a general regulator of stress responses (39). Nevertheless, the regulation of various catalase genes or other antioxidative factors in V. parahaemolyticus has not been investigated.

In conclusion, this work demonstrates that the katE-homologous genes katE1 and katE2 are not critical for the aerobic growth of V. parahaemolyticus in a rich medium but that katE1 was the most important required detoxifier under extrinsic H₂O₂ stress during logarithmic growth. The sensitivity of the ΔkatE1 mutant to H₂O₂ increased as the incubation temperature was lowered below 37°C, and the katE2 and ahpC genes may have alternate or compensatory roles in this mutant.