Abstract
Seafood has been identified as an important source of Vibrio cholerae in Thailand, especially in the Southern coastal region. In this study, we isolated and characterized V. cholerae from seafood obtained from several markets in Hat Yai city, Southern Thailand. A total of 100 V. cholerae isolates were obtained from 55 of 125 seafood samples. The dominant serotype was non-O1/non-O139. Polymerase chain reaction (PCR) analysis was used to detect the presence of pathogenesis-related genes. The stn/sto and hlyA El Tor virulence genes were detected in 20% and 96% of the isolates, respectively. None of the isolates were positive for the ctxA, tcpA, zot, and ace genes. Only 6% of the isolates carried the T3SS gene (vcsV2); however, the majority of the isolates (96%) carried the T6SS gene (vasH). Representative isolates (n=35) that exhibited various virulence gene patterns were randomly selected and analyzed for their hemolytic activity, antibiotic susceptibility, biofilm formation, and genotype. Hemolytic activity using sheep red blood cells was detected in only one of the hlyA-negative isolates. Apart from ampicillin, all isolates were pansusceptible to five test antibiotics. Biofilm production was observed in most of the isolates, and there was no difference in the presence of a biofilm between the smooth and rugose isolates. Using the enterobacterial repetitive intergenic consensus–PCR method, clonal relationships were observed among the isolates that exhibited identical virulence gene patterns.
Introduction
V
In addition to those virulence factors, the type III (T3SS) and type VI (T6SS) secretion systems have been reported to play an essential role in the translocation of the effector proteins into eukaryotic cells and are considered as additional important virulence factors (Chatterjee et al., 2009; Ma et al., 2009). The majority of non-O1/non-O139 V. cholerae isolates from patients in China, Nigeria, Germany, and Austria carried T3SS (Luo et al., 2013; Marin et al., 2013; Schirmeister et al., 2014). One of the T3SS genes of V. cholerae, vcsV2, is homologous to the vcrD2 T3SS2 gene of V. parahaemolyticus, which is associated with enterotoxicity in the rabbit ileal loop model (Dziejman et al., 2005). Regarding T6SS, vasH is one of the regulatory genes required for its expression and vasH detection may be used as a proxy for the presence of T6SS (Kitaoka et al., 2011).
Choleragenic V. cholerae is able to transfer its virulence genes to environmental V. cholerae. Thus, environmental V. cholerae has been reported to be an important genetic reservoir of virulence genes (Faruque et al., 1998; Rivera et al., 2001), and the investigation of isolates from environments such as seafood is one way to assess the health risk associated with its consumption.
Materials and Methods
Bacterial isolation
Seafood including shrimp, squid, mollusk, fish, and crab was purchased from markets in Hat Yai city, Songkhla, Thailand, in 2013. Twenty-five grams of each seafood sample was homogenized in 225 mL of alkaline peptone water and was incubated at 42°C for 6 h (DePaola et al., 1988), followed by subculture onto thiosulfate–citrate–bile salts–sucrose agar. Sucrose-fermenting colonies were confirmed as V. cholerae by polymerase chain reaction (PCR) targeted to the gene for the outer membrane protein, ompW (Supplementary Table S1; Supplementary Data are available online at
PCR detection of virulence genes
All isolates of V. cholerae obtained from seafood were investigated. Chromosomal DNA was extracted by a boiling method (Thaithongnum et al., 2006). The ctxA, tcpA, zot, and ace genes were detected using a multiplex PCR assay with slight modifications (Singh et al., 2002). Briefly, amplification was performed in a 50-μL reaction mixture containing 2.5 μL of template DNA, 1 mM dNTPs, 0.16 μM of ctxA, and 0.3 μM of tcpA, zot, and ace primers, 2.5 U of Taq DNA polymerase (NEB, Ipswich, MA), and 1x Thermopol buffer (NEB). PCR analysis was performed using a T100 Thermal Cycler (Bio-Rad, Hercules, CA), and the reactions comprised the following cycling parameters: 95°C for 2 min, 20 cycles of denaturation at 94°C for 1 min, annealing at 62°C for 1 min, extension at 72°C for 1 min, 10 cycles of denaturation at 94°C for 1 min, annealing at 54°C for 1 min, and extension at 72°C for 1 min, followed by a final 72°C extension for 10 min. A simplex PCR assay was performed for the detection of the stn/sto, hlyA, vcsV2 (T3SS), and vasH (T6SS) genes (Rivera et al., 2001; Chatterjee et al., 2009; Marin et al., 2013). All PCR products were visualized after electrophoresis in 1.8% agarose gel.
Colony hybridization of hlyA El Tor hemolysin gene
Colony hybridization was performed to confirm the PCR hlyA negative isolates. Briefly, V. cholerae isolates were inoculated onto nylon membranes (GE Healthcare, Buckinghamshire, UK) overlying Luria-Bertani agar plates and incubated at 37°C overnight. After incubation, the membrane was treated with NaOH, neutralized, and fixed by ultraviolet cross-linking. Hybridization was performed under high stringency using hlyA-PCR amplification products labeled with digoxigenin, and detection was carried out according to the manufacturer's instructions (Roche Applied Science, Mannheim, Germany).
Determination of hemolytic activities
Hemolytic activities were determined by tube hemolysis assays (Albert et al., 1997). V. cholerae isolates were grown overnight at 37°C in heart infusion broth supplemented with 1% (vol/vol) glycerol. Then, 0.5 mL of bacterial suspension was mixed with 0.5 mL of 1% (vol/vol) sheep red blood cells in phosphate buffered saline. The mixture was incubated at 37°C for 2 h, followed by incubation at 4°C overnight. A positive hemolytic activity result was defined as an absence of red blood cells that settled at the bottom of the tube.
Antimicrobial susceptibility tests
The antibiotic susceptibility tests were carried out based on the standard disk-diffusion method recommended by the Clinical Laboratory and Standards Institute (CLSI, 2012), using ampicillin (10 μg), chloramphenicol (30 μg), ciprofloxacin (5 μg), cotrimoxazole (25 μg), norfloxacin (10 μg), and tetracycline (30 μg).
Biofilm assays
V. cholerae biofilms were assessed using a quantitative measurement of bacterial adherence to a 96-well microplate according to previously described methods (Nesper et al., 2001), with minor modifications. Briefly, overnight cultures of V. cholerae were adjusted to 0.5 McFarland standards (corresponding to 1.5×108 colony-forming units/mL), and 100 μL of the cell suspension was inoculated into the wells of a microplate. After incubation at 30°C for 24 h, unattached bacterial cells were discarded and the plate was gently washed 3 times with water. The cells were fixed and stained with 0.4% (wt/vol) crystal violet for 30 min at room temperature. After rinsing with water, dye was redissolved in ethanol–acetone (80:20), and biofilm production was measured at OD 570 nm. All experiments were done in triplicate. Statistical analysis was performed using the Mann–Whitney test (Mohamed et al., 2004).
Enterobacterial repetitive intergenic consensus PCRs (ERIC-PCR)
DNA fingerprinting of V. cholerae was performed using the ERIC-PCR technique. Briefly, DNA was extracted using a Genomic DNA extraction kit (Geneaid, Taipei, Taiwan). Amplification was conducted using the primer pair, ERIC1R (5′-ATGTAAGCTCCTGGGGATTCAC-3′) and ERIC2 (5′-AAGTAAGTGACTGGGGTGAGCG-3′) (Rivera et al., 1995). A clinical isolate of V. cholerae DMST16261, which contains the ctxA, tcpA, zot, ace, hlyA El Tor, vcsV2, and vasH genes, was used as an outgroup in this study. The amplification products were analyzed by electrophoresis using a 1.5% agarose gel. The experiments were done in duplicate. A dendrogram was constructed using a Bioprofile image analysis system (Vilber Lourmat, Torey, France).
Results
Isolation and identification of V. cholerae from seafood
A total of 125 seafood samples were obtained from various markets in Hat Yai city. Of these, 55 were positive for V. cholerae. One to 4 V. cholerae isolates, totaling 100, were selected from each sample for further analysis. As determined by PCR, all 100 isolates were confirmed as V. cholerae. All were non-O1/non-O139, apart from 1 isolate that was O1. Five rugose colonies, obtained from the different samples, were confirmed as non-O1/non-O139 V. cholerae by PCR targeting the rfb gene.
Characteristics of V. cholerae
None of V. cholerae isolated in this study contained ctxA, tcpA, zot, or ace, according to PCR analysis. The stn/sto, hlyA, vcsV2, and vasH genes were detected in 20%, 94%, 6%, and 96% of the isolates, respectively. Testing of the hlyA-negative V. cholerae by the colony hybridization method indicated that two isolates did possess the hlyA gene (Table 1).
None was detected.
Colony hybridization was used to confirm the hlyA polymerase chain reaction.
Exhibited rugose colony morphology.
ET, El Tor.
We classified the V. cholerae isolates into eight groups (A–H) according to their virulence gene patterns. Most of the isolates (72%) were classified as Group F, which was positive for hlyA and vasH (Supplementary Table S2). Only two isolates were negative for all virulence genes (Group H). Thirty-five isolates that exhibited various virulence gene patterns were selected for investigation of their antibiotic susceptibility, biofilm formation, and genotyping (Table 1).
All 35 V. cholerae isolates were susceptible to chloramphenicol, ciprofloxacin, cotrimoxazole, norfloxacin, and tetracycline. However, 5 isolates (14.3%) were resistant to ampicillin.
Thirty-two isolates produced biofilm with an OD570 value range of 0.18–4.03. There was no significant difference between the smooth and rugose isolates in their ability to produce biofilm (Mann–Whitney test, p=0.314) (Fig. 1).

Quantitative biofilm assays of Vibrio cholerae smooth and rugose isolates. Each dot indicates the median OD570 value from three determinations. Horizontal bars represent median values for the smooth and rugose isolates. No significant differences were observed in the biofilm formation between the smooth and rugose isolates (Mann–Whitney test, p=0.314).
DNA fingerprinting analysis revealed that all isolates displayed different ERIC-PCR patterns (Fig. 2). However, dendrogram analysis indicated a close relationship of the isolates within their virulence gene profile Groups B, E, and F (Fig. 2).

Enterobacterial repetitive intergenic consensus–polymerase chain reaction patterns and the resulting dendrogram of 35 Vibrio cholerae isolates and Vibrio cholerae O1 DMST16261. Numbers 1–35: Each number is correlated with the isolate number of V. cholerae listed in Table 1. A–H indicate the groups of V. cholerae listed in Table 1.
Discussion
Seafood has been reported as an important vehicle for transmission of V. cholerae to humans in countries including Italy, India, and Taiwan (Ko et al., 1998; Saravanan et al., 2007; Ottaviani et al., 2009). In Thailand, many cholera outbreaks have been reported to be associated with the consumption of seafood. In this study, we evaluated seafood in southern Thailand and found 44% of seafood samples to be contaminated with V. cholerae. In comparison, an Italian study found that only 4.7% (n=230) of fresh seafood was positive for V. cholerae (Ottaviani et al., 2009). Furthermore, in our study, all of the isolates except one were serogroup non-O1/non-O139 and none of them possessed the ctxA, tcpA, zot, or ace genes. However, 96% of them harbored hlyA, which is a similar value to the non-O1/non-O139 isolates reported in an Indian study (Singh et al., 2001). V. cholerae HlyA encoded by hlyA has been demonstrated to be associated with enterotoxicity and diarrhea (Menzl et al., 1996; Saka et al., 2008). It is of interest that all hlyA + V. cholerae isolates in this study exhibited hemolysis against sheep erythrocytes, along with three isolates of hlyA− V. cholerae. However, two hlyA− V. cholerae isolates produced a positive colony hybridization with the hlyA probe. This discrepancy could be due to variations in the target sequences of PCR primers, as the PCR results were negative. Virulence factors associated with hemolytic activity of another isolate remained unidentified. Recently, several additional hemolysins produced by V. cholerae, including hlx-coding hemolysin, hemolysin (II), lecithinase LEC, V. cholerae δ-thermostable hemolysin (Vc-δTH), and NAG-rTDH have been reported (Richardson et al., 1986; Yoh et al., 1986; Nagamune et al., 1995; Fiore et al., 1997; Fallarino et al., 2002). However, none of these appear to directly contribute to the virulence of V. cholerae.
In this study, stn/sto and vcsV2 (T3SS) were detected in 20% and 6% of nonO1/nonO139 isolates, respectively. This is similar to the findings of Rivera et al. (2001), who demonstrated that stn/sto were present in 28.2% of environmental non-O1/non-O139 Brazilian V. cholerae isolates (Rivera et al., 2001). In addition, Rahman et al. (2008) demonstrated that 11.9% of environmental non-O1/non-O139 isolates in Bangladesh carried the T3SS genes (Rahman et al., 2008). However, up to 31.5% of clinical non-O1/nonO139 V. cholerae isolates from India were previously found to be positive for T3SS (Chatterjee et al., 2009), suggesting that T3SS might be involved in the pathogenicity of noncholeragenic V. cholerae.
In this work, vasH was detected in almost all environmental V. cholerae isolates. This supports findings from a previous study from Nigeria, which showed that all non-O1/non-O139 V. cholerae isolates recovered from patients were positive for vasH (Marin et al., 2013). Therefore, the role of vasH (T6SS) in pathogenesis warrants future clarification.
Biofilm is an important factor in the persistence of environmental V. cholerae, and it has been found to enhance colonization in the suckling mouse model (Zhu and Mekalanos, 2003). In our present study, the ability of V. cholerae to produce surface-attached biofilm varied (Fig. 1), suggesting that its production might be influenced by environmental factors such as location or host. Isolates with a rugose colony morphology did not produce a higher level of surface-attached biofilm than smooth colony isolates. However, this may be due to the biofilm analysis technique used in this study, which determines surface biofilm formation, whereas rugose-type biofilms (pellicle) mostly form at the liquid–air interface (Yildiz et al., 2004).
ERIC-PCR has been reported to be a useful technique for distinguishing toxigenic V. cholerae O1 and O139 strains (ctx +, zot +) from nontoxigenic strains (ctx− , zot− ) (Rivera et al., 1995). In this study, ERIC-PCR clearly differentiated the 35 environmental nontoxigenic V. cholerae isolates. In addition, this technique revealed a high relatedness between V. cholerae isolates from within the same virulence gene pattern groups, indicating that they probably arise from the same clonal origin. We suggest that the further application of this epidemiological technique to the evaluation of clinical nontoxigenic isolates should now be considered.
Conclusions
Most of the V. cholerae isolates from seafood were non-O1/non-O139, and they were very heterogeneous regarding their virulence gene patterns as well as their DNA fingerprint patterns. Although the pathogenic mechanisms of V. cholerae non-O1/non-O139 have not been completely elucidated, their presence in seafood represents a public health warning.
Footnotes
Acknowledgments
The authors thank Dr. Brian Hodgson and Enago (
Disclosure Statement
No competing financial interests exist.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
