Presence of virulence markers in environmental Vibrio vulnificus strains

Authors


Correspondence

Carlos Vázquez Salinas, Departamento de Biotecnología, DCBS-UAM-Iztapalapa. San Rafael Atlixco #186, Col. Vicentina, 09340 Mexico, DF, Mexico. E-mail: cvs@xanum.uam.mx

Abstract

Aims

This work aims to demonstrate the presence of several genes and factors associated with virulence in strains isolated from the environment at Pueblo Viejo Lagoon, State of Veracruz, Mexico.

Methods and Results

In this study, we investigated the production of V. vulnificus virulence factors, as cytolysin (haemolysin), RTX toxin, metalloprotease, siderophores, capsular polysaccharide, adhesion structures (like type IV pili), and polar and lateral flagella, involved in swimming and swarming (or, at least, the presence of genes encoding some of them) in 40 strains of V. vulnificus isolated from water and food. The results indicate that strains of environmental origin possess potential virulence characteristics.

Conclusions

Caution should be exercised when consuming raw shellfish (especially by those more susceptible risk groups).

Significance and Impact of the Study

This is the first work focused on the evaluation of V. vulnificus virulence factors in Mexico.

Introduction

Vibrio vulnificus is an opportunistic pathogen, often associated with gastroenteritis (and even with primary septicaemia in people with pre-existing chronic conditions) following consumption of seafood or raw bivalve molluscs (Kim and Rhee 2003). They are Gram negative bacilli, straight and curved, movable by the presence of a polar flagellum; they are also oxidase positive, nonspore forming, thermolabile and behave as facultative anaerobes (Janda et al. 1988). Vibrio vulnificus is a common bacterium in the water of estuaries in tropical climates and can be present in oysters, clams and fish (Daniels 2011). During the summer months, it is considered that 100% of oysters contain V. vulnificus (Cook et al. 2002; Pruzzo et al. 2005).

The role of virulence factors of this organism is not well established yet. However, it has been described that the bacterium produces extracellular toxins such as cytolysin, encoded in the gene vvhA (Bang et al. 1999), whose activity has been associated with red cell lysis by the formation of pores in the membrane of these cells (Zhang and Austin 2005). RTX toxin, encoded in the gene rtxA, exhibits activity on the HeLa cell line and causes rounding and actin depolymerization (Kim et al. 2008). Vibrio vulnificus also produces a zinc-dependent metalloprotease, encoded in the gene vvpE, which displays several activities, such as elastase, collagenase and caseinase (Kothary and Kreger 1985; Kawase et al. 2004). Vulnibactin, a siderophore encoded in the viuB gene, allows the micro-organism to compete for iron (both in the environment and in the host) and is pointed out as a major virulence factor in this organism (Ratledge and Dover 2000; Paniker et al. 2004).

The capsule is considered, however, the major virulence factor of V. vulnificus, being a protective shield against the action of the complement and of phagocytosis (Whitfield 2006). The operon involved in the synthesis of this polysaccharide has a high similarity to previously identified operons in Escherichia coli (Livanis et al. 2006). The gene wza encodes a lipoprotein that forms ring-like structures in the outer membrane, where the capsular polysaccharide is secreted (Collins and Derrick 2007).

Vibrio vulnificus has the ability to adhere to epithelial cells by type IV pili (Gander and LaRocco 1989). The operon encoding this pili formation consists of four genes, pilABCD (Paranjpye et al. 2005). It has also been found that flagella allow V. vulnificus the initial adsorption, biofilm formation and the subsequent invasion of the host (McCarter 2001). When a bacterium swims, its flagellum moves the cell through a fluid medium, allowing its displacement in low-viscosity media. The swarming migration is defined as a rapid and coordinated migration of flagellated bacteria through solid surfaces. By itself, this effect is considered also a virulence factor and has been proposed to play an important role in the regulation and expression of virulence factors associated with invasiveness in V. vulnificus (Kim et al. 2006).

In Mexico, only nine case reports related to this organism are known. According to data from the Center for Disease Control and Prevention (CDC) in the United States, approximately 100 people per year are infected with V. vulnificus, causing about 50 deaths (Daniels 2011). This work aimed to demonstrate the presence of several genes and factors associated with virulence in strains isolated from the environment at Pueblo Viejo Lagoon, State of Veracruz, Mexico.

Materials and methods

Bacterial strains and growth medium

We used 40 strains of Vibrio vulnificus, isolated from water samples (10), fish (10) and oysters (20), from Pueblo Viejo Lagoon, Veracruz, Mexico.

The isolated strains of V. vulnificus were seeded in tubes containing trypticase soy broth with 2% NaCl and incubated for 18–24 h at 35–37°C. For enzyme detection tests, strain V. vulnificus ATCC 29397 was used as a positive control.

Analysis of virulence determinants

Proteolytic activity was determined by cross-seeding bacteria on skim milk agar, at a final concentration of 2%. Plates were incubated at 35–37°C for 18–24 h. The presence of a clear halo around the colony was indicative of proteolysis (García Moreno and Landgraf 1998).

To detect haemolysin, strains were cross-seeded on blood agar with 5% sheep and rabbit erythrocytes. Plates were incubated for 18–24 h at 35–37°C, and during this time, the appearance of haemolysis halos was observed (García Moreno and Landgraf 1998).

For detection of siderophores, strains were incubated in nutrient broth for 18–24 h at 37°C. Next, bacteria were seeded in agar CAS and incubated for 24 h at 37°C. Bacterial colonies producing siderophores in this medium were surrounded by a yellow-orange halo (Schwyn and Neilands 1987).

For capsule presence, cultures incubated in brain and heart infusion (BHI) broth for 18–24 h at 35–37°C were used. Strains were cross-over plated on glycerol agar with 2% NaCl and incubated at 35–37°C for 18–24 h. India ink staining was used to observe the capsule.

For swimming (Kim et al. 2007), strains of V. vulnificus were cross-over plated on BHI agar containing 2% NaCl and incubated at 37°C for 24 h. A single isolated colony was selected to inoculate BHI medium plates with 0·3% agar. The Petri dishes were incubated for 24 h at 37°C.

For swarming (Kim et al. 2007), strains of V. vulnificus were cross-over plated on BHI agar containing 2% NaCl and incubated at 37°C for 24 h. A single isolated colony was selected to inoculate BHI medium plates with 1% agar. The plates were incubated for 48 h at 37°C.

To determine the cytotoxic effect of V. vulnificus, each of the isolated strains was seeded in a tube containing 15 ml of AKI broth and incubated for 18 h in a shaker (200 rpm) at 37°C. The culture was centrifuged at 8000× g for 10 min. The supernatant was filtered using a 0·22-μm membrane (Kaysner and DePaola 2004). Vibrio cholerae ATCC 11623 was used as the positive control, and Escherichia coli ATCC 25922 was used as the negative control.

In a 96-well plate, 100 μl of CHO (Chinese hamster ovary) cells in F12 medium supplemented with 10% foetal calf serum was placed per well. The plate was incubated at 37°C with a 5% CO2 atmosphere for 24 h. The medium and serum were removed, and the plate was washed three times with sterile phosphate-buffered saline (PBS). Then, 80 μl of F12 medium and 20 μl of supernatant from each strain culture were added per well. The plate was incubated at 37°C with an atmosphere of 5% CO2. The microplate was observed under a reversed field microscope at hourly intervals to visualize changes in cell morphology (elongation, rounding or monolayer destruction). The medium and the supernatant were then removed, the plate was washed three times with sterile PBS, and absolute methanol was added to cover the wells. Immediately, methanol was removed, and the plate was washed three times with PBS. Cells were stained with Giemsa, covering the wells for 20 min. The dye was recovered, and the plate was washed with distilled water until no leftover dye was observed. The cytotoxic effect was observed with a reversed field microscope (Thielman et al. 1997).

Genetic analysis

Strains were grown in Luria–Bertani broth and incubated at 37°C for 24 h. Wizard ® Genomic DNA Purification kit (Promega, Madison, WI, USA) was used to obtain DNA.

The amplification mix was prepared with 1X buffer, 25 mmol l−1 MgCl2, 40 μmol l−1 of dNTP mix, 50 pmol of each primer, 1·66 U of Taq polymerase and 100 ng of DNA, in a final volume of 50 μl. Amplifications were performed in a Hybaid thermocycler-Omn-E, with the primers and conditions reported in Tables 1 and 2. As positive control, genomic DNA from the reference strain V. vulnificus ATCC 29307 was used. As negative control, V. mimicus ATCC 33653 genomic DNA was chosen.

Table 1. Primers used in this study
TargetPrimerReference
vvhA

F- 5′TTCCAACTTCAAACCGAACTATGA-3′

R-5′ATTCCAGTCGATGCGAATACGTTG-3′

Paniker et al. (2004)
viuB

F- 5′GGTTGGGCACTAAAGGCAGATATA-3′

R-5′TCGCTTTCTCCGGGGCGG-3′

Paniker et al. (2004)
pilA

F- 5′TGGCTGCTGTTGCTATTC-3′

R-5′GGTCCACCACTAGTACCAAC-3′

Paranjpye et al. (2005)
Wza

F- 5′ATTCCGTGACCGATTGAGCGT-3′

R-5′GCAGTAGAAGATACACCTAGG-3′

Wright et al. (2001)
vvpE

F5′-GTCGCGGAAGAAGAGCC-3′

R5′-GGCCGTGAGAGCACTCCGG-3′.

Shao and Hor (2000)
rtxA

F 5′-CGGGATCCTATGGCGTGAACGGCGAAG-3′

R 5′-CGGGATCCAGCAGCCACAAGCGATTC-3′

Kim et al. (2008)
Table 2. Conditions used for the detection of genes vvhA, viuB, pilA, rtxA, vvpE and wza
 vvhA viuB pilA vvpE rtxA wza
Initial denaturing step94°C per 3 min94°C per 3 min92°C per 2 min94°C per 3 min94°C per 3 min92°C per 5 min
Denaturing94°C per 1 min94°C per 1 min92°C per 1 min94°C per 45 s94°C per 30 s92°C per 1 min
Hybridization68°C per 1 min68°C per 1 min50°C per 1 min55°C per 45 s61°C per 30 s63°C per 1 min
Elongation72°C per 1 min72°C per 1 min72°C per 1 min72°C per 1 min72°C per 1 min72°C per 1 min
Number of cycles303030303030
Amplicon size (pb)2053162174261440880

Results

Haemolytic and enzymatic activity, siderophore production and presence of capsule

All strains (40/40) analysed allowed β-haemolysis under the conditions tested; 92·5% (37/40) had proteolytic activity, only 25% (10/40) of strains allowed siderophore detection, and 70% (28/40) of strains showed the presence of capsule.

Swimming and swarming assay

All the 40 strains studied were capable of producing swimming movements, which is characterized by the migration of bacteria in a disorderly manner and spread throughout the agar. All of them were also able to induce the swarming effect in the plate, due to formation of lateral flagella.

Cytotoxicity assay

All the supernatants of V. vulnificus strains were cytotoxic to CHO cell line in an incubation period of 10 h. This effect was characterized by the change in cell morphology (elongation) and destruction of the monolayer (Fig. 1).

Detection of genes vvhA, viuB, pilA, rtxA, vvpE and wza

Table 3 shows the results of the amplification of all the genes tested in this study. By using the PCR technique, the presence of the gene rtxA became apparent in 25% (10/40) of the tested strains. vvhA gene could be amplified from 100% of strains, showing a 205-bp-sized amplified fragment in this case (suggesting that all the strains used in this work are indeed V. vulnificus, as this gene is used as a target to identify this micro-organism).

Table 3. Detection of genes vvhA, viuB, pilA, vvpE, rtxA and wza in tested strains
Source of strainsPositive samples
vvhA viuB pilA vvpE rtxA wza
Water10 410 3 3 6
Oyster20 519 6 516
Fish10 2 9 3 2 6
Total401138121028

The gene viuB was amplified in 27% of the tested strains (Fig. 2). This result correlates with siderophore production in agar CAS. In 95% of the strains tested, pilA presence was revealed; in this case, the amplified fragment was 217 bp in size.

Figure 1.

Effect of a Vibrio vulnificus strain isolated from oyster on Chinese hamster ovary (CHO) cell line. Panel (a) shows the control (noninoculated CHO cells). Panels (b and c) show cell elongation and destruction (monolayer structure is absent). Panel (d) shows destruction of the monolayer at 10 h of incubation.

Figure 2.

Electrophoresis gels of vvhA, viuB, pilA, rtxA, vvpE and wza genes' amplification by PCR in strains of Vibrio vulnificus. (a) vvhA gene. (b) viuB gene. (c) pilA gene. (d) rtxA gene. (e) vvpE gene. (f) wza gene. In all the panels, lane 1 corresponds to the molecular size marker (100-bp ladder or λ bacteriophage DNA), lane 2 corresponds to the positive control, lanes 3–5 to oyster strains and lanes 6–7 to fish strains.

vvpE gene was detected in 30% (12/40) of the strains, and wza in 70% (28/40), which also correlated with negative staining tests performed to observe the capsule.

Discussion

The presence of virulence factors in V. vulnificus is a recognized matter; however, the distribution of the virulence factors described so far is still unknown in strains of environmental origin.

The pathogenicity mechanisms of V. vulnificus involve both structural and extracellular components. The rtxA gene was amplified in 25% (10/40) of the strains obtained in this work. There are no reports in the literature on the distribution of this gene in environmental strains of V. vulnificus. However, Cordero et al. (2007) amplified this gene in 95% of clinical and environmental V. cholera strains. It has been reported that the toxin RtxA of V. cholerae is the most potent cytotoxin from this bacterium, with ‘cross-linking’ activities and being secreted outside the cell by a type I secretion system (TISS) (Boardman and Satchell 2004; Sheahan et al. 2004).

Kim et al. (2008) developed various studies with mutants in genes vvpE, vvhA and rtxA, finding that the RtxA toxin plays the main cytotoxic role in the disease caused by V. vulnificus, whereas the role of VvhA seems to be auxiliary and VvpE probably has an insignificant role. rtxA mutation demonstrates that its encoded toxin is related to cell death (Lee et al. 2008). In general, it seems likely that the RtxA toxin is also the main toxin in V. vulnificus implicated in both cytotoxicity and virulence.

All strains (100%) showed β haemolysis and cytotoxic effect on CHO cell line at 10 h of incubation. This correlated with vvhA gene detection, encoding a haemolysin with cytolysin activity. Kreger and Lockwood (1981) were the first to demonstrate haemolytic and cytolytic activity of V. vulnificus cytolysin; they reported a rounding effect on cells at 24 h of incubation (this effect was not observed in our work, because the monolayer was destroyed at 24 h).

This cytolysin, injected subcutaneously at concentrations of 3 μg kg−1, is lethal in mice, causing structural changes in the skin. If the infection occurs through wounds, the observed damage is similar (Kwon et al. 2001; Rho et al. 2002). A very important feature of this cytolysin is its ability to lyse red cells through pore formation, releasing the haemoglobin's iron and thus providing an additional source of iron to the organism (Kim et al. 1993). The vvhA gene is highly conserved in this species and has been used as a target for the identification of V. vulnificus (Paniker et al. 2004).

Twenty-five percentage of the tested strains showed siderophore production in agar CAS, and PCR amplification of the gene viuB was possible in 27% of them. These data correlate with those described by Paniker et al. (2004), who noted that this gene is poorly distributed in environmental strains. Humans naturally synthesize transferrin or lactoferrin to replenish iron once erythrocytes fulfil their lifecycle. When V. vulnificus affects humans, siderophores capable of competing for iron uptake are produced (Ashrafian 2003), and the siderophore vulnibactin, catechol-type, is the main mechanism by which V. vulnificus acquires iron. To demonstrate the importance of this siderophore in the pathogenesis of V. vulnificus, several studies have been conducted with genes involved in the synthesis of the compound. Mutation of various genes associated with vulnibactin's function (vuuA, venB, vvsA and vvsB) resulted in decreased virulence of this bacterium in animal models (Litwin et al. 1996; Webster and Litwin 2000; Kim et al. 2008).

The pilA gene was amplified in 95% of the tested strains. This gene participates in the operon encoding the type IV pili. Gander and LaRocco (1989) identified the pili of V. vulnificus by electron microscopy and found that 80% (16/20) of their clinical isolates showed this structure, whereas only 30% (3/10) of their environmental isolates showed it. Paranjpye et al. (2005) detected, in 100% (27/27) of their isolates, the presence of the pilA gene and linked it with the ability to adhere to human epithelial cell lines. They also observed that a pilA mutation does not cause pili formation and affects adhesion in HEp-2 cell line. It has been shown that during human infection by V. cholera, a homologous gene for pilA is expressed that is related to colonization of the gastrointestinal tract (Hang et al. 2003).

The vvpE gene was detected in 30% (12/40) of the strains tested in this work. This is the first work describing the presence of this gene in environmental strains. VvpE metalloprotease has broad substrate specificity, and it has been shown that the purified enzyme causes tissue necrosis and skin damage, as well as an increased vascular permeability leading to oedema, characteristic of vesicular lesions caused by systemic diseases (Chang et al. 2005; Miyoshi 2006).

We found that 100% of the tested strains presented swimming and swarming movements, necessary for adhesion not only to biotic but also to abiotic surfaces. Swarming is a multicellular phenomenon, essentially requiring cell–cell or cell–surface interactions (Harshey 1994; Fraser and Hughes 1999). A possible mechanism that explains the interaction of several of these factors would imply VvpE participation in the hydrolysis of cell surface proteins, resulting in modifications that assist cell–cell or cell–surface interactions that facilitate swarming (Dunny and Leonard 1997). Therefore, VvpE may also have an important role in adhesion to the surface, colonization and invasion by V. vulnificus (Kim et al. 2007). During the pathologic process, V. vulnificus must first adhere and colonize the surface of the skin and intestinal tract to establish the infection. For this purpose, V. vulnificus manifests swarming-type mobility (Kim and Rhee 2003). Swarming movement is considered a virulence factor in pathogens such as Salmonella enterica serovar Typhimurium and is accompanied by an increase in the expression of various genes associated with pathogenicity (Connelly et al. 2004; Wang et al. 2004).

In 70% of the analysed strains of V. vulnificus, both the presence of the capsule and wza gene amplification were observed. One function of the capsule is to protect the organism from desiccation effects, forming a hydrated gel structure around the bacterium (Collins and Derrick 2007). But, the main function in the pathogenesis of this structure is to resist the action of the complement, being a barrier that masks structures that can be potent activators of the alternative pathway (Roberts 1996; Collins and Derrick 2007). Because of this, septicaemia can be triggered, with a lethality rate over 50% (Shapiro et al. 1998).

According to the results obtained in this work, environmental strains of V. vulnificus present factors associated with virulence, so caution should be exercised by not consuming raw or inadequately cooked shellfish, especially by those population groups most susceptible to this bacterium.

Acknowledgments

Ivan Natividad-Bonifacio was supported with a grant from the National Council for Science and Technology (CONACyT), with registration number 174703. We thank Ingrid Mascher for the English revision of this manuscript.

Ancillary