Abstract
Vibrio vulnificus is a ubiquitous toxigenic bacterium found in a coastal environment but little is known about its occurrence and seasonality among seaweeds, which are widely consumed as seafood in Japan. Therefore, we have observed the bacterium's abundance in seawater and seaweed samples from three areas of the Kii Channel, Japan, during June 2003 to May 2004. A total of 192 samples were collected: 24 from each source in summer, autumn, winter and spring. The samples were selectively cultivated following the most probable number (MPN) technique. Vibrio vulnificus population ranged from 0 to 103 MPN 100 mL−1 seawater or 10 g seaweeds; higher counts were observed during summer. The optimum temperature, salinity and pH for the bacterium were 20–24 °C, 24–28 p.p.t. and 7.95–8.15, respectively. However, seaweeds always contained higher V. vulnificus than seawater. Among 280 V. vulnificus strains, detected by species-specific colony hybridization and PCR, 78, 74, 11 and 16 were from seaweeds and 46, 42, 2 and 11 were from seawater during summer, autumn, winter and spring, respectively. Ribotyping of 160 selected strains revealed a higher genotypic diversity (18 patterns) among strains from seaweeds than from seawater (10 patterns). Seaweeds can thus act as a potential habitat for V. vulnificus and are more unsafe for consumption during summer.
Introduction
Vibrio vulnificus is a Gram-negative marine gammaproteobacterium that can cause severe human infections such as gastroenteritis, invasive septicemia, necrotizing fasciitis, osteomyelitis, etc., especially among coastal inhabitants, fishermen and tourists. The modes of infections are generally through consumption of raw seafood or when the wounded parts of the skin come in contact with contaminated waters (Dalsgaard et al., 1996). The bacterium is considered to be an emerging pathogen, with a possible link to the global climate change and the resulting increased seawater temperature (Paz et al., 2007). In August 2005, extensive flooding due to Hurricane Katrina caused dozens of Vibrio-related illness including five deaths among the coastal population in the United States; V. vulnificus was the predominant bacterium (Centers for Disease Control and Prevention, 2005). In marine and coastal environments, the bacterium can be found free-living in water or associated with plankton and other aquatic fauna including a variety of fish species, oysters, clams, mussels, scallops, etc. (DePaola et al., 1994, 1998; Wright et al., 1996; Høi et al., 1998). Two major biotypes of V. vulnificus are known: biotype 1 is related to human infection while biotype 2 is exclusively found among diseased fish, particularly eel (Høi et al., 1998). Recent findings indicate the spread of pathogenic V. vulnificus strains into new regions of the earth, e.g. aquaculture settings in Spain and Denmark (Fouz et al., 2006).
The bacterium poses a serious health threat to humans with low immunity or underlying diseases (liver dysfunction, alcoholic cirrhosis or hemochromatosis), with a fatality rate c. 50%, and is thus considered to be one of the most rapidly fatal human pathogens (Hilton et al., 2006). Vibrio vulnificus differs significantly from other vibrios because of its rapid invasive characteristic, which facilitates its pathogenesis as well as association with aquatic organisms. The protease enzyme secreted by the bacterium can cause tissue damage during infection (Miyoshi et al., 1987). Besides, V. vulnificus also secretes chitinase, cytolysin, hemolysin, mucinase and elastase, which can also aid in its intimate association with aquatic organisms (Somerville & Colwell, 1993; Moreno & Landgraf, 1998). Previous investigations have highlighted the role of temperature and salinity in the abundance and distribution of V. vulnificus in coastal waters as well as among shellfish samples (Kaspar & Tamplin, 1993; DePaola et al., 1994; Wright et al., 1996). Microbial ecologists are nowadays highlighting bacterial genotypic diversity as a tool to understand its evolution, competitive interactions and to trace the spread of pathogenic genotypes. Ribotyping of the conserved 16S rRNA gene is considered to be one of the most effective genotyping methods because of its excellent reproducibility, good discriminatory power, ease of interpretation and integration to automation (Grimont & Grimont, 1986; Koblavi et al., 1990).
Japanese coasts are among the few areas in the world where the greatest diversity of seaweeds is found. Seaweeds have been widely used as popular food and medicines in Japan since ancient times, and its contribution is the highest among the dietary fiber intake among Japanese (Fukuda et al., 2007). Our recent studies have shown that seaweeds can act as a potential habitat for Vibrio parahaemolyticus, another coastal Vibrio species, along the Kii Channel, Japan (Mahmud et al., 2006, 2007). Occurrences of V. vulnificus have been reported from different regions of Japan (Venkateswaran et al., 1989; Hara-Kud et al., 2005). However, there is no extensive study on the quantitative abundance, seasonality and genetic diversity of V. vulnificus strains associated with coastal seaweeds in comparison with its free-living population in the water column. Therefore, in the present investigation we have observed these unexplored ecological aspects of V. vulnificus along the Kii Channel to define whether seaweeds can harbor this pathogenic organism and can thus be potentially risky to use as seafood during a certain period of the year. Molecular methods based on a species-specific probe were used for better isolation and specific detection of the bacterium among thousands of other vibrios in the coastal environment.
Materials and methods
Sample collection, processing and selective culture for V. vulnificus
Seawater and seaweed samples were collected seasonally (two times in each of summer, autumn, winter and spring) from June 2003 to May 2004 from three coastal areas (Komatsushima, Tokushima and Naruto) of the Kii Channel, Japan (Fig. 1). In each area, samples were collected at four different sites following the method of American Public Health Association (1998). Subsurface seawaters (0.5 m depth) were collected from sampling positions located 50 meters away from the shore using an engine-driven boat. The seaweed samples were collected during high tide from the submerged seaweed beds on the slope of the banks using a sterile knife. Three subsamples were collected randomly for each environmental setting and then pooled together. Surface water temperature and salinity were monitored at each site using a salinometer (model S-27, Sankyo Pharmacy Co. Ltd, Japan). Water pH was also checked using a pH meter (Horiba pH meter F-22, Japan).
Map showing the location of Kii Channel in Japan (left) with an enlarged image of coastal areas (right) showing the sampling locations (Komatsushima, Tokushima and Naruto).
A total of 192 environmental samples were collected, 96 from each of the seawater and seaweed sources, and in each of the four seasons 24 samples were collected for both the groups. The samples were processed within 12 h of collection following aseptic techniques. For seaweeds, a 10 g portion of each sample was extensively washed with sterile normal saline (0.9% w/v of NaCl, pH 7.5) to remove bacteria in surrounding seawater, then mixed with adequate quantities of sterile normal saline and crushed using a grinder and finally the volume was adjusted to 100 mL. Both seawater and seaweed homogenates were spread plated on thiosulfate-citrate-bilesalt-sucrose (TCBS) agar following a standard method and incubated at 37 °C overnight to obtain the total Vibrio count (CFU mL−1). The most probable number (MPN) culture method was applied in both samples for quantification of the V. vulnificus population according to the method of American Public Health Association (1998). In the MPN culture method (three-tube, three-dilution) samples were enriched in alkaline peptone water (APW) at 37 °C overnight. A portion of the enrichment broth (two to three loopfuls) was then subcultured onto TCBS agar (BD) and CHROMagar Vibrio (CV) (CHROMagar, Paris, France) and grown at 37 °C overnight. The green or blue-green colonies, 2–3 mm in diameter on TCBS agar plates and pale blue colonies on CV agar plates, were presumptively selected as V. vulnificus. Each of the selected colonies was confirmed by their colonial characteristics after they were transferred using sterile toothpicks and freshly grown on both TCBS and CV agar plates. Bacterial colonies having morphological characteristics similar to V. vulnificus were then selected, transferred onto Luria–Bertani (LB) agar plates (2% NaCl) using the patch inoculation technique and then subjected to blot preparation after overnight incubation at 37 °C.
Colony blot hybridization using a species-specific probe
Whatman 541 filter paper (Whatman Int. Ltd, UK) (8 cm diameter) was marked and placed over the LB agar and incubated for 2 h at room temperature. It was then removed, placed with the colony side up on top of a Whatman 3 M filter paper and colony blots were prepared as described previously (Mahmud et al., 2006). Positive and negative controls were prepared on a separate strip of Whatman 541 filter paper.
A synthetic oligonucleotide probe for V. vulnificus cytolysin structural gene vvhA 5′-GAGCTGTCACGGCAGTTGGAACCA-3′, 1857–1880 bp; (Morris et al., 1986), was purchased from Kurabo Biomedical, Japan. Purified vvhA oligonucleotide was labelled at the 5′ end (50 pmol of 5′ end per reaction) by transfer of 32P from [γ-32P] ATP with T4 polynucleotide kinase (Invitrogen, Life Technologies) following the procedures described elsewhere (Mahmud et al., 2006). The labelled probe DNA was then purified by a G-50 column (Sigma) according to the instructions of the manufacturer, and its specific activity ranged from 6 × 108 to 8 × 108 cpm μL−1. The blots were hybridized at 40 °C overnight in a solution consisting of 6 × SSC (1 × SSC containing 0.15 M NaCl, 0.015 M sodium citrate, pH 7.0), 5 × Denhardt solution (1 × Denhardt solution containing 0.02% Ficoll, 0.02% polyvinylpyrrolidone and 0.02% bovine serum albumin), 1 mM EDTA (pH 8.0), 100 μg of boiled salmon sperm DNA mL−1 and probe DNA (106 cpm per blot). Afterwards, the blots were washed two times, 1 h each, in 6 × SSC at 60 °C. Then they were briefly rinsed in 2 × SSC at room temperature and air-dried. The blots were then exposed to a phosphor imaging plate (Fuji, Japan) for 24 h, followed by the detection of radioactivity by a high-resolution imaging scanner (Storm 840, Amersham Biosciences).
Preparation of chromosomal DNA
Chromosomal DNA was extracted from each of the colonies detected by V. vulnificus-specific colony hybridization according to Chowdhuryet al.(2000) with some modifications. In brief, cells from 18-h cultures in LB broth were harvested and resuspended in TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). The cell pellets were treated with 10% sodium dodecyl sulfate (SDS) (Sigma, Japan) and freshly prepared proteinase K (100 μg mL−1) (Sigma, Japan) and incubated for 1 h at 37 °C. Then CTAB–NaCl (10% cetyl trimethyl ammonium bromide in 0.7 M NaCl) was added and incubated at 65 °C for 15 min. The aqueous phase was treated two times with phenol–chloroform and the resulting DNA was precipitated with 0.6 volume of isopropanol and washed with 70% ethanol. RNA contaminations were minimized by resuspending the DNA in TE buffer and treating with RNAse (100 μg mL−1) at 37 °C for 1 h. Then, purified DNA was precipitated, washed with 70% ethanol, air-dried and finally dissolved in TE buffer.
Colony confirmation by PCR
To confirm the presence of the vvh gene in the bacterial colonies detected by colony hybridization, PCR was performed to amplify 340 bp DNA fragment using the primers Vv1 (5′-CGC CGC TCA CTG GGG CAG TGG CTG-3′) and Vv2 (5′-CCA GCC GTT AAC CGA ACC ACC CGC-3′) (Morris et al., 1986). The amplification was performed in a 25 μL volume consisting of 0.5 μg of chromosomal DNA, 0.2 pM of each of the oligonucleotide primers for vvh, 2.5 μL of 10 × PCR reaction buffer (500 mM Tris-Cl, pH 8.9, 500 mM KCl and 25 mM MgCl2), 0.5 μL 10 mM dNTPs, 1.25 U AmpliTaq DNA polymerase (Invitrogen, Life Technologies) and an appropriate volume of sterile MilliQ water. The amplification was carried out in a thermal cycler (Perkin-Elmer, Cetus) with initial denaturation at 95 °C for 5 min, followed by 30 cycles of amplification (denaturation at 95 °C, annealing at 58 °C and extension at 72 °C, each for 1 min). Finally, the samples were kept at 72 °C for 10 min as the final extension step. PCR-amplified products were separated in a 1.5% agarose gel by electrophoresis, followed by staining with ethidium bromide (0.5 μg mL−1) and visualization under a UV transilluminator.
Strain selection for ribotyping
A total of 57 and 103 V. vulnificus strains were selected for ribotyping from water and seaweed sources, respectively. All strains isolated during winter and spring were selected because of their low number, while 50% of the summer and autumn strains were selected because of their large number (see ‘Results and discussion’). Differences in colony morphology, sampling time and source were the selective criteria to obtain diverse and representative strains from the summer and autumn samples, and in case of the same source, time and similar morphology, the strains were chosen randomly.
Preparation of the 16S rRNA gene probe and ribotyping
A universal 16S rRNA gene-specific PCR was conducted using 5′-GGATTA GATACC CTG GTA GTC C-3′ (forward) and 5′-TCG TTG CGG GAC TTA ACC CAA C-3′ (reverse) primers (Talukder et al., 2002); the PCR product was excised from the gel and purified by the GenElute™ Minus EtBr Spin Column (Sigma) according to its instruction manual. The purified amplicon was labelled with digoxigenin (DIG)-dUTP using a random primed DNA labelling Kit (Boehringer Mannheim, Germany) according to the manufacturer's instructions.
At first, several restriction enzymes (HindIII, EcoRI and BglI) were used to digest the chromosomal DNA of some representative strains to find out the most appropriate enzyme for differentiating V. vulnificus strains. Restriction endonuclease HindIII (Takara Shuzo, Japan) was then finally used to digest the selected 160 representative strains. Complete digestion was performed with 2–3 μg of chromosomal DNA at 37 °C for 17 h according to the manufacturer's instructions. Electrophoresis was performed in 0.8% agarose gel (150 mm × 135 mm × 10 mm) at 10 V cm−1 of gel for 3 h for separation of restriction fragments. DNA molecular-weight marker II (digoxigenin-labelled, Boehringer Mannheim, Germany) was used during electrophoresis to compare the size of the restriction fragments.
Agarose gels were denatured in 0.5 M NaOH solution containing 1.5 M NaCl, neutralized in 0.5 M Tris buffer (pH 7.2) containing 1.5 M NaCl and 1 mM EDTA, and washed in 4 × SSC for 10 min (Maniatis et al., 1989). Afterwards, the restriction fragments in the gel were transferred onto a nylon membrane (Hybond-N, Amersham) by capillary transfer, and then rinsed with 2 × SSC, followed by incubation at 80 °C for 2 h for fixation of DNA to the membrane. Hybridization of the membrane with the prepared 16S rRNA gene probe, washing and detection of fragments (immunologically, with nitroblue tetrazolium chloride as a substrate for labelled alkaline phosphatase) were performed according to the manufacturer's instructions (Boehringer Mannheim, Germany).
Statistical analysis
Analysis of the results and trends in V. vulnificus occurrence (both free-living in water and seaweed associated) with respect to changes in the physicochemical properties of water (e.g. salinity, temperature and pH) was conducted using the data desk software version 6.0 (Data Description Inc., NY). The average of the MPN values of V. vulnificus populations had large SDs; thus, median and mode values were calculated to check the bacterium's comparative abundance in each season among water and seaweed samples. During similarity analysis among ribotype patterns, the size of each band in the gels was determined and the data were coded as 0 (negative) or 1 (positive). The relatedness among V. vulnificus strains in different seasons was examined using hierarchical cluster analysis, and dendrograms were produced using the spss software version 10.0 (SPSS Inc., Chicago, IL).
Results and discussion
Seasonal abundances of total Vibrio and V. vulnificus populations
The temporal variations in cultivable Vibrio (TCBS counts) are shown in Fig. 2. The total Vibrio counts were 0–104 CFU mL−1 and 102–106 CFU g−1 in water and seaweeds, respectively. The highest counts were observed in summer and autumn samples. During the winter months, no cultivable Vibrio spp. could be detected from water samples while the seaweed samples still harbored a population of 103–104 CFU g−1. The seaweed samples mainly comprised Porphyra, Undaria, Laminaria and Fucus species.
Vibrio abundance among seawater and seaweed samples in the Kii Channel in different seasons of the year.
Like the total Vibrio population, V. vulnificus abundance was also higher in the summer and autumn months (water temperature 20–29 °C) when all the water and seaweed samples were contaminated with cultivable V. vulnificus (Table 1). During these warmer months, c. 20% of the samples from both water and seaweeds (19 out of 96) yielded >1100 MPN V. vulnificus 100 mL−1 seawater or 10 g seaweeds, while in 36 samples (37.5%) the counts ranged between 251 and 1100 MPN V. vulnificus 100 mL−1 seawater or 10 g seaweeds, respectively. However, V. vulnificus abundances in seaweeds were always higher in comparison with the water samples (Table 1). During summer and autumn, the bacterium's population in seawater had median MPN values of 240 and 290, respectively, when the seaweed-associated population had a median MPN value of 460 in both the seasons. In contrast, during winter and spring (water temperature 10–18 °C) their abundances were very low (<3–150 MPN V. vulnificus 100 mL−1 seawater or 10 g seaweed). The seasonal changes among the observed free-living and seaweed-associated V. vulnificus population in this study are comparable to the seasonal variations of V. parahaemolyticus in the same coast that we have reported previously (Mahmud et al., 2006). During winter, 2 and 11 samples out of 24 were positive for V. vulnificus in water and seaweed sources, respectively. Overall, the presence of V. vulnificus among 61 and 75 out of 96 samples each of seawater and seaweed samples, respectively (Table 1), suggests the bacterium's endemicity in these two habitats along the coast of the Kii Channel. The bacterial populations that were free-living in water during summer and autumn had a wide range of MPN values, e.g. between 3 and >1100 MPN V. vulnificus 100 mL−1, while during summer their distribution was bimodal (460 and >1100 MPN). This type of distribution indicates the nonuniformity in V. vulnificus abundance, which might be attributable to particular environmental influences.
Vibrio vulnificus population abundance among water and seaweed samples at different seasons of the year (sample number, n=24 for each source in each season)
| Category | Summer | Autumn | Winter | Spring | ||||
| Water | Seaweeds | Water | Seaweeds | Water | Seaweeds | Water | Seaweeds | |
| Abundance (MPN 100 mL−1 water or 10 g−1 seaweed) | ||||||||
| >1100 | 3 | 7 | 3 | 6 | 0 | 0 | 0 | 0 |
| 251–1100 | 7 | 9 | 9 | 11 | 0 | 0 | 0 | 0 |
| 151–250 | 4 | 3 | 6 | 2 | 0 | 0 | 0 | 0 |
| 51–150 | 6 | 5 | 4 | 5 | 0 | 1 | 2 | 5 |
| 21–50 | 2 | 0 | 0 | 0 | 0 | 9 | 5 | 11 |
| 3–20 | 2 | 0 | 2 | 0 | 2 | 1 | 4 | 0 |
| <3 | 0 | 0 | 0 | 0 | 22 | 13 | 13 | 8 |
| Median | 240 | 460 | 290 | 460 | <3 | 3 | <3 | 21 |
| Mode | 460, >1100 | >1100 | 460 | >1100 | <3 | 21 | <3 | 21 |
| Category | Summer | Autumn | Winter | Spring | ||||
| Water | Seaweeds | Water | Seaweeds | Water | Seaweeds | Water | Seaweeds | |
| Abundance (MPN 100 mL−1 water or 10 g−1 seaweed) | ||||||||
| >1100 | 3 | 7 | 3 | 6 | 0 | 0 | 0 | 0 |
| 251–1100 | 7 | 9 | 9 | 11 | 0 | 0 | 0 | 0 |
| 151–250 | 4 | 3 | 6 | 2 | 0 | 0 | 0 | 0 |
| 51–150 | 6 | 5 | 4 | 5 | 0 | 1 | 2 | 5 |
| 21–50 | 2 | 0 | 0 | 0 | 0 | 9 | 5 | 11 |
| 3–20 | 2 | 0 | 2 | 0 | 2 | 1 | 4 | 0 |
| <3 | 0 | 0 | 0 | 0 | 22 | 13 | 13 | 8 |
| Median | 240 | 460 | 290 | 460 | <3 | 3 | <3 | 21 |
| Mode | 460, >1100 | >1100 | 460 | >1100 | <3 | 21 | <3 | 21 |
Vibrio vulnificus population abundance among water and seaweed samples at different seasons of the year (sample number, n=24 for each source in each season)
| Category | Summer | Autumn | Winter | Spring | ||||
| Water | Seaweeds | Water | Seaweeds | Water | Seaweeds | Water | Seaweeds | |
| Abundance (MPN 100 mL−1 water or 10 g−1 seaweed) | ||||||||
| >1100 | 3 | 7 | 3 | 6 | 0 | 0 | 0 | 0 |
| 251–1100 | 7 | 9 | 9 | 11 | 0 | 0 | 0 | 0 |
| 151–250 | 4 | 3 | 6 | 2 | 0 | 0 | 0 | 0 |
| 51–150 | 6 | 5 | 4 | 5 | 0 | 1 | 2 | 5 |
| 21–50 | 2 | 0 | 0 | 0 | 0 | 9 | 5 | 11 |
| 3–20 | 2 | 0 | 2 | 0 | 2 | 1 | 4 | 0 |
| <3 | 0 | 0 | 0 | 0 | 22 | 13 | 13 | 8 |
| Median | 240 | 460 | 290 | 460 | <3 | 3 | <3 | 21 |
| Mode | 460, >1100 | >1100 | 460 | >1100 | <3 | 21 | <3 | 21 |
| Category | Summer | Autumn | Winter | Spring | ||||
| Water | Seaweeds | Water | Seaweeds | Water | Seaweeds | Water | Seaweeds | |
| Abundance (MPN 100 mL−1 water or 10 g−1 seaweed) | ||||||||
| >1100 | 3 | 7 | 3 | 6 | 0 | 0 | 0 | 0 |
| 251–1100 | 7 | 9 | 9 | 11 | 0 | 0 | 0 | 0 |
| 151–250 | 4 | 3 | 6 | 2 | 0 | 0 | 0 | 0 |
| 51–150 | 6 | 5 | 4 | 5 | 0 | 1 | 2 | 5 |
| 21–50 | 2 | 0 | 0 | 0 | 0 | 9 | 5 | 11 |
| 3–20 | 2 | 0 | 2 | 0 | 2 | 1 | 4 | 0 |
| <3 | 0 | 0 | 0 | 0 | 22 | 13 | 13 | 8 |
| Median | 240 | 460 | 290 | 460 | <3 | 3 | <3 | 21 |
| Mode | 460, >1100 | >1100 | 460 | >1100 | <3 | 21 | <3 | 21 |
Among seaweeds, the high abundance of V. vulnificus during summer and autumn (mode values >1100 MPN) as well as preservation of a bulk of the bacterium's cultivable population during the winter months indicate that the influence of physico-chemical stress is probably less for the seaweed-associated strains. Despite the washing step adopted in our methodology, the persistence of a large number of cultivable V. vulnificus in seaweed samples indicates its intimate association, which can be facilitated by the bacterium's strong invasive capability to penetrate the epithelial membrane and an outer membrane protein (ompU) has been implicated recently as an important virulence factor (Goo et al., 2006). Seaweed derivatives like galactans of agars and carrageenans are common foods for many bacterial species (Michel et al., 2006). Similar to other vibrios, V. vulnificus may also utilize seaweed extracts including l-fucose, d-mannose, d-galactose, and d-glucuronic acid as nutrients for its survival and growth (Sarker et al., 1994; Sakai et al., 2002). During high tide or in response to environmental stress seaweeds generally release a huge amount of slimy materials from their body surface containing sugars and amino acids, and these dissolved organic materials are very attractive food for bacteria. Vibrio vulnificus contains lecithinase, lipase, caseinolytic protease, chitinase, mucinase and elastase enzymes, which can play a vital role in the degradation of seaweed-derived materials (Somerville & Colwell, 1993; Moreno & Landgraf, 1998). Seaweeds are also known to have extensive spawning events, often turning the sea ‘milky’. Taitet al.(2005) have shown that the seaweed zoospore, produced during spring and early summer, can be induced to interact with the bacterial N-acylhomoserine lactone (AHL) quorum-sensing signal molecules, influencing their attachment (including Vibrio species) on the seaweed surfaces. The dispersion of seaweeds along the coast can thus play an important role in the spread of seaweed-associated V. vulnificus populations into new areas.
Our selective enrichment technique, followed by species-specific detection via colony hybridization and confirmatory PCR, yielded a total of 280 strains, out of which 46, 42, 2 and 11 strains were from water and 78, 74, 11 and 16 strains were from seaweed samples during summer, autumn, winter and spring, respectively. There was no significant deviation in the isolation rate of V. vulnificus among the three selected areas. Among the 280 isolated strains, 64% and 36% were from seaweed and seawater sources, respectively. In another coastal area of Japan, V. vulnificus was isolated from eight (17.2%) of 46 fish samples and 68 (43.8%) of 156 shellfish (Fukushima, 2006). In comparison with V. vulnificus abundance among oysters collected from the Chesapeake and Apalachicola Bay at levels of c. 103 g−1 during summer (Jackson et al., 1997; Motes et al., 1998), the bacterium had similar abundance among seaweed samples of the Kii Channel. A comparatively higher abundance of V. vulnificus has been observed in the constricted area of the inland sea, which is eutrophic as a result of riverine influence, than other coastal areas without fluvial input (Venkateswaran et al., 1989). The Kii Channel also receives nutrient inputs from several small rivers that may favor V. vulnificus persistence in this area.
Role of physicochemical parameters
Water temperature ranged between 20 and 29 °C during summer and autumn while during winter and spring seasons the temperatures ranged between 10 and 18 °C. A higher level of salinity (24–30 p.p.t.) was observed during summer and autumn in comparison with winter and spring (14–25 p.p.t.). The water pH ranged between 7.7 and 8.9, having comparatively higher values during spring and winter (data not shown). Trends in the changes of V. vulnificus population in different areas of the selected sites in relation to variations in water pH, temperature and salinity were nonlinear (Fig. 3).
Vibrio vulnificus population changes in relation to variation in water pH, temperature and salinity. Nonlinear lines representing the trends in bacterial abundance were calculated using median smoothing with 20% Lowess span, 20% Trewess span and 10% Trewess trim.
Vibrio vulnificus abundance showed a drastic increase in both seaweed and water samples when the temperature increased above 20 °C and in contrast, their number decreased drastically during winter and spring. Our results are in concordance with other studies where exclusive isolations and higher abundances of V. vulnificus have been observed during summer months (Wright et al., 1996; Motes et al., 1998; Randa et al., 2004). In the Chesapeake Bay and the Gulf coast of the United States, the V. vulnificus abundances in oysters were positively correlated with water temperature until reaching 26 °C while above this temperature its influence was nominal (Wright et al., 1996; Motes et al., 1998). We have also observed a drastic increase in V. vulnificus numbers in both seawater and seaweed samples within a temperature range between 20 and 24 °C, while a declining trend in the bacterial abundance among seaweeds was prominent at a higher temperature (>25 °C). Owing to the stress arising from a lower temperature, most V. vulnificus strains may enter into a dormant viable but nonculturable (VBNC) state to withstand adverse conditions while under the more favorable summer conditions, some VBNC bacteria may be resuscitated to become cultivable (Oliver et al., 1995; Randa et al., 2004). Besides, both free-living and seaweed associated V. vulnificus populations may be limited by the double-stranded DNA phages (e.g. Podoviridae, Styloviridae and Myoviridae), which are present in abundant numbers in the coastal environment (DePaola et al., 1998).
Both the counts of seaweed-associated and free-living V. vulnificus populations in water were higher when the water pH was 7.75–8.15. This is in concordance with the bacterium's growth in laboratory media (Hsu et al., 1998). The highest number of V. vulnificus was observed at pH 7.9 for the seaweed-associated population and at pH 8.15 for the free-living population in the water column. Interestingly, at a higher pH (>8.20), the counts decreased drastically. But the water pH was higher (>8.20) during cold seasons when a lower abundance of cultivable V. vulnificus could be related to the influence of a lower temperature. Otherwise, at a higher pH level, cultivable V. vulnificus may transform into the VBNC stage, the activities of its phages may increase or other bacteria may out-compete its cultivable population.
The V. vulnificus population gradually increased at a higher salinity level (>20 p.p.t.) in both seaweed and water samples, and the maximum abundance was observed at 24–28 p.p.t. during warmer months. However, in the subtropical Kii Channel, the increase in salinity occurred during warmer seasons when a higher temperature might have also influenced V. vulnificus abundance. Previous studies could not establish a clear-cut influence of water salinity on V. vulnificus occurrence (Oliver et al., 1995; Wright et al., 1996; Høi et al., 1998). According to Moteset al.(1998), changes in salinity could only explain about 10% of the total variability in the cultivable bacteria, a positive correlation existed in the lower salinity range (<15 p.p.t.) but a negative correlation appeared with higher salinity (>25 p.p.t.). The lack of consensus among the various studies may be due to the greater influence of temperature, which had confounding effects (Kaspar & Tamplin, 1993). However, a positive correlation has been observed with salinity and V. vulnificus occurrence in the subtropical coastal water of Barnegat Bay, NJ, and 75% of the strains were isolated between 20 and 28 p.p.t. salinity (Randa et al., 2004).
Genetic diversity and relatedness as revealed from ribotyping
The ribotypes obtained in this study exhibited stable and reproducible patterns, showing 10–18 bands over a size range of 0.8–9.0 kb. Among the selected 160 isolates, 18 different patterns (dissimilarities <10%) and nine different types (dissimilarities >10%) could be differentiated by spss analysis where four types (A, D, E and H) were dominant (Fig. 4). Seaweed strains had a higher genotypic diversity than seawater strains, with a total of 18 and 10 different patterns among them, respectively. A high number of ribotype patterns were observed among the summer and autumn strains. However, the number of ribotype patterns were 16, 14, 4 and 6 among isolates obtained during summer, autumn, winter and spring, respectively. Among both seawater and seaweed strains, one pattern (D1) was dominant during winter and spring while a couple of patterns (C and H1) dominated during summer and autumn. Some patterns occurred only among seaweed strains throughout the year (e.g. A4) or during certain periods (e.g. A1, A2, B1, G2 and H2 in the summer months, and D2 in the winter months), indicating the affinity of some strains to seaweeds. The persistence of some ribotypes throughout the year might be due to their greater resistance power to withstand unfavorable environmental conditions or because of their large abundance. The higher isolation rate of V. vulnificus, along with its higher genetic diversity among seaweeds than water samples, indicates that seaweeds can act as an important niche for the bacterium in the coastal environment.
Vibrio vulnificus ribotype (RT) diversity and their frequency among the seaweed and seawater strains based on the squared Euclidean distance and average linkage clustering method. Abbreviated Sum, Aut, Wint and Spr represent summer, autumn, winter and spring, respectively, beneath which total number of RT patterns in respective seasons (within parentheses) are shown. The corresponding frequencies of the RT patterns are shown in adjacent rows and in the last row the total number of representative strains examined for each source in each season is cited.
Vibrio vulnificus in coastal seaweed and public health
In the Gulf Coast states of the United States, most cases (85%) of V. vulnificus-related primary septicemia occurred in summer months with high fatality rates (Kaspar & Tamplin, 1993; Levine et al., 1993; Shapiro et al., 1998). Human infection due to V. vulnificus drastically increases during summer, which coincides with a higher abundance of the bacterium in shellfish or due to fishing and swimming in estuarine areas (Motes et al., 1998; Ralph & Currie, 2007). Humans can also contract disease from sea foods prepared from seaweeds infected with pathogenic V. vulnificus strains. Moreover, in summer months, before human consumption, V. vulnificus in the harvested stock can grow rapidly during the storage period without refrigeration (Cook, 1997). Immediate chilling after the harvest and storage at a low temperature (0–4 °C) can lower its cultivable population to reduce its threat (Cook & Ruple, 1992). According to previous findings, the V. vulnificus biotype 1 strains, capable of infecting humans, is overwhelmingly dominant (c. 99%) among the total V. vulnificus population in coastal and marine environments (Hor et al., 1995; Høi et al., 1998). However, a latest study has revealed that among coastal biotype 1 strains the virulence-correlated gene responsible for human infection (vcgC) comprises a minor fraction of the total population (c. 15%) but increases its percentage as the water temperature increases (Warner & Oliver, 2008). Serotyping of V. vulnificus strains isolated from the Japanese coasts has revealed the existence of 11 and 18 serotypes in two different studies, respectively, of which O4 and O7 serotypes were the most frequent, and a potential transmission of these pathogenic serotypes from the coastal environment to humans has been suggested (Oonaka et al., 2004; Fukushima, 2006).
In the recent decade, a key scientific focus that has attracted increasing interest is on the emergence and spread of infectious diseases that can be affected by environmental factors, particularly variations in climate, e.g. increase in sea surface temperature. Notably, bacterial pathogens like V. vulnificus have an enormous biological and genetic capacity to adapt with varying environmental conditions and to evolve new strategies to spread epidemics. Moreover, with the enhancement of world globalization interconnectedness, the likelihood of microbial disease emergence and spread is becoming higher. The changes in bacterial abundance among seaweeds may show similar trends in other coastal regions of the world having a similar climate and, expectantly, the bacteriological changes and the associated risks are not unpredictable. Therefore, understanding the relationship between V. vulnificus and climatic parameters (e.g. temperature, salinity, etc.) may help in forecasting the probable concentrations of the bacterium along coastal waters as well as among potential hosts like seaweeds. We can conclude that seaweeds can harbor a large population of V. vulnificus and aid the bacterium to withstand physicochemical stress along the coast.
Acknowledgements
This study was supported by the Ministry of Education, the Government of Japan and Yakult Ltd, Japan. The authors are grateful to Dr Sayed Mohiuddin Abdus Salam, University of Tokushima, Japan, and M.A. Yushuf Sharker, ICDDR, B for their voluntary support.
References
Author notes
Editor: Riks Laanbroek
