Population Relationship of Vibrio parahaemolyticus Isolates Derived from Aquaculture Ponds,a Seafood Market,Restaurants,and Clinical Samples

Abstract

To study the relationship between environmental and clinical populations of Vibrio parahaemolyticus, we collected in total 86 isolates from Southern China during one and a half years. Sixty-eight isolates were recovered from aquaculture ponds, a seafood market, and restaurants, and 18 isolates were recovered from clinical samples. Virulence gene analysis revealed that 25 isolates (14 clinical and 11 environmental) tested positive for tdh, but only 4 carried trh. Interestingly, none of the tdh ⁺ environmental isolates was recovered from ponds. Both environmental and clinical tdh⁺ isolates, except for one clinical isolate, harbor type III secretion system 2α (T3SS2α) and T3SS2β-related genes, including vopB2α, which was previously suggested to be absent from environmental strains. More than 70% of clinical isolates carried the pandemic marker of new toxRS (GS-PCR⁺), which was not present in the environmental isolates. Pulsed-field gel electrophoresis and multilocus sequence typing analysis showed a high degree of genetic diversity within the environmental isolates. In contrast, the clinical population formed a tight cluster that differed from the environmental isolates. These findings suggest that the pandemic strains of V. parahaemolyticus may not directly originate from marine animals. Rather the environments where they are maintained could serve as reservoirs for toxigenic, but not pandemic strains. These environments provide an ideal place for generation of new toxigenic strains through DNA exchange, which was revealed by extensive recombination events in recA sequences of the environmental isolates.

Introduction

V ibrio parahaemolyticus is a halophilic, Gram-negative marine bacterium distributed widely all over the world (Bockemühl and Triemer, 1974; Nair et al., 1980). Most strains are strictly environmental, but some cause serious human gastroenteritis mainly due to consumption of raw seafood contaminated with pathogenic strains (Johnson et al., 1971; Daniels et al., 2000). It has been suggested that V. parahaemolyticus has outnumbered Salmonella as the most frequent agent to cause gastroenteritis in Southern China (Liu et al., 2004).

Virulence factors of V. parahaemolyticus have been extensively studied. These factors include various toxins and effectors of the type III secretion system (T3SS) (Nishibuchi and Kaper, 1995; Izutsu et al., 2008). Thermostable direct hemolysin (TDH) and TDH-related hemolysin are two well-characterized V. parahaemolyticus toxins. Strains that are tdh positive show a hemolytic activity on Wagatsuma agar, which is an important clinical marker to identify pathogenic strains (Honda and Iida, 1993). In previous epidemiological studies, tdh and trh were detected in 99% of clinical strains, but only in 2–3% of environmental strains (Nishibuchi and Kaper, 1995; Roque et al., 2009). T3SSs use a needle-like structure to deliver bacterial effectors directly into host cells (Ham and Orth, 2012). V. parahaemolyticus possesses two T3SS systems that are located on the chromosome 1 and 2, respectively (Makino et al., 2003). T3SS1 has high homologies with that in other vibrio species. T3SS2 is only present in pathogenic strains of Vibrio cholerae and V. parahaemolyticus (Dziejman et al., 2005). Two distinct lineages of T3SS2, T3SS2α, and T3SS2β have been described for V. parahaemolyticus. Presence of tdh correlates with T3SS2α, whereas presence of trh correlates with T3SS2β (Park et al., 2004; Okada et al., 2009).

In China and South East Asia, people prefer to consume fresh instead of frozen seafood. Consequently, aquatic animals are maintained alive shortly before consumption. Clinical strains are consequently considered to have been transmitted through marine animals (Chao et al., 2011). However, there are no data available that describe the relationship among strains recovered from the aquaculture ponds, temporary holding environments, and strains from clinical samples. In this study, we surveyed and isolated V. parahaemolyticus from different consumer food chain environments associated with Southern China, including aquaculture ponds, a seafood market, restaurants, and clinics during a time period of one and a half years. Virulence traits, genetic diversity, and population structure of the isolates were analyzed by multiple approaches. Our results may shed light on the origin and transmission of pathogenic strains of V. parahaemolyticus from marine environment to humans.

Materials and Methods

Sample collection

Samples were collected from three aquaculture ponds, a seafood wholesale market, two restaurants, and four hospitals located in Hainan and Guangdong Provinces during 2010–2012 and are recorded in Supplementary Table S1 (Supplementary Data are available online at www.liebertpub.com/fpd). Aquaculture ponds are located in Hainan province. Restaurants, fish market, and hospitals are located in Guangzhou city. For environmental samples, water samples were taken biweekly and concentrated by filtering through a 0.22 μm Millipore nitrocellulose membrane. For clinical samples, stool samples of the patients were collected. A loop of sample was inoculated in 100 μL Tryptic Soy Broth (TSB; BD Diagnostics) plus 2% NaCl. The inocula were incubated at 30°C for 12–18 h and one loop of the culture was streaked on Tryptic Soy Agar (TSA; Oxoid) supplemented with 2% NaCl. Single colonies were cultured in TSB plus 2% NaCl and stored at −80°C.

Identification of bacterial strains

Bacteria isolated from the samples were subcultured on thiosulfate citrate bile salt sucrose agar (BD Diagnostics). Single green colonies were picked and cultured on TSA with addition of 2% NaCl. Bacterial DNA was obtained with an Axyprep Bacterial Genomic DNA Miniprep Kit (Axygen Scientific) and used as template for amplification of 16S rDNA genes and gyrB. Nucleotide sequences of both genes were sequenced and then run through a megablast search to identify the highly similar sequences (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Only if the BLAST result for both genes matched V. parahaemolyticus, these isolates were further confirmed by biochemical analysis using Biolog GN-3.

Detection of virulence genetic markers

Fifteen sets of oligonucleotide primers (Table 1) were designed to detect the presence of virulence genes as well as the pandemic marker (toxRS). Polymerase chain reaction (PCR) cycles were run as described in Table 1. Amplified products were separated by electrophoresis in 1% agarose gel. DNA from E379 (ATCC33843) and E154 (ATCC17802) that are positive for all the virulence markers served as a positive control and DNA from Escherichia coli DH5α served as negative control. To avoid false-positive reaction, products were purified and sequenced to confirm identity of the respective gene.

Table 1.

Primers and Polymerase Chain Reaction Programs for Detection of the Target Genes

Target gene	Primer/probe sequence(5′-3′)	Amplification procedure	Length (bp)	Reference
tdh	F1: GTAAAGGTCTCTGACTTTTGGAC	95°C 5 min; 95°C 45 s, 55°C 45 s, 72°C 30 s for 35 cycles; 72°C 8 min	270	Bej et al. (1999)
	R1: TGGAATAGAACCTTCATCTTCACC
trh	F2: CATAACAAACATATGCCCATTTCCG	95°C 5 min; 95°C 45 s, 54°C 45 s, 72°C 45 s for 35 cycles; 72°C 8 min	500	Bej et al. (1999)
	R2: CATAACAAACATATGCCCATTTCCG
tlh	F3: AAAGCGGATTATGCAGAAGCACTG	95°C 5 min; 95°C 45 s, 56°C 45 s, 72°C 45 s for 35 cycles; 72°C 8 min	450	Bej et al. (1999)
	R3: GCTACTTTCTAGCATTTTCTCTGC
vopQ	F4: AGGCTACCGAGGAGAACC	95°C 5 min; 95°C 45 s, 54°C 45 s, 72°C 45 s for 35 cycles; 72°C 8 min	427	This study
	R4: CAACATCGCGTGACCAAT
vopR	F5: AACCAGCAAAGCCTCATC	95°C 5 min; 95°C 45 s, 54°C 45 s, 72°C 45 s for 35 cycles; 72°C 8 min	448	This study
	R5: GCCTTCTTGTCTCCCTACC
vopS	F6: TTTGACAACGAACAGGGAC	95°C 5 min; 95°C 45 s, 53°C 45 s, 72°C 30 s for 35 cycles; 72°C 8 min	296	This study
	R6: ACGCCACCACCGTAATCT
VPA 0450	F7: TTGCTGAAGGCTCTGATG	95°C 5 min; 95°C 45 s, 53°C 45 s, 72°C 30 s for 35 cycles; 72°C 8 min	275	This study
	R7: CTGCACTGGCTTATGGTC
vopC	F8: AGCGTGGTGGTTAGTGAA	95°C 5 min; 95°C 45 s, 52°C 45 s, 72°C 30 s for 35 cycles; 72°C 8 min	332	This study
	R8: AACTGTCCAAGAGGCGTA
vopT	F9: ACCTCACGAGACCCAGAA	95°C 5 min; 95°C 45 s, 53°C 45 s, 72°C 30 s for 35 cycles; 72°C 8 min	250	This study
	R9: CAAGTAAGGCAGGCACAT
vopA	F10: AGAGGATACAACCGCCAGAT	95°C 5 min; 95°C 45 s, 53°C 45 s, 72°C 30 s for 35 cycles; 72°C 8 min	358	This study
	R10: AACGGAAACGAGCAAACC
vopL	F11: CACCGCCACTACCATCAG	95°C 5 min; 95°C 45 s, 55°C 45 s, 72°C 45 s for 35 cycles; 72°C 8 min	459	This study
	R11: CGTCAATCACAGAACCTCC
vopB2	F12: TCGCTGCATTCATGGTTTTA	95°C 5 min; 95°C 45 s, 60°C 45 s, 72°C 45 s for 35 cycles; 72°C 8 min	217	This study
	R12: CTGCTGGTGTTGACGAGAAA
vopD2	F13: TCTGCAAGCACGATAGCAAG	95°C 5 min; 95°C 45 s, 60°C 45 s, 72°C 45 s for 35 cycles; 72°C 8 min	197	This study
	R13: CGTCGTTAGAAAAAGCGAGC
VPA1362	F: CTGCAGGTATCGCATCTTCA	95°C 5 min; 95°C 45 s, 60°C 40 s, 72°C 45 s for 33 cycles; 72°C 3 min	218	Noriea et al. (2010)
	R: TTAGAACCAACCGACGAAGC
GS-VP	F: TAATGAGGTAGAAACA	96°C 5 min; 96°C 1 min, 45°C 2 min, 72°C 3 min for 25 cycles; 72°C 7 min	651	Chowdhury et al. (2000)
	R: ACGTAACGGGCCTACA
gyrB	F: ACGGCATCTCGGTAGAAGTG	95°C 5 min; 95°C 45 s, 59°C 45 s, 72°C 45 s for 12 cycles; 95°C 45 s, 57°C 45 s, 72°C 45 s for 12 cycles; 95°C 45 s, 55°C 45 s, 72°C 45 s for 12 cycles; 72°C 8 min	406	This study
	R: AGAGTGCCGGATCTTTTTCCT
recA	F: GAAGTCATCATCACCGTTCTGCGTGARAARCARTTYGGTAAAGG	95°C 5 min, 3 cycles of 45 s at 95°C, 2 min at 55°C, and 1 min at 72°C, 30 cycles of 20 s at 95°C, 1 min at 55°C, 1 min at 72°C, and a final extension of 7 min at 72°C	776	Thompson et al. (2005)
	R: AGCAGGGTACGGATGTGCGAGCTCRCCNTTRTAGCTRTACC
toxR	F: AGTAATTCGCTCGCTGACCA	95°C 5 min; 95°C 45 s, 59°C 45 s, 72°C 45 s for 12 cycles; 95°C 45 s, 57°C 45 s, 72°C 45 s for 12 cycles; 95°C 45 s, 55°C 45 s, 72°C 45 s for 12 cycles; 72°C 8 min	615	This study
	R: AGCTGGTTATTTTGTCCGCC

Molecular typing of V. parahaemolyticus by pulsed-field gel electrophoresis

Pulsed-field gel electrophoresis (PFGE) analysis was performed according to the standard protocol of PulseNet (www.pulsenetinternational.org) using NotI-restricted genomic DNA. Fragments were separated in a 1% agarose gel in 0.5% Tris–borate–EDTA buffer using a CHEFMapper XA system (Bio-Rad Laboratories). Lambda Ladder PFG marker (BioLabs) was used as molecular size marker. The gel images are compared by the software BioNumerics. The phylogenetic trees based on band profiles were created with the MVSP v3.1 software (Kovach, 1999).

Multilocus sequence typing analysis

Multiple sequence alignments (MSAs) for each locus (gyrB, recA, toxR) were aligned and trimmed by Clustal W in MEGA 5 (Tamura et al., 2011). The in-frame sequences at the three loci were concatenated to produce one sequence. A phylogenetic tree was constructed from the allelic profile with the unweighted-pair group method (UPGMA) using average linkages. The tree was subsequently created in MEGA5. Numbers of polymorphic sites, the parameters Pi and dN/dS, as well as the haplotypes were determined for individual loci and concatenated sequences with DnaSP 5 (Librado and Rozas, 2009). Recombination events were detected using the Recombination Detection Program version 3.0 (RDP3) (Martin et al., 2005). The structural genetic relationship between each population was determined with Arlequin3 following the instructions described by Excoffier et al. (2007).

Results

Purification and detection of V. parahaemolyticus

Over 900 isolates were collected during one and a half years. The detection rates of V. parahaemolyticus differed depending on the sampling site. From ponds, the market, and restaurants, 3.4%, 4.41%, and 12.34% of all isolates were identified as V. parahaemolyticus. Eighty-six isolates and two reference strains were included in the genetic analysis. They were assigned to four groups based on their origin: aquaculture ponds (12.5%; 11 isolates), seafood market (20.5%; 18 isolates), restaurants (44.3%; 39 isolates), and clinics (20.5%; 18 isolates).

Virulence traits of the V. parahaemolyticus isolates

The isolates were screened for 15 virulence markers, including genes encoding for hemolysins, T3SS1, T3SS2α, and T3SS2β and the pandemic marker of group-specific PCR (GS-PCR). The major virulence gene tdh was detected in 14 clinical strains (77.8%) and 11 environmental isolates (16.2%), of which 6 came from the market (33.3%), 5 came from the restaurants (13.2%), but none from the ponds. The gene tlh was present in all isolates, except one from the ponds. The trh gene was present in only four isolates (4.5%), of which two were from the environment and the other two from clinical samples. Three of these trh-positive isolates also carried tdh (Table 2).

Table 2.

Distribution of 15 Virulence-Associated Genes in 88 Vibrio parahaemolyticus Isolates

	No. of clinical isolates (n = 18)				No. of isolates from pond (n = 11)				No. of isolates from seafood market (n = 18)				No. of isolates from restaurant (n = 39)
Gene	tdh⁺ trh⁺ (n = 2)	tdh⁺ trh⁻ (n = 12)	tdh⁻ trh⁺ (n = 0)	tdh⁻ trh⁻ (n = 4)	tdh⁺ trh⁺ (n = 0)	tdh⁺ trh⁻ (n = 0)	tdh⁻ trh⁺ (n = 0)	tdh⁻ trh⁻ (n = 11)	tdh⁺ trh⁺ (n = 0)	tdh⁺ trh⁻ (n = 6)	tdh⁻trh⁺ (n = 0)	tdh⁻trh ⁻ (n = 12)	tdh⁺ trh⁺ (n = 1)	tdh⁺trh⁻ (n = 4)	tdh⁻trh⁺ (n = 1)	tdh⁻trh⁻ (n = 33)
T3SS1
All genes	2	12	0	4	0	0	0	6	0	6	0	7	1	4	1	30
vpA0450	2	12	0	4	0	0	0	7	0	6	0	7	1	4	1	33
vopS	2	12	0	4	0	0	0	11	0	6	0	11	1	4	1	32
vopR	2	12	0	4	0	0	0	11	0	6	0	7	1	4	1	31
vopQ	2	12	0	4	0	0	0	10	0	6	0	7	1	4	1	32
T3SS2α
All genes	0	12	0	1	0	0	0	0	0	6	0	0	1	3	0	0
vp1362	1	12	0	4	0	0	0	1	0	6	0	0	1	3	0	0
vopL	1	12	0	3	0	0	0	1	0	6	0	0	1	3	0	0
vopA	0	12	0	2	0	0	0	0	0	6	0	0	1	3	0	0
vopT	1	12	0	2	0	0	0	0	0	6	0	0	1	3	0	0
vopC	1	12	0	3	0	0	0	0	0	6	0	0	1	3	0	0
T3SS2β
All genes	2	10	0	4	0	0	0	8	0	2	0	11	1	3	0	23
vopB2	2	10	0	4	0	0	0	8	0	2	0	11	1	3	0	23
vopD2	2	11	0	4	0	0	0	10	0	6	0	12	1	3	1	32
GS-PCR	0	12	0	1	0	0	0	0	0	0	0	0	0	0	0	0

GS-PCR, group-specific polymerase chain reaction; T3SS, type III secretion system.

To detect presence of T3SS1, four genes, vpa0450, vopS, vopR, and vopQ were used to screen isolates. All isolates, except one from the market possessed at least one of four genes (Table 2). 89.5% of the isolates harbored vpa0450, 97.7% harbored vopS, 93.0% harbored vopR, and 91.9% harbored vopQ. To detect presence of T3SS2α five genes, vopB2α (VP1362), vopL, vopA, vopT, and vopC were used to screen 86 isolates. Of the 68 environmental isolates, the tdh⁺/trh⁻ isolates tested positive for all five genes, including vopB2α (VP1362), which could not be detected in previous studies (Jones et al., 2012). Two tdh⁻/trh⁻ environmental isolates (DDC05 and HS43-2) carried part of the T3SS2α genes. In the remaining environmental isolates, none of the T3SS2α genes were present. Of the clinical samples, all tdh⁺/trh⁻ isolates contained the five genes, whereas all of the tdh⁻/trh⁻ isolates contained at least two of these genes, including vopB2α. One tdh⁺ /trh⁺ isolate did not harbor any of T3SS2α gene. To detect presence of T3SS2β genes, primers for vopB2β and vopD2β were employed. 74.4% carried vopB2β and 97.6% carried vopD2β, of which only four isolates contained trh. To further screen the isolates, a GS-PCR was used, which targets a new toxRS region that is associated with pandemic strains (tdh⁺ /trh⁻ ). None of the environmental isolates were positive for this marker. However, in 13 of the 18 clinical isolates, this region was detected. Of these 13 isolates, 12 were tdh⁺ / trh⁻ and 1 isolate ESS12070 was tdh⁻/trh⁻ , but had previously been tested positive for four genes of T3SS2α (Table 2).

Molecular typing of V. parahaemolyticus by PFGE

PFGE analysis showed a high level of diversity among 86 isolates. Sixty-eight distinct patterns were distinguished according to the criteria suggested by Tenover et al. (1995), where two band profiles are considered to be the same pattern if their difference is not greater than three bands (Fig. 1 and Supplementary Table S1). Pattern P29 was the dominant profile and contained seven isolates, of which all were isolated from ponds and the market (Supplementary Table S1). The tdh⁺ strains were characterized by PFGE patterns (Fig. 2). Five groups were distinguished at a similarity cutoff level of 80%. Group I with three isolates was recovered from the same patient in 2011. However, the isolates were not identical concerning virulence markers (Table 3). Group II with four isolates originated from four different patients. Group III contained three isolates, which were either recovered from the market or restaurants. Group IV contained four strains that had been isolated from four different patients. Group V encompasses four isolates that originated from the seafood market and were sampled on the same day. Interestingly, except group I, the isolates within each group had identical virulence markers (Table 3).

FIG. 1.

PFGE patterns of NotI-digested genomic DNA of Vibrio parahaemolyticus isolates. Isolates were classified into 68 distinct patterns. Two profiles were considered to be the same if the difference did not exceed three bands. PFGE, pulsed-field gel electrophoresis.

FIG. 2.

PFGE pattern of the tdh+ isolates clustered with the UPGMA method. Five groups were distinguished at a similarity cutoff level of 80%. Group I, II, and IV were composed of clinical isolates, but group III and V were composed of environmental isolates. PFGE, pulsed-field gel electrophoresis; UPGMA, unweighted-pair group method.

Table 3.

Molecular Traits of the tdh-Positive Isolates of Vibrio parahaemolyticus

	T3SS1								T3SS2α				T3SS2β
Strain	tlh	tdh	trh	VPA0450	vopS	vopR	vopQ	VPA1362	vopL	vopA	vopT	vopC	vopB2	vopD2	GS-PCR	PFGE patterns	Source
HS51-2	+	+	−	+	+	+	+	+	+	+	+	+	−	+	−	P31	E
HS51-4	+	+	−	+	+	+	+	+	+	+	+	+	−	+	−	P67	E
HS51-5	+	+	−	+	+	+	+	+	+	+	+	+	+	+	−	P62	E
HS53-2	+	+	−	+	+	+	+	+	+	+	+	+	−	+	−	P67	E
HS54-3	+	+	−	+	+	+	+	+	+	+	+	+	−	+	−	P11	E
HS54-4	+	+	−	+	+	+	+	+	+	+	+	+	+	+	−	P67	E
MJ02-3	+	+	−	+	+	+	+	+	+	+	+	+	+	+	−	P62	E
MJ03-3	+	+	−	+	+	+	+	+	+	+	+	+	+	+	−	P62	E
MJ03-4	+	+	−	+	+	+	+	+	+	+	+	+	+	+	−	P66	E
MJ39-4	+	+	−	+	+	+	+	−	−	−	−	−	−	−	−	P53	E
VP13150	+	+	+	+	+	+	+	+	+	+	+	+	+	+	−	P68	E
F09-157	+	+	+	+	+	+	+	−	−	−	−	−	+	+	−	P23	C
F11-30	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P61	C
F11-34	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P59	C
F11-36	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P58	C
F11-68	+	+	−	+	+	+	+	+	+	+	+	+	−	+	+	P55	C
F11-69	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P54	C
F11-70	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P55	C
F12-12	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P60	C
ESS12218	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P56	C
ESS12506	+	+	+	+	+	+	+	+	+	−	+	+	+	+	−	P38	C
ESS12663	+	+	−	+	+	+	+	+	+	+	+	+	−	−	+	P64	C
ESS12664	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P64	C
ESS12665	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P64	C
ESS12666	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	P63	C

+, positive amplification; −, negative amplification; E, environmental; C, clinical.

GS-PCR, group-specific polymerase chain reaction; PFGE, pulsed-field gel electrophoresis; T3SS, type III secretion system.

Multilocus sequence typing analysis of V. parahaemolyticus

MSAs for each locus (gyrB, recA, toxR) were aligned and trimmed by Clustal W in MEGA 5. A majority consensus phylogeny was constructed from concatenated loci (1794 bp). Twelve clades were identified in the dendrogram as shown in Figure 3. Clade 1 comprised nine clinical isolates that were all tdh⁺ , except for ESS12070. Interestingly, these strains were characterized as pandemic strains that carry the GS-PCR marker (Supplementary Table S1). Clade 7 consists of nine isolates, of which all are tdh⁺ , but had been isolated from the environment. Compared to clade 1, they lacked the GS-PCR marker. The rest of clades contained only environmental strains. None of the clades contained both environmental and clinical isolates.

FIG. 3.

Phylogenetic tree derived from concatenated sequences of gyrB, recA, and toxR of 86 V. parahaemolyticus constructed by the UPGMA method. Clustering was performed at similarities of 80% and yielded 12 clusters. UPGMA, unweighted-pair group method.

The genetic diversity of isolates was analyzed with Dnsp5.0. As shown in Table 4, internal fragment sizes of gyrB, recA, and toxR are 406, 776, and 615 bp, respectively. Numbers of alleles are 33, 42, and 36. Polymorphic sites for each locus are 27, 257, and 42. Average nucleotide diversities (pi) are 0.0129, 0.0535, and 0.0096. The rates of nonsynonymous substitutions range from 0.0007 (gyrB) to 0.013 (recA) and the synonymous substitutions range from 0.027 (toxR) to 0.183 (recA). The ratio of nonsynonymous to synonymous mutations show that all loci were subjected to purifying selection since the dN/dS ratios are <1 for each locus.

Table 4.

Nucleotide and Allelic Diversity of Multilocus Sequence Typing Loci for 88 Vibrio parahaemolyticus Isolates

	Fragment length	Alleles	Polymorphic sites	p	dN	dS	dN/dS
gyrB	406	33	27	0.0129	0.00007	0.05386	0
recA	776	42	257	0.0535	0.01344	0.18123	0.07
toxR	615	36	42	0.0096	0.0032	0.02736	0.11
gyrB-recA-toxR	1794	61	326	0.0293	0.0069	0.09938	0.07

A total of 61 haplotypes were detected among the concatenated sequences of 86 strains (Supplementary Table S2). With respect to epidemiological sources, 9 haplotypes were found in the pond (n = 11), 13 in the market (n = 18), 28 in the restaurant (n = 39), and 10 in the clinical samples (n = 18). Hap_15 encompassed environmental tdh ⁺ isolates that are found in Clade 7. Hap_54 encompassed clinical tdh ⁺ isolates that are found in Clade 1. The genetic structures of the four populations were compared by Alequein3 based on the haplotype frequencies. The results showed that the V. parahaemolyticus population associated with the aquaculture ponds was not genetically related to the populations associated with other sites. The same was true for the clinical isolates (p < 0.05), but the genetic differences between the isolates associated with the market and restaurant populations were not significant (p = 0.075) (Table 5). These results suggest that these food chain sites are sites where acquisition of new virulence factor genes may be occurring.

Table 5.

The Population Genetic Differences of Vibrio parahaemolyticus Among Each Pair of Four Populations Based on Haplotype Frequencies

	Pond	Market	Restaurant
Market	0.01086 ± 0.0012^a
Restaurant	0.02278 ± 0.0028^a	0.07477 ± 0.0149
Clinical	0.00137 ± 0.0011^a	0.0000 ± 0.0000^a	0.00017 ± 0.0002^a

Significant difference, p < 0.05.

The recombination events of the gyrB, recA, and toxR and their concatenated sequences were examined. We could not detect recombination events in gyrB and toxR. However, 78 and 142 recombination signals with 2 and 11 unique recombination events were found in recA and the concatenated sequences, respectively. Three highly divergent alleles, recA-QB2, recA-XCC04, and recA-HS31, were identified in eight environmental isolates that with nucleotide Blast best matched counterparts in Vibrio campbelli, Photobacterium damselae, and Shewanella sp., respectively. Likely, those isolates have undergone interspecies DNA exchange involving entire fragments of recA. In addition, various mosaic structures of recA that are caused by intragenic recombination were found in 44 isolates. Interestingly, in 52 isolates with recombinant recA, only three belonged to clinical group and none of them were pandemic isolates (Supplementary Fig. S1).

Discussion

In this study, 68 isolates were recovered from different environments. In the market and restaurant, V. parahaemolyticus was highly prevalent. 19.3% of these isolates were found to carry tdh and other virulence markers. These data suggest that holding conditions for aquatic animals at the market and restaurants are at high risk for V. parahaemolyticus contamination and for consumers. In our investigation, 16.2% of the environmental isolates carried tdh gene, which seems much higher than other reports (Nishibuchi and Kaper, 1995; DePaola et al., 2003). However, in this study, all of the tdh⁺ strains were isolated from the market and restaurants, but not from the ponds, which suggests that toxigenic strains occur at a low rate in aquaculture environments. As a matter of fact, previous studies have reported that occurrences of potential pathogenic strains from fish markets and restaurants range from 8% to 15% (Alam et al., 2009; Duan et al., 2011; Jun et al., 2012), whereas the range is 1–2.5% in the aquaculture environment (Alam et al., 2002; Cook et al., 2002; Wagley et al., 2009).

Interestingly, the distribution of the tdh and trh genes in the pathogenic strains seems to be related to the geographic regions. In our investigation the predominant pathogenic strains, clinical or environmental, were tdh⁺ and trh⁻ , which is in accordance with findings from previous studies in Southeast Asia and South America, where the pandemic strains (tdh ⁺/trh⁻ ) are more frequently distributed (Laohaprertthisan et al., 2003; García et al., 2009; Zhao et al., 2011). However, in Europe, where cases of infection by V. parahaemolyticus have rarely been reported, most of the pathogenic strains were tdh⁻ or trh ⁺ (Robert-Pillot et al., 2004; Ellingsen et al., 2008; Ottaviani et al., 2013). Most potential toxigenic isolates from the U.S. Pacific coast and Gulf of Mexico have both tdh and trh genes (DePaola et al., 2000; Paranjpye et al., 2012).

T3SS1 was present in almost all of our V. parahaemolyticus isolates, regardless of their origin. Presence of T3SS2α was generally associated with tdh gene, which is in agreement with many reports (Meador et al., 2007; Han et al., 2008). In contrast to the observation by Noriea et al. (2010) and Jones et al. (2012), we found vopB2α (T3SS2α) in both environmental and clinical tdh⁺ isolates. This divergence might be due to genetic differences between populations from two geographic areas or sequence variation. Nevertheless, it is clear that diversity of V. parahaemolyticus is still beyond present understanding and thus, it is controversial to utilize vopB2α as a specific indicator of strains virulence as suggested by Noriea et al. (2008) and Jones et al. (2012).

Although environmental and clinical tdh⁺ strains carried similar virulence markers, they exhibited great variation with respect to genetic profiles. Above 70% of the clinical isolates were typical pandemic strains carrying the marker of GS-PCR, while none of the environmental isolates contained this marker, which is in line with previous reports (Noriea et al., 2010; Yu et al., 2013). Moreover, the tdh⁺ strains were clustered into distinct environmental and clinical groups based on PFGE or multilocus sequence typing profiles. These differences indicate that pandemic strains may not arise directly from environmental strains. This hypothesis is supported by previous studies. Yan et al. (2011) suggested that V. parahaemolyticus has evolved as a typical epidemic population with a different origin than the nonclinical isolates. Theethakaew et al. (2013) found that clinical samples formed a tight cluster different from the environmental population, and thus suggested that disease-causing isolates were not a random sample of the environmental reservoir.

Extensive recombination events in recA nucleotide sequences were found by previous studies as well as in this study (Chowdhury et al., 2004; González-Escalona et al., 2008; Yu et al., 2011). In contrast to the observation by Theethakaew et al. (2013), the predominant strains with the recombinant recA alleles were recovered from the environment and not from clinical samples. This could indicate that the human intestinal tract is not, as previously hypothesized, a hot spot driving the emergence of new potentially pathogenic strains (Theethakaew et al., 2013). It is unclear what causes this difference, and further studies are required to explore the origin of pathogenic and pandemic strains of V. parahaemolyticus.

Conclusion

This study has revealed that population structures of V. parahaemolyticus from aquaculture ponds, sea food markets, and restaurants and clinical cases were significantly different. Even though tdh ⁺ isolates from the environment carry a wide range of virulence markers, they are not related to clinical isolates, which suggests that the pandemic strains may not originate from fresh seafood or associated environments. A high rate of recombination events in the recA gene of the environmental strains suggests that V. parahaemolyticus may frequently change its genetic traits in the environment by horizontal gene transfer or by DNA rearrangement.

Footnotes

Acknowledgments

This work was supported by the Programme of Knowledge Innovation of CAS (KZCX2-EW-Q215), the National Natural Science Foundation of China (31272697), 12th five-year-major-projects of China's Ministry of Public Health (2012zx10004-213), the Canton-CSA United Foundation (2009B1300157), the National Natural Science Foundation of China (41276163), and the Project of Science and Technology New Star of Zhujiang in Guangzhou city (2013J2200094), which are acknowledged.

Disclosure Statement

No competing financial interests exist.

References

Alam

, Tomochika

, Miyoshi

, Shinoda

. Environmental investigation of potentially pathogenic Vibrio parahaemolyticus in the Seto-Inland Sea, Japan. FEMS Microbiol Lett, 2002; 208:83–87.

Alam

, Chowdhury

, Bhuiyan

, Islam

, Hasan

, Nair

, Watanabe

, Siddique

, Huq

, Sack

, Akhter

, Grim

, Kam

, Luey

, Endtz

, Cravioto

, Colwell

. Serogroup, virulence, and genetic traits of Vibrio parahaemolyticus in the estuarine ecosystem of Bangladesh. Appl Environ Microbiol, 2009; 75:6268–6274.

Bej

, Patterson

, Brasher

, Vickery

, Jones

, Kaysner

. Detection of total and hemolysin-producing Vibrio parahaemolyticus in shellfish using multiplex PCR amplification of tlh, tdh and trh . J Microbiol Methods, 1999; 36:215–225.

Bockemühl

, Triemer

. Ecology and epidemiology of Vibrio parahaemolyticus on the coast of Togo. Bull World Health Organ, 1974; 5:353–360.

Chao

, Wang

, Zhou

, Jiao

, Huang

, Pan

, Zhou

, Qian

. Origin of Vibrio parahaemolyticus O3:K6 pandemic clone. Int J Food Microbiol, 2011; 145:459–463.

Chowdhury

, Chakraborty

, Ramamurthy

, Nishibu-Chi

, Yamasaki

, Takeda

, Nair

. Molecular evidence of clonal Vibrio parahaemolyticus pandemic strains. Emerg Infect Dis, 2000; 6:631–636.

Chowdhury

, Stine

, Morris

, Nair

. Assessment of evolution of pandemic Vibrio parahaemolyticus by multilocus sequence typing. J Clin Microbiol, 2004; 42:1280–1282.

Cook

, Bowers

, DePaola

. Density of total and pathogenic (tdh+) Vibrio parahaemolyticus in Atlantic and Gulf coast molluscan shell-fish at harvest. J Food Prot, 2002; 65:1873–1880.

Daniels

, MacKinnon

, Bishop

, Altekruse

, Ray

, Hammond

, Thompson

, Wilson

, Bean

, Griffin

. Vibrio parahaemolyticus infections in the United States, 1973–1998. J Infect Dis, 2000; 181:1661–1666.

10.

DePaola

, Kaysner

, Bowers

, Cook

. Environmental investigations of Vibrio parahaemolyticus in oysters after outbreaks in Washington, Texas, and New York (1997 and 1998). Appl Environ Microbiol, 2000; 66:4649–4654.

11.

DePaola

, Ulaszek

, Kaysner

, Tenge

, Nord-strom

, Wells

. Molecular, serological, and virulence characteristics of Vibrio parahaemolyticus isolated from environmental, food, and clinical sources in North America and Asia. Appl Environ Microbiol, 2003; 69:3999–4005.

12.

Duan

, Ju

, Liu

, Yu

, Wen

, Zeng

. Characterization of Vibrio parahaemolyticus obtained from environment and cases of foodborne disease in Shenzhen, during 2006–2008. Wei Sheng Wu Xue Bao, 2011; 51:518–523 (Chinese).

13.

Dziejman

, Serruto

, Tam

, Sturtevant

, Diraphat

, Faruque

, Rahman

, Heidelberg

, Decker

, Li

, Montgomery

, Grills

, Kucherlapati

, Mekalanos

. Genomic characterization of non-O1, non-O139 Vibrio cholerae reveals genes for a type III secretion system. Proc Natl Acad Sci U S A, 2005; 102:3465–3470.

14.

Ellingsen

, Jørgensen

, Wagley

, Monshaugen

, Rørvik

. Genetic diversity among Norwegian Vibrio parahaemolyticus . J Appl Microbiol, 2008; 105:2195–2202.

15.

Excoffier

, Laval

, Schneider

. Arlequin (version 3.0): An integrated software package for population genetics data analysis. Evol Bioinform Online, 2007; 1:47–50.

16.

García

, Torres

, Uribe

, Hernández

, Rioseco

, Romero

, Espejo

. Dynamics of clinical and environmental Vibrio parahaemolyticus strains during seafood-related summer diarrhea outbreaks in southern Chile. Appl Environ Microbiol, 2009; 75:7482–7487.

17.

González-Escalona

, Martinez-urtaza

, Romero

, Espejo

, Jaykus

, Depaola

. Determination of molecular phylogenetics of Vibrio parahaemolyticus strains by multilocus sequence typing. J Bacteriol, 2008; 190:2831–2840.

18.

Ham

, Orth

. The role of type III secretion system 2 in Vibrio parahaemolyticus pathogenicity. J Microbiol, 2012; 50:719–725.

19.

Han

, Wong

, Kan

, Guo

, Zeng

, Yin

, Liu

, Yang

, Zhou

. Genome plasticity of Vibrio parahaemolyticus: Microevolution of the ‘pandemic group’. BMC Genomics, 2008; 9:570.

20.

Honda

, Iida

. The pathogenicity of Vibrio parahaemolyticus and the role of thermostable direct haemolysin and related haemolysins. Rev Med Microbiol, 1993; 4:106–113.

21.

Izutsu

, Kurokawa

, Tashiro

, Kuhara

, Hayashi

. Comparative genomic analysis using microarray demonstrates a strong correlation between the presence of the 80-kilobase pathogenicity island and pathogenicity in Kanagawa phenomenon-positive Vibrio parahaemolyticus strains. Infect Immun, 2008; 76:1016–1023.

22.

Johnson

, Baross

, Liston

. Vibrio parahaemolyticus and its importance in seafood hygiene. J Am Vet Med Assoc, 1971; 159:1470–1473.

23.

Jones

, Lüdeke

, Bowers

, Garrett

, Fischer

, Parsons

, Bopp

, DePaola

. Biochemical, serological, and virulence characterization of clinical and oyster Vibrio parahaemolyticus isolates. J Clin Microbiol, 2012; 50:2343–2352.

24.

Jun

, Kim

, Choresca

Jr. , Shin

, Han

, Chai

, Park

SC.

Isolation, molecular characterization, and antibiotic susceptibility of Vibrio parahaemolyticus in Korean seafood. Foodborne Pathog Dis, 2012; 9:224–231.

25.

Kovach

WL.

MVSP—A Multivariate Statistical Package for Windows, ver 3.1. Pentraeth, Wales, UK: Kovach Computing Services, 1999.

26.

Laohaprertthisan

, Chowdhury

, Kongmuang

, Kalnauwakul

, Ishibashi

, Matsumoto

, Nishibuchi

. Prevalence and serodiversity of the pandemic clone among the clinical strains of Vibrio parahaemolyticus isolated in southern Thailand. Epidemiol Infect, 2003; 130:395–406.

27.

Librado

, Rozas

. DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics, 2009; 25:1451–1452.

28.

Liu

, Chen

, Wang

, Ji

. Foodborne disease outbreaks in China from 1992 to 2001—national foodborne disease surveillance system J. Hygiene Res, 2004; 33:725–727.

29.

Makino

, Oshima

, Kurokawa

, Yokoyama

, Uda

, Tagomori

, Iijima

, Najima

, Nakano

, Yamashita

. Genome sequence of Vibrio parahaemolyticus: A pathogenic mechanism distinct from that of V. cholerae . Lancet, 2003; 361:743–749.

30.

Martin

, Williamson

, Posada

. RDP2: Recombination detection and analysis from sequence alignments. Bioinformatics, 2005; 21:260–262.

31.

Meador

, Parsons

, Bopp

, Gerner-Smidt

, Painter

. Vora GJ. Virulence gene- and pandemic group-specific marker profiling of clinical Vibrio parahaemolyticus isolates. J Clin Microbiol, 2007; 45:1133–1139.

32.

Nair

, Abraham

, Natarajan

. Distribution of Vibrio parahaemolyticus in finfish harvested from Porto Novo (S. India) environs: A seasonal study. Can J Microbiol, 1980; 26:1264–1269.

33.

Nishibuchi

, Kaper

. Thermostable direct haemolysin gene of Vibrio parahaemolyticus: A virulence gene acquired by a marine bacterium. Infect Immun, 1995; 63:2093–2099.

34.

Noriea

III , Johnson

, Griffitt

, Grimes

. Distribution of type III secretion systems in Vibrio parahaemolyticus from the northern Gulf of Mexico. J Appl Microbiol, 2010; 109:953–962.

35.

Okada

, Iida

, Park

, Goto

, Yasunaga

, Hiyoshi

, Matsuda

, Kodama

. Identification and characterization of a novel type III secretion system in trh-positive Vibrio parahaemolyticus strain TH3996 reveal genetic lineage and diversity of pathogenic machinery beyond the species level. Infect Immun, 2009; 77:904–913.

36.

Ottaviani

, Leoni

, Rocchegiani

, Mioni

, Costa

, Virgilio

, Serracca

, Bove

, Canonico

, Di Cesare

, Masini

, Potenziani

, Caburlotto

, Ghidini

, Lleo

. An extensive investigation into the prevalence and the genetic and serological diversity of toxigenic Vibrio parahaemolyticus in Italian marine coastal waters. Environ Microbiol, 2013; 15:1377–1386.

37.

Paranjpye

, Hamel

, Stojanovski

, Liermann

. Genetic diversity of clinical and environmental Vibrio parahaemolyticus strains from the Pacific Northwest. Appl Environ Microbiol, 2012; 78:8631–8638.

38.

Park

, Ono

, Rokuda

, Jang

, Okada

, Iida

, Honda

. Functional characterization of two type III secretion systems of Vibrio parahaemolyticus . Infect Immun, 2004; 72:6659–6665.

39.

Robert-Pillot

, Guénolé

, Lesne

, Delesmont

, Fournier

, Quilici

. Occurrence of the tdh and trh genes in Vibrio parahaemolyticus isolates from waters and raw shellfish collected in two French coastal areas and from seafood imported into France. Int J Food Microbiol, 2004; 91:319–325.

40.

Roque

, Lopez-Joven

, Lacuesta

, Elandaloussi

, Wagley

, Furones

, Ruiz-Zarzuela

, de Blas

, Rangdale

, Gomez-Gil

. Detection and identification of tdh- and trh-positive Vibrio parahaemolyticus strains from four species of cultured bivalve molluscs on the Spanish Mediterranean Coast. Appl Environ Microbiol, 2009; 75:7574–7577.

41.

Tamura

, Peterson

, Stecher

, Nei

, Kumar

. MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol, 2011; 28:2731–2739.

42.

Tenover

, Arbeit

, Goering

, Mickelsen

, Murray

, Persing

, Swaminathan

. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: Criteria for bacterial strain typing. J Clin Microbiol, 1995; 33:2233–2239.

43.

Theethakaew

, Feil

, Castillo-Ramírez

, Aanensen

, Suthienkul

, Neil

, Davies

. Genetic relationships of Vibrio parahaemolyticus isolates from clinical, human carrier, and environmental sources in Thailand, determined by multilocus sequence analysis. Appl Environ Microbiol, 2013; 9:2358–2370.

44.

Thompson

, Gevers

, Thompson

, Dawyndt

, Naser

, Hoste

, Munn

, Swings

. Phylogeny and molecular identification of vibrios on the basis of multilocus sequence analysis. Appl Environ Microbiol, 2005; 71:5107–5115.

45.

Wagley

, Koofhethile

, Rangdale

. Prevalence and potential pathogenicity of Vibrio parahaemolyticus in Chinese mitten crabs (Eriocheir sinensis) harvested from the River Thames estuary. Engl J Food Prot, 2009; 72:60–66.

46.

Yan

, Cui

, Han

, Xiao

, Wong

, Tan

, Guo

, Liu

, Yang

, Zhou

. Extended MLST-based population genetics and phylogeny of Vibrio parahaemolyticus with high levels of recombination. Int J Food Microbiol, 2011; 145:106–112.

47.

, Jong

, Lin

, Tsai

, Tey

, Wong

. Prevalence of Vibrio parahaemolyticus in oyster and clam culturing environments in Taiwan. Int J Food Microbiol, 2013; 160:185–192.

48.

, Hu

, Wu

, Zhang

, Chen

, Wang

, Fang

. Vibrio parahaemolyticus isolates from southeastern Chinese coast are genetically diverse with circulation of clonal complex 3 strains since 2002. Foodborne Pathog Dis, 2011; 8:1169–1176.

49.

Zhao

, Zhou

, Cao

, Ma

, Jiang

. Distribution, serological and molecular characterization of Vibrio parahaemolyticus from shellfish in the eastern coast of China. Food Control, 2011; 22:1095–1100.

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.

0.00 MB

0.09 MB

0.03 MB

0.02 MB