Identification of Vibrio harveyi using PCR amplification of the toxR gene


X.-H. Zhang, Department of Marine Biology, Ocean University of China, 5 Yushan Road, Qingdao 266003, China. E-mail:


Aims:  The aim of this study was to develop an effective method for the identification of Vibrio harveyi based on using the toxR gene as a taxonomic marker.

Methods and Results:  Primers for the toxR gene were designed for specificity to V. harveyi, and incorporated in a polymerase chain reaction (PCR). The results of the PCR, which took <5 h from DNA extraction to amplification, revealed positive amplification of the toxR gene fragment in 20 V. harveyi isolates including type strains, whereas DNA from 23 other Vibrionaceae type strains and 13 Vibrio parahaemolyticus strains were negative. The detection limit of the PCR was 4.0 × 103 cells ml−1. In addition, the technique enabled the recognition of V. harveyi from diseased fish.

Conclusions:  The PCR was specific and sensitive, enabling the identification of V. harveyi within 5 h.

Significance and Impact of the Study:  The PCR allowed the rapid and sensitive detection of V. harveyi.


Vibrio harveyi is a Gram-negative, luminous, marine bacterium, which is ubiquitous in warm near-shore marine waters and sediment, the surfaces of marine animals, light-emitting organs of marine fish and cephalopods, and is also a common member of the natural intestinal microflora of aquatic vertebrates and invertebrates (e.g. O'Brien and Sizemore 1979; Orndorff and Colwell 1980; Ramesh and Venugopalan 1989). The organism is a major causal agent of luminous vibriosis, which affects a diverse range of marine invertebrates, especially penaeids in South America and Southeast Asia, resulting in severe economic losses (e.g. Sunaryanto and Mariam 1986; Liu et al. 1996; Vandenberghe et al. 1998). It is argued that the early diagnosis of V. harveyi infection could facilitate disease surveillance and prevention in cultured marine animals. However, difficulties have been encountered with the identification of isolates because of the phenotypic and genotypic diversity (Austin and Austin 1989).

Vibrios harbour a wealth of diverse genomes as revealed by different genomic techniques, including amplified fragment length polymorphism (AFLP), multilocus sequence typing, repetitive extragenic palindrome polymerase chain reaction (PCR), ribotyping and whole-genome analysis (Thompson et al. 2004). Several PCR methods have been developed to identify V. harveyi isolates by using the 16S rDNA sequence, the vhh haemolysin gene or the toxR gene (Conejero and Hedreyda 2003, 2004; Oakey et al. 2003). However, strains other than V. harveyi have been reported to give false-positive results (e.g. Oakey et al. 2003). Also, false-negative results have been recorded (Conejero and Hedreyda 2003).

The toxR gene was first discovered as a positive transcriptional regulator of the ctx gene encoding the cholera toxin of Vibrio cholerae (Miller and Mekalanos 1984). The gene was subsequently shown to encode a transmembrane protein that plays a pivotal role in the coordinate regulation of ctx and many other genes, including the tcp gene encoding toxin-coregulated pili and the ompU and ompT genes encoding major outer membrane proteins in this micro-organism (Miller et al. 1987; Miller and Mekalanos 1988; DiRita 1992; Crawford et al. 1998). Subsequently, the toxR gene was found to be present in many Vibrio species (Lin et al. 1993; Lee et al. 2000; Osoril and Klose 2000; Conejero and Hedreyda 2003). Comparison of the toxR amino acid sequences from several Vibrio spp. showed the presence of the highly divergent membrane tether region flanked by relatively conserved transcription activation and transmembrane domains. This gene is regarded as an effective taxonomic marker for identification of Vibrio spp. (Kim et al. 1999). Therefore, this study sought to develop a species-specific method to detect V. harveyi based on PCR amplification of the toxR gene.

Materials and methods

Bacterial strains

Twenty-six Vibrionaceae type strains, 18 V. harveyi isolates and 12 Vibrio parahaemolyticus isolates from a diverse range of hosts and geographical locations, were used (Table 1). Most of the cultures were obtained from the Vibrionaceae collection in the School of Life Sciences (Heriot-Watt University, Edinburgh). SF1 was obtained from the Department of Marine Biology (Ocean University of China, China). Authenticity was verified after the characteristics in Bergey's Manual of Systematic Bacteriology (Krieg and Holt 1984). The isolates were cultured on Zobell's 2216E medium (Difco) at 28°C, overnight, with subculturing every 7–14 days.

Table 1.   Bacterial strains
StrainName as receivedSourceCountry (year of isolation)
  1. LMG, Culture collection of the Laboratory voor Microbiologie, University of Gent, Belgium; T, type strain.

VIB 314Grimontia hollisaeLMG17719TUSA
VIB 289Photobacterium damselae ssp. damselaeLMG 7892TUSA
VIB 288Salinivibrio costicola ssp. costicolaLMG 6460TAustralia
VIB 72Vibrio anguillarumLMG 4437TNorway
VIB 281Vibrio aestuarianusLMG 7909TUSA
VIB 283Vibrio alginolyticusLMG 4408T 
VIB 284Vibrio alginolyticusLMG 4409TJapan
VIB 285Vibrio campbelliiLMG 11216TUSA
VIB 286Vibrio harveyiLMG 7890TUSA (1982)
VIB 287Vibrio cincinnatiensisLMG 7891TUSA
VIB 290Vibrio diazotrophicusLMG 7893TCanada
VIB 291Vibrio fischeriLMG 4414TUSA (1933)
VIB 292Vibrio fluvialisLMG 7894TBangladesh
VIB 293Vibrio furnissiiLMG 7910TJapan
VIB 294Vibrio gazogenesLMG 1354TUSA
VIB 295Vibrio harveyiLMG 4044TUSA (1935)
VIB 296Vibrio mediterraneiLMG 11258TSpain
VIB 298Vibrio mimicusLMG 7896TUSA
VIB 299Vibrio natriegensLMG 10935TUSA
VIB 301Vibrio nereisLMG 13543TUSA
VIB 304Vibrio parahaemolyticusLMG 2850TJapan
VIB 305Vibrio pelagiaLMG 3897TUSA
VIB 306Vibrio proteolyticusLMG 3772TUSA
VIB 309Vibrio tubiashiiLMG 10936TUSA
VIB 310Vibrio vulnificusLMG 13545TUSA
VIB 414Vibrio logeiLMG19806TUSA
VIB 351Vibrio harveyiSharkBahamas
VIB 391Vibrio harveyiShrimpThailand (1990)
VIB 395Vibrio harveyiLMG 11225 
VIB 400Vibrio harveyiLMG 11659 
VIB 410Vibrio harveyiATCC 14126 
VIB 571Vibrio harveyiSea bassSpain (1990)
VIB 572Vibrio harveyiSea breamSpain (1990)
VIB 645Vibrio harveyiSea bassTunisia (1993)
VIB 646Vibrio harveyiShark tank waterDenmark (1993)
VIB 647Vibrio harveyiSea breamGreece (1992)
VIB 648Vibrio harveyiShark liverDenmark
VIB 649Vibrio harveyiSea breamMalta (1993)
VIB 651Vibrio harveyiShark tank waterDenmark (1994)
VIB 652Vibrio harveyiSea bassItaly
VIB 653Vibrio harveyiSea bassTurkey
VIB 658Vibrio harveyiSea breamFrance (1990)
VIB 659Vibrio harveyiSea bassTunisia
SF1Vibrio harveyiSea perchChina (2002)
VIB 457Vibrio parahaemolyticusLMG 12093 
VIB 458Vibrio parahaemolyticusLMG 12094 
VIB 459Vibrio parahaemolyticusShrimpThailand
VIB 460Vibrio parahaemolyticusShrimpThailand
VIB 461Vibrio parahaemolyticusShrimpThailand
VIB 463Vibrio parahaemolyticusShrimpThailand
VIB 611Vibrio parahaemolyticusATCC 33844 
VIB 612Vibrio parahaemolyticusATCC 17803 
VIB 797Vibrio parahaemolyticusShrimpChina
VIB 798Vibrio parahaemolyticusShrimpThailand
VIB 799Vibrio parahaemolyticusShrimpThailand
VIB 800Vibrio parahaemolyticusShrimpThailand

DNA extraction

A volume (1.5 ml) of overnight bacterial culture from each Vibrio isolate was harvested by centrifugation at 5000 g for 5 min at 4°C, washed in 0.9% (w/v) saline, re-centrifuged, and the cell pellet was resuspended in 100 μl of distilled water. The mixture was boiled for 10 min, and centrifuged at 1000 g for 10 min to sediment the cell debris. Then, the DNA-containing supernatants were transferred to fresh Eppendorf tubes.

Primer design

Nucleotide sequences of partial toxR from various Vibrio spp. including V. harveyi, V. parahaemolyticus, Grimontia hollisae, V. cholerae, Vibrio fischeri and Vibrio anguillarum were aligned using ClustalW ( Specific primers for PCR were designed from the divergent region of the partial toxR gene of V. harveyi (GenBank accession number AY247418), which is a membrane-tether region of toxR. The forward primer, named toxRF1, was 5′-GAAGCAGCACTCACCGAT-3′; the reverse primer, named toxRR1, was 5′-GGTGAAGACTCATCAGCA-3′. The expected PCR is 382-bp long.

PCR amplification

The PCR procedure was as follows. The PCR mixture consisted of 0.5 μl of DNA sample, 5 μl of 10x PCR buffer, 3 μl of 25 mmol l−1 MgCl2, 1 μl of 2.5 mmol l−1 dNTPs, 0.25 μl of Taq polymerase (Shanghai Sangon Biological Engineering Technology and Services Co. Ltd, Shanghai, China) and 39 μl of distilled water. The amplification conditions were 30 cycles at 94°C for 1 min, 55°C for 1 min, and 72°C for 1 min, and then an extra extension step of 72°C for 10 min in the ABI 2400 PCR machine (Advanced Biotechnologies Inc., Columbia, MD, USA). Volumes (10 μl) of each PCR product were subjected to electrophoresis in a 1% (w/v) agarose gel.

The specificity and sensitivity of the PCR

The specificity of the primers was determined by PCR amplification of the extracted DNA from a wide range of vibrios (Table 1). To determine the detection limit of the PCR, bacterial cultures were grown overnight at 28°C, washed by 0.9% (w/v) saline, and diluted to 105–102 cells ml−1. Each suspension was examined in duplicate by the extraction and amplification procedures outlined above.

Detection of V. harveyi in cultures isolated from diseased fish

Diseased sea perch (Lateolabrax japonicus), flounder (Paralichthys olovaceas) and turbot (Scophthalmus maximus L.) were obtained from a marine aquaculture farm in Shandong Province, China during 1999–2004. To isolate bacterial pathogens, the diseased fish were washed for a few minutes with sterile (121°C, 15 min−1) 0.9% (w/v) saline. The damaged tissue were collected, homogenized and diluted with sterile saline. Then, 0.1 ml of suspensions were spread onto 2216E plates, and cultured for 24 h at 28°C. Colonies from plates displaying dense growth of one dominant colony type were subcultured to achieve purity. Each isolate was examined by the extraction and amplification procedures outlined above. Selected biochemical tests on positive isolates were carried out according to protocols of Alsina and Blanch (1994).

16S rDNA PCR, sequencing and phylogenic analysis

The 16S rDNA of bacterial isolates from diseased fish were amplified by PCR with bacterial universal primers P11: 5′-AGAGTTTGATCCTGGCTCAG-3′ (corresponding to 8–27 of Escherichia coli) and P12: 5′-GGTTACCTTGTTACGACTT-3′ (corresponding to 1492–1510 of E. coli). The expected PCR is c. 1503-bp long. The 16S rDNA fragment obtained from the PCR was purified using the TaKaRa agarose gel DNA purification kit (TaKaRa Biotechnology Co., Ltd, Dalian, China), and subsequently cloned into the PUCm-T vector (Shanghai Sangon). The recombinant plasmids containing the insert of expected size were sequenced by Shanghai Sangon. The 16S rDNA sequences were aligned and compared with available sequences of Vibrio spp. in the NCBI GenBank using BLAST. MegAlign expert sequence analysis software from DNASTAR Inc. (Madison, WI, USA) was used to construct phylograms. Two bacterial 16S rDNA sequences obtained were submitted to GenBank with nucleotide accession numbers DQ304557 and DQ304558.


Amplification of V. harveyi-specific toxR gene fragment

PCR primers (toxRF1 and toxRR1) targeting an internal fragment (382 bp) of the partial V. harveyitoxR gene (GenBank accession number AY247418) were designed. Subsequently, the primers were used in the amplification of toxR gene fragment from two V. harveyi reference strains and 18 V. harveyi strains obtained from a diverse range of hosts and geographical locations. The expected 382-bp fragment was obtained from all 20 V. harveyi strains (Fig. 1).

Figure 1.

 PCR profiles using toxRF1 and toxRR1 primers and DNA templates from 18 Vibrio harveyi isolates, 26 Vibrionaceae type strains and 12 Vibrio parahaemolyticus isolates. (a) Lanes 2–16, DNA templates from VIB72, VIB281, VIB283, VIB284, VIB285, VIB286, VIB287, VIB288, VIB289, VIB290, VIB291, VIB292, VIB293, VIB294 and VIB295 respectively. Lanes 1 and 17 contain the 100-bp DNA ladder. (b) Lanes 2–16, DNA templates from VIB296, VIB298, VIB299, VIB301, VIB304, VIB305, VIB306, VIB309, VIB310, VIB314, VIB400, VIB410, VIB457, VIB458 and VIB459 respectively. Lanes 1 and 17 contain the 100-bp DNA ladder. (c) Lanes 2–16, DNA templates from VIB460, VIB461, VIB463, VIB611, VIB612, VIB797, VIB798, VIB800, VIB351, VIB799, VIB391, VIB395, VIB571, VIB572 and VIB645 respectively. Lanes 1 and 17 contain the 100-bp DNA ladder. (d) Lanes 2–12, DNA templates from VIB646, VIB647, VIB648, VIB649, VIB651, VIB652, VIB653, VIB658, VIB659, SF1 and VIB414 respectively. Lanes 1 and 13 contain the 100-bp DNA ladder.

Specificity of PCR primers for detection of V. harveyitoxR gene

The specificity of the primers that were designed in this study for the amplification of V. harveyitoxR gene was investigated by using DNA templates from a wide range of vibrios. Results showed that the expected 382-bp fragment could only be observed in the PCR profile of the two V. harveyi type strains (Fig. 1). The expected size fragments were not observed from the PCR profiles of other Vibrionaceae type strains including G. hollisae, Photobacterium damselae ssp. damselae, Salinivibrio costicola ssp. costicola, V. anguillarum, Vibrio aestuarianus, Vibrio alginolyticus, Vibrio campbellii, Vibrio cincinnatiensis, Vibrio diazotrophicus, V. fischeri, Vibrio fluvialis, Vibrio furnissii, Vibrio gazogenes, Vibrio mediterranei, Vibrio natriegens, Vibrio nereis, Vibrio pelagia, Vibrio proteolyticus, Vibrio tubiashii and Vibrio logei. The only positives were for the two V. harveyi type strains (Fig. 1). Profiles from PCR using DNA templates from 12 V. parahaemolyticus isolates did not exhibit the 382 bp amplified fragment either (Fig. 1).

Sensitivity of the detection

Results indicated that the assay had a detection limit of 4.0 × 103 cells ml−1, assuming that the DNA extraction process has completely released all the bacterial DNA present in the sample.

Detection of V. harveyi in cultures isolated from diseased fish

Fifty-five isolates were recovered from diseased fish. Moreover, the PCR amplification targeting the toxR gene of V. harveyi showed amplification of 382 DNA fragments in eight of the 55 isolates (14.55%), i.e. XVP5, TA-1, FA-1, FM-1, C-12D, R2-16K, CY-16D and CC-36D. All these isolates were Gram-negative facultatively anaerobic curved rods, which were positive for oxidase production, grew on TCBS (thiosulphate-citrate-bile salts-sucrose) agar, and were sensitive to the vibriostatic agent O/129.

Confirmation of V. harveyi identification by 16S rDNA sequencing and phylogenic analysis

Two isolates (XVP5 and FA-1) were selected for 16S rDNA sequencing and phylogenetic analysis (Fig. 2). The genes from these isolates were 99% similarity with representative V. harveyi isolates already described in GenBank.

Figure 2.

 Phylogenic tree of 16S rDNA sequences from FA-1, XVP5 and other representative Vibrio isolates already described in GenBank by using MegAlign expert sequence analysis software from DNASTAR Inc.


The results of this study reinforce the view that PCR offers a highly sensitive and specific means for the detection and identification of organisms. In this case, the detection limit was 4.0 × 103 cells ml−1. Indeed, the entire procedure including DNA extraction from the cultures, PCR amplification and detection of the amplified DNA was finished within 5 h. It is relevant that the specificity was confirmed by 16S rDNA sequencing. Thus, the PCR amplification of the toxR gene developed in this study is a very effective method in identifying V. harveyi isolate. Moreover, the PCR detected the bacterial isolates from diseased fish obtained from aquaculture sites in China. Subsequently, some of these isolates were used in laboratory-based infectivity experiments to confirm virulence (data not shown).

The toxR genes are universally distributed in the family Vibrionaceae, and the nucleotide sequences reflect their phylogenetic relationship (Lin et al. 1993; Osoril and Klose 2000). This gene is regarded as an effective taxonomic marker for identification of Vibrio spp. (Kim et al. 1999). The outcome of this study was that the PCR showed positive amplification of the toxR gene fragment in all the 20 V. harveyi isolates including two type strains, whereas DNA from 23 other Vibrionaceae type strains and 13 V. parahaemolyticus strains were negative. This shows that the primers are highly specific for toxR gene of V. harveyi. This coincides with the work of Conejero and Hedreyda (2003), who also designed a pair of toxR-targeted PCR primers to detect V. harveyi. Although this study revealed that the expected 390-bp fragment was detected in all V. harveyi strains examined except in two shrimp-derived strains (VIB 391 and STD3-101), we did not experience the problem. This is especially relevant insofar as VIB 391 was common to both studies. The reason may well be that our primers and PCR conditions were different from what used in the other study. Nevertheless, it should be emphasized that Oakey et al. (2003) developed a PCR using 16S rDNA sequences, which gave positive results for all strains of V. harveyi. However, some cultures of V. alginolyticus also gave positive results. The reasons for this apparent anomaly could reflect that the 16S rDNA sequences of V. harveyi are too conserved with that of other Vibrio spp. We did not experience this lack of specificity.

In conclusion, a PCR protocol amplifying a 382-bp fragment of the V. harveyitoxR was established and could be useful in the specific and rapid detection of the species.


This work was supported by the program for New Century Excellent Talents in University (no. NCET-04-0645), a grant from the National Natural Science Foundation of China (no. 30371119), and a grant from the National High Technology Research and Development Program of China (863 program) (no. 2003AA622070).