Introduction

The luminous bacterium Vibrio harveyi was initiallyconsidered a normal component of marine intestinalmicroflora (Baumann et al., 1980). Recently, Vibrioharveyi has appeared to be a primary pathogen infect-ing hatchery-reared fish and shellfish species over awide geographical area (Pass et al., 1987; Pizzuttoand Hirst, 1995; Vandenberghe et al., 1998; Zhangand Austin, 2000). Mortalities in hatchery-reared pe-naeid shrimps caused by Vibrio harveyi infection hasresulted in severe economic losses in producing coun-tries in Asia (Karunasagar et al., 1994; Lavilla-Pitogoet al., 1990; Liu et al., 1996). The lack of tools thatclearly differentiate Vibrio harveyi from its phylogeneti-cally close relatives V. campbellii, V. alginolyticus andV. parahaemolyticus (Kita-Tsukamoto et al., 1993) hashampered the rapid and accurate detection of the lumi-nous bacterium.

A molecular approach to develop detection proce-dures for bacterial pathogens involves targeting one ofthe genes encoding a virulence factor. One such geneimplicated in the virulence of Vibrio species encodesfor hemolysins. Hemolysins are responsible for the dis-ruption of the erythrocyte membrane or hemolysis andgenes encoding for hemolysins have been reported tobe present in several members of the genus includingV. harveyi (Hirono et al., 1996; Nishibuchi and Kaper,1995; Nishibuchi et al., 1990; Zhang et al., 2001). He-molysin genes have been targeted in PCR for the de-tection of pathogenic strains of Vibrio species includinghemolysin genes (tl, tdh and trh) of V. parahaemolyti-cus (Bej et al., 1999) and V. vulnificus cytolysin gene(Lee et al., 1998).

Studies conducted by Zhang and Austin (2000) sug-gested that hemolysins are one of the pathogenic de-terminants of several V. harveyi strains isolated from avariety of hosts and geographic origins. Subsequently,they isolated the duplicated hemolysin genes (vhhAand vhhB) of a strain of V. harveyi pathogenic tosalmons (Zhang et al., 2001). Sequences of the twocopies of the V. harveyi (VIB 645) hemolysin genes(vhhA and vhhB), are almost identical and primers thatcould amplify the 1.3 kb open reading frame of bothcopies are already available but have not been used

PCR detection of hemolysin (vhh) gene in Vibrio harveyi

Mia Judith U. Conejero and Cynthia T. Hedreyda*

National Institute of Molecular Biology and Biotechnology, University of the Philippines, Diliman, Quezon City 1101, Philippines

(Received April 14, 2003; Accepted May 24, 2004)

The Vibrio harveyi hemolysin gene (vhh), which encodes for a virulence factor involved in patho-genicity to fish and shellfish species, may be targeted for species detection or strain differentia-tion. Primers designed for this gene were used in detection studies of V. harveyi strains fromvarious hosts. One primer set among four tested, could amplify the expected gene fragment inPCR using templates from all 11 V. harveyi strains studied. Detection of the presence of the he-molysin gene could therefore serve as a suitable detection marker of Vibrio harveyi isolates po-tentially pathogenic to fish and shrimps.

Key Words——hemolysin; PCR; vhh; Vibrio harveyi

J. Gen. Appl. Microbiol., 50, 137–142 (2004)

* Address reprint requests to: Dr. Cynthia T. Hedreyda, Molec-ular Microbiology Laboratory, National Institute of Molecular Bi-ology and Biotechnology—UP Diliman, Albert Hall, Lakandulacor. E. Jacinto, UP Diliman Campus, Quezon City 1101, Philip-pines.

E-mail: hedreyda@laguna.net

Full Paper

for detection of the species. The vhh gene sequencesof V. harveyi VIB 645 showed the highest homologywith the thermolabile gene (tl) of the close relative V.parahaemolyticus, but with only 85.6% degree of iden-tity, suggesting that the sequence variation in the he-molysin genes of different species may be targeted forspecies-specific detection of Vibrio isolates.

This present study focused on the detection of ten V.harveyi strains from various hosts and a reference V.harveyi strain IFO 15634, using new PCR primers de-signed in this study and targeting the hemolysin gene.A combination of the new PCR primers and a primerset from an earlier study by Zhang and co-workers(2001) was also tested, resulting in the identification ofa primer set specific for amplifying a 308-bp fragmentof the hemolysin gene in all strains of V. harveyitested. The ultimate goal of the study is to develop aprotocol for the rapid and accurate detection of V. har-veyi via hemolysin gene-targeted PCR in order to facil-itate effective disease prevention and surveillance ofhatchery-reared marine animals.

Materials and Methods

Bacterial isolates or strains. Ten Vibrio harveyi iso-lates obtained from various geographic origin andhosts (Table 1) from the Vibrio collection in the Depart-ment of Biological Sciences, Heriot-Watt University,Edinburgh, UK, through the generous assistance of Dr.Brian Austin were used in this study. These isolateswere characterized and identification was verifiedusing the protocol reported by Pedersen and co-work-ers (1998). This research also utilized referencestrains of Vibrio harveyi (IFO 15634). Other closely re-lated Vibrio species (Table 2) from the Institute of Fer-mentation Osaka, (IFO) Japan, and Institute of Molec-ular and Cellular Biosciences (IAM), University ofTokyo, were also used to determine if the primers an-neal to genes of non-target Vibrio isolates. These iso-lates were kindly provided by Dr. Rose Monsalud ofthe Philippine National Collection of Microorganisms,Los Baños, Laguna. Non-Vibrio bacteria were alsoused to test the specificity of primers to V. harveyi.

Vibrio cultures were routinely grown in Marine agar(Pronadisa Hispanlab, Madrid, Spain) or Nutrient agar(Becton Dickinson and Company, Cockeysville, MD,USA) with 1.5% NaCl at 30°C overnight. Non-Vibriobacterial cultures were maintained by growing in Luria-Bertani agar (LB agar, Pronadisa Hispanlab) with

overnight incubation at 37°C.DNA extraction. Genomic DNA extracts from bac-

teria and black tiger shrimp (Penaeus monodon) wereprepared using a commercially available Nucleospin®

Tissue kit (Clontech Laboratories, Inc., Palo Alto, CA,USA). One possible use of the detection procedure isanalysis of shrimp samples or water from shrimpponds; thus, DNA from black tiger shrimp (Penaeusmonodon) was also used as a template for PCR toshow that primers were not annealing to shrimp DNA.

PCR amplification of the V. harveyi hemolysin gene(vhh). Primers targeting different regions of the vhhgene utilized in this study are listed in Table 3.Oligonucleotide primers for the amplification of the he-molysin gene fragment were designed with the aid ofthe DNASIS v2.5 program based on the nucleotide se-quences of reported vhhA and vhhB genes from V.

138 CONEJERO and HEDREYDA Vol. 50

Table 1. V. harveyi cultures from different hosts and geographic origins.

Strain SourceCountry of isolation

VIB 295 LMG 4044T�V. harveyi USAVIB 391 Shrimp Thailand (1990)VIB 572 Sparus aurata (sea bream) Spain (1990)VIB 645 Dicentrarchus labrax (sea bass) Tunisia (1993)VIB 647 S. aurata Greece (1992)VIB 648 Shark liver DenmarkVIB 649 S. aurata Malta (1993)VIB 653 D. labrax TurkeyVIB 658 S. aurata France (1990)STD 3-101 Larval shrimp Ecuador (1994)

LMG�Culture Collection of the Laboratorium voor Microbiolo-gie, Universiteit Gent, Belgium.

T�type strain.

Table 2. Non-target Vibrio species used in this study.

Strain Identification

IFO 12711 V. parahaemolyticusIFO 13287 V. proteolyticusIFO 15630 V. alginolyticusIFO 15631 V. campbelliiIFO 15634 V. harveyiIFO 15635 V. natriegensIFO 15645 V. vulnificusIAM 14402 V. diazotrophicus

harveyi VIB 645 (GenBank Accession numbersAF293430 and AF29343, respectively). These primerstarget unique regions in the V. harveyi vhh genes re-vealed by Clustal W alignment (http://www.ebi.ac.uk/clustalw/) with homologous hemolysin sequences inother Vibrio species. All primers were custom-de-signed by Genset Corp. (Singapore). Using differentprimer combinations, DNA samples from Vibrio andnon-Vibrio species were subjected to 30 cycles of94°C for 1 min, 53°C for 1 min and 72°C for 1 min in aPerkin Elmer Gene Amp System (Perkin Elmer Corp.,Singapore). PCR products were resolved by elec-trophoresis in 1.5% agarose gel with 3.5 mm wells.

Direct sequencing of PCR product. The PCR reac-tion mixture containing the amplified fragments of thehemolysin gene of Vibrio harveyi (IFO 15634) was pu-rified using a Nucleotrap PCR Purification Kit (Clon-tech Laboratories). The purified PCR product was sub-jected to sequencing protocols using the ABI 377 Au-tomated Sequencer (Perkin Elmer, Foster City, CA,USA) of the AMOR Laboratory (NIMBB, University ofthe Philippines, Diliman).

Results

Amplification of the V. harveyi hemolysin gene (vhh)PCR primers reported by Zhang and co-workers

(2001) were used to amplify the 1.3 kb open readingframe of the hemolysin gene in the reference strain V.harveyi IFO 15634 (Fig. 1). The identity of the PCR-amplified 1.3 kb product was verified by direct se-quencing and BLAST homology search. The partialnucleotide sequence displays 95% homology with thereported Vibrio harveyi vhhA and vhhB genes de-posited in the GenBank database (data not shown).Subsequently, other nested primers were designed inthis study in order to amplify internal hemolysin vhhgene fragments in the V. harveyi reference strain.Primer sequences used (Table 3) include a primer setF2 and R2 designed in this study, generating a 216-bphemolysin gene fragment. PCR reactions using combi-nations of the old and new primers and template DNAfrom the reference V. harveyi strain, was observed toyield amplified fragments with the following approxi-mate sizes: 308 bp and 1.2 kb (with primer sets, F1R2

2004 Hemolysin gene detection in Vibrio harveyi 139

Table 3. Nucleotide sequences of primers for amplification of the vhh gene.

Primer Sequence Annealing sitea

F1 or VHF1 (forward) 5�-ATCATGAATAAAACTATTACGTTACT-3� 1→26R1 or VHR1 (reverse) 5�-GAAAGGATGGTTTGACAAT-3� 1257→1238F2 or VhhemoF (forward) 5�-TCAGTGCCTCTCAAGTAAGA-3� 92→111R2 or VhhemoR (reverse) 5�-GCTTGATAACACTTTGCGGT-3� 307→287

a Nucleotide positions are relative to the first nucleotide of the vhh gene of V. harveyi (GenBank Accession number AF293430).

Fig. 1. Primer annealing sites (A) and profile of amplified products of the V. harveyi vhh gene using different primercombinations (B).

Lanes 1–4: 1.3 kb, 216 bp, 308 bp and 1.2 kb amplified fragments using the primer sets F1R1, F2R2, F1R2 andF2R1, respectively. M is the 100 bp DNA ladder (Roche Applied Bioscience, Indianapolis, IN, USA).

and F2R1, respectively; Fig. 1).In order to test the universality of the primer sets in

producing the expected vhh gene amplified fragments,genomic DNA extracts from ten (10) Vibrio harveyi iso-lates from different hosts and geographical locationswere used as PCR templates (Fig. 2). The 1.3 kb PCRproduct generated using the F1R1 primer set (de-signed in an earlier study) was observed in all V. har-veyi strains tested except 2 strains, VIB 391 and STD3-101, isolated from larval shrimp in Thailand andEcuador, respectively. Amplified DNA fragments werealso observed from all isolates tested after PCR usingthe new set of primers F2R2 and the combinedprimers F2R1 (216-bp and 1.2-kb, respectively), ex-cept in two Vibrio harveyi strains that did not yield theexpected products with F1R1 either. Interestingly, the308-bp amplified fragment produced by the F1R2primer set with V. harveyi (IFO 15634) was consis-tently obtained in all ten (10) V. harveyi strains. The

expected 308-bp band and an additional 600-bp ampli-con were observed in profiles of one strain fromEcuador (STD 3-101). The presence of this additionalband can not be explained at this time.

Test for the specificity of the PCR detection protocolfor V. harveyi vhh gene

The specificity of the primer set amplifying 308-bpVibrio harveyi vhh gene fragment was established byusing the target V. harveyi, non-target Vibrio species,and several non-Vibrio DNA templates (data notshown). The expected amplicon was only observed inV. harveyi. Use of DNA templates from non-Vibriospecies and the black tiger shrimp (Penaeusmonodon) did not yield the 308-bp amplicon, indicatingthat no false positive results will be obtained even inthe presence of these DNA in samples to be analyzed.

140 CONEJERO and HEDREYDA Vol. 50

Fig. 2. Amplified fragments of the hemolysin gene (vhh) obtained using different primer sets and eleven V. harveyistrains.

A–D, use of primer sets F1R1, F2R2, F1R2 and F2R1, respectively. Lanes 1–11, V. harveyi DNA templates from Vib-rio harveyi isolates IFO 15634, VIB 295, VIB 391, VIB 572, VIB 645, VIB 647, VIB 648, VIB 649, VIB 653, VIB 658 andSTD 3-101, respectively. M contains 100-bp DNA ladder and lane 12 is the negative control.

Discussion

The primers reported in a previous study (Zhang etal., 2001) were utilized in the PCR amplification of the1.3 kb hemolysin gene from Vibrio harveyi strains ob-tained from different geographic locations (Table 1).Amplification using the reported primer set success-fully generated 1.3 kb putative hemolysin gene fromonly nine out of eleven strains including the referenceV. harveyi strain IFO 15634 (Fig. 2A). This observationsuggests that the other two strains may not possessthe gene, or the possibility that sequence variationswithin the annealing site of at least one of the primersexist. Partial DNA sequence analysis of the resulting1.3 kb fragment from the reference V. harveyi IFO15634 strain and sequence comparison with the previ-ously reported vhhA and vhhB genes, confirmed thatthe amplicon is the gene for V. harveyi hemolysin.

It was necessary to design other primers that couldresult in amplification of a hemolysin gene fragment forall strains studied, including the variant strains of V.harveyi from Thailand and Ecuador. Accordingly, a setof new primers designated as F2R2, F2 for the forwardprimer and R2 for the reverse primer (Table 2 and Fig.1), was designed in this study. The use of F2 and R2primers could amplify a 216-bp internal fragment of thehemolysin gene in nine out of the eleven V. harveyistrains used (Fig. 2B) but did not amplify genes of theisolates from Thailand (VIB 391) or Ecuador (STD 3-101). When the F2 forward primer designed in thisstudy was used in combination with the reverse primerR1 from a previous study (Zhang et al., 2001), the ex-pected 1.2-kb amplicon was observed in nine (9) V.harveyi strains, but still no PCR product was producedin the Thailand or Ecuador isolates (Fig. 2D). Finally, auniversal primer set was discovered in another PCRexperiment that made use of a combination of a for-ward primer F1 from Zhang and co-workers (2001)and a reverse primer R2, designed in this study. TheF1 and R2 primer combination or primer set F1R2,could amplify the expected 308-bp PCR amplicon in allof the eleven (11) V. harveyi strains (Fig. 2C), includingthe isolates from Thailand and Ecuador. This findingsuggests that the annealing sites for primer set F1R2were relatively conserved in all strains of V. harveyiused in this study. The PCR product, which is approxi-mately 308-bp, was amplified using the following opti-mized PCR conditions: denaturation at 94°C for 1 min,annealing at 53°C for 1 min, and extension at 72°C for

1 min. Reaction mixture for optimum amplification in-clude the following components: 1.5 mM MgCl2, 0.2 mM

deoxynucleotide triphosphate (dNTP) mix, 1� PCRbuffer, 0.5 mM of each primer (F1 and R2) and0.025 U/ml of Taq DNA polymerase (Invitrogen, Carls-bad, CA, USA).

All primer sets used in this study could amplify he-molysin gene fragments of all strains exhibiting refer-ence strain characteristics but not the variants fromThailand and Ecuador. These PCR results are consis-tent with results of a previous study that characterizedand grouped these isolates based on amplified frag-ment length polymophism (AFLP) profiles (Pedersenet al., 1998). Strains that could be detected by allprimer sets in this study belonged to AFLP cluster 1,which was considered to be the bonafide V. harveyigroup. The two variant isolates were not grouped inthe same AFLP cluster; the isolate from Thailand (VIB391) belonged to cluster 4 and the isolate fromEcuador (STD 3-101) was unclassified. Both variantswere only detected using the primers for the 308-bpPCR product, suggesting that significant sequencevariation in hemolysin gene sequence from these twostrains of V. harveyi. Moreover, the variant strain fromEcuador possesses an additional 600-bp amplicon, re-quiring further sequence analysis. It would be interest-ing to sequence the complete hemolysin genes of theshrimp-derived strains from Thailand and Ecuador inorder to compare V. harveyi hemolysin sequences withthose reported in the database and in order to investi-gate the possible significance of these variations inshrimp pathogenesis.

The PCR protocol specifically amplifying the 308-bpvhh fragment in Vibrio harveyi may facilitate the rapidand accurate detection of V. harveyi from infected fishand shellfish species. This will allow early diagnosis ofV. harveyi infection for effective disease preventionand surveillance.

Acknowledgments

This research was funded by the Bureau of Agricultural Re-search, Department of Agriculture (DA-BAR) and the NationalInstitute of Molecular Biology and Biotechnology, University ofthe Philippines Diliman (NIMBB, UP Diliman). Authors wish tothank Dr. Brian Austin of Heriot-Watt University, Edinburgh, UKfor the V. harveyi cultures and Dr. Rosario Monsalud of NIMBB,UP Los Baños for the reference strains of different Vibriospecies.

2004 Hemolysin gene detection in Vibrio harveyi 141

References

Baumann, P., Baumann, L., Bang, S. S., and Woolkalis, M. J.(1980) Re-evaluation of the taxonomy of Vibrio, Beneckea,and Photobacterium: Abolition of the genus Beneckea.Curr. Microbiol., 4, 127–132.

Bej, A. K., Patterson, D. P., Brasher, C. W., Vickery, M. C.,Jones, D. D., and Kaysner, C. A. (1999) Detection of totaland hemolysin-producing Vibrio parahaemolyticus in shell-fish using multiplex PCR amplification of tl, tdh and trh. J.Microbiol. Methods, 36, 215–225.

Hirono, I., Masuda, T., and Aoki, T. (1996) Cloning and detec-tion of the hemolysin gene of Vibrio anguillarum. Microb.Pathog., 21, 173–182.

Karunasagar, I., Pai, R., Malathi, G. R., and Karunasagar, I.(1994) Mass mortality of Penaeus monodon larvae due toantibiotic resistant Vibrio harveyi infection. Aquaculture,128, 203–209.

Kita-Tsukamoto, K., Oyaizu, H., Nanba, K., and Shimidu, U.(1993) Phylogenetic relationships of marine bacteria,mainly members of the family Vibrionaceae, determined onthe basis of 16s rRNA sequences. Int. J. Syst. Bacteriol.,43, 8–19.

Lavilla-Pitogo, C. R., Baticados, M. C. L., Cruz-Lacierda, E. R.,and dela Peña, L. D. (1990) Occurrence of luminous bacte-rial disease of Penaeus monodon larvae in the Philippines.Aquaculture, 91, 1–13.

Lee, S. E., Kim, S. Y., Kim, S. J., Kim, H. S., Shin, J. H., Choi,S. H., Chung, S. S., and Rhee, J. H. (1998) Direct Identifi-cation of Vibrio vulnificus in clinical specimens by nestedPCR. J. Clin. Microbiol., 36, 2887–2892.

Liu, P. C., Lee, K. K., Yii, K. C., Kou, G. H., and Chen, S. N.(1996) Isolation of Vibrio harveyi from diseased kurumaprawns Penaeus japonicus. Curr. Microbiol., 33, 129–132.

Nishibuchi, M. and Kaper, J. (1995) Thermostable direct he-molysin gene of Vibrio parahaemolyticus: A virulence geneacquired by a marine bacterium. Infect. Immun., 63,2093–2099.

Nishibuchi, M., Khaeomanee-iam, V., Honda, T., Kaper, J. B.,and Miwatani, T. (1990) Comparative analysis of the he-molysin genes of Vibrio cholerae non-O1, Vibrio mimicus,and Vibrio hollisae that are similar to the tdh gene of Vibrioparahaemolyticus. FEMS Microbiol. Lett., 67, 251–256.

Pass, D. A., Dybdahl, R., and Mannion, M. M. (1987) Investiga-tions into the causes of mortality of the pearl oyster, Pinc-tada maxima (Jamson), in Western Australia. Aquaculture,65, 149–169.

Pedersen, L. V., Austin, B., Austin, D. A., Blanch, A. R., Gri-mont, P. A. D., Jofre, J., Koblavi, S., Larsen, J. L., Tiainen,T., Vigneulle, M., and Swings, J. (1998) Taxonomic evi-dence that Vibrio carchariae Grimes et al. 1985 is a juniorsynonym of Vibrio harveyi (Johnson and Shunk 1936) Bau-mann et al. 1981. Int. J. Syst. Bacteriol., 48, 749–758.

Pizzutto, M. and Hirst, R. G. (1995) Classification of isolates ofVibrio harveyi virulent to Penaeus monodon larvae by pro-tein profile analysis and M13 DNA fingerprinting. Dis.Aquat. Org., 21, 61–68.

Vandenberghe, J., Li, Y., Verdonk, L., Li, J., Sorgeioos, P., Xu,H. S., and Swings, J. (1998) Vibrios associated with Pe-naeus chinensis (Crustacea:Decapoda) larvae in Chineseshrimp hatcheries. Aquaculture, 169, 121–132.

Zhang, X.-H. and Austin, B. (2000) Pathogenicity of Vibrio har-veyi to salmonids. J. Fish Dis., 23, 93–102.

Zhang, X.-H., Meaden, P. G., and Austin, B. (2001) Duplicationof hemolysin genes in a virulent isolate of Vibrio harveyi.Appl. Environ. Microbiol., 67, 3161–3167.

142 CONEJERO and HEDREYDA Vol. 50