SGM Journals
ENVIRONMENTAL MICROBIOLOGY

Molecular identification of Vibrio harveyi-related isolates associated with diseased aquatic organisms

Bruno Gomez-Gil, Sonia Soto-Rodríguez, Alejandra García-Gasca, Ana Roque, Ricardo Vazquez-Juarez, Fabiano L. Thompson, Jean Swings

  • 1CIAD/Mazatlán Unit for Aquaculture, AP 711, Mazatlán Sinaloa, Mexico 82000
  • 2CIBNOR, AP 128, La Paz, Baja California, Mexico 23090
  • 3Laboratory for Microbiology, K. L. Ledeganckstraat 35, Ghent University, Ghent 9000, Belgium
  • Correspondence
    Bruno Gomez-Gil
    bruno{at}victoria.ciad.mx

    • Microbiology 2004; 150(6):1769–1777 · https://doi.org/10.1099/mic.0.26797-0

    View at publisher PubMed

    Abstract

    Fifty strains belonging to Vibrio harveyi, Vibrio campbellii, and the recently described Vibrio rotiferianus, were analysed using phenotypic and genomic techniques with the aim of analysing the usefulness of the different techniques for the identification of V. harveyi-related species. The species V. harveyi and V. campbellii were phenotypically indistinguishable by more than 100 phenotypic features. Thirty-nine experimental strains were phenotypically identified as V. harveyi, but FAFLP, REP-PCR, IGS-PCR and DNA–DNA hybridization proved that they in fact belong to the species V. campbellii. Similar groupings were found among all fingerprinting methodologies (except IGS-PCR). Thirty-two experimental strains clustered with the V. campbellii type and one reference strain; seven strains clustered with the V. harveyi type and three reference strains; and the type and four reference strains of V. rotiferianus grouped together. The correlations between DNA–DNA hybridization and the genomic fingerprinting by FAFLP and (GTG)5-PCR were found to be above 0·68 and statistically significant, suggesting the value of the latter techniques for the reliable identification of V. harveyi-related species. The results presented indicate that strains phenotypically identified as V. harveyi are in fact V. campbellii; these findings position V. campbellii as an important species involved in diseases of reared aquatic organisms.

    INTRODUCTION

    Vibrios are one of the most important pathogens for reared aquatic organisms such as penaeid shrimps (Lightner, 1993), several fish species and molluscs (Austin, 1988), and also for corals (Ben-Haim et al., 2003). Some of the luminescent vibrios, which include Vibrio cholerae (biotype albensis), V. fischeri, V. harveyi, V. logei, V. splendidus, V. mediterranei (Farmer & Hickman-Brenner, 1992), V. orientalis (Yang et al., 1983), Photobacterium leiognathi and P. phosphoreum have been implicated principally with disease outbreaks in shrimp larviculture facilities (Lavilla-Pitogo et al., 1990), and to a lesser degree in grow-out ponds (Lavilla-Pitogo et al., 1998). Only a few Vibrio species have been proven to be pathogens for shrimps; the closely related species V. harveyi and V. campbellii have caused disease in shrimp larvae (Abraham et al., 1999; Karunasagar et al., 1994; Hameed et al., 1996; Prayitno & Latchford, 1995), while V. penaeicida and V. parahaemolyticus affected juveniles and adults (Ishimaru et al., 1995; Roque et al., 1998).

    The taxonomy of Vibrio is in the process of revision due to the increasing data obtained with modern molecular biology techniques, where different genes are examined or where the whole genome is inspected. Special emphasis has been paid to the 16S rRNA, although other genes, such as those for 23S rRNA, 16S–23S intergenic spacer region (IGS), or the gyrB gene, have been employed (Chun et al., 1999; Dorsch et al., 1992; Venkateswaran et al., 1998). Unfortunately, the 16S rRNA is unable to resolve closely related species (Nagpal et al., 1998), such as the ones clustered in the Vibrio core group, namely V. alginolyticus, V. parahaemolyticus, V. harveyi, V. campbellii, V. natriegens and the newly described V. rotiferianus (Gomez-Gil et al., 2003). The identification of vibrios isolated from the aquacultural environment has been imprecise and is labour-intensive, requiring many biochemical and/or physiological tests (Vandenberghe et al., 2003).

    Several highly powerful molecular tools, e.g. amplified fragment length polymorphism (AFLP, Rademaker et al., 2000; Gurtler & Mayall, 2001) and repetitive extragenic palindromic elements polymerase chain reaction (REP-PCR) (Versalovic et al., 1991), have become readily available for the identification of bacteria, including vibrios (Thompson et al., 2001a; Sawabe et al., 2003). Rep-PCR can also differentiate strains of the same species, e.g. Escherichia coli (Dombek et al., 2000), Bradyrhizobium spp. (Vinuesa et al., 1998), Bacillus subtilis (Versalovic et al., 1991) and Vibrio cholerae (Rivera et al., 1995). In this study, we report on the use of FAFLP, REP-PCR and IGS-PCR for the identification of V. harveyi-related species.

    METHODS

    Bacterial strains.

    Eleven type and reference strains were obtained from the BCCM/LMG Bacteria Collection (Ghent University, Belgium); 39 isolates were obtained from the LMG and CAIM (CIAD, Mazatlán Unit, Mexico) collections (Table 1⇓). These isolates originated mainly from aquacultural environments, including penaeid shrimp, fish and molluscs, and seawater from several locations. Isolates were preserved at −70 °C according to the methodology of Gherna (1994).

    Table 1.

    Type, reference and experimental strains employed in this study

    Synonyms used in other publications or catalogues are shown in parentheses. ATCC, American Type Culture Collection; CAIM, Collection of Aquacultural Important Micro-organisms; LMG, Laboratorium voor Microbiologie-Bacteriënverzameling; R, Research Collection, Microbiology Laboratory, University of Ghent, Belgium.

    Phenotypic characterization.

    Phenotypic identification followed the schemes of Alsina & Blanch (1994) and of Holt et al. (1994). Tests performed were Gram stain, growth in thiosulphate/citrate/bile salts/sucrose agar (TCBS; Difco), bioluminescence on luminescence agar (West & Colwell, 1984), oxidase, sensitivity to the vibriostatic agent O/129, OF, arginine dehydrolase, ornithine decarboxylase and lysine decarboxylase. Growth at 0, 2·5 and 8 % NaCl, indole production, gelatinase production, Voges–Proskauer, and utilization of citrate, l-arabinose and d-glucosaminic acid were also determined. A further characterization was done with the Biolog GN2 system; 95 different carbon sources were tested with this system following the manufacturer's instructions but with the addition of NaCl to give a final concentration of 2·5 %. Data for the five strains of V. rotiferianus were obtained from Gomez-Gil et al. (2003). All other tests were done according to the methodologies of West & Colwell (1984) and Cowan et al. (1993) but NaCl was added to a final concentration of 2·5 % to allow the growth of the strains.

    Genomic characterization.

    DNA from each strain was extracted with the Promega Wizard Genomic DNA Purification kit (A1120). For DNA–DNA hybridization experiments, DNA from the selected strains was extracted following the technique of Pitcher et al. (1989). The DNA quality of both extractions was assessed by observations of an electrophoresed sample (1·0 % agarose, 1× TAE buffer, 1 h, 100 V). The same extracted DNA was employed in all genetic fingerprinting methods, except DNA–DNA hybridization, as explained above.

    PCR reaction mix for REP-PCR contained 12·45 μl water, 1·25 μl dNTP mix (25 mM each), 2·5 μl DMSO, 5·0 μl 5× Gitschier buffer (Rademaker et al., 1998), 0·4 μl BSA (10 mg ml−1), 1·0 μl of each primer (0·3 μg μl−1), 0·4 μl Taq polymerase (5 U μl−1, AmpliTaq; Applied Biosystems) and 1·0 μl DNA (50 ng μl−1) for a final volume of 25 μl. For IGS amplification, the reaction mix contained 18·3 μl water, 2·5 μl dNTP mix (2 mM each), 2·5 μl 10× PCR buffer with 25 mM MgCl2 (Applied Biosystems), 0·26 μl of each primer (0·25 μg μl−1), 0·15 μl of Taq polymerase (5 U μl−1, AmpliTaq; Applied Biosystems) and 1·0 μl of DNA (50 ng μl−1) for a final volume of 25 μl.

    Primers for REP-PCR were REP1R (5′-III ICG ICG ICA TCI GGC-3′) and REP2 (5′-ICG ICT TAT CIG GCC TAC-3′). Inosine (I) contains the purine base hypoxanthine, capable of forming Watson–Crick base pairs with A, G, C or T (Versalovic et al., 1991). BOX-PCR employs the primer BOXA1R (5′-CTA CGG CAA GGC GAC GCT GAC G-3′), and (GTG)5-PCR the primer 5′-GTG GTG GTG GTG GTG-3′ (Versalovic et al., 1994). The amplification protocol for BOX-PCR was 95 °C for 2 min, followed by 35 cycles of 94 °C for 3 min, 92 °C for 30 s, 50 °C for 1 min and 65 °C for 8 min with a final extension of 65 °C for 8 min. The amplification protocol for REP- and (GTG)5-PCR was 95 °C for 2 min, followed by 35 cycles of 94 °C for 3 min, 92 °C for 30 s, 40 °C for 1 min and 65 °C for 8 min with a final extension of 65 °C for 8 min. Primers for the IGS-PCR 16S–23S were V16S-1492F (5′-AAG TCG TAA CAA GGT ACG GCT-3′) and V23S-68R (5′-GCC TCA TCT ACG CTT ATC GC-3′). The amplification protocol was 94 °C for 2 min followed by 35 cycles of 94 °C for 1 min, 70 °C for 1 min and 72 °C for 1 min with a final extension of 72 °C for 5 min.

    The amplification products were resolved in a 1·5 % agarose gel in TAE buffer: 5 μl of a gel loading dye was mixed with the 25 μl of the reaction, and 10 μl of the mixture was loaded in the gel; 5 μl of a PCR molecular mass marker (Smartladder; Eurogenetec) was added every five lanes. The gel was electrophoresed at 4–8 °C for 15 h at 55 V. The gel was stained in an ethidium bromide solution with 1× TAE buffer for 20 min, destained for 5 min and photographed with a digital system. The resulting images were processed with Bionumerics 2.5 software (Applied Maths).

    FAFLP (fluorescent amplified fragment length polymorphism) patterns were generated and analysed as described previously (Thompson et al., 2001a). Briefly, 1 μg high-molecular-mass DNA was digested with TaqI and HindIII followed by ligation of restriction half-site specific adapters to all restriction fragments with T4 ligase. Pre-selective PCR amplification was done with H00-ABI primer (5′-GAC TGC GTA CCA GCT T-3′) and T00-ABI primer (5′-CGA TGA GTC CTG ACC GA-3′), and the selective PCR amplification with H01-6FAM primer (5′-GAC TGC GTA CCA GCT TTA-3′) and T03-ABI (5′-CGA TGA GTC CTG ACC GAG-3′).

    DNA–DNA hybridization of representative strains of each cluster was done following the methodology described by Willems et al. (2001) under stringent conditions (39 °C) and data obtained from previous works (Gomez-Gil et al., 2003; Thompson et al., 2001a); in total, values for 13 strains were employed.

    Numerical analyses.

    Similarity among band patterns was calculated with the Dice similarity coefficient and dendrograms were constructed with the Ward algorithm. A band position tolerance of 0·5–0·7 % was allowed to compensate for misalignments of homologous bands due to technical imperfections (Thompson et al., 2001a). Correlation analysis of DNA homologies and similarity values from all the REP-PCR and IGS-PCR experiments were done with the Pearson product-moment coefficient. Data were analysed for normality using SigmaStat for Windows version 2.03 (SPSS). Fifty values were analysed for all tests, except for IGS-PCR, where only 39 values could be obtained.

    RESULTS

    All the strains analysed grew on TCBS agar, were motile, fermented glucose, were oxidase positive, and were sensitive to the vibriostatic agent O/129 at 150 μg. These characteristics placed them within the genus Vibrio (Alsina & Blanch, 1994). Additionally, all the isolates were arginine dehydrolase negative, and lysine decarboxylase and ornithine decarboxylase positive (except R-14901, which was ornithine decarboxylase negative). All the experimental strains were phenotypically identified as V. harveyi according to the scheme proposed by Alsina & Blanch (1994). Luminescence was observed in 74·3 % of the experimental strains. V. harveyi is differentiated from V. campbellii in being positive for nine other characters (Holt et al., 1994). Almost all the strains analysed with the Biolog GN2 system for these other characters were positive for utilization of d-gluconate (95·0 %, 2·4 % doubtful, n=42), l-glutamate (90·0 %, 4·8 % doubtful, n=42), d-glucuronate (74·0 %, n=42), heptanoate (97·2 %, n=36), d-galactose (83·0 %, 2·4 % doubtful, n=42), and growth at 40 °C (80·6 %, n=36). Negative results were obtained for utilization of l-histidine (80·5 %, 2·4 % doubtful, n=42) and l-arabinose (92·9 %, n=42); utilization of sucrose produced varying results (88·9 % positive, n=42).

    FAFLP analysis produced the highest number of bands per strain compared to the other analyses (mean 102·4, max. 147, min. 72, sd 13·41, n=49). Lower numbers were obtained, in descending order, with (GTG)5-PCR (mean 24·1, max. 31, min. 17, sd 2·93, n=49), BOX-PCR (mean 11·0, max. 17, min. 6, sd 2·92, n=48), REP-PCR (mean 14·8, max. 27, min. 5, sd 4·48, n=49) and IGS-PCR (mean 6·5, max. 9, min. 5, sd 1·04, n=44). REP-PCR produced the highest band sizes (mean 2384 bp, max. 9710, min. 153, sd 1806, n=49), followed by (GTG)5-PCR (mean 1414 bp, max. 4560, min. 187, sd 837, n=49), BOX-PCR (mean 1109 bp, max. 5130, min. 191, sd 786, n=48) and IGS-PCR (mean 611 bp, max. 894, min. 351, sd 161, n=44), and the lowest size of bands with FAFLP (215 bp, max. 534, min. 50, sd 120, n=49). No bands could be obtained for strain LMG 16828 with REP- and GTG5-PCR, nor for LMG 16835 with FAFLP. Strains LMG 21457 and CAIM 113 did not produce any bands with BOX-PCR. V. rotiferianus strains (LMG 21456 to LMG 21460T) were not tested in IGS-PCR, and no bands were obtained for strain LMG 16835.

    Clustering of the strains with the FAFLP and REP-PCR methods produced similar groups (Figs 1 and 2⇓⇓). Type and reference strains of the three species analysed were clustered in different groups when a cut-off level of 45 % similarity was applied for the case of FAFLP (Thompson et al., 2001a). In this case, 33 strains were assigned to V. campbellii, 11 to V. harveyi, and five to V. rotiferianus (Fig. 1⇓). The V. campbellii group is very diverse and the FAFLP cluster cutoff value for this species is less than 10 % similarity. V. harveyi and V. rotiferianus present tighter clusters well delimited at or above the 45 % cutoff FAFLP value.

    Figure image not available in archive
    Fig. 1.

    Dendrograms of strains belonging to V. campbellii, V. harveyi and V. rotiferianus with different genetic fingerprinting methods. Strains marked with an asterisk (*) were DNA–DNA hybridized. Similarities were calculated with the Dice coefficient and the dendrograms constructed with the Ward algorithm.

    Figure image not available in archive
    Fig. 2.

    Dendrograms of strains belonging to V. campbellii, V. harveyi and V. rotiferianus with different genetic fingerprinting methods. Strains marked with an asterisk (*) were DNA–DNA hybridized. Similarities were calculated with the Dice coefficient and the dendrograms constructed with the Ward algorithm.

    Cluster analysis with (GTG)5- and REP-PCR produced similar results to FAFLP, but strains R-14905 and LMG 13241 clustered apart from the three main groups (Fig. 1⇑). BOX-PCR also produced a dendrogram very similar to the other methods, but only one strain (LMG 11256) was not assigned to any major group (Fig. 2⇑). Analysis with IGS-PCR produced two big clusters at a cutoff value of 75 % but both clusters contained strains assigned to V. campbellii and to V. harveyi with all the other methods employed here.

    Eight strains were assigned to V. campbellii by DNA–DNA hybridization (Table 2⇓); values above 70 % similarity were obtained between the type strain LMG 11216T and CAIM 372 (79·2 %), R-14899 (78·0 %), CAIM 128 (71·3 %), CAIM 333 (74·0 %), CAIM 415 (81·7 %), LMG 20369 (76·1 %), CAIM 113 (81·8 %) and LMG 16835 (77·6 %). Strains LMG 7890 and LMG 19643 were identified as V. harveyi because the DNA similarity value with the type strain LMG 4044T was 97·9 and 79·6 % respectively, and that between the two reference strains was 82·0 % (Thompson et al., 2001a; Pedersen et al., 1998). Values below the 70 % proposed by Wayne et al. (1987) to delimit a species were obtained between the three type strains of the species analysed: 69·0 between V. campbellii LMG 11216T and V. harveyi LMG 4044T, 65·0 % between V. campbellii LMG 11216T and V. rotiferianus LMG 21460T, and 66·0 % between V. harveyi LMG 4044T and V. rotiferianus LMG 21460T (Gomez-Gil et al., 2003).

    Table 2.

    DNA–DNA hybridization values among selected strains identified as V. harveyi and V. campbellii

    Comparison of the DNA–DNA hybridization values and the similarity values obtained with the DNA fingerprinting methods employed here gave the highest and significant correlation (P<0·05, n=50) between DNA and FAFLP (Table 3⇓). Also a high and significant correlation was obtained with (GTG)5-PCR, and a somewhat lower, but still significant, correlation with REP- and BOX-PCR. No correlation was obtained with IGS-PCR.

    Table 3.

    Pearson product-moment correlation coefficient between similarity values of the genetic fingerprinting methods and DNA–DNA hybridization (n=50, except IGS-PCR=39)

    The analysis with genomic fingerprinting and DNA homologies clearly shows that many strains phenotypically identified as V. harveyi are in fact V. campbellii. Apart from the type strain, only strains LMG 7890, LMG 11226, LMG 19643, LMG 20977, CAIM 2, CAIM 79, CAIM 462, CAIM 463, CAIM 464 and CAIM 689 are V. harveyi (Figs 1 and 2⇑⇑).

    No clear phenotypic differences were found between the strains of V. harveyi and V. campbellii analysed in this study; at most, only a majority of strains had positive or negative reaction to eight tests (Table 4⇓). None of the tests suggested by Holt et al. (1994) to differentiate the two species (mentioned above) were useful because overlapping results were observed.

    Table 4.

    Phenotypic differences between the species analysed

    Percentages of positive strains are tabulated; the percentages of strains with variable results are given in parentheses.

    DISCUSSION

    DNA–DNA reassociation is still the acknowledged standard for species delineation, although it has many technical problems restricting its application to few laboratories (Stackebrandt et al., 2002). Therefore, other techniques are required that can give additional and equally valid information on the classification of bacteria. AFLP has been proposed as an alternative to DNA–DNA hybridization to determine the taxonomic position of bacteria such as Burkholderia (Coenye et al., 1999), Aeromonas (Huys et al., 1996) and Xanthomonas (Rademaker et al., 2000). Significant correlations between DNA similarity values and these whole-genome analysis methods have been reported (Mougel et al., 2002). In this study, a significant correlation was found between DNA–DNA hybridization experiments and FAFLP and REP-PCR; DNA–DNA and FAFLP gave a correlation value of 0·761 (n=50), similar to the values found by Rademaker et al. (2000) of 0·838 (n=80) and by Thompson (2003) of 0·80 (n=234). Rep-PCR DNA fingerprinting has also been used to classify bacterial strains, and high levels of correlation with DNA–DNA hybridization have been obtained. Rademaker et al. (2000) compared BOX-, ERIC- and REP-PCR, and a combination of these (BER) with DNA–DNA hybridization, and obtained Pearson's product-moment values of 0·669, 0·779, 0·777 and 0·808, respectively. In this study, BOX-, REP- and (GTG)5-PCR were used with correlation values of 0·404, 0·449 and 0·680, respectively; these values are lower than those reported by Rademaker et al. (2000), but are still statistically significant. The differences could be due to the lower number of strains used in the correlations (n=50) in this study.

    It was suggested that a high number of fragments (bands) included in an analysis yields a higher correlation (Rademaker et al., 2000). In this study, although FAFLP had the highest number of bands (mean 102·4) and the highest correlation coefficient with DNA–DNA hybridization (0·761), no correlation could be established with the number of bands and hybridization values when all the methodologies were statistically analysed (Spearman correlation coefficient=0·083, P>0·05, n=4). It has been recommended that at least a minimum of 8–15 bands per sample (or lane) must be used for a rigorous comparative analysis (Versalovic et al., 1994). (GTG)5-PCR has been used very rarely, and never, to our knowledge, correlated with DNA–DNA hybridization. Using this method, Gevers et al. (2001) obtained a reliable identification of Lactobacillus species with a mean of 16·5 bands; they also found that combining data from BOX-, REP- and (GTG)5-PCR did not significantly enhance the discriminatory power compared to the increase in the amount of work needed. On the other hand, Nick et al. (1999) preferred a combined dendrogram because a maximized specificity of the patterns was obtained. (GTG)5-PCR has also been useful to identify and describe new species of bacteria such as Vibrio coralliitycus (Ben Haim et al., 2003), and Virgibacillus spp. (Heyrman et al., 2003).

    In our study, analysis of the IGS-PCR results produced an unrealistic dendrogram, which could be explained, in part, because of the low number of bands observed and the region analysed, which is only a small part of the genome, as compared to the other methods used. Apparently, the DNA sequence of these intergenic spacers can provide valuable information for the identification of many vibrios (Lee et al., 2002), although no specific sequences for V. campbellii could be found in the IGS types studied. Unfortunately, in the study of Lee et al. (2002) V. harveyi was not analysed, and therefore it is not known if differences in the DNA sequences are useful to differentiate these very closely related species.

    The evidence provided in this study and others strongly supports the use of AFLP (or FAFLP) and (GTG)5-PCR as alternative or supportive methods to DNA hybridization to delineate a species.

    V. harveyi and V. campbellii are genetically related species with a DNA–DNA similarity value of 69 % (Gomez-Gil et al., 2003) and a 16S rRNA similarity higher than 97 %. The ability of V. harveyi to produce ornithine decarboxylase is a key phenotypic feature to separate it from V. campbellii (Alsina & Blanch, 1994). Based on this and other characters many strains were identified as V. campbellii, although they shared characters with V. harveyi and thus were named V. campbellii-like (Hameed et al., 1996). The results obtained in this study reinforce the findings of Gauger & Gomez-Chiarri (2002) and of Thompson (2003) that phenotypic characters are not useful to differentiate these species. None of the characters could clearly discriminate the two species (Table 4⇑). Therefore it is recommended that molecular fingerprinting methods be used to identify these species, such as rep-PCR [preferably (GTG)5] or FAFLP.

    Sequencing of the 16S rRNA gene of some of the strains employed in this study (Gauger & Gomez-Chiarri, 2002), namely LMG 4044T, LMG 780, LMG 11216T, R-14905 and LMG 16828, and others, produced a dendrogram that supports the results of this study. A similar result was obtained by Pedersen et al. (1998) with some of the strains used here; all the strains identified as V. harveyi here clustered in the AFLP cluster 1 of Pedersen's work, while all the V. campbellii clustered in the AFLP groups 4 and 5. Thompson et al. (2001a) also analysed some of the strains used in this study, and found a similar arrangement, where V. harveyi strains formed a tight cluster (A36) and the V. campbellii strains were divided in two (A14 and A37); the delineation thresholds used by these authors were 50 % and 45 %, respectively, useful for V. harveyi strains but not for the more genetically diverse V. campbellii strains. For V. campbellii strains, a 10 % similarity threshold in the FAFLP dendrogram had to be used to include all the strains that were identified as this species based on the DNA hybridization studies (Fig. 1⇑). It was suggested that identification with FAFLP should be made when a similarity pattern of 60–70 % was adopted (Thompson, 2003), which corresponds to a DNA hybridization value of 70 %; clearly, this threshold can be used for the majority of the Vibrio species. The subdivision of the V. campbellii strains into many clusters could be explained by the great heterogeneity of this species.

    Of particular interest was strain LMG 20977 (=AK2), which was identified as V. shilonii by Kushmaro et al. (2001) due to being phenotypically identical to the type strain AK1T (=LMG 19703T). Later, Thompson et al. (2001b) corrected the identification to V. mediterranei but considering only the type strain and not LMG 20977. In the present study, strain LMG 20977 was grouped, with all methods except IGS-PCR, in the V. harveyi cluster. The FAFLP similarity between LMG 20977 and the V. harveyi type strain (LMG 4044T) was 66·33 % and with the V. campbellii type strain (LMG 11216T) only 49·08 %. It is not clear if strain LMG 20977 is pathogenic for corals (Kushmaro et al., 2001), but if so, then V. harveyi should also be considered as pathogenic for corals, and many other marine organisms.

    The present study highlights the inadequacy of dichotomous keys (Alsina & Blanch, 1994) for the identification of V. harveyi-related species. Presumptive V. harveyi isolates associated with diseased organisms (see Soto-Rodríguez et al., 2003) analysed in this study turned out to be V. campbellii on the basis of both molecular fingerprinting and DNA–DNA hybridization, suggesting that V. campbellii may be an important pathogenic species of aquatic organisms (Gauger & Gomez-Chiarri, 2002; Soto-Rodríguez et al., 2003).

    Acknowledgments

    We thank Margo Cnockaert and Cristiane C. Thompson for their technical support, and Celia Lavilla-Pitogo and Leonardo Lizárraga-Partida for donating some of the strains. We acknowledge financial support from CONACYT project J-28344, FWO grants, and CONACYT sabbatical grant 010175 to B. G.

    References

    1. Abraham, T. J., Palaniappan, R. & Dhevendaran, K. (1999). Simple taxonomic key for identifying marine luminous bacteria. Indian J Mar Sci 28, 35–38.
    2. Alsina, M. & Blanch, A. R. (1994). A set of keys for biochemical identification of environmental Vibrio species. J Appl Bacteriol 76, 79–85.
    3. Austin, B. (1988). Marine Microbiology. Cambridge, UK: Cambridge University Press.
    4. Ben Haim, Y., Thompson, F. L., Thompson, C. C., Cnockaert, M. C., Hoste, B., Swings, J. & Rosenberg, E. (2003). Vibrio coralliilyticus sp. nov., a temperature-dependent pathogen of the coral Pocillopora damicornis. Int J Syst Evol Microbiol 53, 309–315.
    5. Chun, J., Huq, A. & Colwell, R. R. (1999). Analysis of 16S–23S rRNA intergenic spacer regions of Vibrio cholerae and Vibrio mimicus. Appl Environ Microbiol 65, 2202–2208.
    6. Coenye, T., Schouls, L. M., Govan, J. R. W., Kersters, K. & Vandamme, P. (1999). Identification of Burkholderia species and genomovars from cystic fibrosis patients by AFLP fingerprinting. Int J Syst Evol Microbiol 49, 1657–1666.
    7. Cowan, S. T., Steel, K. J., Barrow, G. I. & Feltham, R. K. A. (1993). Cowan and Steel's Manual for the Identification of Medical Bacteria, 3rd edn. Cambridge: Cambridge University Press.
    8. Dombek, P. E., Johnson, L. K., Zimmerley, S. T. & Sadowsky, M. J. (2000). Use of repetitive DNA sequences and the PCR to differentiate Escherichia coli isolates from human and animal sources. Appl Environ Microbiol 66, 2572–2577.
    9. Dorsch, M., Lane, D. & Stackebrandt, E. (1992). Towards a phylogeny of the genus Vibrio based on 16S rRNA sequences. Int J Syst Bacteriol 42, 58–63.
    10. Farmer, J. J. & Hickman-Brenner, F. W. (1992). The genera Vibrio and Photobacterium. In The Prokaryotes – a Handbook on the Biology of Bacteria: Ecophysiology, Isolation, Identification, Applications, pp. 2952–3011. Edited by A. Balows. New York: Springer.
    11. Gauger, E. J. & Gomez-Chiarri, M. (2002). 16S ribosomal DNA sequencing confirms the synonymy of Vibrio harveyi and V. carchariae. Dis Aquat Organ 52, 39–46.
    12. Gevers, D., Huys, G. & Swings, J. (2001). Applicability of rep-PCR fingerprinting for identification of Lactobacillus species. FEMS Microbiol Lett 205, 31–36.
    13. Gherna, L. R. (1994). Culture preservation. In Methods for General and Molecular Bacteriology, pp. 278–292. Edited by P. Gerhardt, R. G. E. Murray, W. A. Wood & N. R. Krieg. Washington, DC: American Society for Microbiology.
    14. Gomez-Gil, B., Thompson, F. L., Thompson, C. C. & Swings, J. (2003). Vibrio rotiferianus sp. nov., isolated from cultures of the rotifer Brachionus plicatilis. Int J Syst Evol Microbiol 53, 239–243.
    15. Gurtler, V. & Mayall, B. C. (2001). Genomic approaches to typing, taxonomy and evolution of bacterial isolates. Int J Syst Evol Microbiol 51, 3–16.
    16. Hameed, A. S. S., Rao, P. V., Farmer, J. J., Brenner, F. W. H. & Fanning, G. R. (1996). Characteristics and pathogenicity of a Vibrio campbellii like bacterium affecting hatchery reared Penaeus indicus (Milne Edwards, 1837) larvae. Aquaculture Res 27, 853–863.
    17. Heyrman, J., Logan, N. A., Busse, H. J., Balcaen, A., Lebbe, L., Rodriguez-Diaz, M., Swings, J. & De Vos, P. (2003). Virgibacillus carmonensis sp. nov., Virgibacillus necropolis sp. nov. and Virgibacillus picturae sp. nov., three novel species isolated from deteriorated mural paintings, transfer of the species of the genus Salibacillus to Virgibacillus, as Virgibacillus marismortui comb. nov. and Virgibacillus salexigens comb. nov., and emended description of the genus Virgibacillus. Int J Syst Evol Microbiol 53, 501–511.
    18. Holt, J. G., Krieg, N. R., Sneath, P. H. A., Staley, T. & Williams, S. T. (1994). Bergey's Manual of Determinative Bacteriology, 9th edn. Baltimore, MD: Williams & Wilkins.
    19. Huys, G., Coopman, R., Janssen, P. & Kersters, K. (1996). High-resolution genotypic analysis of the genus Aeromonas by AFLP fingerprinting. Int J Syst Evol Microbiol 46, 572–580.
    20. Ishimaru, K., Akagawa-Matsushita, M. & Muroga, K. (1995). Vibrio penaeicida sp. nov., a pathogen of kuruma prawns (Penaeus japonicus). Int J Syst Bacteriol 45, 134–138.
    21. Karunasagar, I., Pai, R. & Malathi, G. R. (1994). Mass mortality of Penaeus monodon larvae due to antibiotic-resistant Vibrio harveyi infection. Aquaculture 128, 203–209.
    22. Kushmaro, A., Banin, E., Loya, Y., Stackebrandt, E. & Rosenberg, E. (2001). Vibrio shiloi sp. nov., the causative agent of bleaching of the coral Oculina patagonica. Int J Syst Evol Microbiol 51, 1383–1388.
    23. Lavilla-Pitogo, C. R., Baticados, M. C. L., Cruz-Lacierda, E. R. & Pena, L. D. (1990). Occurrence of luminous bacterial disease of Penaeus monodon larvae in the Philippines. Aquaculture 91, 1–13.
    24. Lavilla-Pitogo, C. R., Leano, E. M. & Paner, M. G. (1998). Mortalities of pond-cultured juvenile shrimp, Penaeus monodon, associated with dominance of luminescent vibrios in the rearing environment. Aquaculture 164, 337–349.
    25. Lee, S. K., Wang, H. Z., Law, S. H., Wu, R. S. & Kong, R. Y. (2002). Analysis of the 16S–23S rDNA intergenic spacers (IGSs) of marine vibrios for species-specific signature DNA sequences. Mar Pollut Bull 44, 412–420.
    26. Lightner, D. V. (1993). Diseases of cultured penaeid shrimp. In CRC Handbook of Mariculture, Crustacean Aquaculture, pp. 393–486. Edited by J. P. Mcvey. Boca Raton, FL: CRC Press.
    27. Mougel, C., Thioulouse, J., Perriere, G. & Nesme, X. (2002). A mathematical method for determining genome divergence and species delineation using AFLP. Int J Syst Evol Microbiol 52, 573–586.
    28. Nagpal, M. L., Fox, K. F. & Fox, A. (1998). Utility of 16S–23S rRNA spacer region methodology: how similar are interspace regions within a genome and between strains for closely related organisms? J Microbiol Methods 33, 211–219.
    29. Nick, G., Jusilla, M., Hoste, B., Niemi, R. M., Kaijalainen, S., de Lajudie, P., Gillis, M., de Bruijn, F. J. & Lindstrom, K. (1999). Rhizobia isolated from root nodules of tropical leguminous trees characterized using DNA–DNA dot-blot hybridisation and rep-PCR genomic fingerprinting. Syst Appl Microbiol 22, 287–299.
    30. Pedersen, K., Verdonck, L., Austin, B. & 9 other authors (1998). Taxonomy evidence that Vibrio carchariae Grimes et al., 1985 is a junior synonym of Vibrio harveyi (Johnson and Shunk 1936) Baumann et al., 1981. Int J Syst Bacteriol 48, 749–758.
    31. Pitcher, D. G., Saunders, N. A. & Owen, R. J. (1989). Rapid extraction of bacterial genomic DNA with guanidium thiocyanate. Lett Appl Microbiol 8, 151–156.
    32. Prayitno, S. B. & Latchford, J. W. (1995). Experimental infections of crustaceans with luminous bacteria related to Photobacterium and Vibrio – effect of salinity and pH on infectiosity. Aquaculture 132, 105–112.
    33. Rademaker, J. L. W., Louws, F. J. & de Bruijn, F. J. (1998). Characterization of the diversity of ecologically important microbes by rep-PCR genomic fingerprinting. In Molecular Microbial Ecology Manual, pp. 1–27. Edited by A. D. L. Akkermans, J. D. van Elsas & F. J. de Bruijn. Dordrecht: Kluwer.
    34. Rademaker, J. L., Hoste, B., Louws, F. J., Kersters, K., Swings, J., Vauterin, L., Vauterin, P. & de Bruijn, F. J. (2000). Comparison of AFLP and rep-PCR genomic fingerprinting with DNA–DNA homology studies: Xanthomonas as a model system. Int J Syst Evol Microbiol 50 Pt 2, 665–677.
    35. Rivera, I. G., Chowdhury, M. A., Huq, A., Jacobs, D., Martins, M. T. & Colwell, R. R. (1995). Enterobacterial repetitive intergenic consensus sequences and the PCR to generate fingerprints of genomic DNAs from Vibrio cholerae O1, O139, and non-O1 strains. Appl Environ Microbiol 61, 2898–2904.
    36. Roque, A., Turnbull, J. F., Escalante, G., Gomez-Gil, B. & Alday-Sanz, M. V. (1998). Development of a bath challenge for the marine shrimp Penaeus vannamei Boone, 1931. Aquaculture 169, 283–290.
    37. Sawabe, T., Setogushi, N., Inoue, S., Tanaka, R., Ootsubo, M., Yoshimizu, M. & Ezura, Y. (2003). Acetic acid production of Vibrio halioticoli from alginate: a possible role for establishment of abalone–V. halioticoli association. Aquaculture 219, 671–679.
    38. Soto-Rodríguez, S., Roque, A., Lizarraga-Partida, M. L., Guerra-Flores, A. L. & Gomez-Gil, B. (2003). Virulence of luminous vibrios to Artemia franciscana nauplii. Dis Aquat Organ 53, 231–240.
    39. Stackebrandt, E., Frederiksen, W., Garrity, G. M. & 10 other authors (2002). Report of the ad hoc committee for the re-evaluation of the species definition in bacteriology. Int J Syst Evol Microbiol 52, 1043–1047.
    40. Thompson, F. L. (2003). Improved taxonomy of the family Vibrionaceae. PhD thesis, Ghent University.
    41. Thompson, F. L., Hoste, B., Vandemeulebroecke, K. & Swings, J. (2001a). Genomic diversity amongst Vibrio isolates from different sources determined by fluorescent amplified fragment length polymorphism. Syst Appl Microbiol 24, 520–538.
    42. Thompson, F. L., Hoste, B., Thompson, C. C., Huys, G. & Swings, J. (2001b). The coral bleaching Vibrio shiloi Kushmaro et al. 2001 is a later synonym of Vibrio mediterranei Pujalte and Garay 1986. Syst Appl Microbiol 24, 516–519.
    43. Vandenberghe, J., Thompson, F. L., Gomez-Gil, B. & Swings, J. (2003). Phenotypic diversity amongst Vibrio isolates from marine aquaculture systems. Aquaculture 219, 9–20.
    44. Venkateswaran, K., Dohmoto, N. & Harayama, S. (1998). Cloning and nucleotide sequence of the gyrB gene of Vibrio parahaemolyticus and its application in detection of this pathogen in shrimp. Appl Environ Microbiol 64, 681–687.
    45. Versalovic, J., Koeuth, T. & Lupski, J. R. (1991). Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Res 19, 6823–6831.
    46. Versalovic, J., Schneider, M., de Bruijn, F. J. & Lupski, J. R. (1994). Genomic fingerprinting of bacteria using repetitive sequence-based polymerase chain reaction. Methods Mol Cell Biol 5, 25–40.
    47. Vinuesa, P., Rademaker, J. L., de Bruijn, F. J. & Werner, D. (1998). Genotypic characterization of Bradyrhizobium strains nodulating endemic woody legumes of the Canary Islands by PCR-restriction fragment length polymorphism analysis of genes encoding 16S rRNA (16S rDNA) and 16S–23S rDNA intergenic spacers, repetitive extragenic palindromic PCR genomic fingerprinting, and partial 16S rDNA sequencing. Appl Environ Microbiol 64, 2096–2104.
    48. Wayne, L. G., Brenner, D. J., Colwell, R. R. & 8 other authors (1987). Report of the ad hoc committee on reconciliation of approaches to bacterial systematics. Int J Syst Bacteriol 37, 463–464.
    49. West, P. A. & Colwell, R. R. (1984). Identification and classification of Vibrionaceae – an overview. In Vibrios in the Environment, pp. 285–363. Edited by R. R. Colwell. New York: Wiley.
    50. Willems, A., Doignon-Bourcier, F., Goris, J., Coopman, R., de Lajudie, P., De Vos, P. & Gillis, M. (2001). DNA–DNA hybridization study of Bradyrhizobium strains. Int J Syst Evol Microbiol 51, 1315–1322.
    51. Yang, Y. K., Yeh, L. P., Cao, Y. H., Baumann, L., Baumann, P., Tang, J. E. & Beaman, B. (1983). Characterization of marine luminous bacteria isolated off the coast of China and description of Vibrio orientalis sp. nov. Curr Microbiol 8, 95–100.