INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Vibrio cholerae is a waterborne pathogen that causes gastrointestinal disorders with a wide range of clinical manifestations, including vomiting and rice-like diarrhea (24). The association of human illness with consumption of V. cholerae-contaminated oysters, seawater, and other shellfish is well documented (29, 37). Consumption of raw oysters correlates strongly with gastrointestinal infections, and several Vibrio species, including strains of Vibrio parahaemolyticus, Vibrio vulnificus, and V. cholerae, have been implicated as the causative agents.

Coastal areas with brackish waters and estuarine regions are niches for many Vibrio species, including strains of toxigenic O1 V. cholerae. Epidemic cholera strains are endemic in several regions, including the U.S. gulf coast and Australia, and are occasionally involved in illnesses in these regions (24). Because Vibrio species attach to material suspended in water, shellfish and mollusks that are in these environments can be expected to consume Vibrio during feeding (24).

Traditional identification methods currently used are time-consuming and laborious, requiring prolonged incubation and selective enrichment to reduce the growth of background flora. Vibrio cells may also enter a viable but not culturable (VBNC) state, caused by nutrient starvation and physical stress. This may explain the failure of traditional culture techniques to isolate this organism from contaminated water and food samples implicated in food-borne outbreaks (10, 27, 47). Several investigators have developed PCR and DNA probe techniques for the detection of pathogenic Vibrio species (7, 12, 14, 19, 23, 25, 40, 41, 44, 46; S. C. Arya, Letter, J. Clin. Microbiol. 35:3364, 1997). DNA-based methods such as PCR have been increasingly used for rapid, sensitive analysis, but are nonquantitative in their detection of V. cholerae cells (7, 12, 14, 19, 23, 25, 34, 40, 41, 44, 46; Arya, Letter). The presence of PCR products must also be verified by subsequent procedures such as gel electrophoresis and Southern hybridization. All of these additional steps are time-consuming and laborious, and the additional procedures add to the overall cost of the test. The TaqMan assay utilizes the 5'-exonuclease activity of Thermus aquaticus DNA polymerase (17, 21, 35, 43; K. J. Livak, L. Marmaro, and S. J. A. Flood, Perkin-Elmer Research News, p. 1-12, Perkin-Elmer Corp., Norwalk, Conn., 1995) to hydrolyze an internal TaqMan probe labeled with a fluorescent reporter dye (FAM-6-carbooxyfluorescein) and a quencher dye (TAMRA-6-carboxy-N,N,N',N'-tetramethylrhodamine) (User bulletin 2, ABI Prism 7700 Sequence Detection System, PE Applied Biosystems, Foster City, Calif., 1997). The probe is designed to hybridize to the DNA sequence between the PCR primers (User bulletin 2, PE Applied Biosystems). During PCR amplification, cleavage of the TaqMan probe separates the reporter dye and quencher dye, which results in increased fluorescence. In contrast, when the probe is intact, the proximity of the reporter dye to the quencher dye results in blockage of the reporter fluorescence (User bulletin 2, PE Applied Biosystems). TaqMan PCR eliminates the need for subsequent PCR product verification that is required by other PCR amplifications, thereby reducing the amount of time needed for sample analysis.

Several investigators have discussed the major obstacles encountered with the current DNA-based tests. These obstacles include the separation of culturable, VBNC, and dead microorganisms (30, 33). In addition to culturable versus VBNC versus dead cells, PCR inhibition is a major problem, especially in samples from the environment or food (30, 33). The nucleic acids in living cells are protected because the cell walls and membranes are intact. In dead cells, the cell membranes are compromised, and the nucleic acids are thus exposed to compounds added to the sample (30, 33). The exposure of DNA in dead cells may be utilized to destabilize or inactivate the nucleic acids, while the nucleic acids within living cells are protected from the treatment by the cell membrane and wall. Nogva et al. (30) also addressed the fact that bacterial DNases may cleave exposed DNA and determined that addition of DNase to the samples prior to DNA extraction did not have any effect on intact live cells, but did result in a 1-log reduction in the cell counts (i.e., dead cells) compared to that of control samples not treated with DNase.

Another major concern with VBNC cells is the potential for reversion to a metabolic state that can cause disease in humans after consumption of VBNC cells. Colwell et al. (10, 11) first reported that V. cholerae cells rendered nonculturable by incubation in estuarine water retained virulence when tested in rabbit ligated ileal loop assays and could cause disease when fed to human volunteers. Therefore, it is important that VBNC cells are enumerated and taken into consideration when evaluating foods that could be potentially contaminated with VBNC Vibrio cells.

Several investigators have developed TaqMan assays for detection of Salmonella species, Listeria monocytogenes, Yersinia enterocolitica, Campylobacter jejuni, Escherichia coli O157:H7, and Shiga-like toxin genes in various foods (5, 9, 30, 31, 45). Among the various quantitative PCR strategies available, those based on real-time monitoring of the amplification reactions are the most accurate (20, 21, 35; Livak et al., Perkin-Elmer Research News). However, a real-time assay is still not available for V. cholerae. TaqMan PCR assays are also desirable for rapid analysis of foods that are consumed raw without further processing. Rapid methods that can quickly address the presence of Vibrio in fresh raw products such as oysters would be beneficial in eliminating release of contaminated products into the marketplace, which could result in potential food-borne outbreaks.

An elegant paper written by Singh et al. (42) evaluated the virulence of V. cholerae O1, O139, non-O1, and non-0139 strains isolated from various environmental sources. They found that environmental V. cholerae strains do not possess the virulence gene cassette (26) that contains the genes that encode the cholera toxin (ctx), zonula occludens toxin (zot), and accessory cholera toxin (ace). Furthermore, environmental vibrios lack the gene coding for toxin-coregulated pilus (tcp), which is recognized as a virulence factor in the pathogenicity of toxigenic V. cholerae serotypes O1 and O139 (26, 42). The genes coding for cholera toxin, the colonization toxin-coregulated pilus, and central regulatory protein (toxR) are required by virulent strains of V. cholerae O1 and O139 for production of the diarrheal syndrome (16, 18, 26, 27). In contrast, V. cholerae non-O1 and non-O139 strains lack the virulence genes (i.e., ctx, zot, and ace); however, these strains do contain genes that encode other products, such as NAG-specific heat-stable toxin (st), thermostable direct hemolysin, shigella-like toxin, and the classical hemolysin (4, 15, 32, 42, 48). The presence of these toxins has been correlated with the disease symptoms produced by V. cholerae non-O1 and non-O139 strains (22, 24). Singh et al. (42) determined that both environmental and clinical V. cholerae O1, O139, non-O1, and non-O139 strains contain the nonclassical fragment of the hemolysin (hlyA) gene. We further evaluated the hlyA gene for the feasibility of using this gene in the development of a TaqMan PCR for V. cholerae strains.

This paper describes the development and evaluation of a primer and probe system that is rapid, sensitive, and quantitative for V. cholerae cells in pure cultures, seawater, and raw oysters. The primers and probe are directed towards the nonclassical specific hlyA gene of V. cholerae O1, O139, non-O1, and non-O139 strains. This primer-probe set can be used in the quantification of V. cholerae in the presence of other contaminating Vibrio species. A TaqMan primer and probe system that is a rapid screening tool for the presence of V. cholerae in oysters and seawater would eliminate the need for isolation and characterization by classical microbiological methods.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains, media, and cultures. Sixty-seven strains of Vibrio were used to test the specificity of the primers and the probe (Table 1). The isolates were collected from patients and food products within the United States, India, Peru, and England. Sixty bacterial strains not belonging to the genus Vibrio were used to evaluate the specificity of the TaqMan primers and probe developed in this study. A complete list of bacterial strains that were evaluated is given in Tables 1 and 2.

View this table: [in this window] [in a new window]   TABLE 1.   Vibrio strains evaluated and RQ values generated in PCR with TaqMan primers designed to detection V. cholerae strains

View this table: [in this window] [in a new window]   TABLE 2.   Bacterial strains evaluated with TaqMan primers and a probe specifically designed for detection of V. cholerae DNA and RQ values generated during PCR amplification

DNA isolation. Three 1.0-ml samples (1.0 ml, 10⁷ CFU ml¹) of 18-h culture were centrifuged at 6,000 × g for 5 min, and the supernatants were discarded. Cell pellets were resuspended with 200 µl of Dynabeads DNA Direct I solution (Dynal AS, Oslo, Norway). The bacterium-bead suspensions were incubated at 65°C for 20 min, followed by incubation at room temperature for another 2 min. DNA bound to magnetic beads was then drawn to the wall of the microcentrifuge tube by a magnet (MPC-E; Dynal AS) for 2 min. The supernatant containing salts, detergent, and cell debris was carefully removed without disrupting the Dynabead-DNA complex. The beads were washed twice with a wash buffer provided in the kit. The DNA was removed from the beads by resuspension of the bead-DNA complex in 20 µl of 10 mM Tris HCl (pH 8.0). The bead complexes were incubated at 65°C for 5 min to release the DNA from the beads. The beads were collected with the magnet, the DNA-containing supernatant was transferred to a fresh tube, and 2.5 µl of the DNA suspension was used directly in PCRs. The DNA extraction efficiency was determined by the method described by Heid et al. (17). DNA was extracted from three different 1.0-ml aliquots of each sample, the and the DNA concentrations were determined with a Hoefer DyNa Quant 200 fluorometer as described by the manufacturer (Amersham Pharmacia Biotech, Piscataway, N.J.).

TaqMan probe and primer design. The probe region used in this study was localized within a sequence region coding for the nonclassical region of hlyA (42) shown in Table 3. BLAST N, BLAST P, and BLAST X database searches (2) were done, and the primers and probes were designed within a region that had no homology with other known proteins in the database. The Primer Express (version l.5) ABI Prism system (PE Applied Biosystems, Foster City, Calif.) was used for the primer-probe design, together with guidelines from PE Applied Biosystems (User bulletin 2, PE Applied Biosystems; Livak et al., Perkin-Elmer Research News).

TaqMan assay probe for detection of V. cholerae. Amplification reaction mixtures (50 µl) contained a DNA sample (2.5 µl, 100 ± 0.08 ng/µl); 1× TaqMan buffer A; 5 mM MgCl₂; 200 µM (each) dATP, dCTP, and dGTP; 400 µM dUTP; 0.02 µM V. cholerae-specific fluorogenic probe; 0.3 µM (each) V. cholerae-specific primers; 1 U of AmpErase uracil N-glycosylase; and 2.5 U of AmpliTaq Gold DNA polymerase. All PCR samples and controls were prepared in triplicate by using 0.2 ml of MicroAmp Optical reaction tubes and MicroAmp Optical tube caps (PE Applied Biosystems).

Specificity studies with pure cultures. Genomic DNA was isolated from 67 Vibrio strains (Table 1) and 60 other bacteria (Table 2). PCR was run with the DNA from each strain to determine the specificity of the TaqMan probe for V. cholerae species. Detection sensitivity studies were done to identify the lower detection limit of the TaqMan PCR analyses. V. cholerae O1 and O139 (ATCC 14035 and ATCC 51394, respectively) were incubated at 37°C in BHI broth containing 3% (wt/vol) NaCl until the mid-logarithmic growth phase (10⁷ CFU ml¹). Three 1.0-ml samples were taken from each of the cultures, serially diluted (10-fold to 10⁷) in SSW, and enumerated on BHI broth supplemented with 3% NaCl. Three 1.0-ml samples were taken from each serial dilution; the cells were pelleted by centrifugation (6,000 × g for 5 min), and the DNA was extracted from each cell pellet.

1

7

DNA extraction from VBNC V. cholerae cells. Vibrio species are known to exist in a VBNC state when subjected to stress (i.e., storage at cold temperature in the presence of NaCl). The following experiments were done to evaluate whether the resuscitation of VBNC cells was necessary prior to DNA extraction and to determine if the DNA extraction procedure was effective in extracting DNA from VBNC cells. DNase was also added to samples to eliminate the contribution of DNA from dead cells during cell quantification.

1

Sensitivity studies with pure cultures spiked into raw oysters and SSW. Freshly harvested shellstock oysters obtained from three different docks in Louisiana were bagged in polyethylene bags, transported on ice to the Louisiana Department of Agriculture & Forestry, Louisiana State University Rapid Microbial Detection Laboratory, stored at refrigeration temperature (5°C), and subjected to microbiological analysis within 4 to 8 h. Fresh shellstock oysters were scrubbed with a brush under running tap water to remove mud and debris from the shells prior to shucking. V. cholerae serotype O1 (ATCC 14035) or V. cholerae serotype O139 (ATCC 51394) cultures were grown overnight in BHI, serially diluted (10-fold) in SSW, and enumerated on BHI supplemented with 3% (wt/vol) NaCl. Twenty-five grams of oysters was spiked with either V. cholerae serotype O1 (6.2 × 10⁷ CFU) or V. cholerae serotype O139 (6.7 × 10⁷ CFU), placed into a sterile stomacher bags (Seward, Inc.; model 400 filter bags), and homogenized with 225 ml of APW for 2 min in a stomacher (Tekmar, Inc.). One-milliliter samples from of the oyster homogenates were serially diluted 10-fold in APW. One-milliliter samples were taken from each of the APW dilutions, the cells were pelleted by centrifugation (6,000 × g for 5 min), and the DNA was extracted from the cells.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

TaqMan PCR specificity. Specific PCR primers and a probe were designed for detection of V. cholerae. The probe region was chosen to optimize specificity and amplification efficiency. The putative primers and probe were constructed with the primer express program, and then these DNA sequences were subjected to BLAST N, BLAST X, and BLAST P database searches (2) to find any sequence similarities. A 70-bp chromosomal DNA fragment of the nonclassical hlyA gene was found to be unique for all V. cholerae O1, O139, non-O1, and non-O139 strains tested. There were no known DNA sequences or protein sequences in the BLAST N, BLASTX, and BLAST P databases with homology to this primer-probe region for Vibrio cholerae O1, O139, non-O1, and non-O139 species.

Effect of externally added DNase to samples during DNA extraction. The effect of externally added DNase was compared to that in control samples to which no DNase was added to the cell pellets prior to extraction of the DNA. Sixteen-hour cultures of V. cholerae serotype O1 (ATCC 14035; 7.2 × 10⁷ CFU ml¹) and V. cholerae O139 (ATCC 51394; 6.4 × 10⁷ CFU ml¹) were serially diluted 10-fold in SSW, and DNA was extracted from 1.0 ml of cells for each Vibrio culture. DNA (2.5 µl) extracted from these cells was amplified by using the TaqMan PCR assay as described earlier. There was a 1-log reduction in CFU per milliliter in DNase-treated samples when compared to the non-DNase-treated control samples (Table 4). DNA extracted from V. cholerae O1 cells (ATCC 14035) that were treated with DNase had an average DNA concentration of 90 ± 5 ng/µl, and the DNA samples from cells not treated with DNase had a DNA concentration of 108 ± 3 ng/µl (Table 4). The RQ values for samples that received DNase treatment were 37.5, 37.8, and 37.0, and samples that did not receive DNase treatment had RQ values of 35.2, 35.4, and 35.8 (Table 4). Similar results were obtained with V. cholerae O139 (ATCC 51394) cells. The DNA concentration of DNase-treated cells was 89 ± 2 ng/µl, and the DNA concentrations of cells not treated with DNase was 100 ± 4 ng/µl (Table 4). The RQ values for samples that received DNase treatment were 35.2, 35.5, and 35.4, and samples that did not receive DNase treatment had RQ values of 37.0, 37.2, and 37.4 (Table 4). TaqMan PCR sensitivity was able to quantify a 1-log reduction in cells; this was indicated by an increase in the RQ values for both V. cholerae O1 and O139 serotypes (Table 4).

Sensitivity studies with pure cultures of V. cholerae. Sensitivity studies were performed with pure cultures of V. cholerae serotypes O1 (ATCC 14035; 6.2 × 10⁷ CFU ml¹) and O139 (ATCC 51394; 6.7 × 10⁷ CFU ml¹) to test the lower detection limit of the TaqMan PCR. All V. cholerae strains tested gave a positive reaction with the RQ values above the threshold of 2.80 (four times the average RQ of the no-template controls; Table 1) (39, 45). Other bacterial and non-V. cholerae species tested gave a negative interpretation, with RQ values below the threshold of 2.80 (Table 1). These results are similar to the data described by Ibrahim et al. and Vishnubhatla et al. (20, 45). The sensitivity curves based on the dilutions of V. cholerae O1 (ATCC 14035) cells and V. cholerae O139 (ATCC 51394) cells are shown in Fig. 1. Division of the lowest dilution of the culture that gave a positive reaction by the approximate final volume of the DNA extraction (CFU per microliter) used in the reaction gave a lower detection limit of >7.0 CFU ml of reaction mixture¹. In our assay, the lowest detection limits were 7.3 and 8.2 CFU ml of reaction mixture¹ for V. cholerae serotype O1 (ATCC 14035) and V. cholerae O139 (ATCC 51394), respectively.

View larger version (15K): [in this window] [in a new window]   FIG. 1.   Sensitivity of the fluorogenic TaqMan PCR assay for detection of pathogenic 10-fold dilutions of V. cholerae serotype O1 (; ATCC 14035) and O139 (; ATCC 51394) in SSW. Experiments were done in triplicate. DNA was amplified with the hlyA TaqMan primers in the presence of the hlyA fluorogenic probe. The average RQ values determined from DNA recovered from both serotypes were plotted against the CFU per milliliter. The average RQ threshold values at four times the average of no-template controls were calculated to be 2.80 and 2.83 for V. cholerae O1 and O139, respectively (line across the graph). Error bars indicate the standard deviations of the means.

Sensitivity of the TaqMan assay for detection of DNA extracted from VBNC V. cholerae cells. V. cholerae cells (O1, ATCC 14035; O139, ATCC 51394) were inoculated in SSW and stored at 5°C. Refrigerated V. cholerae cultures were sampled at days 0 and 7. For V. cholerae O1 cultures, there were 3.8 × 10⁷ and <0.1 CFU ml at days 0 and 7, respectively. For V. cholerae O139 (ATCC 51394) cultures, there were 2.8 × 10⁷ CFU ml and <0.1 CFU g at days 0 and 7, respectively. DNA was extracted from three 1.0-ml samples of each culture at days 0 and 7. TaqMan PCR analysis was run on all of the DNA samples, and the data are shown in Table 5. Each of the DNA preparations (2.5 µl; 100 ± 5 ng µl¹) was analyzed in triplicate, and the experiment was repeated three times (Table 5). These data indicate that the Vibrio cells induced by exposure to NaCl and storage at 5°C were in a VBNC state and that the amounts of DNA isolated were similar to that of day 0. VBNC Vibrio cells were not detected by traditional culturing methods when plated immediately after cold storage (Table 5). However, the VBNC cells were resuscitated on filters placed on HI plates after an incubation of 25 h at 25°C. As expected, these results are similar to the results found by other investigators (27, 47), indicating that when Vibrio strains are placed under physiochemical stresses, they are induced into a VBNC state and can be resuscitated by raising the temperature to 25°C. VBNC V. cholerae cells were detected by using the TaqMan PCR assay (Table 5). The RQ values were similar to that of Vibrio cells that had not been induced into a VBNC state (Table 5).

Sensitivity studies with spiked raw oysters. Raw oysters obtained from three different docks in Louisiana were confirmed to be culture negative for V. cholerae. Twenty-five-gram samples were spiked with 6.2 × 10⁷ CFU of V. cholerae serotype O1 (ATCC 14035). The presence of oyster homogenate did not affect the ability to isolate comparable amounts of Vibrio DNA. The average amounts of DNA isolated from three 1.0-ml samples of pure cultures of V. cholerae O1and O139 were 100 ± 5 and 102 ± 3 ng µl¹, respectively. The average amounts of DNA isolated from three 1.0-ml samples of spiked oyster homogenate were105 ± 2 and 103 ± 4 ng µl¹ for V. cholerae O1 and O139, respectively (data not shown). The contribution of the DNA from the oyster cells was determined by serially diluting the nonspiked oyster homogenate 10-fold in SSW. The DNA concentrations were 0.09 ± 0.1 and 0.001 ± 0.003 ng µl¹ in the 1:10 and 1:100 dilutions, respectively (data not shown). The oyster cell DNA was considered to be an additional control and was run in triplicate with each PCR run. The RQ values for oyster homogenates were 0.30, 0.28, and 0.32. The average RQ value for oyster homogenates was 0.32; this value was subtracted from amplification RQ values obtained for each Vibrio DNA dilution prior to plotting, which was the RQ (Fig 2A).

View larger version (14K): [in this window] [in a new window]   FIG. 2.   Sensitivity of the fluorogenic TaqMan PCR assay for detection of pathogenic V. cholerae O1 and O139 in spiked oyster samples (A) and spiked SSW (B). Tenfold dilutions of V. cholerae serotype O1 (; ATCC 14035) and O139 (; ATCC 51394) in SSW were done in triplicate. DNA was amplified with the hlyA TaqMan primers in the presence of the hlyA fluorogenic probe. The average RQ values determined from DNA recovered from both serotypes were plotted against either the CFU per gram or CFU per milliliter. Delta RQ average threshold values are defined as four times the average of no-template controls and are shown for spiked oyster samples (A) and SSW samples (B) as 3.3 and 2.5 (line across the graphs), respectively. Error bars indicate the standard deviations of the means.

1

1

Sensitivity of the TaqMan PCR for isolation of V. cholerae cells from SSW. Sensitivity studies were performed with pure cultures of V. cholerae serotype O1 to test the lower detection limit of the TaqMan PCR. Division of the lowest dilution of the culture that gave a positive reaction by the approximate final volume of the DNA extraction (CFU per microliter) used in the reaction gave a lower detection limit of approximately 10 CFU ml of reaction mixture¹ with a threshold RQ of 2.50 for V. cholerae serotype O1 cultures (ATCC 14035; 6.2 × 10⁷ CFU) spiked into SSW (Fig. 2B). Similar results were obtained with V. cholerae O139 (ATCC 51394; 6.8 × 10⁷ CFU), which had a detection limit of 10 CFU ml¹ in spiked SSW. The reproducibility was good among three experiments, with RQ threshold values of 2.50, 2.46, and 2.53 and 2.50, 2.46, and 2.56 for V. cholerae O1 and V. cholerae O139, respectively. Theron et al. (44) obtained a comparable detection level of 1 CFU ml¹ for V. cholerae in environmental water samples.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

There is a requirement for rapid, quantitative, and accurate measurements of target organisms responsible for food poisoning. In the present study, a TaqMan PCR system was constructed and applied to specifically detect and quantify V. cholerae strains. The preferred targets for pathogen detection are pathogen virulence genes. However, the need for a specific probe that targets all V. cholerae species would be valuable because of the variation between the different V. cholerae serotypes. For example, the ctx genes are expressed in some of the serotypes of V. cholerae O1/O139 and some strains of non-O1/O139, but not at all in other non-O1/O139 strains (10, 27, 47). Therefore, it appears that the virulence genes (ctx, zot, and ace) would be unsuitable for the design of primers and probes for detection of V. cholerae strains. The sequence region of the nonclassical hlyA region (70 bp) in V. cholerae is specific for this organism and not other Vibrio species or other bacterial genera. The 70-bp target region within the nonclassical hlyA region was compared with the most recently published sequences in the National Center for Biotechnology Information database (BLAST N, BLAST P, and BLAST X), and no homologous protein sequences were found within this region, with the exception of the structural subunit of the longus pili of E. coli, which had less than 38% homology. This region of the hlyA gene was not included in the design of the primers and probe. The specificity of the constructed primers and probe was tested both by homology searches of protein and nucleotide databases and by screening a number of V. cholerae strains isolated from patients, oysters, shellfish, and water from several parts of the world. No false positives were recorded among the 61 bacterial species belonging to other genera, and no false positives were recorded among the other Vibrio strains tested, demonstrating the high specificity of the designed primer-probe set for V. cholerae strains (Tables 1 and 2).

Because PCR operates with constant efficiency, it is well suited for quantitative measurements. The detection limits of the PCR assay were estimated to be approximately 10 CFU ml¹ in synthetic seawater and 6 to 8 CFU ml¹ in oysters. These reported limits of detection are similar to those in other reports obtained with a TaqMan PCR assay for endpoint detection (5, 9, 30, 31, 45). For Listeria monocytogenes, Bassler et al. (5) and Nogva et al. (30) obtained detection levels of approximately 50 CFU ml¹ and 6 CFU ml¹, respectively. Chen et al. (9) and Vishnubhatla et al. (45) showed a detection limit as low as 2 CFU ml¹ from a pure culture of Salmonella enterica serovar Typhimurium and 9.4 CFU ml¹ for Yersinia enterocolitica in spiked ground pork, respectively.

Our data indicate a good correlation between CFU counts and the TaqMan assay for V. cholerae incubated in synthetic seawater at 5°C (Fig. 2B and Table 5). These results are in agreement with earlier experiments in which Vibrio cells (i.e., VBNC cells) were recovered from refrigerated seafoods by incubation in SSW at 5°C (10, 27, 47). After 7 days, there was a 7-log reduction in the direct viable counts and plate counts of V. cholerae O1 and O139 (Table 5). These data indicate that traditional plating can result in underestimation of the number of potentially infectious V. cholerae cells present in samples stored under refrigeration. In contrast, VBNC cells are detectable through the use of the TaqMan PCR (Table 5).

It is also important to know the history of the food product prior to analysis. Products such as raw oysters stored on ice may contain a large numbers of cold stress-induced VBNC Vibrio cells. The history of the food product is especially important if the product was heat treated or underwent a process that could cause cell death of the target organism. Nogva et al. (30) found that use of DNase in the sample prior to isolation of the DNA appears to reduce the amount of nucleic acid isolated from dead bacteria. In this study, it appears that the aspect of living versus dead cells was important, because Vibrio cells isolated from spiked oysters appear to contain both viable and dead cells. Treatment with DNase prior to DNA isolation resulted in a 1-log reduction in the cell counts and subsequently resulted in elimination of DNA from dead Vibrio cells (Table 4).

The use of the TaqMan PCR is a sensitive and quantitative method that is useful for estimating the number of cells of a specific pathogen in a food product (5, 9, 30, 31, 45). The main advantage of the TaqMan PCR is that it is very rapid and is a valuable method for screening a large number of samples. Traditional culturing methods require time-consuming enrichments and labor-intensive procedures that require several days. However, the TaqMan PCR developed in this study requires only 3 h. Other molecular systems for identifying V. cholerae in food and water, such as PCR fingerprinting and ribotyping (1, 41) and detection of virulence genes by PCR ( 6, 8, 28, 36, 44, 46; Arya, Letter), require additional verification steps that increase analysis time. The real value of the TaqMan PCR is the potential for rapid analysis of numerous pathogen-free samples, thereby allowing laboratories the ability to quickly screen products before they are released for human consumption. Putative positive food samples can also be quickly identified and pulled for further analysis.

Future developments. The use of TaqMan technology for quantification of V. cholerae in other foods and environmental water samples should be feasible. Work is under way to develop TaqMan assays for V. vulnificus and V. parahaemolyticus. Once these nuclease assays are developed, it should be a simple procedure to screen for all three species by using one enrichment technique and one DNA sample extracted from bacterial cells in the food sample.