Detection and Enumeration of Vibrio vulnificus by Direct Colony Immunoblot


  • R.N. Senevirathne,

    1. Authors Senevirathne and Janes are with Dept. of Food Science and author Simonson is with Dept. of Agricultural Chemistry, Louisiana State Univ. Ag Center, Baton Rouge, LA 70803, U.S.A. Direct inquiries to author Janes (E-mail:
    Search for more papers by this author
  • M.E. Janes,

    1. Authors Senevirathne and Janes are with Dept. of Food Science and author Simonson is with Dept. of Agricultural Chemistry, Louisiana State Univ. Ag Center, Baton Rouge, LA 70803, U.S.A. Direct inquiries to author Janes (E-mail:
    Search for more papers by this author
  • J.G. Simonson

    1. Authors Senevirathne and Janes are with Dept. of Food Science and author Simonson is with Dept. of Agricultural Chemistry, Louisiana State Univ. Ag Center, Baton Rouge, LA 70803, U.S.A. Direct inquiries to author Janes (E-mail:
    Search for more papers by this author


ABSTRACT:  A direct colony immunoblot method (DCI) for the enumeration of Vibrio vulnificus was developed. Bacterial colonies were transferred from agar plates to membranes, which were then dried and blocked with bovine serum albumin. Subsequently, the membranes were treated with anti-V. vulnificus H antibodies, washed and incubated with peroxidase-conjugated goat anti-rabbit IgG. After a final wash, the membranes were exposed to a substrate mixture containing H2O2 which resulted in the development of a purple color by V. vulnificus colonies. The DCI detected all clinical and environmental V. vulnificus strains tested and did not cross-react with other Vibrio species including V. cholerae, V. parahaemolyticus, or V. fluvialis. The DCI was then compared to the DNA hybridization procedure (DNAH) using V. vulnificus agar plates inoculated with mixed cultures of V. vulnificus and V. parahaemolyticus and V. vulnificus-seeded oyster homogenates. Both DCI and DNAH detected 1 to 2 log colony forming units (CFU)/mL V. vulnificus mixed with 4 log CFU/mL V. parahaemolyticus. Both methods were comparable and demonstrated no significant statistical differences when enumerating V. vulnificus in mixed cultures or in oyster homogenates seeded with levels of V. vulnificus from 2 to 6 log CFU/mL. The DCI demonstrated clearer color development and was less time consuming than the DNAH.


Vibrio vulnificus is ubiquitous in the marine environment where it can be isolated from water, sediment, and shellfish (Kelly 1982; Oliver and others 1983). It is postulated that less than 100 virulent cells can cause disease in high-risk individuals and the pathogen is responsible for 95% of all seafood-related deaths in the United States (Oliver 2005). During warmer months in the Gulf of Mexico, levels of V. vulnificus can often reach 103 to 104 most probable number per gram (MPN/g) of oyster meat (Motes and others 1998; Cook and others 2002). Due to the rapid progression and high mortality rates of V. vulnificus infection in humans, especially those with underlying chronic disease, oysters are being examined for the presence and level of this pathogen. Conventional detection and enumeration of V. vulnificus in oysters by the MPN method (Kaysner and DePaola 2004) can be overwhelming in logistics, material, and time, while the expeditious and specific identification of V. vulnificus in the laboratory is desirable.

To save assay time and materials, the Food and Drug Administration (FDA) adopted a direct plating/DNA hybridization (DNAH) method using a nonradioactive labeled probe, which targets the V. vulnificus cytolysin gene (Kaysner and DePaola 2004). This enumeration method is specific and has been shown to be equivalent to the MPN procedure (Wright and others 1993; DePaola and others 1997). However, the method requires expensive reagents, stringent temperature requirements, 8 to 36 h to perform and often it can be difficult to differentiate between positive and negative colonies. The variable time required to complete the DNAH protocol is dependent on the time required for signal development which, in turn, is dependent on the probe source (DePaola and others 1997).

Many vibrios have been shown to express flagellar (H) antigens unique to the species (Miwatani and Shinoda 1971; Bhattacharyya 1975; Tassin and others 1983). Based on this knowledge, anti-V. vulnificus H antibodies were developed and used to construct a coagglutination reagent that agglutinated V. vulnificus cells within 1 to 2 min (Simonson and Siebeling 1986). The anti-H antibodies exhibited species specificity in that, other than V. pelagius, only V. vulnificus cells were coagglutinated from among 19 Vibrio spp. examined including 723 V. cholerae, V. mimicus, V. parahaemolyticus, and V. fluvialis isolates recovered from seafood and the marine environment. However, since colonies were transferred from selective plates and grown overnight on slants, the total V. vulnificus isolation and identification time was about 2 to 3 d depending on whether an enrichment or direct plating was employed.

To further shorten identification time and facilitate V. vulnificus detection, a direct colony immunoblot (DCI) using anti-H antibodies could be employed. Direct colony immunoblots have been successfully used to identify a variety of bacteria including Salmonella, Listeria, and Streptococcus (DeSoet and others 1990; Bhunia and others 1991; Bhunia and Johnson 1992; Petter 1993). For these assays, colonies were transferred directly from an agar plate to a membrane where they were detected using specific antibodies and antibody–enzyme conjugates. Like the direct plating/DNAH method, this results in the direct enumeration of specific bacteria with a single, simple, and rapid assay. A DCI procedure for V. vulnificus would further reduce assay time and would be inexpensive since the only equipment required would be a shaker. The method would also be simple in that it would require no cell lysis, less time and manipulations, and no stringent temperature requirements.

The objective of this study was to examine the possibility of using a DCI to establish a faster, user-friendly method, equivalent to the direct plating/DNAH procedure for the detection and enumeration of V. vulnificus in oysters.

Materials and Methods

Bacterial cultures

The vibrio strains used in this study are listed in Table 1. Stock cultures were stored at −70 °C and subcultures were maintained at room temperature on agar deeps containing 8 g tryptone, 22.5 g NaCl, 4 g nutrient broth, 4 g agar, 4 g KCl, and 4 g MgCl2.6H2O per liter of distilled water. Cultures were transferred monthly to maintain viability.

Table 1—.  Different Vibrio species tested by DCI.
OrganismCulture designationType of strainSourcea
  1. aATCC = American Type Culture Collection; LDHH = Louisiana Dept. of Health and Hospitals; LSU = Louisiana State Univ.; UNCC = Univ. of North Carolina Charlotte; NCS = North Carolina State Univ.

V. alginolyticusATCC® 33787EnvironmentalATCC
V. choleraeATCC® 14035ClinicalATCC
V. damselaATCC® 35083EnvironmentalATCC
V. fluvialisATCC® 33809ClinicalATCC
V. mimicusATCC® 33653ClinicalATCC
V. parahaemolyticusATCC® 17802ClinicalATCC
V. vulnificusVv 1001ClinicalLDHH
V. vulnificusVv 1002ClinicalLDHH
V. vulnificusVv 1004ClinicalLDHH
V. vulnificusVv 1005ClinicalLDHH
V. vulnificusVv 1006ClinicalLDHH
V. vulnificusVv 1007ClinicalLDHH
V. vulnificusVv 1009ClinicalLDHH
V. vulnificusATCC® 27562ClinicalATCC
V. vulnificusC7184ClinicalUNCC
V. vulnificusLSU0106VV12 (EN1)EnvironmentalLSU
V. vulnificusLSU0106VV14 (EN2)EnvironmentalLSU
V. vulnificusLSU0541VV49C (EN3)EnvironmentalLSU
V. vulnificus515 4C2 (EN4)EnvironmentalLSU
V. vulnificusWR1 (EN5)EnvironmentalNCS

Flagellar core purification

Flagellar cores were isolated from a motile strain of V. vulnificus ATCC 27562 by methods described previously (Simonson and Siebeling 1986). Briefly, the bacterial cells were propagated on brain heart infusion agar supplemented with 1.5% NaCl (BHI+). The cells were harvested from the agar surface in 0.15 M NaCl and were homogenized for 90 s in a Waring commercial blender at medium speed. The bacterial cells were sedimented by centrifugation at 10000 ×g for 10 min and the sheared flagella were obtained by centrifugation of the remaining supernatant fluid at 30000 ×g for 2 h. The flagella were suspended in 0.1 M Tris buffer, pH 7.8, containing 0.1 mM EDTA, 1% Triton X-100, and 0.001% thimerosal and the differential centrifugation cycle was repeated 3 times. The H cores were further isolated from residual cellular debris by cesium chloride density ultracentrifugation at 64000 ×g for 18 h. The flagella were examined by electron microscopy to verify that flagellar cores, free of sheath material, were present. The purified cores were stored at 4 °C in Tris buffer containing EDTA and thimerosal. Total protein was determined by the bicinchoninic acid (BCA) method (Pierce, Rockford, Ill., U.S.A.).

Production of anti-H serum

A New Zealand White rabbit was immunized by subcutaneous injection of 125 μg of core protein followed by alternating intravenous injections of 60 μg protein and subcutaneous injections of 125 μg protein at 3-d intervals over a 40-d period. The rabbit was exsanguinated 45 d after the initial injection, the V. vulnificus H antiserum was separated from the red blood cells by centrifugation and stored frozen in small aliquots until used.

Direct colony immunoblot (DCI)

Colonies were directly transferred from incubated V. vulnificus agar (VVA; BAM manual,, thiosulfate-citrate bile salts agar (TCBS; Acumedia Manufactures Inc., Lansing, Mich., U.S.A.), modified cellobiose-polymyxin-colistin agar (mCPC; BAM manual,, BHI+ (Acumedia Manufactures Inc.) or nutrient agar with 2% NaCl (NA+) plates to polyvinylidene fluoride membranes (Pall Corp., Fla., U.S.A.) and air dried for 10 min. The membranes were notched and plates marked at identical positions for proper orientation. Using forceps, the membranes were placed colony side up in separate Petri dishes and washed with gentle agitation at 25 °C on a shaker in 20 mL 0.067 M phosphate buffered saline, pH 7.2 to 7.3 containing 0.05% Tween 20 (PBS T-20) or Triton X-100 for 10 min. The membranes were moved to clean Petri dishes and incubated with 1% bovine serum albumin (BSA) in PBS for 30 min to block nonspecific binding sites. The membranes were washed once and treated with 20 mL rabbit anti-H V. vulnificus serum diluted 1: 2000 in PBS for 30 min. Unbound antibody was removed by washing 4 times for 5 min each in PBS T-20 and the membranes were then incubated with 20 mL peroxidase-conjugated goat anti-rabbit IgG (Sigma Chemical Co., St. Louis, Mo., U.S.A.) diluted 1: 2000 in PBS containing 1% BSA for 30 min. The membranes were again washed 4 times for 5 min each and a color development solution was added (20 mL 0.05 M Tris, pH 7.6, 10 mg 3, 3′-diaminobenzidine tetrahydrochloride, 1 mL 8% NiCl2, and 0.1 mL 30% H2O2) for 5 min. Color development was stopped by washing the filters in distilled water. All washing solutions were collected and treated with chlorine before discarding.

DNA probe hybridization (DNAH)

Colonies were directly transferred from incubated VVA plates to Whatman nr 541 filter paper disks (Whatman Intl. Ltd., Springfield Mill, Maidstone, U.K.) and V. vulnificus was identified by DNA hybridization using an alkaline phosphatase-labeled probe (DNA Technology A/S, Risskov, Denmark) specific for the cytolysin gene (Kaysner and DePaola 2004). Briefly, after lifting colonies onto filter paper, the cells were disrupted by placing the filters in Petri dishes containing lysis solution (0.5 M NaOH, 1.5 M NaCl) and by drying in a microwave oven. The filters were neutralized using 4 mL of 2 M ammonium acetate buffer, rinsed twice, and treated with proteinase K for 30 min at 42 °C with agitation on an orbital shaker. The filters were rinsed 3 times, incubated with hybridizing buffer (0.5 g BSA, 1 g sodium dodecyl sulfate, 0.5 g of polyvinylpyrrolidone, and 100 mL of 5× standard saline citrate solution) at 55 °C for 30 min after which the DNA probe in fresh buffer was added and incubation at 55 °C was continued for 1 h. The filters were washed 5 times for 5 min each and incubated with substrate solution (2 nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate ready to use tablets (Roche Diagnostics, Indianapolis, Ind., U.S.A.) in 20 mL distilled water). Color development was checked with control strips and was stopped by washing the filters in distilled water.

Enumeration of V. vulnificus in mixed cultures

Vibrio vulnificus ATCC 27562 and V. parahaemolyticus ATCC 17802 were grown separately in 10 mL of tryptic soy broth containing 3% NaCl (T1N3) overnight at 37 °C. Ten microliters of the overnight culture were transferred to 10 mL of T1N3 and incubated at 37 °C for 16 h. The bacterial cells were collected by centrifugation, washed once with PBS (pH 7.3), and were suspended in 10 mL PBS. Serial 10-fold dilutions of V. vulnificus and V. parahaemolyticus were made separately in PBS. Ten microliters of each dilution of V. vulnificus and V. parahaemolyticus were mixed with 80 μL of PBS and the combined cultures were then spread on VVA plates. The plates were incubated at 35 °C for 16 h and the number of colony forming units (CFU) for V. vulnificus and V. parahaemolyticus was determined. The DCI immediately followed by DNAH was performed on each mixed culture plate as described previously. The concentration of the vibrios in the initial pure cultures was determined by separately plating aliquots of the V. vulnificus and V. parahaemolyticus serial dilutions on VVA plates.

Enumeration of V. vulnificus in oysters

Oysters were collected in January, May, and July of 2006 from the Gulf of Mexico. The oysters were immersed in ice just after harvest. Upon arrival at the laboratory, the oyster shells were scrubbed under running water to remove debris and kept for 48 to 64 h in a freezer to minimize natural microflora. The oyster shells were then opened aseptically using sterilized oyster knives. Serial dilutions of a 16 h V. vulnificus ATCC 27562 culture were made as described previously. Ten milliliters from each 10−1 to 10−5 dilution were mixed with 40 mL of alkaline peptone water (APW) and 50 g of oyster meat and the solution was homogenized in a stomacher for 1 to 2 min. An oyster aliquot (50 g) was mixed with 50 mL of APW as a control. Serial 10-fold dilutions in PBS were made from each spiked oyster homogenate and 100 μL aliquots were plated on VVA plates. The plates were incubated for 16 h at 35 °C and colonies were counted. Vibrio vulnificus was enumerated directly from each VVA plate by DCI followed by DNAH as described previously using the same VVA plate.

Statistical analysis

Both the DCI method and the DNAH method were analyzed by statistical comparisons of all pairs using Student's t-test following one-way analysis of the variance (ANOVA) (JUMP In version 4.0.3, SAS Inst. Inc., Cary, N.C., U.S.A.). Statistical significance occurs for P < 0.05. All experiments were repeated 3 times with 2 replications per experiment.

Results and Discussion

Detection of V. vulnificus by DCI

We have developed a DCI method for specific detection of V. vulnificus in oysters that was completed in less than 4 h at room temperature whereas the DNAH method takes about 6 h. Initially, different selective (VVA, TCBS, and mCPC) and nonselective (BHI+ or NA+) media were tested for use in development of the procedure. Although detection of V. vulnificus by DCI was possible on all media assayed (data not shown), it was determined that the most effective medium for enumeration of V. vulnificus in oyster homogenates by DCI was VVA because of its selective and differential nature and ease of preparation. Vibrio colonies also appeared smaller and less mucoid on VVA plates when compared to TCBS, BHI+, or NA+. In addition, the FDA recommends employment of VVA plates for direct plating of oyster homogenates in conjunction with the DNAH protocol (Kaysner and DePaola 2004).

The rabbit anti-flagellar serum exhibited a titer of 6400 against the vaccine strain by tube H flocculation (Tassin and others 1983) and was diluted 1: 2000 in PBS for use in the DCI, since this was the highest dilution that produced optimal color development of V. vulnificus colonies within 30 min. In addition, substrate color development was complete, without nonspecific antibody binding, within 5 min using this serum dilution. In an effort to improve antibody reactivity and DCI sensitivity by stripping the flagellar sheath from the core (Simonson and Siebeling 1986), Triton X-100 was substituted for Tween 20 in the washing buffer. Even though good color development of V. vulnificus colonies was observed by using Triton X-100 in the wash solution, the chemical damaged the membranes, rendering them almost transparent. Thus, Tween 20 was incorporated into the washing buffer for all subsequent studies.

The sensitivity and specificity of the assay were tested using various Vibrio species and V. vulnificus clinical and environmental strains (Table 1). The Vibrio isolates were stab inoculated in a grid pattern on VVA plates and the plates were incubated for 16 h at 35 °C and assayed by DCI. All V. vulnificus strains were easily differentiated from other Vibrio species by the production of dark purple spots on the developed membranes (data not shown). Vibrio fluvialis and V. damsela colonies exhibited slight color reactions, but the intensity of the color was negligible. This weak cross-reaction was insignificant when compared to nonspecific color development exhibited by DNAH at lower hybridizing temperatures (Wright and others 1993). Anti-V. vulnificus H polyclonal antibody obtained from immunization of rabbits with purified intact flagellar cores has been shown to exhibit excellent specificity for and reactivity with motile V. vulnificus strains (Simonson and Siebeling 1986). Of 434 V. vulnificus isolates identified bacteriologically, 432 (99.5%) were detected with the polyclonal anti-V. vulnificus antibody which, other than a low titer against V. pelagius, did not demonstrate any detectable cross-reactivity with 19 heterologous Vibrio spp. Thus, employment of V. vulnificus anti-H antibody in a DCI would result in a low number of false negative reactions. Similarly, the occurrence of false positives, due to cross-reaction of the primary antibody with other Vibrio species, should be very low.

Enumeration of V. vulnificus in mixed cultures

Vibrio vulnificus was easily detected and enumerated by both V. vulnificus-specific DCI and DNAH in mixed V. vulnificus and V. parahaemolyticus cultures spread on VVA plates (Figure 1). Vibrio vulnificus colonies exhibited a strong purple color development on membranes and filters using both methods. However, V. vulnificus colonies demonstrated a more intense color difference against the background of the DCI membrane (Figure 1B) when compared to the appearance of V. vulnificus colonies detected by DNAH (Figure 1C). No cross-reaction was observed for V. parahaemolyticus on DCI membranes (Figure 1B) while V. parahaemolyticus colonies produced a brown color on DNAH filters (Figure 1C).

Figure 1—.

Detection of V. vulnificus using DCI and DNAH. (A) VVA plate inoculated with mixed culture of V. vulnificus and V. parahaemolyticus, (B) V. vulnificus-specific direct colony immunoblot of VVA plate, (C) V. vulnificus-specific DNA probe hybridization of VVA plate.

The sensitivity of the DCI compared to DNAH was tested by mixing V. vulnificus with increasing concentrations of V. parahaemolyticus (100 to 104 CFU/mL). The 2 methods were equally sensitive since V. vulnificus colonies were readily counted by DCI and DNAH when levels of V. vulnificus were inoculated at 1 or 2 log CFU/mL and V. parahaemolyticus levels were increased (Table 2). When the V. parahaemolyticus concentration reached 4.5 log CFU/mL, V. vulnificus could still be enumerated by DCI and DNAH even though V. vulnificus colonies were overgrown by V. parahaemolyticus colonies on the VVA plate (unable to count). There were no significant differences observed in the results when V. vulnificus was enumerated by DCI and DNAH at different V. parahaemolyticus challenge levels.

Table 2—.  Enumeration of V. vulnificus in mixed V. vulnificus and V. parahaemolyticus cultures (log CFU/mL).a
V. vulnificusbV. parahaemolyticuscDCIdDNAHe
  1. aAll analyses were based on 3 separate experiments with each mean ± standard deviation being average of 3 determinations. Means with each vertical column followed by the same letter are not significantly different (P ≥ 0.05) from each other.

  2. bNumber of V. vulnificus as determined by growth and appearance on VVA plates.

  3. cNumber of V. parahaemolyticus added to mixed culture.

  4. dNumber of V. vulnificus as determined by direct colony immunoblot.

  5. eNumber of V. vulnificus as determined by DNA hybridization.

2 logUnable to count4.50 ± 0.412.02 ± 0.19 A2.01 ± 0.17 A
2.163.55 ± 0.402.14 ± 0.19 A2.07 ± 0.28 A
2.062.27 ± 0.162.10 ± 0.09 A2.09 ± 0.09 A
1.991.49 ± 0.402.17 ± 0.10 A2.18 ± 0.16 A
2.0402.15 ± 0.21 A2.10 ± 0.19A
1 logUnable to count4.50 ± 0.411.20 ± 0.15 B1.18 ± 0.18 B
1.323.55 ± 0.401.29 ± 0.35 B1.19 ± 0.24 B
1.222.27 ± 0.161.16 ± 0.26 B1.16 ± 0.26 B
1.131.49 ± 0.401.17 ± 0.05 B1.18 ± 0.03 B
1.1201.00 ± 0.24 B0.97 ± 0.25 B

Enumeration of V. vulnificus in oysters

To further test the accuracy, sensitivity, and specificity of the DCI, oyster homogenates were spiked with different levels of V. vulnificus and the organism was then enumerated from dilutions of the seeded oyster slurry by DCI and DNAH (Table 3). Both methods exhibited comparable results and demonstrated no significant differences when oysters were seeded with levels of V. vulnificus from 2 to 6 log CFU/mL. There were no difficulties encountered when enumerating V. vulnificus from oyster homogenates with either DCI or DNAH. The naturally occurring V. vulnificus counts in the oysters remained low, averaging around 1.5 log CFU/g. Vibrio vulnificus colonies were easily differentiated as being dark purple spots on the DCI membranes and dark brown to purple spots on the DNAH membranes (data not shown).

Table 3—.  Enumeration of V. vulnificus by DCI and DNAH in seeded oyster homogenate (log CFU/g).a
InoculumbTotal countcDCIdDNAHe
  1. aAll analyses were based on 3 separate experiments with each mean ± standard deviation being average of 3 determinations. Means with each vertical column followed by the same letter are not significantly different (P ≥ 0.05) from each other.

  2. bNumber of V. vulnificus added per milliliter of undiluted oyster homogenate.

  3. cNumber of bacteria as determined by growth on VVA plates.

  4. dNumber of V. vulnificus as determined by direct colony immunoblot.

  5. eNumber of V. vulnificus as determined by DNA hybridization.

6.37 ± 0.246.66 ± 0.456.52 ± 0.34 A6.42 ± 0.32 A
5.55 ± 0.315.67 ± 0.305.34 ± 0.22 B5.35 ± 0.22 B
4.08 ± 0.074.84 ± 0.254.56 ± 0.35 C4.54 ± 0.27 C
3.00 ± 0.174.78 ± 0.583.77 ± 0.20 D3.76 ± 0.17 D
1.79 ± 0.303.31 ± 0.262.65 ± 0.35 E2.63 ± 0.33 E
02.08 ± 0.151.52 ± 0.29 F1.58 ± 0.32 F


It is postulated that the V. vulnificus DCI employing specific anti-H primary antibody could be a useful addition to the methods available for enumerating this pathogen in oyster homogenates and other environmental samples. The procedure is faster and simpler than other methods and demonstrates good correlation with the DNAH method. Vibrio vulnificus colonies were easily differentiated on the DCI membranes whereas, depending on the quality of the cytolysin probe used in the DNAH, V. vulnificus colonies were sometimes difficult to differentiate from V. parahaemolyticus colonies. The DCI would also require less equipment and supplies than real-time PCR methods and V. vulnificus detection is not inhibited by the oyster matrix.


This study was supported by grant NA16RB2249 from the Louisiana Sea Grant College Program, a part of the Natl. Sea Grant College Program maintained by the Natl. Oceanic and Atmospheric Administration, U.S. Dept. of Commerce.