Differential specificity of selective culture media for enumeration of pathogenic vibrios: Advantages and limitations of multi-plating methods

Differential specificity of selective culture media for enumeration of pathogenic vibrios: Advantages and limitations of multi-plating methods

Journal of Microbiological Methods 111 (2015) 24–30 Contents lists available at ScienceDirect Journal of Microbiological Methods journal homepage: w...

990KB Sizes 0 Downloads 9 Views

Journal of Microbiological Methods 111 (2015) 24–30

Contents lists available at ScienceDirect

Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth

Differential specificity of selective culture media for enumeration of pathogenic vibrios: Advantages and limitations of multi-plating methods Olivia D. Nigro ⁎, Grieg F. Steward Department of Oceanography, University of Hawaii at Manoa, 1950 East West Road, C-MORE Hale, Honolulu, HI 96822, United States

a r t i c l e

i n f o

Article history: Received 24 September 2014 Received in revised form 13 January 2015 Accepted 16 January 2015 Available online 17 January 2015 Keywords: Vibrios Identification Enumeration Environmental microbiology Culture media Pathogens

a b s t r a c t Plating environmental samples on vibrio-selective chromogenic media is a commonly used technique that allows one to quickly estimate concentrations of putative vibrio pathogens or to isolate them for further study. Although this approach is convenient, its usefulness depends directly on how well the procedure selects against false positives. We tested whether a chromogenic medium, CHROMagar Vibrio (CaV), used alone (single-plating) or in combination (double-plating) with a traditional medium thiosulfate-citrate-bile-salts (TCBS), could improve the discrimination among three pathogenic vibrio species (Vibrio cholerae, Vibrio parahaemolyticus, and Vibrio vulnificus) and thereby decrease the number of false-positive colonies that must be screened by molecular methods. Assays were conducted on water samples from two estuarine environments (one subtropical, one tropical) in a variety of seasonal conditions. The results of the double-plating method were confirmed by PCR and 16S rRNA sequencing. Our data indicate that there is no significant difference in the false-positive rate between CaV and TCBS when using a single-plating technique, but determining color changes on the two media sequentially (double-plating) reduced the rate of false positive identification in most cases. The improvement achieved was about two-fold on average, but varied greatly (from 0- to 5-fold) and depended on the sampling time and location. The double-plating method was most effective for V. vulnificus in warm months, when overall V. vulnificus abundance is high (false positive rates as low as 2%, n = 178). Similar results were obtained for V. cholerae (minimum false positive rate of 16%, n = 146). In contrast, the false positive rate for V. parahaemolyticus was always high (minimum of 59%, n = 109). Sequence analysis of false-positive isolates indicated that the majority of confounding isolates are from the Vibrionaceae family, however, members of distantly related bacterial groups were also able to grow on vibrio-selective media, even when using the double-plating method. In conclusion, the double-plating assay is a simple means to increase the efficiency of identifying pathogenic vibrios in aquatic environments and to reduce the number of molecular assays required for identity confirmation. However, the high spatial and temporal variability in the performance of the media mean that molecular approaches are still essential to obtain the most accurate vibrio abundance estimates from environmental samples. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Members of the bacterial genus Vibrio are important constituents of coastal microbial communities. Some vibrio species, such as Vibrio cholerae, Vibrio parahaemolyticus, and Vibrio vulnificus, are also opportunistic human pathogens. Infections with these species may occur by exposure of open wounds to marine or estuarine water, or by ingestion of contaminated water or raw or poorly cooked seafood (Pacini, 1854; Blake et al., 1979; Colwell and Huq, 1994; Wittman and Flick, 1995; Oliver, 2005). The resulting infections can be mild to severe, with the worst cases being fatal (Blake et al., 1979; Faruque et al., 1998). The ability to quickly isolate and identify pathogenic vibrios is important to environmental microbiologists trying to assess the risk of vibrio

⁎ Corresponding author. E-mail address: [email protected] (O.D. Nigro).

http://dx.doi.org/10.1016/j.mimet.2015.01.014 0167-7012/© 2015 Elsevier B.V. All rights reserved.

infection from coastal waters, and to clinicians trying to diagnose infections that have already occurred. Vibrio-selective media have been developed (Oliver, 2012) and several are commercially available. Although pathogenic vibrio species are the primary target of these media, the growth of other vibrios and even non-vibrio species has been reported (Gomez-Gil and Roque, 2006). Thiosulfate-citrate-bile salts-sucrose (TCBS) is primarily used to culture pathogenic vibrio strains and it is the most widely used medium for vibrio isolation (Gomez-Gil and Roque, 2006). The medium contains sucrose as well as a pH indicator that is green at alkaline pH and turns yellow in the presence of acid (Table 1). Vibrios that grow on TCBS and do not ferment sucrose (sucrose negative) form green colonies, while those that do ferment sucrose (sucrose positive) produce local areas of reduced pH and form yellow colonies (Fig. 1). This medium does not allow discrimination between many vibrio species. For example, V. cholerae and V. alginolyticus are both predominantly sucrose positive, and V. parahaemolyticus and V. vulnificus are both predominantly

O.D. Nigro, G.F. Steward / Journal of Microbiological Methods 111 (2015) 24–30 Table 1 Composition of TCBS and CaV media. TCBS (g l−1) Agar Proteose peptone Yeast extract Sodium citrate Sodium thiosulfate Ox gall Saccharose Sodium chloride Ferric ammonium citrate Bromothymol blue Thymol blue

CaV (g l−1) 15 10 5 10 10 8 20 10 1 0.04 0.04

Agar Peptone and yeast extract Salts Chromogenic mix

15 8 51.4 0.3

sucrose negative. Non-Vibrio genera reported to grow on TCBS include Staphylococcus, Flavobaterium, Pseudoalteromonas, Streptococcus, Aeromonas, and Shewanella (Nicholls et al., 1976). The medium CHROMagar Vibrio (CaV) selects for vibrios using a high pH (9.0) and discriminates among strains based on differences in their ability to metabolize chromogenic substrates (Fig. 1). Some of the medium ingredients are known (Table 1), but the chromogenic mixture that causes the color change is proprietary. Two studies investigating the accuracy and specificity of TCBS and CaV for isolating V. parahaemolyticus from seafood samples have both indicated that CaV is more accurate and specific than TCBS (Hara-Kudo et al., 2001; Di Pinto et al., 2011). This is expected, because V. parahaemolyticus is not distinguishable from V. vulnificus on TCBS (most strains of each species form green colonies), but these species form different colors on CaV. Conversely, V. vulnificus and V. cholerae are not discriminated from one another on CaV, but can be discriminated on TCBS. We took advantage of the complementary capabilities of CaV and TCBS in studies of vibrios in Lake Pontchartrain, LA (Nigro et al., 2011) and in the Ala Wai Canal, HI (Nigro, 2012) by employing a doubleplating procedure to obtain isolates. We reasoned that streaking colonies on these two chromogenic media sequentially and using the color change information from each would significantly decrease the percentage of false positive colonies for all three of the species in which we were interested (V. cholerae, V. parahaemolyticus, and V. vulnificus), both by discriminating among them and by eliminating a greater proportion of non-target species. A similar approach using three media was later described as a means to minimize false positives when isolating V. vulnificus (Williams et al., 2013), but the method described is specific to that species and was

25

tested in one location over a limited period. Our objectives with this study were 1) quantify the false positive identification rates of three pathogenic vibrio species based on color changes on TCBS and CaV when used alone and together, and 2) identify, by 16S rRNA sequencing, some of the species that present as false positives for V. cholerae, V. parahaemolyticus, or V. vulnificus using the double-plating method. 2. Materials and methods 2.1. Sampling Whole water samples were collected from each of two environments, one subtropical (Lake Pontchartrain, New Orleans, LA) and one tropical (Ala Wai Canal, Honolulu, HI). Whole water samples were taken from 15 stations in Lake Pontchartrain on four occasions (October 2005, January, March, and September 2006) as previously described (Sinigalliano et al., 2007; Nigro et al., 2011). Water was collected from the Ala Wai Canal on five occasions (March, June, September, and December 2008, and March 2009) at 15 stations (Nigro, 2012). In all cases, whole water samples were filtered in duplicate or triplicate through sterile 0.45 μm filters. Duplicate or triplicate filters from each station were placed face up on each of two different, chromogenic, vibrio-selective media (TCBS and CaV) immediately after filtering (four to six replicates in total). Putative Vibrio spp. were enumerated as colony-forming units (CFUs) after 12–18 h of incubation at 37 °C. Selected colonies picked from one medium were streaked onto plates of the other medium and incubated for 12–18 h at 37 °C and color changes on the two media were used to make preliminary species assignments. 2.2. Isolate identification: DNA extraction and PCR Following preliminary identification, colonies were streak purified three times, alternating between TCBS and CaV agar plates for each streaking. Template DNA was prepared by dispersing a single colony in TE buffer (10 mmol l− 1Tris, 1 mmol l− 1 EDTA, pH 8), heating the suspension to 100 °C in a thermal cycler with a heated lid for 10 min, followed by centrifugation at 5000 ×g for 10 min. PCR analysis was performed on DNA from isolated colonies from Lake Pontchartrain that were presumed to be V. cholerae (n = 591), V. parahaemolyticus (n = 401), or V. vulnificus (n = 570). From the Ala Wai Canal, only putative V. vulnificus isolates (n = 1084) were collected. The Ala Wai isolates were used to determine only the false positive rate of the double-plating method. For each PCR reaction, 1 μl

Fig. 1. Appearance of Vibrio isolates on TCBS and CaV. On TCBS, V. cholerae appears yellow (Y). V. parahaemolyticus and V. vulnificus appear green (G). On CaV, V. parahaemolyticus appears mauve (M), while V. cholerae and V. vulnificus appear blue (B). Non-target species may appear white (W) or a variety of other colors.

26

O.D. Nigro, G.F. Steward / Journal of Microbiological Methods 111 (2015) 24–30

of the DNA-containing supernatant was used as template. A preliminary PCR reaction was performed on each extract using 16S rRNA oligonucleotides primers (Lane, 1991) (Table 2) to ensure that samples were amplifiable. Extracts that did not amplify with the 16S rRNA primers were cleaned using InstaGene Matrix (Bio-Rad) according to the manufacturer's instructions and tested again. DNA extracts were then assayed by PCR with species-specific primers (Table 2) for V. cholerae (internal transcribed spacer region of rRNA gene) (Chun et al., 1999), V. vulnificus (vvhA) (Brasher et al., 1998; Morris et al., 2004; Panicker et al., 2004), and V. parahaemolyticus (tlh) (Bej et al., 1999). PCR was performed using 2.6 units of Taq polymerase and 5 μl of 5 × PCR buffer per reaction (both from Roche Applied Science) per 25 μl reaction. Primer and MgCl2 concentrations in each reaction are listed in Table 2. PCR conditions were as follows: initial denaturation at 95 °C for 5 min, followed by 40 cycles of 95 °C for 1 min, the appropriate annealing temperature (Table 2) for 1.5 min and 72 °C for 1.5 min in a DNA Engine thermal cycler (MJ Research, Inc.). Strains used as positive controls were V. vulnificus YJ016, V. cholerae 0395, and V. parahaemolyticus AW1.

Phylogenetic trees were annotated and displayed using the Interactive Tree of Life v. 2 (Letunic and Bork, 2011). 2.4. Statistical treatment of data Data were analyzed using the SPSS statistics software package (PASW Statistics version 22.0), and the cutoff for statistical significance was set at p b 0.05. False-positive rates for each medium were calculated as the percentage of colonies that appeared the correct color on either TCBS or CaV medium, that were determined by PCR to be something other than the presumptive species. False-positive rates for the double-plating method were calculated as the percentage of colonies having the correct color on both media that were determined by PCR to be something other than the presumptive species. Statistical significance of the differences in false positive rates between single- and double-plating methods was determined using a dependent t-test. Bivariate correlation analysis was performed using Pearson's correlation coefficient. 3. Results

2.3. Isolate identification confirmation: DNA sequence analysis 3.1. Efficacy of CHROMagar Vibrio A subset of all of the isolates was selected for sequencing of the 16S rRNA gene in order to confirm results of the species-specific PCR and to determine the identities of false-positive isolates. From Lake Pontchartrain, the 16S rRNA genes of 54 isolates were sequenced. From the Ala Wai Canal, 82 isolates were sequenced. For all isolates, DNA was amplified with the 16S rRNA primers (Table 2) using the Expand High Fidelity PCR system (Roche Applied Science). For each reaction, 2.6 units of enzyme mix were used to amplify products under the following conditions: initial denaturation at 95 °C for 5 min, followed by 40 cycles of 95 °C for 30 s, 50 °C for 1 min, and 72 °C for 2 min, followed by a final extension at 72 °C for 7 min. PCR products were purified for sequencing using PureLink 96-well PCR Kit (Invitrogen), according to the manufacturer's instructions. Direct sequencing of the amplicons was performed in both directions using the 27B-F and the 1492-R primers (Table 2). Sequences were aligned using Sequencher (Gene Codes Corporation) and Geneious (Drummond et al., 2011) software. Sequence alignments were constructed using the SILVA INcremental Aligner (SINA) (Pruesse et al., 2012) and then imported into release 106 of the ‘All-Species Living Tree’ project SSU rRNA gene database (Yarza et al., 2010) using the ARB version 5.3 software (Ludwig, 2004). Alignments were refined manually by including nearest neighbors and taking into account the secondary structure information of the rRNA gene. Clone sequences were identified to the closest recognized species or lineage. Phylogenetic trees were built using the ARB version 5.3 software (Ludwig, 2004). A database was created that contained the DNA sequences generated in this study and their nearest neighbor sequences, along with selected sequences from other vibrios for context. A phylogenetic tree was generated using RAxML maximum-likelihood method using a GTR gamma model and a rapid hill-climbing algorithm.

Of all of the blue colonies isolated on CaV from four samplings of Lake Pontchartrain over the course of a year (n = 302), 32.5% were neither V. cholerae nor V. vulnificus. The false positive rate varied among samplings, however, with very low false positive rates (b 5.4%) in September and October when the lake was warmer and high false positive rates (N98%) in January and March when the lake temperature was lower (Fig. 2). Of the mauve colonies isolated from Lake Pontchartrain on all four occasions (n = 133) 91.7% of the isolates were not V. parahaemolyticus, but the false positive rate for mauve colonies also varied among samplings from 58 to 100% (Fig. 2). 3.2. Effectiveness of using the double-plating method to identify species When plating on TCBS, the percentage of yellow colonies that turned out to be V. cholerae varied from 2.9 to 57%. The percentage of green colonies that turned out to be V. parahaemolyticus varied from 0 to 7.1% and the percentage that turned out to be Vibrio vulnificus ranged from 0 to 72% (Table S1). When plating on CaV, the percentage of blue colonies confirmed to be V. cholerae varied from 2.0 to 26% and those confirmed to be V. vulnificus ranged from 0 to 88%. The percentage of mauve colonies on CaV that were confirmed to be V. parahaemolyticus ranged from 0 to 42% (Table S2). In all cases where one of the three pathogenic vibrios was detected on a single medium, screening on a second medium (double-plating method) increased the percentage of the isolates correctly identified (Fig. 3). On two occasions (winter and spring in Lake Pontchartrain), no V. parahaemolyticus or V. vulnificus colonies were detected on either medium, so no improvement was possible by the double-plating method. When considering the average for each species over all

Table 2 PCR primer names, sequences, and concentrations at which they are used along with the MgCl2 concentrations and the annealing temperatures used for the reactions in which the primers were applied. Name

Sequence (5′ to 3′)

Primer (μmol l−1)

MgCl2 (mmol l−1)

Temp (°C)

Reference

27B F 1492 R VVHA F VVHA R TLH F TLH R Chol ITS F Chol ITS R

AGRGTTYGATYMTGGCTCAG GGYTACCTTGTTACGACTT TTCCAACTTCAAACCGAACTATGAC ATTCCAGTCGATGCGAATACGTTG AAAGCGGATTATGCAGAAGCACTG GCTACTTTCTAGCATTTTCTCTGC TTAAGCSTTTTCRCTGAGAATG AGTCACTTAACCATACAACCCG

0.6 0.6 0.6 0.6 0.5 0.5 0.6 0.6

15 15 15 15 25 25 15 15

50 50 63 63 58 58 60 60

Morris et al. (2004) Lane (1991) Brasher et al. (1998) Panicker et al. (2004) Bej et al. (1999) Bej et al. (1999) Chun et al. (1999) Chun et al. (1999)

O.D. Nigro, G.F. Steward / Journal of Microbiological Methods 111 (2015) 24–30

27

3.3. Variation in the incidence of false positives

Fig. 2. False positive rate of blue and mauve isolates on CaV for samples collected from Lake Pontchartrain on four occasions. The numeral above each bar indicates the number of isolates analyzed. Also shown is the average temperature of the lake at each sampling as reported previously (Nigro et al., 2011).

sampling times, the use of two media to identify species resulted in a 1.9- to 2.7-fold increase in the percent correctly identified compared to use of either medium alone (Tables S1 and S2). There was no significant difference in the degree of improvement whether colonies were first isolated on CaV or on TCBS (paired t-test; p ≥ .16, n = 34). Although the relative improvement in identification using the double- vs. single-plating method was similar across species, the absolute percentage of colonies correctly identified by double-plating varied. In subtropical Lake Pontchartrain, up to 41% (mean 13%) of V. parahaemolyticus was correctly identified, but up to 84% (mean 51%) of V. cholerae, and up to 98% (mean 93%) of V. vulnificus was correctly identified. In the tropical estuary, up to 87% (mean 53%) of V. vulnificus was correctly identified (Fig. 3).

The percentages of false positive V. cholerae and V. parahaemolyticus isolates obtained on TCBS were significantly negatively correlated with the log total abundance of yellow; r = −.955, p (2-tailed) = 0.045 or green; r = − .96, p (2-tailed) = 0.04 colonies on TCBS, respectively (Fig. S1). A negative relationship was also observed between log green colony counts and false positive V. vulnificus rate, however it was not statistically significant; r = −.781, p (2-tailed) = 0.219. For the sample collected in Lake Pontchartrain, no significant correlation between false positive rate and the count of colonies of the appropriate color on CaV was detected for V. cholerae: r = − .672, p (2-tailed) = 0.328, V. parahaemolyticus: r = − .234, p (2-tailed) = 0.766, or V. vulnificus: r = −.397, p (2-tailed) = 0.603. For the Ala Wai Canal samples, only V. vulnificus isolates were investigated and the percentage of false positives of V. vulnificus was significantly negatively correlated with total green colony counts on TCBS; r = − .932, p (2-tailed) = 0.021 and total blue colony counts on CaV; r = − .967, p (2-tailed) = 0.007 (Fig. S2).

3.4. 16S rRNA sequence analysis of false positive isolates Sequence analysis of the 16S rRNA gene was performed on a selection of 101 isolates from Lake Pontchartrain and the Ala Wai Canal that were identified as putative pathogenic vibrio species based on their color changes on chromogenic media, but were PCR-negative for the species based on failure to amplify a species-specific marker gene. Thirty-one of these putative false-positive isolates were analyzed from Lake Pontchartrain (LP) and 70 isolates were analyzed from the Ala Wai Canal (AW). Among these, three isolates were false positive for V. cholerae, 24 for V. parahaemolyticus and 74 for V. vulnificus. Of the isolates that were false positive for V. cholerae, phylogenetic analysis indicates that all three belong to the genus Vibrio: two are closely related

Lake Pontachartrain

Ala Wai Canal

100

Correctly identified by color change (%)

90 80

Fall Summer Spring Winter

70 60 50 40 30 20 10 0 one medium

two media

V. parahaemolyticus

one medium

two media

V. cholerae

one medium

two media

V. vulnificus

one medium

two media

V. vulnificus

Fig. 3. Percentage of isolates confirmed by PCR to be V. parahaemolyticus (Vp), V. cholerae (VC), or V. vulnificus (Vv) when selected based on color change on a single medium (either TCBS or CaV) or when selected based on color change on both media by sequential streaking.

28

O.D. Nigro, G.F. Steward / Journal of Microbiological Methods 111 (2015) 24–30

to Vibrio aestuarianus (LP) while the third is most closely related to Vibrio diazotrophicus (AW) (Fig. 4). All of the 24 isolates analyzed that were false positives for V. parahaemolyticus using the double-plating method were isolated from Lake Pontchartrain. According to phylogenetic analysis, only one of these isolates groups within the genus Vibrio. Fifteen of the isolates grouped outside of the genus, but were in the Vibrionaceae family. Fourteen of these sequences were most closely related to Photobacterium damselae and one was closely related to Photobacterium ganghwense. Of the remaining eight isolates, four grouped closely with members of the clade that contains the Gram-negative genus Aeromonas, and four others grouped most closely with members of the Gram-positive Exiguobacterium genus (Fig. 4). Of the 74 isolates that were identified as false-positives for V. vulnificus, and subsequently investigated by 16S rRNA sequencing and analysis, 69 were isolated from the Ala Wai Canal and five were isolated from Lake Pontchartrain. Phylogenetic analysis showed that the five false-positive isolates from Lake Pontchartrain all grouped within the genus Vibrio, and were most closely related to V. aestuarianus. Of the 69 false-positive isolates that were investigated from the Ala Wai

Canal, 61 grouped most closely with members of the genus Vibrio. Eight isolates did not group with members of the Vibrio genus, but grouped with members of the family Vibrionaceae, and one isolate grouped most closely with the gammaproteobacterium Enterobacter ludwigii (Fig. 4). 4. Discussion 4.1. Efficacy of CHROMagar Vibrio medium for identifying pathogens CaV seems an attractive alternative to TCBS for isolating pathogenic vibrios, since it provides multi-color discrimination among a number of species. However, our data and that of other studies that focused on V. vulnificus (Froelich et al., 2012; Williams et al., 2013) indicate that using CaV alone to enumerate pathogenic vibrios from natural environmental samples is unreliable without confirmation by additional methods. Not only are V. vulnificus and V. cholerae indistinguishable from one another, but these pathogens are indistinguishable from a variety of other vibrio species, some of which can be present at relatively high concentrations. The latter problem was most pronounced for

cu

a

tes

teri

mi

bac

teo

Pro

Fir

Species Labels Correct ID False Vc False Vp False Vv

Outer Ring domain Bacteria class Gammaproteobacteria order Vibrionales genus Vibrio 0.001

Fig. 4. Phylogenetic tree based on the 16S rRNA sequences of isolates and types strains. Isolate names highlighted in colors indicate those that were correctly identified by the double-plating method as one of the three species (Correct ID, green) and those that were false positives for V. cholerae (Vc, yellow), V. parahaemolyticus (Vp, mauve) or V. vulnificus (Vv, blue). The code names of isolates correctly identified begin with “V” and those that were false positives begin with “FP”. The bar at the end of each of these species represents the relative percent of isolates with that identical sequence, relative only within each species/color group. The ring around the outside of the tree indicates the nearest taxonomic level that is common between the non-target and target species. Genus names in the tree are abbreviated as follows: Aeromonas (A.), Aliivibrio (Al.), Enterovibrio (E.), Enterobacterium (Eb.), Exiguobacterium (Ex.), Grimontia (G.), Listonella (L.), Photobacterium (P.), Shewanella (S.), Salinivibrio (Sv.), and Vibrio (V.). Scale bar for interpreting branch lengths indicates nucleotide substitutions per site.

O.D. Nigro, G.F. Steward / Journal of Microbiological Methods 111 (2015) 24–30

V. parahaemolyticus, which can be distinguished from V. vulnificus and V. cholerae on CaV, but was always a minority of the total mauve CFUs in Lake Pontchartrain. These conclusions differ from those found by Di Pinto et al. (2011), who determined CaV to be 88% accurate and 95% specific when determining the identities of isolates from shellfish samples. These results were significantly higher than the accuracy and specificity they observed with TCBS, which were 51% and 71%, respectively. That study demonstrates that there can be situations in which CaV is superior to TCBS for identifying pathogens, but for the suite of environmental samples we analyzed, there was no significant difference in the false-positive rates for the two media for any of the three pathogens we targeted. 4.2. Efficacy of the double-plating method Streaking presumptive vibrio isolates sequentially on CaV and TCBS media results in a two-color combination signature that is unique among the three species we targeted. False positives can, and did, still occur, but the incidence of false positives was substantially reduced for all species when using both media together compared to using either medium alone. Although the average percentage of isolates correctly identified by double-plating varied by species, the relative advantage provided by using the two media, as opposed to only one, was remarkably similar for all three targeted species. Streaking colonies from one medium type to the other is relatively simple and inexpensive. It also does not add much time, since colonies often need to be serially streaked anyway to obtain clonal isolates. Cross-streaking colonies onto a second medium is therefore an effective way to reduce the number of organisms that are non-target species. This saves time and money when subsequent techniques, such as PCR, must be employed to confirm the identity of isolates. The same conclusion was reached in a recent paper describing a triple-plating method for V. vulnificus (Williams et al., 2013) which, in addition to CaV and TCBS, included plating on CPC +, a V. vulnificusselective medium. The authors reported that, compared to plating on CaV alone, the triple-plating method resulted in a 2.3-fold increase in accurate prediction of V. vulnificus isolates with up to 93% of environmental isolates identified correctly. We found a very similar degree of relative improvement in the identification of V. vulnificus (1.9-fold) in Lake Pontchartrain, and an average absolute correct percentage that was identical (93%) using only two media. However, by comparing data from multiple sampling sites collected at various times of year, we show that the time and location of sampling can have a large influence on the apparent efficacy of multi-plating methods. We did not explicitly evaluate the incidence of false negatives in this study, but previously estimated that the incidence of false negatives by the double-plating method among the Lake Pontchartrain isolates was relatively low (1.5%, 1.7%, and 8.1% for V. cholerae, V. parahaemolyticus, and V. vulnificus, respectively; Nigro et al., 2011). A trade-off that one should consider when adopting multi-plating methods is the diminishing returns in eliminating false positives vs. the increasing chance of generating false negatives with each discriminating criterion added. The negative correlation we observed between total CFUs on TCBS and the percentage of false positives from the double-plating method implies that vibrio blooms were dominated by the human pathogens targeted in this study. If generally true, this suggests that the concentration of CFUs on TCBS will tend to overestimate the concentrations of human pathogenic species when conditions are unfavorable (low temperature, high salinity) and total CFUs are low, but may provide more accurate estimates of their abundance when total CFUs are relatively high. A similar conclusion was reached from others investigating the incidence of false positives on V. vulnificus-selective media (Macián et al., 2000; Froelich et al., 2012). The same relationship was not observed when using CaV to enumerate pathogenic vibrios from Lake Pontchartrain, perhaps because other

29

species that form mauve or blue colonies on CaV were sometimes abundant when pathogenic species were low, particularly in winter and spring when temperatures were low. During the sampling event on January 27, 2006, for example, higher CFUs were observed on CaV plates as compared to TCBS plates. A large portion of these CaVderived isolates would not grow when transferred to TCBS. Previous studies have shown that a variety of bacterial genera will grow on TCBS (Nicholls et al., 1976), however this is the first report of the identities of non-vibrio species capable of growth on CaV. This includes organisms outside of the order Vibrionales and even outside the class Gammaproteobacteria. The isolates that we analyzed that were most distantly related to the genus Vibrio were those that turned mauve on CaV. It thus appears that, in environmental settings, the ability of CaV to distinguish V. parahaemolyticus from even distantly related organisms is weak. 5. Conclusions In conclusion, this study demonstrates that isolation of pathogenic vibrios by culturing sequentially on TCBS and CaV decreases the number of false-positives, although the absolute efficacy is variable among the three target species we tested. Adopting a multi-plating method like this one, or the triple-plating method for V. vulnificus, is simple and inexpensive, and should save time and money (Williams et al., 2013) by reducing the number of false positive isolates that need to be screened by molecular methods. Enumeration by plating is still being used to report semi-discriminatory counts of culturable pathogenic vibrios without confirmation by molecular techniques (e.g., PCR or probing) (Presley et al., 2006; Hsieh et al., 2007; Lara et al., 2009) and multiplating methods can improve such estimates. However, because of the high variability and differential specificity observed using culturebased techniques with environmental samples, we recommend that even a multi-plating method be followed by molecular confirmation (Nigro et al., 2011) to ensure accurate counts. Acknowledgments We are grateful to Edward Laws and Aixin Hou for logistical support of the sampling in Lake Pontchartrain and to Gordon Walker, La'Toya James, and Brett Marchant for assistance in sampling the Ala Wai Canal. We acknowledge James Oliver, Doug Bartlett, and Roger Fujioka for providing V. vulnificus, V. cholerae, and V. parahaemolyticus strains, respectively. This work was supported by grants to G. F. Steward from NSF (OCE05-54768) and Hawaii Sea Grant (NOAA NA09OAR4170060). Additional support was provided by the Pacific Research Center for Marine Biomedicine (NSF OCE04-32479; NIEHS 1P50EF012740-01), the Pacific Island Ocean Observing System (NOAA NA07NOS4730207 and NA08NOS4730299), and the Center for Microbial Oceanography (NSF EF04-24599). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.mimet.2015.01.014. References Bej, A., Patterson, D., Brasher, C., Vickery, M., Jones, D., Kaysner, C., 1999. Detection of total and hemolysin-producing Vibrio parahaemolyticus in shellfish using multiplex PCR amplification of tl, tdh and trh. J. Microbiol. Methods 36, 215–225. Blake, P.A., Merson, M.H., Weaver, R.E., Hollis, D.G., Heublein, P.C., 1979. Disease caused by a marine vibrio. N. Engl. J. Med. 300, 1–5. Brasher, C., DePaola, A., Jones, D., Bej, A., 1998. Detection of microbial pathogens in shellfish with multiplex PCR. Curr. Microbiol. 37, 101–107. Chun, J., Huq, A., Colwell, R., 1999. Analysis of 16S–23S rRNA intergenic spacer regions of Vibrio cholerae and Vibrio mimicus. Appl. Environ. Microbiol. 65, 2202–2208. Colwell, R.R., Huq, A., 1994. Environmental reservoir of Vibrio cholerae: the causative agent of cholera. Ann. N. Y. Acad. Sci. 740, 44–54.

30

O.D. Nigro, G.F. Steward / Journal of Microbiological Methods 111 (2015) 24–30

Di Pinto, A., Terio, V., Novello, L., Tantillo, G., 2011. Comparison between thiosulphatecitrate-bile salt sucrose (TCBS) agar and CHROMagar Vibrio for isolating Vibrio parahaemolyticus. Food Control 22, 124–127. Drummond, A., Ashton, B., Buxton, S., Cheung, M., Cooper, A., Duran, C., et al., 2011. Geneious v5.4. Available from. http://www.geneious.com/. Faruque, S.M., Albert, M.J., Mekalanos, J.J., 1998. Epidemiology, genetics, and ecology of toxigenic Vibrio cholerae. Microbiol. Mol. Biol. Rev. 62, 1301–1314. Froelich, B.A., Williams, T.C., Noble, R.T., Oliver, J.D., 2012. Apparent loss of Vibrio vulnificus from North Carolina Oysters coincides with a drought-induced increase in salinity. Appl. Environ. Microbiol. 78, 3885–3889. Gomez-Gil, B., Roque, A., 2006. Isolation, enumeration and preservation of the Vibrionaceae. In: Thompson, F., Austin, B., Swings, J. (Eds.), The Biology of Vibrios. American Society for Microbiology, Washington, D.C., pp. 15–26. Hara-Kudo, Y., Nishina, T., Nakagawa, H., Konuma, H., Hasegawa, J., Kumagai, S., 2001. Improved method for detection of Vibrio parahaemolyticus in seafood. Appl. Environ. Microbiol. 67, 5819–5823. Hsieh, J.L., Fries, J.S., Noble, R.T., 2007. Dynamics and predictive modelling of Vibrio spp. in the Neuse River Estuary, North Carolina, USA. Environ. Microbiol. 10, 57–64. Lane, D., 1991. 16S/23S rRNA sequencing. In: Stackebrandt, E., Goodfellow, M. (Eds.), Nucleic Acid Techniques in Bacterial Systematics. John Wiley & Sons, New York, pp. 115–175. Lara, R.J., Neogi, S.B., Islam, M.S., Mahmud, Z.H., Yamasaki, S., Nair, G.B., 2009. Influence of catastrophic climatic events and human waste on Vibrio distribution in the Karnaphuli Estuary, Bangladesh. EcoHealth 6, 279–286. Letunic, I., Bork, P., 2011. Interactive Tree Of Life v2: online annotation and display of phylogenetic trees made easy. Nucleic Acids Res. 39, W475–W478 (Suppl.). Ludwig, W., 2004. ARB: a software environment for sequence data. Nucleic Acids Res. 32, 1363–1371. Macián, M.C., Arias, C.R., Aznar, R., Garay, E., Pujalte, M.J., 2000. Identification of Vibrio spp. (other than V. vulnificus) recovered on CPC agar from marine natural samples. Int. Microbiol. 3, 51–53. Morris, R., Rappé, M., Urbach, E., Connon, S., Giovannoni, S., 2004. Prevalence of the Chloroflexi-related SAR202 bacterioplankton cluster throughout the mesopelagic zone and deep ocean. Appl. Environ. Microbiol. 70, 2836–2842. Nicholls, K.M., Lee, J.V., Donovan, T.J., 1976. An evaluation of commercial thiosulphate citrate bile salt sucrose agar (TCBS). J. Appl. Bacteriol. 41, 265–269.

Nigro, O.D., 2012. Environmental Controls on Vibrio Vulnificus and other Pathogenic Vibrios in Tropical and Subtropical Coastal Waters. (Doctoral Dissertation). University of Hawaii at Manoa, Honolulu, HI. Nigro, O.D., Hou, A., Vithanage, G., Fujioka, R.S., Steward, G.F., 2011. Temporal and spatial variability in culturable pathogenic Vibrio spp. in Lake Pontchartrain, Louisiana, following Hurricanes Katrina and Rita. Appl. Environ. Microbiol. 77, 5384–5393. Oliver, J.D., 2005. Wound infections caused by Vibrio vulnificus and other marine bacteria. Epidemiol. Infect. 133, 383–391. Oliver, J.D., 2012. Culture media for the isolation and enumeration of pathogenic vibrio species in foods and environmental samples. In: Corry, E.L., Curtis, G.D.W., Baird, R.M. (Eds.), Handbook of Culture Media for Food and Water Microbiology, third edition Royal Society of Chemistry, Cambridge, pp. 377–402. Pacini, F., 1854. Osservazioni microscophiche e deduzione patologiche sul colera asiatico. Gazz. Med. Ital. 6, 405–412. Panicker, G., Call, D., Krug, M., Bej, A., 2004. Detection of pathogenic Vibrio spp. in shellfish by using multiplex PCR and DNA microarrays. Appl. Environ. Microbiol. 70, 7436–7444. Presley, S., Rainwater, T., Austin, G., Platt, S., Zak, J., Cobb, G., et al., 2006. Assessment of pathogens and toxicants in New Orleans, LA following Hurricane Katrina. Environ. Sci. Technol. 40, 468–474. Pruesse, E., Peplies, J., Glöckner, F.O., 2012. SINA: accurate high throughput multiple sequence alignment of ribosomal RNA genes. Bioinformatics 1–7. Sinigalliano, C.D., Gidley, M.L., Shibata, T., Whitman, D., Dixon, T.H., Laws, E., Hou, A., Bachoon, D., Brand, L., Amaral-Zettler, L., Gast, R.J., Steward, G.F., Nigro, O.D., Fujioka, R., Betancourt, W.Q., Vithanage, G., Matthews, J., Fleming, L.E., SoloGabriele, H.M., 2007. Impacts of Hurricanes Katrina and Rita on the microbial landscape of the New Orleans area. Proc. Natl. Acad. Sci. U. S. A. 104, 9029–9034. Williams, T.C., Froelich, B., Oliver, J.D., 2013. A new culture-based method for the improved identification of Vibrio vulnificus from environmental samples, reducing the need for molecular confirmation. J. Microbiol. Methods 93, 277–283. Wittman, R.J., Flick, G.J., 1995. Microbial contamination of shellfish: prevalence, risk to human health, and control strategies. Ann. Rev. Public Health 16, 123–140. Yarza, P., Ludwig, W., Euzéby, J., Amann, R., Schleifer, K.-H., Glöckner, F.O., et al., 2010. Update of the all-species living tree project based on 16S and 23S rRNA sequence analyses. Syst. Appl. Microbiol. 33, 291–299.