Article pp.397-411 du Vol.27 n°6 (2007)

(1)

doi:10.3166/sda.27.397-412 SCIENCES DES ALIMENTS, 27(2007) 397-411

ARTICLE ORIGINAL ORIGINAL PAPER

Optimisation of selective enrichment for the simultaneous isolation

of Vibrio parahaemolyticus and Vibrio cholerae in frozen raw seafood products

S. Copin¹, A. Rouxel², M. Cornu³, A. Robert-Pillot⁴, M.-L. Quilici⁴, P. Malle¹, J. Lesne²

RÉSUMÉ

Optimisation de l’enrichissement sélectif pour l’isolement simultané de Vibrio parahaemolyticus et Vibrio cholerae dans les produits de la mer crus congelés

L’augmentation de la consommation des produits de la mer importés justifie une surveillance de ces denrées. Le but de notre étude est d’optimiser l’étape d’enrichissement sélectif, avant l’isolement et l’identification, pour la détection simultanée de Vibrio parahaemolyticus et Vibrio cholerae dans les produits de la mer congelés. Dix protocoles de détection ont été comparés dans deux laboratoires sur 59 échantillons naturellement contaminés d’ori- gine géographique variée. L’identification de l’espèce et la recherche des facteurs de pathogénicité ont été réalisées par PCR. Au total, 37 échan- tillons étaient positifs à l’une ou/et l’autre de ces deux espèces ; aucun iso- lat n’était potentiellement pathogène. L’analyse des résultats met en évidence l’importance de la phase d’enrichissement, principalement la durée d’incubation du milieu d’enrichissement. La méthode proposée est appro- priée à la détection simultanée de Vibrio parahaemolyticus et Vibrio cholerae dans les produits de la mer congelés. L’utilisation d’un protocole unique de détection présente un réel avantage lors d’analyses de routine dans les laboratoires de microbiologie. Dans l’attente de méthodes d’identification et de quantification rapides et spécifiques grâce au développement de la biolo-

1. Agence Française de Sécurité Sanitaire des Aliments, Laboratoire d’Études et de Recherches sur les Pro- duits de la Pêche, Boulogne-sur-Mer, France.

2. Ecole Nationale de la Santé Publique, Laboratoire d’Études et de Recherche en Environnement et Santé, Rennes, France.

3. Agence Française de Sécurité Sanitaire des Aliments, Laboratoire d’Études et de Recherches sur la Qua- lité des Aliments et sur les Procédés Agro-alimentaires, Maisons Alfort, France.

4. Institut Pasteur, Centre National de Référence des Vibrions et du Choléra, Paris, France.

* Correspondence : Stéphanie Copin, Agence Française de Sécurité Sanitaire des Aliments, Laboratoire d’Études et de Recherches sur les Produits de la Pêche, rue Huret-Lagache, 62200 Boulogne-sur-Mer, France.

e-mail: [email protected] – Telephone: 33 (0)3-21-99-25-00 – Fax: 33 (0)3-21-30-95-47 SDA27_6.book Page 397 Lundi, 18. août 2008 4:26 16

(2)

398 Sci. Aliments 27(6), 2007 S. Copin et al.

gie moléculaire, la méthode de routine proposée devrait permettre de mieux isoler les principaux Vibrio potentiellement pathogènes dans les produits de la mer congelés.

Mots clés

enrichissement, isolement, Vibrio parahaemolyticus, Vibrio cholerae, produit de la mer congelé.

SUMMARY

As consumption of imported seafood products increases, makes the moni- toring of those foods more essential. We sought to optimise selective enrichment before isolation and identification for the simultaneous detection of Vibrio parahaemolyticus and Vibrio cholerae in frozen seafood products. Ten detection protocols were compared in two laboratories using 59 natural (unspiked) samples from various geographic areas. Polymerase chain reac- tion (PCR) was used to identify the species and search for pathogenicity factors. In all, 37 samples were positive for one or both of these vibrio species.

No isolate was potentially pathogenic. Data analysis underscored the importance of enrichment, notably its duration. The method proposed is suitable for the simultaneous detection of Vibrio parahaemolyticus and Vibrio cholerae in frozen seafood products. The use of a single detection protocol is a real advantage in routine analyses in microbiology laboratories. Until develop- ments in molecular biology yield rapid and specific methods of identification and quantification, the routine method proposed should improve isolation of the main potentially pathogenic vibrios in frozen seafood products.

Keywords

enrichment, isolation, Vibrio parahaemolyticus, Vibrio cholerae, frozen sea- food products.

1 – INTRODUCTION

Vibrios are Gram-negative, halophilic or halotolerant, aerobic/anaerobic, oxidase-positive bacilli. They are found in marine environments and more particu- larly in coastal waters and estuaries. Their numbers vary seasonally and peak in the warmest months (Cook et al., 2002a; Cook et al., 2002b; Croci et al., 2001;

Depaola et al., 2003).

The Vibrio genus is highly diverse and includes to date over 70 known species (Fournier and Quilici, 2002). Two potentially pathogenic Vibrio populations can be distinguished epidemiologically. The first is of cholera vibrios and com- prises strains of two serogroups (O1 and O139) of the species Vibrio cholerae which cause cholera pandemics (Kaper et al., 1995). The second population is of non-cholera vibrios and includes non-O1 and non-O139 Vibrio cholerae strains and 11 other species including Vibrio parahaemolyticus and Vibrio vuln-

SDA27_6.book Page 398 Lundi, 18. août 2008 4:26 16

(3)

Detection of potentially pathogenic vibrios in seafood products 399

ificus. Vibrio cholerae, Vibrio parahaemolyticus and Vibrio vulnificus are the three main species implicated in human disease around the world (Lesne and Fournier, 2005; Morris, 2003). Humans are exposed to these vibrios through contact with the sea and its products. Vibrio parahaemolyticus and non-O1/

non-O139 Vibrio cholerae are a major cause of collective food poisoning worldwide (Balakrish Nair and Hormazabal Opazo, 2005; Martinez-Urtaza et al., 2005;

Morris, 2003; Quilici et al., 2005). These two species cause gastroenteritis after the eating of seafood products that are raw, insufficiently cooked or cooked and then recontaminated. As yet there is no clear evidence worldwide for gastrointestinal infection by Vibrio vulnificus. Certain Vibrio parahaemolyticus isolates produce two hemolysins: a thermostable direct hemolysin (TDH) and a TDH-related hemolysin or TRH (Honda and Iida, 1993; Shirai et al., 1990). Only strains that produce one of these two hemolysins (or more rarely both) are human pathogens. Therefore, the presence of Vibrio parahaemolyticus does not necessarily involve a health risk as the vast majority of strains isolated in the environment and in seafood products are not pathogenic (Robert-Pillot et al., 2004). The virulence of O1 and O139 Vibrio cholerae is associated with the production of cholera toxin (Lipp et al., 2003; Rivera et al., 2001).

To minimize the risks of food poisoning by vibrios, reliable analytical methods are needed to check the bacteriological quality of seafood products.

There are various methods of detecting these bacteria, including those of the Food and Drug Administration (FDA) (Elliot et al., 1995), the International Organi- zation for Standardization (ISO) (ISO8914, 1990) and the Nordic Committee on Food Analysis (NMKL) (NMKL, 1997). In 2007, the ISO published a standard that uses a horizontal method for the detection of vibrios presumed to be pathogenic by the gastrointestinal route (ISO-TS21872, 2007). In the first part of this standard, a single protocol is described for the detection of Vibrio parahae- molyticus and Vibrio cholerae (a protocol for the detection of other species is given in the second part). This standard applies to foods in general and to environmental samples collected in production and processing areas. In practice, it concerns fresh and frozen fishery products, and shellfish. A single, reliable protocol usable in all laboratories for routine detection of vibrios pathogenic by the gastrointestinal route (principally Vibrio parahaemolyticus and Vibrio cholerae) must be based on experimental data. We compared several selective enrichment protocols to determine which best simultaneously detects Vibrio parahae- molyticus and Vibrio cholerae in frozen raw seafood products (our aim was not to observe the prevalence of these two species in sea products). Enrichment must favour the selective growth of the target microorganism while inhibiting accompanying marine flora, one of the main obstacles to the analysis of Vibrio spp. in sea products. Nevertheless, the liquid medium used must not be too selective as we need to recover stressed or damaged target bacteria: as Vibrio are cold-sensitive, if they come from a frozen sample they may be in a physiological state that does not enable them to grow. Official methods differ widely in sample dilution factor. The enrichment medium most frequently used to detect vibrios is alkaline peptone water (APW) whose NaCl concentration varies accor- ding to the protocol. The most common incubation temperature is 37°C but a higher temperature (41.5°C or even 42°C) would give better results by impro- ving the selectivity of the medium regarding certain vibrio species (Depaola and Hwang, 1995; Nmkl, 1997). Also, as vibrios grow fast, short (6 hours) incubation is often recommended for the enrichment, although some researchers report

(4)

that recovery of vibrios is better after 18-24 hours (Depaola and Hwang, 1995;

Hara-Kudo et al., 2001; Ottaviani et al., 2003; Parvathi et al., 2004). The purpose of the present study was to evaluate different culture parameters in order to help in the improvement of current standards for simultaneous isolation of Vibrio parahaemolyticus and Vibrio cholerae in frozen seafood products.

2 – MATERIALS AND METHODS

2.1 Sampling

Fifty-nine samples of frozen raw prawns from various geographic areas were collected from an import-export company and several French food testing laboratories. The products were transported to their final destination at – 18°C. Each batch received was divided equally between the Laboratoire d’Etudes et de Recherches sur les Produits de la Pêche, AFSSA (Agence Française de Sécurité Sanitaire des Aliments) at Boulogne-sur-Mer (France) and the Laboratoire d’Étu- des et de Recherche en Environnement et Santé, ENSP (École Nationale de la Santé Publique) in Rennes (France). The samples were stored at – 18°C until analysis.

2.2 Experimental design

We compared ten combinations of the following culture parameters: homogenate-enrichment broth dilution factor, enrichment medium composition, the temperature and length of incubation. The experimental design is shown in figure 1. The raw prawn samples were thawed for 2 hours at room temperature.

A 1/2 dilution homogenate was prepared by homogenising (2 x 1 min) in a sterile Waring Blender 120 g of each sample and 120 mL of tryptone salt broth.

This 1/2 dilution homogenate was then diluted five-fold to give a 1/10 dilution by transferring under sterile conditions 50 g of the homogenate into a sterile glass 500 mL bottle (Schott Duran) with 200 mL of APW containing 1% NaCl (1% APW) (peptone: 10 g/L; NaCl: 10 g/L; pH: 8.6 ± 0.2) (homogenate A₁₀) or with APW containing 2% NaCl (2% APW) (peptone: 10 g/L; NaCl: 20 g/L; yeast extract: 3 g/L; pH: 8.6 ± 0.2)) (homogenate B₁₀). The 1/2 dilution homogenate was also diluted 50-fold to give a final dilution of 1/100by adding 5 g of the homogenate under sterile conditions to 245 mL of the same enrichment media (1% APW and 2% APW) to give respectively homogenates A₁₀₀ andB₁₀₀. The homogenates in 1% APW (A₁₀ and A₁₀₀) were incubated for 5 to 7 hours at 37°C ± 1°C to produce the first enrichments P1 and P’1. The homogenates in 2% APW (B₁₀ and B₁₀₀) were incubated for 5 to 7 hours at 41.5°C ± 0.5°C to produce the first enrichments P3 and P’3 (Pi indicates the enrichment protocols at 1/10 dilution and P’i the enrichment protocols at 1/100 dilution). These enrichment cultures (P1 and P’1, P3 and P’3) were then extended for 12 to 18 hours at their respective temperatures to give the second enrichments, respectively P2 and P’2, P4 and P’4. A subculture was done in parallel from each of the first enrichments in 2% APW (P3 and P’3). Three successive 1 µL sam-

(5)

ples were taken from the surface of enrichment cultures P3 and P’3 using a calibrated loop and transferred to 100 mL of 2% APW in 250 mL sterile glass bottles, which were incubated for 12 to 18 hours at 41.5 ˚C ± 0.5°C to give the second enrichments P5 and P’5. The two laboratories applied each protocol (P1 to P5, P’1 to P’5) in the same way to the samples of frozen raw prawns.

Each protocol was evaluated using one negative control and two positive controls. These controls were prepared using a sample of frozen raw prawns not contaminated by Vibrio spp. which has not been exposed to ionising radia- tion so as not to alter competition from the other flora present in this type of product. The negative control was prepared by adding 120 g of this uncontami-

5 g of homogenate + 245 ml of 1% APW For 5 –7 H at 37°C

P1

P2

P’1

P’2 For 5 –7 H at 37°C

For 12 –18 H at 37°C

50 g of homogenate + 200 ml of 2% APW

For 5 –7 H at 41.5°C

Subculture P5

P3

P4 For 12 –18 H

at 41.5°C

For 5 –7 H at 41.5°C

Subculture P’5

P’3

P’4 50 g of homogenate +

200 ml of 1% APW

0 0 1 0 B

1 0 B

1

A A100

1/2 homogenate

120 g of sample + 120 ml of tryptonesalt broth Homogenised (2 x 1 min) in a Waringblender Frozen raw prawns thawed for 2 hours at room temperature

1^stenrichment

2^ndenrichment

subculture 5 g of homogenate + 245 ml of 2% APW

For 12 –18 H at 41.5°C For 12 –18 H

at 37°C

Figure 1 Experimental design.

A: homogenates in APW containing 1% NaCl; A₁₀: 1/10 homogenate, A₁₀₀: 1/100 homogenate (A₁₀₀= 1/10 of A10).

B: homogenates in APW containing 2% NaCl; B₁₀: 1/10 homogenate, B₁₀₀: 1/100 homogenate (B₁₀₀ = 1/10 of B₁₀).

Pi: enrichment protocol at 1/10 dilution.

P'i: enrichment protocol at 1/100 dilution.

Subculture: 3 times 1 µl collected using a 1µ loop and seeded in 100 mL of APW containing 2% NaCl in a 250 mL bottle.

Représentation schématique du plan d’expérience.

A : broyats en EPSA 1 % NaCl ; A10 : broyat au 1/10ê, A100 : broyat au 1/100ê (A100 = 1/10ê de A10).

B : broyats en BSA 2 % NaCl ; B10 : broyat au 1/10ê, B100 : broyat au 1/100 ê (B100 = 1/10ê de B10).

Pi : protocole d'enrichissement à la dilution 1/10^e. P'i : protocole d'enrichissement à la dilution 1/100^e. Subculture : prélever 3 fois 1 µl à l'aide d'une öse de 1 µl en surface à ensemencer dans 100 mL de BSA 2 % NaCl réparti en flacon de 250 mL.

(6)

nated sample to 120 mL of tryptone salt broth to obtain a 1/2 dilution homogenate. The positive controls were prepared in the same way and then artificially contaminated by Vibrio cholerae and Vibrio parahaemolyticus using strains chosen randomly from those that were isolated and identified during this study.

Each bacterial strain was inoculated in 5 mL of APW containing 3% NaCl, incubated overnight at 37°C, then inoculated on a nutrient agar slant containing salt and incubated for 24 h at 37°C. The bacteria grown were suspended in 2 mL of physiological saline. After measuring optical density (OD) at 620 nm, the bacterial suspension was diluted to 100 CFU/mL and 1000 CFU/mL in 240 mL of the 1/2 dilution homogenate. The 10 protocols (P1 to P5, P’1 to P’5) were then applied to the negative control and the two positive controls (figure 1).

2.3 Treatment of each enrichment

From each enrichment broth (P1 to P5, P’1 to P’5) an isolation was done on a selective medium, thiosulphate-citrate-bile-salts-sucrose (TCBS; Oxoid Ltd, Basingstoke, Hampshire, England) agar, either by spreading decimal dilutions of the broth, or by streaking with a calibrated loop on agar. The growth of bacteria on TCBS agar was observed after 24 hours at 37°C ± 1°C and the presumed Vibrio spp. colonies were selected: large, smooth green-blue colonies (sucrose-negative) for Vibrio parahaemolyticus and colonies circular, flat and yellow (sucrose-positive) for Vibrio cholerae. To guide the selection, a maximum of 32 sucrose-negative colonies were seeded on nutrient agar containing 6%

NaCl (growth of Vibrio parahaemolyticus) and a maximum of 32 sucrose-positive colonies were seeded on nutrient agar containing 0% NaCl (growth of Vibrio cholerae) using sterile toothpicks. These nutrient agars were incubated for 24 hours at 41.5°C ± 0.5°C or 37°C ± 1°C, the temperatures chosen for the enrichments. Lastly, the oxidase test was done on the colonies that grew on nutrient agar 6% NaCl or 0% NaCl. Four oxidase-positive and sucrose-negative colonies and four oxidase-positive and sucrose-positive colonies were selected and isolated on nutrient agar containing 2% NaCl to give pure cultures after 24 hours at 37°C ± 1°C.

After study of ADH (Arginine Dihydrolase) and ONPG (Orthonitro-phenyl- beta-D-galactopyranoside), only sucrose-negative/oxidase-positive/ONPG- negative/ADH-negative colonies (suspicion of Vibrio parahaemolyticus) and sucrose-positive/oxidase-positive/ONPG-positive/ADH-negative colonies (suspicion of Vibrio cholerae) were kept for the following tests. The two species of Vibrio detected were then identified by means of the API 20E test system (Bio- mérieux, France) and a growth test in tubes of APW containing 0, 8 and 10%

NaCl.

2.4 Confirmation of identification of strains and detection of pathogenicity factors by PCR

Molecular techniques were used to confirm the biochemical identification of the species and to detect pathogenicity factors, as recommended by a joint FAO/WHO Expert Consultation (FAO/WHO, 2002). Antigenic identification of Vibrio cholerae was done using the polyvalent anti-O1 serum (Sanofi Diagnos- tics Pasteur, ref: 57 142) and anti-O139 serum (Institut Pasteur, Paris). Toxino- genic power was studied by amplification of the ctx A and ctx B genes

(7)

encoding cholera toxin. This study was done at the Centre National de Réfé- rence des Vibrios et du Choléra (Institut Pasteur, Paris).

Identification of Vibrio parahaemolyticus was confirmed by amplification of the sequence R72H (PCR R72H) and screening for pathogenicity factors by amplification of the genes tdh and trh (PCR tdh and trh), at the Centre National de Référence des Vibrios et du Choléra (Institut Pasteur, Paris) and the Labora- toire d’Études et de Recherches sur les Produits de la Pêche (AFSSA, Boulo- gne-sur-Mer).

DNA extractions and PCR amplifications were done using published methods. The sequences of the primers used for PCR and the references are presented in table 1.

Table 1

Target genes and oligonucleotide primer sequences for PCR.

Tableau 1

Gènes cibles et séquences des amorces oligonucléotidiques utilisées pour la PCR.

2.5 Statistical analysis

A test was considered positive for a species (Vibrio cholerae or Vibrio para- haemolyticus) if the vibrio was detected by the protocol and if its identification was confirmed by PCR.

A prawn sample was positive for a species (Vibrio cholerae or Vibrio para- haemolyticus) if at least one of the 20 analyses done (10 protocols in each of the 2 laboratories) was positive.

Two parameters were defined to evaluate and compare the protocols. Pro- tocol sensitivity (%) with regard to a species is the number of analyses indica- ted as positive by this protocol (in the two laboratories) over the total number of positive samples. The choice of this uncommon definition is due to the fact that we preferred to base the evaluation of sensitivity mostly on the natural unspiked samples. Protocol reproducibility (%) is the number of results on which the two laboratories were in agreement (“absence/absence” and “presence/presence”)

Target

gene Primers Sequences Amplicon

size (bp) References pR72H VP 32 5’ - CGA ATC CTT GAA CAT ACG CAG C - 3’ 320 or 387

VP 33 5’ - TGC GAA TTC GAT AGG GTG TTA ACC - 3’

tdh VP 21 5’ - TGG TTG ACA TCC TAC ATG ACT GTG - 3’ 400 VP 22 5’ - GGG GAT CCC TCA GTA CAA AGC CTT - 3’

trh Trh 1 5’ - CTC TAC TTT GCT TTC AGT - 3’ 460 Trh 2 5’ - AAT ATT CTG GAG TTT CAT - 3’

ctx A CTX2 5’ - CGG GCA GAT TCT AGA CCT CCT G - 3’ 564 CTX3 5’ - CGA TGA TCT TGG AGC ATT CCC AC - 3’

ctx B CTX7 5’ - GGT TGC TTC TCA TCA TCG AAC CAC - 3 460 CTX9B 5’ - GAT ACA CAT AAT AGA ATT AAG GAT - 3’

OLSVIK et al., 1993 LEE and PAN, 1995 LEE et al., 1993

SUTHIENKUL et al., 1995 FIELDS et al., 1992 SDA27_6.book Page 403 Lundi, 18. août 2008 4:26 16

(8)

over the total number of results. For protocol evaluation, reproducibility cannot be considered independently of sensitivity. The closer the two percentages are to 100%, the better the protocol.

For each percentage (sensitivity or reproducibility), a 95% confidence interval was calculated, where necessary taking into account the Clopper-Pearson interval for the low numbers (Lamy et al., 2004).

3 – RESULTS

Each of 59 samples of whole frozen prawns naturally contaminated by Vibrio spp. was analysed in parallel in the two study laboratories, so 118 analyses in all were done. Of the 59 samples, 37 were positive for Vibrio cholerae or Vibrio parahaemolyticus (or for both). Of the 118 analyses, 28 were positive for Vibrio cholerae and 37 positive for Vibrio parahaemolyticus. These results, which were confirmed by PCR or the agglutination test, are presented in table 2. As no more than four characteristic colonies were selected after isolation, the 10 protocols isolated 95 sucrose-positive strains, which the API 20E test system identified as Vibrio cholerae. After confirmation, 93 were found to be non-O1/non- O139 Vibrio cholerae strains. No strain had the ctxA and ctxB genes coding respectively for the A and B subunits of the cholera toxin. The API 20E test identified 161 sucrose-negative strains as Vibrio parahaemolyticus, and this was confirmed by PCR for 154 of the 161 strains. No strain had the genes coding for TDH or TRH. We also detected other species in our samples, such as Vibrio alginolyticus and Vibrio vulnificus (data not shown).

3.1 Between-laboratory comparison of protocol efficiency

Figure 2 presents pooled sensitivity data from the two laboratories for each of the 10 protocols (P1 to P5 and P’1 to P’5) used in simultaneous detection of Vibrio cholerae and Vibrio parahaemolyticus in frozen naturally contaminated raw prawns. Protocol P2 (incubation at 37°C) was clearly the most efficient at simultaneous detection of the two species, with 79% sensitivity for Vibrio chole- rae and 73% for Vibrio parahaemolyticus. Protocol P4 (incubation at 41.5°C) did not differ significantly from P2, given the overlap of the confidence intervals, and can therefore also be used for the simultaneous detection of the two species in frozen seafood products. Figure 2 shows that protocols P3 and P5 did not differ significantly from P2 and P4 in isolation of Vibrio parahaemolyticus, or from P4 in isolating Vibrio cholerae (overlap of confidence intervals). Protocols P3 and P5, however, differed significantly from P2 in the detection of Vibrio cholerae. Protocols P’2 and P’4 did not differ significantly from P2 and P4 for the detection of Vibrio cholerae but were less efficient in detection of Vibrio parahaemolyticus. These four protocols (P3, P5, P’2, P’4) are therefore not suitable for simultaneous isolation of Vibrio cholerae and Vibrio parahaemolyticus.

(9)

Table 2

Results of isolation of V. cholerae and V. parahaemolyticus from samples of frozen raw whole prawns (results confirmed by the agglutination test or by PCR).

Tableau 2

Résultats de l’isolement de V. cholerae et V. parahaemolyticus dans les échantillons de crevettes crues congelées entières (résultats confirmés par le test de l’agglutination ou par PCR).

L1 L2 L1 L2

1 Bangladesh – – + +

2 Ecuador + + – –

3 New caledonia + – – –

4 South America + – – –

7 Honduras + + – –

9 Ecuador + + + +

10 Ecuador – + – –

11 Brazil – + – –

12 Africa – – + +

13 Ecuador + + + +

14 Ecuador + – + –

15 Ecuador + – – –

17 Ecuador + – – –

18 Ecuador + + – –

19 Ecuador + + + +

20 Brazil + – – –

21 Madagascar + – – +

22 Unknown + + – –

23 Unknown + – – –

24 Unknown – + – –

25 Unknown + + – –

26 Unknown + – – –

27 Thailand – – + +

28 Thailand – – + +

29 Thailand + – + +

30 Malaysia + – – –

31 Malaysia + – – –

32 Indonesia – – + +

33 Brazil – – + –

34 Indonesia – – + +

35 Malaysia – – + +

36 Bangladesh – – + –

37 Ecuador – – + +

* Frozen raw headless prawns L1 : AFSSA laboratory L2 : ENSP laborator

+ : at least one positive result in 10 tests –

* L1 L2 + – : no positive result in 10 tests

Sample number

Geographical origin

Identification Vibrio

parahaemolyticus

Vibrio cholerae

*

Crevettes crues congelées sans têtes : laboratoire de l'AFSSA

: laboratoire de l'ENSP

: au moins une analyse positive parmi les 10 essais : aucune analyse positive parmi les 10 essais SDA27_6.book Page 405 Lundi, 18. août 2008 4:26 16

(10)

The artificial contamination data presented in table 3 confirm that the 10 protocols detected the two vibrio species in the two positive controls.

Nevertheless, three protocols (P1, P3, P’5) failed in simultaneous detection of the two species, and so can be excluded. All the protocols gave a negative result for the negative control.

When considering protocols P (1/10 dilution) and P’ (1/100 dilution) in figure 2, detection of Vibrio cholerae and Vibrio parahaemolyticus was better with the 1/10 dilution of homogenate in the enrichment broth.

The percentage agreement between the two laboratories for the 10 protocols (P1 to P5 and P’1 to P’5) is presented in figure 3. The percentages are high for all protocols with regard to Vibrio cholerae, but more irregular for Vibrio parahaemolyticus, the best being P1 and all the P’ protocols. These results do not call into question the choice of protocols P2 and P4, which was made on the basis of the best compromise in the light of the sensitivity results.

3.2 Comparison of protocol efficiency

When considering only similar results per sample in the two laboratories, for one or other or both of the two species, 18 results were in agreement, of which 11 for Vibrio cholerae and 10 for Vibrio parahaemolyticus. Analysis of percentage sensitivity of the P protocols (1/10 dilution) in figure 4 shows that protocols P2 and P4 were the most efficient in simultaneous detection of Vibrio cholerae (respectively 77 and 55%) and Vibrio parahaemolyticus (respectively 80 and 70%). Percentage agreement between the protocols was also high: 83 and 71% (P2), 78 and 71% (P4) respectively for the detection of Vibrio cholerae and Vibrio parahaemolyticus (data not shown).

0 10 20 30 40 50 60 70 80 90 100

P1 P2 P3 P4 P5 P’1 P’2 P’3 P’4 P’5 Protocol

% sensitivity

Vibrio cholerae Vibrio parahaemolyticus

Figure 2

Sensitivity (%) of different protocols in simultaneous isolation of Vibrio parahaemolyticus and V. cholerae from frozen naturally contaminated raw prawns: pooled results

from the two laboratories.

Sensibilité (en %) des différents protocoles pour l’isolement simultané de Vibrio parahaemolyticus et V. cholerae dans des crevettes crues congelées

naturellement contaminées : résultats groupés des deux laboratoires.

(11)

Table 3

Simultaneous detection of Vibrio cholerae and Vibrio parahaemolyticus in artificially contaminated controls.

Tableau 3

Détection simultanée de Vibrio cholerae et de Vibrio parahaemolyticus dans les témoins artificiellement contaminés.

V.c V.p V.c V.p V.c V.p

P1 - - * + * +

P2 - - + + + +

P3 - - + + * +

P4 - - + + + +

P5 - - + + + +

P’1 - - + + + +

P’2 - - + + + +

P’3 - - + + + +

P’4 - - + + + +

P’5 - - + + + ††

Enrichment protocols

Control - Control + 100 CFU/mL Control + 1000 CFU/mL

V.c : Vibrio cholerae ; V.p : Vibrio parahaemolyticus

* : no characteristic colonies on TCBS agar

* : no characteristic sucrose-positive on TCBS agar

+ : characteristic colonies on TCBS agar confirmed as being Vibrio cholerae or Vibrio parahaemolyticus

†† : presence of characteristic colonies on TCBS agar not confirmed as being Vibrio cholerae or Vibrio parahaemolyticus V.c : Vibrio cholerae ; V.p : Vibrio parahaemolyticus

- : absence de colonies caractéristiques sur les boîtes de TCBS

* : absence de colonies caractéristiques saccharose + sur les boîtes de TCBS

+ : présence de colonies caractéristiques sur TCBS confirmées appartenir aux identifications des espèces Vibrio cholerae ou Vibrio parahaemolyticus

†† : présence de colonies caractéristiques sur TCBS non confirmées appartenir aux espèces Vibrio cholerae ou de Vibrio parahaemolyticus

0,00 10,00 20,00 30,00 40,00 50,00 60,00 70,00 80,00 90,00 100,00

Protocol

% agreement

P1 P2 P3 P4 P5 P’1 P’2 P’3 P’4 P’5

Figure 3

Reproducibility (%) of different protocols in simultaneous isolation of Vibrio parahaemolyticus and V. cholerae from frozen naturally contaminated

raw prawns: agreement of the two laboratories.

Reproductibilité (en %) des différents protocoles pour l’isolement simultané de Vibrio parahaemolyticus et V. cholerae dans des crevettes crues congelées

naturellement contaminées : concordance des deux laboratoires.

(12)

4 – DISCUSSION

We studied the effect on enrichment of various parameters (homogenate dilution factor, medium composition, temperature and length of incubation) so as optimise conditions for simultaneous isolation on selective medium of Vibrio cholerae and Vibrio parahaemolyticus from frozen seafood products. We com- pared different combinations of these parameters in a large number of naturally contaminated samples. Significantly better simultaneous detection of Vibrio cholerae and Vibrio parahaemolyticus was noted with 1/10 dilution in APW con- taining 1% NaCl and incubation for 18-24 hours at 37°C ± 1°C (protocol P2) and 1/10 dilution in APW containing 2% NaCl incubated for 18-24 hours at 41.5°C ± 0.5°C (protocol P4). Between-protocol and between-laboratory comparison of the results shows that heterogenous contamination is possible within a given batch of seafood products, as already described for Listeria monocyto- genes (Lappi et al., 2004; Midelet-Bourdin et al., 2007).

Our results indicate that for the type of product we have analysed, a 1/100 dilution of homogenate in enrichment broth reduces the proportion of positive results compared with the 1/10 dilution. We often observed little or even no growth when TCBS agar was seeded with a 1/100 dilution. These results contradict those obtained with oyster samples in a study of detection of Vibrio cholerae where a 1/100 dilution gave better results than a 1/10 dilution (Depaola and Hwang, 1995).

0 20 40 60 80 100

P1 P2 P3 P4 P5 Protocol

% sensitivity

Figure 4

Sensitivity (%) of different protocols (1/10 dilution) in simultaneous isolation of Vibrio parahaemolyticus and V. cholerae from frozen naturally contaminated

raw prawns: between-laboratory agreement in results for each sample.

Sensibilité (en %) des différents protocoles (dilution au 1/10^e) pour l’isolement simultané de Vibrio parahaemolyticus et V. cholerae dans des crevettes crues congelées

naturellement contaminées : résultats concordants pour chaque échantillon dans les deux laboratoires.

(13)

An initial experimental study (data not shown) showed that the composition of the enrichment medium (1% APW or 2% APW) did not significantly affect isolation of the bacteria.

Although a previous study on oysters showed that incubation of the enrichment medium at 42°C was better for isolation of O1 Vibrio cholerae (Depaola et al., 1987), our results show, for frozen raw seafood products, that isolation was less efficient after incubation at 41.5°C ± 0.5°C (P4) than at 37°C ± 1°C (P2). It should be remembered that we worked with samples of frozen raw prawns. Our results are compatible with the ISO technical specification (TS 21872) which recommends a temperature of 37°C ± 1°C for the enrichment phase of frozen products and a temperature of 41.5°C ± 0.5°C for fresh products. Our findings confirm that the incubation temperature of the enrichment medium is important and will depend on the products analysed (fresh, processed, frozen…).

Finally, although several official methods recommend 6- to 8-hour incubation for the enrichment, our results show that 18-24 hours was better for detection of V. cholerae and V. parahaemolyticus in frozen raw prawns. These data tally with literature findings, in particular for V. cholerae isolated from water (Spira and Ahmed, 1981) and for O1 Vibrio cholerae enumerated in samples of sterile ice (Cava et al., 2006). Our results clearly show that neither the enrichments of 5-7 hours (P1 and P3) nor the subcultures of 18-24 hours done after 6 hours (P5) significantly improve detection of the two vibrio species in frozen raw prawns. These differences may be due to the low concentrations of bacteria at the start of enrichment or to better repair of freeze-injured bacteria after 18-24 hours of enrichment in APW.

5 – CONCLUSIONS

The length of enrichment appears to be a key parameter before isolation on selective medium for simultaneous detection of Vibrio cholerae and Vibrio para- haemolyticus in frozen seafood products. We have shown that a single protocol can efficiently isolate and reliably identify these bacteria. The use of a common protocol and of 18- to 24-hour incubation has a clear advantage for routine analysis of frozen products in a microbiology laboratory. In the case of frozen seafood products, we recommend a 1/10 dilution in APW containing 1% NaCl and incubation for 18-24 hours at 37°C ± 1°C, Those results, pending the even- tual publication of similar observations, are expected to improve the current standards.

ACKNOWLEDGEMENTS

The authors would like to thank the Direction Générale de l’Alimentation, the Ministry of Agriculture, France for funding this study. Our thanks also go to

(14)

Dr. Graziella Midelet-Bourdin for help with the analysis of the results. Lastly, we thank the companies and laboratories that supplied us with samples of frozen raw prawns.

REFERENCES

BALAKRISH NAIR G., HORMAZABAL OPAZO J. C., 2005. The Vibrio parahaemolyticus pandemic. Rev Chilena Infectol, 22, 125- 130.

CAVA R., SANGRONIS E., MARIN-INIESTA F., 2006. Comparison of methods for recov- ering Vibrio cholerae O1 from ice. J Food Prot, 69, 217-220.

COOK D. W., BOWERS J. C., DEPAOLA A., 2002a. Density of total and pathogenic (tdh+) Vibrio parahaemolyticus in Atlantic and Gulf coast molluscan shellfish at har- vest. J Food Prot, 65, 1873-1880.

COOK D. W., OLEARY P., HUNSUCKER J. C., SLOAN E. M., BOWERS J. C., BLOD- GETT R. J., DEPAOLA A., 2002b. Vibrio vulnificus and Vibrio parahaemolyticus in U.S. retail shell oysters: a national survey from June 1998 to July 1999. J Food Prot, 65, 79-87.

CROCI L., SERRATORE P., COZZI L., STAC- CHINI A., MILANDRI S., SUFFREDINI E., TOTI L., 2001. Detection of Vibrionaceae in mussels and in their seawater growing area. Lett Appl Microbiol, 32, 57-61.

DEPAOLA A., HWANG G. C., 1995. Effect of dilution, incubation time, and temperature of enrichment on cultural and PCR detection of Vibrio cholerae obtained from the oyster Crassostrea virginica. Mol Cell Probes, 9, 75-81.

DEPAOLA A., KAYSNER C. A., MCPHEAR- SON R. M., 1987. Elevated temperature method for recovery of Vibrio cholerae from oysters (Crassostrea gigas). Appl Environ Microbiol, 53, 1181-1182.

DEPAOLA A., NORDSTROM J. L., BOWERS J. C., WELLS J. G., COOK D. W., 2003.

Seasonal abundance of total and pathogenic Vibrio parahaemolyticus in Alabama oysters. Appl Environ Microbiol, 69, 1521- 1526.

ELLIOT E. L., KAYSNER C. A., JACKSON L., TAMPLIN M. L., 1995. Chapter 9. Vibrio cholerae, V. parahaemolyticus, V. vulnifi- cus, and other Vibrio spp. FDA Bacterio- logical analytical manual, 8^th edition.

FAO/WHO, (2002). Risk assessment of Cam- pylobacter spp. in broiler chickens and Vibrio spp. in seafood. In. Report of a joint FAO/WHO Expert consultation, 5-9 August 2002, Bangkok, Thailand.

FOURNIER J.-M., QUILICI M.-L., 2002. Infec- tions à vibrions non cholériques. Encyclo- pédie Médico-chirurgicale (Ed.

Scientifiques et Médicales Elsevier SAS), Maladies infectieuses, 8-026-F-15, 7 p.

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.

HONDA T., IIDA T., 1993. The pathogenicity of Vibrio parahaemolyticus and the role of the thermostable direct haemolysin and related haemolysins. Reviews in Medical Microbiology, 4, 106-113.

ISO8914, 1990. Microbiology – General Gui- dance for the Detection of Vibrio parahe- molyticus. International Organisation for Standardisation, Geneva.

ISO-TS21872, 2007. Microbiology of food and animal feeding stuffs – Horizontal method for the detection of potentially enteropa- thogenic Vibrio spp. International Organi- sation for Standardisation.

KAPER J. B., MORRIS J. G., JR., LEVINE M.

M., 1995. Cholera. Clin Microbiol Rev, 8, 48-86.

LAMY B., DELIGNETTE-MULLER M. L., BATY F., CARRET G., 2004. Simple table for estimating confidence interval of discre- pancy frequencies in microbiological safety evaluation. Journal of Microbiologi- cal Methods, 56, 137-139.

(15)

LAPPI V. R., HO A., GALL K., WIEDMANN M., 2004. Prevalence and growth of Listeria on naturally contaminated smoked sal- mon over 28 days of storage at 4 degrees C. J Food Prot, 67, 1022-1026.

LESNE J., FOURNIER J.-M., 2005. Vibrio In Bactériologie alimentaire. FEDERIGHI Michel ed., Economica, pp.189-217.

LIPP E. K., RIVERA I. N., GIL A. I., ESPELAND E. M., CHOOPUN N., LOUIS V. R., RUSSEK-COHEN E., HUQ A., COLWELL R. R., 2003. Direct detection of Vibrio cholerae and ctxA in Peruvian coastal water and plankton by PCR. Appl Environ Microbiol, 69, 3676-3680.

MARTINEZ-URTAZA J., SIMENTAL L., VELASCO D., DEPAOLA A., ISHIBASHI M., NAKAGUCHI Y., NISHIBUCHI M., CARRERA-FLORES D., REY-ALVAREZ C., POUSA A., 2005. Pandemic Vibrio parahaemolyticus O3:K6, Europe. Emerg- ing Infectious Diseases, 11, 1319-1320.

MIDELET-BOURDIN G., LELEU G., MALLE P., 2007. Evaluation of the international refe- rence methods NF EN ISO 11290-1 and 11290-2 and an in-house method for the isolation of Listeria monocytogenes from retail seafood products in france. J Food Prot, 70, 891-900.

MORRIS J. G., JR., 2003. Cholera and other types of vibriosis: a story of human pandemics and oysters on the half shell. Clin Infect Dis, 37, 272-280.

NMKL, 1997. Method n˚ 156 (UDC 576.851).

Nordic Committee on Food Analysis.

OTTAVIANI D., MASINI L., BACCHIOCCHI S., 2003. A biochemical protocol for the isolation and identification of current species

of Vibrio in seafood. J Appl Microbiol, 95, 1277-1284.

PARVATHI A., KUMAR H. S., KARUNASAGAR I., 2004. Detection and enumeration of Vibrio vulnificus in oysters from two estuaries along the southwest coast of India, using molecular methods. Appl Environ Microbiol, 70, 6909-6913.

QUILICI M. L., ROBERT-PILLOT A., PICART J., FOURNIER J. M., 2005. Pandemic Vibrio parahaemolyticus O3:K6 spread, France [letter]. Emerging Infectious Dis- eases, 11, 1148-1149.

RIVERA I. N., CHUN J., HUQ A., SACK R. B., COLWELL R. R., 2001. Genotypes associated with virulence in environmental isolates of Vibrio cholerae. Applied and Environmental Microbiology, 67, 2421- 2429.

ROBERT-PILLOT A., GUENOLE A., LESNE J., DELESMONT R., FOURNIER J. M., QUI- LICI M. L., 2004. Occurrence of the tdh and trh genes in Vibrio parahaemolyticus isolates from waters and raw shellfish collected in two French coastal areas and from seafood imported into France. Int J Food Microbiol, 91, 319-325.

SHIRAI H., ITO H., HIRAYAMA T., NAKA- MOTO Y., NAKABAYASHI N., KUMAGAI K., TAKEDA Y., NISHIBUCHI M., 1990.

Molecular epidemiologic evidence for association of thermostable direct hemolysin (TDH) and TDH-related hemolysin of Vibrio parahaemolyticus with gastroenteri- tis. Infect Immun, 58, 3568-3573.

SPIRA W. M., AHMED Q. S., 1981. Gauze fil- tration and enrichment procedures for recovery of Vibrio cholerae from contaminated waters. Appl Environ Microbiol, 42, 730-733.

(16)