Di erences in the API 20E biochemical patterns of clinical and environmental Vibrio parahaemolyticus isolates

Transcription

1 Di erences in the API 20E biochemical patterns of clinical and environmental Vibrio parahaemolyticus isolates Jaime Martinez-Urtaza, Antonio Lozano-Leon, Alejandro Viña-Feas, Jacobo de Novoa & Oscar Garcia-Martin Instituto de Acuicultura, Universidad de Santiago de Compostela, Campus Universitario Sur, Santiago de Compostela, Spain Correspondence: Jaime Martinez-Urtaza, Instituto de Acuicultura, Universidad de Santiago de Compostela, Santiago de Compostela, Spain. Tel.: ext ; fax: ; Received 11 July 2005; revised 19 October 2005; accepted 14 November First published online 15 December doi: /j x Editor: Reggie Lo Key words biochemical characterization; API 20E; tdh; food safety; Vibrio parahaemolyticus. Abstract Genetic differences in clinical and environmental strains of Vibrio parahaemolyticus have been widely used as criteria in identifying pathogenic isolates. However, few studies have been carried out to assess the differences in biochemical characteristics of V. parahaemolyticus isolates from human and environmental sources. We compared the biochemical profiles obtained by the characterization of V. parahaemolyticus isolates from human infections and the marine environment using the API 20E system. Environmental and clinical isolates showed significant differences in the gelatin and arabinose tests. Additionally, clinical isolates were correctly identified according to the API 20E profile using 0.85% NaCl diluent, but they presented nonspecific profiles with 2% NaCl diluent. In contrast, use of 2% NaCl diluent facilitated correct identification of the environmental isolates. Clinical isolates showed significant differences in up to five biochemical tests with respect to the API 20E database. The API 20E system is widely used in routine identification of bacteria in clinical laboratories, and this discrepancy in an important number of biochemical tests may lead to misidentification of V. parahaemolyticus infection. Introduction Vibrio parahaemolyticus is a halophilic member of the family Vibrionaceae that inhabits estuarine areas of temperate and tropical marine environments worldwide. Some strains of this organism cause enteric infections in humans, which are mainly associated with the consumption of raw shellfish and undercooked seafood. Vibrio parahaemolyticus infections have increased globally in recent years (Chowdhury et al., 2000). This organism is the leading cause of seafoodassociated bacterial gastroenteritis in the United States (Mead et al., 1999) and causes approximately one half of the food-borne outbreaks that occur in some Asian countries (Joseph et al., 1982). An increasing number of cases of V. parahaemolyticus infections have been reported recently in Europe (Martinez-Urtaza et al., 2004). The pathogenicity of strains of V. parahaemolyticus has been unequivocally related to the presence of two hemolysins, thermostable direct hemolysin (TDH) and TDH-related hemolysin (TRH). The presence of genes encoding these hemolysins is widely used as a distinguishing feature in the identification of pathogenic isolates. While nearly all clinical strains possess one or both of these virulence factors, these strains represent less than 1% of all environmental strains (DePaola et al., 2003). Although the genetic differences between clinical and environmental isolates are well established, few studies on the biochemical characteristics of V. parahaemolyticus isolates from different sources have been carried out. Conventional methods based on a presumptive incubation in a nonselective broth, followed by confirmative culture on agar have been most widely used in identifying V. parahaemolyticus. Traditionally, colony confirmation was carried out using a panel of biochemical tests. In recent years, a significant number of molecular methods have been specifically developed for both V. parahaemolyticus identification and enumeration. Although the use of these procedures has spread rapidly, the application of molecular method is still restricted to some research groups and reference laboratories. Phenotypic-based procedures continue to be used in most food and clinical laboratories for routine analysis and identification of V. parahaemolyticus isolates. The API 20E system (BioMérieux, Marcy-l Etoile, France) is a standardized bacterial identification system that involves 21 miniaturized biochemical tests and database for species identification. It is considered one of the gold standard

2 76 J. Martinez-Urtaza et al. methods for biochemical identification of enteric and Gram-negative bacteria in clinical and food laboratories. The system was primarily designed for clinical application, although its use has subsequently spread to other microbiological areas, including environmental and food microbiology. The manufacturer s instructions for halophilic Vibrio investigation recommend the use of a 0.85% NaCl diluent to prepare the bacterial inoculum. However, a significant number of misidentifications have been reported in association with the use of this salt concentration for identification of marine bacteria (Seidler et al., 1980; Mac- Donell et al., 1982). Such problems in the identification of bacteria from marine environments using the API 20E system have been related to the Na 1 -dependence of some enzymatic reactions present in marine bacteria (MacDonell et al., 1982). On the basis of this observation, a 2% saline diluent has been proposed as more suitable for identification of halophilic bacteria with the API 20E system (MacDonell et al., 1982). However, the effect of salt concentration on the identification of human isolates of V. parahaemolyticus has been largely overlooked. In the present study, we characterized a series of environmental and clinical isolates of V. parahaemolyticus using the API 20E system, and analysed the different biochemical profiles for any distinctive traits among isolates. In addition we evaluated the effect of saline concentration on the accuracy of identification of the isolates to species level. Materials and methods Bacterial isolates Biochemical profiles of 169 isolates of V. parahaemolyticus from the marine environment, obtained during previous studies carried out in northwestern Spain, Central American countries and Senegal, were included in the comparison. Additionally, all of the clinical isolates in our strain collection were characterized. These include five human isolates from Spain (Martinez-Urtaza et al., 2004), 16 clinical isolates from Asia (Nishibuchi et al., 1989; Matsumoto et al., 2000; Laohaprertthisan et al., 2003) and one clinical isolate from the USA (Matsumoto et al., 2000). The Spanish isolates belonged to serotype O4:K11 and O4:KUntypeable (UT), whereas the non-european clinical strains comprised isolates belonged to the pandemic clone (n = 15) of V. parahaemolyticus, as well as nonpandemic isolates (n = 2). The reference strains ATCC43996 (tdh positive), isolated from cockles involved in fatal food poisoning in Cornwall (UK), and ATCC17802 (tdh negative), isolated from a patient with food poisoning in Japan, were also characterized. A selection of 11 V. vulnificus isolates (two clinical and nine environmental) and nine V. alginolyticus isolates (four clinical and five environmental) were additionally investigated to assess the effect of the saline concentration of the bacterial inoculum on the correct species identification of close related Vibrio species by the API 20E system. Environmental Vibrio vulnificus and Vibrio alginolyticus isolates were retrieved from our strain collection and obtained during previous studies carried out in the northwest coast of Spain. V. vulnificus and V. alginolyticus clinical isolates were recovered from hospitals respectively located in Zaragoza and Getafe (Spain). Biochemical characterization All of the isolates were processed using API 20E strips (BioMérieux), following the manufacturer s instructions. The API 20E system consists of strips of 20 microtubes containing dehydrated substrates that are inoculated with a bacterial suspension. The biochemical tests investigated with API 20E system are: b-galactosidase (ONPG), arginine dihydrolase (ADH), lysine decarboxylase (LDC), ornithine decarboxylase (ODC), citrate utilization (CIT), H 2 Sproduction (H2S), urease (URE), tryptophane deaminase (TDA), indole production (IND), Voges Proskauer (VP), gelatinase (GEL), glucose (GLU), mannitol (MAN), inositol (INO), sorbitol (SOR), rhamnose (RHA), saccharose (SAC), melibiose (MEL), amygdalin (AMY), arabinose (ARA) and cytochrome oxidase (OX). All the strains investigated were retrieved from our strains collection (stored at 70 1C) and cultured onto tryptic soy agar plates supplemented with 1% NaCl. An overnight culture was used to prepare the bacterial inoculum. Each bacterial inoculum was prepared in 0.85% NaCl diluent following the manufacturer s instructions. A further 84 selected environmental isolates were also characterized using 2% NaCl as diluent. Clinical isolates were exclusively investigated with 0.85% diluent because of the high number of unspecific profiles obtained with 2% saline diluent. All of the inoculated API 20E strips were incubated at 37 1C for 24 h. Supplementary investigation of cytochrome oxidase was performed using Bactident Oxidase strips (Merck, Darmstadt, Germany). Results were recorded according to the reading table provided by the manufacturer and the species identification was obtained using the API 20E database included in APILAB Plus software (BioMérieux). Confirmation of isolate identification Identification of all of the isolates included in the present study was confirmed by the presence of the species-specific Vp-toxR and tlh genes, by PCR. Chromosomal DNA isolation was performed from an overnight culture of V. parahaemolyticus isolates in Luria Bertani medium with 1% NaCl incubated at 37 1C with moderate shaking. A 1 ml aliquot of the culture was transferred to a tube and

3 Biochemical characterization of V. parahaemolyticus 77 centrifuged at a g at 4 1C for 5 min. The supernatant was carefully discarded and the pellet was suspended in 300 ml of Tris-EDTA buffer (10 mm Tris-HCl, 0.1 mm EDTA [ph 8.0]). The tubes were incubated in boiling water for 10 min and centrifuged for 5 min at g at 4 1C. Supernatants were carefully transferred to a new tube and 5 ml aliquot was used as template for PCR amplifications. The presence of the Vp-toxR gene was investigated by PCR using the primers 5 0 -GTCTTCTGACGCAATCGTTG-3 0 and 5 0 -ATACGAGTGGTTGCTGTCATG-3 0, as described previously (Kim et al., 1999). The PCR mixture (50 ml) consisted of 5 ml of template genomic DNA, 0.4 mm of each oligonucleotide primer, 1 PCR buffer, 2 mm MgCl 2, 200 m M of each dntp and 2 U of AmpliTaq DNA Polymerase (Roche Diagnostics, Indianapolis, IN). The amplification conditions consisted of a initial denaturation at 94 1C for 5 min, followed by 20 cycles of denaturation at 94 1C for 1 min, primer annealing at 63 1C for 1.5 min, and primer extension at 72 1C for 1.5 min, followed by a final extension at 72 1C for 7 min. Presence of tlh gene was investigated simultaneously to that of the virulence genes tdh and trh by multiplex PCR, according to the procedure described by Bej et al. (1999). Multiplex PCR assay was preformed using the oligonucleotide primer pairs for tdh (5 0 -GTAAAG- GTCTCTGACTTTTGGAC-3 0 and 5 0 -TGGAATAGAACC- TTCATC TTCACC-3 0 ), trh (5 0 -TTGGCTTCGATATTTT- CAGTATCT-3 0 and 5 0 -CATAACAAACATATGCCCATTTC- CG-3 0 ) and tl (5 0 -AAAGCGGATTATGCAGAAGCACTG-3 0 and 5 0 -GCTACTTTCTAGCATTTTCTCTGC-3 0 ). Each 50 ml of PCR mixture contained 5 ml of template genomic DNA, 1 mm of each oligonucleotide primer, 1 PCR buffer, 2 mm MgCl 2, 200 mm of each dntp and 2 U of AmpliTaq DNA Polymerase (Roche Diagnostics). PCR reactions were performed with the following reaction conditions: denaturation at 94 1C for 3 min, followed by 30 cycles of denaturation at 94 1C for 1 min, primer annealing at 58 1C for 1 min, and primer extension at 72 1C for 1 min. A final extension was performed at 75 1C for 5 min. PCR reactions were carried out with a PT-200 DNA Engine Thermal Cycler (M&J Research, Waltham, MA) and amplicons were analyzed in a 1.8% agarose gel. Statistical analysis Significant differences between the biochemical API 20E profile for V. parahaemolyticus and the percentage of isolates in each group (environmental with 0.85% NaCl, environmental with 2% NaCl and clinical strains) showing positive biochemical reactions were determined using the Chi- Square test. All statistical analysis was carried out with SPSS version (SPSS Inc., Chicago, IL) and the level of significance was set at P o Results The results of the biochemical investigation of the isolates included in the study are summarized in Table 1. Biochemical results obtained from the analysis of the 169 environmental isolates analysed using a bacterial inoculum prepared with 0.85% NaCl diluent showed significant differences (P o 0.001) in two of the biochemical tests with respect to the Vibrio parahaemolyticus API 20E profile. Positive citrate utilization and amygdalin occurred in, respectively, 4% and 40% of the environmental isolates in contrast with the 50% and 12% corresponding to the API 20E profile. In addition to the differences shown by the environmental isolates, all of the clinical isolates investigated with 0.85% diluent also showed significantly less positive results of gelatinase and arabinose utilization than in the API 20E profile. Biochemical investigation of the 84 selected environmental isolates characterized using 2% NaCl diluent Table 1. Biochemical characterization of the different groups of Vibrio parahaemolyticus investigated in the study and API 20E biochemical profile for this organism Biochemical tests (% positive isolates) No. Data source ONPG ADH LDC ODC CIT H2S URE TDA IND VP GEL GLU MAN INO SOR RHA SAC MEL AMY ARA OX strains API 20E Profile Environmental 0.85% Environmental 2% Clinical strains 0.85% Reference strains Statistically significant differences (P o 0.001). All the data represent the percentage of isolates showing a positive reaction after 24 h incubation at 37 1C. Significant statistical differences (P o 0.001) between the results of the tests and the API 20E profile are also shown. ONPG, b-galactosidase; ADH, arginine dihydrolase; LDC, lysine decarboxylase; ODC, ornithine decarboxilase; CIT, citrate utilization; H2S, H 2 S production; URE, urease; TDA, tryptophane deaminase; IND, indole production; VP, Voges Proskauer; GEL, gelatinase; GLU, glucose; MAN, mannitol; INO: inositol; SOR: sobitol; RHA, rhamnose; SAC; saccharose; MEL, melibiose; AMY: amygdalin; ARA, arabinose; OX, cytochrome oxidase.

4 78 J. Martinez-Urtaza et al. showed significant differences in five biochemical tests. Apart from the significant differences in citrate and amygdalin utilization, these isolates showed a significantly higher number of positive reactions in the urease and melibiose tests than in the API 20E profile, and a lower number of positive results in the ornithine decarboxylase test. All of the isolates included in the study were previously confirmed as V. parahaemolyticus by PCR. Clinical isolates obtained from human infections were characteristically tdh positive and trh negative, whereas all of the environmental isolates, except the isolate, lacked these two genes. The isolate was obtained from environmental sources but presented the typical genetic virulence attributes of clinical isolates. Four V. parahaemolyticus isolates from each clinical and environmental source were selected for evaluation of the effect of the salt concentration of the diluent used to prepare the inoculum on the accuracy of the identification to species level (Table 2). Biochemical investigation of the four environmental isolates with 0.85% NaCl diluent identified two isolates as Vibrio (Grimontia) hollisae, while the other two isolates presented nonspecific profiles. One isolate showed unexpected results in six of the biochemical tests compared with those corresponding to the API 20E profile. Two isolates showed negative or unclear results for the glucose test. However, when these environmental isolates were analysed using 2% NaCl diluent, all four isolates were correctly identified as V. parahaemolyticus, although they showed differences in two (two isolates) and four (two isolates) biochemical traits. By contrast, the analysis of the clinical isolates revealed contrary results. Human isolates were only correctly identified when the 0.85% NaCl diluent was used. With exception of the CIT test for which negative results were obtained in all cases most of the isolates showed the expected biochemical profiles. When the human isolates were analysed using 2% NaCl diluent, three results were recorded as nonspecific, while another presented an unacceptable profile. All of the clinical isolates exhibited different results from those obtained in the API 20E system in three tests (ODC, CIT, and AMY), whereas the isolate with an unacceptable profile also showed a different result in the MEL test. Table 2. Biochemical profiles of clinical and environmental Vibrio parahaemolyticus isolates obtained with 0.85% and 2% NaCl diluent Biochemical tests (% positive isolates) Strain ONPG ADH LDC ODC CIT H2S URE TDA IND VP GEL GLU MAN INO SOR RHA SAC MEL AMY ARA OX Identification API profile Clinical 0.85% Vph Vph Vph 447/ Vph 2% Unacceptable Vv/Vph Vv/Vph 447/ Vv/Vph Environmental 0.85% Vv/Vph Vph/Val Vholl Vholl 2% Vph Vph Vph Vph ONPG, b-galactosidase; ADH, arginine dihydrolase; LDC, lysine decarboxylase; ODC, ornithine decarboxylase; CIT, citrate utilization; H 2 S, H 2 S production; URE, urease; TDA, tryptophane deaminase; IND, indole production; VP, Voges Proskauer; GEL, gelatinase; GLU, glucose; MAN, mannitol; INO, inositol; SOR, sobitol; RHA, rhamnose; SAC, saccharose; MEL, melibiose; AMY, amygdalin; ARA, arabinose; OX, cytochrome oxidase; Vph, Vibrio parahaemolyticus; Vv,Vibrio vulnificus; Vholl, Vibrio hollisae., unclear result.

5 Biochemical characterization of V. parahaemolyticus 79 Table 3. API 20E identifications of the clinical and environmental Vibrio vulnificus and Vibrio alginolyticus isolates obtained with 0.85% and 2% NaCl diluent API 20E result Organism Isolate number Source 0.85% NaCl 2% NaCl Vibrio alginolyticus Environmental Val 98.2% Val 92% Environmental Val 97.7% Val 97.7% Ac-14b Environmental Val 92.6% Val 98.5% Environmental Val 65% Val 99.7% Ahy 30% val219 Environmental Val 98.2% Val 89.6% Ahy 8.6% Cli2/51291 Clinical Val 97.7% Val 97.7% Cli2/54741 Clinical Val 98.6% Val 99.5% Cli2/67369 Clinical Val 98.5% Val 97.6% Cli99/51535 Clinical Val 97.7% Val 97.7% Vibrio vulnificus Ac01 Environmental Vv 99.7% Vv 99.7% Environmental Vv UP Vv 77.3% 9531 Environmental Burkholderia 76% Vv 99.5% Pasteurella 18% Environmental Pasteurella 64% Vv 98.2% Chryseomonas 26% 19/84b Environmental Vv 57.2% Vv 57.2% Vph 26.5% Vph 26.5% Environmental Vv 90.5% Vv 99.9% Environmental Vv 98.2% Vv 99.5% 9715 Environmental Vholl 75.4% Vv 99.5% Pasteurella 11.4% Environmental Burkholderia UP Vv 99.5% cliz-01 Clinical Vv 99.9% Vv 99.9% vlt2003 Clinical Ahy 67% Vv 94.1% Vv 17% Vph, Vibrio parahaemolyticus; Vv, Vibrio vulnificus; Val, Vibrio alginolyticus; Ahy, Aeromonas hydrophyla; Vholl, Vibrio hollisae; UP, unacceptable profile. The investigation of the other related Vibrio species included in the study showed that the 2% NaCl diluent was the most suitable for the correct identification of Vibrio vulnificus and Vibrio alginolyticus isolates obtained both from environmental and clinical sources (Table 3). All the nine Vibrio alginolyticus investigated were correctly identified using both 0.85% and 2% NaCl diluent, although 2% NaCl diluent gave a higher discriminatory profile in the identification of the isolate (99.7% compared with 65% at 0.85% NaCl). The suitability of the 2% NaCl diluent for species identification was most evident when V. vulnificus isolates were investigated. Ten V. vulnificus isolates were correctly identified at species level using 2% NaCl, with a percentage of identification ranging from 99.9% to 77.3%. The isolate number 19/84b presented identical low discriminatory biochemical profile at 2% and 0.85% NaCl diluent, although it was firstly identified as V. vulnificus with a percentage of identification of 57.2%. By contrast, six isolates (five environmental and one clinical) could not be correctly identified as V. vulnificus using 0.85% NaCl diluent, and they showed a variety of non- Vibrio species identifications and unacceptable profiles (Table 3). Discussion The difficulty in obtaining a reliable identification of Vibrio parahaemolyticus isolates by biochemical procedures is widely recognized among technicians working on this organism. The particular salt requirement of Vibrio parahaemolyticus frequently causes limited growth in standard microbiological media and a deficient or unclear reaction when some of these media are supplemented with salts. Problems associated with the correct identification of V. parahaemolyticus have been pointed out during the evaluation of biochemical galleries for bacterial identification (MacDonell et al., 1982). However, differences between clinical and environmental Vibrio parahaemolyticus populations, isolated from such different environments as humans and marine waters have never been taken into consideration. Genetic differences between clinical and environmental isolates are well established and are mainly based on the

6 80 J. Martinez-Urtaza et al. presence of tdh gene, present in most clinical isolates but absent from environmental isolates (DePaola et al., 2003). The results included in the present study indicated that these differences also extend to their biochemical features. Clinical isolates showed statistically significant differences in gelatin utilization (GEL) and arabinose fermentation (ARA) with respect to the environmental strains. While most of the environmental isolates were positive for these reactions using both 0.85% and 2% NaCl as diluent, most of the clinical isolates showed negative results in the biochemical tests. This different pattern was also evident in the response of biochemical activity to different salinities. Clinical isolates showed typical biochemical reactions in 0.85% NaCl diluent and were correctly identified using the API 20E profile. The use of 2% NaCl diluent facilitated the correct identification of the environmental isolates as indicated by the higher number of positive reactions. Environmental isolates analysed using 2% diluent showed 20%, 6%, 12% and 82% positive reactions, respectively, for the CIT, URE, MEL and AMY tests, in contrast with the 4%, 2%, 0% and 40% observed when the 0.85% diluent was used. Some environmental isolates were unable to ferment glucose with 0.85% NaCl diluent, although they were able to when 2% NaCl diluent was used. These results are consistent with previous data that demonstrated that 2% saline diluent allowed characterization of environmental strains of V. parahaemolyticus that are normally nonreactive when standard API 20E diluents are used (MacDonell et al., 1982). In the present study, the 2% saline diluent has also shown to be the most suitable one for identification of V. vulnificus and V. alginolyticus isolates from environmental and clinical sources. The suitability of 0.85% and 2% NaCl diluents for the respective biochemical investigations of clinical and environmental V. parahaemolyticus isolates was also demonstrated by the correct identification of unspecific profiles. If we consider all the V. parahaemolyticus isolates to be equally genetically competent for an enzymatic pathway, regardless of their source, the differences could be exclusively ascribed to variations in the efficiency of an enzymatic process due to adaptation to particular environments with different salinities (humans and seawater). Nevertheless, the presence of some distinctive traits among isolates from a single group may also be based on a particular genetic feature related to a V. parahaemolyticus population, which may be used in the future for characterization and typing purposes. The investigation of the suitability of API 20E system for identification of V. parahaemolyticus has shown some limitations in the accuracy of the V. parahaemolyticus data included in the API 20E database. Significant differences in the biochemical results of CIT and AMY tests were detected using a diverse panel of isolates from different sources and origin. Although the API 20E database was initially elaborated using biochemical patterns of clinical isolates, the comparative analysis of the API 20E profiles for V. parahaemolyticus and the results from environmental and clinical isolates revealed a higher degree of agreement between the API 20E pattern and the biochemical patterns corresponding to environmental isolates. Clinical isolates showed significant differences in up to 5 biochemical tests compared with the results included in the API 20E database (Spanish clinical isolates). Biochemical reactions in CIT and GEL tests are almost nonexistent among clinical isolates, although they have been assigned positive values of 50% and 75% respectively in the API 20E profile. The widespread use of the API 20E system in the routine identification of bacteria in clinical laboratories may therefore lead to misidentification of V. parahaemolyticus infection. Data included in the present study showed a preliminary evidence of differences between human and environmental isolates in some biochemical reactions. Furthermore, results from biochemical characterization at different salinities have revealed that the isolates enhanced their biochemical fitness at salinities close to those existing in their source habitat (marine environments, with salinities around 2 3%, and human extracellular fluids, with salinities around 0.9%), highlighting the biological differences between pathogenic and environmental specimens of V. parahaemolyticus. Acknowledgements We greatly appreciate the technical help of Veronica Blanco Abad. References Bej AK, Patterson DP, Brasher CW, Vickery MC, Jones DD & Kaysner CA (1999) Detection of total and hemolysinproducing Vibrio parahaemolyticus in shellfish using multiplex PCR amplification of tl, tdh and trh. J Microbiol Methods 36: Chowdhury NR, Chakraborty S, Ramamurthy T, Nishibuchi M, Yamasaki S, Takeda Y & Nair GB (2000) Molecular evidence of clonal Vibrio parahaemolyticus pandemic strains. Emerg Infect Dis 6: DePaola A, Ulaszek J, Kaysner CA, Tenge BJ, Nordstrom JL, Wells J, Puhr N & Gendel SM (2003) Molecular, serological, and virulence characteristics of Vibrio parahaemolyticus isolated from environmental, food, and clinical sources in North America and Asia. Appl Environ Microbiol 69: Joseph SW, Colwell RR & Kaper JB (1982) Vibrio parahaemolyticus and related halophilic Vibrios. Crit Rev Microbiol 10: Kim YB, Okuda J, Matsumoto C, Takahashi N, Hashimoto S & Nishibuchi M (1999) Identification of Vibrio parahaemolyticus strains at the species level by PCR targeted to the toxr gene. J Clin Microbiol 37: Laohaprertthisan V, Chowdhury A, Kongmuang U, Kalnauwakul S, Ishibashi M, Matsumoto C & Nishibuchi M (2003) Prevalence

7 Biochemical characterization of V. parahaemolyticus 81 and serodiversity of the pandemic clone among the clinical strains of Vibrio parahaemolyticus isolated in southern Thailand. Epidemiol Infect 130: MacDonell MT, Singleton FL & Hood MA (1982) Diluent composition for use of API 20E in characterizing marine and estuarine bacteria. Appl Environ Microbiol 44: Martinez-Urtaza J, Lozano-Leon A, DePaola A, Ishibashi M, Shimada K, Nishibuchi M & Liebana E (2004) Characterization of pathogenic Vibrio parahaemolyticus isolates from clinical sources in Spain and comparison with Asian and North American pandemic isolates. J Clin Microbiol 42: Matsumoto C, Okuda J, Ishibashi M, et al. (2000) Pandemic spread of an O3: K6 clone of Vibrio parahaemolyticus and emergence of related strains evidenced by arbitrarily primed PCR and toxrs sequence analyses. J Clin Microbiol 38: Mead PS, Slutsker L, Dietz V, McCaig LF, Bresee JS, Shapiro C, Griffin PM & Tauxe RV (1999) Food-related illness and death in the United States. Emerg Infect Dis 5: Nishibuchi M, Taniguchi T, Misawa T, Khaeomanee-Iam V, Honda T & Miwatani T (1989) Cloning and nucleotide sequence of the gene (trh) encoding the hemolysin related to the thermostable direct hemolysin of Vibrio parahaemolyticus. Infect Immun 57: Seidler RJ, Allen DA, Colwell RR, Joseph SW & Daily OP (1980) Biochemical characteristics and virulence of environmental group F bacteria isolated in the United States. Appl Environ Microbiol 40: