The ability of two different Vibrio spp. bacteriophages to infect Vibrio harveyi, Vibrio cholerae and Vibrio mimicus

AIMS: To determine the host range of the Vibrio harveyi myovirus-like bacteriophage (VHML) and the cholera toxin conversion bacteriophage (CTX Phi) within a range of Vibrio cholerae and V. mimicus and V. harveyi, V. cholerae and V. mimicus isolates respectively. METHODS AND RESULTS: Three V. harveyi, eight V. cholerae and five V. mimicus isolates were incubated with VHML and CTX Phi. Polymerase chain reaction (PCR) was used to determine the presence of VHML and CTX Phi in infected isolates. We demonstrated that it was possible to infect one isolate of V. cholerae (isolate ACM #2773/ATCC #14035) with VHML. This isolate successfully incorporated VHML into its genome as evident by positive PCR amplification of the sequence coding part of the tail sheath of VHML. Attempts to infect all other V. cholerae and V. mimicus isolates with VHML were unsuccessful. Attempts to infect V. cholerae non-01, V. harveyi and V. mimicus isolates with CTX Phi were unsuccessful. CONCLUSIONS: Bacteriophage infection is limited by bacteriophage-exclusion systems operating within bacterial strains and these systems appear to be highly selective. One system may allow the co-existence of one bacteriophage while excluding another. VHML appears to have a narrow host range which may be related to a common receptor protein in such strains. The lack of the vibrio pathogenicity island bacteriophage (VPI Phi) in the isolates used in this study may explain why infections with CTX Phi were unsuccessful. SIGNIFICANCE AND IMPACT OF THE STUDY: The current study has demonstrated that Vibrio spp. bacteriophages may infect other Vibrio spp.     

Vibrio mimicus diarrhea following ingestion of raw turtle eggs.

Clinical and epidemiological characteristics of diarrhea associated with Vibrio mimicus were identified in 33 hospitalized patients referred to the Costa Rican National Diagnostic Laboratory Network between 1991 and 1994. The relevant symptoms presented by patients included abundant watery diarrhea, vomiting, and severe dehydration that required intravenous Dhaka solution in 83% of patients but not fever. Seroconversion against V. mimicus was demonstrated in four patients, from whom acute- and convalescent-phase sera were obtained. Those sera did not show cross-reaction when tested against Vibrio cholerae O1 strain VC-12. All the V. mimicus isolates from these cases produced cholera toxin (CT) and were susceptible to commonly used antibiotics. Attempts to isolate this bacterium from stool samples of 127 healthy persons were not successful. Consumption of raw turtle eggs was recalled by 11 of the 19 (58%) individuals interviewed. All but two V. mimicus diarrheal cases were sporadic. These two had a history of a common source of turtle (Lepidochelys olivacea) eggs for consumption, and V. mimicus was isolated from eggs from the same source (a local market). Among the strains, variations in the antimicrobial susceptibility pattern were observed. None of the strains recovered from market turtle eggs nor the four isolates from river water showed CT production. Further efforts to demonstrate the presence of CT-producing V. mimicus strains in turtle eggs were made. Successful results were obtained when nest eggs were tested. In this case, it was possible to isolate CT- and non-CT-producing strains, even from the same egg. For CT detection we used PCR, enzyme-linked immunosorbent assay (ELISA), and Y-1 cell assay, obtaining a 100% correlation between ELISA and PCR results. Primers Col-1 and Col-2, originally described as specific for the V. cholerae O1 ctxA gene, also amplified a 302-bp segment with an identical restriction map from V. mimicus. These results have important implications for epidemiological surveillance in tropical countries where turtle eggs are used for human consumption, serving as potential sources of cholera-like diarrhea.     

Medium-dependent production of extracellular enterotoxins by non-O-1 Vibrio cholerae, Vibrio mimicus, and Vibrio fluvialis.

Fluid accumulation at 4 h in the intestines of suckling mice enabled us to distinguish non-O-1 Vibrio cholerae, V. mimicus, and V. fluvialis clinical isolates from environmental isolates. Enterotoxin production was culture medium dependent. Filtrates of cultures grown in tryptic soy broth without glucose but with added 0.5% NaCl did not exhibit marked enterotoxin activity in the assay. Culture filtrates of all clinical strains grown in brain heart infusion broth supplemented with 0.5% NaCl induced large amounts of fluid accumulation in mouse intestines. However, most environmental strains grown in brain heart infusion broth amended as described above were unable to induce fluid accumulation. The enterotoxin present in culture filtrates lost activity at 56 degrees C and appeared to be distinct from previously described virulence factors, including the well-described cholera toxin. The new enterotoxin could represent an important virulence mechanism common to all three species.     

Sugar composition of the polysaccharide portion of lipopolysaccharides of Vibrio fluvialis, Vibrio vulnificus, and Vibrio mimicus

A chemotaxonomic study was carried out on Vibrio fluvialis and V. vulnificus on the basis of the sugar composition of the polysaccharide portion of their lipopolysaccharides (LPS). A previously developed rapid method of preparing samples for compositional sugar analysis was employed. Nineteen O-serogroups of V. fluvialis were divided into 14 chemotypes while seven O-serogroups of V. vulnifucus were divided also into seven chemotypes since the polysaccharide portion of LPS of each serogroup has a different sugar composition from that of the other serogroups. Close similarities in the sugar composition of the same portion were demonstrated between serologically cross-reacting non-O1 group V. cholerae and V. fluvialis, and non-O1 V. cholerae and V. mimicus.     

Seasonal distribution of facultatively enteropathogenic vibrios (Vibrio cholerae, Vibrio mimicus, Vibrio parahaemolyticus) in the freshwater of the Elbe River at Hamburg

Between June 1981 and December 1982 the incidence of Vibrio cholerae, V. mimicus and V. parahaemolyticus was determined at two sampling sites on the Elbe River at Hamburg. A total of 183 strains was isolated from 147 water samples. Of these, 107 belonged to non-01 V. cholerae (ten strains producing a cholera-like enterotoxin); 33 were identified as V. mimicus, including two enterotoxin producers; 42 strains were Kanagawa-negative cultures of V. parahaemolyticus; and one was V. fluvialis. The highest incidence was observed from June to September with about 10(2) organisms/l. Halophilic vibrios, less than five organisms/l, were detectable during the period June/July to October. The vibrio incidence was not influenced by the numbers of aerobic heterotrophic bacteria, coliforms or faecal bacteria. In general water temperature correlated with the seasonal variation. Thus, a temperature rise over 10 degrees to 20 degrees C was followed by a distinct increase in vibrio numbers. Of 14 chemical parameters only chloride concentration might have had an influence on the seasonal variation. It is concluded that the three Vibrio species are indigenous organisms of the Elbe River.     

Incidence of toxigenic vibrios in foods available in Taiwan.

A total of 1088 vibrios and related species were isolated from seafood and aquacultured foods available in Taiwan. They were identified as Vibrio alginolyticus, V. cholerae, V. fluvialis I, V. fluvialis II, V. parahaemolyticus, V. mimicus, Aeromonas caviae, A. hydrophila, A. sobria and other species. Incidence of these Vibrio and Aeromonas species in these foods was high. Vibrio parahaemolyticus was frequently found in seawater and in foods of freshwater origin. The Vibrio isolates were examined for enzymatic and toxigenic activities. Most of them showed strong lipase or protease activities. Haemolytic activities of V. cholerae, V. fluvialis I and V. fluvialis II isolates were mostly strong. About 49% showed cytotoxic activity and 5% cytotonic activity in Chinese hamster ovary cell culture assay. Nevertheless, only three non-O1 V. cholerae (2.07%) and two V. parahaemolyticus isolates (1.65%) produced cholera toxin and thermostable direct haemolysin activity, respectively. Various toxigenic vibrios may be important food-borne pathogens in this region because of their high incidence in foods.     

Potentially pathogenic vibrios in brackish waters and mussels

Water and mussel samples were collected from two brackish lakes, used as mussel farms, at different times of the year, for the quantitative analysis of Vibrio spp. and for the isolation of potentially pathogenic species. The isolates underwent cultural and biochemical tests selected for rapid identification. Glucose oxidizing-fermenting and O/129 sensitive strains were distinguished on the basis of the following tests: sucrose and cellobiose utilization, sulphatase activity and polymyxin B resistance performed, respectively, on TCBS, CPC and SPS media. Responses to the presence of beta-galactosidase, salt requirement and growth on triple sugar iron medium were also added. A total of 125 from 152 isolates were referred to the species Vibrio fluvialis (55 strains), V. alginolyticus (40), V. parahaemolyticus (11), V. vulnificus (10) and V. mimicus (9). The remaining 27 isolates were not identified. The isolation of potentially pathogenic vibrios from cultivated mussels is a risk for health of people consuming raw seafood. Therefore, a long-term monitoring programme should also include the search for these bacterial species.     

Distribution of virulence-associated genes in Vibrio mimicus isolates from clinical and environmental origins

Distribution of virulence-associated genes in Vibrio mimicus was studied including the toxin genes ctxA, tdh, st and vmh and the genes necessary for regulation of toxin production, toxR, toxS, toxT, tcpA and tcpP. Approximately half of clinical V. mimicus isolates possessed one or more genes encoding V. cholerae enterotoxic factors such as ctxA, tdh and st. All of the clinical and environmental isolates possessed vmh encoding V. mimicus hemolysin (VMH). The ctxA encoding cholera toxin was detected in only 2 strains, 5% of the clinical isolates. Furthermore, there were very few strains possessing tcpP and toxT needed for the expression of ctxA. These results may suggest that VMH is a more important pathogenic factor than well recognized toxins such as cholera toxin (CT) in V. mimicus infection.     

An Electrophoretic Analysis of Variation in the Glucose-6-Phosphate Dehydrogenase and Malate Dehydrogenase of Vibrio cholerae, Vibrio parahaemolyticus and Vibrio fluvialis

A total of 59 isolates of Vibrio cholerae, V. parahaemolyticus and V. fluvialis were studied using the techniques of enzyme electrophoresis. The enzymes used were malate dehydrogenase (E.C.I. 1.1.37) and glucose-6-phospate dehydrogenase (E.C.I. 1.1.49). The results in general confirmed classical methods of vibrio taxonomy. The three species could be separated from each other and identified by their enzyme variant types.
The term zymovar, a group of strains having similar enzyme variants, is introduced. In the V. cholerae isolates six zymovars were found. All V. cholerae serovar 01 strains were classified in the same zymovar except for some strains of environmental origin, which occurred in another zymovar. In V. fluvialis two zymovars were detected corresponding with biovars 1 and 2 of this organism. All isolates of V. parahaemolyticus gave the same enzyme type.
The technique of enzyme electrophoresis appears to be a useful tool in vibrio taxonomy and used in conjunction with other methods may aid in the elucidation of the systematics of the genus.     

Incidence of toxigenic vibrios in foods available in Taiwan

A total of 1088 vibrios and related species were isolated from seafood and aquacultured foods available in Taiwan. They were identified as Vibrio alginolyticus, V. cholerae, V. fluvialis I, V. fluvialis II, V. parahaemolyticus, V. mimicus, Aeromonas caviae, A. hydrophila, A. sobria and other species. Incidence of these Vibrio and Aeromonas species in these foods was high. Vibrio parahaemolyticus was frequently found in seawater and in foods of freshwater origin. The Vibrio isolates were examined for enzymatic and toxigenic activities. Most of them showed strong lipase or protease activities. Haemolytic activities of V. cholerae, V. fluvialis I and V. fluvialis II isolates were mostly strong. About 49% showed cytotoxic activity and 5% cytotonic activity in Chinese hamster ovary cell culture assay. Nevertheless, only three non-1 V. cholerae (2.07%) and two V. parahaemolyticus isolates (1.65%) produced cholera toxin and thermostable direct haemolysin activity, respectively. Various toxigenic vibrios may be important food-borne pathogens in this region because of their high incidence in foods.     

Environmental Vibrio spp., isolated in Mozambique, contain a polymorphic group of integrative conjugative elements and class 1 integrons

Circulation of mobile genetic elements linked to drug resistance spread was studied in Vibrio strains isolated from surface urban water (river and sea) and shellfish samples in 2002-2003 in Maputo, Mozambique. Class 1 integrons and integrating conjugative elements (ICE) were investigated by PCR and mating experiments in strains of major health interest: 10 Vibrio cholerae, six Vibrio parahaemolyticus, two Vibrio alginolyticus and one Vibrio fluvialis. Resistance to at least two antibiotics (predominantly beta-lactams) was detected in all the strains, with additional resistances to sulfamethoxazole, spectinomycin, streptomycin and/or trimethoprim. Class 1 integrons contributed partially to the expression of drug resistance and were found in five isolates: four V. cholerae (blaP1 cassette, one strain also contained the dfrA15 cassette) and one V. alginolyticus (aadA2 cassette). ICEs, apparently devoid of resistance genes, were found in eight V. cholerae, three V. parahaemolyticus and one V. fluvialis isolates. A wide variability was observed by molecular characterization of ICEs. Five ICEs were included in the SXT/R391 family and seven ICEs were not classified. Our results indicate that the SXT/R391 family and related ICEs comprise a large class of polymorphic genetic elements widely circulating in environmental Vibrio strains in Africa, beside those evidently linked to drug resistance in clinical isolates.     

Seasonal distribution of facultatively enteropathogenic vibrios (Vibrio cholerae, Vibrio mimicus, Vibrio parahaemolyticus) in the freshwater of the Elbe River at Hamburg

Between June 1981 and December 1982 the incidence of Vibrio cholerae, V. mimicus and V. parahaemolyticus was determined at two sampling sites on the Elbe River at Hamburg. A total of 183 strains was isolated from 147 water samples. Of these, 107 belonged to non-01 V. cholerae (ten strains producing a cholera-like enterotoxin); 33 were identified as V. mimicus , including two enterotoxin producers; 42 strains were Kanagawa-negative cultures of V. parahaemolyticus ; and one was V. fluvialis. The highest incidence was observed from June to September with about 10² organisms/1. Halophilic vibrios, less than five organisms/1, were detectable during the period June/July to October. The vibrio incidence was not influenced by the numbers of aerobic heterotrophic bacteria, coliforms or faecal bacteria. In general water temperature correlated with the seasonal variation. Thus, a temperature rise over 10° to 20°C was followed by a distinct increase in vibrio numbers. Of 14 chemical parameters only chloride concentration might have had an influence on the seasonal variation. It is concluded that the three Vibrio species are indigenous organisms of the Elbe River.     

Organization of the CTX prophage in environmental isolates of Vibrio mimicus

The organization of the CTX prophage in environmental strains of Vibrio mimicus was investigated. Sixteen hundred non-sucrose fermenting vibrios were examined for ctx gene by hybridization. Out of 1,600 isolates, 6 V. mimicus isolates contained ctxA gene. The organization of CTX prophage was determined by RFLP using ctxA probe. The CTX prophage integrated at a single site in V. mimicus genome which was present as a single copy flanked by at least a single RS element. Ribotype pattern revealed that a particular clone of V. mimicus acquired the CTXPhi in the aquatic environment. This study demonstrated that V. mimicus could act as a reservoir of CTXPhi in the aquatic environment.     

Differentiation of environmental and clinical isolates of Vibrio mimicus from Vibrio cholerae by multilocus enzyme electrophoresis

In this study, we demonstrated that analyzed strains of Vibrio mimicus and Vibrio cholerae could be separated in two groups by using multilocus enzyme electrophoresis (MEE) data from 14 loci. We also showed that the combination of four enzymatic loci enables us to differentiate these two species. Our results showed that the ribosomal intergenic spacer regions PCR-mediated identification system failed, in some cases, to differentiate between V. mimicus and V. cholerae. On the other hand, MEE proved to be a powerful molecular tool for the discrimination of these two species even when atypical strains were analyzed.     

Ecology of Vibrio mimicus in aquatic environments

An environmental study was done to examine the prevalence of Vibrio mimicus in some aquatic environments of Dhaka, Bangladesh, and of Okayama, Japan. Water samples from Dhaka environments and water and plankton samples from Okayama environments were quantitatively as well as qualitatively analyzed throughout the seasons for V. mimicus. The organism was isolated from Bangladesh environments throughout the year, whereas it was not isolated in Okayama when the water temperature fell below 10 degrees C. Samples with as many as 9.0 x 10(2) CFU of V. mimicus per 100 ml of water in Dhaka and 1.5 x 10(4) CFU of V. mimicus per 100 ml of water in Okayama were detected during the study period. V. mimicus was not found in any environment with an average salinity of 10% or more. Brackish environments with an average salinity of 4% were observed to be the optimal natural condition for the pathogen. Using the API 20E system with the conventional test methods, we observed variations in biochemical properties within the V. mimicus species. This study reveals the inefficacy of the API 20E system to identify a significant percentage of V. mimicus. Therefore, in addition to the API 20E system, a salt tolerance test and a string test are recommended for identification of this species. Susceptibility testing of strains isolated from Okayama environments showed higher resistance to ampicillin and susceptibility to trimethoprim-sulfamethoxazole when compared with environmental isolates of V. mimicus from Bangladesh.     

Ecology of Vibrio mimicus in aquatic environments.

An environmental study was done to examine the prevalence of Vibrio mimicus in some aquatic environments of Dhaka, Bangladesh, and of Okayama, Japan. Water samples from Dhaka environments and water and plankton samples from Okayama environments were quantitatively as well as qualitatively analyzed throughout the seasons for V. mimicus. The organism was isolated from Bangladesh environments throughout the year, whereas it was not isolated in Okayama when the water temperature fell below 10 degrees C. Samples with as many as 9.0 x 10(2) CFU of V. mimicus per 100 ml of water in Dhaka and 1.5 x 10(4) CFU of V. mimicus per 100 ml of water in Okayama were detected during the study period. V. mimicus was not found in any environment with an average salinity of 10% or more. Brackish environments with an average salinity of 4% were observed to be the optimal natural condition for the pathogen. Using the API 20E system with the conventional test methods, we observed variations in biochemical properties within the V. mimicus species. This study reveals the inefficacy of the API 20E system to identify a significant percentage of V. mimicus. Therefore, in addition to the API 20E system, a salt tolerance test and a string test are recommended for identification of this species. Susceptibility testing of strains isolated from Okayama environments showed higher resistance to ampicillin and susceptibility to trimethoprim-sulfamethoxazole when compared with environmental isolates of V. mimicus from Bangladesh.     

Detection and identification of Vibrio species using whole-cell protein pattern analysis

Outbreaks of foodborne diseases associated with Vibrio species such as V. parahaemolyticus, V. vulnificus, and V. cholerae frequently occur in countries having a dietary habit of raw seafood consumption. For rapid identification of different Vibrio species involved in foodborne diseases, whole-cell protein pattern analysis for 13 type strains of 12 Vibrio species was performed using SDS-PAGE analysis. Pathogenic Vibrio species such as V. parahaemolyticus, V. vulnificus, V. cholerae, V. alginolyticus, V. fluvialis, and V. mimicus were included in the 12 Vibrio species used in this study. Each of the 12 Vibrio species showed clearly specific band patterns of its own. Two different strains of V. parahaemolyticus showed two different SDS-PAGE wholecell protein patterns, giving the possibility of categorizing isolated strains in the same V. parahaemolyticus species into two subgroups. The 36 Vibrio isolates collected from sushi restaurants in Busan were all identified as V. parahaemolyticus by comparing their protein patterns with those of Vibrio type strains. The identified isolates were categorized into two different subgroups of V. parahaemolyticus. The whole-cell protein pattern analysis by SDS-PAGE can be used as a specific, rapid, and simple identification method for Vibrio spp. involved in foodborne diseases at the subspecies level.     

Rapid serological identification of Vibrio vulnificus by anti-H coagglutination.

Staphylococcus aureus Cowan 1 cells were armed with anti-flagellar (anti-H) antibody produced in rabbits immunized with flagellar core protein prepared from Vibrio vulnificus. This reagent was assessed by coagglutination for its capacity to agglutinate and identify V. vulnificus. A species-specific H antigen is expressed in the core proteins of the polar flagella of V. vulnificus. Of 435 V. vulnificus isolates identified bacteriologically, 432 (99.3%) were agglutinated in the slide test within 2 min after the addition of the anti-V. vulnificus H coagglutination reagent. Other than Vibrio pelagius, the reagent did not agglutinate 19 heterologous Vibrio spp. tested, including 290 V. cholerae, 22 V. mimicus, 395 V. parahaemolyticus, and 16 V. fluvialis isolates recovered from seafood and the marine environment. The serological resolution of the coagglutination reaction was enhanced if the organism under test was suspended in 0.1 M Tris buffer-0.1 mM EDTA-1.0% Triton X-100 (TET) for 24 h before serological examination. The TET buffer also increased the sensitivity of the coagglutination reaction 100-fold over that for isolates suspended in 0.3% formalinized phosphate-buffered saline before testing. The anti-H coagglutination test is a rapid, serologically specific, and inexpensive procedure for identifying V. vulnificus one step beyond primary isolation.     

Molecular typing of epidemic and nonepidemic Vibrio cholerae isolates and differentiation of V. cholerae and V. mimicus isolates by PCR-single-strand conformation polymorphism analysis

AIMS: To examine the utility of polymerase chain reaction (PCR)-single-strand conformation polymorphism (SSCP) analysis to differentiate epidemic and nonepidemic Vibrio cholerae isolates as well as to differentiate V. cholerae and Vibrio mimicus isolates. METHODS AND RESULTS: By both PCR-restriction fragment length polymorphism (RFLP) and PCR-SSCP analysis of groEL-I on chromosome 1 and groEL-II on chromosome 2, V. cholerae isolates gave distinct profiles compared with V. mimicus isolates. In addition, PCR-SSCP analysis of groEL-I and groEL-II could differentiate between V. cholerae epidemic and nonepidemic isolates. Interestingly, the relationships among strains based on groEL-I from chromosome 1 and groEL-II from chromosome 2 were congruent with each other, highlighting the conserved evolutionary history of both chromosomes in this species. CONCLUSIONS: PCR-SSCP is a powerful typing technique, which has the ability to differentiate V. cholerae and V. mimicus isolates. The epidemic V. cholerae O1/O139 serogroup isolates represent a clonal complex distinct from non-O1/non-O139 isolates that can be identified by PCR-SSCP analysis. SIGNIFICANCE AND IMPACT OF THE STUDY: This study highlights the effectiveness of using reliable molecular typing methods and in particular PCR-SSCP, to identify genetic variation among V. cholerae and V. mimicus isolates.     

Analysis of 16S-23S rRNA intergenic spacer regions of Vibrio cholerae and Vibrio mimicus

Vibrio cholerae identification based on molecular sequence data has been hampered by a lack of sequence variation from the closely related Vibrio mimicus. The two species share many genes coding for proteins, such as ctxAB, and show almost identical 16S DNA coding for rRNA (rDNA) sequences. Primers targeting conserved sequences flanking the 3' end of the 16S and the 5' end of the 23S rDNAs were used to amplify the 16S-23S rRNA intergenic spacer regions of V. cholerae and V. mimicus. Two major (ca. 580 and 500 bp) and one minor (ca. 750 bp) amplicons were consistently generated for both species, and their sequences were determined. The largest fragment contains three tRNA genes (tDNAs) coding for tRNAGlu, tRNALys, and tRNAVal, which has not previously been found in bacteria examined to date. The 580-bp amplicon contained tDNAIle and tDNAAla, whereas the 500-bp fragment had single tDNA coding either tRNAGlu or tRNAAla. Little variation, i.e., 0 to 0.4%, was found among V. cholerae O1 classical, O1 El Tor, and O139 epidemic strains. Slightly more variation was found against the non-O1/non-O139 serotypes (ca. 1% difference) and V. mimicus (2 to 3% difference). A pair of oligonucleotide primers were designed, based on the region differentiating all of V. cholerae strains from V. mimicus. The PCR system developed was subsequently evaluated by using representatives of V. cholerae from environmental and clinical sources, and of other taxa, including V. mimicus. This study provides the first molecular tool for identifying the species V. cholerae.