Cloning of enterotoxin gene from Aeromonas hydrophila provides conclusive evidence of production of a cytotonic enterotoxin

Culture filtrates of two Aeromonas hydrophila strains which were isolated from patients with diarrhea and assumed to be causative agents of the infections were shown to contain enterotoxic, cytotoxic, and hemolytic activities. Modest heat treatment of the filtrates inactivated the cytotoxic and cytolytic activities, but not the enterotoxic activity. The construction of cosmid gene banks in Escherichia coli of DNA from both A. hydrophila strains demonstrated that the determinants of the three activities are located on three different segments of the A. hydrophila chromosome. Both heated culture filtrates of A. hydrophila and nonheated filtrates of an E. coli clone containing the A. hydrophila enterotoxin gene provoked fluid accumulation in the rabbit ileal loop and suckling mouse models and caused elongation of Chinese hamster ovary cells. Differences in the responses of the models to the A. hydrophila enterotoxin and to the heat-labile and heat-stabile toxins of E. coli indicated that the former is distinct from the latter two types of toxin. These results constitute conclusive evidence for the production by A. hydrophila of a cytotonic enterotoxin that is distinct from the A. hydrophila cytotoxin and hemolysin and known E. coli enterotoxins.     

Purification and chemical characterization of a cholera toxin-cross-reactive cytolytic enterotoxin produced by a human isolate of Aeromonas hydrophila.

A bacterial protein toxin possessing hemolytic, enterotoxic, and cytotoxic activities as well as cross-reactivity to cholera toxin was purified from culture filtrates of a human diarrheal isolate of Aeromonas hydrophila (SSU). This cytolytic enterotoxin was purified by ammonium sulfate precipitation, hydrophobic chromatography using phenyl-Sepharose, anion-exchange chromatography on DEAE-Bio-Gel A, and size-exclusion high-performance liquid chromatography. The factor was a single polypeptide with an apparent molecular weight of 52,000 as determined by polyacrylamide gel electrophoresis. Automated amino acid sequence analysis confirmed that the toxin was a single chain and established a 25-residue N-terminal segment which was identical to that of aerolysin purified from culture supernatants of A. hydrophila isolate Ah65 originally obtained from rainbow trout as reported by Howard et al. (S. P. Howard, W. J. Garland, M. J. Green, and J. T. Buckley, J. Bacteriol. 169:2869-2871, 1987). However, the amino acid compositional analysis of the toxin produced by our human isolate (SSU) differed significantly from that of the Ah65 isolate. Taken together, these results strongly indicated that several toxic phenomena associated with A. hydrophila (SSU) culture filtrates, including hemolysis, cytotoxicity, and enterotoxicity as well as cross-reactivity to cholera toxin, all can occur on a single polypeptide. In addition, these results underline the fact that although aerolysin-related toxins isolated from culture filtrates of A. hydrophila are biologically similar, significant chemical and immunological differences may exist between toxins produced by individual isolates.     

Purification of cytotoxic enterotoxin of Aeromonas sobria by use of monoclonal antibodies

Cytotoxic enterotoxin of Aeromonas sobria was purified by affinity chromatography with monoclonal antibodies. The purified enterotoxin gave a single protein band in polyacrylamide gradient gel electrophoresis and its mol. wt estimated by this technique was 63,000; it had a pI of 6.2. The purified enterotoxin caused fluid accumulation in rat ileal loops and in infant mice, was cytotoxic to cultured cells, was haemolytic to human erythrocytes, and was lethal to mice after intravenous injection. The relative concentrations of enterotoxic, cytotoxic and haemolytic activities were approximately the same in a culture filtrate and in purified, electrophoretically homogeneous enterotoxin. The three activities were also inactivated to the same extent after incubation for 10 min at 56 degrees C. There was no immunological cross-reactivity with cholera toxin (CT) nor did antiserum to CT neutralise the biological effects of the toxin.     

Aeromonas cytotonic enterotoxin cross reactive with cholera toxin

Isolation by affinity chromatography from crude culture filtrate of Aeromonas sobria of protein that cross reacted with cholera toxin (CT) revealed a toxin that produced fluid accumulation in rat ileal loops and in infant mice and caused rounding of Y1 adrenal cells. All these activities were neutralised by antiserum to CT. There was no haemolytic or cytotoxic activity associated with this CT-cross reactive cytotonic enterotoxin. CT-cross reactive material detected in enzyme linked immunosorbent assay (ELISA) was produced by 25% of Aeromonas isolates from faeces of children with or without diarrhoea-26% of A. sobria, 20.0% of A. hydrophila and 24% of A. caviae tested gave positive ELISA results. Most strains that produced this cytotonic enterotoxin but no cytotoxic enterotoxin were isolated from children without diarrhoea. Toxin preparations from Aeromonas spp. that completely inhibited adenosine-5'-diphosphate-induced platelet aggregation, an effect related to elevation of intracellular cAMP, were, with one exception, cross reactive with CT in ELISA.     

Evidence for production of an enterotoxin and cholera toxin cross-reactive factor by Aeromonas hydrophila.

The enterotoxins produced by Aeromonas hydrophila were examined for biological activity by the rabbit ileal loop and suckling mouse assays, as well as by elongation of CHO cells. Antigenic evaluation of the culture filtrates from various isolates of A. hydrophila was performed by enzyme-linked immunosorbent assay with anti-cholera toxin and anti-Aeromonas enterotoxin. Heat stability data demonstrated the presence of a heat-labile cholera toxin cross-reactive factor and a heat stable non-cholera toxin cross-reactive enterotoxin. The biological activities of both enterotoxins were heat labile at 56 degrees C for 20 min.     

Production of cholera-like enterotoxin by a Vibrio cholerae non-O1 strain isolated from the environment.

Vibrio cholerae non-O1 strain E8498, isolated in 1978 from fresh water in Louisiana, produced a vascular permeability factor when cultured in shallow resting cultures of Casamino Acids-yeast extract-glucose medium for 24 h at 30 degrees C. Undiluted resting culture filtrates contained heat-labile permeability factor activity which was only partially neutralized by cholera antitoxin and GM1 ganglioside. Supernatants concentrated with PM-10 membranes caused hemorrhage and necrosis in rabbits within 1 h after intracutaneous injection, whereas appropriate dilutions of both filtrates and concentrates demonstrated delayed permeability factor activity, without hemorrhage or necrosis, which was indistinguishable in appearance from that caused by purified cholera enterotoxin produced by V. cholerae O1 Inaba strain 569B. Crude E8498 filtrates contained the biological equivalent of about 5 ng/ml of purified enterotoxin. Permeability factor activity in the fraction obtained by 20 to 50% saturation of filtrate concentrate with ammonium sulfate could be completely neutralized by reference standard cholera antitoxin prepared against purified 569 B enterotoxin. Hemorrhagic activity was unaffected by cholera antitoxin. A 5,000-fold concentrate of the culture supernatant yielded a line of identity with purified cholera enterotoxin in an agar gel double-diffusion test against cholera antitoxin purified by affinity column chromatography with BrCN-activated Sepharose 4B-linked purified cholera enterotoxin as the adsorbent. These findings indicate that V. cholerae non-O1 E8498 produces a permeability factor which is immunologically and biologically indistinguishable from that produced by a strain of V. cholerae O1 classical biotype.     

Purification and some properties of Aeromonas hydrophila hemolysin

A hemolysin produced by a strain of Aeromonas hydrophila isolated from a patient with diarrhea was purified by acid precipitation and quarternary aminoethyl-Sephadex chromatography. The molecular weight of the hemolysin was estimated at 50,000 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and at 48,000 by Sephadex G-100 gel filtration. In polyacrylamide gel electrophoresis at pH 4.0 and sodium dodecyl sulfate-polyacrylamide gel electrophoresis, the hemolysin migrated as a single band, whereas electrophoresis at pH 9.4 and thin-layer isoelectric focusing demonstrated multiple bands. The results may indicate charge isomers of the hemolysin. The purified hemolysin had a hemolytic activity of 134 hemolytic units per microgram of protein on rabbit erythrocytes. It caused fluid accumulation in infant mouse intestines and rabbit ligated ileal loops. Purified hemolysin also elicited cytotoxicity to Vero cells and lethal toxicity to mice. All these biological activities were lost on heating for 5 min at 56 degrees C. These findings support the notion that A. hydrophila hemolysin is a cytotoxic enterotoxin.     

Purification and characterization of an extracellular secretogenic non-membrane-damaging cytotoxin produced by clinical strains of Vibrio cholerae non-O1.

Some clinical strains of Vibrio cholerae non-O1 produce an extracellular factor that evokes a rapid and dramatic cytotoxic response which manifests as cell rounding of Chinese hamster ovary (CHO) and HeLa cells without accompanying membrane damage. This study was performed to establish the identity of the non-membrane-damaging cytotoxin (NMDCY), which was not inhibited by antitoxins against cholera toxin, heat-labile toxin of enterotoxigenic Escherichia coli, El Tor hemolysin, Shiga-like toxin I, and Shiga-like toxin II, indicating that NMDCY did not bear an apparent immunological relationship with the above toxins and hemolysin. Brain heart infusion broth and AKI medium supported the maximal production of NMDCY; culture supernatant of AKI medium was found to be free of hemolysin activity, whereas in brain heart infusion broth hemolysin was coproduced with NMDCY. Maximal production of NMDCY in AKI medium was observed at 37 degrees C under shaking conditions with the pH of the medium adjusted to 8.5. NMDCY was purified to homogeneity by a three-step purification procedure which increased the specific activity of the cytotoxin by 1.7 X 10(5)-fold. The denatured molecular weight of the purified toxin was 35,000, and the cytotoxin was heat labile and sensitive to trypsin. Purification of the cytotoxin revealed an enterotoxic activity as reflected by its ability to accumulate fluid in the rabbit ileal loop. Both the cytotoxic and enterotoxic activities of NMDCY could be inhibited or neutralized by antiserum raised against purified cytotoxin but not by preimmune serum. Immunodiffusion test between purified NMDCY and antiserum gave a single well-defined precipitin band which showed reactions of complete identity, while, in an immunoblot assay, a well-defined single band was observed in the 35-kDa region. Our results indicate that the cytotoxic and enterotoxic activities expressed by NMDCY appear to contribute to the pathogenesis of the disease associated with V. cholerae non-O1 strains which produce this cytotoxin.     

Production of "Asao toxin" by Aeromonas strains isolated from feces and drinking water

Cultures of Aeromonas species were tested for production of a toxin recently purified by Asao et al. (T. Asao, Y. Kinoshita, S. Kozaki, T. Uemura, and G. Sukaguchi, Infect. Immun. 46:122-127, 1984) and described as a hemolysin with enterotoxic and cytotoxic activity. The toxin was produced by only 63% of Aeromonas sobria strains and by 93% of Aeromonas hydrophila strains. Also, 54% of A. hydrophila strains produced another cytotoxic entity.     

The type III secretion system and cytotoxic enterotoxin alter the virulence of Aeromonas hydrophila

Many gram-negative bacteria use a type III secretion system (TTSS) to deliver effector proteins into host cells. Here we report the characterization of a TTSS chromosomal operon from the diarrheal isolate SSU of Aeromonas hydrophila. We deleted the gene encoding Aeromonas outer membrane protein B (AopB), which is predicted to be involved in the formation of the TTSS translocon, from wild-type (WT) A. hydrophila as well as from a previously characterized cytotoxic enterotoxin gene (act)-minus strain of A. hydrophila, thus generating aopB and act/aopB isogenic mutants. The act gene encodes a type II-secreted cytotoxic enterotoxin (Act) that has hemolytic, cytotoxic, and enterotoxic activities and induces lethality in a mouse model. These isogenic mutants (aopB, act, and act/aopB) were highly attenuated in their ability to induce cytotoxicity in RAW 264.7 murine macrophages and HT-29 human colonic epithelial cells. The act/aopB mutant demonstrated the greatest reduction in cytotoxicity to cultured cells after 4 h of infection, as measured by the release of lactate dehydrogenase enzyme, and was avirulent in mice, with a 90% survival rate compared to that of animals infected with Act and AopB mutants, which caused 50 to 60% of the animals to die at a dose of three 50% lethal doses. In contrast, WT A. hydrophila killed 100% of the mice within 48 h. The effects of these mutations on cytotoxicity could be complemented with the native genes. Our studies further revealed that the production of lactones, which are involved in quorum sensing (QS), was decreased in the act (32%) and aopB (64%) mutants and was minimal (only 8%) in the act/aopB mutant, compared to that of WT A. hydrophila SSU. The effects of act and aopB gene deletions on lactone production could also be complemented with the native genes, indicating specific effects of Act and the TTSS on lactone production. Although recent studies with other bacteria have indicated TTSS regulation by QS, this is the first report describing a correlation between the TTSS and Act of A. hydrophila and the production of lactones.     

Changes in Intestinal Fluid Transport and Immune Responses to Enterotoxins Due to Concomitant Parasitic Infection

The effect of a parasitic infection on enterotoxic diarrhea and on local and systemic formation of antibody to the toxin after immunization was studied in mice. Trichinella spiralis infection was chosen as the model, since the effects of the parasite when residing in both intestinal and extraintestinal sites can be studied. It was found that during the intestinal stage of the infection, the fluid response to cholera toxin as well as dibutyryl-cyclic adenosine 3′,5′ -monophosphate was greatly enhanced and that this was associated with a marked reduction in the absorption of fluid from the intestine. Later in the infection (migration stage), fluid accumulation in response to cholera toxin was significantly reduced, whereas absorption was normal and secretion in response to dibutyryl-cyclic adenosine 3′,5′-monophosphate was somewhat increased. Still later in the infection (muscular stage), the fluid-secretory response to cholera toxin was normal. There was a drastic depression of local formation of antitoxin of both immunoglobulin and immunoglobulin classes in mice given the first two of four oral immunizations with cholera toxin during the intestinal stage of T. spiralis infection. When the priming was given before or after the intestinal stage, the local antitoxin response was not affected. The titers of circulating antibodies were also depressed in mice given the first immunizations during the intestinal stage. In addition, significant though less pronounced depression of the serum antibody response was observed in mice primed during the extraintestinal stage.     

Isolation of Aeromonas hydrophila in diarrhea. Characterization of enterotoxinogenic strains and clinical relations

Aeromonas hydrophila is isolated from diarrhoea specimens with increasing frequency. The interest in this organism at the present time is related to the fact that it can produce a number of toxins, in particular alpha and beta cytotoxic haemolysins, an enterotoxin and various enzymes. The authors determined the frequency of isolation of this organism and tested the haemolytic, cytotoxic and enterotoxic effects of culture filtrates in all of the stool specimens received in their laboratory over a period of 9 months. At the same time, the clinical context was defined in order to demonstrate a relation between the aptitude of the strains to produce toxins and the presence of diarrhoea. The frequency of isolation of A. hydrophila was 0.88 per cent, which corresponds to 67 strains. 38 strains presented a haemolytic and/or enterotoxic activity, i.e. 57 per cent of the strains isolated. In diarrhoeal stools, 67 per cent of the A. hydrophila isolated produced at least one of the toxins, while in the group of patients without diarrhoea, only 38 per cent of the strains isolated produced toxins. The results obtained reveal a statistically significant correlation between the production of cytotoxic haemolysin and the presence of diarrhoea. In contrast, there was no correlation between the production of enterotoxin and the presence of diarrhoea. Twenty of the 67 strains ware isolated from children under the age of 2 years. In 40 per cent of cases, no other aetiology could be found for the diarrhoea, apart from the isolation of A. hydrophila.(ABSTRACT TRUNCATED AT 250 WORDS)     

Characterization of Shigella dysenteriae 1 (Shiga) toxin purified by anti-Shiga toxin affinity chromatography.

Shigella dysenteriae 1 (Shiga) toxin was purified from whole-cell lysates by antitoxin affinity column chromatography, radioiodination, and Sephacryl S-200 gel filtration of 125I-labeled affinity column eluates. Two chromatographic peaks were observed. The percentage of radioactivity in peak I samples immunoprecipitated with antitoxin ranged from 95 to 100%. A pool of samples from this first peak contained over 90% of the HeLa-cell-cytotoxic units applied to the column and was enterotoxic for rabbit ileal loops and lethal for rabbits. This radiolabeled material migrated as a single cytotoxic band after nondenaturing polyacrylamide gel electrophoresis, but formed three bands, of 33,000, 29,000, and 4,000 to 7,000 daltons, after sodium dodecyl sulfate-polyacrylamide gel electrophoresis. In addition, material estimated as 7,000 daltons by Bio-Gel P-10 chromatography could be generated by treatment of S-200 peak I samples with 8 M urea. Pooled fractions from the second S-200 peak were separable into several low-molecular-weight peaks on a P-10 column. One of these P-10 peaks (7,000 daltons) was 27% immunoprecipitable with antitoxin. These data indicate that three of the known biological activities of Shiga toxin are associated with a 33,000-dalton substance which can be dissociated into 29,000- and 4,000- to 7,000-dalton components.     

Purification and characterization of an enterotoxin from Bacteroides fragilis.

An enterotoxin produced by Bacteroides fragilis was purified to homogeneity and characterized as to its biological activity and basic molecular properties. Toxin preparations were prepared by growing B. fragilis VPI 13784 in brain heart infusion broth to early stationary phase, immediately precipitating the culture supernatant fluid with 70% ammonium sulfate, and stabilizing the precipitate with the protease inhibitor TPCK (tolylsulfonyl phenylalanyl chloromethyl ketone). The toxin was sequentially purified by anion-exchange chromatography on Q-Sepharose, hydrophobic interaction chromatography on phenyl-agarose, and high-resolution ion-exchange chromatography on Mono Q. The toxin appeared homogeneous as judged by polyacrylamide gel electrophoresis. The estimated molecular weight of the highly purified toxin as determined by gel filtration chromatography on Superose-12 and sodium dodecyl sulfate-polyacrylamide gel electrophoresis is 19,000. It has an isoelectric point of approximately 4.5 and is stable at pHs 5 to 10. The purified toxin is stable at -20 and 4 degrees C and upon freeze-drying, but it is unstable at temperatures above 55 degrees C. It is sensitive to proteinase K and Streptomyces protease but is resistant to trypsin and chymotrypsin. The activity of the purified toxin is neutralized by antiserum to a toxigenic strain of B. fragilis but not by antiserum to nontoxigenic strains. N-terminal amino acid analysis reveal an unambiguous sequence of Ala-Val-Pro-Ser-Glu-Pro-Lys-Thr-Val-Tyr-Val-Ile-Xxx-Leu-Arg-Glu-Asn-Gly- Ser-Thr . The highly purified toxin induced a strong fluid accumulation response in the lamb ileal-loop assay as well as a cytotoxic response (cell rounding) on HT-29 colon carcinoma cells. Thus, the purified toxin can cause both enterotoxic and cytotoxic activities.     

DNA adenine methyltransferase influences the virulence of Aeromonas hydrophila

Among the various virulence factors produced by Aeromonas hydrophila, a type II secretion system (T2SS)-secreted cytotoxic enterotoxin (Act) and the T3SS are crucial in the pathogenesis of Aeromonas-associated infections. Our laboratory molecularly characterized both Act and the T3SS from a diarrheal isolate, SSU of A. hydrophila, and defined the role of some regulatory genes in modulating the biological effects of Act. In this study, we cloned, sequenced, and expressed the DNA adenine methyltransferase gene of A. hydrophila SSU (dam(AhSSU)) in a T7 promoter-based vector system using Escherichia coli ER2566 as a host strain, which could alter the virulence potential of A. hydrophila. Recombinant Dam, designated as M.AhySSUDam, was produced as a histidine-tagged fusion protein and purified from an E. coli cell lysate using nickel affinity chromatography. The purified Dam had methyltransferase activity, based on its ability to transfer a methyl group from S-adenosyl-l-methionine to N(6)-methyladenine-free lambda DNA and to protect methylated lambda DNA from digestion with DpnII but not against the DpnI restriction enzyme. The dam gene was essential for the viability of the bacterium, and overproduction of Dam in A. hydrophila SSU, using an arabinose-inducible, P(BAD) promoter-based system, reduced the virulence of this pathogen. Specifically, overproduction of M.AhySSUDam decreased the motility of the bacterium by 58%. Likewise, the T3SS-associated cytotoxicity, as measured by the release of lactate dehydrogenase enzyme in murine macrophages infected with the Dam-overproducing strain, was diminished by 55% compared to that of a control A. hydrophila SSU strain harboring the pBAD vector alone. On the contrary, cytotoxic and hemolytic activities associated with Act as well as the protease activity in the culture supernatant of a Dam-overproducing strain were increased by 10-, 3-, and 2.4-fold, respectively, compared to those of the control A. hydrophila SSU strain. The Dam-overproducing strain was not lethal to mice (100% survival) when given by the intraperitoneal route at a dose twice that of the 50% lethal dose, which within 2 to 3 days killed 100% of the animals inoculated with the A. hydrophila control strain. Taken together, our data indicated alteration of A. hydrophila virulence by overproduction of Dam.     

Purification and properties of Klebsiella pneumoniae heat-stable enterotoxin.

The enterotoxic material in cell-free growth preparations of Klebsiella pneumoniae serotype 5 was purified by sequential ultrafiltration and gel filtration (GF) procedures and the fractions were assayed for enterotoxic activity by determining their ability to induce in vivo net water secretion in the rat jejunum. Whole-cell lysates were inactive. Anaerobic broth culture conditions yielded a 10-fold increase in toxin production over aerobic conditions. Enterotoxic activity was absent in the UM-10 retentate of the broth filtrate but present in both the retentate and filtrate of the UM-2 membrane. GF of the two UM-2 ultrafiltration fractions through a Sephadex G-25 column yielded an active eluate, whose potency was increased by 10- or 200-fold, in or adjacent to the void volume. When subsequently passed through a G-50 column, these pools eluted at a Kav of between 0.4 and 0.6 and were further increased in potency by two- or fivefold. A second equally potent fraction was also recovered in the void volume of the G-50 eluate of the UM-2 filtrate; this may represent a polymer. Progressive purification by GF was associated with an increased protein and decreased carbohydrate content of the most active fractions. The most active G-50 eluate of the UM-2 retentate had a minimal effective enterotoxic dose of 5 mug/ml and that of the filtrate was less than 0.1 mug/ml. Heating the active GF eluates to 100 C for 30 min did not abolish enterotoxic activity and lowering the pH to 1 or incubation with either Pronase or trypsin had no effect on activity. These observations indicate that K. pneumoniae heat-stable enterotoxin is probably a single toxin with an apparent molecular weight in the range of 5,000. The elution characteristics during GF as well as the chemical composition of the most purified enterotoxin fractions indicate that the toxin is not associated with endotoxin.     

Production of antisera against the enterotoxin of Bacteroides fragilis and their use in a cytotoxicity neutralization assay of HT-29 cells.

To study the enterotoxin of Bacteroides fragilis, the colon carcinoma cell line HT-29 was used in a standard cytotoxicity assay. We produced high-titer neutralizing antisera in rabbits and goats against both crude and purified toxin and developed a cytotoxicity neutralization assay for use in confirming enterotoxin activity in culture filtrates and stools. The neutralization titers of the antisera on the colon carcinoma cell line HT-29 ranged from 1,600 to 2,400. In an antibody screening enzyme-linked immunosorbent assay, titers ranged from 10(4) to 10(5). The antisera produced against the highly purified toxin also neutralized the enterotoxic activity of the toxin and were monospecific by immunoelectrophoresis.     

Prevalence of enterotoxin genes in Aeromonas spp. isolated from children with diarrhea, healthy controls, and the environment

Aeromonads are causative agents of a number of human infections. Even though aeromonads have been isolated from patients suffering from diarrhea, their etiological role in gastroenteritis is unclear. In spite of a number of virulence factors produced by Aeromonas species, their association with diarrhea has not been clearly linked. Recently, we have characterized a heat-labile cytotonic enterotoxin (Alt), a heat-stable cytotonic enterotoxin (Ast), and a cytotoxic enterotoxin (Act) from a diarrheal isolate of Aeromonas hydrophila. Alt and Ast are novel enterotoxins which are not related to cholera toxin; Act is aerolysin related and has hemolytic, cytotoxic, and enterotoxic activities. We studied the distribution of the alt, ast, and act enterotoxin genes in 115 of 125 aeromonads isolated from 1, 735 children with diarrhea, in all 27 aeromonads isolated from 830 control children (P = 7 x 10(-4) for comparison of rates of isolation of aeromonads from cases versus those from controls), and in 120 randomly selected aeromonads from different components of surface water in Bangladesh. Aeromonas isolates which were positive only for the presence of the alt gene had similar distributions in the three sources; the number of isolates positive only for the presence of the ast gene was significantly higher for the environmental samples than for samples from diarrheal children; and isolates positive only for the presence of the act gene were not found in any of the three sources. Importantly, the number of isolates positive for both the alt and ast genes was significantly higher for diarrheal children than for control children and the environment. Thus, this is the first study to indicate that the products of both the alt and ast genes may synergistically act to induce severe diarrhea. In 26 patients, Aeromonas spp. were isolated as the sole enteropathogen. Analysis of clinical data from 11 of these patients suggested that isolates positive for both the alt and ast genes were associated with watery diarrhea but that isolates positive only for the alt gene were associated with loose stools. Most of the isolates from the three sources could be classified into seven phenospecies and eight hybridization groups. For the first time, Aeromonas eucrenophila was isolated from two children, one with diarrhea and another without diarrhea.