versión impresa ISSN 0001-3714
Rev. Microbiol. v. 29 n. 3 São Paulo Sept. 1998
Vera Lúcia Mores Rall, Sebastião Timo Iaria*, Sandra Heidtmann, Fabiana Cristina Pimenta, Rosa Carvalho Gamba, Débora Midori Myaki Pedroso
Departamento de Microbiologia, Instituto de Ciências Biomédicas, Universidade de São Paulo, São Paulo, Brasil Submitted: February 20, 1997; Returned to authors for corrections: April 23, 1997;
Approved: July 23, 1998
Aeromonas has been described as an emergent foodborne pathogen of increasing importance. In this study, we report that 48% of 50 Pintado fish samples collected at the retail market of São Paulo city were positive for Aeromonas sp, as detected by the direct plating method. When the presence/absence method was used, the positivity was 42%. A. caviae was the most frequent species, followed by A. hydrophila and A. sobria. Production of cytotoxic enterotoxin, observed in suckling mouse assay, was detected in 67% of A. sobria strains, in 60% of A. hydrophila strains and in 40% of A. caviae strains. In vitro tests, performed with HEp-2 cells, showed that 88% of A. hydrophila, 27% of A. sobria and 13% of A. caviae strains were positive for this toxin. The in vivo production of cytotonic enterotoxin, tested after heating the filtrates at 56ºC for 20 minutes, was detected in 17% of A. sobria, in 10% of A. caviae and in none of A. hydrophila strains in vivo. All analyzed strains did not alter HEp-2 cells. 20% and 16% of A. sobria and A. caviae isolates, respectively, presented capacity to adhere to HEp-2 cells. In counterpart, invasion of HEp-2 cells was not observed in any isolate. The Aeromonas isolates were sensitive to the majority of the antimicrobiol agents tested.
Kew words: Aeromonas sp, virulence factors, drug susceptibility, fish
INTRODUCTION Aeromonas spp belong to the Vibrionaceae family and some species, such as A. hydrophila, A. sobria and A. caviae, have been described as emergent foodborne pathogens causing gastroenteritis. These species have been isolated from fresh food, seafood, milk, some kinds of meats, vegetables, ice cream and salads, among others.
So far, the mechanisms by which these microorganisms do cause diarrhea have been only partially elucidated but it is known that they produce enterotoxins and certain enzymes and are able to adhere to cell membranes and invade them (9,16,17).
Keusch and Donta (14) divided the enterotoxins into two types: cytotonic and cytotoxic ones. These two types can be differentiated by heating them at 56ºC for 20 minutes which inactivates the cytotoxic activity but not the cytotonic one. The cytotonic enterotoxin (heat lable), like the E. coli LT and the coleric toxins, stimulates cAMP that leads to the activation of adenylate cyclase in the intestine, stimulating salt and water secretion and causing diarrhea. The cytotoxic enterotoxin (heat stable) causes cell death and produces dysentery-like symptoms. The enterotoxin activity of both toxins can be demonstrated using the suckling mouse test or the ligated rabbit ileal loop system. The cytotoxic and cytotonic effects can be seen through "in vitro" tests using several cells lineages like Vero, CHO, Y-1 or HEp-2.
In order to assure the Aeromonas pathogenicity the verification of other factors like ability to adhere to the intestinal mucosa and invasiveness is also required. Aeromonas strains have been shown to adhere and invade HEp-2 cells (6, 18, 27). One of the most common mechanism of bacterial adhesion is mediated by pili. Other mechanisms depend on bacterial LPS or non fimbrial proteins. Aeromonas isolates adhere to human erythrocytes and bucal epithelium and adhesion has been associated to the presence of pili (2).
In Brazil, there are only few studies about Aeromonas and its virulence factors. So, the present work is intended to detect the occurrence of Aeromonas sp in fresh fish (Pseudoplatystoma sp) retailed in São Paulo city, the production of some virulence factors (adhesion, invasion and toxins production) and the susceptibility of these isolates to antimicrobial agents.
MATERIALS AND METHODS Food Samples:
A total of 50 fish samples (Pseudoplatystoma sp) were purchased in the period of November 1992 to September 1993 in supermarkets in São Paulo city. The samples were transported to the laboratory under refrigeration and analysed immediately.
Using a sterile blender, 25 g of fish sample were homogenized with 225 ml of 0.1% peptone saline solution (3-4 minutes). Then, tenfold serial dilutions up to 10-5 using the same diluent (5) were done.
The strains were isolated using the direct plating method and Presence/Absence test (P/A). The selective agar used was starch ampicillin agar (SAA) according to Palumbo et al (22).
In the direct plating method used for Aeromonas enumeration, 0.1 ml of each dilution was spreaded onto the surface of SAA plates and incubated at 28ºC for 24 hours (22).
The P/A test was performed using 10 ml of tripticase soy broth (TSB) (Difco), added with 20 µg/ml of ampicillin (Sigma). Ten ml of the initial dilution (10-1) was inoculated in this broth and incubated at 28ºC for 24 hours. After incubation, it was streaked onto the surface of SAA and incubated under the same conditions.
After incubation, colonies were transferred to tripticase soy agar (TSA) (Difco) and triple-sugar-iron slants (TSI) (Difco) and incubated at 28ºC for 24 hours.
The colonies that showed typical reaction in TSI and were oxidase and catalase positive were confirmed as Aeromonas spp according to POPOFF (1984) (23) The confirmed strains were maintained in TSA at room temperature and also in TSB with 20% of glycerol (Synth) at - 70ºC.
Preparation of cell-free supernatant. Strains of Aeromonas spp were inoculated in 20 ml of TSB and incubated at 28ºC in stationary culture for 24 hours. The cultures were centrifuged at 10,000 rpm for 30 minutes (Sorvall GLC-2B/Du Pont Instruments) and 0.45 µm filter sterilized (Millipore). Ten ml were boiled at 56ºC during 20 minutes for inactivation of heat-labile enterotoxins. The sterile filtrates were frozen and stored at - 70ºC (27).
Exotoxins assay. The heat-lable and heat-stable enterotoxins were both tested by means of two tests. a) suckling mouse test: two drops of 2.5% pontamine sky blue were added to one ml of each cell-free supernatant and 100 µl were injected into the stomach of 2-4 day-old suckling mice, at the C3H/HePas isogenic lineage. Three mice were used for test. After 4 hours at room temperature, the animals were killed with chloroform. The entire intestinal tract of each mouse was removed and immediately weighed. The entire remaining body was weighed as well. The ratio between the intestinal weight and the body weight was determined. The 0.08 ratio was considered positive. A positive control was included in each test. b) cultures of cells:. Semiconfluent monolayers of 104 to 105 HEp-2 cells per ml, grown in Eagle Minimal Essential Medium (MEM) with 10% fetal calf serum (FCS) in a 96 well tissue culture tray were obtained by incubation for 24 hours at 37ºC in a humidified 5% CO2 incubator. The cells were washed with Hanks balanced salt solution (HBSS) (Merck). 100 µl of pure culture filtrates were added to MEM to obtain 1:5, 1:8 and 1:10 final dilutions. Two samples of each dilution were made and the tissue culture tray was incubated overnight at 37ºC in 5% CO2 atmosphere. The plates were examined using inverted light microscopy for cytotonic and cytotoxic effects. A negative control of not inoculated MEM and a positive cytotoxic strain were used in each test. Positive cytotoxic activity was taken as 50% cell rounding and detachment. Positive cytotonic activity was taken as cellular elongation (16).
Adhesion test. For the adhesion assay, strains were grown on blood agar (BA) (Difco) incubated at 28ºC overnight. Isolated colonies were inoculated into 10 ml of BHI and incubated for 16-18 hours at 28ºC. Log phase cultures were prepared adding 0.5 ml of the overnight culture to 10 ml of BHI and incubation for 3 hours at 28ºC. A final suspension containing 2-3x106 UFC/ml was obtained by dilution with MEM. Semiconfluent monolayers of HEp-2 cells were grown in 24 well plates with a cover slip in each well for 24 hours at 37ºC in a humidified 5% CO2 incubator. The wells were washed with 2 ml of HBSS before the addition of bacterial cells. All strains were tested three times each, including a positive control. After a 90 minutes incubation at 37ºC, the wells were washed four times with HBSS in order to remove non adherent bacterial cells. The cover slips were fixed with 3:1 methanol: acetic acid for 5 minutes, stained by Gram method and then mounted on glass microscope slides (6,9).
The adhesion was assessed in bright-field microscope at 1000x under oil immersion and the adhesion pattern was determined as described by Grey and Kirov (9).
Invasion test. This test was performed according to Watson et al. (27) with modifications. The strains of Aeromonas sp were cultured in BA plates at 28ºC for 24 h, and a single colony was inoculated into 10 ml of BHI and incubated under the same conditions. Afterwards, 0.5 ml from this culture was inoculated into 10 ml of BHI and reincubated. Bacterial cultures were diluted in MEM with 1% FBS (fetal bovine serum), to a concentration of 5x105 cells per ml, semiconfluent monolayers of 5x104 to 1x105 HEp-2 cells were grown in a 24 well tray at 37ºC for 24 h in 5% CO2 atmosphere and each well was washed three times with 1 ml of HBSS. One ml of bacterial cultures were added to each well with three repetitions. The tray was incubated for 3 hours at 37ºC in a humidified 5% CO2 incubator; subsequently, the monolayers were washed three times with 1 ml of HBSS and a similar volume of a MEM solution containing 5% FCS and 10 µg/ml gentamicin, was added to each well. The tray was incubated for 2 hours under the same conditions and again the monolayers were washed three times with HBSS and reincubated for more two hours in fresh MEM with 5% FCS. After washing the monolayers four times with HBSS, 1 ml of lysing solution (0.01 M NaH2PO4, Tween 20 1% and 0.025% tripsin at pH 8.0) was added to each well and the tray was incubated for 30 minutes at 35ºC. The lysate was serial diluted in BHI broth and 0.1 ml were plated on BHI agar and incubated at 28ºC for 24 hours to determine the number of CFU per ml. The strains were considered invasive when the concentration was 5x106 CFU/ml, which is one log higher than the initial inoculum. Strains were classified as noninvasive when not recovered or when the concentration did not exceed 2x103 CFU/ml.
The positive control was a Aeromonas strain supplied by Federal University of São Paulo.
Drug susceptibility: The susceptibility to antibiotic was tested using satured paper discs (Laborclin). The cultures were spread on Mueller-Hinton agar (105 UFC/ml), the discs were set onto the agar surface and incubated at 28ºC for 24 hours. The resistance of the strains were determined according to the size of the inhibition halo (7).
The drugs used were: gentamicin (10µg), choramphenicol (30mg), tetracyclin (30mg), penicillin (10µg), carbecillin (100µg), cephalotin (30µg), cephoxetin (30µg), kananicin (30µg), nalidixic acid (30µg) and streptomycin (10µg) (8).
RESULTS AND DISCUSSION Using the direct plating method 24 (48%) fish samples were positive for Aeromonas spp, whereas 21 (42%) were positive when P/A test was used. A. caviae was the most frequent species (30 strains), followed by A. hydrophila (25 strains) and A. sobria (6 strains). These results agree with other reports, where A. caviae was more frequent in fresh water (fish natural habitat) and in other foods with direct contact with water through irrigation, like vegetables.
Table 1 shows the in vivo and in vitro results of the cytotoxic and cytotonic toxin assays. The production of a cytotoxic enterotoxin in the suckling mouse assay was detected in 67% of A. sobria strains, in 60% of A. hydrophila strains and in 40% of A. caviae strains. In vitro tests were performed with HEp-2 cells, in which 88% of A. hydrophila, 17% of A. sobria and 13% of A. caviae were positive. Some isolates produced toxin with concomitant in vivo enterotoxic effects and in vitro cytotoxic effects. Only 48% of A. hydrophila strains, 17% of A. sobria and 3% of A. caviae isolates presented this pattern.
Table 1. Positivity of A. caviae, A. hydrophila and A. sobria strains isolated from fish (Pseudoplatystoma sp), for cytotoxic and cytotonic enterotoxins production.
LT: heat lable
N: number of isolates tested
n: number of positive isolates
Burke et al. (3) reported that A. sobria was the most frequently isolated species from children with diarrhea and the production of cytotoxic enterotoxin associated with diarrhea is more often related to A. sobria and A. hydrophila and seldom occurs with A. caviae. Kirov et al. (15) also observed a greater toxin production by A. sobria strains using Vero cells. However, we found the cytotoxic effect in HEp-2 cells to be more often detected in A. hydrophila strains.
The distinct results in in vitro and in vivo tests can be explained by the production of two kinds of cytotoxic toxins. Burke et al. (4), working with a A. sobria strain, reported the production of a cholerae-like enterotoxin and a cytotoxic toxin. Potomski et al. (24) reported that A. sobria produced two types of supernatants, one of them containing a toxin with a molecular weight of 63Kda not didnt cross react with cholerae toxin antisera, and a second one, combining a toxin that cross reactive with this antisera. Others authors, like Rose et al. (25) and Houston et al. (12), also reported a toxin that presented cross reaction with choleric toxin antisera and had a molecular weight of 52 kDa.
The positivity for production of cytotonic enterotoxin was low: only one (17%) out of 6 A. sobria strains, three (10%) out of 30 A. caviae strains and none out of 25 A. hydrophila strains were positive in the suckling mouse assay. None of the analyzed strains did alter HEp-2 cells. These results differ from those reported by Chopra and Houston (7), who observed that 48% of A. hydrophila clinical strains were cytotonic, and Majeed et al. (19), who detected positive results in 75% of the A. sobria strains, 67% of the A. hydrophila and 5% in the A. caviae strains tested.
The results for the adhesion assay in HEp-2 cells showed that only 4 out of 26 isolates of A. caviae (15,4%) and 1 out of 5 isolates of A. sobria were positive in this test. None of 3 A. hydrophila strains tested were positive. However, several Aeromonas sp strains were toxin producers, making the evaluation of results impossible because cellular death occurred before the invasion test was completed. This characteristic was also described by Carrelo et al. (6) and by Grey and Kirov (9).
According to criteria of Carrelo et al. (6) four adherent A. caviae strains presented low adherence pattern (between 1- 2,6 bac/cell. However, according to these same criteria, the only one adherent A. sobria strain showed a high capacity of adherence to HEp-2 cells (33.4 bac/cells). Carrelo et al. (6) and Grey and Kirov (9) suggest a positive association between the ability to cause diarrhea and the high adhesion level. These authors observed that A. sobria and A. hydrophila showed greater adhering capacity than A. caviae. These positive isolates in the adhesion assay did not invade HEp-2 cells, which led the authors to conclude that adhesion to HEp-2 cells alone could be a virulence mechanism in A. caviae, just as in E. coli.
Grey and Kirov (9) found that 13% of A. caviae strains were positive in the adhesion assay using HEp-2 cells, a result close to ours (15.4%).
Previously, Carrelo et al. (6) had shown that 71% of A. caviae strains were positive in the adhesion test with HEp-2 cells but the level of adhesion was always low. The same authors reported a 15% adherent A. sobria strains, always with a high adhesion level. Only three A. hydrophila isolates were tested for adhesion capacity and none presented the characteristic, which is in agreement with Grey and Kirov (9), who observed little positivity for this species, independently from the source.
None of the Aeromonas sp strains was capable of invading HEp-2 cells. Some factors must be herein considered as related to this topic. First of all, only a few strains of A. hydrophila and A. sobria were tested (three and five, respectively) and A. caviae is considered the species with the lowest invasion capacity (8,9). Besides, the most frequently tested strains in invasion studies have been recovered from diarrhea stools. Lawson et al. (18) have observed that ten A. hydrophila strains isolated from fresh water were not invaders, while five strains of the same species that had been isolated from stools presented invasion capacity in HEp-2 cells. In the invasion assay destruction due to toxin production occurred in 27 (44.3%) strains.
Table 2 shows that all isolates of Aeromonas sp were resistant to penicillin, a property due to beta-lactamase production and all strains presented a high resistance to cefalotin (84%), which agree with other authors (10). All strains were sensitive to cephoxetin and gentamicin, as previously reported by Masher et al. (20). MacCracken (21) observed that a great number of strains seemed to be more sensitive to gentamicin and kanamicin than to streptomycin. These data are in accordance with our findings, inasmuch as 100% of the Aeromonas isolates were sensitive to gentamicin, 93% to kanamicin and 89% to streptomycin. About this last drug, it was observed that A. sobria resulted the highest resistance (33%) when compared to A. hydrophila and A. caviae. Aeromonas strains showed high sensitivity to chloramphenicol: all the isolates of A. sobria and A. caviae and 96% of A. hydrophila were sensitive to this drug. Resistance to this antibiotic seems to be related to the action of an enzyme, the chloramphenicol acetiltranferase. The majority (95%) of Aeromonas sp isolates was sensitive to nalidixic acid, as previously reported by Sleider et al. (26). In the present study, little variation in sensitivity to tetracycline was detected, since 100% of A. sobria isolates, 93% of A. caviae and 92% of A. hydrophila were sensitive. Janda et al. (13) found similar results with A. hydrophila, 98% of which were sensitive. Hedges et al. (11) suggested that such a low resistance could be due to a plasmid.
Table 2. Antibiotic sensitivity presented by Aeromonas strains isolated from fish (Pseudoplatystoma sp).
N: number of isolates tested
n: number of sensitive isolates
ACKOWLWDGMENTS The authors thank to Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) for financial support of this research (Grant # 93/3293-3). The authors are grateful to Dr. L.R. Trabulsi, Department of Microbiology, São Paulo University, Brazil, and Dr. R. Toledo, Department of Microbiology, Federal University of São Paulo, Brazil, for supplying strains of Aeromonas sp used as controls in the tests. We also thank Dr. C. Harsi, Department of Microbiology for providing HEp-2 cells, T. Marques, Department of Immunology for supplying the isogenic mice and C.A. Silva, Department of Microbiology of São Paulo University for technical help with the animals used in this study.
Espécies de Aeromonas sp, isoladas de peixe Pintado (Pseudoplatystoma sp): fatores de virulência e sensibilidade à drogas
Bactérias do gênero Aeromonas têm sido descritas como patógenos emergentes de importância crescente em alimentos. Neste estudo, relatamos que 48% das amostras de peixe "Pintado" coletado no comércio de São Paulo, foram positivas para Aeromonas sp quando isoladas pelo método de plaqueamento direto. Quando o método Presença/Ausência foi utilizado, a porcentagem de positividade foi de 42%. A. caviae foi a espécie mais freqüente, seguida por A. hydrophila e A. sobria. Produção de enterotoxina citotóxica, determinada em camundongos recém-nascidos, foi observada em 67% das cepas de A. sobria, em 60% das de A. hydrophila e em 40% das de A. caviae. No teste in vitro em células HEp- 88% das cepas de A. hydrophila, 27% das cepas de A. sobria e 13% das cepas de A. caviae revelaram-se positivas. Com relação a produção de enterotoxina citotônica, testada após o aquecimento do sobrenadante a 56ºC por 20 minutos, 17% das cepas de A. sobria, 10% das de A. caviae e nenhuma das de A. hydrophila foram positivas in vivo e para todas as cepas analisadas, os testes foram negativos em cultura de célula HEp-2. Quanto a capacidade de adesão, 20% das 5 cepas de A. sobria e 16% das 20 cepas de A. caviae aderiram a células HEp-2. A capacidade de invasão em células HEp-2 não foi detectada em nenhuma das cepas testadas. As cepas isoladas foram sensíveis a maior parte dos antimicrobianos testados.
Palavras-chave: Aeromonas sp, fatores de virulência, sensibilidade à drogas, peixe
REFERENCES 1. Abeyta, C.; Weagant, S.D.; Kaysner, C.A.; Ewkell, M.M.; Stott, R.F.; Krane, M.H.; Peeler, J.T. Aeromonas hydrophila in shellfish growing waters: incidence and media evaluation. J. Food Protect., 52: 7-12, 1898 [ Links ]
2. Atksin, H.M.; Trust, T.J. Haemagglutiotion properties and adherence ability of Aeromonas hydrophila. Infect. Immun., 27: 938-46, 1980. [ Links ]
3. Burke, V.; Robinson, J.; Gracey, M.; Peterson, D.; Patridge, K. Isolation of Aeromonas hydrophila from a metropolitan water supply: seasonal correlation with clinical isolates. App. Environ. Microbiol., 48: 361-6, 1984. [ Links ]
4. Burke, V.; Robinson, J.; Gracey, M. Enterotoxins of Aeromonas species. Experientia, 43: 369-91, 1987. [ Links ]
5. Callister, AJ.; Agger, W.A Enumeration and characterization of Aeromonas hydrophila and Aeromonas caviae isolated from grocery store produce. App. Environ. Microbiol., 96: 1-16, 1987 [ Links ]
6. Carrelo, A.; Silburn, K.A.; Budden, J.R.; Chang, B.J. Adhesion of clinical and environmental Aeromonas isolates to HEp-2 cells. J. Med. Microbiol., 26: 19-27, 1988. [ Links ]
7. Chopra, A, A.K.; Houston, C.W. Purification and partial characterization of a cytotonic enterotoxin produced by Aeromonas hydrophila. Can. J. Microbiol, 35: 719-27, 1989. [ Links ]
8. Gray, S.J.; Stickler, D.J.; Bryant, T.N. The incidence of virulence factors in mesophila Aeromonas species isolated from animals and their environment. Epidemiol. Infect., 105: 277-94, 1990. [ Links ]
9. Grey, P.A.; Kirov, M. Adherence to HEp-2 cells and enteropathogenic potential of Aeromonas spp. Epidemiol. Infect., 110: 279-87, 1993. [ Links ]
10. Gosling, P.J. Biochemical characteristics, enterotoxigenicity, susceptibility to antimicrobial agents of clinical isolates of Aeromonas species encountered in the western region of Saudi Arabia. J. Med. Microbiol., 22: 51-5, 1986. [ Links ]
11. Hedges, R.W.; Medeiros, A.A.; Cohenford, M.; Jacoby, G.A. Genetic and biochemical properties of AER-1 a novel carbenicillin hydrolyzing beta -lactamase from Aeromonas hydrophila. Antimicrob. Agents Chemother., 27:479-84, 1985. [ Links ]
12. Houston, C.W.; Chopra, A K., Rose, J.M.; Kurosky, A. Review of Aeromonas enterotoxin. Experientia, 47: 424-6, 1991. [ Links ]
13. Janda, J.M.; Reitano, M.; Bottone, E.J. Biotyping of Aeromonas isolates as a correlate delineating a species- associated disease spectrum. J. Clin. Microbiol., 19:44- 7, 1984. [ Links ]
14. Keusch, G.T.; Donta, S.T. Classification of enterotoxins on the basis of activity in cell culture. J. Infect. Dis., 131: 58-63, 1975. [ Links ]
15. Kirov, S.M.; Anderson, M.J.; MCMeekin, T.A. A note on Aeromonas spp from chickens as possible food-borne pathogens. J. Applied Bacteriol, 68: 327-34, 1990. [ Links ]
16. Kirov, S.M.; Hui, D.S.; Hayward, J. Milk as a potential source of Aeromonas gastrointestinal infection, J. Food Protection, 56: 306-12, 1993. [ Links ]
17. Kirov, S.M.; Hudson, J.A; Hayward, J; Mott, S.J. Distribution of Aeromonas hydrophila hybridization groups and their virulence properties in Australasian clinical and environmental strains. Lett. Appl. Microbiol., 18: 71-3, 1994. [ Links ]
18. Lawson, M.A.; Burke, V.; Chang, B. Invasion of Hep-2 cells by fecal isolates of Aeromonas hydrophila. Infect. Immun., 47: 680-3, 1985. [ Links ]
19. Majeed, K.N.; Egan, A F.; Mac Era, I.C. Enterotoxigenic aeromonads on retail lamb meat and offal. J. Appl. Bacteriol, 67: 165-70, 1989. [ Links ]
20. Mascher, F.; Reinthler, F.F.; Steinzner, D.; Lamberger, B. Aeromonas species in a municipal water supply of a central European city: biotyping of strains and detection toxins. Zbl. Bakt. Hyg. B, 186: 333-7, 1988. [ Links ]
21. Macraken, A.W.; Barkley, R. Isolation of Aeromonas species from clinical sources. J. Clin. Pathol., 25: 970, 1972. [ Links ]
22. Palumbo, S.A.; Marino, C.; Willians, A.C.; Buchanan, R.L.; Thaõer,D.W. Starch ampicillin agar for the quantitative detection of Aeromonas hydrophila. App. Environ. Microbiol., Oct 50: 1027-30, 1985. [ Links ]
23. Popoff, M. in Bergeys Manual of Systematic Bacteriology., ed. 8, vol I, 545-8. Baltimore, Williams and Wilkins., 1984. [ Links ]
24. Potomski, J.; Burke, V.; Robinson, J.R.; Fuarola, D.; Miragliotta, G. Aeromonas cytotonic enterotoxin cross reactive with cholera toxin. J. Med. Microbiol., 23:179-86, 1987. [ Links ]
25. Rose, J.M.; Houston, C.W.; Coppenhaver, D.H.; Dixon, J.D.; Kurosky, A.Purification and chemical characterization of a cholera toxin cross reactive cytolytic enterotoxin produced by a human isolate of Aeromonas hydrophila. Infect Immun., 57: 1176-9, 1989.[ Links ]26. Seidler, R.J.; Allen, D.A.; Lockman, H.; Colwell, R.R.; Joseph, S.W.; Daily, O.P. Isolation, enumeration and characterization of Aeromonas from polluted waters encountered in diving operations. Appl. Environ. Microbiol., 39: 1010-8, 1980. [ Links ]
27. Watson, I.M., Robinson, J.O.; Burke, V.; Gravey, M. Invasiveness of Aeromonas spp in relation to biotype, virulence factors and clinical features. J. Clin. Microbiol., 22: 48-51, 1985.[ Links ]
* Corresponding author. Mailing address: Departamento de Microbiologia, Instituto de Ciências Biomédicas, Universidade de São Paulo, Av. Prof. Lineu Prestes, 1374, Cidade Universitária, CEP 05508-900, São Paulo, SP, Brasil. Fax: (+5511) 818-7234