Biotype and antibiotic sensitivity of 100 clinical isolates of Yersinia enterocotitica

Transcription

1 Journal of Antimicrobial Chemotherapy (1991) 28, Biotype and antibiotic sensitivity of 100 clinical isolates of Yersinia enterocotitica Jeannette N. Pham, Sydney M. Bell* and Josette Y. M. Lanzarone Department of Microbiology, The Prince of Wales Hospital, Randwick, NSW 2031, Australia One hundred clinical isolates of Yersinia enterocolitica were biotyped, serotyped and tested for in-vitro sensitivity to seven /J-lactam antibiotics and six other antibiotics. Twenty-four of the isolates belonged to Nilehn's and Wauters' biotype 1A, 12 to biotype 3 and 64 to biotype 4. No biotypes IB or 2 strains were found. All isolates were sensitive to chloramphenicol, ciprofloxacin, gentamicin, tetracycline, trimethoprim and all but one to sulphafurazole. All were resistant to ampicillin with MICs ranging from 32 to 256 mg/l, but were susceptible to ticarcillin/clavulanic acid and imipenem. With other 0-lactam agents there were different patterns of sensitivity, spedfic to each of the three biotypes. All biotype 4 strains were resistant to carbenicillin and ticarcillin but sensitive to amoxycillin/clavulanic acid and cefoxitin. Biotype 3 strains, by contrast, were sensitive to carbenicillin and ticarcillin but resistant to amoxycillin/clavulanic acid and cefoxitin, whilst biotype 1A strains were resistant to each of these four /J-lactam agents. Introduction The recognition of Yersinia enterocolitica as a human pathogen (Bottone, 1977; Mollaret, Bercovier & Alonso, 1979) was soon followed by the successful isolation of this organism from clinical specimens in many countries including Australia. The availability of Yersinia selective media and the comprehensive identification and biotype schemes of Bercovier et al. (1980) and Wauters, Kandolo & Janssens (1987) also contributed to the increase in the number of Y. enterocolitica isolated from stool specimens. The results of in-vitro antimicrobial susceptibility testing of Y. enterocolitica have been reported in several studies (Hausnerova,, Hausner & Pauckova, 1973; Hammerberg, Sorger & Marks, 1977; Raevuori et al., 1978; Gaspar & Soriano, 1981; Juhlin & Winblad, 1981; Scribner et al., 1982; Soriano & Vega, 1982; Ahmedy et al., 1985; Hornstein et al., 1985; Kanazawa, Ikemura & Kuramata, 1987). Some authors suggested a certain relationship between antibiotic sensitivity and serotype or biotype of Y. enterocolitica. Others have found little difference in susceptibility to antibiotics in biochemically diverse strains within the species. Hausnerova et al. (1973) in a study of 212 biotype 4 strains and 64 biotype 1 strains, found that the susceptibility to chloramphenicol, tetracycline, neomycin and colistin exhibited by strains of the two biotypes was similar. However biotype 4 strains as a group, were relatively less sensitive to the aminoglycosides than biotype 1 strains and were more sensitive to carbenicillin 'Corresponding author /91/ $02.00/ The British Society for Antimicrobial Chemotherapy

2 14 J. N. Pluun et al and cephaloridine. Ahmedy et al. (1985) also noted that biotype 1 strains of Y. enterocolitica were resistant to ampicillin and carbenicillin whilst strains which belonged to biotype 3 were sensitive. Similarly Hornstein et al. (1985) observed clusters in the antibiotic susceptibility patterns corresponding to major serotypes. On the other hand, Raevuori et al. (1978) found that 25% of 67 indole-negative strains and 26% of 123 indole-positive strains were susceptible to ampicillin and they concluded that the biochemically diverse strains of Y. enterocolitica possessed relatively uniform susceptibility patterns. These authors also made reference to the results of Toma & Lafleur (1974) who reported that none of their indole-negative serotype 0:3 strains, which constituted 78% of their 278 isolates, was susceptible to ampicillin. One likely explanation for the lack of agreement concerning the distribution of antibiotic sensitivities into well denned groups was the absence of a uniform categorization of members of this species. Some reports relied on single biochemical differences whilst others used a serological techique for further classification within the species. In the present study of biotyping and antibiotic susceptibility testing of Y. enterocolitica, we paid special attention to precision in the determination of biochemical types and adhered to the well documented biotyping schemes of Bercovier et al. (1980) and Wauters et al. (1987). We also examined, in particular, the relationship between the sensitivity to several /Mactam antibiotics and the biotype of 100 clinical isolates of Y. enterocolitica. Bacterial strains and culture conditions Materials and methods One hundred distinct isolates of Y. enterocolitica, cultured from patients attending a group of hospitals in the metropolitan area of Sydney, were used in this study. All isolates were obtained from patients with clinical conditions consistent with yersinia infections. Seven isolates were from blood cultures (three of biotype 3 and four of biotype 4), one isolate (biotype 3) was cultured from a lymph node and the remaining 92 isolates were from stool specimens of patients presenting with gastrointestinal symptoms. Forty-three isolates were from children under the age of six and 72 isolates were from patients under the age of 16. Cultures were incubated at 28 C in air in all experiments since, at this temperature, Y. enterocolitica strains are normally prototrophic (Bercovier et al., 1979). Twenty-hour cultures of each isolate grown on horse blood agar were used in biotyping, serotyping and antibiotic sensitivity testing. Escherichia coli NCTC and E. coli NCTC were the control organisms in all antibiotic susceptibility tests. Antibiotics and chemicals Clavulanic acid, amoxycillin, carbenicillin and ticarciuin (Beecham Research Laboratories), ciprofloxacin (Bayer), cefoxitin and imipenem (Merck, Sharp & Dohme), chloramphenicol (Parke Davis) and tetracycline (Cyanamid) were kindly supplied by the manufacturers or their agents. Nitrocefin was purchased from Oxoid Laboratories, ampicillin from Commonwealth Serum Laboratories, gentamicin from David Bull Laboratories, sulphafurazole from Roche Laboratories and trimethoprim from Wellcome Co. All working solutions were prepared fresh before use.

3 Biotypcs of Y. emterocoetica 15 Biochemical identification and biotyping Each of the 100 isolates was identified and biotyped by the methods described by Nil6hn (1969), reviewed by Bercovier et al. (1980) and Wauters el al. (1987). The carbohydrate fermentation tests were performed in 3 ml of peptone water in test tubes using Andradc's indicator (Wilkinson, 1965). The results were read at 24 and 48 h incubation at 28 C. Positive tests gave a dark pink colour and negative tests were colourless after 48 h. The esculin hydrolysis test was done using Bacto Bile Esculin agar (Difco) slopes. The slopes were incubated at 28 C for up to 48 h. Positive hydrolysis of esculin showed a definite black colour of the slant and in at least half of the butt after 24 h of incubation. A negative test presented no blackening of the medium after 48 h incubation. Pyrazinamidase and /f-d-glucosidase activities were determined by the methods described by Kandolo & Wauters (1985) and Wauters et al. (1987). Serotyping Serotyping was performed by slide agglutination of live cultures grown overnight on horse blood agar on all the 100 isolates using Y. enterocolitica O Antisera kits (Denka Scikcn Co Ltd, Japan) which contained antisera 0: 1 and O : 2 (mixed), 0: 3, 0: 5, O : 8 and O : 9. Biotype 3 strains were tested, in addition, with O : 5, 27 antiscrum kindly supplied by Dr K. Bettelheim of the Australian Yersinia Reference Laboratory. Quantitative antibiotic susceptibility testing The minimum inhibitory concentration (MIC) of each antibiotic was performed by an agar dilution technique which conformed with the recommendations of the International Collaborative Study on Antibiotic Sensitivity Testing (Ericsson & Sherds, 1971). A 24 h culture from each isolate grown on horse blood agar and incubated at 28 C C was used to prepare a suspension in normal saline with an absorbance of 0-78 at 640 nm. This suspension was diluted 1/100 in saline to produce the final inoculum of 1 x 10 7 organisms per ml which yielded lo^cfu per replicator drop of 4 nl. The suspensions were then inoculated onto agar plates containing serial two-fold dilutions of antibiotics as described previously (Bell, Pham & Lanzarone, 1985). The plates were read after 24 h incubation at 28 C in air. Biotyping Results Of the 100 clinical isolates, 64 strains belonged to biotype 4 of Nilehn and Wauters. That is they produced DNA'se but not lipase or indole, reduced nitrate to nitrite, fermented sucrose and D-trehalose but not D-xylosc, did not hydrolyse esculin or ferment salicin and did not show pyrazinamidase or 0-D-glucosidase activity. Twenty-four isolates were biotype 1A and produced lipase, pyrazinamidase, /?-D-glucosidase and indole but not DNA'se; fermented salicin, D-xylose, sucrose and D-trehalose, hydrolysed esculin and reduced nitrate to nitrite. Only 12 belonged to biotype 3; they did not produce lipase or indole, showed weak or no DNA'se activity, fermented 4-xylose, sucrose and D-trehalose, reduced nitrate to nitrite, did not hydrolyse esculin or ferment salicin; all 100 isolates were Voges-Proskauer positive. Neither biotype IB nor biotype 2 was found amongst the 100 isolates.

4 16 J. N. Pham et al Table. Susceptibility of 100 isolates of Y. enterocolitica to /Mactam agents (range of MIC in mg/l) Antibiotic Biotype 1A («= 24) Biotype 3 (n = 12) Biotype 4 (n»64) Serotyping Amoxycillin Amoxycillin/clavulanate* Carbenicillin Cefoxitin Imipenem Ticarcillin Ticarcillin/clavulanatc* 'Clavulanic acid 2 mg/l O All 64 isolates of biotype 4 agglutinated with the antiserum O : 3 and all 12 isolates of biotype 3 agglutinated with antisera O : 5 and O : 5, 27. Seventeen out of 24 biotype 1A isolates agglutinated with either 0: 5 or 0: 8 antisera or with both antisera and the remainder failed to agglutinate with any of the somatic antisera 0 : 1 & 0: 2 (mixed), O : 3, O : 5, O : 8, O : 9. Antibiotic susceptibility Antibiotics other than fi-lactams. AH 100 strains were sensitive to five of the antibiotics tested. The range of MIC in mg/l of each antibiotic is shown in parenthesis after each drug; chloramphenicol (4-0-80), ciprofloxacin ( ), gentamicin ( ), tetracycline () and trimethoprim (0-5-). Ninety-one isolates were sensitive to sulphafurazole (32-64) whilst only one strain was resistant to sulphafurazole with an MIC of 256 mg/l. P-Lactam antibiotics The MIC of the seven /Mactam antibiotics is shown in the Table. All isolates were resistant to ampicillin and the MICs ranged from 32 to 512 mg/l. All 64 strains of biotype 4 showed a consistently high susceptibility to amoxycillin/clavulanate with an MIC of 0-5 mg/l amoxycillin in the presence of clavulanic acid at the concentration of 2 mg/ml, whilst all strains of biotype 1A and biotype 3 were resistant to this combination. All biotype 4 strains were also highly sensitive to cefoxitin whilst strains of biotypes 1A and 3 were resistant to this /Mactam. In contrast to biotypes 1A and 4 strains, all 12 strains belonging to biotype 3 were sensitive to carbencillin and ticarcillin. All isolates showed in-vitro sensitivity to imipenem and ticarcillin/clavulanate. Discussion The biotyping and serotyping results observed with the 100 clinical isolates of Y. enterocolitica in the present study were similar to those previously reported from

5 Biotype* of Y. enterocobtica 17 Australia by Marriott (1987). In that study, of the 38 isolates of Y. enterocolitica cultured from clinical specimens from children, 19 were biotyped and serotyped. Fourteen isolates were biotype 4, serotype 0: 3. The remainder were biotype 1 with different serotypes, including serotype O : 5, 27. The absence of biotype 3 in Marriott's study probably could be explained by the low prevalence of this particular biotype and the use of a relatively small sample. We also found that biotype 4, serotype O : 3, was the predominant biotype in clinical specimens in Australia. A similar predominance of this biotype has been reported elsewhere in the world, including the United States where in some states, biotype 4, serotype O : 3 replaced biotype 1B, serotype O : 8, as the most common biotype of Y. enterocolitica encountered after 1978 (Bottone, Gullans & Sierra, 1987). The 100 isolates in our series were uniformly susceptible to chloramphenicol, ciprofloxacin, gentamicin, tetracycline and trimethoprim and only one strain was found to be resistant to sulphafurazole. Our findings with these antibiotics agree with published data on sensitivity patterns in other parts of the world (Hausnerova et al., 1973; Hammerberg et al., 1977; Raevuori et al., 1978; Soriano & Vega, 1982; Kanazawa et al., 1987). Thus, at this stage, Y. enterocolitica does not appear to have acquired the multiple resistance to antibiotics which is often seen within other Enterobacteriaceae. With the /Mactam agents, we observed sensitivity patterns which appeared to be biotype specific. All isolates were resistant to ampicillin but sensitive to ticarcillin/ clavulanate and imipenem. However only biotype 4 strains were also sensitive to cefoxitin and amoxycillin/clavulanate. On the other hand biotypes 1A and 3 were resistant to both these antibiotics. Sensitivity to ticarcillin further distinguished the biotypes. Whereas all strains of biotype 3 were sensitive to ticarcillin, strains belonging to biotypes 1A and 4 were resistant. The biotype specificity of the sensitivity patterns to the /Mactams which was demonstrated in the present study further elucidates the apparent clustering of antibiotic sensitivities corresponding to the major serotypes reported by Hornstein et al. (1985). The differential sensitivity to the members of the /Mactam class of antibiotics which we observed in the three biotypes of Y. enterocolitica can be explained on the basis of the production of different /Mactamases by members of each biotype. Y. enterocolitica, as a species, was shown previously to produce more than one type of /Mactamase (Cornells, 1975; Cornells & Abraham, 1975). In the present study, the differences in distribution of these enzymes within the three biotypes were demonstrated not only by the grouping of /Mactam sensitivity but also by the different responses in each biotype to the /Mactams when those were combined with clavulanic acid, a powerful /Mactamase inhibitor. References Ahmedy, A., Vidon, D. J., Dclmas, C. L. & Lett, M. C. (1985). Antimicrobial susceptibilities of food-isolated strains of Yersmia enterocolitica, Y. intermedia, Y. frederiksenii, and Y. kristensenii. Antimicrobial Agents and Chemotherapy 28, Bell, S. M., Pham, J. N. & Lanzarone, J. Y. M. (1985). Mutation of Pseudomonas aeruginosa to piperaciuin resistance mediated by /J-lactamase production. Journal of Antimicrobial Chemotherapy 15, Bercovier, H., Alonso, J. M., Bentaiba, Z. N., Brault, J. & Mollaret, H. H. (1979). Contribution to the definition and the taxonomy of Yersinia enterocolitica. Contributions to Microbiology and Immunology 5, 1V22.

6 18 J. N. Pham et ol Bcrcovier, H., Brenner, D. J., Urging, J., Steigerwalt, A. G., Fanning, R. G., Alonso, J. M. et al. (1980). Characterization of Yersinia enterocolitica sensu stricto. Current Microbiology 4, Bottone, E. J. (1977). Yersinia enterocolitica: a panoramic view of a charismatic organism. CRC Critical Reviews in Microbiology 5, Bottone, E. J., Gullans, C. R. & Sierra, M. F. (1987). Disease spectrum of Yersinia enterocolitica serogroup O : 3, the predominant cause of human infection in New York City. Contributions to Microbiology and Immunology 9, Cornelis, G. (1975). Distribution of /Mactamases A and B in some groups of Yersinia enterocolitica and their role in resistance. Journal of General Microbiology 91, Cornelis, G. & Abraham, E. P. (1975). /J-Lactamases from Yersinia enterocolitica. Journal of General Microbiology 87, Ericsson, H. M. & Shcrris, J. C. (1971). Antibiotic sensitivity testing. Report of an international collaborative study. Acta Pathologica et Microbiologica Scandinavica, Suppl. 217, Gaspar, M. C. & Soriano, F. (1981). Susceptibility of Yersinia enterocolitica to eight 0-lactam antibiotics and clavulanic acid. Journal of Antimicrobial Chemotherapy 8, Hammerberg, S., Sorger, S. & Marks, M. I. (1977). Antimicrobial susceptibilities of Yersinia enterocolitica biotype 4, serotype O : 3. Antimicrobial Agents and Chemotherapy 11, Hausnerova, S., Hausner, O. & Pauckova, V. (1973). Antibiotic sensitivity of Yersinia enterocolitica strains isolated in two regions of Czechoslovakia. Contributions to Microbiology and Immunology 2, Hornstein, M. J., Jupeau, A. M., Scavizzi, M. R., Philippon, A. M. & Grimont, P. A. (1985). In vitro susceptibilities of 126 clinical isolates of Yersinia enterocolitica to 21 /Mactam antibiotics. Antimicrobial Agents and Chemotherapy Tl, Juhlin, I. & Winblad, S. (1981). Susceptibility to mecillinam and other antibiotics of 28 O-serotypes of Yersinia enterocolitica. Journal of Antimicrobial Chemotherapy 8, Kanazawa, Y. Ikemura, K. & Kuramata, T. (1987). Drug susceptibility of Yersinia enterocolitica and Yersinia pseudotuberculosis. Contributions to Microbiology and Immunology 9, Kandolo, K. & Wautcrs, G. (1985). Pyrazinamidase activity in Yersinia enterocolitica and related organisms. Journal of Clinical Microbiology 6, Marriott, D. (1987). Yersinia enterocolitica infection in children in New South Wales. Contributions to Microbiology and Immunology 9, Mollaret, H. H., Bercovier, H. & Alonso, J. M. (1979). Summary of the data received at the WHO Reference Center for Yersinia enterocolitica. Contributions to Microbiology and Immunology 5, Nilehn, B. (1969). Studies on Yersinia enterocolitica with special reference to bacterial diagnosis and occurrence in human acute enteric disease. Acta Pathologica et Microbiologica Scandinavica, Suppl. 206, Raevuori, M., Harvey, S. M., Pickett, M. J. & Martin, W. J. (1978). Yersinia enterocolitica: in vitro antimicrobial susceptibility. Antimicrobial Agents and Chemotherapy 13, Scribner, R. K., Marks, M. I., Weber, A. & Pai, C. H. (1982). Yersinia enterocolitica:. comparative in vitro activities of seven new /Mactam antibiotics. Antimicrobial Agents and Chemotherapy 22, Soriano, F. & Vega, J. (1982). The susceptibility of Yersinia to eleven antimicrobials. Journal of Antimicrobial Chemotherapy 10, Toma, S. & Lafleur, L. (1974). Survey on the incidence of Yersinia enterocolitica infection in Canada. Applied Microbiology 28, Wauters, G., Kandolo, K. & Janssens, M. (1987). Revised biogrouping scheme of Yersinia enterocolitica. Contributions to Microbiology and Immunology 9, Wilkinson, J. F. (1965). Tests employed in bacterial identification. In Medical Microbiology, 11th edn (Cruickshank, R., Ed.), pp E. & S. Livingstone, Edinburgh. (Received 19 November 1990; revised version accepted 12 March 1991)