Jpn. J. Infect. Dis., 60, 217-219, 2007

Short Communication Isolation and Drug-Resistant Patterns of Campylobacter Strains Cultured from Diarrheic Children in Tehran Mohammad Mehdi Feizabadi*, Samaneh Dolatabadi** and Mohammad Reza Zali1 Department of Microbiology, School of Medicine, Tehran University, and 1Research Center for and , Shaheed Beheshti University of Medical Sciences, Tehran, Iran (Received December 22, 2006. Accepted April 5, 2007)

SUMMARY: To detect and determine the drug susceptibility of causative organisms, we acquired 500 diarrheic samples in Cary-Blair transfer medium from two pediatric hospitals in Tehran between October 2004 and October 2005. The samples were also enriched in Preston broth (with supplements) and defibrinated sheep (7%). They were plated from both media on Brucella agar containing and blood. Isolates were identified through biochemical tests and by the polymerase chain reaction method. Drug susceptibility testing was performed using the disk diffusion method. In total, 40 Campylobacter strains were isolated (8%). C. jejuni was the dominant species (85.8%) followed by C. coli (14.2%). The rates of resistance to antimicrobial agents were as follows: ciprofloxacin (61.7%), ceftazidime (47%), carbenicillin (35%), tetracycline (20.5%), cefotaxime (14.7%), ampicillin (11.7%), neomycin, erythromycin and chloramphenicol (2.9%), gentamicin, streptomycin, imipenem and colistin (0%). Campylobacter is an important cause of among Iranian children. The detection of Campylobacter increases by 25% if samples are treated in enrichment broth prior to plating. The high rate of resistance to ciprofloxacin is alarming, and further investigation into the possible reasons for this is imperative.

Campylobacter is a leading cause of acute bacterial gastro- strains isolated from children in Tehran. Moreover, polymerase intestinal infection worldwide (1). The genus Campylobacter chain reaction (PCR) was used to confirm the identification consists of at least 20 species including C. jejuni and C. coli, of the organism involved at the species level. This is the first which are the major causes of diarrhea in Europe, the United study of the species identification and profiling of resistance States and other industrialized countries (2). Importantly, patterns within the clinical isolates of Campylobacter cul- Campylobacter infections are hyper-endemic among infant tured from diarrheic children in Iran. populations in developing countries. The syndromes asso- To isolate Campylobacter, diarrheic stool samples (n = 500) ciated with C. jejuni infection range from mild to were collected from two pediatric hospitals in Tehran and severe invasive disease. In addition, sequelae can occur, includ- transported to the laboratory in Cary-Blair transport medium ing the autoimmune-mediated demyelinating neuropathies between October 2004 and October 2005. Samples were Guillain-Barré and Miller Fisher syndromes (3,4). Therefore, also enriched in Preston enrichment broth (Himedia, Mumbai, it is necessary to determine the susceptibilities of India) supplemented with polymyxin B (2,500 IU/L), rifampi- these enteropathogens in order to pinpoint the best mode of cin (5.0 mg/L), trimethoprim lactate (5.0 mg/L), amphotericin therapy. B (5.0 mg/L) and 7% defibrinated sheep blood for 24 h at Strains of C. jejuni and C. coli are intrinsically resistant to 42°C in microaerophilic condition (2,8). Patients had not been penicillin G and cephalosporines. Macrolides, in particular treated with antibiotics before the specimens were collected. erythromycin, are the usual drugs of choice (1,5). However, All samples were received within 2 h after being collected. for the empiric treatment of adults with suspected bacte- Samples received in Cary-Blair medium as well as those rial , the preferred drug typically includes a enriched in Preston broth were plated on Brucella agar fluoroquinolone (ciprofloxacin) because of its broad scope containing defibrinated sheep blood (7%) and the same anti- of efficacy against almost all enteric bacterial (6). biotics as added to the enrichment broth (9-11). The plates Campylobacter spp. are asaccharolytic, fastidious bacteria. were incubated at 42°C for 2 to 3 days under microaerophilic Their identification at the species level is limited and un- conditions. reliable; hence they are identified only at the genus level (7). Typical colonies were subcultured on Brucella agar. One Campylobacter spp. are frequently isolated from livestock, isolate from each suspected colony was identified at the meat and chicken throughout Iran. However, little informa- species level by conventional bacteriological tests (12,13). tion is available on the prevalence of Campylobacter as an To confirm the identification of isolated organisms at the etiologic agent of diarrhea among children in the country. species level, DNA was extracted by phenol-chloroform (14) This study was conducted to determine the prevalence and the and subjected to PCR assay. The primers CCCJ609f (5´-AAT antibiotic resistance patterns of thermophilic Campylobacter CTA ATG GCT TAA CCA TTA-3´) and CCCJ1442R (5´-GTA ACTAGT TTA GTA TTC CGG-3´) were used to amplify the *Corresponding author: Mailing address: Department of Micro- 16S rRNA gene (15). This area is conserved for both species biology, School of Medicine, Tehran University, Tehran, Iran. of C. jejuni and C. coli. The second pair of primers (5´-GAA Tel & Fax: +98-21-88955810, E-mail: [email protected] GAG GGT TTG GGTGGT G-3´ and 5´-AGC TAG CTT CGC **Present address: Islamic Azad University, Neishaboor Branch, ATA ATA ACT TG-3´) targeted the hippuricase gene (16), Neishaboor, Iran. and the third (5´-GGT ATG ATT TCT ACA AAGCGA G-3´

217 and 5´-ATA AAA GAC TAT CGTCGC GTG-3´) was based species-specific primers for hippuricase and aspartokinase on the sequence of the aspartokinase gene from C. coli (15). genes, PCR identified 85.8% of the isolates as C. jejuni and C. jejuni ATCC27853 was used as a control. The PCR condi- 14.2% as C. coli. This shows the power of PCR for differ- tions and cycling program were the same as described by entiating C. coli from C. jejuni and the advantage of PCR others (15,16). over biochemical tests. However, we did not observe any Of 500 diarrheic samples, 40 (8%) were found to be posi- differences between C. jejuni and C. coli in terms of clinical tive in culture, and these were either recovered by plat- symptoms such as diarrhea, blood in stool, abdominal pain, ing from Cary-Blair medium (n = 26, 64%) or plating from fever, vomiting, mean duration of illness or admission to hos- Preston enrichment broth (n = 10, 25.6%). Four samples were pitals. found to be positive with both methods. Most isolates were To determine the susceptibility of isolates to antimicrobial recovered in late summer (n = 11) and autumn (n = 23). The agents, isolates were inoculated on Mueller-Hinton agar remaining isolates were cultured in spring (n = 5). Only one with 5% sheep blood. Disks containing ampicillin (25 μg), isolate was detected by culture in the winter season. chloramphenicol (30 μg), ciprofloxacin (5 μg), colistin (20 Failure to diagnose acute capmpylobacteriosis in diarrheic μg), nalidixic acid (3 μg), gentamicin (10 mg), erythromycin cases may be attributed to the low sensitivity of the existing (15 μg), imipenem (20 μg) ofloxacin (5 μg), streptomycin routine methods. The inclusion of an enrichment protocol (100 μg), tetracycline (30 μg) cephalotin (30 μg), cepha- resulted in a 25% increase in the detection of Campylobacter lexin (30 μg), ceftazidime (30 μg), carbenicillin (100 μg), in comparison with transportation in Cary-Blair medium. cefotaxime (10 μg), and neomycin (30 μg) (Oxoid, Hamp- Based on this finding, we now inoculate the diarrheic samples shire, UK) were placed on the inoculated plates. The plates directly into the enrichment medium and subsequently culture were then incubated in a microaerophilic atmosphere at 42°C them in Brucella agar rather than transporting them in Cary- for 48 h. Susceptibility categorization was carried out accord- Blair medium. ing to the recommendations of the National Committee The age group with the highest rate of diarrhea was the for Clinical Laboratory Standards (20). β-lactamase activ- group including children aged <1 year (n = 228). The other ity was assessed using nitrocefin as a substrate (Oxoid). age groups were as follows: 1-2 (n = 85), 2-3 (n = 46), 3-4 ATCC29213, ATCC (n = 40), 4-5 (n = 22), 5-6 (n = 10), 6-7 (n = 14), 7-8 (n = 25922, and Pseudomonas aeruginosa ATCC 27853 were used 12), 8-12 (n = 37), 12-13 (n = 4), and 13-14 years (n = 2). as controls. The numbers of positive cultures for each group were 16, 7, The results of the susceptibility testing for the C. jejuni 5, 2, 3, 0, 3, 2, 0, 1, and 1, respectively. The rate of positive and C. coli strains are shown in Table 1. All Campylobacter cultures for Campylobacter was higher for children aged isolates in this study were susceptible to colistin, imipenem, group 7-8 years (25%), followed by those aged 6-7 (21.4%), getamicin and streptomycin. All C. coli isolates were resist- 2-3 (15.2%), and 4-5 years (13.6%). None of the specimens ant to cephalexin and nalidixic acid. Cross-resistance between from diarrheic children aged 5-6 or 8-12 years were posi- nalidixic acid and ciprofloxacin was found in all quinolone- tive for Campylobacter in culture. resistant isolates. Production of β-lactamase was found in Of the 40 isolates, 19 were cultured from male children three isolates. and 21 from females. Five isolates did not grow in subsequent The rate of Campylobacter isolation from diarrheic chil- culture. Using biochemical tests, the remaining isolates (n = dren (8%) found in the present study was higher than that 35) were identified as C. jejuni (n = 26, 74.3%) and C. coli found in Saudi Arabia (4.5%) (21). This result shows the (n = 9, 25.7%). However, 4 isolates that failed to hydrolyze importance of these organisms as an etiologic agent of gastro- hippurate were positive for the hip gene but negative for the intestinal disease in Tehran. aspartokinase gene in PCR aasay. Therefore, PCR identified 30 (85.8%) and 5 (14.2%) of the isolates as C. jejuni and C. Table 1. Rates of resistance to different antibiotics for isolates of coli, respectively. C. jejuni (n = 29) and C. coli (n = 5) cultured from diarrheic In one study, C. jejuni was isolated in 99% of the children at two hospitals of Tehran during a year ended October Campylobacter infections in humans (17). However, Van 2005 Looveren et al. isolated C. jejuni and C. coli isolates in 79 C. jejuni C. coli Total and 21% cases of gastroenteritis, respectively (18). Eyigor et No. (%) No. (%) No. (%) al. have reported the prevalence of C. coli in such cases to be Cephalotin 28 (96.5) 5 (100) 33 (97) 33% (19). Based on the results of biochemical tests, 74.3% of Cephalexin 27 (93) 5 (100) 32 (94) the strains belong to C. jejuni and 25.7% to C. coli. In current Ofloxacin 24 (82.7) 4 (80) 28 (87.5) laboratory practices, the identification of Campylobacter at Nalidixic acid 22 (75.8) 5 (100) 27 (79.4) the species level relies on relatively few phenotypic tests. Ciprofloxacin 17 (58.6) 4 (80) 21 (61.7) For example, C. jejuni and C. coli are distinguished only by Ceftazidime 12 (41.3) 4 (80) 16 (47) hippurate hydrolysis, while C. coli and C. upsaliensis are Carbenicillin 9 (31) 3 (60) 12 (35) distinguished by the weak catalase activity and sensitivity to Tetracycline 7 (24.3) 0 7 (20.5) cephalotin of the latter species. Even when a rapid hippurate Cefotaxime 3 (10.3) 2 (40) 5 (14.7) hydrolysis phenotypic test is performed to identify C. jejuni Ampicillin 3 (10.3) 1 (20) 4 (11.7) isolates, significant difficulties remain in the identification Chloramphenicol 1 (3.4) 0 1 (2.9) of any hippurate-negative isolate since such isolates may be Neomycin 1 (3.4) 0 1 (2.9) hippurate-negative strains of C. jejuni or may belong to other Erythromycin 1 (3.4) 0 1 (2.9) species. Therefore, the PCR assay described in this study was Colistin 0 0 0 used for the rapid and definitive identification of C. jejuni Gentamicin 0 0 0 and C. coli. The first pair of primers produced an amplicon Imipenem 0 0 0 850 bp in size for both species of C. jejuni and C. coli. Using Streptomycin 0 0 0

218 Our study shows an alarmingly high rate of resistance to of six media, including a semisolid agar, for the isolation of various ciprofloxacin (61.7%), which is higher than that found in Campylobacter species from stool specimens. J. Clin. Microbiol., 29,1007-1010. Kuwait (53%) (22). This finding means that the majority of 11. Goossens, H., Pot, B., Vlaes, L., et al. (1990): Characterization and our patients probably suffer from side effects due to infection description of “Campylobacter upsaliensis” isolated from human with ciprofloxacin-resistant strains. The duration of diarrhea feces. J Clin. Microbiol., 28, 1039-1046. lasts twice as long in patients infected with these isolates 12. Engberg, J., On, S.L., Harrington, C.S., et al. (2000): Prevalence of Campylobacter, Arcobacter, Helicobacter, and Sutterella spp. in human than in those infected with the susceptible isolates (23). An fecal samples as estimated by a reevaluation of isolation methods for increase in the rate of resistance to ciprofloxacin, from 11% Campylobacters. J. Clin. Microbiol., 38, 286-291. in 1994 to 29% in 1997, has already been reported for 13. Aarestrup, F.M., Nielsen, E.M., Madsen, M., et al. (1997): Antimicrobial Campylobacter strains in the Netherlands (24). Ciprofloxacin- susceptibility patterns of thermophilic Campylobacter spp. from humans, resistant Campylobacters have recently been reported in pigs, cattle, and broilers in Denmark. Antimicrob. Agents Chemother., 41, 2244-2250. Canada, Spain, and the USA (25-27). C. jejuni and C. coli 14. Oyofo, B.A., Thornton, S.A., Burr, D.H., et al. (1992): Specific detection strains showing resistance to ciprofloxacin are frequently of and Campylobacter coli by using polymerase isolated from different animal sources, and they can be trans- chain reaction. J. Clin. Microbiol., 30, 10:2613-2619. ferred from animal products to humans. However, the reasons 15. Linton, D., Lawson, A.J., Owen, R.J., et al. (1997): PCR detection, identi- fication to species level, and fingerprinting of Campylobacter jejuni and for the high rate of resistance to ciprofloxacin among the Campylobacter coli direct from diarrheic samples. J. Clin. Microbiol., Campylobacter population in Iran require further investiga- 35, 2568-2572. tion. 16. Hani, E.K. and Chan V.L. (1995): Expression and characterization of Campylobacter jejuni benzoylglycine amidohydrolase (hippuricase) gene in Escherichia coli. J. Bacteriol., 77, 2396-2402. ACKNOWLEDGMENTS 17. Ng, L.K., Kingombe, C.I., Yan, W., et al. (1997): Specific detection and The authors thank Haudi Dehdashti for editing this manuscript. Thanks confirmation of Campylobacter jejuni by DNA hybridization and PCR. should be extended to S. Ardalan for her technical assistance. Appl. Environ. Microbiol., 63, 4558-4563. This project was partly supported by RCGLD. 18. van Looveren, M., Daube, G., de Zutter, L., et al. (2001): Antimicrobial susceptibilities of Campylobacter strains isolated from food animals in Belgium. J. Antimicrob. Chemother., 48, 235-240. REFERENCES 19. Eyigor, A., Dawson, K.A., Langlois, B.E., et al. (1999): Detection of 1. Allos, B.M. (2001): Campylobacter jejuni infections: update on emerg- cytolethal distending toxin activity and cdt genes in Campylobacter spp. ing issues and trends. Clin. Infect. Dis., 32, 1201-1216. isolated from chicken carcasses. Appl. Environ. Microbiol., 65, 1501- 2. Barros-Velazquez, J., Jimenez, A. and Villa, T.G. (1999): Isolation and 1505. typing methods for the epidemiologic investigation of thermotolerant 20. National Committee for Clinical Laboratory Standards (2002): Perfor- campylobacters. Int. Microbiol., 2, 217-226. mance Standards for Antimicrobial Susceptibility Testing: 12th Infor- 3. Dingle, K.E., Colles, F.M., Ure, R., et al. (2002): Molecular characteriza- mation Supplement. Wayne, Pa., USA. tion of Campylobacter jejuni clones: a basis for epidemiologic investiga- 21. Zaman, R. (1992): Campylobacter enteritis in Saudi Arabia. Epidemiol. tion. Emerg. Infect. Dis., 8, 949-955. Infect., 108, 51-58. 4. Nachamkin, I., Allos, B.M. and Ho, T. (1998): Campylobacter species 22. Albert, M.J., Neil, L., Pazhoor, A.A., et al. (2005): Ciprofloxacin resist- and Guillain-Barre syndrome. Clin. Microbiol. Rev., 11, 555-567. ance and its molecular mechanism in Campylobacter spp. isolated in 5. Tajada, P., Gomez-Graces, J.L., Alos, J.I., et al. (1996): Antimicrobial Kuwait. Microb. Drug Resist., 11, 266-270. susceptibilities of Campylobacter jejuni and Campylobacter coli to 12 23. Nelson, J.M., Smith, K.E., Vugia, D.J., et al. (2004): Prolonged diarrhea beta-lactam agents and combinations with beta-lactamase inhibitors. due to ciprofloxacin-resistant Campylobacter infection. J. Infect. Dis., Antimicrob. Agents Chemother., 40, 1924-1925. 190, 1150-1157. 6. Engberg, J., Neimann, J., Nielsen, E.M., et al. (2004): Quinolone-resistant 24. Talsma, E., Goettsch, W.G., Nieste, H.L., et al. (1999): Resistance in Campylobacter infections: risk factors and clinical consequences. Emerg. Campylobacter species: increased resistance to fluoroquinolones and Infect. Dis., 10, 1056-1063. seasonal variation. Clin. Infect. Dis., 29, 845-848. 7. Kulkarni, S.P., Lever, S., Logan, J.M., et al. (2002): Detection of 25. Gaudreau, C. and Gilbert, H. (1998): Antimicrobial resistance of clinical campylobacter species: a comparison of culture and polymerase chain strains of Campylobacter jejuni subsp. jejuni isolated from 1985 to 1997 reaction based methods. J. Clin. Pathol., 55, 749-753. in Quebec, Canada. Antimicrob. Agents Chemother., 42, 2106-2108. 8. Chuma, T., Ikeda, T., Maeda, T., et al. (2001): Antimicrobial suscep- 26. Prats, G., Mirelis, B., Llovet, T., et al. (2000): Antibiotic resistance trends tibilities of Campylobacter strains isolated from broilers in the southern in enteropathogenic bacteria isolated in 1985-1987 and 1995-1998 in part of Japan from 1995 to 1999. J. Vet. Med. Sci., 63, 1027-1029. Barcelona. Antimicrob. Agents Chemother., 44, 1140-1145. 9. Corry, J.E., Post, D.E., Colin, P., et al. (1995): Culture media for the 27. Smith, K.E., Besser, J.M., Hedberg, C.W., et al. (1999): Quinolone- isolation of Campylobacters. Int. J. Food. Microbiol., 26, 43-76. resistant Campylobacter jejuni infections in Minnesota, 1992 - 1998. 10. Endtz, H.P., Ruijs, G.J., Zwinderman, A.H., et al. (1991): Comparison Investigation Team. N. Engl. J. Med., 340, 1525-1532.

219