J Infect Chemother 20 (2014) 243e249

Contents lists available at ScienceDirect

Journal of Infection and Chemotherapy journal homepage: http://www.elsevier.com/locate/jic

Original article

Comparison of broad-spectrum cephalosporin-resistant Escherichia coli isolated from dogs and humans in Hokkaido, Japan Torahiko Okubo a, Toyotaka Sato a, Shin-ichi Yokota b, Masaru Usui a, Yutaka Tamura a, * a Laboratory of Food Microbiology and Food Safety, Department of Health and Environmental Sciences, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, Hokkaido, Japan b Department of Microbiology, Sapporo Medical University School of Medicine, Sapporo, Hokkaido, Japan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 September 2013 Received in revised form 29 November 2013 Accepted 1 December 2013

Resistance to broad-spectrum cephalosporins (BSCs) in Enterobacteriaceae in companion animals has become a great concern for public health. To estimate the dissemination of BSC-resistant bacteria between dog and human, we examined the BSC-resistance determinants of and genetic similarities between 69 BSC-resistant Escherichia coli isolates derived from canine rectal swabs (n ¼ 28) and human clinical samples (n ¼ 41). Some E. coli isolates possessed blaTEM-1b (14 canine and 16 human isolates), blaCTx-M-2 (6 human isolates), blaCTx-M-14 (3 canine and 14 human isolates), blaCTx-M-27 (1 canine and 15 human isolates), and blaCMY-2 (11 canine and 3 human isolates). The possession of CTX-M-type b-lactamases was significantly more frequent in human isolates, whereas CMY-2 was more common in canine isolates. Bacterial typing methods (phylogenetic typing, O-antigen serotyping, and pulsed-field gel electrophoresis) showed little clonal relationship between canine isolates and human isolates. Plasmid analysis and Southern blotting indicated that the plasmids encoding CMY-2 were similar among canine and human isolates. Based on the differences in the major b-lactamase and the divergence of bacterial types between canine and human isolates, it seems that clonal dissemination of BSC-resistant E. coli between canines and humans is limited. The similarity of the CMY-2-encoding plasmid suggests that plasmid-mediated b-lactamase gene transmission plays a role in interspecies diffusion of BSC-resistant E. coli between dog and human. Ó 2013, Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.

Keywords: Escherichia coli Antimicrobial resistance Extended spectrum b-lactamase AmpC b-lactamase Dog

1. Introduction

b-lactam antibiotics, including broad-spectrum cephalosporins (BSCs), are widely used for treatment of bacterial infections in both veterinary and human medicine. In accordance with the increased usage of BSCs, cephalosporin-resistant Gram-negative bacteria have increasingly been isolated from animals and humans throughout the world [1]. The predominant cause of resistance to BSCs in Enterobacteriaceae is the production of b-lactamases, a group of enzymes rendering b-lactam antibiotics harmless to bacteria [2]. Among human clinical isolates, the production of extended-spectrum blactamases (ESBLs), such as TEM, SHV, and CTX-M-type enzymes * Corresponding author. 582 Midorimachi-Bunkyoudai, Ebetsu, Hokkaido 0698501, Japan. Tel./fax: þ81 11 388 4890. E-mail address: [email protected] (Y. Tamura).

were major BSC-resistant determinants [3]. Recent studies showed the international dissemination of a specific Escherichia coli clone, which produces the CTX-M-15 enzyme, and belonging to phylogenetic group B2, serogroup O25:H4, and sequence type 131 (ST131), among human communities [4]. Although this E. coli clone was recently detected in Japan, an O25:H4-ST131 clone that does not produce CTX-M-15 was predominant in Hokkaido Island [5]. BSC-resistant E. coli have been isolated from companion animals [6,7]. The CTX-M-15-producing B2-O25-ST131 E. coli clone has also been isolated from companion animals in several countries [8e10]. In Japan, the CTX-M-15-producing B2-O25:H4-ST131 clone has not yet been derived from companion animals, but blaCTx-M-27e harboring E. coli O25-ST131 has been isolated from dogs and cats [11]. It is suspected that animals can serve as a reservoir of antimicrobial-resistant bacteria [12], and evidence of transfer of ESBL-producing bacteria from animals to humans has been reported [13]. Companion animals are thought to pose a high risk of

1341-321X/$ e see front matter Ó 2013, Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jiac.2013.12.003

244

T. Okubo et al. / J Infect Chemother 20 (2014) 243e249

transmission of antimicrobial-resistant bacteria to humans via the faecaleoral or oraleoral route, because of their high frequency of close contact with humans [12]. Although BSC-resistance in companion animals has become a great concern for public health, interspecies dissemination of BSC-resistant E. coli between companion animals and humans is still unclear. In our previous study, we isolated CTX-M-producing BSC- and fluoroquinolone-resistant E. coli from canine rectal swabs in Hokkaido Island, Japan [14]. We also detected the presence of the CTXM-15-nonproducing O25:H4-ST131 E. coli clone in clinical isolates from humans living in the same area [5]. Here, we examined and compared the resistant determinants and genetic similarity of cephalosporin-resistant E. coli isolates from dogs and humans to estimate the possibility of the transmission of BSC-resistant bacteria between these species. 2. Methods A total of 69 E. coli, including 28 canine and 41 human isolates, which could colonize on Mueller-Hinton agar (Oxoid, Basingstoke, UK) containing 4 mg ml1 of cefpodoxime, were evaluated in this study. Canine isolates were collected between April and June 2003, from rectal swabs at Rakuno Gakuen University Veterinary Teaching Hospital and from 8 companion animal hospitals in Ebetsu, Hokkaido, Japan. Isolates of human origin were collected and stocked in Sapporo Clinical Laboratories Inc. (Sapporo, Hokkaido, Japan) in 2008 and 2009. Ebetsu and Sapporo are neighboring cities and are located in the central region of Hokkaido Island. Human isolates were derived from submitted clinical samples such as urine, pharyngeal samples, phlegm, feces, aspirator samples, intestinal juices, wounds, vaginal secretions, and blood. These isolates were shared with a previous study and their phylogenetic type, O serotype, and sequence type (ST) have already been identified before [5]. Each isolate was obtained from a single dog or patient. Background information of the donors, such as medical records, antimicrobial treatment history, and the history of rearing dogs and humans was unavailable. This study was approved by the ethical review boards of all participating institutions. The MICs of cefazolin, cefpodoxime, cefotaxime, ceftazidime, and cefepime were determined by the agar dilution method in accordance with Clinical and Laboratory Standard Institute (CLSI) criteria [15]. Cefpodoxime and cefepime were kindly provided by Professor Ishii (Department of Microbiology, Toho University School of Medicine, Tokyo, Japan), and other antimicrobials were purchased from SigmaeAldrich (St. Louis, MO, USA). Resistance to these antibiotics was defined on the basis of conventional breakpoints [15]. Total bacterial DNA was extracted from each isolate using InstaGene DNA Purification Matrix (Bio-Rad Japan, Tokyo, Japan) according to the manufacturer’s instructions. PCR amplification of blaTEM and blaSHV was performed as described before [16]. Multiplex PCR was also carried out to detect the genes encoding CTX-M and plasmid-mediated AmpC b-lactamases (PMACs), as described earlier [17,18]. CTX-M- or PMAC-positive isolates were amplified by simplex PCRs to obtain the total sequence [19,20]. PCR products were purified using a High Pure PCR Cleanup Micro Kit (Roche Diagnostics K.K. Tokyo, Japan) and sequenced using a BigDye Terminator v1.1 Cycle Sequencing Kit with a 3130 Genetic Analyzer (Life Technologies Japan, Tokyo, Japan). Association of the insertion sequence (IS) element with blaCTx-M and blaAmpC was determined by PCR using primers for ISEcp1 [19] and ISCR1 [21]. The chromosomal ampC promoter region of isolates was amplified by PCR using primers and conditions previously reported [22]. Amplicons were purified and sequenced in the same manner as described above.

The resulting sequences were compared with the ampC promoter of E. coli K-12 MG1655 (GenBank accession number J01611, considered as wild-type) to detect nucleotide mutations in the region. Isolates suspected to be hyperproducers of chromosomal AmpC were further investigated by a double-disk synergy test for the detection of AmpC production [23]. The phylogenetic type of the isolates was determined by PCRbased methods [24]. O-antigen serotyping was performed using E. coli antisera ‘SEIKEN’ Set 1 (Denka Seiken Co., Ltd, Tokyo, Japan) according to the manufacturer’s instructions. The clonal relationship between isolates was studied by pulsed-field gel electrophoresis (PFGE) according to the Pulse Net CDC protocol [25]. Genomic DNA in each agarose plug was digested with XbaI restriction enzyme (Roche Diagnostics K.K.) [26]. The fingerprints were analyzed using BioNumerics software, version 4.6 (Applied Maths, Sint-Martens-Latem, Belgium). Filter mating assays were performed with E. coli ML4909 (F galK2 galT22 hsdR metB1 relA supE44; rifampicin resistant) as recipient strain [27]. Donor and recipient strains were cultured in LuriaeBertani broth (Invitrogen, Carlsbad, CA, USA) for 16 h at 37  C and then mixed together in the proportion of 1:9 (0.5 mL of donor: 4.5 mL of recipient) on a sterile 0.45-mm filter (Advantec Toyo Kaisha, Ltd., Tokyo, Japan). Filters were placed on Mueller-Hinton agar, incubated for 3 h at 37  C, and washed by vortexing in 1.5 mL sterile saline. Cell suspensions were serially diluted to 108 and 100 mL of the cell dilutions were applied to modified Drigalski agar (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan) containing cefpodoxime (4 mg ml1) and rifampicin (50 mg ml1; Sigmae Aldrich), and then incubated overnight at 37  C. Donors and recipients were also inoculated on this selective media to confirm putative mutants. DNA of transconjugants was extracted and PCR for bla genes was performed to confirm whether transconjugants possessed the same resistance genes as the donors. Transfer frequencies of bla genes were estimated as the counts of colony forming units (CFUs) of transconjugants per donor. Conjugations were carried out twice, independently, and the average of transfer frequencies was calculated. MICs of recipient and transconjugant strains against BSCs were examined in the same manner as above. After conjugation, we selected 12 strains, including 6 randomly selected donors (3 canine and 3 human isolates) and their respective transconjugants for plasmid analysis. PCR-based replicon typing of plasmids was carried out to classify plasmids into incompatibility groups [28]. To detect the bla gene-encoding plasmid, PFGE of whole bacterial DNA digested with S1 nuclease (S1-PFGE) and Southern blot hybridization were performed as described previously [29]. PFGE was carried out under the following conditions: pulse at 6 V/cm, pulse time of 5e45 s, runtime for 17 h, and temperature of 14  C. Lambda DNA Standards in InCert Agarose Gel Plugs (Lonza Rockland Inc., Rockland, USA) were used as size standard. DNA probes specific for bla genes were obtained using a PCR DIG Probe Synthesis Kit (Roche Diagnostics K.K.), with DNA templates extracted from E. coli strains harboring blaCTxM-2, blaCTx-M-27, or blaCMY-2 and primers for the respective b-lactamase genes [19,20]. We also performed Southern blotting with probes created from PCR-based replicon typing product of FIA, FIB, and N replicons in the same manner. Statistical significance was determined by the Chi-square test or Fisher’s exact test. A p-value of