Detection of a single isolate of CTX-M-1-producing Escherichia coli from healthy pigs in Denmark

Sir, Extended-spectrum β-lactamase (ESBL)-mediated resistance is an increasing concern in human clinical settings. In Denmark, only a few cases of ESBL-producing Escherichia coli have been reported from food animals,1–3 however, there was no baseline study on ESBL prevalence among the healthy pig populations in Denmark. In this study, we investigated the prevalence of ESBL-mediated resistance in E. coli isolates obtained from faecal samples of healthy pigs in Denmark. Furthermore, ESBL-related genes and mutations were determined and cephalosporin consumption in pig farms associated with ESBL-mediated resistance was investigated. As part of the DANMAP surveillance programme (The Danish Integrated Antimicrobial Resistance Monitoring and Research Programme), a total of 137 faecal samples were randomly selected among the healthy pigs at farm level between November 2005 and March 2006.1 Faecal sample was enriched in MacConkey broth containing 2 mg/L cefotaxime and 3M™ Petrifilms™ Select E. coli Count Plates (SEC plates) with 2 mg/L cefotaxime were used to identify E. coli with reduced susceptibility to cefotaxime. E. coli appears on SEC plates as dark green to light blue-green colonies and was subcultured on Mueller–Hinton II agar plates supplemented with 2 mg/L cefotaxime. Susceptibility testing for 17 antimicrobials1 was carried out using a commercially dehydrated panel (Trek Diagnostic Systems, UK) and E. coli strains that showed broad-spectrum cephalosporin (ceftiofur and cefpodoxime) resistance were selected and studied further. Only one colony per sample was further investigated and a total of four broad-spectrum-cephalosporin-resistant E. coli isolates were obtained from the 137 samples. ESBL production was determined by the disc diffusion test using five oxyimino-cephalosporins including cefotaxime and ceftazidime with or without clavulanic acid as described by the CLSI guidelines (Table 1).4 Based on the obtained phenotype, the presence of TEM, SHV, CTX-M, ACT, FOX and CMY β-lactamase-encoding genes was studied using PCR and obtained amplicons were sequenced. One isolate (E. coli IV) out of the four was highly resistant to cefotaxime and ceftiofur but susceptible to ceftazidime and cefoxitin. Sequencing of PCR products obtained using primers targeting TEM and CTX-M genes detected the presence of blaCTX-M-1 and blaTEM-1b.5,6E. coli IV was able to transfer its cephalosporin resistance to the recipient E. coli K-12 HEHA4 in a conjugation experiment. The remaining three isolates were all susceptible to cefotaxime as well as ceftiofur and did not have an ESBL phenotype by the disc diffusion test. PCR and sequence analysis detected mutations in the promoter of the chromosomal ampC gene, determined at positions −42(C→T) and −18(G→A). In addition, two of the isolates carried the blaTEM gene (Table 1). PFGE analysis of the four E. coli strains revealed significantly distinguishable PFGE patterns suggesting that they were not clonally related. Phenotypic confirmation test, resistance to non-β-lactams, plasmidic β-lactamase genes and mutations in the promoter of the chromosomal ampC gene in the four E. coli isolates in the study FOX, cefoxitin; CXM, cefuroxime; CAZ, ceftazidime; FEP, cefepime; CTX, cefotaxime; STR, streptomycin; TET, tetracycline; SMX, sulfamethoxazole; TMP, trimethoprim; SPE, spectinomycin. aNon-β-lactam antimicrobials tested: apramycin, chloramphenicol, ciprofloxacin, florfenicol, gentamicin, nalidixic acid, neomycin, spectinomycin, streptomycin, sulfamethoxazole, tetracycline and trimethoprim. bCompared with ampC sequence of E. coli K-12. Phenotypic confirmation test, resistance to non-β-lactams, plasmidic β-lactamase genes and mutations in the promoter of the chromosomal ampC gene in the four E. coli isolates in the study FOX, cefoxitin; CXM, cefuroxime; CAZ, ceftazidime; FEP, cefepime; CTX, cefotaxime; STR, streptomycin; TET, tetracycline; SMX, sulfamethoxazole; TMP, trimethoprim; SPE, spectinomycin. aNon-β-lactam antimicrobials tested: apramycin, chloramphenicol, ciprofloxacin, florfenicol, gentamicin, nalidixic acid, neomycin, spectinomycin, streptomycin, sulfamethoxazole, tetracycline and trimethoprim. bCompared with ampC sequence of E. coli K-12. Data on cephalosporin consumption from the Danish Veterinary Medicines Statistics Programme (VetStat) indicated that the plotting of cephalosporin consumption of the 137 pig farms did not reveal any clear association between farms with high cephalosporin usage and occurrence of ESBL-producing E. coli. This study is the first national surveillance of ESBL-producing E. coli from randomly sampled pig farms all over Denmark using selective enrichment. On a positive note, only one positive strain was detected out of 137 pig farms examined over half a year, suggesting a very low prevalence (0.7%) of ESBL-producing E. coli in the Danish pig populations. But it should be noted that the actual occurrence of ESBL-producing E. coli in Danish pig populations could have been underestimated by only testing a single colony from each sample. The first ESBL-producing E. coli from the Danish primary production were isolated in 2005 when two E. coli isolates carrying the blaCTX-M-1 gene were isolated from diseased pigs as part of the routine diagnostics performed in Denmark.3 In 2006, this increased to 10 ESBL-producing pathogenic E. coli isolated from diseased Danish pigs and cattle as well as the first ESBL-producing Salmonella Typhimurium isolated from a healthy pig in Denmark. All these isolates carried versions of the blaCTX-M gene (blaCTX-M-1, blaCTX-M-2 or blaCTX-M-9) and were from farms that had used cephalosporins previously (H. Hasman, Technical University of Denmark, unpublished results). However, a very recent Danish study demonstrated 19 CTX-M-1-carrying E. coli isolates from two pig farms with a history of ceftiofur usage. No statistical significance between usage of ceftiofur and occurrence of ESBL-producing E. coli could be concluded, due to the limited number of farms investigated.7 Likewise, the number of positive isolates obtained from healthy pigs in our study (where cross-contamination between animals might have taken place) is probably too low to make firm conclusions about a clear association between ESBL production and cephalosporin usage. This study was supported by a grant from the EU Marie Curie Programme TRAINAU (MEST-CT-2004-007819). The technicians in the Unit of Antimicrobial Resistance are acknowledged for their training of Miss E. Chouliara and Miss S. Wu. Furthermore Erik Jacobsen is acknowledged for providing information on cephalosporin usage. None to declare.