Detection of Escherichia coli harbouring extended-spectrum -lactamases of the CTX-M classes in faecal samples of common buzzards (Buteo buteo)

Sir, In recent years, infections due to Escherichia coli harbouring extended-spectrum β-lactamases (ESBLs) of the CTX-M classes have dramatically increased among human populations, particularly in the community setting.1 There exist just a few previous reports in which ESBLs were detected in faecal E. coli isolates of wild animals,2 but, to our knowledge, never in faecal E. coli isolates of buzzards (Buteo buteo). Thirty-three faecal samples from buzzards of Portugal were recovered from September 2007 to February 2008 and were studied for the presence of ESBL-producing E. coli isolates. All the faecal samples were collected individually from each buzzard and obtained in collaboration with CRATAS (Centre of Collecting, Welcome and Handling of Wild Animals). This centre is located in the University of Trás-os-Montes and Alto Douro and receives injured animals. None of the buzzards had been previously fed by humans or had received antibiotics. Most of the animals inhabited the Pêneda Gêres Natural Park or other rural conservation areas of Portugal. Faecal samples were screened for the presence of ESBLs using Levine agar (Oxoid Limited, UK) supplemented with 2 mg/L cefotaxime (Levine–CTX) (Sigma–Aldrich, USA). Two colonies with typical E. coli morphology were selected and identified by classical biochemical methods (Gram, catalase, oxidase, indole, Methyl Red–Voges–Proskauer, citrate and urease) and by the API 20E system (BioMérieux, La Balme Les Grottes, France) from each positive faecal sample. Susceptibility to 16 antibiotics (ampicillin, amoxicillin/clavulanic acid, cefoxitin, cefotaxime, ceftazidime, aztreonam, imipenem, gentamicin, amikacin, tobramycin, streptomycin, nalidixic acid, ciprofloxacin, sulfamethoxazole/trimethoprim, tetracycline and chloramphenicol) (Oxoid Limited, UK) was determined by the CLSI disc diffusion method3 for all recovered E. coli isolates. E. coli ATCC 25922 was used as a quality-control strain. Isolates resistant to third-generation cephalosporins (i.e. cefotaxime or ceftazidime) were selected for further studies (one per faecal sample, or two if they presented different phenotypes of antibiotic resistance). The double disc diffusion test (cefotaxime, ceftazidime and aztreonam in the presence or absence of amoxicillin/clavulanic acid)3 was performed to detect ESBL production. The presence of genes encoding TEM-, SHV-, OXA- and CTX-M-type β-lactamases was studied by specific PCRs,2 and positive amplicons were sequenced to determine the specific type of β-lactamase gene. The genetic environment of blaCTX-M genes was studied by PCR and sequencing in all blaCTX-M-containing isolates using previously reported primers.4 The following antibiotic resistance genes were also sought by PCR:2tet(A) and tet(B) (in tetracycline-resistant isolates); aadA (in streptomycin-resistant isolates); aac(3)-II and aac(3)-IV (in gentamicin-resistant isolates); and sul1, sul2 and sul3 (in sulfamethoxazole/trimethoprim-resistant isolates). The presence of the intI1 and intI2 genes, encoding class 1 and 2 integrases, respectively, and the detection of phylogenetic groups of E. coli isolates were studied by PCR.2 E. coli colonies were isolated from 5 of the 33 (15.2%) faecal samples in the Levine–CTX screen. Two E. coli isolates from each positive sample were recovered, showing different phenotypic and genomic profiles. All 10 of these isolates exhibited a resistant phenotype to cefotaxime and/or ceftazidime, and gave a positive ESBL production test. The β-lactamase genes detected in the ESBL-positive E. coli isolates were the following: blaCTX-M-32 + blaTEM-1 (seven isolates); and blaCTX-M-1 + blaTEM-1 (three isolates) (Table 1). The high prevalence of blaCTX-M-32 in commensal E. coli isolates of buzzards in this study is remarkable (15.2% of total buzzards), as this gene is not frequently found in animal isolates, having been reported in only a few studies previously.2,5 Additionally, some of the birds were colonized with both CTX-M-1- and CTX-M-32-producing E. coli. The orf477 sequence was detected downstream of the blaCTX-M-1 and blaCTX-M-32 genes, and ISEcp1 was found upstream of the blaCTX-M-1 gene in all our isolates; similar molecular arrangements incorporating CTX-M genes have been previously reported.2,4 The association of insertion sequences (e.g. ISEcp1) upstream of the CTX-M β-lactamase genes may be involved in their dissemination and expression. Our CTX-M-containing isolates were all multiresistant, with resistance to tetracycline and sulfamethoxazole/trimethoprim due to tet(A) [associated or not with tet(B)] and different combinations of the three sul genes, respectively. In addition, the aadA gene was detected in the six streptomycin-resistant isolates and the aac(3)-IV gene in the two gentamicin-resistant isolates (Table 1). Characteristics of the ESBL-positive faecal E. coli isolates recovered from buzzards GEN, gentamicin; TOB, tobramycin; AMK, amikacin; NAL, nalidixic acid; CIP, ciprofloxacin; TET, tetracycline; STR, streptomycin; SXT, sulfamethoxazole/trimethoprim. Characteristics of the ESBL-positive faecal E. coli isolates recovered from buzzards GEN, gentamicin; TOB, tobramycin; AMK, amikacin; NAL, nalidixic acid; CIP, ciprofloxacin; TET, tetracycline; STR, streptomycin; SXT, sulfamethoxazole/trimethoprim. All the blaCTX-M-1-producing isolates were classified in the B1 phylogroup. The blaCTX-M-32-producing isolates belonged to the B2 (four isolates), A (two isolates) or B1 (one isolate) phylogroups. The B2 phylogroup has been associated in previous reports with more-virulent isolates and has also been linked with some specific mechanisms of resistance, as is the case of the β-lactamase CTX-M-15.6 The possible association of blaCTX-M-32 in isolates of the B2 phylogroup should be tracked in the future. This is the first time, to our knowledge, that CTX-M-producing E. coli isolates have been detected in buzzards. Buzzards are carnivorous birds, and they can fly large distances searching for food and territory. It is possible that these feeding habits could expose them to faecal material of farm animals or even of humans. This might explain the acquisition and dissemination of bacteria harbouring antibiotic resistance genes in the buzzard population, even in the absence of direct antibiotic pressure. This study highlights that ESBLs are found in ecosystems other than those closely related to humans or containing obvious antibiotic resistance selection pressures. More studies should be carried out in the future with different kinds of wild animals to confirm the dissemination of ESBLs in other ecosystems and animal populations. This study was carried out as part of the routine work of our Department. None to declare. We thank the CRATAS (Centre of Collecting, Welcome and Handling of Wild Animals) for their contribution to the collection of samples.