Ivan Literak, Department of Biology and Wildlife Diseases, Faculty of Veterinary Hygiene and Ecology, University of Veterinary and Pharmaceutical Sciences, Palackeho 1–3, 612 42 Brno, Czech Republic. E-mail: email@example.com
Aims: To determine the presence of antibiotic-resistant faecal Escherichia coli in populations of wild mammals in the Czech Republic and Slovakia.
Methods and Results: Rectal swabs or faeces collected during 2006–2008 from wild mammals were spread on MacConkey agar and MacConkey agar containing 2 mg l−1 of cefotaxime. From plates with positive growth, one isolate was recovered and identified as E. coli. Susceptibility to 12 antibiotics was tested using the disk diffusion method. Resistance genes, class 1 and 2 integrons and gene cassettes were detected in resistant isolates by polymerase chain reaction (PCR). Extended-spectrum beta-lactamases (ESBL) were further characterized by DNA sequencing, macrorestriction profiling and determination of plasmid sizes. Plasmid DNA was subjected to EcoRV digestion, transferability by conjugation and incompatibility grouping by multiplex PCR. The prevalence of resistant isolates was 2% in small terrestrial mammals (rodents and insectivores, nE. coli = 242), 12% in wild ruminants and foxes (nE. coli = 42), while no resistant isolates were detected in brown bears (nE. coli = 16). In wild boars (Sus scrofa) (nE. coli = 290), the prevalence of resistant isolates was 6%. Class 1 and 2 integrons with various gene cassettes were recorded in resistant isolates. From wild boars, five (2%, nrectal smears = 293) multiresistant isolates producing ESBL were recovered: one isolate with blaCTX-M-1 + blaTEM-1, three with blaCTX-M-1 and one with blaTEM-52b. The blaCTX-M-1 genes were carried on approx. 90 kb IncI1 conjugative plasmids.
Conclusions: Antibiotic-resistant E. coli occured in populations of wild mammals in various prevalences.
Significance and Impact of the Study: Wild mammals are reservoirs of antibiotic-resistant E. coli including ESBL-producing strains which were found in wild boars.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
The increasing antimicrobial resistance in bacteria is a current problem in medicine. In the veterinary area, the alarming state of bacterial antibiotic resistance is seen in examining the Escherichia coli isolates, where attention has been given especially to food-producing animals such as pigs, cattle and domestic fowl (White 2006). A phenomenon studied rather more recently is the spreading of resistant E. coli strains into the environment beyond the populations directly influenced by the antibiotic practice and the subsequent colonization of wild animal populations.
We have studied the occurrence of resistant commensal E. coli in the populations of various wild bird species in the Czech Republic (Dolejska et al. 2007, 2008, 2009; Literak et al. 2007). The prevalence of resistant E. coli was found to be high in omnivorous synanthropic birds like black-headed gulls (Larus ridibundus) and rooks (Corvus frugilegus), while in house sparrows (Passer domesticus) from cattle stables the prevalence was substantially lower. Wild birds colonized by resistant E. coli, and particularly the synanthropic bird species with high prevalence of resistant E. coli, can thus become important reservoirs and vectors of these strains that are potentially concerning because of the presence of conjugatively transferable determinants of resistance.
The goal of the present study is to characterize the central European (Czech and Slovak) populations of wild mammals regarding the occurrence of commensal antimicrobial-resistant E. coli.
Material and methods
Small terrestrial mammals (Rodentia, Insectivora) were captured in a suburban and forest environment. They were snap trapped and dissected. During dissection, samples of the large intestines’ contents were collected for the subsequent E. coli isolation. In the suburban environment of the village of Kunin (north-eastern Czech Republic), small terrestrial mammals were captured near fields and family houses with gardens in May 2007 and 2008. In a forest environment frequented by people but not by farm animals near the village of Vracov (south-eastern Czech Republic), small terrestrial mammals were captured in September 2008.
Large wild mammals (Cervidae, Canidae) were examined in Poloniny National Park (NP) in north-eastern Slovakia. Faecal samples of these mammals were collected in winter of 2006/2007 and 2007/2008. The brown bear (Ursus arctos) is a carnivore living in the mountains of Slovakia. Its present population density is maintained by regulated culling. Rectal swabs were collected from brown bears that were shot in the Low Tatras NP in central Slovakia during 2005–2007. The wild boar (Sus scrofa) is common in central Europe and is a frequently hunted game species. Rectal swabs were collected from wild boars that were shot near Litomysl and Opava (central and north-eastern Czech Republic) in 2006 and 2007 (Fig. 1).
Sample collection and processing
Individual rectal swabs from wild pigs, samples of intestinal content from small terrestrial mammals or samples of faeces were placed overnight in buffered peptone water (BPW) at 37°C, then cultivated for E. coli and tested for susceptibility to 12 antimicrobial agents as previously described (Literak et al. 2007). Briefly, one colony of each plate was tested for susceptibility to antimicrobial agents in accordance with CLSI (The Clinical and Laboratory Standards Institute, Wayne, PA, USA). Antibiotic susceptibility was tested by disk diffusion method on Mueller–Hinton agar (CM337; Oxoid, UK) using antibacterial substances: amoxicillin–clavulanic acid (30 μg), ampicillin (10 μg), cephalothin (30 μg), ceftazidime (30 μg), chloramphenicol (30 μg), ciprofloxacin (5 μg), gentamicin (10 μg), nalidixic acid (30 μg), streptomycin (10 μg), trimethoprim–sulfamethoxazole (25 μg), sulfonamides compounds (300 μg) and tetracycline (30 μg) (Oxoid). In E. coli isolates found to be resistant to one or more of the antibiotics listed above, polymerase chain reaction (PCR) was used to detect specific antibiotic-resistant genes, integrase genes int1 and int2, and gene cassettes within class 1 and 2 integrons (Dolejska et al. 2007; Literak et al. 2007). The presence of genes tetA, tetB, tetC, tetD, tetE, tetG, blaTEM, blaSHV, blaOXA-2-like, blaOXA-1/-30, cat, cmlA, floR, sul1, sul2, sul3, strA, int1, int2, variable region of class 1 integron, variable region of class 2 integron, dhfr1, dhfr12, dhfr17, aadA1, aadA2, aadA5, estX and sat1/2 was tested with primers and under the conditions listed in Table 1. The control strains E. coli F134, H195, HR17, M2, M44, M66, M148, OP1, R128, Aeromonas sp. A233, Salmonella enterica serovar Typhimurium DT104S1 and Klebsiella pneumoniae ATTC700603 were used. All E. coli strains were identified by the API 10S test (BioMerieux, France).
Table 1. List of primers for antibiotic resistance genes used in the study
The samples from BPW were enriched with MacConkey broth and subcultivated onto MacConkey agar containing cefotaxime (2 mg l−1) to detect E. coli strains with extended-spectrum beta-lactamases (ESBLs) (Wu et al. 2008). The colonies were examined using the double-disc synergy test for the production of ESBL (Thomson and Sanders 1992; CLSI 2008a) and identified by the API 10S test. The genes responsible for the ESBL phenotype (blaTEM, blaSHV and blaCTX-M) were identified by PCR, and the products were further analysed using sequencing (ABI 310 Genetic Analyser; Applied Biosystems). The primers 1537 and 1580 were used for sequencing of blaCTX-M-1 group, and the primers 686 and 757 were used for sequence analysis of blaTEM (Table 1).
Antimicrobial susceptibility of all the ESBL-positive isolates was tested quantitatively by broth microdilution with cation-adjusted Mueller–Hinton broth, according to CLSI guidelines (CLSI 2008b). Microtitre trays were used with dehydrated dilution ranges of custom-made panels of antibiotics (Trek Diagnostic System, East Grinstead, UK). The following antimicrobial agents were included in the panel: amoxicillin–clavulanic acid, ampicillin, apramycin, cefpodoxime, ceftiofur, cephalotin, chloramphenicol, ciprofloxacin, colistin, florfenicol, gentamicin, nalidixic acid, neomycin, spectinomycin, streptomycin, sulfamethoxazole, tetracycline and trimethoprim with the test ranges and Subcommittee on Veterinary Antimicrobial Susceptibility Testing (CLSI/VAST) breakpoints as described previously (CLSI 2008b). Identification of E. coli phylogenetic groups was performed using a multiplex PCR assay (Clermont et al. 2000) in all the ESBL isolates. By this method, E. coli isolates can be devided into four main phylogenetic groups (A, B1, B2 and D) according to the presence of chuA and yjaA genes and DNA fragment TSPE4.C2. The isolates were typed by XbaI-pulsed field gel electrophoresis (PFGE) (CDCP 2004). The samples with no band differences were designed as indistinguishable and possibly epidemiologically linked, and the samples with the different PFGE patterns (more than three band changes) were epidemiologically unrelated (Tenover et al. 1995).
Transferability of bla genes was tested by conjugation. Plate mating experiments were performed using plasmid-free, rifampicin-resistant and nalidixic acid-resistant E. coli MT102RN and Salm. Typhimurium SL5325 as recipients (Caroff et al. 1999; Olesen et al. 2004). The strains were grown to exponential phase, mixed (1 : 1), and 500 μl of the donor and recipient mixture was incubated using a bacteriological filter on the surface of blood agar at 37°C overnight. Transconjugants were selected on brain heart infusion (BHI) medium supplemented with 25 mg l−1 rifampicin, 32 mg l−1 nalidixic acid and 2 mg l−1 cefotaxime.
Plasmids of ESBL strains were characterized by replicon typing and EcoRV digestion. Primarily plasmid DNA was extracted using the Qiagen plasmid midi kit (Qiagen, Germany). Plasmid DNA was introduced to competent E. coli Genehogs® (Invitrogen, USA) by electroporation followed by the selection of transformants on BHI agar supplemented with cefotaxime (2 mg l−1). The presence of relevant bla gene in transformants was confirmed by PCR. The size of the plasmid with ESBL gene from transformants was designated by S1-PFGE. Plasmid DNA from transformants was digested with EcoRV and subjected to gel electrophoresis in 0·8% agarose gel for 4 h at 4·0 V cm−1. Plasmids were replicon typed as previously described (Carattoli et al. 2005).
The following abbreviations are used for resistance phenotype in this study: ampicillin (A), amoxicillin–clavulanic acid (Ac), chloramphenicol (C), cephalotin (Cf), ceftazidime (Cfz), ciprofloxacin (Cip), gentamicin (Gn), nalidixic acid (Na), streptomycin (S), sulfonamides cp. (Su), trimethoprim–sulfamethoxazole (Sxt) and tetracycline (T).
In the suburban environment of the village of Kunin, 75 small terrestrial mammals were examined (Mus musculus– 4 specimens, Apodemus agrarius– 24, Apodemus flavicollis– 20, Apodemus sylvaticus– 2, Microtus arvalis– 6, Microtus subterraneus– 13, Arvicola terrestris– 1, Sorex araneus– 1, Crocidura suaveolens– 4). Escherichia coli was recovered from 43 animals. Resistance to antibiotics was recorded in only one isolate (1/43, 2%) from striped field mouse (A. agrarius). It was a multiresistant isolate with the phenotype ACSSuSxtT and with resistance genes blaTEM, cat, strA, sul1, sul2 and tetB. Also detected was the class 1 integron (2·0 kb) with gene cassette dhfr12-orf-aadA2. No isolates with ESBL production were found.
In the forest environment near the village of Vracov, 235 small terrestrial mammals were examined (A. flavicollis– 30 specimens, A. sylvaticus – 51, A. microps– 22, Microtus arvalis– 3, Microtus subterraneus– 11, Clethrionomys glareolus– 103, Sorex araneus– 12, Sorex minutus– 2, Crocidura suaveolens– 1). Escherichia coli was recovered from 199 animals. There were four resistant isolates (4/199, 2%). Three resistant isolates originated from bank voles (Clethrionomys glareolus): E. coli of the phenotype A (with the gene blaTEM), phenotype T (with the gene tetA) and phenotype Na. One resistant isolate was from A. sylvaticus: phenotype AT (with genes blaTEM and tetB). No isolates with ESBL production were found.
Cervidae, Canidae, Ursus arctos
Sixty samples of faeces from Poloniny NP were examined (Cervus elaphus– 53, Capreolus capreolus– 2, Vulpes vulpes– 5). Escherichia coli was isolated from 42 samples. Five isolates were resistant (5/42, 12%). Four resistant E. coli isolates originated from Cervus elaphus with phenotype T (with tetA gene) and one isolate originated from Vulpes vulpes with phenotype T (with tetB gene). A total of 21 rectal swabs were collected from U. arctos, and 16 E. coli isolates were recovered. In no U. arctos isolate was there recorded a resistance to the antimicrobials tested. No ESBL-producing isolates were found.
A total of 293 rectal swabs were collected from S. scrofa, and 290 E. coli were recovered. Antibiotic resistance was recorded in 17 isolates (17/290, 6%) (Table 2).
Table 2. Overview of antibiotic-resistant Escherichia coli isolates from wild boars
Of the 293 rectal swabs, five E. coli strains (5/293, 2%) were isolated on the medium with cefotaxime. All five strains produced ESBL (they were positive in the double disk synergy test). In all cases, the strains were multiresistant (with resistance to 3–8 antibiotics), and in two strains, class 2 integrons were found (Table 3). The genes blaCTX-M-1 (three strains), blaTEM-52b (one strain) and the combination of blaCTX-M-1 and blaTEM-1 (one strain) were found by PCR and sequencing. PFGE analysis of all five wild type isolates showed that each isolate produced a distinct pulse-type. All the isolates belonged to phylogenetic group A. The plasmids with blaCTX-M-1 were transferred by conjugation to the E. coli MT102RN and Salm. Typhimurium SL5325, while the plasmid with blaTEM-52b was transferred only to E. coli. The sul2 gene transferred together with blaCTX-M-1 genes by conjugation as well as transformation in all the strains, thus showing the presence of this gene on the same plasmid encoding blaCTX-M-1. All four plasmids encoding blaCTX-M-1 had identical EcoRV RFLP profile designated A. They belonged to the IncI group and had a size of approx. 90 kb. The blaTEM-52b-harbouring plasmid had a distinct RFLP profile (B), and the size of the plasmid was approx. 45 kb. In this plasmid, however, it was not possible to specify the Inc group by the method used.
Table 3. Phenotypic and genotypic characteristics of ESBL-positive Escherichia coli strains from wild boars
Nonbeta-lactam antibiotic resistance phenotype*
Other antibiotic resistance genes and integrons†
Plasmid type (EcoRV)
bla gene on plasmid
Plasmid size (approximate kb)
Salmonella enterica serovar Typhimurium
ESBL, extended-spectrum beta-lactamase.
*S, streptomycin; Spe, spectinomycin; Su, sulfonamides cp.; T, tetracycline.
†I2, class 2 integron.
‡The phylogenetic groups were determinated by PCR for chuA and yjaA genes and DNA fragment TSPE4.C2 according to Clermont et al. (2000).
§NT, incompatibility group not specified by the primers used in the study.
strA, sul2, tetA
tetB, sul2, int2, I2: 1·5 kb sat-aadA1
tetA, sul2, int2, I2: 1·5 kb sat-aadA1
The low prevalence (2%) of resistant E. coli isolates in the populations of small terrestrial mammals at two sites in Czech Republic was probably caused by an exceptional contact between the hosts and resistant E. coli strains produced by either humans or domestic animals that were capable of colonizing new hosts or at least of enabling the horizontal transfer of genetic determinants of antibiotic resistance to the E. coli strains that colonize small terrestrial mammals.
By examination of small terrestrial mammals originated from two pig farms in the Czech Republic, we found much higher prevalences of resistant E. coli in these mammals (11% and 54%, respectively) depending probably on the high prevalences of resistant E. coli isolates in pigs reared and treated with antibiotics in those farms (Literak et al. 2009). Recently, it was also suggested that E. coli isolates from wild small terrestrial mammals living on swine farms in Canada have higher rates of resistance and are more frequently multiresistant than E. coli isolates from environments, such as natural areas, that are less impacted by human and agricultural activities (Kozak et al. 2009).
Exceptional, yet still most frequent, was the resistance to tetracycline in E. coli from Czech small terrestrial mammals. Resistance to tetracycline had been frequent (unlike other antibiotics tested) also in E. coli isolates from bank voles examined in Poland (Swiecicka et al. 2003). A study of antibiotic resistance in E. coli from rats in Indonesia has suggested that tetracycline-resistant E. coli has become established in rats in West Java, whereas rats on the nearby uninhabited Krakatau Islands do not contain these bacteria (Graves et al. 1988). It was concluded that excessive or uncontrolled use of antibiotics by human population in areas where faecal contamination of the environment is inevitable leads to changes in the enteric flora of wild mammals living in these areas. We concur with that conclusion. Role of environmental bacteria as natural resistance reservoirs and possible source of genetic elements (Chopra and Roberts, 2001) could be another explanation for the frequent occurrence of tetracycline resistance in E. coli from wild mammals.
One E. coli strain isolated from the striped field mouse showed a multiresistant pattern. It is an exceptional finding in our study of small terrestrial mammals. However, the colonization of other small terrestrial mammals – 13-lined ground squirrels (Spermophilus tridecemlineatus) – by multiresistant isolates of other Gram-negative members of the γ-proteobacteria group was recorded in the USA (Cloud-Hansen et al. 2007).
In free-living ruminants, e.g. the Japanese serow (Capricornis crispus), the prevalence of resistant E. coli isolates was markedly lower compared to individuals of the same species in captivity (Kinjo et al. 1992). At Poloniny NP in Slovakia, the prevalence of resistant E. coli isolates from wild mammals, particularly ruminants but also the red fox, was 12%. This finding suggests the existence of important sources of colonization by resistant E. coli isolates in Poloniny NP. Meadows that are a food source for wild ruminants in the NP are also used for grazing of cattle. Cattle, especially calves, could be the source of resistant isolates (Dolejska et al. 2008). Foxes searching for food in the immediate vicinity of human habitations, and particularly in winter, can easily become infected with food contaminated by resistant isolates of human origin, or domestic animals could be the source. At villages within Poloniny NP, resistant E. coli isolates were found also in domestic dogs (data not shown). Both in wild ruminants and foxes from Poloniny NP, only monoresistant isolates with resistance to tetracycline were recorded, and we have not found the presence of ESBL in E. coli isolates. The occurrence of ESBL-producing E. coli from two deer (blaTEM-52) and one fox (blaCTX-M-14 + blaTEM-1) has recently been recorded in Portuguese NPs (Costa et al. 2006) suggesting the spreading of such strains also into the wildlife of NPs.
Antibiotic-resistant E. coli isolates were found among wild ruminants in the Stelvio NP, Italy (Caprioli et al. 1991). In Norway, antibiotic resistance was also found in E. coli isolates from wild cervids, and most of the resistant isolates were resistant to one type of antibiotics only (Lillehaug et al. 2005). Our results from Poloniny NP were similar. On the other hand, E. coli isolates from wild cervids and bank voles from remote areas of Finland were almost free of antimicrobial resistance (Osterblad et al. 2001), and thus it was suggested that the widespread resistance found in E. coli populations must be caused by human activities.
Class 1 and 2 integrons are genetic elements that play an important role in the development of antibiotic resistance. They have a worldwide distribution and are described from Gram-negative bacteria colonizing both humans and animals (Sunde 2005). Integrons so far characterized have originated mostly from bacteria isolated from environments where antibiotics are heavily used (Sunde 2005). Recently, a class 1 integron was found in E. coli isolated from wild reindeer (Rangifer tarandus) in a remote mountain area in Norway (Sunde 2005). In the present study, we have recorded class 1 and 2 integrons in striped field mouse and wild boars. In our other studies, we have recorded the occurrence of E. coli isolates with class 1 and 2 integrons in both domestic and wild animals as well as in surface waters in the Czech Republic (Dolejska et al. 2007, 2008, 2009; Literak et al. 2007). Class 1 integrons with the same gene cassettes as in wild boars were found in cattle, in black-headed gulls and in water from the pond where these gulls lived. Class 2 integrons with the same gene cassettes as in wild boars were also recorded in black-headed gulls. It seems that E. coli isolates possessing integrons with various gene cassettes are emerging more and more frequently in synanthropic wild animal populations, although these animals are not directly influenced by antibiotics.
The occurrence of multiresistant E. coli isolates from wild boars with ESBL-producing genes is a novelty in Europe. In central Europe, the wild boar is a common and widespread large mammal that lives in forest, field and suburban habitats and is intensively hunted in many countries (Herre 1986). Wild boars are omnivores and can visit communal refuse sites as well as the proximity of animal farms and consume animal waste containing resistant E. coli strains. We assume that the presence of multiresistant E. coli isolates with ESBL in wild boars is caused by these sources. Bacterial resistance to multiple antibiotics was recently documented in bacteria, including E. coli, colonizing the avian scavengers Egyptian vultures (Neophron percnopterus) in Spain (Blanco et al. 2007). The prevalence of antibiotic-resistant vulture bacteria was higher in areas where vultures consumed the highest proportion of carrion from stabled livestock, especially fattening pigs and poultry.
Most of the ESBL isolates from wild boars contained the blaCTX-M-1 gene. CTX-M-1 is one of most prevalent ESBLs in Europe. It has been frequently detected in humans (Novais et al. 2007) as well as in food-producing animals (Meunier et al. 2006). ESBL-producing E. coli with blaCTX-M-1 has also been recently detected in wild birds (Poeta et al. 2008; Dolejska et al. 2009). The blaCTX-M-1 in all the isolates was carried on large 90 -kb plasmids. This plasmid also carried the sulfonamide-resistance gene sul2. This plasmid seems to be of relatively broad host range because of conjugation to the Salmonella recipient. All the plasmids had the same RFLP profile and belonged to the incompatibility group IncI1. This type of plasmid has recently been associated with blaCTX-M-1 in poultry in France (Girlich et al. 2007) and Denmark (data not published). It provides evidence that the ESBL-producing E. coli selected for in humans and food-producing animals are colonizing wild animals, which can thus become reservoirs and possible vectors of these strains into the environment.
In central Europe, we can observe a widespread occurrence of antimicrobial resistance in E. coli isolates from wildlife. We suppose similarly as do Skurnik et al. (2006) that antimicrobial resistance found in E. coli from wild mammals is anthropogenic. Among wild mammals, the wild boar plays an important role in the circulation of resistant E. coli isolates including isolates with integrons and ESBL production. The wild boar can be considered an important reservoir and vector of these strains in the environment.
We are grateful to K. Dibdakova, L. Hanusova, N. Horakova, L. Chudobova, F. Kupka, T. Najer, J. Mikulska, S. Ondrus, P. Slamova, M. Slavikova and E. Suchanova for their assistance in the field and/or in the laboratory. This study was funded by Grant No. MSM6215712402 of the Ministry of Education, Youth and Sports of the Czech Republic. Capture, euthanasia and examination of small terrestrial mammals were approved by the Ethical Commission of the Ministry of Education, Youth and Sports of the Czech Republic.