Abstract
We conducted a cross-sectional study to determine the prevalence of antimicrobial resistance (AMR) in fecal Escherichia coli isolates from raccoons (Procyon lotor) living in Ontario, Canada. From June to October 2007, we trapped raccoons in three areas: one primarily urban site around Niagara, one primarily rural site north of Guelph, and one at the Toronto Zoo. In addition, we conducted a longitudinal study at the Toronto Zoo site to investigate the temporal dynamics of fecal E. coli and AMR in raccoons. Reduced susceptibility to ≥1 antimicrobial agent was detected in E. coli isolates from 19% of 16 raccoons at the urban site, 17% of 29 raccoons from the rural site, and 42% of 130 samples collected from 59 raccoons at the zoo site. Raccoons from the zoo site were significantly more likely to shed E. coli with reduced susceptibility to ≥1 antimicrobial agent than animals from the rural site (odds ratio [OR], 3.41; 95% confidence interval [CI], 1.17 to 12.09; P = 0.02). Resistance to expanded-spectrum cephalosporins (and the associated blaCMY-2 gene) was detected in two animals from the zoo site and one animal from the rural site. Serotyping and pulsed-field gel electrophoresis analysis show that raccoons on the zoo grounds harbor a diverse assemblage of E. coli, with rapid bacterial turnover within individuals over time. Our study indicates that raccoons may shed resistant bacteria of public health significance and that raccoons have the potential to disseminate these bacteria throughout their environment.
INTRODUCTION
Antimicrobial resistance (AMR) is a global health issue, affecting human, animal, and environmental health (13, 18). Wild animals, particularly species that live in close association with humans, may be exposed to resistant bacteria in their environment, and antimicrobial resistance has been detected in fecal bacteria from a variety of wild animals, including birds, reptiles, mammals, and fish, throughout the world (2, 15, 17, 24). Wild animals have been implicated as potential reservoirs of resistant bacteria, and it has been hypothesized that wild animals might act as vehicles for the dissemination of resistant bacteria throughout the environment (4). However, since most studies of antimicrobial resistance in wildlife have been cross-sectional, focused primarily on identifying resistance phenotypes and genotypes, the role of wildlife in maintaining antimicrobial-resistant bacteria in the environment remains largely unknown. Understanding the role of wildlife in the complex epidemiology of antimicrobial resistance may help us identify potential gaps in current strategies aimed at reducing the burden of antimicrobial resistance in the environment.
Small wild mammals, such as mice and voles, are often targeted for studies investigating the impact of human activity on the development of antimicrobial resistance the environment and associated wildlife (2, 8, 12); however, because of the limited home range of these species, they are unlikely to be involved in the widespread dissemination of resistant bacteria. For this study, we targeted raccoons (Procyon lotor), which are common, midsized mammals that live in close association with humans in both urban and agricultural areas. Raccoons are more likely to be exposed to anthropogenic sources of resistant bacteria than species living in more-remote areas. In addition, their larger home range size (1 to 4 km2, depending on habitat, age, and sex [22]) indicates that they have the potential to be involved in the dissemination of resistant bacteria throughout the environment.
The objectives of this study were to (i) compare the prevalences and patterns of antimicrobial resistance in Escherichia coli isolates from raccoons at different sites, including an urban site and a rural site, and (ii) examine the role of raccoons in maintaining resistant E. coli by investigating the temporal dynamics of fecal E. coli and of antimicrobial resistance in E. coli in individual raccoons from a single site.
MATERIALS AND METHODS
Trapping and specimen collection.
Procedures for trapping and handling raccoons were approved by the Animal Care Committee at the University of Guelph in accordance with the guidelines of the Canadian Committee on Animal Care. The trapping methods and sites used in this study have been described previously (10). Briefly, raccoons were live trapped using Tomahawk traps (Tomahawk Live Trap Co., Tomahawk, WI) set in three areas: an urban setting in Niagara, Ontario, Canada (43°3′N, 79°2′W), a rural setting north of Guelph, Ontario, Canada (43°57′N, 80°24′W), and on the grounds of the Toronto Zoo (43°49′N, 79°11′W).
At the zoo site, a total of 40 traps were set in two areas for three nights each month from June to October 2007. Traps were set at sites where raccoons were known to be present and where there was limited public access. Traps were baited with cat food, set in the evening, and checked the following morning. Captured raccoons were brought to a centralized holding area for processing, unless they had already been caught that month, in which case they were released immediately at the point of capture.
Raccoons were anesthetized using an intramuscular injection of medetomidine hydrochloride (0.05 mg/kg of body weight) (Domitor [1 mg/ml]; Pfizer Animal Health, Pfizer Canada, Inc., Kirkland, Quebec, Canada) and ketamine hydrochloride (5 mg/kg) (Ketaset [100 mg/ml]; Wyeth Animal Health, Guelph, Ontario, Canada) prior to removal from traps. A numbered metal ear tag (no. 1005-3; National Band and Tag Co., Newport, KY) was placed in one ear, and a passive integrated transponder (PIT) tag (AVID Canada, Calgary, Alberta, Canada) was injected subcutaneously between the shoulder blades for subsequent identification. Sex, age, class (adult or juvenile; based on animal size and tooth wear/staining), and mass were recorded for each animal. Fecal samples were collected per rectum. After sampling was complete, animals were given an anesthetic reversal agent, atipamazole (0.25 mg/kg) (Antisedan [5 mg/ml]; Pfizer Animal Health, Pfizer Canada, Inc., Kirkland, Quebec, Canada), and placed back in the traps to recover from the anesthetic prior to release at the point of capture. Individual animals were sampled only once per monthly trapping session; however, multiple samples were collected from the same individual if it was caught in subsequent months.
As part of separate, ongoing management procedures at the Toronto Zoo, the majority of raccoons were sterilized (vasectomy or tubal ligation), vaccinated for distemper and rabies, and treated for parasites upon first capture. Nine animals were given 1 ml/animal of penicillin (penicillin G, benzathine, and procaine sterile aqueous suspension; Vétoquinol Canada, Inc., Lavaltrie, Quebec, Canada) as a precaution against infection after surgery in June. Animals that underwent surgery were released at the point of capture the following morning.
Samples from the other two sites were obtained opportunistically from raccoons trapped as part of the Ontario Ministry of Natural Resources Rabies Research and Control Program using the same methods that were used for the raccoons at the zoo site. Samples from 20 urban animals were obtained from animals trapped between 29 and 31 August 2007, and samples from 30 rural animals were obtained from animals trapped from 2 to 4 October 2007. All fecal samples were placed in sterile vials and kept in a cooler in the field. Samples were kept refrigerated at the laboratory prior to submission for E. coli culture (within 3 days of sample collection).
E. coli isolation.
Fecal specimens ranged from 0.5 g to 5 g. A portion of the specimen was suspended in buffered peptone water (Becton Dickinson, Sparks, MD) at a 1:10 ratio and incubated at 37°C overnight. The broth suspension was homogenized using a vortex, and 2 ml was dispensed into an equal amount of double-strength EC (Escherichia coli) broth (Becton Dickinson) and incubated at 37°C overnight. Following incubation, a loopful of cultured EC broth was inoculated onto MacConkey agar (Becton Dickinson) and incubated at 37°C overnight. Six lactose-fermenting colonies were subcultured for purity onto secondary MacConkey plates. Once purified, each of the six presumptive E. coli isolates was plated onto tryptic soy agar (Becton Dickinson) and incubated at 37°C overnight. Biochemical confirmation was conducted using an indole spot reagent test (Remel, Inc., Lenexa, KS) and a citrate utilization test using citrate agar (Becton Dickinson). Once confirmed, at most five positive E. coli isolates per raccoon examined were preserved in 2 ml of brucella broth (Becton Dickinson) with 50% glycerol and stored at −80°C.
Susceptibility testing.
Three E. coli isolates from each fecal sample were submitted to the Laboratory for Food-Borne Zoonoses (Guelph, Ontario, Canada) for susceptibility testing. MICs for the following antimicrobial agents, representing six antimicrobial classes (β-lactams [ampicillin, amoxicillin-clavulanic acid, cefoxitin, ceftiofur, and ceftriaxone], aminoglycosides [streptomycin, kanamycin, gentamicin, and amikacin], tetracyclines [tetracycline], phenicols [chloramphenicol], inhibitors of the folic acid pathway [sulfisoxazole and trimethoprim-sulfamethoxazole], and quinolones [nalidixic acid and ciprofloxacin]), were determined using broth microdilution (NCCLS/CLSI document M7-A7) (Sensititre system; TREK Diagnostics, Cleveland, OH) in accordance with the protocols of the Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS) (7). Isolates were classified as susceptible, intermediate, or resistant based on standardized breakpoints used by the CIPARS (7). For this study, we considered all isolates classified as intermediate or resistant to have reduced susceptibility.
Antimicrobial resistance gene testing.
Regardless of resistance pattern, all resistant isolates were tested for the presence of major antimicrobial resistance genes by use of single and multiplex PCR as described previously (11, 12, 26). These genes were as follows: sul1, sul2, and sul3 for inhibitors of the folic acid pathway; tet(A), tet(B), and tet(C) for tetracycline; aadA, aadB, aphA1, aphA2, aac(3)IV, and strA and strB for aminoglycosides; and blaTEM, blaSHV, and blaCMY-2 for β-lactams.
Serotyping.
Isolates from 10 individual raccoons trapped at the zoo site in ≥4 trapping sessions were submitted for serotyping to investigate E. coli persistence over time. Identification of somatic (O) and flagellar (H) antigens was performed by standard agglutination methods that identified O1 to O181 and H1 to H56 (5). The isolates were serotyped at the Public Health Agency of Canada, Laboratory for Food-Borne Zoonoses.
PFGE.
Isolates from four individuals trapped at the zoo site in ≥4 trapping sessions with the same (or unidentified) serotypes during different sessions were submitted for pulsed-field gel electrophoresis (PFGE) to investigate E. coli persistence over time. PFGE was conducted using standard protocols (23). Briefly, after genomic DNA preparation and restriction digestion with XbaI (New England BioLabs, Ipswich, MA), electrophoresis was performed with 1% SeaKem Gold agarose gels (Cambrex Bio Science Rockland, Rockland, ME) in 0.5× Tris-borate-EDTA (Fisher Scientific, Fair Lawn, NJ) containing 200 μM thiourea (Fisher Scientific) at 12°C for 15.5 h. Pulse times started at 1 s and ended at 40 s, with linear ramping, a field of 6 V/cm, and an angle of 120°, in a Bio-Rad CHEF-III electrophoresis unit (Bio-Rad Laboratories, Hercules, CA). PFGE gels were analyzed using BioNumerics software version 5.10 (Applied Maths, Austin, TX). Band matching was performed using a 1.0% position tolerance, and cluster analysis was performed using the Dice similarity coefficient.
Statistical analysis.
Initially, a generalized linear mixed model was fit to the data using Proc GLIMMIX (SAS 9.2; SAS Institute, Inc., Cary, NC). The fixed effects include age, sex, site, and month. All 2-way interactions were included in the initial model (except site by month, because we collected data only during a single month at two sites). Because there were repeated measures on individual animals (nested within sites), there could be an autocorrelation that would need to be accounted for. This was modeled by trying various error structures offered by SAS [ar(1), arh(1), toep, toep(2)-toep(4), toeph, toeph(2)-toeph(4), un, un(2)-un(4)] as well as fitting a variance component model. Based on the Akaike information criterion, there appeared to be no autocorrelation; hence, we used standard logistic regression methods. Any terms that were not significant were removed from the model. We used exact methods, such as Fisher exact tests with exact confidence intervals (CIs) or extensions of Fisher's method (in the cases of tables larger than 2 by 2) applied to standard 2-way contingency tables to obtain P values and CIs. Differences were considered significant if the P value was <0.05.
RESULTS
Capture and E. coli recovery.
One hundred eighty-two fecal samples were collected from 109 apparently healthy raccoons between June and October 2007. Of 59 animals trapped on the grounds of the Toronto Zoo, 33 were trapped on more than one occasion, and 11 animals were trapped in at least four of five consecutive trapping periods (Table 1). At the rural and urban sites, animals were trapped only once.
Table 1.
Recovery rates and frequencies of resistance to ≥1 antimicrobial agent and ≥2 antimicrobial classes among E. coli isolates from raccoons trapped at three sites in Ontario, Canada
| Sampling site (mo) | No. of animals with E. coli/total no. sampled | % recapturesa | No. of isolates tested | No. (%) with resistance to ≥1 antimicrobial agent |
No. (%) with resistance to ≥2 antimicrobial classesb |
||
|---|---|---|---|---|---|---|---|
| Isolates | Animals | Isolates | Animals | ||||
| Rural (October) | 29/30 | NA | 87 | 10 (12) | 5 (17) | 1 (1) | 1 (3) |
| Urban (August) | 16/20 | NA | 48 | 3 (6) | 3 (19) | 1 (2) | 1 (6) |
| Zoo (June) | 18/18 | NA | 54 | 10 (19) | 8 (44) | 4 (7) | 3 (17) |
| Zoo (July) | 35/35 | 29 | 105 | 12c (10) | 9 (26) | 6 (6) | 5 (14) |
| Zoo (August) | 26/27 | 81 | 78 | 18 (23) | 10 (38) | 4 (5) | 3 (12) |
| Zoo (September) | 28/28 | 71 | 82d | 21e (25) | 16 (54) | 6 (7) | 6 (21) |
| Zoo (October) | 23/24 | 87 | 69 | 14f (20) | 11 (48) | 5 (7) | 4 (17) |
| Zoo total | 130/132 | 78 | 388 | 75 (19) | 54 (42) | 25 (6) | 21 (16) |
Percentage of animals sampled each month that had previously been sampled. NA, not applicable.
Antimicrobial classes (sulfonamides, aminoglycosides, tetracyclines, phenicols, β-lactams, and quinolones).
Includes one isolate with intermediate resistance to tetracycline and one isolate with intermediate resistance to chloramphenicol.
One sample yielded only one isolate.
Includes one isolate with intermediate resistance to cefoxitin.
Includes one isolate with intermediate resistance to chloramphenicol.
E. coli was detected in 130 of 132 fecal samples collected from the 59 individual raccoons trapped on the grounds of the Toronto Zoo, 29 of 30 animals from the rural area, and 16 of 20 animals from the predominantly urban site (Table 1). E. coli was detected in four or more consecutive samples from 10 of the 11 animals trapped in at least four of five consecutive trapping periods. A total of 523 E. coli isolates were recovered and submitted for susceptibility testing (one sample yielded only a single isolate) (Table 1).
E. coli susceptibility.
A total of 88 E. coli isolates (17% of 523) from 62 raccoons (35% of 175) showed reduced susceptibility to ≥1 antimicrobial agent. There was no evidence of autocorrelation for samples taken from the same individual at different times; hence, we used standard logistic regression methods to examine associations between reduced susceptibility to ≥1 antimicrobial agent in E. coli isolates from raccoons and age, sex, location, and month. Age and sex were not significantly associated with reduced susceptibility to ≥1 antimicrobial agent (P > 0.4), and there was no significant association between the occurrence of reduced susceptibility to ≥1 antimicrobial agent and month (P = 0.08). However, reduced susceptibility to ≥1 antimicrobial agent was significantly associated with site (P = 0.02). Raccoons from the zoo site were significantly more likely to shed E. coli with reduced susceptibility to ≥1 antimicrobial agent than animals from the rural site (odds ratio [OR], 3.41; 95% CI, 1.17 to 12.09; P = 0.02). The odds of detecting reduced susceptibility to ≥1 antimicrobial agent in E. coli isolates from raccoons at the zoo site was not statistically different from what was observed for the urban site (OR, 3.08; 95% CI, 0.79 to 17.53; P = 0.10), and there were no differences in the frequencies of reduced susceptibility to ≥1 antimicrobial agent in raccoons between the rural and urban sites (OR, 1.11; 95% CI, 0.15 to 6.79; P = 1.00).
Multiclass resistance (reduced susceptibility to ≥2 antimicrobial classes) occurred in E. coli isolates from raccoons at all sites (3%, 6%, and 16% of raccoons in rural, urban, and zoo sites, respectively) (Table 1). There was no significant difference in the frequencies of multiclass resistance between sites (P = 0.14).
E. coli isolates from 14%, 19%, and 39% of raccoons from the rural, urban, and zoo sites, respectively, were resistant to tetracycline, making this resistance phenotype the most commonly detected one in this study (Table 2). Resistance to tetracycline was significantly associated with site (P = 0.01). Raccoons from the zoo site were significantly more likely to shed E. coli with reduced susceptibility to tetracycline than animals from the rural site (OR, 4.00; 95% CI, 1.35 to 13.25; P = 0.009); however, there was no difference in the frequencies of reduced susceptibility to tetracycline in E. coli isolates from raccoons at the urban and zoo sites (OR, 0.36; 95% CI, 0.08 to 1.36; P = 0.17) or between the rural and urban sites (OR, 0.70; 95% CI, 0.13 to 4.04; P = 0.69).
Table 2.
Frequencies of antimicrobial resistance in E. coli isolates from raccoons trapped at three sites in Ontario, Canada
| Antimicrobial agent(s) | No. (%) resistant in indicated area and mo |
|||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Rural, October |
Urban, August |
Zoo grounds |
||||||||||||
| June |
July |
August |
September |
October |
||||||||||
| Isolates (n = 87) | Animals (n = 29) | Isolates (n = 48) | Animals (n = 16) | Isolates (n = 57) | Animals (n = 18) | Isolates (n = 105) | Animals (n = 35) | Isolates (n = 78) | Animals (n = 26) | Isolates (n = 82) | Animals (n = 28) | Isolates (n = 69) | Animals (n = 23) | |
| Amoxicillin-clavulanic acid | 1 (1) | 1 (3) | 1 (2) | 1 (6) | 1 (2) | 1 (6) | 1 (1) | 1 (3) | 0 | 0 | 1 (1) | 1 (4) | 1 (1) | 1 (4) |
| Ampicillin | 1 (1) | 1 (3) | 1 (2) | 1 (6) | 4 (7) | 3 (17) | 1 (1) | 1 (3) | 1 (1) | 1 (4) | 3 (4) | 3 (11) | 2 (3) | 2 (9) |
| Cefoxitin | 1 (1) | 1 (3) | 0 | 0 | 1 (2) | 1 (6) | 1 (1) | 1 (3) | 0 | 0 | 1 (1) | 1 (4) | 1 (1) | 1 (4) |
| Ceftiofur | 1 (1) | 1 (3) | 0 | 0 | 1 (2) | 1 (6) | 1 (1) | 1 (3) | 0 | 0 | 0 | 0 | 0 | 0 |
| Ceftriaxone | 1 (1) | 1 (3) | 0 | 0 | 0 | 1 (6) | 1 (1) | 1 (3) | 0 | 0 | 0 | 0 | 0 | 0 |
| Chloramphenicol | 0 | 0 | 0 | 0 | 0 | 0 | 1 (1) | 1 (3) | 0 | 0 | 0 | 0 | 2 (3) | 2 (9) |
| Ciprofloxacin | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 (1) | 1 (4) |
| Gentamicin | 0 | 0 | 1 (2) | 1 (6) | 0 | 0 | 0 | 0 | 0 | 1 (1) | 1 (4) | 0 | 0 | |
| Nalidixic acid | 0 | 0 | 1 (2) | 1 (6) | 1 (2) | 1 (6) | 0 | 0 | 0 | 0 | 0 | 0 | 1 (1) | 1 (4) |
| Streptomycin | 0 | 0 | 0 | 0 | 3 (5) | 2 (11) | 2 (2) | 2 (6) | 1 (1) | 1 (4) | 2 (2) | 2 (7) | 2 (3) | 2 (9) |
| Sulfisoxazole | 0 | 0 | 1 (2) | 1 (6) | 3 (5) | 2 (11) | 4 (4) | 3 (9) | 2 (3) | 1 (4) | 3 (4) | 33 (11) | 4 (6) | 3 (13) |
| Tetracycline | 9 (10) | 4 (14) | 3 (6) | 3 (19) | 10 (18) | 8 (44) | 11 (10) | 8 (23) | 18 (23) | 10 (38) | 19 (23) | 15 (54) | 12 (17) | 10 (43) |
| Trimethoprim-sulfamethoxazole | 0 | 0 | 1 (2) | 1 (6) | 1 (2) | 1 (6) | 1 (1) | 1 (3) | 2 (3) | 1 (4) | 2 (2) | 2 (7) | 4 (6) | 3 (13) |
We detected reduced susceptibility to four antimicrobials that are considered to be of very high importance to human health by Health Canada (9) in E. coli isolates from raccoons in this study. Reduced susceptibility to amoxicillin-clavulanic acid was detected in E. coli isolates from animals at all sites, and reduced susceptibilities to cefoxitin, ceftiofur, and ceftriaxone were detected in E. coli isolates from animals at the rural and zoo sites, while reduced susceptibility to ciprofloxacin was detected in an E. coli isolate from the zoo site only. There were no significant differences in the frequencies of resistance to any of these antimicrobials or any other antimicrobial tested among sites (P ≥ 0.25).
Isolates from six raccoons treated with penicillin in June and sampled in July were pan-susceptible (the other three animals were not trapped and sampled in July).
Resistance genes.
Resistance genes were detected in 84 of 86 phenotypically resistant isolates tested (Table 3). The most commonly detected resistance genes were tet(A) and tet(B), which were detected in 54% and 36% of resistant isolates, respectively (Table 3). The blaCMY-2 gene, which codes for resistance to ceftiofur, was detected in E. coli isolates from raccoons at both the rural and the zoo sites. The sul3, blaSHV, aadB, aphA1, aphA2, and aac(3)IV genes were not detected in any isolates. Because we have only a small number of samples at the urban and rural sites, we did not have enough statistical power to make meaningful comparisons of the frequencies of resistance genes in E. coli isolates from raccoons between sites.
Table 3.
Frequencies of antimicrobial resistance genes in resistant E. coli isolates from raccoons trapped at three sites in Ontario, Canada
| AMR gene(s)a | No. positive in indicated area and mo |
|||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Rural, October |
Urban, August |
Zoo grounds |
||||||||||||
| June |
July |
August |
September |
October |
||||||||||
| Isolates (n = 10) | Animals (n = 5) | Isolates (n = 3) | Animals (n = 3) | Isolates (n = 10) | Animals (n = 8) | Isolates (n = 12) | Animals (n = 9) | Isolates (n = 18) | Animals (n = 10) | Isolates (n = 20b) | Animals (n = 15b) | Isolates (n = 13c) | Animals (n = 11) | |
| blaCMY-2 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 |
| blaTEM | 0 | 0 | 1 | 1 | 3 | 2 | 0 | 0 | 1 | 1 | 2 | 2 | 2 | 1 |
| strA and strB | 2 | 1 | 0 | 0 | 4 | 3 | 2 | 2 | 1 | 1 | 2 | 2 | 3 | 2 |
| aadA | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 | 1 | 1 | 0 | 0 |
| sul1 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 2 | 1 | 1 | 1 | 3 | 2 |
| sul2 | 0 | 0 | 0 | 0 | 4 | 3 | 3 | 2 | 1 | 1 | 2 | 2 | 3 | 2 |
| tet(A) | 3 | 1 | 3 | 3 | 5 | 3 | 7 | 5 | 6 | 4 | 15 | 12 | 8 | 8 |
| tet(B) | 0 | 0 | 0 | 0 | 5 | 5 | 3 | 3 | 12 | 6 | 5 | 5 | 6 | 5 |
| tet(C) | 3 | 2 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 |
Resistance genes sul3, blaSHV, aadB, aphA1, aphA2, and aac(3)IV were not detected in any isolates.
One isolate with intermediate resistance to cefoxitin was not tested.
One isolate with intermediate resistance to chloramphenicol was not tested.
Serotyping.
Eighty-one unique E. coli serotypes were detected in 126 isolates from 42 samples (3 isolates per sample) collected from 10 raccoons that were positive for E. coli in four or more consecutive sessions on zoo grounds. Three distinct serotypes were detected in the majority of samples (25 of 42). Three identical serotypes were detected in 4 of 42 samples, and two serotypes were detected in 13 of 42 samples (Table 4).
Table 4.
Serotypes and antibiotic resistance and PFGE profiles of E. coli isolates from raccoons caught in ≥4 consecutive trapping sessions on the grounds of the Toronto Zoo
| Animal | Serotype (antibiotic resistance profile and PFGE profile)a for indicated mo |
||||
|---|---|---|---|---|---|
| June | July | August | September | October | |
| 504 | O?:H7 (a) | O5:H11 (a) | O22:H16 (b) | O1:H25 (a, I) | O1:H25 (a, I) |
| O18ac:H7 (a) | O5:H11 (a) | O22:H16 (b) | O?:H10 (c, I) | O19:H10 (a, IV) | |
| O6:H16 (b) | O29:H25 (a) | O175:H15 (a) | O?:H10 (a, III) | O19:H10 (a, IV) | |
| 507 | O8:H19 (a) | O22:H16 (b) | O?:H32 (b) | 15a O41:NM (a) | |
| O8:H19 (a) | O22:H16 (b) | O?:H4 (a) | O156:H30 (b) | ||
| O8:H19 (a) | O22:H16 (b) | O103:H21 (a) | O172:H45 (a) | ||
| 508 | O124:H18 (a, V) | O128:H20 (a) | O29:H5 (a) | O?:H42 (b) | O?:H4 (a, VIII) |
| O124:H18 (a, V) | O?:H18 (a, VII) | O29:H5 (a) | O?:H48 (b) | O71:H53 (b) | |
| O?:H4 (a, VI) | O5:H11 (a) | O29:H5 (a) | O?:H32 (a) | O141:H5 (a) | |
| 514 | O?:H10 (c) | O124:H18 (a) | O?:H9 (d) | O100:NM (a) | |
| O15:H11 (d) | O8:H20 (a) | O106:NM (a) | O71:H53 (b) | ||
| O87:H7 (a) | O22:H16 (b) | O88:H16 (a) | O100:NM (a) | ||
| 515 | O2:H56 (a) | O6:H7 (a) | O73:H1 (a) | O71:H11 (b) | |
| O29:H25 (a) | O?:H2 (a) | O4:H45 (a) | O84:H14 (a) | ||
| O91:H9 (b) | O?:H23 (e) | O4:H45 (a) | O?:H14 (a) | ||
| 516 | O9:H21 (a, IX) | O159:H28 (a) | O71:NM (b) | O68:H12 (a) | |
| O4:H5 (a) | O?H21 (a, X) | O55:H12 (b) | O1:H5 (a) | ||
| O4:H5 (a) | O6:H1 (f) | O?:H36 (a) | O?:H27 (a) | ||
| 526 | O29:H25 (a) | O?:H34 (a) | O102:H6 (g) | O8:NM (d) | |
| O?:H23 (a) | O175:H15 (a) | O6:H7 (a) | O9:H19 (a) | ||
| O?:H23 (a) | O175:H15 (a) | O?:H16 (b) | O126:H27 (a) | ||
| 527 | O27:H? (a) | O?:H16 (b, XI) | O9:NM (a) | O93:H28 (a) | |
| O126:H21 (a) | O?:H16 (b, XI) | O?:NM (a) | O22:H16 (b, XI) | ||
| O36:H42 (a) | O?:H16 (b, XI) | O9:NM (a) | O136:H18 (a) | ||
| 532 | O36:H42 (a) | O111:H9 (b) | O18ac:H45 (a) | O158:H23 (a) | |
| O84:H14 (a) | O73:H10 (a) | O18ac:H45 (a) | O18:acH7 (a) | ||
| O45:H9 (a) | O157:NM (b) | O9:H21 (a) | O43:NM (a) | ||
| 543 | O?:H32 (h) | O7:H4 (a) | O180:H14 (a) | O71:H53 (h) | |
| O66:H10 (a) | O7:H? (a) | O102:H6 (a) | O21:H15 (a) | ||
| O117:H10 (a) | O7:H4 (a) | O136:H18 (a) | O8:H9 (a) | ||
Antibiotic resistance profiles (a to h) and PFGE profiles (I to XI) have been given arbitrary designations. Underlining shows the isolates tested using PFGE profiles. Boldface highlights two cases where the same serotypes were found in the same animal in different trapping sessions.
Nine identical serotypes were detected in different animals during the same trapping session, 11 identical serotypes were detected in different animals in different sessions, and 2 identical serotypes were detected in the same animal in different sessions (Table 4).
PFGE.
Seventeen isolates from 4 individuals trapped at the zoo site in ≥4 trapping sessions with the same (or unidentified) serotypes during different sessions were submitted for PFGE typing (Table 4). Isolates identified as serotype O1:H25 from individual 504 sampled in session 4 and 5 had identical PFGE profiles; however, isolates identified as serotype O?:H4 from individual 508 in sessions 1 and 5 had distinct PFGE profiles (Table 4). Isolates identified as O?:H16 in animal 527 sampled in session 3 had the same PFGE profile as isolates identified as O22:H16 from the same animal in session 5 (Table 4). The PFGE profiles in three other cases, where similar serotypes were found in the same individual in different sessions, were distinct, with more than nine band differences.
DISCUSSION
The prevalences of AMR in fecal E. coli isolates from raccoons in this study ranged from 17% to 54%, depending on location and month. These values are consistent with previous studies looking at the prevalences of AMR in wildlife in Ontario and around the world (5% to 48% in Ontario, Canada [2, 12], 13.7% in the Czech Republic [14], and 20.1% in the United Kingdom [28]). We found that the prevalence of AMR in fecal E. coli isolates from raccoons in rural areas was lower than what has been reported for small mammals (e.g., mice, rats, and chipmunks) living on swine farms (48%) and in contrast with previous studies, and we detected no significant difference in the prevalences of resistance between urban and rural environments (2). We did not trap raccoons directly on farm sites, so the raccoons trapped in rural areas for this study may not have used livestock farms as part of their home range and therefore may not have been exposed to the same conditions as small mammals from those previous studies.
We were significantly more likely to detect resistance to ≥1 antimicrobial agent in E. coli isolates from raccoons at the zoo site than in isolates from raccoons at the rural site. Antimicrobial resistance is generally thought to emerge in bacterial populations as a consequence of selective pressure resulting from exposure to antimicrobials. Wild animals are unlikely to be deliberately exposed to antimicrobials at any location; however, tetracycline-containing oral vaccination baits have been used in a number of locations across Ontario to control rabies (21). An area near the zoo site was baited from 1989 to 1991 (21), and an area near the rural site was baited immediately prior to and during the period of this study (20). The impact of this type of exposure to tetracycline on the development of resistance is unknown; however, it is interesting to note that tetracycline resistance was detected in E. coli isolates from raccoons significantly less frequently in the rural area that was baited most recently than at the zoo site. These results suggest that oral baiting with tetracycline was unlikely to be a primary factor associated with the occurrence of resistance in E. coli isolates from raccoons in this study.
Wild animals may be unintentionally exposed to antimicrobials at the zoo site via contact with medicated feed/water or via exposure to waste from treated zoo animals. This may be particularly likely with raccoons, with their scavenger behavior. Antimicrobial-resistant bacteria have been isolated from zoo animals (1, 6) and could be the source of the resistant isolates obtained in this study from raccoons. However, we detected no difference in the frequencies of reduced susceptibility to ≥1 antimicrobial agent in fecal E. coli isolates from raccoons between the urban and zoo sites. Previous studies have shown that some populations of wild animals in frequent contact with human habitations and refuse had a significantly greater prevalence of antimicrobial-resistant bacteria than populations of the same species without access to human refuse (19). Raccoons living at the zoo and urban sites would likely have had more access to human refuse than raccoons living at the more rural site; however, because we looked only at a small number of raccoons at single urban, rural, and zoo sites in this study, further studies are required to determine if the prevalences of AMR in E. coli isolates from raccoons differ between environments.
Serotyping and PFGE typing of E. coli isolates from raccoons that were part of the longitudinal study at the zoo site showed that, similar to other animals, including humans (3, 25), individual raccoons can concurrently shed a diverse array of E. coli strains. In addition, we found only two instances where the same serotype of E. coli was detected in the same individual in different months, indicating that there is probably a rapid turnover of E. coli serotypes within individual animals. A previous study conducted in the same population of animals showed similar patterns of high turnover and only short-term shedding with Salmonella serotypes (10). Thus, the results of this study suggest that raccoons are not likely to maintain specific serotypes of E. coli, including resistant serotypes, over long periods. However, they can shed resistant E. coli for short periods and therefore have the potential to spread resistant bacteria throughout their environment. Further studies investigating the degree of dissemination of AMR into the environment by raccoons are required to determine if raccoons play a key role in disseminating AMR into the environment.
We detected resistance to amoxicillin-clavulanic acid in E. coli isolates from raccoons at all sites and resistance to expanded-spectrum cephalosporins (with the corresponding blaCMY-2 gene) in E. coli isolates from raccoons trapped at the zoo and rural sites. These antimicrobials are important therapeutic agents in veterinary and human medicine and are considered to be of very high importance to human health by Health Canada (9).
There is evidence that the plasmids carrying the blaCMY-2 gene can be transferred between different bacterial species (16) and between food animals and humans (29). Raccoons are commonly colonized with Salmonella in some areas (10), so there is the potential for transmission of resistant determinants from E. coli to pathogenic bacteria, such as Salmonella, that pose a direct threat to human and animal health. In addition, because raccoons often live in close association with humans and livestock, there is the potential for raccoons to be involved in the transfer of resistance determinants between bacteria of different host species.
The findings of this study show that raccoons can shed and have the potential to disseminate resistant bacteria, including bacteria with resistance to antimicrobials considered to be of very high importance to human health, at least over the short term. In addition, the results of this study indicate that the potential role of wildlife in the complex epidemiology of AMR should be taken into account as control and management strategies are developed to reduce the burden associated with antimicrobial resistance for human, animal, and environmental health.
ACKNOWLEDGMENTS
We thank K. Winger, A. Nguyen, the staff and veterinarians at the Toronto Zoo, and the staff at the Ontario Ministry of Natural Resources for assistance with trapping and processing animals. We also thank B. Jefferson and students for assistance with sample processing. We thank W. Sears for assistance with statistical analyses. The Canadian Research Institute for Food Safety and the Laboratory for Food-Borne Zoonoses provided laboratory support.
Footnotes
Published ahead of print 23 March 2012
REFERENCES
- 1. Ahmed AM, et al. 2007. Zoo animals as reservoirs of gram-negative bacteria harboring integrons and antimicrobial resistance genes. Appl. Environ. Microbiol. 73:6686–6690 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Allen S, et al. 2011. Antimicrobial resistance in generic Escherichia coli isolated from wild small mammals living in farm, residential, landfill and natural environments in southern Ontario. Appl. Environ. Microbiol. 77:882–888 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Bettleheim KA, Kuzevski A, Gilbert RA, Krause DO, McSweeney CS. 2005. The diversity of Escherichia coli serotypes and biotypes in cattle feces. J. Appl. Microbiol. 98:699–709 [DOI] [PubMed] [Google Scholar]
- 4. Dolejska M, Cizek A, Literak I. 2007. High prevalence of antimicrobial-resistant genes and integrons in Escherichia coli isolates from black-headed gulls in the Czech Republic. J. Appl. Microbiol. 103:11–19 [DOI] [PubMed] [Google Scholar]
- 5. Ewing EH. 1986. Genus Escherichia, p 96–134 In Edwards PR, Ewing WH. (ed), Identification of Enterobacteriaceae, 4th ed. Elsevier Science, New York, NY [Google Scholar]
- 6. Gopee NV, Adesiyun AA, Caesar K. 2000. A longitudinal study of Escherichia coli strains isolated from captive mammals, birds, and reptiles in Trinidad. J. Zoo Wildl. Med. 31:353–360 [DOI] [PubMed] [Google Scholar]
- 7. Government of Canada 2011. Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS) 2008. Public Health Agency of Canada, Guelph, Ontario, Canada [Google Scholar]
- 8. Guenther S, et al. 2010. First insights into antimicrobial resistance among faecal Escherichia coli isolates from small wild mammals in rural areas. Sci. Total Environ. 408:3519–3522 [DOI] [PubMed] [Google Scholar]
- 9. Health Canada 24 June 2011, accession date Categorization of antimicrobial drugs based on importance in human medicine. April 2009. Health Canada, Ottawa, Ontario, Canada: http://www.hc-sc.gc.ca/dhp-mps/consultation/vet/consultations/amr_ram_hum-med-rev-eng.php [Google Scholar]
- 10. Jardine C, Reid-Smith RJ, Janecko N, Allan M, McEwen SA. 2011. Salmonella in raccoons (Procyon lotor) in southern Ontario, Canada. J. Wildl. Dis. 47:344–351 [DOI] [PubMed] [Google Scholar]
- 11. Khashayar B. 2009. Prevalence of antimicrobial resistance and of selected beta-lactams resistance genes in Escherichia coli isolates from canine urinary tract infections in Ontario and Quebec, p 35–41 M.S. thesis University of Guelph, Guelph, Ontario, Canada [Google Scholar]
- 12. Kozak GK, Boerlin P, Janecko N, Reid-Smith RJ, Jardine C. 2009. Antimicrobial Resistance in Escherichia coli isolates from swine and wild small mammals in the proximity of swine farms and in natural environments in Ontario, Canada. Appl. Environ. Microbiol. 75:559–566 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Kummerer K. 2004. Resistance in the environment. J. Antimicrob. Chemother. 54:311–320 [DOI] [PubMed] [Google Scholar]
- 14. Literák I, Vanko R, Dolejská M, Cízek A, Karpísková R. 2007. Antibiotic resistant Escherichia coli and Salmonella in Russian rooks (Corvus frugilegus) wintering in the Czech Republic. Lett. Appl. Microbiol. 45:616–621 [DOI] [PubMed] [Google Scholar]
- 15. Madsen M, Hangartner P, West K, Kelly P. 1998. Recovery rates, serotypes, and antimicrobial susceptibility patterns of salmonellae isolated from cloacal swabs of wild Nile crocodiles (Crocodylus niloticus) in Zimbabwe. J. Zoo Wildl. Med. 29:31–34 [PubMed] [Google Scholar]
- 16. Martin LC, Weir EK, Poppe C, Reid-Smith RJ, Boerlin P. 2012. Characterization of blaCMY-2 plasmids in Salmonella and Escherichia coli isolates from food animals in Canada. Appl. Environ. Microbiol. 78:1285–1287 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Miranda CD, Zemelman R. 2001. Antibiotic resistant bacteria in fish from the Concepción Bay, Chile. Mar. Pollut. Bull. 42:1096–1102 [DOI] [PubMed] [Google Scholar]
- 18. Mølbak K. 2004. Spread of resistant bacteria and resistance genes from animals to humans—the public health consequences. J. Vet. Med. B 51:364–369 [DOI] [PubMed] [Google Scholar]
- 19. Rolland RM, Hausfater G, Marshall B, Levy SB. 1985. Antibiotic-resistant bacteria in wild primates: increased prevalence in baboons feeding on human refuse. Appl. Environ. Microbiol. 49:791–794 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Rosatte RC, et al. 2009. Aerial distribution of ONRAB baits as a tactic to control rabies in raccoons and striped skunks in Ontario, Canada. J. Wildl. Dis. 45:363–374 [DOI] [PubMed] [Google Scholar]
- 21. Rosatte RC, Power MJ, MacInnes CD, Campbell JB. 1992. Trap-vaccinate-release and oral vaccination for rabies control in urban skunks, raccoons and foxes. J. Wildl. Dis. 28:562–571 [DOI] [PubMed] [Google Scholar]
- 22. Rosatte R, et al. 2010. Density, movements, and survival of raccoons in Ontario, Canada: implications for disease spread and management. J. Mammal. 91:122–135 [Google Scholar]
- 23. Sambrook J, Russel DW. 2001. Molecular cloning, a laboratory manual, p 5.55–5.82 Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY [Google Scholar]
- 24. Sayah RS, Kaneene JB, Johnson Y, Miller R. 2005. Patterns of antimicrobial resistance observed in Escherichia coli isolates obtained from domestic- and wild-animal fecal samples, human septage, and surface water. Appl. Environ. Microbiol. 71:1394–1404 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Shooter RA, Bettelheim KA, Lennox-King SMJ, O'Farrell S. 1977. Escherichia coli serotypes in the faeces of health adults over a period of several months. J. Hyg. (Lond.) 78:95–98 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Travis RM, et al. 2006. Chloramphenicol and kanamycin resistance among porcine Escherichia coli in Ontario. J. Antimicrob. Chemother. 58:173–177 [DOI] [PubMed] [Google Scholar]
- 27. Reference deleted.
- 28. Williams NJ, et al. 2011. The prevalence of antimicrobial-resistant Escherichia coli in sympatric wild rodents varies by season and host. J. Appl. Microbiol. 110:962–970 [DOI] [PubMed] [Google Scholar]
- 29. Winokur PL, Vonstein DL, Hoffman LJ, Uhlenhopp EK, Doern GV. 2001. Evidence for transfer of CMY-2 AmpC β-lactamase plasmids between Escherichia coli and Salmonella isolates from food animals and humans. Antimicrob. Agents Chemother. 45:2716–2722 [DOI] [PMC free article] [PubMed] [Google Scholar]