Antimicrobial Resistance and Resistance Genes in Escherichia coli Isolated from Retail Meat Purchased in Alberta,Canada

Abstract

This study analyzed antimicrobial resistance (AMR) and resistance genes in generic Escherichia coli isolated from retail meat samples purchased (2007–2008) in Alberta, Canada, and determined potential associations between resistance phenotypes and resistance genes with relation to the meat types. A total of 422 E. coli isolates from retail chicken, turkey, beef, and pork meats were tested for antimicrobial susceptibility. Multiplex PCRs were used to detect major resistance genes for tetracyclines [tet(A), tet(B), tet(C)], sulfonamides (sul1, sul2, sul3), aminoglycosides (strA/B, aadA, aadB, aac(3)IV, aphA1, aphA2), and β-lactamase (bla _CMY-2 , bla _TEM , bla _SHV , bla _PSE-1). Resistance to ciprofloxacin was not found in any isolate. Overall resistances to clinically important antimicrobials amoxicillin-clavulanic acid (16.8% of isolates) and ceftriaxone (12.6% isolates) were observed. These resistances were observed more frequently (p<0.0001) in chicken-derived E. coli than those from the other meat types. Resistance to multiple antimicrobials (≥5) was found in more chicken derived E. coli (32%) than E. coli from other meat types. The β-lactamase genes of clinical importance, including bla _CMY-2 and bla _TEM, were found in about 18% of poultry-derived E. coli and in only 5% of ground beef. The bla _CMY-2 gene was more likely to be found in E. coli from chicken than turkey, beef, or pork meats. The tet(A) gene was associated with bla _CMY-2, whereas bla _CMY-2 and bla _TEM genes were associated with strA/B genes. Resistance genes for tetracycline, sulfonamides, and aminoglycosides were associated with the phenotypic expression of resistance to unrelated classes of antimicrobials. These data suggest the prevalence of AMR and select resistance genes were higher in poultry-derived E. coli. The multiple associations found between AMR phenotypes and resistance genes suggest a complex nature of resistance in E. coli from retail meat, and hence the use of a single antimicrobial could result in the selection of resistant E. coli not only to the drug being used but to other unrelated classes of antimicrobials as well.

Introduction

T he emergence of antimicrobial resistance among meatborne pathogens and commensals has become an issue in meat safety that poses a serious public health risk. Resistance monitoring programs in various countries have generally focused on analyzing phenotypic antimicrobial resistance (AMR) in bacteria, and thus AMR prevalence in bacteria isolated from retail meat has been well documented (Chen et al., 2004; CIPARS, 2009; Li et al., 2011; M'ikanatha et al., 2010; NARMS, 2008; Schroeder et al., 2003; Yang et al., 2010). A few studies from Europe, Asia, and United States have shown the presence of resistance genes in meatborne E. coli in order to determine the genetic mechanisms underlying the prevalence of AMR (Sunde et al., 2006; Yan et al., 2004; Zhao et al., 2009).

Commensal E. coli can serve as a reservoir of resistance genes with the ability to transfer these genes to pathogens in the hosts as well as in the human intestinal tract after the consumption of contaminated foods of animal origin (Blake et al., 2003). Furthermore, a number of studies have demonstrated the transfer of antimicrobial resistance between commensal bacteria and zoonotic pathogens in various ecological environments (Mathew et al., 2009; Poppe et al., 2005; Walsh et al., 2008).

With the use of β-lactam antibiotics in animal production E. coli carrying extended-spectrum β-lactamase (ESBL) genes has been emerging rapidly (Liu et al., 2007). A higher prevalence of ESBL genes in E. coli isolated from retail chicken meat has been reported, and these strains showed a higher degree of similarity to human isolates (Overdevest et al., 2011). Therefore, the presence of β-lactamase genes coding for resistance to extended spectrum cephalosporins in E. coli for instance may represent a potential human health risk (Randall et al., 2011). With the changing epidemiology of ESBL genes in E. coli and geographic difference in their distribution, information is needed about the prevalence of β-lactamase genes in E. coli from retail meats. Furthermore, for better understanding the epidemiology of AMR in meatborne bacteria, information about the genes coding for resistances to other clinically important antimicrobials is also needed.

Genotyping shows higher diversity than phenotypes and consequently allows for more accurate comparisons between resistant bacterial populations (Gow et al., 2008; Rosengren et al., 2009). To date, most information about the distribution of resistance genes in E. coli and statistical association between AMR phenotypes and resistance genes in Canada has been reported from animals (Kozak et al., 2009a; Boerlin et al., 2005; Rosengren et al., 2009). Different livestock management practices, exposure to antimicrobials, and other environmental factors may affect the frequency of AMR and genes that may found in E. coli from retail meat. Therefore, in addition to resistance phenotypes data, information is also needed about resistance genes found in E. coli isolated from retail meat available for purchase and consumption by consumers.

Therefore, the objectives of this study were to assess the prevalence of AMR and major resistance genes in E. coli isolated from retail meat samples purchased in Alberta, Canada, to determine unconditional statistical associations between resistance phenotypes and genes in relation to meat types.

Methods

Sampling and E. coli isolates

The sampling strategy established by the Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS) was used (CIPARS, 2007). This sampling plan involved continuous weekly sampling over a period of 1 year (2007–2008), from 19 census divisions in Alberta representing 40 sampling days. A total of 564 samples, including chicken (n=206), turkey (n=91), beef (n=134), and pork (n=133), were purchased from retail stores across Alberta, Canada. Samples were collected aseptically in sterile plastic bags, and transported on ice to the laboratory for isolation and identification of E. coli.

Standard protocols of CIPARS (CIPARS, 2007) were followed for sample handling, preparation, and isolation procedures. E. coli isolates considered positive by conventional cultural and biochemical tests were also confirmed by a polymerase chain reaction (PCR) using uspA gene, as described by Chen and Griffiths (1998). One confirmed E. coli isolate from each positive sample was used for further analysis.

Antimicrobial susceptibility testing

The minimum inhibitory concentration (MIC) values were determined using an automated broth microdilution method (Sensititre^®; Trek Diagnostic Systems Inc., Westlake, OH) with CMV1AGNF plates. Results were interpreted according to the guidelines set by the Clinical and Laboratory Standards Institute (CLSI, 2010). Escherichia coli American Type Culture Collection (ATCC) 25922, Pseudomonas aeruginosa ATCC 27853, and Staphylococcus aureus ATCC 29213 were used as quality control organisms.

Antimicrobial susceptibility testing was performed for the following antimicrobial agents (antimicrobial abbreviations and breakpoints (CLSI) are shown in parentheses): amoxicillin/clavulanic acid (AMC; ≥32/16 μg/mL), ceftiofur (TIO; ≥8 μg/mL), ceftriaxone (CRO; ≥4 μg/mL), ciprofloxacin (CIP; ≥4 μg/mL), amikacin (AMK; ≥64 μg/mL), ampicillin (AMP; ≥32 μg/mL), cefoxitin (FOX; ≥32 μg/mL), gentamicin (GEN; ≥16 μg/mL), kanamycin (KAN; ≥64 μg/mL), nalidixic acid (NAL; ≥32 μg/mL), streptomycin (STR; ≥64 μg/mL), trimethoprim/sulfamethoxazole (SXT; ≥4/76 μg/mL), chloramphenicol (CHL; ≥32 μg/mL), sulfisoxazole (FIS; ≥512 μg/mL), and tetracycline (TET; ≥16 μg/mL). These antimicrobials are representative of the major classes of antimicrobial drugs important to both veterinary and human medicine.

Detection of antimicrobial resistance genes

A set of multiplex PCRs was used for identifying major resistance genes following the procedure described by Kozak et al. (2009a). The major genes conferring resistance for tetracycline [tet(A), tet(B), tet(C)], streptomycin (strA, strB, aadA), kanamycin and neomycin (aphA1, aphA2), gentamicin and apramycin (aac(3)IV), kanamycin and gentamicin (aadB), sulfonamides (sul1, sul2, and sul3), and β-lactamases (bla _CMY-2, bla _TEM, bla _SHV , bla_PSE-1) were targeted. Appropriate positive and negative controls were included in each PCR run.

Statistical analysis

Descriptive analyses and unconditional univariable logistic regression were used to determine prevalence and examine associations between resistance genes, resistance phenotypes, and meat types using SPSS 17.0 for Windows (SPSS Inc., Chicago, IL). To examine unconditional associations between resistance genes, outcome and predictor variables included resistance genes with a prevalence of over 5%. Unconditional associations were examined for resistance phenotypes with a prevalence of over 5% and meat type as predictor variables.

Model specifications included a binomial distribution and logit link function. Associations that were statistically significant were reported as odd ratios (OR) with 95% confidence intervals (CI). Bonferroni corrections were used as a conservative adjustment for multiple comparisons setting the level of statistical significance at p<0.05/n, where n is the number of comparisons made for each outcome (Gow et al., 2008). Chi-squared analysis (PROC FREQ/CHISQ EXACT; SAS version 9.2 for Windows, SAS Institute, Cary, NC) was used to test differences among AMR frequencies for each meat type with a p value of <0.05 to declare significance.

Results

Prevalence of resistance in E. coli

Escherichia coli were isolated from 96%, 86%, 82%, and 40% of chicken, ground turkey, beef, and pork samples, respectively. The majority of E. coli (81%) from beef, and about 28%, 36%, and 58% of E. coli from chicken, turkey, and pork, respectively, were susceptible to all tested antimicrobials (Table 1). Resistance to CIP was not found in any E. coli isolate, whereas amikacin (AMK) resistance was found only in 2.4% of E. coli isolated from pork. Resistance to antimicrobials very important in human medicine (AMC, CRO) was higher (p<0.0001) in E. coli isolates from chicken than isolates from the other three meat types (Table 1). Resistance to ≥2 antimicrobials was found in about 55% of E. coli from chicken, 44% of E. coli from turkey, 14% of E. coli from beef, and 32% of E. coli from pork meats (Table 1). Resistance to at least five antimicrobials was found in more E. coli from chicken (32%) than E. coli from other meat types.

Table 1.

Prevalence of Antimicrobial Resistance and Multiple Resistances in Escherichia coli Recovered from Retail Meats

Category^a of antimicrobials	Antimicrobials	Overall,n=422 (%)	Beef,n=110 (%)	Chicken,n=193 (%)	Pork,n=41 (%)	Turkey,n=78 (%)	p-value
I	AMC	71 (16.8)	3 (2.7)	60 (31.1)	0 (0)	8 (10.3)	<0.0001
I	TIO	53 (12.6)	2 (1.8)	43 (22.3)	0 (0)	8 (10.3)	<0.0001
I	CRO	66 (15.6)	2 (1.8)	55 (28.5)	1 (2.4)	8 (10.3)	<0.0001
I	CIP	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	NC
II	AMK	1 (0.2)	0 (0)	0 (0)	1 (2.4)	0 (0)	NC
II	AMP	113 (26.8)	6 (5.5)	83 (43.0)	5 (12.2)	19 (24.3)	<0.0001
II	FOX	70 (16.6)	3 (2.7)	58 (30.1)	1 (2.4)	8 (10.3)	<0.0001
II	GEN	9 (2.1)	0 (0)	8 (4.1)	0 (0)	1 (1.3)	0.068
II	KAN	34 (8.1)	3 (2.7)	18 (16.4)	4 (9.8)	9 (11.5)	0.1081
II	NAL	13 (3.1)	1 (0.9)	8 (4.1)	1 (2.4)	3 (3.8)	0.4437
II	STR	73 (17.3)	7 (6.4)	46 (23.8)	5 (12.2)	15 (19.2)	0.0012
II	SXT	11 (2.6)	0 (0)	8 (4.1)	1 (2.4)	2 (2.6)	0.1912
III	FIS	77 (18.2)	7 (6.4)	41 (21.2)	10 (24.4)	19 (24.4)	0.0022
III	CHL	14 (3.3)	1 (0.9)	8 (4.1)	2 (4.9)	3 (3.8)	0.4267
III	TET	186 (44.1)	18 (16.4)	110 (57.0)	13 (31.7)	45 (57.7)	<0.0001

No. of antimicrobials
0	197 (46.7)	90 (81.8)	55 (28.5)	24 (58.5)	28 (35.9)
1	56 (13.3)	5 (4.5)	31 (16.1)	4 (9.8)	16 (20.5)
2	51 (12.1)	10 (9.1)	25 (13.0)	6 (14.6)	10 (12.8)
3	25 (5.9)	2 (1.8)	11 (5.7)	3 (7.3)	9 (11.5)
4	19 (4.5)	1 (0.9)	9 (4.7)	2 (4.9)	7 (9.0)
5	25 (5.9)	0 (0)	23 (11.9)	1 (2.4)	1 (1.3)
6	20 (4.7)	0 (0)	16 (8.3)	1 (2.4)	3 (3.8)
7	11 (2.6)	0 (0)	10 (5.2)	0 (0)	1 (1.3)
8	8 (1.9)	1 (0.9)	6 (3.1)	0 (0)	1 (1.3)
9	4 (0.9)	0 (0)	3 (1.6)	0 (0)	1 (1.3)
10	6 (1.4)	1 (0.9)	4 (2.1)	0 (0)	1 (1.3)

Antimicrobials have been grouped into three categories based on their importance in human medicine (CIPARS, 2007).

AMC, amoxicillin/clavulanic acid; TIO, ceftiofur; CRO, ceftriaxone; CIP, ciprofloxacin; AMK, amikacin; AMP, ampicillin; FOX, cefoxitin; GEN, gentamicin; KAN, kanamycin; NAL, nalidixic acid; STR, streptomycin; SXT, trimethoprim/sulfamethoxazole; FIS, sulfisoxazole; CHL, chloramphenicol; TET, tetracycline; NC, not calculated.

Prevalence of resistance genes

The prevalence of select resistance genes in E. coli varied by meat type and was higher in E. coli from chicken (75% isolates) and turkey (69% of isolates), which is consistent with the phenotypic data (Table 2). Overall, the most common resistance genes were strA/B (28% of isolates), followed by tet(A) (27% of isolates) and tet(B) (23% of isolates).

Table 2.

Distribution of Resistance Genes and Multiple Resistance Genes in Escherichia coli Isolates Recovered from Retail Meat

Resistance genes	Overall, n=422 (%)	Beef, n=110 (%)	Chicken, n=193 (%)	Pork, n=41(%)	Turkey, n=78 (%)
No gene	170 (40.3)	75 (68.2)	48 (24.9)	23 (56.1)	24 (30.8)
sul1	34 (8.1)	2 (1.8)	21 (10.9)	4 (9.8)	7 (9.0)
sul2	52 (12.3)	4 (3.6)	33 (17.1)	2 (4.9)	13 (12.3)
sul3	4 (0.9)	0 (0)	0 (0)	4 (9.8)	0 (0)
tet(A)	113 (26.8)	9 (8.2)	73 (37.8)	10 (24.4)	21 (26.9)
tet(B)	98 (23.2)	10 (9.1)	56 (29.0)	5 (12.2)	27 (34.6)
tet(C)	18 (4.3)	13 (11.8)	2 (1.0)	1 (2.4)	2 (2.6)
bla _CMY-2	62 (14.7)	1 (0.9)	52 (26.9)	1 (2.4)	8 (10.3)
bla _SHV	5 (1.2)	0 (0)	4 (2.1)	0 (0)	1 (1.3)
bla _TEM	58 (13.7)	5 (4.5)	34 (17.6)	6 (14.6)	13 (16.7)
aphA1	28 (6.6)	2 (1.8)	15 (7.8)	2 (4.9)	9 (11.5)
aphA2	7 (1.7)	0 (0)	5 (2.6)	2 (4.9)	0 (0)
aadB	2 (0.5)	0 (0)	1 (0.5)	1 (2.4)	0 (0)
aac(3)IV	3 (0.7)	1 (0.9)	0 (0)	0 (0)	2 (2.6)
strA/B	118 (28.0)	13 (11.8)	76 (39.4)	4 (9.8)	25 (32.1)
aadA	44 (10.4)	6 (5.5)	24 (12.4)	9 (22.0)	5 (6.4)

No. of resistance genes
0	170 (40.3)	75 (68.2)	48 (24.9)	23 (56.1)	24 (30.8)
1	69 (16.4)	16 (14.5)	31 (16.1)	5 (12.2)	17 (21.8)
2	72 (17.1)	9 (8.2)	49 (25.4)	2 (4.9)	12 (15.4)
3	55 (13.0)	8 (7.3)	29 (15.0)	5 (12.2)	13 (16.7)
4	33 (7.8)	2 (1.8)	17 (8.8)	4 (9.8)	10 (12.8)
5	9 (2.1)	0 (0)	8 (4.1)	1 (2.4)	0 (0)
6	9 (2.1)	0 (0)	7 (3.6)	1 (2.4)	1 (1.3)
7	4 (0.9)	0 (0)	3 (1.6)	0 (0)	1 (1.3)
9	1 (0.2)	0 (0)	1 (0.5)	0 (0)	0 (0)
≥2	183 (43.4)	19 (17.3)	114 (59.1)	13 (31.7)	37 (47.4)

The bla _TEM gene was present in about 18%, 17%, and 15% of E. coli recovered from chicken, turkey, and pork, respectively, and only in about 5% of E. coli from beef (Table 2). The bla _CMY-2 gene was found in 27% of E. coli from chicken, while in only 1–2% of E. coli from beef or pork. About 43% of isolates carried ≥2 AMR genes (Table 2), and ≥5 resistance genes were only found in E. coli from chicken (10% isolates), turkey (3% isolates), and pork (5% isolates) meats.

Unconditional associations between resistance genes

Unconditional statistical associations between eight selected genes (sul1, sul2, tet(A), tet(B), bla _CMY-2 , bla _TEM , aphA1, strA/B) with a prevalence of >5% were explored. The tet(A) gene was 5.5 times more likely to be found with bla _CMY-2 gene and 3.6 time more likely with bla _TEM gene (Table 3). The bla _TEM gene was 7.4 times more likely to be found with strA/B genes. The aadA or strA/B genes responsible for streptomycin resistance were associated with all the other resistance genes tested for and detected in this study (Table 3).

Table 3.

Unconditional Associations ^a Observed Between Resistance Genes Found in Escherichia coli Isolates Recovered from Retail Meat

Outcome gene	Predictor gene	Odds ratio	95% CI	p-value
sul1	tet(A)	11.2	4.9–25.7	<0.001
	aadA	108.5	39.4–299.0	<0.001
sul2	tet(B)	4.5	2.5–8.3	<0.001
	aphA1	9.4	4.2–21.1	<0.001
	strA/B	9.8	5.1–19.1	<0.001
tet(A)	tet(B)	0.3	0.2–0.6	<0.001
	bla _CMY-2	5.5	3.1–9.8	<0.001
	bla _TEM	3.6	2.1–6.4	<0.001
	strA/B	3.7	2.3–5.9	<0.001
	aadA	7.6	3.9–15.0	<0.001
tet(B)	aphA1	15.3	6.0–39.1	<0.001
	strA/B	3.7	2.3–5.9	<0.001
bla _CMY-2	strA/B	2.5	1.4–4.3	<0.001
bla _TEM	strA/B	7.4	4.1–13.5	<0.001
aphA1	strA/B	9.2	3.8–22.3	<0.001

Only statistically significant associations are reported (Bonferroni corrected p-value of 0.05/36<0.001).

CI, confidence interval.

Unconditional associations between resistance genes and meat type

Table 4 shows the unconditional associations between resistance genes and meat type with beef isolates used as a reference category. The bla _CMY-2 gene is more likely to be found in E. coli isolated from chicken meat (OR, 40.2; 95% CI, 5.5–295.4; p<0.001).

Table 4.

Unconditional Associations Observed Between Resistance Genes and Meat Types ^a

Resistance gene	Sample type	Odds ratio	95% CI	p-value
sul2	Overall			0.005
	Chicken	5.5	1.9–15.9	0.002
	Pork	na	na	0.729
	Turkey	5.3	1.7–16.9	0.005
tet(A)	Overall			<0.001
	Chicken	6.8	3.3–14.3	<0.001
	Pork	na	na	0.011
	Turkey	4.1	1.8–9.6	<0.001
tet(B)	Overall			<0.001
	Chicken	4.1	2.0–8.4	<0.001
	Pork	na	na	0.572
	Turkey	5.3	2.4–11.8	<0.001
bla _CMY-2	Overall			<0.001
	Chicken	40.2	5.5–295.4	<0.001
	Pork	na	na	0.482
	Turkey	12.5	1.5–101.8	0.019
strA/B	Overall			<0.001
	Chicken	4.8	2.5–9.3	<0.001
	Pork	na	na	0.722
	Turkey	3.5	1.7–7.4	<0.001

Beef isolates were used as the reference group in the analysis.

Data shown here include only those genes that were significantly associated with meat type.

CI, confidence interval; na, non-significant associations (therefore, odd ratios and 95% CI are not presented).

Unconditional associations between resistance genes and resistance phenotypes

The presence of β-lactamase genes in E. coli isolates was associated with the expression of phenotypic resistance to STR, TET, and FIS (Table 5). Similarly, the presence of resistance genes for tetracycline, sulfonamides, and aminoglycosides was associated with phenotypic expression of resistance to different classes of antimicrobials.

Table 5.

Unconditional Associations ^a Observed Between Resistance Gene Families and Phenotypic Antimicrobial Resistance in Escherichia coli Isolates Recovered from Retail Meats

Outcome gene family	Predictor AMR phenotype	Odds ratio	95% CI	p-value
β-lactamase	AMC	68.4	28.1–166.5	<0.001
	AMP	308.0	124.6–761.0	<0.001
	FOX	54.4	23.6–125.4	<0.001
	TIO	126.3	29.8–532.4	<0.001
	CRO	92.7	32.3–265.9	<0.001
	STR	6.6	3.9–11.4	<0.001
	FIS	2.3	1.4–3.9	0.001
	TET	4.6	2.9–7.4	<0.001
Tetracycline	AMC	3.8	2.1–6.9	<0.001
	AMP	4.9	3.0–8.0	<0.001
	FOX	3.4	1.9–6.1	<0.001
	TIO	3.1	1.6–5.9	<0.001
	CRO	2.3	1.3–4.1	<0.001
	KAN	7.7	3.7–16.0	<0.001
	STR	6.6	3.8–11.5	<0.001
	FIS	6.5	3.4–12.2	<0.001
	TET	9.9	5.2–18.6	<0.001
Sulfonamides	AMP	3.4	2.0–5.6	<0.001
	KAN	7.7	3.7–16.0	<0.001
	STR	6.6	3.8–16.5	<0.001
	FIS	2118.8	419.4–10703.3	<0.001
	TET	9.9	5.2–18.6	<0.001
Aminoglycosides	AMC	2.4	1.4–4.0	<0.001
	AMP	3.8	2.4–5.9	<0.001
	KAN	35.3	8.3–149.7	<0.001
	STR	115.6	27.7–481.5	<0.001
	FIS	9.6	5.3–17.2	<0.001
	TET	12.6	7.8–20.5	<0.001

Only statistically significant associations are reported (Bonferroni corrected p-value of 0.05/36<0.001).

Abbreviations are same as presented in Table 1.

Discussion

Data from this study suggests that overall more E. coli from retail poultry (chicken and turkey) were resistant to the tested antimicrobials, whereas prevalence of resistant E. coli was less in beef and pork samples. Consequently, resistance to antimicrobials deemed to be very important in human medicine (AMC and CRO) was also found more frequently in E. coli from retail poultry samples. Concurrent resistances to clinically important antimicrobials have been reported, and that may be due to the presence of bla _CMY-2 gene as suggested by previous studies (Kozak et al., 2009a; Rosengren et al., 2009). Resistance to these antimicrobials raises concerns about their effectiveness and ultimately therapeutic failure in patients should they be exposed to these bacteria through the food chain. Exposure to resistant generic E. coli poses public health risks as they can transfer genetic elements of resistance to pathogens in the human gut or while sharing a same niche.

The results of this study are mostly in agreement with reports from other provinces of Canada where a only phenotypic resistance monitoring program has been in place for the past several years (CIPARS, 2007, 2009). However, a varying degree of resistance was reported to clinically important antimicrobials in E. coli isolated from retail meats purchased in the United States (NARMS, 2008). In our study, resistance to CIP was not found in E. coli, which is a positive finding, as CIP is a fluoroquinolone, which is a class of drug used for treating severe Salmonella infections in human.

Similar to our study, very low prevalences of resistances to GEN, KAN, CHL, NAL, and SXT have been also reported in E. coli isolated from retail meat (CIPARS, 2009; NARMS, 2008). However, there was a small variation in the prevalence of resistance in E. coli from retail meats purchased in other Canadian provinces. These differences in the prevalence of resistance are difficult to interpret because data about the source of retail meat was not obtained. Interestingly, low prevalences of resistance to NAL, SXT, and KAN were found in E. coli isolated from chicken, turkey, beef, and pork samples collected in the NARMS surveillance program (NARMS, 2008). Further studies are needed to explore differences in the prevalence of resistance in E. coli from retail meat from various geographical regions.

The β-lactamase genes are important from a human health perspective, because some of them confer resistance to extended spectrum cephalosporins such as CRO and TIO. In our study, β-lactamase genes such as bla _CMY-2 and bla _TEM were commonly found in E. coli isolates. The prevalence of resistance genes was generally correlated with the phenotypic AMR to their corresponding antimicrobials. However, it is possible that resistance genes are present in some E. coli without the expression of phenotypic resistance (Rosengren et al., 2009). Furthermore, the presence of resistance phenotype without genotype pinpoints to the resistance determinants not being investigated. Despite this, the importance of cross-linked resistance cannot be ruled out. Therefore, in future, the panel of genes to be tested can be extended, and new emerging mechanism of resistance need to be explored.

Results of this study also suggest that poultry-derived E. coli are more likely to carry β-lactamase genes than pork- or beef-derived E. coli. This is conceivable because more E. coli from poultry meat were resistant to β-lactam antimicrobials, and hence, the genes responsible for resistance are present as well.

The sul3 gene conferring resistance to sulfonamides was found only in pork isolates that were resistant to SXT; this gene has been previously reported in E. coli isolated from swine (Boerlin et al., 2005). The reason for this observation is not entirely clear, but it is possible that this gene is linked with different genes than in other species and consequently a differential selection because of different antimicrobial use practices in the different commodities (Kozak et al., 2009b).

Interestingly, tet(C) was the most common resistance gene found in E. coli from beef but was found in <2% of E. coli from other meat types. This gene has been reported in E. coli isolates recovered from samples obtained from beef processing plant (Aslam et al., 2009). Previous reports suggested that tet(C) gene appears to occur more commonly in E. coli isolates from human and pigs than cattle (Boerlin et al., 2005; Wilkerson et al., 2004). In past few years, this gene has also been reported in E. coli isolated from dairy and beef cattle (Alexander et al., 2009; Srinivasan et al., 2007). Gow et al., (2008) has showed a strong association between tet(A) and tet(C) genes, suggesting their presence on the same genetic element (Roberts, 2005).

The resistance genes can be co-located on the same genetic mobile elements such as plasmids or transposons (Carattoli, 2001; Boerlin et al., 2005), and selective pressure due to the use of one antimicrobial may result in the selection of co-located genes as well. In some instances, multiple resistance genes can be present on the same genetic element (Rosengren et al., 2009), and often more than one gene can play a role in the appearance of similar resistant phenotypes (Chopra and Roberts, 2001). In such cases, the use of unrelated antimicrobials may select bacteria resistant to antimicrobials that are important in human medicine. Although integrons or plasmids were not analyzed in this study, several previous studies have established their presence in E. coli and the role they may play in the transfer of resistance genes (Carattoli, 2009; van Essen-Zandbergen et al., 2007). Further studies are needed to explore the role of transmissible genetic elements in the spread of resistance genes in various ecological environments.

The presence of antimicrobial-resistant bacteria in food animals raises serious human health concerns because of the possibility of their transfer to retail meat during commercial animal slaughter and meat processing (Aslam and Service, 2006; Aslam et al., 2009; Wu et al., 2009). Cross contamination and consumption of undercooked meat can result in ingestion of resistant bacteria by human and opportunities may exist in the human gastrointestinal tract for these bacteria to transfer resistance genes to pathogens. Such mechanisms of transfer of resistance have been demonstrated previously (Ramchandani et al., 2005; Trobos et al., 2009). Therefore, retail meat, particularly poultry meat, contaminated with antimicrobial-resistant E. coli could pose a risk to human health if improperly cooked meats are consumed or through poor food handling in the kitchen (i.e., cross-contamination). Of particular concern is the transmission of strains carrying β-lactamase genes.

Conclusion

Antimicrobial resistance was higher in E. coli isolated from retail poultry meat than those from ground beef or pork. Significant statistical associations observed between resistance genes detected in E. coli isolated from retail meats may suggest their presence on common genetic elements such as transposons, plasmids, or integrons. The associations observed between some resistance phenotypes and seemingly unrelated resistance genes suggest that the use of one antimicrobial may result in the selection of resistant E. coli to other antimicrobials of unrelated classes. Results of this study also underscore the importance of meat safety implications of antimicrobial resistant E. coli carrying genes for resistance to clinically important antimicrobials. The genotypic and phenotypic AMR data assessed in this study would help estimating the transmission of multiple linked antimicrobial resistance genes to human pathogens and possible consumer exposure to resistant strains.

Footnotes

Acknowledgments

We would like to extend our thanks to Cheryl Turner, Barbara Dakin, Catherine Taylor, Jovana Kovacevic, Kyla Kennedy, and Deana Rolheiser from the Agri-Food Laboratories Branch, Alberta Agriculture and Rural Development for their technical support for initial bacterial isolation and confirmation. The authors also thank Cara Service and Gerard Bedie for their technical assistance. We greatly acknowledge the clerical support of Loree Verquin from Lacombe Research Centre. This study was supported by the Alberta Livestock and Meat Agency and Matching Investment Initiative funds from Agriculture and Agri-Food Canada.

Disclosure Statement

No competing financial interests exist.

References

Alexander

, Reuter

, Sharma

, Yanke

, Topp

, McAllister

. Longitudinal characterization of resistant Escherichia coli in fecal deposits from cattle fed subtherapeutic levels of antimicrobials. Appl Environ Microbiol, 2009; 75:7125–7134.

Aslam

, Service

. Antimicrobial resistance and genetic profiling of Escherichia coli from a commercial beef packing plant. J Food Prot, 2006; 69:1508–1513.

Aslam

, Diarra

, Service

, Rempel

. Antimicrobial resistance genes in Escherichia coli isolates recovered from a commercial beef processing plant. J Food Prot, 2009; 72:1089–1093.

Blake

, Hillman

, Fenlon

, Low

. Transfer of antibiotic resistance between commensal and pathogenic members of the Enterobacteriaceae under ideal conditions. J Appl Microbiol, 2003; 95:428–436.

Boerlin

, Travis

, Gyles

, Reid-Smith

, Janecko

, Lim

, Nicholson

, McEwen

, Friendship

, Archambault

. Antimicrobial resistance and virulence genes of Escherichia coli isolates from swine in Ontario. Appl Environ Microbiol, 2005; 71:6753–6761.

Carattoli

. Importance of integrons in the diffusion of resistance. Vet Res, 2001; 32:243–260.

Carattoli

. Resistance plasmid families in enterobacteriaceae. Antimicrob Agents Chemother, 2009; 53:2227–2238.

Chen

, Griffiths

. PCR differentiation of Escherichia coli from other Gram-negative bacteria using primers derived from the nucleotide sequences flanking the gene encoding the universal stress protein. Lett Appl Microbiol, 1998; 27:369–371.

Chen

, Zhao

, White

, Schroeder

, Lu

, Yang

, McDermott

, Ayers

, Meng

. Characterization of multiple-antimicrobial-resistant Salmonella serovars isolated from retail meats. Appl Environ Microbiol, 2004; 70:1–7.

10.

Chopra

, Roberts

. Tetracycline antibiotics: Mode of action, applications, molecular biology, and epidemiology of bacterial resistance. Microbiol Mol Biol Rev, 2001; 65:232–260.

11.

[CIPARS] Canadian Integrated Program for Antimicrobial Resistance Surveillance. Annual report. 2007. Public Health Agency of Canada Guelph: Ontario, Canada www.phac-aspc.gc.ca/cipars-picra/2007/pdf/cipars-picra-2007-eng.pdf. 2012 January 1.

12.

[CIPARS] Canadian Integrated Program for Antimicrobial Resistance Surveillance. Final report. 2009. Public Health Agency of Canada Guelph: Ontario, Canada www.phac-aspc.gc.ca/cipars-picra/2009/pdf/2009pr-eng.pdf. 2012 February 1.

13.

[CLSI] Clinical and Laboratory Standards Institute. 20th Informational Supplement, 2010; 30,1:M100–S20. Wayne, PA: CLSI.

14.

Dutil

, Irwin

, Finley

, Ng

, Avery

, Boerlin

, Bourgault

A-M

, Cole

, Daignault

, Desruisseau

, Demczuk

, Hoang

, Horsman

, Ismail

, Jamieson

, Maki

, Pacagnella

, Pillai

. Ceftiofur resistance in Salmonella enterica serovar Heidelberg from chicken meat and humans, Canada. Emerg Infect Dis, 2010; 16:48–54.

15.

Gow

, Waldner

, Harel

, Boerlin

. Associations between antimicrobial resistance genes in fecal generic Escherichia coli isolates from cow-calf herds in Western Canada. Appl Environ Microbiol, 2008; 74:3658–3666.

16.

Kozak

, Boerlin

, Janecko

, Reid-Smith

, Jardine

. Antimicrobial resistance in Escherichia coli of swine and wild small mammals in the proximity of swine farms and in natural environments in Ontario. Appl Environ Microbiol, 2009a. 75:559–566.

17.

Kozak

, Pearl

, Parkman

, Reid-Smith

, Deckert

, Boerlin

. Distribution of sulfonamide resistance genes in Escherichia coli and Salmonella isolates from swine and chickens at abattoirs in Ontario and Québec, Canada. Appl Environ Microbiol, 2009b. 75:5999–6001.

18.

M-C

, Wang

, Li

. Identification and molecular characterization of antimicrobial-resistant shigatoxin-producing Escherichia coli isolated from retail meats products. Foodborne Pathog Dis, 2011; 8:489–493.

19.

Liu

, Wei

, Ma

, Zeng

, Lu

, Yang

, Chen

. Detection and characterisation of CTX-M and CMY-2 beta- lactamases among Escherichia coli isolates from farm animals in Guangdong Province of China. Int J Antimicrob Agents, 2007; 29:576–581.

20.

Mathew

, Liamthong

, Lin

, Hong

. Evidence of class 1 integron transfer between Escherichia coli and Salmonella spp. on livestock farms. Foodborne Pathog Dis, 2009; 6:959–964.

21.

M'ikanatha

, Sandt

, Localio

, Tewari

, Rankin

, Whichard

, Altekruse

, Lautenbach

, Folster

, Russo

, Chiller

, Reynolds

, McDermott

. Multidrug-resistant Salmonella isolates from retail chicken meat compared with human clinical isolates. Foodborne Pathog Dis, 2010; 7:929–934.

22.

[NARMS] National Antimicrobial Resistance Monitoring System. Retail meat annual report. 2008. www.fda.gov/downloads/AnimalVeterinary/SafetyHealth/AntimicrobialResistance/NationalAntimicrobialResistanceMonitoringSystem/UCM237111.pdf. 2011 July 20.

23.

Overdevest

, Willemsen

, Rijnsburger

, Eustace

, Xu

, Hawkey

, Heck

, Savelkoul

, Vandenbroucke-Grauls

, van der Zwaluw

, Huijsdens

, Kluytmans

. Extended-spectrum β-lactamase genes of Escherichia coli in chicken meat and humans, The Netherlands. Emerg Infect Dis, 2011; 17:1216–1222.

24.

Poppe

, Martin

, Gyles

, Reid-Smith

, Boerlin

, McEwen

, Prescott

, Forward

. Acquisition of resistance to extended-spectrum cephalosporins by Salmonella Newport and Escherichia coli in the intestinal tract of turkey poults. Appl Environ Microbiol, 2005; 71:1184–1192.

25.

Ramchandani

, Manges

, DebRoy

, Smith

, Johnson

, Riley

. Possible animal origin of human-associated multidrug-resistant uropathogenic Escherichia coli. Clin Infect Dis, 2005; 40:251–257.

26.

Randall

, Clouting

, Horton

, Coldham

, Wu

, Clifton-Hadley

, Davies

, Teale

. Prevalence of Escherichia coli carrying extended-spectrum β-lactamases (CTX-M and TEM-52) from broiler chickens and turkeys in Great Britain between 2006 and 2009. J Antimicrobiol Chemother, 2011; 66:86–95.

27.

Roberts

. Update on acquired tetracycline resistance genes. FEMS Microbiol Lett, 2005; 245:195–203.

28.

Rosengren

, Waldner

, Reid-Smith

. Associations between antimicrobial resistance phenotypes and antimicrobial resistance genes and virulence genes of fecal Escherichia coli isolates from healthy grow-finish pigs. Appl Environ Microbiol, 2009; 75:1373–1380.

29.

Schroeder

, White

, Ge

, Zhang

, McDermott

, Ayers

. Isolation of antimicrobial-resistant Escherichia coli from retail meats purchased in Greater Washington, D.C., USA. Int J Food Microbiol, 2003; 85:197–202.

30.

Srinivasan

, Gillespie

, Lewis

, Nguyen

, Headrick

, Schukken

, Oliver

. Phenotypic and genotypic antimicrobial resistance patterns of Escherichia coli isolated from dairy cows with mastitis. Vet Microbiol, 2007; 124:319–328.

31.

Sunde

, Norström

. The prevalence of, associations between and conjugal transfer of antibiotic resistance genes in Escherichia coli isolated from Norwegian meat and meat products. J Antimicrob Chemother, 2006; 58:741–747.

32.

Trobos

, Lester

, Olsen

, Frimodt-Møller

, Hammerum

. Natural transfer of sulphonamide and ampicillin resistance between Escherichia coli residing in the human intestine. J Antimicrobial Chemother, 2009; 63:80–86.

33.

van Essen-Zandbergen

, Smith

, Veldman

, Mevius

. Occurrence and characteristics of class 1, 2 and 3 integrons in Escherichia coli, Salmonella and Campylobacter spp. in The Netherlands. J Antimicrobial Chemother, 2007; 59:746–750.

34.

Walsh

, Duffy

, Nally

, O'Mahoney

, McDowell

, Fanning

. Transfer of ampicillin resistance from Salmonella Typhimurium DT104 to Escherichia coli K12 in food. Lett Appl Microbiol, 2008; 46:210–215.

35.

Wilkerson

, Samadpour

, van Kirk

, Roberts

. Antibiotic resistance and distribution of tetracycline resistance genes in Escherichia coli O157:H7 isolates from humans and bovines. Antimicrobiol Agents Chemother, 2004; 48:1066–1067.

36.

, Dalsgaard

, Vieira

, Emborg

H-D

, Jensen

. Prevalence of tetracycline resistance and genotypic analysis of populations of Escherichia coli from animals, carcasses and cuts processed at a pig slaughterhouse. Int J Food Microbiol, 2009; 135:254–259.

37.

Yan

J-J

, Hong

C-Y

, Ko

W-C

, Chen

Y-J

, Tsai

S-H

, Chuang

C-L

. Dissemination of blaCMY-2 among Escherichia coli isolates from food animals, retail ground meats and humans in southern Taiwan. Antimicrobiol Agents Chemother, 2004; 48:1353–1356.

38.

Yang

, Qu

, Zhang

, Shen

, Cui

, Shi

, Xi

, Sheng

, Zhi

, Meng

. Prevalence and characterization of Salmonella serovars in retail meats of marketplace in Shaanxi, China. Int J Food Microbiol, 2010; 141:63–72.

39.

Zhao

, Blickenstaff

, Glenn

, Ayers

, Friedman

, Abbott

, McDermott

. β-Lactam resistance in Salmonella strains isolated from retail meats in the United States by the National Antimicrobial Resistance Monitoring System between 2002 and 2006. Appl Environ Microbiol, 2009; 75:7624–7630.