Next Article in Journal
New Antimicrobial Resistance Strategies: An Adaptive Resistance Network Conferring Reduced Glycopeptide Susceptibility in VISA
Next Article in Special Issue
Antibiotic Resistance in Selected Emerging Bacterial Foodborne Pathogens—An Issue of Concern?
Previous Article in Journal
Clinical Diagnostics of Bacterial Infections and Their Resistance to Antibiotics—Current State and Whole Genome Sequencing Implementation Perspectives
Previous Article in Special Issue
Assessment of Heavy Metals Contamination and Antimicrobial Drugs Residue in Broiler Edible Tissues in Bangladesh
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Antibiotics
Volume 12
Issue 4
10.3390/antibiotics12040782
antibiotics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Assessment of Prevalence and Diversity of Antimicrobial Resistant Escherichia coli from Retail Meats in Southern California

by
Katie Yen Lee
1,2,*,
Kurtis Lavelle
2,
Anny Huang
1,
Edward Robert Atwill
1,
Maurice Pitesky
1 and
Xunde Li
2,*
1
Department of Population Health and Reproduction, School of Veterinary Medicine, University of California, Davis, Davis, CA 95616, USA
2
Western Institute for Food Safety and Security, University of California, Davis, Davis, CA 95616, USA
*
Authors to whom correspondence should be addressed.
Antibiotics 2023, 12(4), 782; https://doi.org/10.3390/antibiotics12040782
Submission received: 13 March 2023 / Revised: 10 April 2023 / Accepted: 17 April 2023 / Published: 19 April 2023
(This article belongs to the Special Issue Research of Antimicrobial Resistance in the Food Chain)
Browse Figures
Versions Notes

Abstract

:
Retail meat products may serve as reservoirs and conduits for antimicrobial resistance, which is frequently monitored using Escherichia coli as indicator bacteria. In this study, E. coli isolation was conducted on 221 retail meat samples (56 chicken, 54 ground turkey, 55 ground beef, and 56 pork chops) collected over a one-year period from grocery stores in southern California. The overall prevalence of E. coli in retail meat samples was 47.51% (105/221), with E. coli contamination found to be significantly associated with meat type and season of sampling. From antimicrobial susceptibility testing, 51 isolates (48.57%) were susceptible to all antimicrobials tested, 54 (51.34%) were resistant to at least 1 drug, 39 (37.14%) to 2 or more drugs, and 21 (20.00%) to 3 or more drugs. Resistance to ampicillin, gentamicin, streptomycin, and tetracycline were significantly associated with meat type, with poultry counterparts (chicken or ground turkey) exhibiting higher odds for resistance to these drugs compared to non-poultry meats (beef and pork). From the 52 E. coli isolates selected to undergo whole-genome sequencing (WGS), 27 antimicrobial resistance genes (ARGs) were identified and predicted phenotypic AMR profiles with an overall sensitivity and specificity of 93.33% and 99.84%, respectively. Clustering assessment and co-occurrence networks revealed that the genomic AMR determinants of E. coli from retail meat were highly heterogeneous, with a sparsity of shared gene networks.

1. Introduction

The emergence and dissemination of antimicrobial resistance (AMR) is of worldwide public health concern [1]. As a habitant of the endogenous microbiota of both humans and animals, Escherichia coli is both a commensal enteric bacterium and a pathogen responsible for various nosocomial, foodborne, and waterborne infections [2,3,4,5,6,7,8,9,10,11,12]. The increasing global incidence of multidrug resistant E. coli, particularly those resistant to therapeutically important drugs such as cephalosporins and fluoroquinolones, and last resort antibiotics such as carbapenem and colistin, prompts the need for integrated initiatives to monitor and reduce the spread of resistant organisms and their AMR genetic determinants [13,14,15,16]. Enhancing our understanding of AMR in E. coli is important as their ubiquity and genomic plasticity enables high frequency of AMR mobilization, which promotes the acquisition and transfer of resistance to other bacterial species [17,18]. Hence, commensal E. coli is also frequently utilized as indicator organisms of AMR for the broader microbial community [19,20,21].
Antimicrobial use in human and veterinary medicine are perceived as key drivers of AMR emergence [22,23], with selective pressures imposed amongst food animals comprising one avenue of public health concern [24,25]. Raw foods of animal origin such as meat products may serve as reservoirs and conduits for AMR [26,27], and have been included in integrated monitoring efforts due to their epidemiological linkage to foodborne outbreaks involving zoonotic pathogens [28,29] and the need to better understand the maintenance and dissemination of AMR along the food chain [30]. In the United States, the National Antimicrobial Resistance Monitoring System (NARMS) monitors AMR in various foodborne bacteria—including commensal E. coli—from humans, food-producing animals, and retail meat [31]. The distribution of AMR has been suggested to vary geographically [1,32,33], with human activity and movement of food and animals contributing to the evolution of bacterial populations [34,35,36,37].
In this study, we present the first available data on E. coli from retail meats in southern California as part of expanded surveillance coverage of NARMS retail meat sampling in 2018. The objective of this study was to determine the prevalence, the distribution of AMR and associated genetic determinants, and the potential drivers of AMR variability in E. coli from retail meat. This study aims to enhance our understanding of AMR in foodborne E. coli, and to provide insight on the clinical and epidemiologic risks associated with retail meat products in California.

2. Results

2.1. Risk Factors Associated with the Presence of E. coli in Retail Meat Products

Escherichia coli isolates were recovered from 47.51% (105/221) of samples, with the highest frequency observed in ground turkey (70.37%, 38/54), followed by chicken (67.86%, 38/56), pork chop (32.14%, 18/56), and ground beef (20.00%, 11/55) (Table 1).
From multivariable logistic regression, risk factors which were significantly associated with the presence of E. coli in retail meat products included meat type and season. The odds of E. coli isolation were 9.43 (95% CI 3.84–23.21) and 11.00 (95% CI 4.39–27.54) times higher in skin-on/bone-in chicken and ground turkey compared to ground beef products, respectively. Samples purchased in the spring, summer, and autumn had a 0.22 (95% CI 0.083–0.55), 0.36 (95% CI 0.14–0.91), and 0.36 (95% CI 0.14–0.91) times odds of E. coli isolation compared to those purchased in the winter. (Table 1).

2.2. Phenotypic Antimicrobial Resistance of E. coli from Retail Meat Products

From antimicrobial susceptibility testing (AST) of 105 E. coli isolates, 51 (48.57%) were susceptible to all antimicrobials tested, 54 isolates (51.43%) were resistant to at least 1 drug, 39 (37.14%) to 2 or more, and 21 (20.00%) to 3 or more drugs. All isolates were susceptible to azithromycin and meropenem, and 15 (14.29%) were multidrug resistant (MDR). Of the MDR isolates, 10 (66.67%) were from ground turkey, 3 (20%) were from chicken, and 2 (13.33%) were from pork chops.
E. coli isolates exhibited the highest overall frequency of resistance to tetracycline (43.81%), followed by streptomycin (30.48%), ampicillin (20.95%), and gentamicin (16.19%) (Table 2). From exact logistic regression, isolates from ground turkey had a significantly higher odds of resistance to ampicillin (OR 4.94, 95% CI 1.17–30.11), streptomycin (OR 3.81, 95% CI 1.10–15.56), and tetracycline (OR 5.24, 95% CI 1.65–18.55) compared to isolates from non-poultry meat types (beef and pork). Isolates from chicken and ground turkey products also exhibited higher odds of resistance to gentamicin compared to those from non-poultry origin; however, the association was only significant for isolates from chicken (OR 8.47, 95% 1.05–394.18) (Table 3). Collectively, E. coli isolates in this study had high diversity of phenotypic AMR patterns, with a total of 21 unique antibiogram patterns identified amongst the 54 non-susceptible isolates. Contributing to this diversity in distribution of antibiogram patterns were three isolates—two MDR isolates from pork chops and one non-MDR from ground turkey—with decreased susceptibility to ciprofloxacin (Table 4).

2.3. Genetic Determinants of AMR in E. coli from Retail Meat Products

Whole-genome sequencing (WGS) was conducted on a random subset of E. coli isolates in this study (n = 52). The genomes of these 52 E. coli isolates were screened for AMR genetic determinants including point mutations conferring quinolone resistance. A total of 27 AMR determinants were identified, including mdf(A), which was detected in all isolates, and other commonly occurring genes corresponding to tetracycline (tet(A), tet(B), and tet(C)) and sulfonamide (sul1 and sul2) resistance. The largest diversity of genes was observed for aminoglycoside resistance (nine genes encoding for various acetyltransferases, nucleotidyltransferases, or phosphotransferases) followed by those for beta-lactam resistance (blaCMY-2, blaTEM-1A, blaTEM-1B, blaHERA-3, and blaSHV-187). Notably, a variant of a colistin resistance gene, mcr-9, was detected in one MDR E. coli isolate from a chicken sample. Quinolone resistance genetic determinants were detected in two isolates including (1) a plasmid-mediated quinolone resistance (PMQR) gene, qnrB19, from an MDR isolate from pork chop; and (2) two chromosomal mutations in gyrA encoding S83L and D87N amino acid substitutions from an MDR isolate from ground turkey. Other resistance genes detected included those conferring resistance to macrolide (ere(A)), phenicol (floR), and folate synthesis inhibitor (dfrA14) drug(s) (Figure 1a).

2.4. Concordance of AMR Phenotype and Genotype

Based on the 52 E. coli isolates that underwent both WGS and AST, the drug-specific and overall sensitivity and specificity was determined to assess WGS predictions of resistant and susceptible AMR phenotypes, respectively. In this study, AMR genetic determinants identified through WGS predicted phenotypic AMR with an overall sensitivity of 93.33% and specificity of 99.84%. The greatest discordance for sensitivity was observed for chloramphenicol (50%), followed by amoxicillin/clavulanic acid (66.67%), ampicillin (84.62%), and streptomycin (93.33%). Discordance for specificity was observed only for streptomycin (97.30%) (Figure 1b, Table S3).

2.5. Clustering Analysis of E. coli Isolates

From the presence and absence of AMR genetic determinants, E. coli isolates of retail meat origin in this study did not differ significantly by meat type, season, packaging type, and label claim (PERMANOVA and ANOSIM p > 0.05; Figure 2 and Table 5). Tests for differences in E. coli AMR genetic compositions indicated that dispersion differences were present among retail meat types (PERMDISP2 p < 0.05; Table 5). The results of PERMANOVA and/or ANOSIM by the grouping factor of meat type should thus be interpreted with care due to the assumption of equal variance among meat types not being met. By season, packaging type, and label claim, the compositional variance among the groups for each of these factors was not significantly different (PERMDISP2 p > 0.05; Table 5). Overall, the grouping factors assessed in this study each accounted for a relatively low proportion of the variance in E. coli AMR genetic composition (PERMANOVA R2 0.0094–0.088), with indications of even distributions of ranks within and between groups for packaging types (ANOSIM R = 0.038) and greater dissimilarities in the average of ranks within-group than those of between-groups for meat type, season, and label claim (ANOSIM R < 0, Table 5) [38].

2.6. Co-Occurrence Networks of AMR Genetic Determinants in E. coli Isolates

Despite the diversity in AMR genetic determinants present in E. coli isolates in this study (Figure 3a), the co-occurrence network of AMR genes was sparse, with the most commonly co-occurring genes being mdf(A) with tet(A) and mdf(A) with tet(B) at a co-occurrence frequency of 15 (28.85%) and 12 (23.08%) genomes, respectively (Figure 3c). At a lower co-occurrence threshold of ≥5 isolates, the networks of mdf(A) with tet(A) and/or tet(B) genes co-occurred with either a gene cluster comprised of aph(3″)-Ib and aph(6)-Id or another including aac(3)-VIa, ant(3″)-Ia, and sul1 (Figure 3b).

3. Discussion

The data presented in this study were collected from raw meat products from retail stores in southern California using the same sampling and processing protocols from FDA NARMS, thus enabling comparisons between our data and previous/concurrent NARMS data collected. No NARMS data have been available on E. coli from retail meats in California, as only certain pathogens of interest (Salmonella and Campylobacter) have been included in routine AMR assessment for California prior to this study.
Of the 221 retail meat products collected from southern California in 2018 and assessed in this study, E. coli was isolated from around half of the samples. Compared to the national average for NARMS retail testing for the same year, isolation of E. coli from samples in California was higher overall (47.51% vs. national average 28.75%) with higher recovery of E. coli in our study for chicken (67.86% vs. national average 19.22%), ground turkey (70.37% vs. national average 45.65%), and pork chop (32.14% vs. national average 25.82%), and lower recovery in ground beef (20.00% vs. national average 28.54%) [39].
Previous studies assessing the presence of enteric bacteria in meat products have found substantial variability in the prevalence of E. coli [26,40,41,42], which could be due to differences in sampling methodology, sample composition and origin, and processing protocols. Nevertheless, the significant association between E. coli contamination and meat type observed in our study is congruent with findings from studies conducted in multiple countries, where chicken and/or turkey products were found to have comparably higher prevalence of E. coli relative to non-poultry meat types such as beef and pork [26,41,42,43]. Additionally, season of retail meat purchase in our study was also significantly associated with the presence of E. coli, which could be explained by season serving as a surrogate for other unmeasured but correlated variables such as temperature, climate, and other temporal factors that more directly reflect the abiotic and biotic drivers of enteric bacteria persistence and proliferation in meat products [44]. This is substantiated by findings from a previous study which evaluated the relationship between weather variables and zoonotic foodborne pathogen contamination of meat products along the food chain in Canada. Although no definite seasonal trend was identified in their study for generic E. coli, Smith et al. found correlations between increased total precipitation and increased average/monthly temperature and E. coli in retail beef and pork, respectively [45].
The elevated frequencies of ampicillin, streptomycin, and tetracycline resistance in E. coli from ground turkey and gentamicin resistance in chicken compared to non-poultry meats (beef and pork) in our study are consistent with previous data collected in the United States [26,46,47]. While E. coli resistance to certain antimicrobials appears to be globally ubiquitous—for instance, high tetracycline resistance in E. coli from retail meats documented in Canada, India, Korea, and China—the overall heterogeneity across global frequencies of resistance is most likely attributed to differences in antimicrobial usage and the variability in selective pressures imposed across different countries and food animal production sectors [41,42,48,49]. For instance, a previous study reported 75.7% of E. coli from poultry meat in Korea being resistant to nalidixic acid, which, while consistent with national data of veterinary antibiotics sold for use in Korean poultry production [50], is in contrast to findings in our study of low nalidixic acid (1.90%) and ciprofloxacin (0.95%) resistance that are likely reflective of the ban of fluoroquinolone use in poultry production and restrictions in fluoroquinolone use for other food animal species in the US by the Food and Drug Administration [51,52].
The detection of mcr-9—a gene encoding a putative phosphoethanolamine transferase that reduces affinity for colistin—from an MDR E. coli isolate from retail chicken in this study is noteworthy as colistin is not used for treatment of food animals in the United States [53,54]. Colistin is a polymyxin antimicrobial and one of the few last-resort drugs available to treat life-threatening multidrug drug-resistant (MDR) and extensively drug-resistant (XDR) infections caused by Gram-negative bacteria such as carbapenem resistant E. coli, P. aeruginosa, and K. pneumoniae [55,56]. It was previously believed that colistin resistance was solely mediated by chromosomal genes (phoPQ, pmrAB, and mgrB) until the plasmid-mediated mcr-1 gene from China was first reported in 2015 [57], and mcr-1 variants (mcr-2 through 10) were subsequently identified in over two dozen bacterial species across six continents [58,59,60]. In the United States, mcr-9 was first identified in a Salmonella enterica serotype Typhimurium clinical isolate in 2019 [61]. Since then, analysis of previously collected bacterial isolates from routine NARMS retail meat surveillance has traced mcr-9 in isolates collected as far back as 2002, with findings of its occurrence in a pronounced proportion of Salmonella isolates in the US (28.6%, 2002–2019)—particularly S. Saintpaul from ground turkey—and also in a few E. coli isolates from samples collected in 2018 and 2019 [54]. Epidemiologically, mcr-9 has not been linked to clinical resistance of colistin [54,61], which is consistent with findings here of mcr-9 carriage in a colistin-susceptible E. coli isolate. It has, however, been shown that the presence of sub-inhibitory levels of colistin is sufficient to induce mcr-9 expression through mediation by the two-component regulatory system qseBC, resulting in elevated MIC levels [53]. This highlights the importance of bacterial genomic surveillance efforts as disseminated mcr genes and other resistant determinants of high public health concern can remain undetected and/or unexpressed in non-resistant isolates until induced by antimicrobial exposure [53].
Consistent with the diversity of phenotypic AMR profiles of E. coli isolates, whole-genome sequencing identified several other AMR genetic determinants, including those mediating resistance to other drugs of high clinical importance such as fluoroquinolones and cephalosporins. Genetic characterization of fluoroquinolone resistance has been well documented to occur through combinations of chromosomal mutations within DNA gyrase (e.g., gyrA) and/or topoisomerase IV genes (e.g., parC) [62], as observed in the MDR ciprofloxacin resistant isolate from ground turkey in this study (gyrA mutations S82L and D87N). Plasmid-mediated quinolone resistance determinants (e.g., qnrB19 from an MDR isolate from pork chop with decreased susceptibility to ciprofloxacin in this study) confer low-level resistance but nevertheless still impart concerns due to the reported frequencies of their of co-occurrence with other AMR genetic determinants—for instance, in ESBL-producing E. coli—and their high propensity to simultaneously disseminate multiple resistances [63,64,65,66]. To this point, we observed low prevalence of cephalosporin resistant E. coli in our study, which was mediated by AmpC-type beta-lactamase gene, blaCMY-2, but these isolates were MDR with carriage of over eight other AMR genes.
Previous studies have evaluated the ability of WGS to predict E. coli phenotypic resistance, with findings of imperfect but high overall concordance between genotypic and phenotypic resistance [20,67,68], as also observed in this study. The discrepancies in sensitivity for amoxicillin/clavulanic acid and chloramphenicol in our study could be attributed to the very small number of resistant isolates for these drugs which the analysis was based on. Moreover, the concordance of WGS with AMR phenotype is heavily dependent on the categorization of susceptible and resistant isolates, with grouping of intermediate isolates potentially affecting results, alongside any ambiguity in breakpoints used. An example of the latter is the lack of Clinical and Laboratory Standards Institute (CLSI) criteria for streptomycin, a veterinary drug which exhibited both imperfect sensitivity and specificity here and discordance in prior studies [67,69,70,71]. Additionally, AMR genetic determinants conferring less definitive phenotypic resistances such as those encoding multidrug efflux pump genes (e.g., mdf(A) in this study) add complexity to both data analysis and interpretation. Other considerations have been detailed previously and include the impacts of technical processes in conducting AST/WGS, database selection, and thresholds used for determining the presence or absence of AMR genetic determinants [72]. Lastly, it should be noted that our study evaluated a small number of isolates, all of which were derived from retail meats limited in one geographic region. Thus, while this study finds WGS to be a robust tool for phenotypic AMR predictions in foodborne E. coli, its utility for other bacteria from different sources should be considered with caution and supplemented with phenotypic testing to ensure comprehensive AMR assessment.
E. coli isolates in this study were genomically heterogeneous with respect to AMR, with meat type, season, packaging, and label claim accounting for very little of the variability in AMR genetic determinants and a sparsity of shared gene networks observed. Our results suggest that AMR acquisition in E. coli from retail meat exhibits greater complexities that could not be fully explained by the retail-level factors assessed and/or by the data collected, as the small number of isolates in this study are likely not representative of the diversity of E. coli as a whole from retail meats. Moreover, a limitation of retail level surveillance conducted in this study is that other factors along the farm-to-fork continuum which could potentially contribute to AMR could not be evaluated. Nevertheless, our findings from AMR gene co-occurrence networks reflect certain intricacies of AMR dynamics. For instance, we observe elevated co-occurrence of gene networks corresponding to a broad multidrug transporter (mdf(A)) and to tetracycline, sulfonamide, and aminoglycoside genes which confer resistance to antimicrobials that are conventionally used in food animal production in the US. The higher frequencies of co-occurrence of these genes (tet(A), tet(B), aph(3″)-Ib, aph(6)-Id, aac(3)-VIa, ant(3″)-Ia, sul1, and sul2) could result from direct exposure to the corresponding antimicrobials at some point along the food chain, or persistence as a result of co-selection from these genes occurring on the same mobile genetic element (e.g., plasmids). We did not assess AMR gene carriage on plasmids in this study due to the limited capacity of short-read sequencing data to fully resolve plasmid structures [73]; however, other studies employing long-read sequencing have confirmed the occurrence of these genes on the same plasmid(s) in E. coli [74,75,76]. Lastly, the diversity of AMR genetic determinants identified in this study—including those corresponding to antimicrobials not used in food animals in the US (e.g., mcr-9)—suggests that the accumulation of AMR reservoirs could occur even in the absence of direct selective pressures, with the acquisition and loss of certain AMR genes in E. coli possibly attributed to the presence or absence of fitness costs that are associated with the maintenance of these genes [77,78].

4. Materials and Methods

4.1. Study Area and Sampling

Samples in this study were collected as part of routine NARMS retail meat surveillance in 2018, when the program expanded geographical coverage to include southern California. From January to December 2018, a total of 480 fresh retail meat samples consisting of 240 skin-on/bone-in chicken, 120 ground turkey, 60 ground beef, and 60 pork chops were purchased from retail grocery stores in southern California twice each month. Sampling locations were selected based on the NARMS retail store sampling plan through random selection of grocery stores within zip codes corresponding to West Los Angeles, East Los Angeles, Ontario, and Irvine. Samples were transported on ice to the laboratory, refrigerated, and processed within 72 h of purchase.

4.2. Sample Processing and E. coli Isolation

A random selection of 221 samples (56 chicken, 54 ground turkey, 55 ground beef, and 56 pork chops) was processed for isolation of E. coli per the 2018 NARMS Retail Meat Surveillance protocol [79]. Briefly, 25 g of each sample was placed in Whirl-Paks containing 250 mL buffered peptone water (Becton Dickinson, Franklin Lakes, NJ, USA) and hand massaged for 3 min. A total of 50 mL of rinsate was then added to 50 mL double-strength MacConkey broth (Becton Dickinson, Franklin Lakes, NJ, USA) and incubated at 35 °C for 24 h. Following overnight enrichment, a loopful (10 µL) was streaked to a MacConkey plate and incubated at 35 °C for 24 h. One suspect E. coli colony based on typical colony morphology was streaked to purity on blood agar plates and incubated overnight at 35 °C. Isolates were confirmed as E. coli using biochemical tests (indole positive and oxidase negative), and banked in Brucella broth with 15% glycerol, frozen, and shipped on dry-ice to the FDA’s Center for Veterinary Medicine (CVM) for antimicrobial susceptibility testing and whole-genome sequencing.

4.3. Antimicrobial Susceptibility Testing

E. coli isolates were tested using a broth microdilution method for 14 antimicrobial drugs using the NARMS Gram-negative plates (Thermo Fisher Scientific, Waltham, MA, USA) per standard protocols [80]. NARMS breakpoints were used to classify isolates into susceptible, intermediate, and resistant categories based on the minimum inhibitory concentration (MIC) values for each drug; these breakpoints are based on the CLSI guidelines with the exception of streptomycin and azithromycin, where NARMS consensus interpretive criteria were used due to absence of available CLSI breakpoints for these two drugs (Table S2) [81]. Due to the limited range of dilutions in the drug panel, resistance could not be determined for azithromycin and sulfisoxazole and only susceptible classification were determined for these two drugs. For analysis, intermediate and susceptible isolates were grouped together. Decreased susceptibility (DSC) to ciprofloxacin (≥0.12 µg/mL) was also noted in descriptive analyses due to the expanded definition from CLSI for its intermediate susceptibility MIC range [81]. Multidrug resistance was defined as resistance to one or more drugs in three or more antimicrobial classes [82].

4.4. Whole-Genome Sequencing and Identification of Resistance Genes

A subset of E. coli isolates (n = 52) was randomly selected to undergo whole-genome sequencing (WGS) by short-read sequencing on the Illumina MiSeq using v2 or v3 chemistry for 2 × 250-bp paired end reads, and identification of resistance genes was conducted as previously described [67]. Briefly, genomic DNA was extracted using the DNeasy Blood and Tissue kit (Qiagen, Valencia, CA, USA), and libraries were prepared using the Illumina Nextera XT kit per manufacturer’s protocols. Sequences were demultiplexed using MiSeq Reporter and assembled using the CLC Genomics Workbench. The ResFinder database (Center for Genomic Epidemiology, DTU) was used to identify resistance gene hits (≥85% amino acid identity and ≥50% sequence length), and Perl scripts were used to extract and analyze the gyrA gene at amino acid position 83 and 87 to assess chromosomal mutations associated with quinolone resistance.

4.5. Data Analysis

Descriptive statistics (prevalence of E. coli, distribution of predictor variables, antimicrobial susceptibility testing results, and prevalence of antimicrobial resistance genes) and binary logistic regression models were conducted using SAS On-Demand for Academics. Predictor variables evaluated in this study include meat type (ground turkey, chicken, pork chop, and ground beef), time of year of sample purchase, packaging type (modified atmosphere packaging, plastic film, vacuum, chub, paper, and plastic bag), label claims, packaging type, and presence of Salmonella. For label claims, reduced antibiotic use included samples labeled organic or reduced/no antibiotic usage; all other samples with absence of such label claims were classified as conventional. Retail meat samples in this study were concurrently processed for isolation of Salmonella, with data on the presence of Salmonella obtained through methods detailed previously [69].
The association between the presence of E. coli in retail meat samples and predictor variables were evaluated using logistic regression models. Univariable logistic regression models were used to evaluate the crude association between each predictor variable and the outcome binary variable, which was designated as the presence or absence of a recovered E. coli isolate from the retail meat sample. A multivariable logistic regression model was then fitted based on the retention of significant variables, assessment of collinearity, testing of all two-way interactions, and best model fit as determined by the lowest Akaike’s Information Criterion (AIC). The association between predictor variables and whether an E. coli isolate from retail meat was multidrug resistant or not, and whether it was resistant or not to ampicillin, gentamicin, streptomycin, and tetracycline, were individually evaluated using exact logistic regression models. These four drugs were selected for evaluation due to the higher frequency of observed resistance.
Prediction of phenotypic antimicrobial resistance (AMR) from antimicrobial susceptibility testing based on genotypic AMR from the presence of antimicrobial resistance genes was evaluated as previously described [69]. Briefly, phenotype and genotype concordance for each drug included true positives (TP: resistant isolate with corresponding AMR genetic determinants) and true negatives (TN: susceptible isolates with absence of corresponding AMR genetic determinants). Discordance included false negatives (FN: resistant isolates with absence of corresponding AMR genetic determinants) and false positives (FP: susceptible isolates with presence of corresponding AMR genetic determinants). Sensitivity and specificity were then calculated as TP/(TP+FN) and TN/(TN+FP), respectively. Due to the absence of phenotypically resistant isolates to sulfisoxazole, azithromycin, and meropenem, sensitivity could not be assessed for them, so these three drugs were omitted from the overall calculation for sensitivity. Multidrug resistance gene, mdf(A), was omitted from the concordance analysis due to ambiguity in its AMR phenotype conferral.
To assess the collective AMR gene profiles of E. coli isolates, clustering based on the presence and absence of AMR genetic determinants was evaluated using functions in the vegan package [83] in R by grouping factors of meat type, season of retail meat purchase, packaging type, and label claim. Non-metric multidimensional scaling (NMDS) was performed using the metaMDS function with a Jaccard distance metric and in two dimensions. The permutest and betadisper functions were used to conduct a PERMDISP2 procedure to evaluate if dispersions of groups for each of the grouping factors were homogenous [84,85]. Permutational analysis of variance (PERMANOVA) was conducted to test the equivalence of centroids of groups for each grouping factor using the adonis2 function. Additionally, analysis of similarity (ANOSIM) was performed to evaluate for each grouping factor whether the average of ranks within-group distances was greater or equal to that of between-group distances [38]. The aforementioned tests (PERMDISP2, PERMANOVA, and ANOSIM) were performed using 10,000 permutations and a Jaccard distance metric.
From the presence and absence data of AMR determinants in E. coli isolates, a pairwise co-occurrence matrix was constructed by transforming the binary data. The resulting co-occurrence data were then visualized as networks of co-occurring genes using Gephi [86], with nodes representing genes and edges representing the frequency of co-occurrence. Networks were evaluated by frequency of co-occurrence thresholds based on ≥1 genome (all E. coli isolates), ≥5 genomes, and ≥10 genomes.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/antibiotics12040782/s1, Table S1: List of E. coli isolates in this study; Table S2: NARMS breakpoints used to classify minimum inhibitory concentrations of E. coli isolates; Table S3: Genotypic prediction of phenotypic resistance in E. coli isolates from retail meat (n = 52).

Author Contributions

Conceptualization, K.Y.L., E.R.A., M.P. and X.L.; methodology, K.Y.L., E.R.A. and X.L.; formal analysis, K.Y.L.; data curation, K.Y.L., K.L. and A.H.; visualization, K.Y.L.; project administration, K.Y.L. and X.L.; resources, E.R.A., M.P. and X.L.; writing—original draft preparation, K.Y.L.; writing—review and editing, K.L., A.H., E.R.A., M.P. and X.L.; funding acquisition, E.R.A., M.P. and X.L. All authors have read and agreed to the published version of the manuscript.

Funding

This work was made possible, in part, by the U.S. Food and Drug Administration through grant 1U01FD006248-01. The views expressed in this article are those of the authors and do not necessarily reflect the official policies of the Department of Health and Human Services; nor do any mention of trade names, commercial practices, or organization imply endorsement by the United States Government. The Article Processing Charge (APC) was partially funded by the UC Davis Library Open Access Fund (UCD-OAF).

Data Availability Statement

Whole-genome sequences of E. coli isolates in this study are deposited under BioProject PRJNA292663.

Acknowledgments

We gratefully thank Jennifer Chase and Jessica Vandeleest for their technical support, and Gordon Martin, Cong Li, and Sonya Bodeis-Jones from the U.S. Food and Drug Administration for performing and sharing results from antimicrobial susceptibility testing and whole-genome sequencing.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Murray, C.J.; Ikuta, K.S.; Sharara, F.; Swetschinski, L.; Robles Aguilar, G.; Gray, A.; Han, C.; Bisignano, C.; Rao, P.; Wool, E.; et al. Global burden of bacterial antimicrobial resistance in 2019: A systematic analysis. Lancet 2022, 399, 629–655. [Google Scholar] [CrossRef] [PubMed]
  2. Mills, E.G.; Martin, M.J.; Luo, T.L.; Ong, A.C.; Maybank, R.; Corey, B.W.; Harless, C.; Preston, L.N.; Rosado-Mendez, J.A.; Preston, S.B.; et al. A one-year genomic investigation of Escherichia coli epidemiology and nosocomial spread at a large US healthcare network. Genome Med. 2022, 14, 147. [Google Scholar] [CrossRef]
  3. Yang, X.; Guo, R.; Xie, B.; Lai, Q.; Xu, J.; Hu, N.; Wan, L.; Dai, M.; Zhang, B. Drug resistance of pathogens causing nosocomial infection in orthopedics from 2012 to 2017: A 6-year retrospective study. J. Orthop. Surg. 2021, 16, 100. [Google Scholar] [CrossRef] [PubMed]
  4. Sikora, A.; Zahra, F. Nosocomial infections. In StatPearls; StatPearls Publishing: Treasure Island, FL, USA, 2022. [Google Scholar]
  5. Emori, T.G.; Gaynes, R.P. An overview of nosocomial infections, including the role of the microbiology laboratory. Clin. Microbiol. Rev. 1993, 6, 428–442. [Google Scholar] [CrossRef] [PubMed]
  6. Daga, A.P.; Koga, V.L.; Soncini, J.G.M.; De Matos, C.M.; Perugini, M.R.E.; Pelisson, M.; Kobayashi, R.K.T.; Vespero, E.C. Escherichia coli bloodstream infections in patients at a university hospital: Virulence factors and clinical characteristics. Front. Cell. Infect. Microbiol. 2019, 9, 191. [Google Scholar] [CrossRef]
  7. Rangel, J.; Sparling, P.; Crowe, C.; Griffin, P.; Swerdlow, D. Epidemiology of Escherichia coli O157:H7 outbreaks, United States, 1982–2002. Emerg. Infect. Dis. 2005, 11, 603–609. [Google Scholar] [CrossRef]
  8. Carstens, C.K.; Salazar, J.K.; Darkoh, C. Multistate outbreaks of foodborne illness in the United States associated with fresh produce from 2010 to 2017. Front. Microbiol. 2019, 10, 2667. [Google Scholar] [CrossRef]
  9. Luna-Guevara, J.J.; Arenas-Hernandez, M.M.P.; Martínez De La Peña, C.; Silva, J.L.; Luna-Guevara, M.L. The role of pathogenic E. coli in fresh vegetables: Behavior, contamination factors, and preventive measures. Int. J. Microbiol. 2019, 2019, 2894328. [Google Scholar] [CrossRef]
  10. Olsen, S.J.; Miller, G.; Breuer, T.; Kennedy, M.; Higgins, C.; Walford, J.; Mckee, G.; Fox, K.; Bibb, W.; Mead, P. A waterborne outbreak of Escherichia coli O157:H7 infections and hemolytic uremic syndrome: Implications for rural water systems. Emerg. Infect. Dis. 2002, 8, 370–375. [Google Scholar] [CrossRef]
  11. Craun, G.F.; Brunkard, J.M.; Yoder, J.S.; Roberts, V.A.; Carpenter, J.; Wade, T.; Calderon, R.L.; Roberts, J.M.; Beach, M.J.; Roy, S.L. Causes of outbreaks associated with drinking water in the United States from 1971 to 2006. Clin. Microbiol. Rev. 2010, 23, 507–528. [Google Scholar] [CrossRef]
  12. Lim, J.Y.; Yoon, J.W.; Hovde, C.J. A brief overview of Escherichia coli O157:H7 and its plasmid O157. J. Microbiol. Biotechnol. 2010, 20, 5–14. [Google Scholar] [CrossRef] [PubMed]
  13. Tadesse, D.A.; Zhao, S.; Tong, E.; Ayers, S.; Singh, A.; Bartholomew, M.J.; McDermott, P.F. Antimicrobial drug resistance in Escherichia coli from humans and food animals, United States, 1950–2002. Emerg. Infect. Dis. 2012, 18, 741–749. [Google Scholar] [CrossRef] [PubMed]
  14. Levy, S.B.; Bonnie, M. Antibacterial resistance worldwide: Causes, challenges and responses. Nat. Med. 2004, 10, S122–S129. [Google Scholar] [CrossRef] [PubMed]
  15. Pires, J.; Huisman, J.S.; Bonhoeffer, S.; Van Boeckel, T.P. Increase in antimicrobial resistance in Escherichia coli in food animals between 1980 and 2018 assessed using genomes from public databases. J. Antimicrob. Chemother. 2022, 77, 646–655. [Google Scholar] [CrossRef] [PubMed]
  16. MacKinnon, M.C.; McEwen, S.A.; Pearl, D.L.; Lyytikäinen, O.; Jacobsson, G.; Collignon, P.; Gregson, D.B.; Valiquette, L.; Laupland, K.B. Increasing incidence and antimicrobial resistance in Escherichia coli bloodstream infections: A multinational population-based cohort study. Antimicrob. Resist. Infect. Control 2021, 10, 131. [Google Scholar] [CrossRef] [PubMed]
  17. Dobrindt, U.; Agerer, F.; Michaelis, K.; Janka, A.; Buchrieser, C.; Samuelson, M.; Svanborg, C.; Gottschalk, G.; Karch, H.; Hacker, J. Analysis of genome plasticity in pathogenic and commensal Escherichia coli isolates by use of DNA arrays. J. Bacteriol. 2003, 185, 1831–1840. [Google Scholar] [CrossRef]
  18. Braz, V.S.; Melchior, K.; Moreira, C.G. Escherichia coli as a multifaceted pathogenic and versatile bacterium. Front. Cell. Infect. Microbiol. 2020, 10, 548492. [Google Scholar] [CrossRef]
  19. Anjum, M.F.; Schmitt, H.; Börjesson, S.; Berendonk, T.U.; Donner, E.; Stehling, E.G.; Boerlin, P.; Topp, E.; Jardine, C.; Li, X.; et al. The potential of using E. coli as an indicator for the surveillance of antimicrobial resistance (AMR) in the environment. Curr. Opin. Microbiol. 2021, 64, 152–158. [Google Scholar] [CrossRef]
  20. Stubberfield, E.; AbuOun, M.; Sayers, E.; O’Connor, H.M.; Card, R.M.; Anjum, M.F. Use of whole genome sequencing of commensal Escherichia coli in pigs for antimicrobial resistance surveillance, United Kingdom, 2018. Euro Surveill. Bull. Eur. Mal. Transm. Eur. Commun. Dis. Bull. 2019, 24, 1900136. [Google Scholar] [CrossRef]
  21. Karp, B.E.; Tate, H.; Plumblee, J.R.; Dessai, U.; Whichard, J.M.; Thacker, E.L.; Hale, K.R.; Wilson, W.; Friedman, C.R.; Griffin, P.M.; et al. National Antimicrobial Resistance Monitoring System: Two Decades of advancing Public Health through Integrated Surveillance of Antimicrobial Resistance. Foodborne Pathog. Dis. 2017, 14, 545–557. [Google Scholar] [CrossRef]
  22. Prestinaci, F.; Pezzotti, P.; Pantosti, A. Antimicrobial resistance: A global multifaceted phenomenon. Pathog. Glob. Health 2015, 109, 309–318. [Google Scholar] [CrossRef] [PubMed]
  23. Dadgostar, P. Antimicrobial resistance: Implications and costs. Infect. Drug Resist. 2019, 12, 3903–3910. [Google Scholar] [CrossRef] [PubMed]
  24. Palma, E.; Tilocca, B.; Roncada, P. Antimicrobial resistance in veterinary medicine: An overview. Int. J. Mol. Sci. 2020, 21, 1914. [Google Scholar] [CrossRef] [PubMed]
  25. Kasimanickam, V.; Kasimanickam, M.; Kasimanickam, R. Antibiotics use in food animal production: Escalation of antimicrobial resistance: Where are we now in combating AMR? Med. Sci. 2021, 9, 14. [Google Scholar] [CrossRef]
  26. Zhao, S.; Blickenstaff, K.; Bodeis-Jones, S.; Gaines, S.A.; Tong, E.; McDermott, P.F. Comparison of the prevalences and antimicrobial resistances of Escherichia coli isolates from different retail meats in the United States, 2002 to 2008. Appl. Environ. Microbiol. 2012, 78, 1701–1707. [Google Scholar] [CrossRef] [PubMed]
  27. Economou, V.; Gousia, P. Agriculture and food animals as a source of antimicrobial-resistant bacteria. Infect. Drug Resist. 2015, 8, 49–61. [Google Scholar] [CrossRef]
  28. Omer, M.K.; Álvarez-Ordoñez, A.; Prieto, M.; Skjerve, E.; Asehun, T.; Alvseike, O.A. A systematic review of bacterial foodborne outbreaks related to red meat and meat products. Foodborne Pathog. Dis. 2018, 15, 598–611. [Google Scholar] [CrossRef]
  29. Bryan, F.L. Foodborne diseases in the United States associated with meat and poultry. J. Food Prot. 1980, 43, 140–150. [Google Scholar] [CrossRef]
  30. Zawack, K.; Li, M.; Booth, J.G.; Love, W.; Lanzas, C.; Gröhn, Y.T. Monitoring antimicrobial resistance in the food supply chain and its implications for FDA policy initiatives. Antimicrob. Agents Chemother. 2016, 60, 5302–5311. [Google Scholar] [CrossRef]
  31. Food and Drug Administration. The National Antimicrobial Resistance Monitoring System. Available online: https://www.fda.gov/animal-veterinary/antimicrobial-resistance/national-antimicrobial-resistance-monitoring-system (accessed on 6 February 2023).
  32. Nyirabahizi, E.; Tyson, G.H.; Tate, H.; Kabera, C.; Crarey, E.; Ayers, S.; Strain, E. Northeastern U.S. Salmonella strains from retail meat are more prevalent and more resistant to antimicrobials. J. Food Prot. 2020, 83, 849–857. [Google Scholar] [CrossRef]
  33. Aronin, S.I.; Gupta, V.; Dunne, M.W.; Watts, J.A.; Yu, K.C. Regional differences in antibiotic-resistant Enterobacterales urine isolates in the United States: 2018–2020. Int. J. Infect. Dis. 2022, 119, 142–145. [Google Scholar] [CrossRef] [PubMed]
  34. Frost, I.; Van Boeckel, T.P.; Pires, J.; Craig, J.; Laxminarayan, R. Global geographic trends in antimicrobial resistance: The role of international travel. J. Travel Med. 2019, 26, taz036. [Google Scholar] [CrossRef]
  35. Strachan, N.J.C.; Rotariu, O.; Lopes, B.; Macrae, M.; Fairley, S.; Laing, C.; Gannon, V.; Allison, L.J.; Hanson, M.F.; Dallman, T.; et al. Whole genome sequencing demonstrates that geographic variation of Escherichia coli O157 genotypes dominates host association. Sci. Rep. 2015, 5, 14145. [Google Scholar] [CrossRef] [PubMed]
  36. Achtman, M. Evolution, population structure, and phylogeography of genetically monomorphic bacterial pathogens. Annu. Rev. Microbiol. 2008, 62, 53–70. [Google Scholar] [CrossRef] [PubMed]
  37. Fletcher, S. Understanding the contribution of environmental factors in the spread of antimicrobial resistance. Environ. Health Prev. Med. 2015, 20, 243–252. [Google Scholar] [CrossRef] [PubMed]
  38. Anderson, M.J.; Walsh, D.C.I. PERMANOVA, ANOSIM, and the Mantel test in the face of heterogeneous dispersions: What null hypothesis are you testing? Ecol. Monogr. 2013, 83, 557–574. [Google Scholar] [CrossRef]
  39. Food and Drug Administration. 2018 NARMS Update: Integrated Report Summary Interactive Version. Available online: https://www.fda.gov/animal-veterinary/national-antimicrobial-resistance-monitoring-system/2018-narms-update-integrated-report-summary-interactive-version (accessed on 6 February 2023).
  40. Zhao, C.; Ge, B.; De Villena, J.; Sudler, R.; Yeh, E.; Zhao, S.; White, D.G.; Wagner, D.; Meng, J. Prevalence of Campylobacter spp., Escherichia coli, and Salmonella serovars in retail chicken, turkey, pork, and beef from the Greater Washington, D.C., area. Appl. Environ. Microbiol. 2001, 67, 5431–5436. [Google Scholar] [CrossRef]
  41. Li, H.; Liu, Y.; Yang, L.; Wu, X.; Wu, Y.; Shao, B. Prevalence of Escherichia coli and antibiotic resistance in animal-derived food samples—six districts, Beijing, China, 2020. China CDC Wkly. 2021, 3, 999–1004. [Google Scholar] [CrossRef]
  42. Sheikh, A.A.; Checkley, S.; Avery, B.; Chalmers, G.; Bohaychuk, V.; Boerlin, P.; Reid-Smith, R.; Aslam, M. Antimicrobial resistance and resistance genes in Escherichia coli isolated from retail meat Purchased in Alberta, Canada. Foodborne Pathog. Dis. 2012, 9, 625–631. [Google Scholar] [CrossRef]
  43. Farhoumand, P.; Hassanzadazar, H.; Soleiman Soltanpour, M.; Aminzare, M.; Abbasi, Z. Prevalence, genotyping and antibiotic resistance of Listeria monocytogenes and Escherichia coli in fresh beef and chicken meats marketed in Zanjan, Iran. Iran. J. Microbiol. 2020, 12, 537–546. [Google Scholar] [CrossRef]
  44. Van Elsas, J.D.; Semenov, A.V.; Costa, R.; Trevors, J.T. Survival of Escherichia coli in the environment: Fundamental and public health aspects. ISME J. 2011, 5, 173–183. [Google Scholar] [CrossRef] [PubMed]
  45. Smith, B.A.; Meadows, S.; Meyers, R.; Parmley, E.J.; Fazil, A. Seasonality and zoonotic foodborne pathogens in Canada: Relationships between climate and Campylobacter, E. coli and Salmonella in meat products. Epidemiol. Infect. 2019, 147, E190. [Google Scholar] [CrossRef] [PubMed]
  46. Food and Drug Administration. NARMS Now: Integrated Data. Available online: https://www.fda.gov/animal-veterinary/national-antimicrobial-resistance-monitoring-system/narms-now-integrated-data (accessed on 6 February 2023).
  47. Davis, G.S.; Waits, K.; Nordstrom, L.; Grande, H.; Weaver, B.; Papp, K.; Horwinski, J.; Koch, B.; Hungate, B.A.; Liu, C.M.; et al. Antibiotic-resistant Escherichia coli from retail poultry meat with different antibiotic use claims. BMC Microbiol. 2018, 18, 174. [Google Scholar] [CrossRef]
  48. Hussain, A.; Shaik, S.; Ranjan, A.; Nandanwar, N.; Tiwari, S.K.; Majid, M.; Baddam, R.; Qureshi, I.A.; Semmler, T.; Wieler, L.H.; et al. Risk of transmission of antimicrobial resistant Escherichia coli from commercial broiler and free-range retail chicken in India. Front. Microbiol. 2017, 8, 2120. [Google Scholar] [CrossRef] [PubMed]
  49. Kim, S.; Kim, H.; Kim, Y.; Kim, M.; Kwak, H.; Ryu, S. Whole-genome sequencing-based characteristics in extended-spectrum beta-lactamase-producing Escherichia coli isolated from retail meats in Korea. Microorganisms 2020, 8, 508. [Google Scholar] [CrossRef] [PubMed]
  50. Kim, S.; Kim, H.; Kim, Y.; Kim, M.; Kwak, H.; Ryu, S. Antimicrobial resistance of Escherichia coli from retail poultry meats in Korea. J. Food Prot. 2020, 83, 1673–1678. [Google Scholar] [CrossRef]
  51. Food and Drug Administration. Withdrawal of Enrofloxacin for Poultry. Available online: https://www.fda.gov/animal-veterinary/recalls-withdrawals/withdrawal-enrofloxacin-poultry (accessed on 6 February 2023).
  52. Food and Drug Administration. Extralabel Use and Antimicrobials. Available online: https://www.fda.gov/animal-veterinary/antimicrobial-resistance/extralabel-use-and-antimicrobials (accessed on 6 February 2023).
  53. Kieffer, N.; Royer, G.; Decousser, J.W.; Bourrel, A.S.; Palmieri, M.; De La Rosa, J.M.O.; Jacquier, H.; Denamur, E.; Nordmann, P.; Poirel, L. Mcr-9, an inducible gene encoding an acquired phosphoethanolamine transferase in Escherichia coli, and its origin. Antimicrob. Agents Chemother. 2019, 63, e00965-19. [Google Scholar] [CrossRef]
  54. Tyson, G.H.; Li, C.; Hsu, C.H.; Ayers, S.; Borenstein, S.; Mukherjee, S.; Tran, T.T.; McDermot, P.F.; Zhao, S. The mcr-9 gene of Salmonella and Escherichia coli is not associated with colistin resistance in the United States. Antimicrob. Agents Chemother. 2020, 64, e00573-20. [Google Scholar] [CrossRef]
  55. El-Sayed Ahmed, M.A.E.G.; Zhong, L.L.; Shen, C.; Yang, Y.; Doi, Y.; Tian, G.B. Colistin and its role in the era of antibiotic resistance: An extended review (2000–2019). Emerg. Microbes Infect. 2020, 9, 868–885. [Google Scholar] [CrossRef]
  56. Andrade, F.F.; Silva, D.; Rodrigues, A.; Pina-Vaz, C. Colistin update on its mechanism of action and resistance, present and future challenges. Microorganisms 2020, 8, 1716. [Google Scholar] [CrossRef]
  57. Liu, Y.Y.; Wang, Y.; Walsh, T.R.; Yi, L.X.; Zhang, R.; Spencer, J.; Doi, Y.; Tian, G.; Dong, B.; Huang, X.; et al. Emergence of plasmid-mediated colistin resistance mechanism mcr-1 in animals and human beings in China: A microbiological and molecular biological study. Lancet Infect. Dis. 2016, 16, 161–168. [Google Scholar] [CrossRef] [PubMed]
  58. Gogry, F.A.; Siddiqui, M.T.; Sultan, I.; Haq, Q.M.R. Current update on intrinsic and acquired colistin resistance mechanisms in bacteria. Front. Med. 2021, 8, 677720. [Google Scholar] [CrossRef]
  59. Li, Y.; Dai, X.; Zeng, J.; Gao, Y.; Zhang, Z.; Zhang, L. Characterization of the global distribution and diversified plasmid reservoirs of the colistin resistance gene mcr-9. Sci. Rep. 2020, 10, 8113. [Google Scholar] [CrossRef] [PubMed]
  60. Kananizadeh, P.; Oshiro, S.; Watanabe, S.; Iwata, S.; Kuwahara-Arai, K.; Shimojima, M.; Ogawa, M.; Tada, T.; Kirikae, T. Emergence of carbapenem-resistant and colistin-susceptible Enterobacter cloacae complex co-harboring blaIMP-1 and mcr-9 in Japan. BMC Infect. Dis. 2020, 20, 282. [Google Scholar] [CrossRef]
  61. Carroll, L.M.; Gaballa, A.; Guldimann, C.; Sullivan, G.; Henderson, L.O.; Wiedmann, M. Identification of novel mobilized colistin resistance gene mcr-9 in a multidrug-resistant, colistin-susceptible Salmonella enterica serotype Typhimurium isolate. ASM J. 2019, 10, e00853-19. [Google Scholar] [CrossRef]
  62. Hooper, D.C.; Jacoby, G.A. Topoisomerase inhibitors: Fluoroquinolone mechanisms of action and resistance. Cold Spring Harb. Perspect. Med. 2016, 6, a025320. [Google Scholar] [CrossRef] [PubMed]
  63. Juraschek, K.; Malekzadah, J.; Malorny, B.; Käsbohrer, A.; Schwarz, S.; Meemken, D.; Hammerl, J.A. Characterization of qnrB-carrying plasmids from ESBL- and non-ESBL-producing Escherichia coli. BMC Genom. 2022, 23, 365. [Google Scholar] [CrossRef]
  64. Ríos, E.; Rodríguez-Avial, I.; Rodríguez-Avial, C.; Hernandez, E.; Picazo, J.J. High percentage of resistance to ciprofloxacin and qnrB19 gene identified in urinary isolates of extended-spectrum β-lactamase-producing Escherichia coli in Madrid, Spain. Diagn. Microbiol. Infect. Dis. 2010, 67, 380–383. [Google Scholar] [CrossRef] [PubMed]
  65. Coppola, N.; Freire, B.; Umpiérrez, A.; Cordeiro, N.F.; Ávila, P.; Trenchi, G.; Castro, G.; Casaux, M.L.; Fraga, M.; Zunino, P.; et al. Transferable resistance to highest priority critically important antibiotics for human health in Escherichia coli strains obtained from livestockfeces in Uruguay. Front. Vet. Sci. 2020, 7, 588919. [Google Scholar] [CrossRef]
  66. Chen, Y.; Liu, Z.; Zhang, Y.; Zhang, Z.; Lei, L.; Xia, Z. Increasing prevalence of ESBL-producing multidrug resistance Escherichia coli from diseased pets in Beijing, China from 2012 to 2017. Front. Microbiol. 2019, 10, 2852. [Google Scholar] [CrossRef]
  67. Tyson, G.H.; McDermott, P.F.; Li, C.; Chen, Y.; Tadesse, D.A.; Mukherjee, S.; Bodeis-Jones, S.; Kabera, C.; Gaines, S.A.; Loneragan, G.H.; et al. WGS accurately predicts antimicrobial resistance in Escherichia coli. J. Antimicrob. Chemother. 2015, 70, 2763–2769. [Google Scholar] [CrossRef]
  68. Jeamsripong, S.; Li, X.; Aly, S.S.; Su, Z.; Pereira, R.V.; Atwill, E.R. Antibiotic resistance genes and associated phenotypes in Escherichia coli and Enterococcus from cattle at different production stages on a dairy farm in central California. Antibiotics 2021, 10, 1042. [Google Scholar] [CrossRef] [PubMed]
  69. Lee, K.Y.; Atwill, E.R.; Pitesky, M.; Huang, A.; Lavelle, K.; Rickard, M.; Shafii, M.; Hung-Fan, M.; Li, X. Antimicrobial resistance profiles of non-typhoidal Salmonella from retail meat products in California, 2018. Front. Microbiol. 2022, 13, 835699. [Google Scholar] [CrossRef]
  70. McDermott, P.F.; Tyson, G.H.; Kabera, C.; Chen, Y.; Li, C.; Folster, J.P.; Ayers, S.L.; Lam, C.; Tate, H.P.; Zhao, S. Whole-genome sequencing for detecting antimicrobial resistance in nontyphoidal Salmonella. Antimicrob. Agents Chemother. 2016, 60, 5515–5520. [Google Scholar] [CrossRef]
  71. Neuert, S.; Nair, S.; Day, M.R.; Doumith, M.; Ashton, P.M.; Mellor, K.C.; Jenkins, C.; Hopkins, K.L.; Woodford, N.; de Pinna, E.; et al. Prediction of phenotypic antimicrobial resistance profiles from whole genome sequences of non-typhoidal Salmonella enterica. Front. Microbiol. 2018, 9, 592. [Google Scholar] [CrossRef]
  72. Carroll, L.M.; Buehler, A.J.; Gaballa, A.; Siler, J.D.; Cummings, K.J.; Cheng, R.A.; Wiedmann, M. Monitoring the microevolution of Salmonella enterica in healthy dairy cattle populations at the individual farm level using whole-genome sequencing. Front. Microbiol. 2021, 12, 763669. [Google Scholar] [CrossRef] [PubMed]
  73. Arredondo-Alonso, S.; Willems, R.J.; van Schaik, W.; Schürch, A.C. On the (im)possibility of reconstructing plasmids from whole-genome short-read sequencing data. Microb. Genomics 2017, 3. [Google Scholar] [CrossRef] [PubMed]
  74. Tyson, G.H.; Li, C.; Hsu, C.H.; Bodeis-Jones, S.; McDermott, P.F. Diverse fluoroquinolone resistance plasmids from retail meat E. coli in the United States. Front. Microbiol. 2019, 10, 2826. [Google Scholar] [CrossRef]
  75. Long, Y.; Lu, X.; Ni, X.; Liu, J.; Wang, M.; Li, X.; Li, Z.; Zhou, H.; Li, Z.; Wu, K.; et al. High carriage rate of the multiple resistant plasmids harboring quinolone resistance genes in Enterobacter spp. isolated from healthy individuals. Antibiotics 2022, 11, 15. [Google Scholar] [CrossRef] [PubMed]
  76. Wang, Q.; Lei, C.; Cheng, H.; Yang, X.; Huang, Z.; Chen, X.; Ju, Z.; Zhang, H.; Wang, H. Widespread dissemination of plasmid-mediated tigecycline resistance gene tet(X4) in Enterobacterales of porcine origin. Microbiol. Spectr. 2022, 10, e01615-22. [Google Scholar] [CrossRef]
  77. Melnyk, A.H.; Wong, A.; Kassen, R. The fitness costs of antibiotic resistance mutations. Evol. Appl. 2015, 8, 273–283. [Google Scholar] [CrossRef]
  78. Schrag, S.; Perrot, V.; Levin, B. Adaptation to the fitness costs of antibiotic resistance in Escherichia coli. Proc. R. Soc. B Biol. Sci. 1997, 264, 1287–1291. [Google Scholar] [CrossRef] [PubMed]
  79. Food and Drug Administration. NARMS Retail Meat Surveillance Laboratory Protocol. Available online: https://www.fda.gov/media/93332/download (accessed on 6 February 2023).
  80. Food and Drug Administration. The National Antimicrobial Resistance Monitoring System Manual of Laboratory Methods. Available online: https://www.fda.gov/media/101423/download (accessed on 6 February 2023).
  81. Food and Drug Administration. 2021 NARMS Interpretative Criteria for Susceptibility Testing. Available online: https://www.fda.gov/media/108180/download (accessed on 6 February 2023).
  82. Magiorakos, A.P.; Srinivasan, A.; Carey, R.B.; Carmeli, Y.; Falagas, M.E.; Giske, C.G.; Harbarth, S.; Hindler, J.F.; Kahlmeter, G.; Olsson-Liljequist, B.; et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: An international expert proposal for interim standard definitions for acquired resistance. Clin. Microbiol. Infect. 2012, 18, 268–281. [Google Scholar] [CrossRef] [PubMed]
  83. Oksanen, J.; Simpson, G.L.; Blanchet, F.G.; Kindt, R.; Legendre, P.; Minchin, P.R.; O’Hara, R.B.; Solymos, P.; Stevens, M.H.H.; Szoecs, E.; et al. Vegan: Community Ecology Package 2022. Available online: https://cran.r-project.org/web/packages/vegan/vegan.pdf (accessed on 6 February 2023).
  84. Anderson, M.J. Distance-based tests for homogeneity of multivariate dispersions. Biometrics 2006, 62, 245–253. [Google Scholar] [CrossRef]
  85. Anderson, M.J.; Ellingsen, K.E.; McArdle, B.H. Multivariate dispersion as a measure of beta diversity. Ecol. Lett. 2006, 9, 683–693. [Google Scholar] [CrossRef] [PubMed]
  86. Bastian, M.; Heymann, S.; Jacomy, M. Gephi: An open source software for exploring and manipulating networks. Proc. Int. AAAI Conf. Web Soc. Media 2009, 3, 361–362. [Google Scholar] [CrossRef]
Antibiotics 12 00782 g001 550
Figure 1. Genomic antimicrobial resistance in E. coli isolates from retail meat (n = 52), by (a) distribution of antimicrobial resistance genes and (b) concordance with phenotypic antimicrobial resistance. STR: streptomycin; GEN: gentamicin; AMC: amoxicillin/clavulanic acid; FOX: cefoxitin; AXO: ceftriaxone; COT: trimethroprim-sulfamethoxazole; AMP: ampicillin; CHL: chloramphenicol; NAL: nalidixic acid; CIP: ciprofloxacin; TET: tetracycline.
Figure 1. Genomic antimicrobial resistance in E. coli isolates from retail meat (n = 52), by (a) distribution of antimicrobial resistance genes and (b) concordance with phenotypic antimicrobial resistance. STR: streptomycin; GEN: gentamicin; AMC: amoxicillin/clavulanic acid; FOX: cefoxitin; AXO: ceftriaxone; COT: trimethroprim-sulfamethoxazole; AMP: ampicillin; CHL: chloramphenicol; NAL: nalidixic acid; CIP: ciprofloxacin; TET: tetracycline.
Antibiotics 12 00782 g001
Antibiotics 12 00782 g002 550
Figure 2. Non-metric multidimensional scaling (NMDS) performed using presence and absence of antimicrobial resistance genes from E. coli isolates from retail meat (n = 52). Individual points represent an isolate, with convex hulls displaying grouping factors of (a) meat type; (b) season; (c) packaging type; and (d) label claim.
Figure 2. Non-metric multidimensional scaling (NMDS) performed using presence and absence of antimicrobial resistance genes from E. coli isolates from retail meat (n = 52). Individual points represent an isolate, with convex hulls displaying grouping factors of (a) meat type; (b) season; (c) packaging type; and (d) label claim.
Antibiotics 12 00782 g002
Antibiotics 12 00782 g003 550
Figure 3. Co-occurrence networks of antimicrobial resistance genes in E. coli isolates (n = 52) by frequency thresholds of (a) 1 isolate; (b) 5 isolates; and (c) 10 isolates. Each node represents a gene and is color coded by antimicrobial class. Edges indicate frequency of co-occurrence, with a light to dark gradient representing low to high co-occurrence, respectively.
Figure 3. Co-occurrence networks of antimicrobial resistance genes in E. coli isolates (n = 52) by frequency thresholds of (a) 1 isolate; (b) 5 isolates; and (c) 10 isolates. Each node represents a gene and is color coded by antimicrobial class. Edges indicate frequency of co-occurrence, with a light to dark gradient representing low to high co-occurrence, respectively.
Antibiotics 12 00782 g003
Table
Table 1. Risk factors for presence of E. coli in retail meat products from southern California.
Table 1. Risk factors for presence of E. coli in retail meat products from southern California.
Risk Factor Univariable ModelsMultivariable Model
E. coli Positive n/N (%)OR (95% CI)p-ValueOR (95% CI)p-Value
Meat Type
        Chicken 38/56 (67.86%)8.44 (3.55, 20.09)<0.00019.43 (3.84, 23.21)<0.0001
        Ground Turkey38/54 (70.37%)9.50 (3.93, 22.95)<0.000111.00 (4.39, 27.54)<0.0001
        Pork Chop18/56 (32.14%)1.90 (0.80, 4.51)0.151.93 (0.79, 4.71)0.15
        Ground Beef11/55 (20.00%)1.00-1.00-
Season
        Spring22/60 (36.67%)0.30 (0.13, 0.69)0.00460.22 (0.083, 0.55)0.0015
        Summer28/60 (46.67%)0.45 (0.20, 1.03)0.05920.36 (0.14, 0.91)0.031
        Autumn28/60 (46.67%)0.45 (0.20, 1.03)0.05920.36 (0.14, 0.91)0.031
        Winter27/41 (65.85%)1.00-1.00-
Packaging type
        Modified atmosphere packaging (MAP)55/85 (64.71%)3.40 (1.55, 7.48)0.0023--
        Plastic film26/64 (40.63%)1.27 (0.56, 2.88)0.57--
        Other (Vacuum, chub, paper)10/32 (31.25%)0.84 (0.31, 2.27)0.74--
        Plastic bag14/40 (35.00%)1.00---
Label Claim
        Reduced antibiotic claim27/37 (72.97%)3.67 (1.68, 8.02)0.0011--
        Conventional78/184 (42.39%)1.00---
Presence of Salmonella
        Yes8/11 (72.73%)3.11 (0.80, 12.04)0.10--
        No97/210 (46.19%)1.00---
Overall prevalence105/221 (47.51%)----
Table
Table 2. Distribution of phenotypic antimicrobial resistance in E. coli isolates from retail meat samples (n = 105).
Table 2. Distribution of phenotypic antimicrobial resistance in E. coli isolates from retail meat samples (n = 105).
Number of Isolates Resistant to Antimicrobial Drugs (%)
CLSI ClassAntimicrobial AgentChicken
(n = 38)		Ground Turkey
(n = 38)		Pork Chop (n = 18)Ground Beef (n = 11)All Samples
(n = 105)
AminoglycosidesSTR10 (26.32%)17 (44.74%)5 (27.78%)0 (0%)32 (30.48%)
GEN9 (23.68%)7 (18.42%)1 (5.56%)0 (0%)17 (16.19%)
Beta-lactam combination agentsAMC1 (2.63%)2 (5.26%)0 (0%)0 (0%)3 (2.86%)
CephemsFOX1 (2.63%)1 (2.63%)0 (0%)0 (0%)2 (1.90%)
AXO1 (2.63%)1 (2.63%)0 (0%)0 (0%)2 (1.90%)
Folate pathway antagonistsCOT0 (0%)2 (5.26%)0 (0%)0 (0%)2 (1.90%)
MacrolidesAZI0 (0%)0 (0%)0 (0%)0 (0%)0 (0%)
PenemsMER0 (0%)0 (0%)0 (0%)0 (0%)0 (0%)
PenicillinsAMP5 (13.16%)14 (36.84%)3 (16.67%)0 (0%)22 (20.95%)
PhenicolsCHL0 (0%)2 (5.26%)0 (0%)0 (0%)2 (1.90%)
Quinolones NAL0 (0%)1 (2.63%)1 (5.56%)0 (0%)2 (1.90%)
CIP0 (0%)1 (2.63%)0 (0%)0 (0%)1 (0.95%)
TetracyclineTET15 (39.47%)24 (63.16%)7 (38.89%)0 (0%)46 (43.81%)
STR: streptomycin; GEN: gentamicin; AMC: amoxicillin/clavulanic acid; FOX: cefoxitin; AXO: ceftriaxone; COT: trimethroprim-sulfamethoxazole; AZI: azithromycin; MER: meropenem; AMP: ampicillin; CHL: chloramphenicol; NAL: nalidixic acid; CIP: ciprofloxacin; TET: tetracycline.
Table
Table 3. Association between retail meat type and phenotypic antimicrobial resistance.
Table 3. Association between retail meat type and phenotypic antimicrobial resistance.
Meat TypeAntimicrobial Drug (Abbreviation)
Ampicillin (AMP)Gentamicin (GEN)Streptomycin (STR)Tetracycline (TET)
OR (95% CI)p-ValueOR (95% CI)p-ValueOR (95% CI)p-ValueOR (95% CI)p-Value
Chicken1.31 (0.23, 9.20)1.008.47 (1.05, 394.18)0.0421.70 (0.45, 7.27)0.562.03 (0.63, 7.08)0.29
Ground turkey4.94 (1.17, 30.11)0.0256.18 (0.72, 294.92)0.133.81 (1.10, 15.56)0.0325.24 (1.65, 18.55)0.003
Non-poultry (beef, pork)1.00-1.00-1.00-1.00-
Table
Table 4. Distribution of phenotypic antibiogram patterns in E. coli isolates (n = 105).
Table 4. Distribution of phenotypic antibiogram patterns in E. coli isolates (n = 105).
Antibiogram PatternNumber of Isolates (n/N %)Multidrug Resistant
Susceptible to all drugs in MIC panel51 (48.57%)No
TET12 (11.43%)No
STR-TET8 (7.62%)No
AMP-TET5 (4.76%)No
GEN-STR-TET5 (4.76%)No
GEN-STR4 (3.81%)No
AMP2 (1.90%)No
AMP-GEN-STR-TET2 (1.90%)Yes
AMP-GEN-TET2 (1.90%)Yes
AMP-STR-TET2 (1.90%)Yes
AMC-AMP-FOX-AXO-GEN-STR-TET1 (0.95%)Yes
AMC-AMP-FOX-AXO-GEN-TET1 (0.95%)Yes
AMC-AMP-STR-TET1 (0.95%)Yes
AMP-CHL-STR-TET1 (0.95%)Yes
AMP-CHL-STR-TET-COT1 (0.95%)Yes
AMP-GEN-NAL-STR-TET-CIP1 (0.95%)Yes
AMP-GEN-STR1 (0.95%)No
AMP-NAL-STR-TET-CIPdsc *1 (0.95%)Yes
AMP-STR-TET-CIPdsc * 1 (0.95%)Yes
STR1 (0.95%)No
STR-TET-CIPdsc1 (0.95%)No
STR-TET-COT1 (0.95%)Yes
* CIPdsc denotes decreased susceptibility to ciprofloxacin. STR: streptomycin; GEN: gentamicin; AMC: amoxicillin/clavulanic acid; FOX cefoxitin: AXO ceftriaxone; COT: trimethroprim-sulfamethoxazole; AZI: azithromycin: MER: meropenem; AMP: ampicillin; CHL: chloramphenicol; NAL: nalidixic acid; CIP: ciprofloxacin; TET: tetracycline.
Table
Table 5. Results of PERMDISP2, PERMANOVA, and ANOSIM tests based on the presence and absence of AMR genetic determinants. All tests were performed using a Jaccard distance metric and 10,000 permutations.
Table 5. Results of PERMDISP2, PERMANOVA, and ANOSIM tests based on the presence and absence of AMR genetic determinants. All tests were performed using a Jaccard distance metric and 10,000 permutations.
Grouping FactorPERMDISP2 p-Value (F)PERMANOVA p-Value (R2)ANOSIM p-Value (R)
Meat type0.010 (4.25)0.10 (0.088)0.89 (−0.059)
Season0.41 (0.97)0.73 (0.044)0.80 (−0.022)
Packaging type0.78 (0.36)0.55 (0.053)0.24 (0.038)
Label claim0.39 (0.78)0.80 (0.0094)0.87 (−0.11)
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Lee, K.Y.; Lavelle, K.; Huang, A.; Atwill, E.R.; Pitesky, M.; Li, X. Assessment of Prevalence and Diversity of Antimicrobial Resistant Escherichia coli from Retail Meats in Southern California. Antibiotics 2023, 12, 782. https://doi.org/10.3390/antibiotics12040782

AMA Style

Lee KY, Lavelle K, Huang A, Atwill ER, Pitesky M, Li X. Assessment of Prevalence and Diversity of Antimicrobial Resistant Escherichia coli from Retail Meats in Southern California. Antibiotics. 2023; 12(4):782. https://doi.org/10.3390/antibiotics12040782

Chicago/Turabian Style

Lee, Katie Yen, Kurtis Lavelle, Anny Huang, Edward Robert Atwill, Maurice Pitesky, and Xunde Li. 2023. "Assessment of Prevalence and Diversity of Antimicrobial Resistant Escherichia coli from Retail Meats in Southern California" Antibiotics 12, no. 4: 782. https://doi.org/10.3390/antibiotics12040782

APA Style

Lee, K. Y., Lavelle, K., Huang, A., Atwill, E. R., Pitesky, M., & Li, X. (2023). Assessment of Prevalence and Diversity of Antimicrobial Resistant Escherichia coli from Retail Meats in Southern California. Antibiotics, 12(4), 782. https://doi.org/10.3390/antibiotics12040782

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop