Prevalence and Molecular Epidemiology of Extended-Spectrum-β-Lactamase (ESBL)-Producing Escherichia coli from Multiple Sectors of Poultry Industry in Korea

The aim of this study was to investigate the molecular epidemiology of extended-spectrum-β-lactamase producing Escherichia coli (ESBL-EC) from poultry, the poultry farm environment, and workers in Korea. A total of 1376 non-duplicate samples were collected from 21 poultry farms, 20 retail stores, 6 slaughterhouses, and 111 workers in a nationwide study in Korea from January 2019 to August 2019. The overall positive rate of ESBL-EC was 6.8%, with variable positive rates according to sources (0.9% of worker, 5.2% of poultry, 10.0% of chicken meat, and 14.3% of environment). Common ESBL types were CTX-M-55 and CTX-M-14 in a total of 93 ESBL-EC isolates. Whole genome sequencing revealed that 84 ESBL-EC isolates had an outstanding accumulation of numerous antimicrobial resistance (AMR) genes associated with resistance to various classes of antimicrobials for human use and well-known antimicrobial gene (ARG)-carrying plasmids. Core gene multi locus sequence typing, using 2390 core genes, indicated no dominant clone or common type in each province. In conclusion, the isolation rates of ESBL-EC were not negligible in the poultry industry-related samples, sharing common ESBL types of human ESBL-EC isolates in Korea.


Introduction
Bacteria from livestock may carry clinically relevant resistance genes associated with resistance to veterinary and human antimicrobials [1]. Exposure to multidrug-resistant bacteria during farming, slaughtering, and distribution can serve as a vector of transmission to occupationally exposed workers [1]. To establish an effective strategy for the control of AMR, it is very important to understand the current epidemiology and transmission mode of extended-spectrum-β-lactamase-producing Escherichia coli (ESBL-EC) among humans, food animals, retail meat, and environments.
E. coli is now the most frequently isolated ESBL-carrier, and CTX-Ms have become the most frequently isolated ESBLs [2]. CTX-M β-lactamases were first reported in Japan in 1986 [3]. Since 2000, CTX-M β-lactamases have been reported increasingly in both humans and animals and are now the dominant type of ESBL, replacing classic TEM-and SHV-type ESBLs in most parts of the world [4]. Currently, there are more than 120 different CTX-M β-lactamases clustered in five groups (CTX-M-1, 2, 8, 9, and 25) [5]. Among the CTX-M-type enzymes, CTX-M-15 and CTX-M-14 have repeatedly been reported in most areas of the world, including in South Korea [6].
ESBL-EC has been isolated from cloacal and boot swabs in farms [7], wastewater and process water from slaughterhouses [8], and commercialized chicken carcasses [9].
However, information about the current epidemiology is very limited in South Korea except for a few reports about ESBL-EC in the commercial layer of chicken farms [10] and poultry slaughterhouses [11].
Thus, the purpose of this study was to evaluate the current prevalence and molecular epidemiology of ESBL-EC in multiple sectors of the poultry-related industry in Korea. Data about the transmission of antimicrobial resistance mechanisms among three sectors (humananimal-environment) could be used to introduce a monitoring system for surveillance and develop an effective strategy for the control of AMR.
ST 93 is the most common (17.9%) with many heterogeneous sequence types (STs). The worldwide pandemic clone, ST131 E. coli, was detected from only one sample of the farm environment (Table S3). Statistically, the distribution of STs were different by provinces, but not by sources (Table S3). Pairwise post-hoc analysis showed significant differences in the distribution of STs between Seoul/Gyeonggi and Gangwon (p-value < 0.0001, Table S4). Core gene multi locus sequence typing (cgMLST), using 2390 core genes, indicated no dominant clone or common type in each province, meaning that most of the ESBL-EC isolates were not associated with each other ( Figure S1).

Discussion
The resistance network among environmental, community, and nosocomial superbugs has been suggested [1]. The widespread use of antibiotics has resulted in the generation of antibiotic concentration gradients in humans and livestock. It has accelerated the emergence and spread of antibiotic-resistant bacteria among humans and animals [12]. Antibiotic use in food-producing animals can cause the selection of antibiotic-resistant bacteria in animals, contaminate surrounding environments, and change the environment microbiome, which may serve as a resistance pool [13].
In this study, ESBL-EC isolation from the swine industry-related samples was not negligible, with an overall positive rate of 6.8%. Especially, positive rates of ESBL-EC isolation from chicken meat samples were as high as 10.0%. This is worrisome in aspects of food safety. Many poultry-related ESBL-EC can produce CTX-M-14 and CTX-M-15 commonly detected in clinical samples [14,15].
We could not detect carbapenemase-producers and colsitisn-resistant E. coli in this study. The spread of carbapenemase-producing enterobacteriaceae (CPE) from foodanimals or the environment to humans is also an essential concern of public health. Recently, NDM-5-producing E. coli isolated from a dog and a cat in Korea has been reported [16]. However, the information about the isolation of the mcr gene from food animal-originated E. coli is still limited in Korea [17,18].
Plasmids as diverse and self-replicating extrachromosomal elements and specific plasmid types have been associated with virulence and/or AMR [19]. Interestingly, IncB/O/K/Z, IncFIA, IncFII, and IncQ1 plasmids known to carry carbapenamase genes were detected in this study. For example, an E. coli isolate carrying an IncB/O/K/Z-like plasmid encoding a bla CMY-2 gene confers resistance to carbapenems with an overexpression of CMY-2 and high plasmid copies [20]. IncFIA plasmid-carrying bla NDM-1 among Klebsiella pneumoniae and Enterobacter cloacae isolates [21], and IncFII plasmid encoding NDM-1 from ST131 E. coli have been reported [22]. IncQ1 plasmids harbor bla KPC-2 in high-risk lineages of K. pneumoniae CG258 [23]. Considering the spread of these plasmids in poultry-related ESBL-EC, continuous monitoring of CPE from food-animals or the environment is essential.
The cgMLST indicated no dominant clone or common type in each province, meaning that most ESBL-EC isolates were not associated with each other. This suggests that there is no definite evidence about the emergence of animal-specific clones like livestock-associated methicillin-resistant Staphylococcus aureus [25]. In this study, clinically common genotypes of ESBL were prevalent in poultry and chicken meat samples. Thus, an indirect contact transmission between humans and chickens should be considered.

Study Design, Duration, Sampling and Isolation
We collected 1376 non-duplicate samples from 21 poultry farms, 20 retail stores, 6 slaughterhouses, and 111 workers in a nationwide study in Korea from January 2019 to August 2019. These samples included swabs from the following: nose, groin, axillary, antecubital fossa, inter-finger spaces, and stool of workers; human pooled stool in a toilet; nose, skin, rectum, and stool of poultry; fence, floor, ventilation fan of poultry farms; knife, cutting board, and floor of slaughterhouse; knife, cutting board, and showcase of retail shop. Swabbed samples were put into transport media (Copan diagnostics, Murrieta, CA, USA). Soil (25 g), wastewater (1 L), slaughtered poultry (10 g), and chicken meat for sale (10 g) were sampled in sterile containers. Samples were stored at 4 • C before analysis and inoculated in media within 24 h of sampling.
The soil was mixed with 225 mL of buffered peptone water (BD Biosciences, San Jose, CA, USA). The wastewater was filtered with a 0.2 µm filter. The content on the surface of the filter was inoculated into 10 mL of Mueller-Hinton (MH) broth. The slaughtered poultry and chicken meat were inoculated into 10 mL of MH broth under sterile conditions. All inoculated broths were incubated at 36 ± 1 • C for 18-24 h. After enrichment, 10-100 µL of the liquid culture was used for inoculation onto ChromeIDESBL (bioMérieux, Marcy l'E'toile, France) and MacConkey agar (Oxoid, Basingstoke, UK). Swabbed samples were directly inoculated on ChromeIDESBL (bioMérieux) and MacConkey agar (Oxoid). All inoculated agars were incubated at 36 ± 1 • C for 18-24 h.

Strain Identification, Colistin Resistance Screening and Phenotypic Detection of Carbapenemases
The identification of E. coli was conducted with the MALDI Biotyper (Bruker Daltonik, Bremen, Germany). The disk diffusion method was used for antimicrobial susceptibility. The diameter of the inhibition zone was interpreted according to CLSI criteria [26]. To detect colistin-resistant isolates, test organisms were screened on MH agar (Oxoid, Basingstoke, UK) containing colistin (0, 1, 2, and 4 µg/mL) with an E. coli ATCC™25922 strain. If MIC was >2 mg/L, the isolate was regarded as a colistin-resistant organism [26]. The CarbaNP [27] or modified carbapenem inactivation method was applied [28] if the isolates exhibited carbapenem resistance without compatible carbapenemase genes.

Molecular Characterization of ESBL Genes and Whole Genome Sequencing of ESBL-E. coli Isolates
The PCR and sequencing for ESBL genes (bla TEM , bla SHV , and bla CTX-M ) (Table S5) were performed as described in a previous study [29] for the isolates showing resistance to cefotaxime or ceftazidime. We performed WGS for a total of 84 ESBL-EC. Nine ESBL-EC isolates were excluded because of the repeated failure of the subculture. DNAs of freshly sub-cultured isolates were extracted using a GenElute™ Bacterial Genomic DNA Kit (Sigma-Aldrich, St. Louis, MO, USA) and 8 µg of input genomic DNA was used. Entire genomes of ESBL-EC isolates were sequenced using a NextSeq 550 instrument (Illumina, San Diego, CA, USA). Sequences were assembled with Spades (version 3.11.1) and annotated with Prokka (version 1.13.7). Data of MLST, cgMLST, plasmid typing, and resistance genes were obtained from the website of the Center for Genomic Epidemiology, including MLST 2.0, ResFinder 3.2, and PlasmidFinder 2.0 [30].

Statistical Analysis
Categorical variables (source, province, ESBL type) were described using count and percentage of the group from which they were derived. The relative proportion was defined as the percentage of the total. The prevalence or positive rate was derived by comparing the number of samples that had ESBL-EC with the total number of samples studied. It was expressed as a percentage. Multiple samples from the same poultry were calculated as one sample.
The Chi-square test was used for comparative analysis of categorical variables using IBM SPSS Statistics for Windows software version 23.0 (IBM Corp., Armonk, NY, USA). The statistical significance of the results was defined at p < 0.05. Post-hoc pairwise Fisher's exact tests with Bonferroni's method were performed to identify which provinces and sources differed significantly, when significant differences (p-value < 0.05) were detected. Bonferroni's correction method was applied to correct multiple tests; the corrected significance limit was a p-value < 0.005 (0.05/10).
This study was approved by the Institutional Review Board of National Health Insurance Service Ilsan Hospital, Goyang, Korea as required by hospital policy (approval number: NHIMC 2017-07-041).

Conclusions
This study aimed to investigate the molecular epidemiology of extended-spectrum-βlactamase producing Escherichia coli (ESBL-EC) from poultry, the poultry farm environment, and workers in Korea. The overall positive rate of ESBL-EC was 6.8%, with variable positive rates according to sources (0.9% in workers, 5.2% in poultry, 10.0% in chicken meat, and 14.3% in the environment). Common ESBL types were CTX-M-55 and CTX-M-14 in a total of 93 ESBL-EC isolates. In conclusion, the proportion of ESBL-EC of poultry industry-related samples was not negligible in Korea. Such epidemiologic data could be used to develop evidence-based policies to reduce AMR and improve animal and human health with the 'one health' concept.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/antibiotics10091050/s1, Table S1: Antimicrobial susceptibility results of E. coli isolates (n = 165) from poultry industry-related samples, Table S2: Resistance genes, and plasmid types of ESBL-EC isolates from poultry industry-related samples (N = 84), Table S3: The distribution of sequence types among different provinces and sources, Table S4: Post-hoc pairwise Fisher's exact tests p-value in the distribution of sequence types by different provinces. Table S5: ESBL genotype pcr-sequencing primers and reaction conditions. Figure S1: The distribution of cgMLST of ESBL-EC isolates from poultry industry-related samples (N = 84) according to provinces.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data would be available upon request.

Conflicts of Interest:
The authors declare no conflict of interest.