Genome-Based Characterization of Hybrid Shiga Toxin-Producing and Enterotoxigenic Escherichia coli (STEC/ETEC) Strains Isolated in South Korea, 2016–2020

The global emergence of hybrid diarrheagenic E. coli strains incorporating genetic markers from different pathotypes is a public health concern. Hybrids of Shiga toxin-producing and enterotoxigenic E. coli (STEC/ETEC) are associated with diarrhea and hemolytic uremic syndrome (HUS) in humans. In this study, we identified and characterized STEC/ETEC hybrid strains isolated from livestock feces (cattle and pigs) and animal food sources (beef, pork, and meat patties) in South Korea between 2016 and 2020. The strains were positive for genes from STEC and ETEC, such as stx (encodes Shiga toxins, Stxs) and est (encodes heat-stable enterotoxins, ST), respectively. The strains belong to diverse serogroups (O100, O168, O8, O155, O2, O141, O148, and O174) and sequence types (ST446, ST1021, ST21, ST74, ST785, ST670, ST1780, ST1782, ST10, and ST726). Genome-wide phylogenetic analysis revealed that these hybrids were closely related to certain ETEC and STEC strains, implying the potential acquisition of Stx-phage and/or ETEC virulence genes during the emergence of STEC/ETEC hybrids. Particularly, STEC/ETEC strains isolated from livestock feces and animal source foods mostly exhibited close relatedness with ETEC strains. These findings allow further exploration of the pathogenicity and virulence of STEC/ETEC hybrid strains and may serve as a data source for future comparative studies in evolutionary biology.


Introduction
Escherichia coli is commonly regarded as a nonpathogenic beneficial inhabitant of the gastrointestinal tract. However, several pathogenic strains have acquired specific virulence factors that are responsible for various intestinal and extraintestinal diseases, including diarrhea, acute inflammation, hemorrhagic colitis, urinary tract infections, septicemia, and neonatal meningitis. Diarrheagenic Escherichia coli (DEC) causes 30-40% of acute diarrhea episodes in children <5 years in developing countries [1]. According to the WHO Global Burden of Foodborne Diseases report, >300 million illnesses and nearly 200,000 deaths are caused by DEC globally each year [2]. Major diarrheagenic E. coli (DEC) strains are subdivided into several pathotypes based on the presence of specific virulence traits directly related to disease development [3][4][5][6]. The DEC pathotypes include enteropathogenic E. coli (EPEC), Shiga toxin-producing E. coli (STEC), enterotoxigenic E. coli (ETEC), enteroinvasive E. coli (EIEC), and enteroaggregative E. coli (EAEC). Many of these pathotypes are foodborne pathogens that raise public health concerns and cause several outbreaks in industrialized and developing countries [7][8][9]. STEC and ETEC are major causes of diarrhea in humans and animals worldwide. STEC is characterized by the presence of the Shiga toxin 1 or 2 genes (stx 1 or stx 2 ), which are generally acquired by a lambda-like bacteriophage [10]. Shiga toxins 1 and 2 (Stx1 and Stx2, respectively) differ in their virulence and host specificity, with Stx2 being most commonly associated with severe illnesses (hemolytic uremic syndrome (HUS), hospitalization, and bloody diarrhea) in humans [11,12]. ETEC is characterized by its ability to produce either a heat-labile (LT) or heat-stable (ST) enterotoxin and carries a diverse set of colonization factors (CFs) for adherence to the intestinal epithelium [13]. It is a major cause of diarrhea among children living in and tourists traveling to developing countries.
Hybrid DEC strains that combine genetic markers belonging to different pathotypes have emerged worldwide and are a public health concern [14]. Numerous virulence markers are frequently carried on mobile genetic elements (MGEs), such as phages and plasmids, allowing the transmission of virulence genes via horizontal gene transfer, leading to the emergence of hybrid pathotypes [3,[15][16][17][18]. Hybrid E. coli strains comprising genetic markers of different pathotypes have been identified owing to the technological advances that provide a better understanding of the genomic and virulence mechanisms of DEC [19].
The most well-documented example is the E. coli O104:H4 strain, which caused a severe outbreak of acute gastroenteritis and HUS in Germany in 2011 [20]. This strain produced Stx2, a signature feature of the STEC pathotype, and it carried a plasmid containing the genes encoding aggregative adherence fimbriae (AAF), which mediate aggregative adherence in EAEC [21][22][23]. Furthermore, hybrids of STEC and ETEC strains (STEC/ETEC) have been recently reported in various countries, including Bangladesh, Sweden, and South Korea, some of which have been associated with diarrheal diseases and HUS in humans [24][25][26][27][28].
Few studies have reported the virulence and antibiotic resistance profiles of STEC/ETEC hybrid strains isolated from livestock feces (cattle and pigs) and animal source foods (beef, pork, and meat patties) in South Korea. This study investigated the genomes of STEC/ETEC hybrid strains to identify the virulence and antibiotic resistance genes they harbored and to determine their phylogenetic position among other E. coli strains. The genomic properties of these strains were investigated via real-time PCR and whole-genome sequencing (WGS). Phylogenetic analysis was performed to assess their phylogeny in a collection of E. coli strains from diverse pathotypes. Based on our findings, we addressed the potential importance of these hybrid E. coli strains for public health.

Bacterial Strains and Serotyping
Pathogenic E. coli strains that originated from the Korean Culture Collection for Foodborne Pathogens (Ministry of Food and Drug Safety) were identified. All 1025 pathogenic E. coli strains isolated in South Korea between 2016 and 2020 were analyzed. Twenty-seven hybrid Shigatoxigenic and enterotoxigenic Escherichia coli (STEC/ETEC) strains were isolated from livestock feces (cattle and pigs) and animal source foods (beef, pork, and meat patties). The strains selected for this study are listed in Table 1. Typical E. coli colonies (blue-green color) on 5-bromo-4-chloro-3-indolyl-β-D-glucuronide (BCIG) agar (Oxoid, UK) were sub-cultured on Tryptic Soy Agar (Oxoid, UK) and then incubated at 37 • C for 18-24 h. The isolates were identified using VITEK MS (BioMerieux Inc., Marcy-l'Etoile, France). The serotype was determined by the agglutination of the bacteria with specific somatic (O1 to O181) antisera [Laboratorio de Referencia de E. coli (LREC), Lugo, Spain] to identify variants of the somatic (O) antigens [29][30][31].

Antimicrobial Susceptibility Tests
Antimicrobial susceptibility tests were performed using Sensititre KRN6F panels (Trek Diagnostic Systems, Cleveland, OH, USA) following the manufacturer's instructions. The antimicrobial susceptibility of the isolated strains was determined using the 16 antimicrobials described as follows: amoxicillin-clavulanic acid, ampicillin, cefoxitin, cefotaxime, ceftazidime, cefepime, chloramphenicol, ciprofloxacin, colistin, gentamicin, meropenem, nalidixic acid, streptomycin, sulfisoxazole, tetracycline, and trimethoprimsulfamethoxazole. The MIC (minimum inhibitory concentration) value of these antimicrobials was determined with the microbroth dilution method. The Clinical and Laboratory Standards Institute guidelines and the U.S. National Antimicrobial Resistance Monitoring System were used to interpret susceptibility results expressed as MICs. For these agents, the degree of increase in resistance was determined by referring to the resistance level of the standard strain, ATCC 25922.

Real-Time PCR-Based Identification of Hybrid Strains
DNA was extracted from the bacterial cultures using automated equipment (EZ1 Advanced XL, Qiagen, Germantown, MD, USA) according to the manufacturer's instructions. The extracted DNA was used as a template for real-time PCR, which was performed using a PowerCheck TM 20/15 Pathogen Multiplex Real-time PCR kit (Kogene Biotech Co., Ltd., Seoul, Korea) to detect virulence genes. Amplification was performed using an ABI 7500 Fast Real-time PCR system (Applied Biosystems, Waltham, MA, USA) at 50 • C for 2 min for 1 cycle, 95 • C for 10 min for 1 cycle, followed by 40 cycles at 95 • C for 15 s, 60 • C for 1 min. The following genes from different DEC pathogens were detected: VT1 and VT2 (STEC); bfpA and eaeA (EPEC); LT, STh, and STp (ETEC); aggR (EAEC); ipaH (EIEC).

Genome Sequencing, Assembly, and Annotation
Genomic DNA was extracted using the MagListo TM 5M Genomic DNA Extraction Kit (Bioneer, Daejeon, Korea) according to the manufacturer's protocol. DNA integrity and concentration were determined using standard agarose gel electrophoresis and a Qubit TM 3.0 Fluorometer (Life Technologies, Carlsbad, CA, USA), respectively. A DNA library was prepared using a Nextera DNA Flex Library Prep Kit (Illumina, San Diego, CA, USA). Sequencing was performed using a MiSeq sequencing system (Illumina) and MiSeq Reagent Kit v3 (600-cycle) (Illumina). The contigs (FASTQ sequence files) were assembled de novo using the CLC Workbench (version 12.0; Qiagen, Hilden, Germany). To obtain high-quality data and determine the complete genomic sequence, the hybrid genome was assembled using additional long-read sequence data obtained from PacBio Sequel (Pacific Bioscience, Menlo Park, CA, USA). Hybrid assembly of raw FASTQ PacBio sequence long-read sequence data and Illumina MiSeq short-read FASTQ sequence data was performed using Unicycler (v0.4.9; https://github.com/rrwick/Unicycler (accessed on 1 January 2023); default settings). The assembled genome was annotated using the Rapid Annotation using Subsystem Technology (RAST) toolkit in the PATRIC genome annotation web service (v3.6.12).

Prophage Prediction and Analysis
Bacteriophage sequences within the unique genome sequence were identified using PHAge Search Tool Enhanced Release (PHASTER) [37]. PHASTER was used to predict putative prophage regions as "intact (score > 90)," "questionable (score 70-90)," or "incomplete (score < 70)" based on the proportion of phage-related genes in the identified phage region of the assembled hybrid genome. The extracted prophage sequences were annotated to identify virulence genes using the RAST toolkit in the PATRIC genome annotation web service (v3.6.12).

Identification of Plasmid-Associated Sequences
Plasmid features of the assembled hybrid genomes were analyzed using PlasmidFinder 2.1 [38]. The threshold and minimum coverage for identification were set to 95% and 60%, respectively. The identification was based on the detection of replicon sequences belonging to several known plasmid incompatibility (Inc) groups. The extracted plasmid sequences were annotated to identify virulence genes using the RAST toolkit in the PATRIC genome annotation web service (v3.6.12).

Phylogenetic Analysis and Population Structure Analysis
Comparative genomic analysis was performed on 27 STEC/ETEC hybrid strains isolated from livestock feces (cattle, pigs) and animal source foods (beef, pork, meat patties) in South Korea and 187 pathogenic E. coli strains. The genomic sequences of 160 strains isolated from food and the environment in South Korea and 27 other pathogenic E. coli strains are available at the National Center for Biotechnology Information (NCBI). The genomes analyzed in this study are summarized in Supplementary Table S5. The pangenome was analyzed using the bacterial pan-genome analysis (BPGA) tool (v1.3; default parameters). The USEARCH tool was used for clustering with 95% sequence identity as the cut-off value. The phylogenetic tree was clustered using the neighbor-joining method and visualized using the Interactive Tree of Life (iTOL) v6. The population structure analysis was performed using RhierBAPs [39].

Genome Assemblies of STEC/ETEC Hybrids
Twenty-seven STEC/ETEC hybrid strains isolated from livestock feces and animal source foods in South Korea were sequenced. All strains had one chromosome and one plasmid. The genomic characteristics of the hybrid STEC/ETEC strains are summarized in Table 1. The genome lengths of these isolates ranged from 5,064,469 to 5,865,149 bp, with coverage ranging from 126× to 524×. In addition, the G + C content of the genomes of these strains was between 50.3% and 50.9%, the length of the coding DNA sequences (CDSs) was between 5081 and 6141 bp, and the number of tRNA and rRNA genes was 82-107 and 22, respectively.

In Silico Identification of Virulence and CRISPR-Associated (Cas) Genes
The initial screening of pathogenic E. coli strains was conducted using real-time PCR. The hybrid STEC/ETEC isolates harbored both Shiga toxin 2 (stx2) and heat-stable enterotoxins (est) encoding genes. Subsequently, we performed virulence gene mapping to identify the various virulence factors present in the STEC/ETEC hybrid genomes ( Figure 1). Multiple virulence factors have been implicated in E. coli pathogenesis, including the Shiga toxin and enterotoxins, as well as other factors such as adhesion factors, colonization factors (CFs), non-LEE-encoded TTSS effectors, and secretion systems. Detailed results of virulence gene mapping of these hybrid genomes are shown in Supplementary Table S1. Importantly, we also detected genes encoding LAA, such as hes, iha, lesP, and agn43, which are related to STEC pathogenicity. The most prevalent gene was iha (59.3%), followed by agn43, hes, and lesP that were present in 25.9, 18.5, and 11.1% of the hybrid strains, respectively. In addition, the CRISPRFinder server identified a type I CRISPR/Cas system in all hybrid strains. Additionally, most of the STEC/ETEC hybrid strains (26/27) identified the type I-E system. All hybrid strains harbored the cas3 gene, which is the signature of type I CRISPR/Cas systems, responsible for target DNA cleavage and degradation [40]. Furthermore, most of the STEC/ETEC hybrid strains (26/27) harbored the cas1, cas2, cas5, cas6, and cas7 genes as well as the cas3 gene.
The hybrid STEC/ETEC isolates harbored both Shiga toxin 2 (stx2) and heat-stable enterotoxins (est) encoding genes. Subsequently, we performed virulence gene mapping to identify the various virulence factors present in the STEC/ETEC hybrid genomes ( Figure  1). Multiple virulence factors have been implicated in E. coli pathogenesis, including the Shiga toxin and enterotoxins, as well as other factors such as adhesion factors, colonization factors (CFs), non-LEE-encoded TTSS effectors, and secretion systems. Detailed results of virulence gene mapping of these hybrid genomes are shown in Supplementary  Table S1. Importantly, we also detected genes encoding LAA, such as hes, iha, lesP, and agn43, which are related to STEC pathogenicity. The most prevalent gene was iha (59.3%), followed by agn43, hes, and lesP that were present in 25.9, 18.5, and 11.1% of the hybrid strains, respectively. In addition, the CRISPRFinder server identified a type I CRISPR/Cas system in all hybrid strains. Additionally, most of the STEC/ETEC hybrid strains (26/27) identified the type I-E system. All hybrid strains harbored the cas3 gene, which is the signature of type I CRISPR/Cas systems, responsible for target DNA cleavage and degradation [40]. Furthermore, most of the STEC/ETEC hybrid strains (26/27) harbored the cas1, cas2, cas5, cas6, and cas7 genes as well as the cas3 gene.

In Silico Identification of Antimicrobial Resistance Genes
ResFinder was used to predict antimicrobial resistance genes in the hybrid genomes ( Figure 2). Most of the genomes (16/27) contained at least two antibiotic (ampicillin, piperacillin, streptomycin, and ticarcillin) resistance genes, whereas nine isolates were negative for them. The comprehensive results of the antimicrobial resistance gene mapping of the hybrid E. coli genomes are shown in Supplementary ampicillin (44.4%), and piperacillin (44.4%). Especially, the tetA and tetB genes were found at the highest frequency in hybrid E. coli strains. We additionally performed antimicrobial susceptibility tests to determine the phenotypic profile of antimicrobial resistance in STEC/ETEC hybrid strains. The phenotypic profile of antimicrobial resistance is described in detail in Supplementary Table S3. Comparing the WGS-based AMR genotype to the antimicrobial susceptibility testing-based phenotype for 13 antibiotics (ampicillin, cefepime, cefoxitin, cefotaxime, ceftazidime, chloramphenicol, ciprofloxacin, colistin, nalidixic acid, meropenem, gentamicin, streptomycin, tetracycline) revealed concordant results for 22 of the 27 STEC/ETEC hybrid strains (81.5%). In five STEC/ETEC hybrid strains, only antibiotic resistance genes were identified, but no phenotypes. Although WGS can provide more information about isolates, genomic approaches cannot always predict phenotypes because the level of gene expression and protein production from identified genes may differ between strains. Bacteria have gene-silencing mechanisms, and mutations may generate stop codons in the data [41].

Serotyping and Sequence Types of the Hybrids
The serotype and sequence type results for the 27 hybrid STEC/ETEC isolates are summarized in Table 2

Serotyping and Sequence Types of the Hybrids
The serotype and sequence type results for the 27 hybrid STEC/ETEC isolates are summarized in Table 2

Phage Characterization
To investigate the phage-mediated horizontal gene transfer of stx genes in hybrid STEC/ETEC isolates, we identified the bacteriophage sequences using PHASTER. The results obtained for the 27 hybrid STEC/ETEC strains are summarized in Table 3. The presence of the majority of stx2 gene sequences was confirmed from phage sequence regions corresponding to "intact" (26/27). Additionally, it was confirmed that the sequences corresponding to the stx2 gene in one STEC/ETEC hybrid strain genome (MFDS1012367; score 90) were found in the "questionable" phage region.

Plasmid-Associated Sequence
To investigate the plasmid-mediated horizontal gene transfer of est in the hybrid STEC/ETEC isolates, we analyzed plasmid-associated sequences using PlasmidFinder 2.1. The plasmid replication results for the 27 hybrid STEC/ETEC isolates are summarized in Table 3. PlasmidFinder identified several plasmid replicon sequences of known Inc groups in all the STEC/ETEC genomes. We elucidated that each of the 27 genomes had an IncFIl or IncFIB plasmid origin and in some cases, such as MFDS1016200, both. The genes encoding heat-stable enterotoxin STa (estA) or STb (estB) were placed in the same contig as IncFIl and IncFIB. The mobile genetic elements proximal to estA and estB are shown in Supplementary Table S4.

Phylogenetic Analysis and Population Structure Analysis
The genomes of 187 isolates, comprising 41 STEC, 46 ETEC, 72 EPEC, 18 EAEC, and 10 EIEC strains, were used for the phylogenetic analysis to determine the genomic relationship between the STEC/ETEC hybrids and other pathogenic E. coli isolates. The 187 genome datasets included the sequencing results of 160 pathogenic E. coli as well as 27 hybrid STEC/ETEC genomes, which were deposited in the NCBI database. Phylogenetic tree analysis revealed that these hybrids were closely related to certain ETEC (21 strains, 77.8%) and STEC (six strains, 22.2%) strains, implying the potential acquisition of Stxphages and/or ETEC virulence genes during their emergence ( Figure 3A). In addition, the population structure of the 187 genome datasets was defined using the RhierBAPS, which divided the genome datasets into six primary sequence clusters (Bayesian analysis of population structure [BAPS] hierarchical level 1). These were further subdivided into 28 lineages (BAPS level 2) ( Figure 3B). The results showed that of the total, 21 hybrid strains closely related to ETEC were divided into six groups (three level 1, five level 2), while the remaining six hybrids correlated with STEC were divided into two groups (two level 1, two level 2). Strains are colored based on the pathogenic E. coli groups: blue, STEC strains; pink, ETEC strains; gray, EPEC strains; green, EAEC strain; and brown, EIEC strains. The pink and blue bars represent 21 and 6 STEC/ETEC hybrid strains that are closely related to specific ETEC and STEC strains, respectively. (B) Sequence clusters (1 to 6) are indicated in the outer colored dot, which are further divided into 28 lineages (inner ring). Of the total, 21 hybrid strains closely related to ETEC were divided into six groups (three level 1, five level 2), while the remaining 6 hybrids correlated with STEC were divided into two groups (two level 1, two level 2).

Discussion
STEC/ETEC hybrids have been recovered from various sources, including humans, animals, food, and water, some of which have been associated with diarrheal diseases and HUS in humans. In South Korea, the STEC/ETEC hybrid strain was first isolated from a patient suffering from diarrhea in 2014 [23]. Here, we report 27 STEC/ETEC hybrid strains among 1025 pathogenic E. coli strains identified in South Korea between 2016 and 2020. This study characterized the virulence and antibiotic resistance genes harbored by these hybrid strains, to further determine their phylogeny among other pathogenic E. coli strains. The molecular properties of these strains were investigated using real-time PCR followed by whole-genome sequencing (WGS). Phylogenetic analysis was performed to assess the phylogenetic positions of these hybrids in a diverse collection of pathogenic E. coli representing all the major pathotypes.
For the initial molecular characterization of all pathogenic E. coli strains, real-time PCR and serotyping were employed. Subsequent WGS analysis of these hybrids yielded results that were consistent with serotyping and the presence of virulence factors. The presence of genes encoding Shiga toxin 2 and heat-stable enterotoxin, namely, stx2 and est, respectively, was confirmed in all 27 STEC/ETEC hybrid strains. Most of the STEC/ETEC hybrid strains among human and animal isolates in Finland [25] harbored the stx2 gene without the stx1 gene. In addition, STEC/ETEC hybrid strains from diarrheal patients in South Korea [27] and Sweden [28] harbored the stx2 gene. However, the majority of STEC/ETEC hybrid strains in livestock of Bangladesh [26] carried the stx1 gene. Our results suggest that the STEC/ETEC hybrid strains isolated in South Korea that contained the stx2 gene may be more dangerous to humans. Shiga toxins are major factors contributing to the virulence of STEC; however, adhesion and colonization to the human intestine are required for STEC pathogenesis [42,43,44]. Some STEC strains carry the locus of enterocyte effacement (LEE-positive) [45,46,47], whereas those that do not carry the LEE (LEE-negative) and mainly harbor the locus of adhesion and autoaggregation (LAA) have coli groups: blue, STEC strains; pink, ETEC strains; gray, EPEC strains; green, EAEC strain; and brown, EIEC strains. The pink and blue bars represent 21 and 6 STEC/ETEC hybrid strains that are closely related to specific ETEC and STEC strains, respectively. (B) Sequence clusters (1 to 6) are indicated in the outer colored dot, which are further divided into 28 lineages (inner ring). Of the total, 21 hybrid strains closely related to ETEC were divided into six groups (three level 1, five level 2), while the remaining 6 hybrids correlated with STEC were divided into two groups (two level 1, two level 2).

Discussion
STEC/ETEC hybrids have been recovered from various sources, including humans, animals, food, and water, some of which have been associated with diarrheal diseases and HUS in humans. In South Korea, the STEC/ETEC hybrid strain was first isolated from a patient suffering from diarrhea in 2014 [23]. Here, we report 27 STEC/ETEC hybrid strains among 1025 pathogenic E. coli strains identified in South Korea between 2016 and 2020. This study characterized the virulence and antibiotic resistance genes harbored by these hybrid strains, to further determine their phylogeny among other pathogenic E. coli strains. The molecular properties of these strains were investigated using real-time PCR followed by whole-genome sequencing (WGS). Phylogenetic analysis was performed to assess the phylogenetic positions of these hybrids in a diverse collection of pathogenic E. coli representing all the major pathotypes.
For the initial molecular characterization of all pathogenic E. coli strains, real-time PCR and serotyping were employed. Subsequent WGS analysis of these hybrids yielded results that were consistent with serotyping and the presence of virulence factors. The presence of genes encoding Shiga toxin 2 and heat-stable enterotoxin, namely, stx2 and est, respectively, was confirmed in all 27 STEC/ETEC hybrid strains. Most of the STEC/ETEC hybrid strains among human and animal isolates in Finland [25] harbored the stx2 gene without the stx1 gene. In addition, STEC/ETEC hybrid strains from diarrheal patients in South Korea [27] and Sweden [28] harbored the stx2 gene. However, the majority of STEC/ETEC hybrid strains in livestock of Bangladesh [26] carried the stx1 gene. Our results suggest that the STEC/ETEC hybrid strains isolated in South Korea that contained the stx2 gene may be more dangerous to humans. Shiga toxins are major factors contributing to the virulence of STEC; however, adhesion and colonization to the human intestine are required for STEC pathogenesis [42][43][44]. Some STEC strains carry the locus of enterocyte effacement (LEE-positive) [45][46][47], whereas those that do not carry the LEE (LEE-negative) and mainly harbor the locus of adhesion and autoaggregation (LAA) have also been associated with illness [48][49][50][51]. LAA is found either as a "complete" structure with four modules (module I (hes and other genes), module II (iha, lesP, and others genes), module III (pagC, tpsA, and other genes), and module IV (agn43 and other genes)) or as an "incomplete" structure if one of the modules is missing [48]. In this study, 27 STEC/ETEC hybrid strains carried one copy of the stx2 gene, lacked eae (E. coli attaching and effacing) gene and were LEE-negative STEC strains. Consistent with findings of previous studies, we observed some genes encoding LAA in some of the identified STEC/ETEC hybrid strains, except for seven strains (MFDS  1009773, 1012367, 1014122, 1016183, 1016224, 1016228, and 1016229).
In addition, the colonization of ETEC on the surface of the intestinal epithelium is a critical step in exerting its toxicity [52]. In addition to heat-labile (LT) and/or heat-stable (ST) enterotoxins, colonization factors (CFs) are major virulence factors in ETEC. Once ETEC colonizes the small intestinal epithelia through CFs, effective enterotoxin delivery commences, which is responsible for the secretion of water and electrolytes from the intestinal lumen [53,54]. These factors are referred to as colonization factor antigen I (CFA/I) or coli surface (CS) antigen [55]. The CFA/I of ETEC-related genes, such as cfa A, cfa B, cfa C, cfa D, and cfa E, was detected in 16 STEC/ETEC hybrid strains.
This study describes the virulence gene transfer of STEC/ETEC hybrid strains isolated from livestock feces and animal source foods in South Korea. It emphasizes that WGS is a powerful tool to analyze bacterial genomes for the presence of regions of MGEs, such as phages and plasmids, in them. Furthermore, the genomic information obtained in this study can significantly contribute to a better understanding of the genomic characteristics of hybrid E. coli strains in the future. In further studies, it may be necessary to investigate the genomic and transcriptome characteristics of STEC/ETEC hybrid strains isolated from diverse ecological and geographical sources in Korea.

Conclusions
In conclusion, we are the first to report the virulence and antibiotic resistance profiles of STEC/ETEC hybrid strains isolated from livestock feces (cattle and pigs) and animal source foods (beef, pork, and meat patties) in South Korea. Through genome-based characterization, we confirmed that virulence markers present in STEC/ETEC pathotypes were carried by MGEs, such as phages and plasmids. In addition, we identified adhesion and colonization factors in the human intestine required for STEC pathogenesis. Most DEC is subdivided into several pathotypes based on the presence of specific virulence traits directly related to disease development [3,4]. Importantly, our results emphasize that the hybrid strains of E. coli with STEC and other DEC-associated virulence factors may be more dangerous than STEC alone [14,[20][21][22][23][24][25][26][27][28]. Thus, the emergence of hybrid DEC strains may have severe consequences for public health and should be considered in patient care and epidemiological surveillance.