Characterization of Extended-Spectrum β-Lactamase-Producing and AmpC β-Lactamase-Producing Enterobacterales Isolated from Companion Animals in Korea.

The emergence of extended-spectrum cephalosporin (ESC)-resistant Gram-negative bacteria is of great concern in both human and veterinary medicine. The aim of this study was to investigate ESC-resistant bacterial isolates from companion animals in South Korea between 2017 and 2019. Isolates with ESC resistance genes, which were identified by PCR, were assessed for genetic relatedness by multi-locus sequence typing (MLST) and pulsed-field gel electrophoresis (PFGE). In total, 91 ESC-resistant Escherichia coli, Klebsiella spp., Serratia spp., and Enterobacter cloacae isolates harbored the blaTEM gene. Among other ESC resistance genes, blaCTX-M-15, blaCIT, and blaCTX-M-55 were predominantly detected in E. coli isolates, whereas blaSHV and blaDHA were more frequently detected in Klebsiella pneumoniae isolates. In addition, all blaEBC-positive isolates were classified as E. cloacae. From the MLST results, blaCTX-M-9-carrying ST131, blaCIT-carrying ST405, and blaCTX-M-1-carrying ST3285 strains were dominant among E. coli isolates. ST273 and ST275 strains harboring blaSHV were frequently detected in K. pneumoniae isolates. Various sequence types were obtained in E. cloacae and Klebsiella oxytoca isolates. All isolates demonstrated unique PFGE profiles (<57-98% similarity) and were unlikely to be derived from a single clone. The present study reveals the presence and wide genetic distribution of ESC-resistant bacterial species in South Korean companion animals.

1. Introduction

The emergence and prevalence of β-lactam resistance in Gram-negative bacteria has increased consistently over the past few decades [1,2]. Resistance to β-lactams is mostly caused by bacterially produced β-lactamases that hydrolyze and inactivate extended-spectrum cephalosporins (ESCs), such as third and fourth generation cephalosporins [1]. ESC resistance is mainly caused by the expression of extended-spectrum β-lactamase (ESBL) and AmpC β-lactamase (AmpC) genes that are normally encoded on mobile genetic elements, mostly plasmids [1].

The first ESBLs have evolved from the native β-lactamases TEM and SHV via genetic mutations [3]. CTX-M β-lactamases, a new group of plasmid-mediated ESBLs, were first reported in Japan in 1986 [4]. However, since 2000, CTX-M β-lactamases have increasingly been reported in both human and animal populations and are now the dominant type of ESBL, replacing classical TEM- and SHV-type ESBLs in most areas of the world [5]. Currently, there are >120 different CTX-M β-lactamases that are clustered into five groups (CTX-M-1, 2, 8, 9, and 25) [6]. Among CTX-M-type enzymes, the presence of CTX-M-15 and CTX-M-14 has increasingly been reported in most areas of the world, including South Korea [5,7].

Since the detection of CMY-1, the first reported AmpC type β-lactamases, in 1989 [6], various types of AmpCs have been identified in clinical isolates of Enterobacterales around the world [8]. Among AmpCs, CIT- and DHA-type enzymes are the most prevalent [2]. Especially, DHA-producing Klebsiella spp. isolates and CIT-producing Escherichia coli isolates have been repeatedly reported for Enterobacterales in South Korea [2,9]. Despite many studies on ESBL- and AmpC-producing bacteria from human or livestock isolates, studies concerning antimicrobial resistance (AMR) bacteria associated with companion animals are lacking, especially with respect to Serratia spp. and Enterobacter spp. [10,11,12].

The popularity of companion animals in South Korea has been growing, which provides a potential reservoir of AMR bacteria, as pets are closely associated with humans, living in their homes and near their food [9]. Thus, the importance of profiling AMR bacteria was emphasized in the “One Health” initiative, which integrates veterinary medicine, human health, animal-production systems, and the environment [13]. Systematic control and prevention, through implementation of a national AMR surveillance program, is greatly needed and should be applied in both human and veterinary clinical medicine. The goal of the current study was to investigate AMR among bacterial isolates belonging to Enterobacterales in companion animals within the province of South Korea, with an emphasis on ESC resistance genes in E. coli, Klebsiella spp., Serratia spp., and Enterobacter cloacae.

2. Results

2.1. ESC Resistance Gene Detection

Among the 91 ESC-resistant Enterobacterales isolates analyzed in this study, all of the isolates harbored the blaTEM gene. blaCTX-M (n = 42, 82.4%) was abundantly detected in E. coli isolates, whereas blaSHV (n = 16, 94.1%) was mainly detected in Klebsiella pneumoniae isolates (Table 1). None of the Serratia spp. and E. cloacae isolates, except one S. marcescens isolate from Serratia spp. that harbored blaSHV, were positive for blaSHV or blaCTX-M (Table 1). Among 55 blaCTX-M-positive Enterobacterales isolates, blaCTX-M-15 (n = 23, 41.8%) and blaCTX-M-55 (n = 12, 21.8%) were the most commonly detected, followed by blaCTX-M-14 (n = 7, 12.7%) (Table 2). Enterobacterales isolates carrying genes of blaCTX-M-3, blaCTX-M-61, blaCTX-M-27, and blaCTX-M-65 were also identified (Table 2). blaCTX-M-2, blaCTX-M-8, and blaCTX-M-25 gene clusters were not detected in any isolates. Among blaAmpC genes, blaCIT was prevalent in E. coli isolates, blaDHA was predominantly detected in Klebsiella spp. isolates, and blaEBC was common in E. cloacae isolates (Table 3). No isolates carrying genes for blaMOX or blaACC were detected.

Table 1

Distribution of extended-spectrum β-lactamase (ESBL) genes among 91 Enterobacterales isolates from companion animals.

Organism	Hospital-Admitted Dogs (n = 56)			Stray Dogs (n = 23)			Hospital-Admitted Cats (n = 11)			Stray Cats (n = 1)			Total
	bla CTX-M	bla SHV	bla TEM	bla CTX-M	bla SHV	bla TEM	bla CTX-M	bla SHV	bla TEM	bla CTX-M	bla SHV	bla TEM	bla CTX-M	bla SHV	bla TEM
E. coli (n = 51)	21	−	22	15	−	23	6	−	6	−	−	−	42 (82.4%)	−	51 (100%)
K. pneumoniae (n = 17)	7	15	15	−	−	−	1	1	1	1	−	1	9 (17.6%)	16 (94.1%)	17 (100%)
K. oxytoca (n = 5)	4	1	4	−	−	−	−	1	1	−	−	−	4 (80.0%)	2 (40.0%)	5 (100%)
S. marcescens (n = 4)	−	1	3	−	−	−	−	−	1	−	−	−	−	1 (45.0%)	4 (100%)
S. liquefaciens (n = 7)	−	−	7	−	−	−	−	−	−	−	−	−	−	−	7 (100%)
E. cloacae (n = 7)	−	−	5	−	−	−	−	−	2	−	−	−	−	−	7 (100%)

Table 2

CTX-M subtype detection of 55 blaCTX-M-positive Enterobacterales isolates.

Organism	Sample Origin	blaCTX-M Subtype Group									Unidentified
		CTX-M-1					CTX-M-9
		CTX-M-15	CTX-M-55	CTX-M-3	CTX-M-61	Total	CTX-M-14	CTX-M-27	CTX-M-65	Total
E. coli (n = 42)	Hospital-admitted dogs (n = 21)	7	6	1	1	15	3	1	−	4	2
	Stray dogs (n = 15)	8	3	−	−	11	3	1	−	4	0
	Hospital-admitted cats (n = 6)	2	2	−	−	4	−	−	−	0	2
K. pneumoniae (n = 9)	Hospital-admitted dogs (n = 7)	4	1	−	−	5	−	1	1	2	0
	Hospital-admitted cats (n = 1)	−	−	1	−	1	−	−	−	0	0
	Stray cats (n = 1)	−	−	−	−	0	−	−	−	0	1
K. oxytoca (n = 4)	Hospital-admitted dogs (n = 4)	2	−	−	−	2	1	−	−	1	1
Total		23 (41.8%)	12 (21.8%)	2 (3.6%)	1 (1.8%)	38 (69.1%)	7 (12.7%)	3 (5.5%)	1 (1.8%)	11 (20.0%)	6 (10.9%)

Table 3

Distribution of AmpC β-lactamase genes among ESBL-producing Enterobacterales isolates.

Organism	Sample Origin	AmpC β-Lactamases Gene
		MOX	CIT	DHA	ACC	EBC	FOX	EBC+CIT	EBC+DHA	Total
E. coli	Hospital-admitted dogs (n = 22)	−	5	−	−	−	−	−	−	5
	Stray dogs (n = 23)	−	13	−	−	−	1	−	−	14
K. pneumoniae	Hospital-admitted dogs (n = 15)	−	1	4	−	1	−	−	−	6
	Hospital-admitted cats (n = 1)	−	−	1	−	−	−	−	−	1
K. oxytoca	Hospital-admitted dogs (n = 4)	−	−	2	−	−	−	−	−	2
E. cloacae	Hospital-admitted dogs (n = 5)	−	−	1	−	2	−	−	−	3
	Hospital-admitted cats (n = 2)	−	−	−	−	−	−	1	1	2
Total		−	19	8	−	3	1	1	1	33

2.2. Multi-Locus Sequence Typing (MLST)

Various sequence types (STs) were revealed among E. coli, K. pneumoniae, K. oxytoca, and E. cloacae isolates. STs of Serratia spp. isolates were not defined due to the lack of an MLST scheme for this genus. Among E. coli isolates, five ST131 and two ST3285 strains were detected from hospital-admitted dogs, with one of each of the following STs: ST372, ST457, ST648, ST1981, ST2179, ST2505, ST4616, ST5150, ST5667, ST8451, ST8885, ST8908, ST10207, ST10220, and ST11000. Accordingly, five ST405, three ST354, two each of ST3285, ST410, ST448, and ST457, and one each of ST68, ST38, ST648, ST1193, ST2541, ST7644, and ST10459 strains were detected in stray dogs. Among six E. coli isolates from hospital-admitted cats, two ST131, two ST156, and one each of ST1262 and ST6105 strains were obtained. In a comprehensive analysis of 19 Klebsiella spp. isolates from hospital-admitted dogs, one ST285, eight ST275, and six ST273 stains were identified among K. pneumoniae isolates, whereas one each of ST34, ST145, ST273, and ST293 strains were identified among K. oxytoca isolates. All Klebsiella spp. isolates from hospital-admitted and stray cats were identified as ST273 strains. Among seven E. cloacae isolates, two ST114 and one each of ST110, ST171, ST198, ST1252, and ST1303 strains were identified (Table 4).

Table 4

Multi-locus sequence typing (MLST) results based on the spread of β-lactamase resistance genes.

Organism	blaCTX-M Cluster	ESC Resistance Gene				No. Isolation
		ESBLs			AmpCs	ST Type
		bla CTX-M	bla SHV	bla TEM		Hospital-Admitted Dogs	Stray Dogs	Hospital-Admitted Cats	Stray Cats
E. coli	CTX-M-1	CTX-M-15	−	+	−	648 (n = 1)	354 (n = 1)	131 (n = 1)	−
						1981 (n = 1)	448 (n = 2)	1262 (n = 1)	−
						2179 (n = 1)	7644 (n = 1)	−	−
						10,207 (n = 1)	10,459 (n = 1)	−	−
			−	+	CIT	457 (n = 1)	354 (n = 2)	−	−
						5667 (n = 1)	410 (n = 1)	−	−
						11,000 (n = 1)	−	−	−
		CTX-M-55	−	+	−	4616 (n = 1)	−	131 (n = 1)	−
						5150 (n = 1)	−	156 (n = 1)	−
						8451 (n = 1)	−	−	−
						8908 (n = 1)	−	−	−
			−	+	CIT	2505 (n = 1)	410 (n = 1)	−	−
						3285 (n = 1)	3285 (n = 2)	−	−
	CTX-M-9	CTX-M-61	−	+	−	3285 (n = 1)	−	−	−
		CTX-M-3	−	+	−	8885 (n = 1)	−	−	−
		CTX-M-14	−	+	−	10,220 (n = 1)	38 (n = 1)	−	−
						131 (n = 2)	68 (n = 1)	−	−
						−	648 (n = 1)	−	−
		CTX-M-27	−	+	−	131 (n = 1)	−	−	−
		CTX-M-27	−	+	FOX	−	1193 (n = 1)	−	−
						−	−	−	−
	Unidentified	−	−	+	−	131 (n = 2)	−	156 (n = 1)	−
						−	−	6105 (n = 1)	−
	Negative	−	−	+	−	372 (n = 1)	2541 (n = 1)	−	−
		−	−	+	CIT	−	405 (n = 5)	−	−
						−	457 (n = 2)	−	−
Total						22	23	6	0
K. pneumoniae	CTX-M-1	CTX-M-15	+	+	−	273 (n = 2)	−	−	−
						275 (n = 1)	−	−	−
			+	+	EBC	285 (n = 1)	−	−	−
		CTX-M-55	+	+	DHA	275 (n = 1)	−	−	−
		CTX-M-3	+	+	DHA	−	−	273 (n = 1)	−
	CTX-M-9	CTX-M-27	+	+	DHA	273 (n = 1)	−	−	−
		CTX-M-65	+	+	DHA	275 (n = 1)	−	−	−
	Unidentified	−	-	+	−	−	−		273 (n = 1)
	Negative	−	+	+	−	273 (n = 2)	−	−	−
						275 (n = 4)	−	−	−
		−	+	+	CIT	273 (n = 1)	−	−	−
		−	+	+	DHA	275 (n = 1)	−	−	−
Total						15	0	1	1
K. oxytoca	CTX-M-1	CTX-M-15	−	+	−	293 (n = 1)	−	−	−
			+	+	DHA	273 (n = 1)	−	−	−
	CTX-M-9	CTX-M-14	−	+	−	145 (n = 1)	−	−	−
	Unidentified	−	−	+	DHA	34 (n = 1)	−	−	−
	Negative	−	+	+	−	−	−	273 (n = 1)	−
Total						4	0	1	0
S. liquefaciens	Negative	−	−	+	−	a ND (n = 7)	−	−	−
S. marcescens	Negative	−	−	+	−	ND (n = 2)	−	ND (n = 1)	−
			+	+	−	ND (n = 1)	−	−	−
Total						10	0	1	0
E. cloacae	Negative	−	−	+	DHA	198 (n = 1)	−	−	−
		−	−	+	EBC	114 (n = 1)	−	−	−
						110 (n = 1)	−	−	−
		−	−	+	CIT+EBC	−	−	171 (n = 1)	−
		−	−	+	DHA+EBC	−	−	1303 (n = 1)	−
		−	−	+	−	1252 (n = 1)	−	−	−
						114 (n = 1)	−	−	−
Total						5	0	2	0

a ND, non-defined.

2.3. Pulsed-Field Gel Electrophoresis (PFGE)

PFGE analysis was only conducted for 46 E. coli, 13 K. pneumoniae, five K. oxytoca, and six E. cloacae isolates; no or few banding patterns were obtained for five E. coli, four K. pneumoniae, and one E. cloacae isolates. Using a >85% similarity cut-off, 31 pulsotypes in E. coli (e1 to e31), nine pulsotypes in K. pneumoniae (kp1 to kp9), four pulsotypes in K. oxytoca (ko1 and ko4), and four pulsotypes in E. cloacae (ec1 to ec4) were identified. Generally, all isolates demonstrated unique PFGE profiles (57–95% similarity), indicating genetic heterogeneity in ESBL- or AmpC-producing strains (Figure 1).

Figure 1

Dendrogram of pulsed-field gel electrophoresis (PFGE) patterns showing the genetic relatedness of ESBL-producing Enterobacterales isolates. (a) Escherichia coli, (b) Klebsiella pneumoniae, (c) K. oxytoca, and (d) Enterobacter cloacae.

2.4. Genetic Relatedness

In E. coli PFGE analysis, the e1 group consisted of three ST3285 strains containing blaCTX-M-55, blaTEM, and blaCIT from hospital-admitted and stray dogs and showed high similarity (>90%) (Figure 1a). Seven ST131 strains of E. coli isolates from five hospital-admitted dogs and two hospital-admitted cats were identified. PFGE results involved only five hospital-admitted dogs and one hospital-admitted cat, because the banding pattern for one hospital-admitted cat isolate was not defined. Two of them, which belonged to the e18 group, showed more than 93% similarity, compared with the remaining isolates in which low similarity was observed (<85%) (Figure 1a). e3 and e25 groups contained two ST448 strains with blaCTX-M-55 and blaTEM and two ST457 strains with blaTEM and blaCIT from the same shelter and showed high similarity (>87%) (Figure 1a).

All blaTEM-positive K. pneumoniae isolates from hospital-admitted dogs harbored the blaSHV gene and exhibited 60–95% similarity (Figure 1b). One K. pneumoniae ST273 strain from stray cats belonged to the kp8 group and showed 87% similarity with hospital-admitted dog isolates (Figure 1b). Two ST275 strains co-carrying blaSHV and blaTEM, which belonged to the kp5 group, showed high similarity (>90%) (Figure 1b). The same pulsotype belonged to both ST273 and ST275, which were single-locus variants at the tonB allele; for example, kp1 and kp8. The remaining strains were independent of the groups obtained. Various STs and PFGE profiles were obtained for K. oxytoca and E. cloacae isolates, and genetic relatedness was not revealed for those strains (Figure 1c,d).

3. Discussion

This study presents the characteristics of 91 Enterobacterales isolates harboring ESC resistance genes, including E. coli, Klebsiella spp., Serratia spp., and E. cloacae, collected from South Korean companion animals between 2017 and 2019. All isolates harbored the blaTEM gene and demonstrated unique PFGE profiles. Similarly, a study by Shin et al. revealed that all E. coli isolates from beef cattle harbored the blaTEM gene [12]. Recent reports have identified CTX-M-type β-lactamases as the most widespread ESBL type, replacing classical TEM and SHV-type ESBLs [14]; however, TEM-type β-lactamases remained the most prevalent ESBL type identified in the current study. We observed varying predominant β-lactamase gene types in different Enterobacterales species, summarized as follows: blaCTX-M and blaCIT in E. coli isolates, blaSHV and blaDHA in K. pneumoniae isolates, and blaEBC in E. cloacae. β-Lactamase gene distribution for each Enterobacterales species was similar to that described in a previous study with human samples [2]. These findings reveal that ESC resistance gene variants are not limited to certain hosts, emphasizing the need for coordinated control in both humans and animals.

In E. coli isolates in this study, blaTEM and blaCTX-M were most frequently detected, followed by blaCIT. Among blaCTX-M positive isolates, blaCTX-M-15 was the most commonly detected gene followed by blaCTX-M-55 in both dogs and cats. A previous study investigating E. coli isolates from dogs reported that blaCTX-M-15, blaCTX-M-14, and blaCIT were the most prevalent β-lactamase genes, whereas blaCTX-M-55 was rarely detected in South Korea [15]. However, blaCTX-M-55-carrying E. coli has become increasingly prevalent in dogs in South Korea [9]. The present study revealed that blaCTX-M-55 was predominantly detected rather than blaCTX-M-14 in E. coli from companion animals, which concurred with the results of the study by Hong et al. [9]. All five E. coli ST405 strains investigated in the current study harbored both blaTEM and blaCIT and were collected from stray dogs in the same shelter. The spread of blaCIT-carrying E. coli ST405 was described in a previous study, which suggested the possibility of direct transmission between humans and companion animals [9]. The spread of E. coli ST405 is usually described in humans harboring blaCTX-M-15 [16]. However, E. coli ST405 did not harbor blaCTX-M-15 in the current study. The increasing prevalence of E. coli ST131 carrying blaCTX-M-15 has been described in humans and animals [15,17]. Unexpectedly, only one blaCTX-M-15-carrying E. coli ST131 strain was detected from hospital-admitted cats in this study. From E. coli PFGE results, the two ST3285 strains showing the same PFGE pattern were both from the same shelter and isolated on the same date (Figure 1a). In this case, it could be the result of contaminated samples during sampling or transmission of a same clone between the two stray dogs in a shared place.

Among the seven K. pneumoniae and four K. oxytoca isolates harboring blaCTX-M from hospital-admitted dogs in this study, the CTX-M-15 genotype accounted for a large proportion. A recent study also reported that K. pneumoniae isolates from companion animals producing CTX-M-15 either alone or in combination with DHA were frequently detected in South Korea [9]. The present MLST results revealed that ST275 and ST273 strains carrying both blaSHV and blaTEM were most commonly identified among K. pneumoniae isolates from hospital-admitted dogs. ST275 and ST273 are differentiated by one allele of the seven house-keeping genes, indicating that they are genetically close sequence types. Recently, the blaSHV- and blaTEM-co-carrying K. pneumoniae ST273 strain has emerged in human patients in Italy and is being disseminated, whereas ST273 and ST275 Klebsiella spp. isolates carrying both blaSHV and blaTEM have not yet been reported in South Korea [18,19]. Moreover, ST11, ST15, ST307, and ST392 strains have been globally identified as β-lactamase-producing Klebsiella spp. [9,20,21]. β-Lactamase-producing ST275 or ST273 strains among

Klebsiella spp. isolates were newly discovered in South Korean companion animals in the current study.

All E. cloacae isolates in this study harbored blaTEM and some carried blaEBC, whereas none carried blaCTX-M and blaSHV. E. cloacae isolates from companion animals combined with more than two ESBL-type genes were described in Germany [22]. Meanwhile, E. cloacae isolates investigated in the present study harbored only one type of ESBL gene. In Serratia spp. isolates, only blaTEM was detected, except one S. marcescens isolate that harbored both blaTEM and blaSHV. An S. marcescens isolate carrying blaTEM was previously identified in South Korea that caused urinary infections in humans [23]. However, in many countries including South Korea, the status of emerging AMR among Serratia spp. and E. cloacae in companion animals remains unknown. To our knowledge, this is the first report of ESC-resistant Serratia spp. and E. cloacae isolates from companion animals in South Korea.

In conclusion, we illustrated the presence and genetic heterogeneity of ESC-resistant Gram-negative bacteria in companion animals in South Korea, providing a potential reservoir of ESC-resistant bacteria and a transmission pathway. More organized surveillance is required to prevent and control the spread of ESC-resistant bacteria between companion animals and humans, in accordance with the “One Health” initiative.

4. Materials and Methods

4.1. Bacterial Characterization

Sampling, isolation, identification, antimicrobial susceptibility tests, and phenotypic characterization of Enterobacterales were previously studied [24,25]. In total, 91 Enterobacterales isolates carrying ESC resistance genes (51 E. coli, 17 K. pneumoniae, five K. oxytoca, four S. marcescens, seven S. liquefaciens, and seven E. cloacae) were collected from companion animals (56 hospital-admitted dogs, 23 stray dogs, 11 hospital-admitted cats, and one stray cat, Table 5). The antimicrobial resistance profile of the 91 bacterial isolates used in this study were summarized against cephalosporins (Table 5).

Table 5

The source and antimicrobial resistance characteristics of 91 bacterial isolates used in this study.

Organism	Origin	Antimicrobial Resistant Rate against Cephalosporins (%)					Reference
		Cephalexin	Cefoxitin	Ceftiofur	Ceftriaxone	Cephalothin
E. coli (n = 51)	Hospital-admitted dogs (n = 22)	95.5	27.3	95.5	95.5	100	[25]
	Stray dogs (n = 23)	100	47.8	100	100	100
	Hospital-admitted cats (n = 6)	100	16.7	100	100	100	[24]
K. pneumonia (n = 17)	Hospital-admitted dogs (n = 15)	53.3	60.0	33.3	40	53.3	[25]
	Hospital-admitted cats (n = 1)	100	100	100	100	100	[24]
	Stray cat (n = 1)	100	100	100	100	100
K. oxytoca (n = 5)	Hospital-admitted dogs (n = 4)	75.0	50.0	100	100	100	[25]
	Hospital-admitted cat (n = 1)	100	100	0	0	100	[24]
S. marcescens (n = 4)	Hospital-admitted dogs (n = 3)	100	100	0	0	100	[25]
	Hospital-admitted cat (n = 1)	100	100	0	0	100	[24]
S. liquefaciens (n = 7)	Hospital-admitted dogs (n = 7)	0	14.3	28.6	0	100	[25]
E. cloacae (n = 7)	Hospital-admitted dogs (n = 5)	100	100	60.0	80.0	100	[25]
	Hospital-admitted cats (n = 2)	100	100	100	100	100	[24]

4.2. Characterization of β-Lactamase Genes

PCR amplification of entire blaCTX-M, blaTEM, and blaSHV genes was performed as previously described [24,25]. For blaCTX-M-positive isolates, PCR and DNA sequencing were carried out for CTX-M-subtype detection. blaCTX-M group-specific primers for five clusters (CTX-M-1, 2, 8, 9, and 25) were used following Kor-GLASS (Korea Global Antimicrobial Resistance Surveillance System) guidelines and previously published protocols [26,27]. DNA sequencing was performed by Intron Biotechnology (Seongnam, South Korea) and homologous sequences were searched against the GenBank database using the BLAST tool of the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov/BLAST (accessed on 15 December 2020)). In ESBL-positive strains, six groups of AmpC β-lactamases (MOX, CIT, DHA, ACC, EBC, and FOX) were screened by PCR amplification [28].

4.3. Multi-Locus Sequence Typing

Multi-locus sequence types (STs) were based on the allelic profile of seven housekeeping genes. For E. coli, K. pneumoniae, K. oxytoca, and E. cloacae isolates, MLST for each bacterial strain was carried out in reference to previous studies [29,30,31]. PCR was performed using primers for the following: adk, fumC, gyrB, icd, purA, mdh, and recA (E. coli); gapA, infB, mdh, pgi, phoE, rpoB, and tonB (K. pneumoniae and K. oxytoca); danA, fusA, gyrB, leuS, pyrG, rplB, and rpoB (E. cloacae). Allelic profile and ST determinations were performed according to web-based MLST databases (https://pubmlst.org/databases/ (accessed on 15 December 2020) and https://bigsdb.pasteur.fr/klebsiella/klebsiella.html (accessed on 15 December 2020)).

4.4. Pulsed-Field Gel Electrophoresis

PFGE of XbaI (Takara Bio Inc., Shiga, Japan)-digested genomic DNA was carried out for E. coli, K. pneumoniae, K. oxytoca, and E. cloacae isolates according to the CDC PulseNet standardized procedure using the Chef Mapper system (Bio-Rad Laboratories, Hercules, CA, USA) [32]. PFGE analysis for Serratia spp. was ignored because different digested genomic DNA samples were employed for Serratia spp. Similarities between restriction fragment length polymorphisms were analyzed using GelCompar II software v. 6.5 (Applied Maths NV, St-Martens-Latem, Belgium) to produce a dendrogram. The unweighted-pair group method using average linkages (UPGMA) cluster analysis was conducted based on an 85% similarity cut-off with 0.5% optimization and 2.0% band tolerance.