Prevalence of Plasmid-Mediated Quinolone Resistance Determinants Among Escherichia coli Isolated from Food Animals in Korea

Abstract

The aim of this study was to investigate the prevalence of plasmid-mediated quinolone resistance (PMQR) determinants in Escherichia coli isolated from food-producing animals and to characterize the PMQR-positive isolates. A total of 365 E. coli isolates which were either nalidixic acid resistant and ciprofloxacin susceptible (NAL^R-CIP^S; n=185), or nalidixic acid and ciprofloxacin resistant (NAL^R-CIP^R; n=180) were assessed for the presence of PMQR determinants by polymerase chain reaction. PMQR-positive isolates were further characterized by mutation analysis within the quinolone resistance-determining region (QRDR) of gyrA, gyrB, parC, and parE, phylogenetic group analysis, and pulsed-field gel electrophoresis (PFGE). Fourteen NAL^R-CIP^S (n=8) and NAL^R-CIP^R (n=6) E. coli isolates were positive for PMQR genes. Among them, qnrB4, qnrS1, and aac(6’)-Ib-cr genes were detected in two (0.5%), eight (2.2%), and four (1.1%) isolates, respectively. None of the isolates harbored qnrA, qnrC, qnrD, and qepA genes. All but one PMQR-positive isolates harbored one or more point mutations in the QRDR of gyrA, and five of these isolates had additional mutations in the parC gene. Furthermore, one isolate each had additional substitutions in gyrB and parE genes, respectively. The most prevalent mutation was Ser83-Leu within the QRDR of gyrA. Phylogenetic analysis identified three major phylogenetic lineages, with phylogroups A (n=7) and D (n=4) being the most common phylogroups. None of the isolates belonged to virulent phylogroup B2. PFGE demonstrated that a combination of clonal and horizontal gene transmission is disseminating PMQR genes among the veterinary E. coli isolates in Korea. To our knowledge, this is the first report of occurrence of qnrB, qnrS, and aac(6’)-Ib-cr genes in E. coli isolated from food-producing animals in Korea. Isolation of PMQR genes from food animals is a matter of concern since they could be transmitted to humans via food animals.

Introduction

Q uinolone resistance in Enterobacteriaceae results mostly from mutations in chromosomal genes encoding DNA gyrase and/or topoisomerase IV or by mutations of genes that encode efflux pumps or outer membrane proteins or their regulatory mechanisms (Ruiz, 2003). However, plasmid-mediated quinolone resistance (PMQR) has emerged in Enterobacteriaceae, and the prevalence of PMQR genes is increasing worldwide including in Korea (Strahilevitz et al., 2009). Since the first PMQR determinant, qnrA1, was reported from the United States in 1988 (Martínez-Martínez, 1998), five major families of qnr determinants (qnrA, qnrB, qnrC, qnrD, and qnrS) and their several allelic variants have been identified in various enterobacterial isolates from humans (Strahilevitz et al., 2009). Moreover, two additional PMQR determinants, aac(6’)-Ib-cr (Robicsek et al., 2006) and qepA (Yamane et al., 2007) have been identified. However, despite several reports (Strahilevitz et al., 2009; Poirel et al., 2012) of various PMQR determinants among clinical isolates worldwide, only limited data on the prevalence of these determinants in isolates from animals are available (Cerquetti et al., 2009; Ma et al., 2009; Poirel et al., 2012; Szmolka et al., 2011).

Besides conferring a low-level resistance to (fluoro)quinolones, PMQR genes may also facilitate selection of additional chromosomal resistance mechanisms, leading to the emergence of bacterial strains with a high level of (fluoro)quinolone resistance (Strahilevitz et al., 2009; Poirel et al., 2008). Moreover, reduced susceptibility to fluoroquinolones in Enterobacteriaceae is an important cause of clinical treatment failures in humans which cause significant therapeutic problems worldwide (Aarestrup et al., 2003; Molbak et al., 1999; Strahilevitz et al., 2009). Furthermore, PMQR genes are usually carried on mobile genetic elements and the transferability of resistant bacteria or mobile resistance genes between animals and humans via the food chain or direct contact has been well documented (Guardabassi et al., 2004; Simjee et al., 2002; Molbak et al., 1999). Thus, resistance to (fluoro)quinolones due to PMQR determinants is a matter of concern both in human and veterinary medicine.

During the past decade, the PMQR gene alone or in association with plasmid-mediated extended-spectrum β-lactamase (ESBL) or AmpC β-lactamase have been increasingly reported among the clinical isolates of Enterobacteriaceae from humans in Korea (Tamang et al., 2008; Kang et al., 2009; Kim et al., 2009). However, data on the occurrence of PMQR determinants among bacteria from animals in Korea is very rare (Tamang et al., 2011). Therefore, the objective of the present study was to determine the prevalence of PMQR genes in E. coli strains isolated from food-producing animals during 2003–2011 and to characterize the PMQR determinant-positive E. coli isolates.

Materials and methods

Bacterial strains

A total of 365 nalidixic acid resistant and ciprofloxacin susceptible (NAL^R-CIP^S; n=185) and nalidixic acid and ciprofloxacin resistant (NAL^R-CIP^R; n=180) Escherichia coli isolated from food-producing animals in Korea were investigated. Among them, 187 isolates were obtained from diseased cattle (n=67) and diseased pigs (n=120). These E. coli from sick animals were isolated from clinical samples at Diagnostic Laboratory of Animal, Plant, and Fisheries Quarantine and Inspection Agency (QIA), Anyang, Korea during 2003–2009 or received from 16 local veterinary service centers during 2010–2011. Remaining 178 isolates examined were obtained from healthy cattle (n=42) and healthy pigs (n=136). These commensal E. coli were isolated from fecal samples collected from various farms located throughout Korea. The collection of samples, culture, and isolation of E. coli isolates were done as described previously (Lim et al., 2007). During strain selection, 10 E. coli isolates (one E. coli per fecal sample) per farm were randomly selected for initial screening of quinolone resistance and one E. coli isolate per farm was randomly selected for further analysis. Identification of E. coli isolates was done by standard biochemical tests and confirmed by the Vitek system (BioMerieux, Hazelwood, MO).

Detection of PMQR genes

The qnrA, qnrB, qnrC, qnrD, qnrS, aac(6’)-Ib, and qepA genes were detected by two sets of multiplex polymerase chain reactions (PCRs) using primers (Table 1) and conditions as described previously (Jacoby et al., 2009; Kim et al., 2009; Park et al., 2006; Tamang et al., 2011). All aac(6’)-Ib positive PCR amplicons were further subjected to digestion with BtsCI (New England Biolabs, Ipswich, MA) restriction enzyme to identify aac(6’)-Ib-cr, which does not have the BtsCI restriction site present in the wild-type gene as described previously (Park et al., 2006). For amplification of the entire qnr genes, qnrB- or qnrS-positive isolates were reamplified and sequenced using qnrB and qnrS full-length primers as described previously (Tamang et al., 2008; Hopkins et al., 2007).

Table 1.

Sequence of Oligonucleotide Primers Used in This Study

Primers	Primer sequences (5′→3′)	Target gene	Amplicon size (bp)	Reference
QnrA-F	ATTTCTCACGCCAGGATTTG	qnrA	574	Jacoby et al., 2009
QnrA-R	TGCCAGGCACAGATCTTGAC
QnrB-F	CGACCTKAGCGGCACTGAAT	qnrB	513	Jacoby et al., 2009
QnrB-R	GAGCAACGAYGCCTGGTAGYTG
QnrC-F	GGGTTGTACATTTATTGAATCG	qnrC	307	Kim et al., 2009
QnrC-R	CACCTACCCATTTATTTTCA
QnrD-F	TGTGATTTTTCAGGGGTTGA	qnrD	520	Tamang et al., 2011
QnrD-R	CCTGCTCTCCATCCAACTTC
QnrS-F	ACTGCAAGTTCATTGAACAG	qnrS	431	Jacoby et al., 2009
QnrS-R	GATCTAAACCGTCGAGTTCG
AAC(6’)-Ib-F	TTGCGATGCTCTATGAGTGGCTA	aac(6’)-Ib	482	Park et al., 2006
AAC(6’)-Ib-R	CTCGAATGCCTGGCGTGTTT
QnrS FL-F	TGGAAACCTACAATCATACATATCG	qnrS	680	Hopkins et al., 2007
QnrS FL-R	TTAGTCAGGATAAACAACAATACCC
QnrB-1-10-F	ATGACGCCATTACTGTATAAAAAA	qnrB	681	Tamang et al., 2008
QnrB-1-10-R	CTAGCCAATAATCGCGATGCCA
QepA-F	AACTGCTTGAGCCCGTAGAT	qepA	596	Kim et al., 2009
QepA-R	GTCTACGCCATGGACCTCAC
GyrA-F	TGTCCGAGATGGCCTGAAGC	gyrA	470	Giraud et al., 1999
GyrA-R	CGTTGATGACTTCCGTCAG
GyrB-F	AACGGTCTGCTCATCAGAAAGG	gyrB	711	Giraud et al., 1999
GyrB-R	GAAATGACCCGCCGTAAAGG
ParC-F	TGTATGCGATGTCTGAACTG	parC	265	Everett et al., 1996
ParC-R	CTCAATAGCAGCTCGGAATA
ParE-F	CCGTATGCGTGCGGCCAAA	parE	641	This study
ParE-R	CGATAGTCAACTGCACCAGAC

Antimicrobial susceptibility testing

Antimicrobial susceptibility test was performed by standard disc diffusion method according to the guidelines of Clinical Laboratory Standards Institute (CLSI, 2009) using Mueller Hinton agar (Becton-Dickinson, Sparks, MD). The antimicrobial drugs tested were ampicillin, amoxicillin-clavulanic acid, apramicin, cephalothin, ceftiofur, cefoxitin, colistin, chloramphenicol, florfenicol, gentamicin, neomycin, streptomycin, tetracycline, and trimethoprim-sulfamethoxazole (Becton-Dickinson). Minimum inhibitory concentrations (MICs) of nalidixic acid, oxolinic acid, ciprofloxacin, enrofloxacin, norfloxacin, ofloxacin, and danofloxacin (Sigma Chemical Co., St. Louis, MO) were determined for all the PMQR-positive isolates following the CLSI guidelines (CLSI, 2009). Escherichia coli ATCC 25922 was used as the quality control strain.

Sequencing of gyrA, gyrB, parC, and parE

Quinolone resistance-determining regions (QRDRs) of gyrA, gyrB, and parC genes in PMQR gene-positive isolates were amplified using previously described primers and protocols (Giraud et al., 1999; Everett et al., 1996). The primer set for amplification of QRDR of parE was designed using the parE nucleotide sequence reported in the GenBank. Purified PCR products were sequenced with specific primers using an automated ABI Prism 3700 Analyzer (Applied Biosystems, Foster City, CA). The QRDR DNA sequences of gyrA, gyrB, parC, and parE for each of the isolates examined were compared with those of E. coli strain K-12. Analysis and comparison were performed with BLAST program at the National Center for Biotechnology Information (www.ncbi.nlm.nih.gov/BLAST) and ExPASY proteomics tools (www.expasy.ch/tools/#similarity).

Phylogenetic grouping

The major phylogenetic group of the E. coli strains was determined by multiplex PCR based on the presence of chuA, yjaA, and TSpE4, as described previously (Clermont et al., 2000).

Pulsed-field gel electrophoresis (PFGE)

PFGE of XbaI (Takara Bio Inc., Shiga, Japan) digested genomic DNA was carried out according to the Pulse Net Standardized Laboratory Protocol as described previously (Gautom, 1997) using a CHEF MAPPER apparatus (Bio-Rad, Laboratories, Hercules, CA). The similarities of the restriction fragment length polymorphisms were analyzed using Bionumerics software, version 4.0 (Applied Maths, Sint-Martens-Latem, Belgium) to produce a dendogram. Clustering was carried out by the unweighted pair-group method with arithmetic averages (UPGMA), based on the Dice similarity index.

Results

Prevalence of PMQR determinants

A total of 14 E. coli isolates representing 3.8% of the 365 E. coli isolates investigated were positive for PMQR genes. Among them, 10 were from samples from sick animals submitted for diagnostic investigation, and four were commensal isolates from healthy animals. Over all, qnrB, qnrS, and aac(6’)-Ib genes were detected in two (0.5%), eight (2.2%), and 10 (2.7%) isolates, respectively. Of the 10 aac(6’)-Ib genes detected, four (1.1%) were aac(6’)-Ib-cr. None of the isolates harbored qnrA, qnrC, qnrD, and qepA genes. The PMQR determinants were detected as early as 2005. All qnrB and qnrS determinants belonged to the qnrB4 and qnrS1 alleles, respectively, on subsequent sequencing. The prevalence of PMQR genes was slightly higher in NAL^R-CIP^S isolates (4.3%) compared to that in NAL^R-CIP^R isolates (3.3%), whereas prevalence of PMQR genes was higher in cattle (7.3%) than in pigs (2.3%). The occurence of various PMQR genes in NAL^R-CIP^S and NAL^R-CIP^R swine and cattle E. coli isolates is compared in Table 2.

Table 2.

Prevalence of Plasmid-Mediated Quinolone-Resistant Genes in Escherichia coli Isolates from Food Animals

	No. of PMQR gene-positive isolates (%)
	NAL^R - CIP^S (n=185)			NAL^R - CIP^R (n=180)
PMQR gene	Cattle (n=44)	Pig (n=141)	Total	Cattle (n=65)	Pig (n=115)	Total
qnrA	0	0	0	0	0	0
qnrB	2	0	2	0	0	0
qnrC	0	0	0	0	0	0
qnrD	0	0	0	0	0	0
qnrS	4	1	5	0	3	3
aac(6’)-Ib-cr	1	0	1	1	2	3
qepA	0	0	0	0	0	0
Total	7 (15.9)	1 (0.7)	8 (4.3)	1 (1.5)	5 (4.3)	6 (3.3)

PMQR, plasmid-mediated quinolone resistance; NAL^R, nalidixic acid resistant; CIP^S, ciprofloxacin susceptible; and CIP^R, ciprofloxacin resistant.

Antimicrobial resistance phenotype

The MICs for nalidixic acid, oxolinic acid, ciprofloxacin, enrofloxacin, norfloxacin, ofloxacin, and danofloxacin against PMQR-positive E. coli isolates are shown in Table 3. Of them, 2.7% (10/365) and 0.8% (3/365) of the isolates exhibited high level resistance to nalidixic acid (MIC≥256 mg/L) and ciprofloxacin (MIC≥64 mg/L), respectively. All PMQR-positive isolates were co-resistant to at least four additional classes of antimicrobial agents other than quinolones and fluroquinolones. The resistance patterns of antimicrobials other than quinolones are shown in Figure 1.

FIG. 1.

Dendogram generated by Bionumerics software showing the cluster analysis of XbaI- PFGE patterns of PMQR-positive Escherichia coli strains isolated from food-producing animals. Similarity analysis was performed by using the Dice coefficient, and clustering was done by the unweighted-pair group method with arithmetic averages (UGPMA). Details given include the strain, pulso-type of each strain, geographical origin of host animal, and resistance pattern of antimicrobials other than quinolones shown by each strain. XbaI- macrorestriction yielded a few DNA banding patterns in an E. coli strain 05-M-7-1 due to the autodegradation of the genomic DNA during agarose plug preparation and hence a cluster formed by this strain is ignored. NA, not applicable; AMP, ampicillin; CEP, cephalothin; FOX, cefoxitin; CHL, chloramfenicol; FFC, florfenicol; NEO, neomycin; GEN, gentamicin; STR, streptomycin; TET, tetracycline; SXT, trimethoprim-sulfamethoxazole.

Table 3.

Phenotypic and Genotypic Characteristics of PMQR-Positive Escherichia coli Isolates from Food Animals Described in This Study (n=14)

	Isolation			MIC (mg/L)								QRDR mutations in
Isolate	Source	Year	Phylogroup	NAL	OXA	CIP	ENR	NOR	OFX	DAN	PMQR gene	gyrA	gyrB	ParC	ParE
09-G-CF-141	Healthy cattle	2009	A	64	4	0.25	2	2	2	1	qnrS1	D87Y	WT	WT	WT
09-D-SF-3	Healthy pig	2009	B1	32	4	2	2	4	4	2	qnrS1	WT	WT	WT	WT
10-CU-4	Sick cattle	2010	A	256	32	2	4	0.5	4	4	qnrS1	S83L	WT	WT	WT
10-CU-3	Sick cattle	2010	A	256	4	0.5	4	1	8	4	qnrS1	S83L	WT	WT	WT
10-CU-8	Sick cattle	2010	A	256	32	1	4	2	4	4	qnrS1	S83L	WT	WT	WT
09-C-CF-19	Healthy cattle	2009	A	256	4	0.5	0.25	2	0.5	0.5	aac(6’)-Ib-cr	S83L	WT	WT	WT
10-V06-PE-08	Sick pig	2010	D	>256	128	4	1	32	2	4	aac(6’)-Ib-cr	S83L	WT	S80R	WT
11-CU-64	Sick cattle	2011	B1	32	64	0.125	0.5	0.5	0.5	0.5	qnrB4	S83A	WT	WT	WT
11-CU-44	Sick cattle	2011	B1	32	4	<0.125	0.5	0.5	1	0.5	qnrB4	S83A	WT	WT	WT
05-M-7-1	Sick pig	2005	D	>256	64	4	8	4	8	32	qnrS1	S83L	WT	WT	WT
05-C-1-2	Sick pig	2005	D	>256	128	4	4	4	4	32	qnrS1	S83L	WT	S80R	WT
07-DS-2-3-3	Sick pig	2007	A	>256	>256	64	64	64	64	128	qnrS1	S83L, D87N	WT	A56T, S80I	WT
7-DS-6-1-1	Sick pig	2007	D	>256	>256	128	128	>256	128	256	aac(6’)-Ib-cr	S83L, D87N	S492N	S80I, E84G	WT
08-EM-CF-2	Healthy cattle	2008	A	>256	>256	128	128	256	64	256	aac(6’)-Ib-cr	S83L, D87N	WT	S80I	S458A

PMQR, plasmid-mediated quinolone resistance; NAL, nalidixic acid; OXA, oxolinic acid; CIP, ciprofloxacin, ENR, enrofloxacin; NOR, norfloxacin; OFX, ofloxacin; DAN, danofloxacin; WT, wild-type gene; QRDR, quinolone resistance-determining region.

Mutations in the topoisomerase genes

All but one PMQR-positive isolates harbored one or more point mutations in the QRDR of topoisomerase genes (Table 3). Within the QRDR of gyrA, they showed single amino acid substitutions at codon Ser83 to Leu (n=5) or Ala (I=2) or at codon Asp87 to Tyr (n=1) as well as double mutations at codons Ser83 to Leu and Asp87 to Asn (n=3). Besides mutations within gyrA, five isolates revealed additional mutations in the parC: Ser80 to Arg (n=2) or Ile (n=1); Ser80 to Ile and Glu84 to Gly (n=1); and Ala56 to Thr and Ser80 to Ile (n=1). In addition, one isolate each had additional substitution in parE (Ser458-Ala) and gyrB (Ser492-Asn) genes, respectively. There were no strains with a gyrB, parC, or parE QRDR mutation alone. Overall, the most prevalent mutation in the QRDR was Ser83-Leu.

Phylogenetic grouping of PMQR-positive isolates

Three major phylogenetic lineages (A, B1, and D) were found among the isolates harboring PMQR genes. Phlygroups A (n=7) and D (n=4) were the most common (Table 3). None of the isolates belonged to virulent phylogroup B2.

PFGE analysis

In order to compare genetic relatedness of PMQR-positive E. coli strains, molecular typing was done by PFGE of XbaI-digested genomic DNA. Among them, at least 10 arbitrary pulso-types (designated as A through J) were observed (Fig. 1). While Pulso-type A (3 strains) and Pulso-type B (2 strains) consisted of identical clones, the eight strains belonging to the remaining Pulso-types C–J demonstrated<70% similarity and were unlikely to be derived from a single clone of E. coli. It is to be noted that only a few banding patterns were obtained from an E. coli strain (05-M-7-1) due to constant autodigestion of genomic DNA during agrose plug preparation, and thus a cluster formed by this strain is ignored throughout this article.

Discussion

In the present study, the prevalence of PMQR determinants in E. coli isolated from food-producing animals was investigated and the PMQR-positive isolates were characterized using molecular methods. This study provides the first report of plasmid mediated quinolone resistance in E. coli isolates from animals in Korea. Our study yielded a prevalence of 3.8% for PMQR genes among 365 nalidixic resistant and ciprofloxacin susceptible or resistant E. coli isolated from pigs and cattle during the 2003–2011 period. Since only a subset of isolates from the total E. coli isolated during this period was examined, this does not represent the true prevalence of PMQR determinants among the E. coli strains in 2003–2007. Nevertheless, our results are similar to those of previous studies on prevalence of PMQR genes in E. coli isolated from human feces (Jeong et al., 2011) and other clinical specimens (Kim et al., 2009) in Korea. In contrast, a previous study from China reported that 17.2% of the 64 E. coli isolates from food-producing animals contained one or more PMQR determinants (Ma et al., 2009). Similarly, our results demonstrated lower (3.8%, 14/365) prevalence of PMQR genes compared to that of a recent international collaborative study in E. coli isolated from animals, humans, food, and the environment in 13 European countries in which PMQR genes were identified in 15% of the E. coli isolates showing reduced susceptibility to fluoroquinolones and nalidixic acid (Veldman et al., 2011). The lower prevalence of PMQR genes in our study may be related to study design of this work where nalidixic acid susceptible isolates were excluded from PMQR screening.

In Korea, qnr genes were first detected in 0.8% (2/260) of human clinical E. coli isolates collected during 2001–2003 period (Jeong et al., 2005). We identified qnr genes in 2.7% of E. coli isolates and among them qnrS1 was the most predominant qnr subtype which was detected in eight (2.2%) of 365 E. coli isolates investigated. This is in contrast to previous findings from human clinical E. coli isolates (Tamang et al., 2008; Kim et al., 2009) and human fecal E. coli isolates (Jeong et al., 2011) in Korea in which the most predominant qnr subtype was qnrB4. However, like in our study, qnrS1 was the most predominant qnr subtype identified in E. coli isolates in recent reports from China (Huang et al., 2009) and Europe (Veldman et al., 2011). Moreover, qnrS1 has also been reported in E. coli from healthy piglets from Hungary (Szmolka et al., 2011) and in poultry E. coli from Italy (Cerquetti et al., 2009). As shown in other similar studies (Ma et al., 2009; Cerquetti et al., 2009), qnrA was not detected in this study. Furthermore, none of the studied isolates were positive for qnrC or qnrD gene. Thus, our data and those of others support the suggestion that that the most prevalent qnr determinant in isolates of animal origin may be qnrS rather than other PMQR determinant (Strahilevitz et al., 2009; Poirel et al., 2012). Although, the qnrS gene by itself confers low-level resistance to (fluoro)quinolones, it may facilitate the selection of mutants under selective pressure of antimicrobials (Strahilevitz et al., 2009).

In this study, aac(6’)-Ib-cr gene was detected in 1.1% (4/365) of the E. coli isolates examined. In a previous study in Korea, the aac(6’)-Ib-cr gene was detected in 2% of the total clinical Enterobacteriaceae studied, most of which were E. coli (Kim et al., 2009). Recently, we identified this gene in four (0.7%) nalidixic acid resistant non-typhoid Salmonella from livestock in Korea (Tamang et al., 2011), but to our knowledge, it has not been reported in E. coli from animals in Korea. Nevertheless, this gene is geographically widespread and has been reported from several countries of Asia, North America, and Europe with high frequency among both human and veterinary E. coli isolates (Strahilevitz et al., 2009; Poirel et al., 2012). In contrast, in a recent study aac(6’)-Ib-cr was rarely found in Salmonella and E. coli isolates from 13 European countries (Veldman et al., 2011) which might be explained by the inclusion criteria (reduced susceptibility to fluoroquinolones and nalidixic acid) defined for this study. Moreover, there was no isolate that contained multiple PMQR genes which suggests that PMQR determinants in animals in Korea circulate independently. Furthermore, despite previous reports of qepA in human clinical isolates from Korea (Kang et al., 2009; Kim et al., 2009) and Japan (Yamane et al., 2008) or in animal isolates from China (Ma et al., 2009; Huang et al., 2009), none of our studied isolates was qepA positive. The reason for this difference is unclear. Nevertheless, our findings indicate that the prevalence of PMQR determinants in nalidixic acid resistant E. coli from animals from Korea is relatively low and less diverse.

PMQR genes are known to facilitate selection of additional chromosomal resistance mutations in the presence of quinolones (Strahilevitz et al., 2009). We determined mutations in the QRDR of gyrA, gyrB, parC, and parE in all PMQR-positive isolates. Except one wild-type gyrA, all NAL^R-CIP^S isolates harbored single point mutation in the QRDR of gyrA either at codon Ser83 or Asp87, whereas all but one NAL^R-CIP^R isolates contained double point mutations in gyrA and additional single or double mutations in parC in majority of them which support the fact that single mutation in gyrA of E. coli may be sufficient to cause resistance to nalidixic acid but additional mutations in gyrA and/or the topoisomerase IV genes are required for fluoroquinolone resistance. The various mutation types observed in gyrA, gyrB, parC, or parE in this study had been described previously and most of them are associated with increase in MICs of nalidixic acid and/or fluoroquinolones (Hopkins et al., 2005). However, we could not support the role of PMQR determinants in promoting mutations in topoisomerase genes as has been previously described (Kim et al., 2009), since the PMQR-negative isolates were not analyzed in this study.

Phylogenetic analysis showed that a majority of PMQR-positive isolates were of avirulent phylogroups A (n=7) and B1 (n=3) which have been reported as animal and human commensal E. coli strains previously (Jakobsen et al., 2010). The PFGE analysis of the PMQR-positive E. coli revealed great genomic diversity in strains from pigs but the PMQR-positive strains from cattle were relatively homogenous. Among them, three of the eight qnrS1-positive strains or two strains carrying qnrB4 genes isolated from cattle were derived from a particular E. coli clone, respectively, whereas the remaining strains from cattle or pig carrying qnrS1 or aac(6’)-Ib-cr genes were clonally unrelated which indicate that the dissemination of PMQR genes in E. coli from pigs may be attributed to horizontal transmission and in E. coli from cattle was owing to combination of both horizontal and clonal spread.

In this study, a majority of PMQR-positive isolates were recovered during the 2009–2011 period. This can be explained by the fact that more clinical samples were submitted to the diagnostic laboratory of QIA, and also more E. coli isolates were received from the local veterinary service centers for diagnostic investigation during this period. Furthermore, a majority (10/14, 71.4%) of the isolates with PMQR genes were derived from sick animals compared to healthy animals (4/14, 28.6%). These findings suggest that PMQR genes are more frequently present in isolates from sick animals compared to commensal E. coli from the healthy food-producing animals. Nevertheless, this constitutes a potential public health concern since the recovered sick animals or the healthy animals enter the food chain.

There are a few limitations of this study. Since the basis of PMQR gene screening in this study was resistance to nalidixic acid and resistance or susceptibility to ciprofloxacin, this study was limited to investigation of PMQR determinants among the nalidixic acid resistant isolates. In addition, the prevalence of the PMQR determinants among E. coli strains during 2003–2011 reported in this study should be considered a minimum estimation since some of the new qnrB variants would not be detected by the PCR methods used in this study. Furthermore, we neither investigated the transferability of plasmids carrying PMQR genes nor examined the incompatibility group of these plasmids. Since E. coli isolates with PMQR genes demonstrated different pulso-types, further studies are needed to determine transferability and incompatibility of these plasmids because of the possibility of horizontal transfer of these resistance genes to humans via food chain.

In conclusion, this is the first report of occurrence of qnrB, qnrS, and aac(6’)-Ib-cr genes in E. coli isolates from animals in Korea. Isolation of PMQR genes in E. coli from food animals is a matter of concern, since they could be transmitted to humans via food animals. Thus, it is essential to carry out continuous surveillance and monitoring of spread of PMQR genes among bacteria from food animals and food-animal products.

Footnotes

Acknowledgments

We are grateful to Prof. Jungmin Kim of the Department of Microbiology, Kyungpook National University School of Medicine, for kindly providing positive control strains for various PMQR determinants. This work was supported by a grant from the Animal, Plant, and Fisheries Quarantine and Inspection Agency, Ministry of Food, Agriculture, Forestry, and Fisheries, Republic of Korea.

Disclosure Statement

No competing financial interests exist.

References

Aarestrup

, Wiuff

, Mølbak

, Threlfall

. Is it time to change fluoroquinolone breakpoints for Salmonella spp.? Antimicrob Agents Chemother, 2003; 47:827–829.

Cerquetti

, Garcia-Fernandez

, Giufre

, Fortini

, Accogli

, Graziani

, Luzzi

, Caprioli

, Carattoli

. First report of plasmid-mediated quinolone resistance determinant qnrS1 in an Escherichia coli strain of animal origin in Italy. Antimicrob Agents Chemother, 2009; 53:3112–3114.

Clermont

, Bonacorsi

, Bingen

. Rapid and simple determination of the Escherichia coli phylogenetic group. Appl Environ Microbiol, 2000; 66:4555–4558.

[CLSI] Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing; 20th Informational SupplementCLSI Document M100-S20-UWayne, PA: CLSI, 2009.

Everett

, Jin

, Ricci

, Piddock

LJV

. Contributions of individual mechanisms to fluoroquinolone resistance in 36 Escherichia coli strains isolated from humans and animals. Antimicrob Agents Chemother, 1996; 40:2380–2386.

Gautom

. Rapid pulsed-field gel electrophoresis protocol for typing of Escherichia coli O157:H7 and other Gram-negative organisms in 1 day. J Clin Microbiol, 1997; 35:2977–2980.

Giraud

, Brisabois

, Martel

, Chaslus-Dancla

. Comparative studies of mutations in animal isolates and experimental in vitro– and in vivo–selected mutants of Salmonella spp. suggest a counterselection of highly fluoroquinolone-resistant strains in the field. Antimicrob Agents Chemother, 1999; 43:2131–2137.

Guardabassi

, Schwarz

, Lloyd

. Pet animals as reservoirs of antimicrobial-resistant bacteria. J Antimicrob Chemother, 2004; 54:321–332.

Hopkins

, Davies

, Threlfall

. Mechanisms of quinolone resistance in Escherichia coli and Salmonella: Recent developments. Int J Antimicrob Agents, 2005; 25:358–373.

10.

Hopkins

, Wootton

, Day

, Threlfall

. Plasmid-mediated quinolone resistance determinant qnrS1 found in Salmonella enterica strains isolated in the UK. J Antimicrob Chemother, 2007; 59:1071–1075.

11.

Huang

, Dai

, Xia

, Du

, Qi

, Liu

, Wu

, Shen

. Increased prevalence of plasmid-mediated quinolone resistance determinants in chicken Escherichia coli isolates from 2001 to 2007. Foodborne Pathog Dis, 2009; 6:1203–1209.

12.

Jacoby

, Gacharna

, Black

, Miller

, Hooper

. Temporal appearance of plasmid-mediated quinolone resistance genes. Antimicrob Agents Chemother, 2009; 53:1665–16.

13.

Jakobsen

, Kurbasic

, Skjot-Rasmussen

, Ejrnaes

, Porsbo

, Pedersen

, Jensen

, Emborg

, Agerso

, Olsen

, Aarestrup

, Frimodt-Moller

, Hammerum

. Escherichia coli isolates from broiler chicken meat, broiler chickens, pork, and pigs share phylogroups and antimicrobial resistance with community-dwelling humans and patients with urinary tract infection. Foodborne Pathog Dis, 2010; 7:537–547.

14.

Jeong

, Yoon

, Kim

, Lee

, Choi

, Kim

, Woo

, Kim

. Detection of qnr in clinical isolates of Escherichia coli from Korea. Antimicrob Agents Chemother, 2005; 49:2522–2524.

15.

Jeong

, Bae

, Shin

, Kim

, Chang

, Jeong

, Kim

, Lee

, Ryoo

, Lee

. Fecal colonization of Enterobacteriaceae carrying plasmid-mediated quinolone resistance determinants in Korea. Microb Drug Resist, 2011; 17:507–512.

16.

Kang

, Tamang

, Seol

, Kim

. Dissemination of plasmid-mediated qnr, aac(6′)-Ib-cr, and qepA genes among 16S rRNA methylase producing Enterobacteriaceae in Korea. J Bacteriol Virol, 2009; 39:173–182.

17.

Kim

, Park

, Kim

E-C

, Jacoby

, Hooper

. Prevalence of plasmid-mediated quinolone resistance determinants over a nine-year period. Antimicrob Agents Chemother, 2009; 53:639–645.

18.

Lim

, Lee

, Nam

, Cho

, Kim

, Song

, Park

, Jung

. Antimicrobial resistance observed in Escherichia coli strains isolated from fecal samples of cattle and pigs in Korea during 2003–2004. Int J Food Microbiol, 2007; 116:283–286.

19.

, Zeng

, Chen

, Xu

, Wang

, Deng

, Lu

, Huang

, Zhang

, Liu

, Wang

. High prevalence of plasmid-mediated quinolone resistance determinants qnr, aac(6′)-Ib-cr, and qepA among ceftiofur-resistant Enterobacteriaceae isolates from companion and food-producing animals. Antimicrob Agents Chemother, 2009; 53:519–524.

20.

Martínez-Martínez

, Pascual

, Jacoby

. Quinolone resistance from a transferable plasmid. Lancet, 1998; 351:797–799.

21.

Molbak

, Baggesen

, Aarestrup

, Ebbesen

, Engberg

, Frydendahl

, Gernersmidt

, Petersen

, Wegener

. An outbreak of multidrug-resistant, quinolone-resistant, Salmonella enterica serotype Typhimurium DT104. N Engl J Med, 1999; 341:1420–1425.

22.

Park

, Robicsek

, Jacoby

, Sahm

, Hooper

. Prevalence in the United States of aac(6′)Ib-cr encoding a ciprofloxacin-modifying enzyme. Antimicrob Agents Chemother, 2006; 50:3953–3955.

23.

Poirel

, Cattoir

, Nordmann

. Is plasmid-mediated quinolone resistance a clinically significant problem? Clin Microbiol Infect, 2008; 14:295–297.

24.

Poirel

, Cattoir

, Nordmann

. Plasmid-mediated quinolone resistance; interactions between human, animal, and environmental ecologies. Front Microbiol, 2012; 3:24.

25.

Robicsek

, Strahilevitz

, Jacoby

, Macielag

, Abbanat

, Park

, Bush

, Hooper

. Fluoroquinolone-modifying enzyme: A new adaptation of a common aminoglycoside acetyltransferase. Nat Med, 2006; 12:83–88.

26.

Ruiz

. Mechanisms of resistance to quinolones: Target alterations, decreased accumulation and DNA gyrase protection. J Antimicrob Chemother, 2003; 51:1109–1117.

27.

Simjee

, White

, McDermott

, Wagner

, Zervos

, Donabedian

, English

, Hayes

, Walker

. Characterization of Tn1546 in vancomycin-resistant Enterococcus faecium isolated from canine urinary tract infections: Evidence of gene exchange between human and animal enterococci. J Clin Microbiol, 2002; 40:4659–4665.

28.

Strahilevitz

, Jacoby

, Hooper

, Robicsek

. Plasmid-mediated quinolone resistance: A multifaceted threat. Clin Microbiol Rev, 2009; 22:664–689.

29.

Szmolka

, Fortini

, Villa

, Carattoli

, Anjum

, Nagy

. First report on IncN plasmid-mediated quinolone resistance gene qnrS1 in porcine Escherichia coli in Europe. Microb Drug Resist, 2011; 17:567–573.

30.

Tamang

, Nam

, Kim

, Lee

, Kim

, Jang

, Jung

, Lim

. Prevalence and mechanisms of quinolone resistance among selected nontyphoid Salmonella isolated from food Animals and humans in Korea. Foodborne Pathog Dis, 2011; 8:1199–1206.

31.

Tamang

, Seol

, Oh

, Kang

, Lee

, Cho

, Kim

. Plasmid-mediated quinolone resistance determinants qnrA, qnrB, and qnrS among clinical isolates of Enterobacteriaceae in a Korean hospital. Antimicrob Agents Chemother, 2008; 52:4159–4162.

32.

Veldman

, Cavaco

, Mevius

, Battisti

, Franco

, Botteldoorn

, Bruneau

, Perrin-Guyomard

, Cerny

, De Frutos Escobar

, Guerra

, Schroeter

, Gutierrez

, Hopkins

, Myllyniemi

, Sunde

, Wasyl

, Aarestrup

. International collaborative study on the occurrence of plasmid-mediated quinolone resistance in Salmonella enterica and Escherichia coli isolated from animals, humans, food, and the environment in 13 European countries. J Antimicrob Chemother, 2011; 66:1278–1286.

33.

Yamane

, Wachino

J-I

, Suzuki

, Kimura

, Shibata

, Kato

, Shibayama

, Konda

, Arakawa

. New plasmid-mediated fluoroquinolone efflux pump, QepA, found in an Escherichia coli clinical isolate. Antimicrob Agents Chemother, 2007; 51:3354–3360.

34.

Yamane

, Wachino

, Suzuki

, Arakawa

. Plasmid-mediated qepA gene among Escherichia coli clinical isolates from Japan. Antimicrob Agents Chemother, 2008; 52:1564–1566.