Characterization of Extended-Spectrum β-Lactamase-Carrying Plasmids in Clinical Isolates of Klebsiella pneumoniae from Taiwan

Abstract

We analyzed the replicon types, sizes, and restriction fragment length polymorphism (RFLP) typing of plasmids carrying extended-spectrum β-lactamase (ESBL) genes in Klebsiella pneumoniae isolates from Taiwan. Fifty-one Escherichia coli transconjugant strains with plasmids from ESBL-producing K. pneumoniae from the Taiwan Surveillance of Antimicrobial Resistance III Program in 2002 were included. All the 51 plasmids carried a bla_CTX-M gene, the majority of which were bla_CTX-M-3 (28/51 [54.9%]). Plasmids ranged in size from 126 to 241 kb by S1 nuclease digestion and subsequent pulsed-field gel electrophoresis, and the most common plasmid size (37.3%) was 161–170 kb. The most common replicon type of plasmids was incompatibility group (Inc)A/C (60.8%). The IncA/C plasmids all carried bla_CTX-M (bla_{CTX-M-3, -14, -15}), and some also carried bla_SHV (bla_{SHV-5, -12}) genes. All 51 plasmids could be typed with PstI, and 27 (52.9%) belonged to 10 clusters. Thirty-eight of the 51 plasmids were typable with BamHI, and 21 plasmids (55.3%) fell into 7 clusters. Plasmids in the same cluster belonged to the same incompatibility group, with the exception of cluster C6. In conclusion, IncA/C plasmids are the main plasmid type responsible for the dissemination of ESBL genes of K. pneumoniae from Taiwan. RFLP with PstI possessed better discriminatory power than that with BamHI and PCR-based replicon typing for ESBL-carrying plasmids in K. pneumoniae in this study. Greater than 50% of plasmids fell into clusters, and >60% of cluster-classified plasmids were present in clonally unrelated isolates, indicating that horizontal transfer of plasmids plays an important role in the spread of ESBL genes.

Introduction

Klebsiella pneumoniae is an important pathogen that frequently causes community-acquired and nosocomial infections, including pneumonia, urinary tract infections, pyogenic liver abscesses, bloodstream infections, and septic shock.^1,2 The increasing resistance of K. pneumoniae to many antimicrobial agents has compromised the treatment of infections. Extended-spectrum β-lactamases (ESBLs) hydrolyze penicillins, third-generation cephalosporins, and aztreonam are the reason that Enterobacteriaceae, including K. pneumoniae, are resistant to extended-spectrum β-lactam treatment.³ ESBL genes are carried on plasmids and distributed by plasmids among different species.^4,5 Classification of plasmids on the basis of molecular typing and phylogenetic relatedness may help understand the distribution of plasmid types, the relationships involving plasmids, and the interspecies transmission of plasmids carrying antimicrobial resistance genes.^6–8 Plasmids can be classified into incompatibility groups (Inc) by replicon typing or types (clusters) by restriction fragment length polymorphism (RFLP) analysis.⁹ The PCR-based replicon typing (PBRT) method developed by Carattoli et al.⁶ can be applied to a large number of strains.

The prevalence of ESBL-producing K. pneumoniae (ESBL-KP) increased from 8% to 18.3% between 2002 and 2009 in Taiwan¹⁰; however, an analysis of the replicon types of ESBL gene-bearing plasmids in ESBL-KP collected in 2002 has not been conducted, and very few relevant studies for K. pneumoniae were reported in Taiwan.^11,12 This study determined the replicon type, plasmid size, clonal relatedness, and classification of these plasmids into RFLP-based clusters to reveal the role of plasmids in the spread of ESBL-associated antibiotic resistance and the epidemiologic implications.

Materials and Methods

Bacterial strains

Clinical isolates of ESBL-KP were collected by the Taiwan Surveillance of Antimicrobial Resistance (TSAR) III Program in 2002. As previously described in the Lin study,¹⁰ the conjugation of the ESBL genes was carried out on 66 ESBL-KP isolates by broth mating out with the rifampicin-resistant strain of Escherichia coli (JP-995) or azide-resistant strain of E. coli (J53) as recipients. Transconjugants were selected on MacConkey agar plates containing rifampicin (100 mg/L) or azide (100 mg/L), as appropriate, and cefotaxime (5 mg/L) or ceftazidime (5 mg/L). Fifty-seven (86.4%) strains were transferable, and the ESBL genes in E. coli transconjugants were characterized by Lin et al.¹⁰ Fifty-one E. coli transconjugant strains harboring plasmids from clinical isolates of ESBL-KP were included in this study. These ESBL-KP strains were isolated from 14 hospitals located in the 4 geographic regions (north, west, south, and east) of Taiwan. To characterize the clonality of ESBL-KP strains, pulsed-field gel electrophoresis (PFGE) groups were previously determined in the Lin study.¹⁰

Antimicrobial susceptibility testing

Antimicrobial susceptibility testing of transconjugant strains was performed by the standard broth microdilution method according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI).¹³ The following antimicrobial agents were tested: ampicillin, amikacin, ceftazidime, cefmetazole, cefazolin, ertapenem, cefepime, cefoxitin, gentamicin, levofloxacin, imipenem, ampicillin–sulbactam, trimethoprim–sulfamethoxazole, and piperacillin–tazobactam.

Plasmid replicon typing

In all 51 transconjugant strains, the plasmid incompatibility group was determined by a PBRT scheme proposed by Johnson and Nolan,¹⁴ who modified the procedure described by Carattoli et al.⁶ Three primer panels, including 18 primer pairs, were used to perform 3 multiplex PCRs. Primers used in plasmid replicon typing were described previously.^6,14 Bacterial template DNA was prepared using a Geneaid DNA Isolation Kit (Geneaid, Taipei, Taiwan) according to the manufacturer's instructions. For each PCR panel, equal amounts of each primer (0.1 mM) were combined in a pooled primer tube. The PCR amplification was performed in 20 μl volumes containing 200 ng bacterial template DNA, 0.2 mM dNTP, 2 mM MgCl₂, 10× PCR buffer, 1 μl primer pool, and 1 U Taq DNA polymerase (Invitrogen, Carlsbad, CA) in a GeneAmp PCR System 2720 (Applied Biosystems, Foster City, CA). The PCR conditions were as follows: 5 min at 94°C; 30 cycles of 30 sec at 94°C, 30 sec at 60°C, and 90 sec at 72°C; and a final extension of 5 min at 72°C. Amplicons were visualized after separation by electrophoresis on a 2% agarose gel.

The IncF replicon sequence typing (RST) scheme was performed according to the scheme proposed by Villa et al.¹⁵ Alleles were assigned by submitting the amplicon sequence to the Web site www.pubmlst.org/plasmid/.

S1 nuclease digestion and subsequent PFGE to determine plasmid size

The sizes of the plasmids were determined by PFGE after S1 nuclease digestion (S1-PFGE), as previously described,¹⁶ with some modifications. S1-digested slices were loaded onto an agarose gel, and the gel was run on a CHEF-DR III system (Bio-Rad, Hercules, CA) at 14°C, with a switch time of 7–35 sec for 20 hr at 6 V/cm on a 120° angle in 0.5× TBE buffer.

Plasmid RFLP analysis

Plasmids were extracted using a Qiagen Plasmid Midi Kit (Qiagen, Hilden, Germany) following the manufacturer's instructions. For RFLP analysis, purified plasmid DNA was digested with BamHI or PstI restriction enzyme (New England Biolabs, Ipswich, MA) and subjected to electrophoresis on a 1% agarose gel at 100 V for 3 hr. Plasmid RFLP fingerprints were analyzed using GelCompar II software 6.5 (Applied Maths, Austin, TX), and dendrograms of the patterns were constructed using the unweighted pair group method with arithmetic mean (UPGMA) based on the Dice similarity index.

Results

ESBL genes in E. coli transconjugants

The ESBL genes in E. coli transconjugants were previously characterized by Lin et al.¹⁰ All 51 E. coli transconjugants had bla_CTX-M genes, including the most prevalent (bla_CTX-M-3: 28/51 [54.9%]), followed by bla_CTX-M-14 (12/51 [23.5%]) and bla_CTX-M-15 (11/51 [21.6%]; Table 1). Twenty-five of the 51 E. coli transconjugants carried bla_SHV-12 in addition to bla_CTX-M genes: 17 of them carried bla_CTX-M-3, 3 carried bla_CTX-M-14, and 5 carried bla_CTX-M-15. Furthermore, bla_SHV-5 coexisted with bla_CTX-M-3 or bla_CTX-M-15 in four and six transconjugants, respectively (Table 1).

Table 1.

Characterization of Extended-Spectrum β-Lactamase-Carrying Plasmids

Plasmid replicon (n = 51)						Plasmid RFLP clusters ^c
Type	n (%)	Transconjugant number	ESBL genes ^a	Resistance profile of transconjugants ^b	Plasmid size, kb	BamHI (n = 38)^e	PstI (n = 51)	PFGE groups (hospital) ^d
A/C	31 (60.8)	45	bla _CTX-M-3	AM AN CZ GM SAM SXT	161–170	—	C2
		214	bla _CTX-M-3	AM CAZ CZ GM SAM	151–160	Noncluster	C6
		46	bla_CTX-M-3, bla_SHV-5	AM AN CZ GM SAM SXT	151–160	—	C2
		86	bla_CTX-M-3, bla_SHV-5	AM AN CZ GM SAM SXT	151–160	—	Noncluster
		102	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ GM	161–170	Noncluster	C8	G (S2)
		105	bla_CTX-M-3, bla_SHV-12	AM AN CAZ CZ GM	151–160	CVII	C8	G (S2)
		107	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ GM	141–150	CVII	C8	G (S2)
		110	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ GM	141–150	CVII	Noncluster	G (S2)
		119	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ GM	151–160	CVII	Noncluster	E (E1)
		122	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ GM	161–170	CI	C9	E (E1)
		132	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ FOX SXT TZP	161–170	CI	C9
		156	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ SXT	151–160	CI	C9	A (N4)
		182	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ GM TZP	131–140	Noncluster	Noncluster	F (S3)
		55	bla _CTX-M-14	AM AN CZ GM SAM SXT	161–170	—	C1
		65	bla _CTX-M-14	AM AN CZ GM SAM SXT	161–170	—	C1
		70	bla _CTX-M-14	AM AN CZ GM SAM SXT	171–180	—	Noncluster
		74	bla _CTX-M-14	AM AN CZ GM SAM SXT	161–170	—	Noncluster
		106	bla _CTX-M-14	AM CZ SAM	171–180	—	C3
		145	bla _CTX-M-14	AM AN CZ GM SAM SXT	161–170	—	C2
		172	bla _CTX-M-14	AM AN CZ GM SAM SXT	131–140	—	C2
		205	bla _CTX-M-14	AM AN CZ GM SAM SXT	141–150	—	C3
		67	bla_CTX-M-14, bla_SHV-12	AM AN CZ GM SAM SXT	161–170	—	C1
		127	bla_CTX-M-14, bla_SHV-12	AM CZ GM SAM	161–170	—	Noncluster
		92	bla_CTX-M-15, bla_SHV-5	AM CAZ CZ GM	161–170	CIII	Noncluster	G (S2)
		99	bla_CTX-M-15, bla_SHV-5	AM CAZ CZ GM	161–170	CI	C9	G (S2)
		100	bla_CTX-M-15, bla_SHV-5	AM CAZ CZ GM	151–160	CIII	C10	G (S2)
		112	bla_CTX-M-15, bla_SHV-5	AM CAZ CZ GM	141–150	CIII	C10	G (S2)
		189	bla_CTX-M-15, bla_SHV-5	AM CAZ CZ GM SXT	>200	CI	Noncluster	F (S3)
		118	bla_CTX-M-15, bla_SHV-12	AM CAZ CZ GM	>200	CI	C9	E (E1)
		124	bla_CTX-M-15, bla_SHV-12	AM AN CAZ GM	171–180	CII	Noncluster	E (E1)
		197	bla_CTX-M-15, bla_SHV-12	AM AN CAZ CZ GM TZP	171–180	CII	Noncluster
L/M	5 (9.8)	53	bla _CTX-M-3	AM AN CZ GM SXT	151–160	CVI	C4
		146	bla _CTX-M-3	AM CZ GM SAM	141–150	Noncluster	Noncluster
		174	bla _CTX-M-3	AM AN CZ GM SXT	141–150	Noncluster	C5
		125	bla_CTX-M-3, bla_SHV-12	AM CZ GM	141–150	CVI	C4	D (E1)
		215	bla_CTX-M-3, bla_SHV-12	AM CZ GM SAM SXT	161–170	Noncluster	C5
FII_K^f	5 (9.8)	103	bla _CTX-M-3	AM CZ	131–140	Noncluster	Noncluster
		104	bla_CTX-M-3, bla_SHV-5	AM CAZ CZ GM	121–130	CV	C7
		114	bla_CTX-M-3, bla_SHV-5	AM CAZ SAM	121–130	CV	C7
		199	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ	161–170	Noncluster	Noncluster
		196	bla_CTX-M-15, bla_SHV-5	AM CAZ CZ FOX	171–180	Noncluster	Noncluster
N	2 (3.9)	191	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ GM SAM SXT	161–170	Noncluster	Noncluster
		59	bla_CTX-M-14, bla_SHV-12	AM CAZ CZ	141–150	Noncluster	Noncluster
FII^g	1 (2.0)	136	bla _CTX-M-3	AM CAZ CZ GM SAM	161–170	Noncluster	C6
F	1 (2.0)	155	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ FOX SAM SXT TZP	151–160	Noncluster	Noncluster	A (N4)
A/C+N	1 (2.0)	80	bla _CTX-M-14	AM CAZ CZ FOX SAM SXT	161–170	Noncluster	Noncluster
L/M+N	1 (2.0)	58	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ	161–170	Noncluster	Noncluster
Nontypable	4 (7.8)	24	bla_CTX-M-3, bla_SHV-12	AM CAZ CZ SAM SXT	171–180	CIV	Noncluster
		123	bla_CTX-M-3, bla_SHV-12	AM CAZ CMZ CZ FOX IPM SAM SXT	151–160	Noncluster	Noncluster	D (E1)
		120	bla_CTX-M-15, bla_SHV-12	AM CAZ CMZ CZ FOX SXT	161–170	Noncluster	Noncluster	D (E1)
		137	bla_CTX-M-15, bla_SHV-12	AM CAZ CZ GM SXT	121–130	CIV	Noncluster

ESBL genes were characterized in the Lin study.¹⁰

Non-β-lactams are underlined. AM, ampicillin; AN, amikacin; CAZ, ceftazidime; CMZ, cefmetazole; CZ, cefazolin; FOX, cefoxitin; GM, gentamicin; IPM, imipenem; SAM, ampicillin–sulbactam; SXT, trimethoprim–sulfamethoxazole; TZP, piperacillin–tazobactam.

Plasmids with ≥80% similarity in the RFLP patterns were grouped into clusters.

PFGE groups of ESBL-KP were determined in the Lin study.¹⁰ The locations of hospitals from which the ESBL-KP were isolated were in the northern (N), western (W), southern (S), and eastern (E) regions of Taiwan.

“—” Indicates that plasmids did not yield RFLP patterns with BamHI.

FAB formulae of plasmids by IncF RST were K2:A-:B- (transconjugants 103, 104, 114, and 199) and K9:A-:B- (transconjugant 196).

FAB formula of the IncFII plasmid was F2:A-:B- (transconjugant 136).

ESBL, extended-spectrum β-lactamase; ESBL-KP, ESBL-producing Klebsiella pneumoniae; PFGE, pulsed-field gel electrophoresis; RFLP, restriction fragment length polymorphism; RST, replicon sequence typing.

Resistance profile of E. coli transconjugants

The resistance profile of E. coli transconjugants is shown in Table 1. All strains were resistant to ampicillin. Most of the strains were resistant to cefazolin (49/51 [96.1%]), followed by gentamicin (37/51 [72.5%]) and ceftazidime (32/51 [62.7%]). One (2.0%) strain (transconjugant 123) was resistant to imipenem. Carbapenemase genes of transconjugant 123 were detected by PCR amplification as previously described.¹⁷ The bla_IMP-8 was harbored in this transconjugant strain. Among non-β-lactams, resistance to gentamicin (72.5%) was most often cotransferred with the ESBL genes, followed by trimethoprim–sulfamethoxazole (24/51 [47.1%]) and amikacin (16/51 [31.4%]).

Plasmid replicon typing

A three-panel multiplex PCR protocol was used to examine the 51 E. coli transconjugants carrying ESBL genes for the presence of 18 plasmid replicon types. RST for IncF plasmids was performed to subcategorize IncF plasmids. The most frequently detected replicon types were A/C (60.8%), followed by L/M (9.8%), FII_K (9.8%), N (3.9%), FII (2.0), and F (2.0%; Table 1). Among the five IncFII_K replicons, the FAB formulae by IncF RST were K2:A-:B- in four plasmids (transconjugants 103, 104, 114, and 199) and K9:A-:B- in one plasmid (transconjugant 196). The FAB formula of the IncFII plasmid was F2:A-:B- (transconjugant 136). Two transconjugants carried two different replicon types on a single plasmid, one with A/C and N and the remaining transconjugant with L/M and N. Additionally, the replicon type could not be determined in four transconjugants (7.8%).

Table 1 reveals the association of the predominant IncA/C replicon with bla_CTX-M alone or bla_CTX-M coexisting with bla_SHV genes, mainly bla_CTX-M-3 and bla_SHV-12 (9/31 [29%]), followed by bla_CTX-M-14 (8/31 [25.8%]), bla_CTX-M-15 and bla_SHV-5 (5/31 [16.1%]), bla_CTX-M-15 and bla_SHV-12 (3/31 [9.7%]), bla_CTX-M-3 (2/31 [6.5%]), bla_CTX-M-3 and bla_SHV-5 (2/31 [6.5%]), and bla_CTX-M-14 and bla_SHV-12 (2/31 [6.5%]). The IncL/M and IncFII_K replicons were associated with bla_CTX-M-3 or bla_CTX-M-3 coexisting with bla_SHV-5 or bla_SHV-12 (except the plasmid with bla_CTX-M-15 and bla_SHV-5 in transconjugant 196).

Plasmid size determined by S1 nuclease digestion and subsequent PFGE

S1 nuclease–PFGE analysis indicated that all E. coli transconjugants contained a single plasmid (size range = 121–180 kb), with the exception that two transconjugants harbored plasmids larger than 200 kb (Table 1). The most common plasmid size (19/51 [37.3%]) was 161–170 kb. The A/C replicon, bla_CTX-M, and bla_SHV were distributed among various-sized plasmids.

RFLP analysis

Thirty-eight of the 51 plasmids were successfully characterized using BamHI, and 21 plasmids (21/38 [55.3%]) were classified into 7 clusters (CI–CVII), with at least 80% similarity (Fig. 1 and Table 1). The similarities in RFLP patterns of 17 unclassified plasmids were smaller than 75%. All plasmids in the same cluster belonged to the same incompatibility group, with the exception of two plasmids in cluster IV in which incompatibility groups could not be determined.

FIG. 1.

Dendrogram of RFLP patterns of plasmids digested with BamHI. Plasmids with ≥80% similarity in their RFLP patterns were grouped into clusters. Twenty-one of 38 plasmids belonged to clusters (CI–CVII). RFLP, restriction fragment length polymorphism.

Although 13 of the 51 plasmids did not yield RFLP patterns with BamHI, all plasmids were successfully typed using PstI. Twenty-seven (52.9%) of the 51 plasmids were separated into 10 clusters (C1–C10) based on at least 80% RFLP similarity (Fig. 2 and Table 1). IncA/C plasmids were the most prevalent in these 10 clusters (74.1% [20/27]). Plasmids in the same cluster belonged to the same incompatibility group, except for the plasmids in cluster C6. ESBL genes (bla_CTX-M and bla_SHV) were present in both cluster-classified and cluster-unclassified plasmids (Table 1).

FIG. 2.

Dendrogram of RFLP patterns of plasmids digested with PstI. Plasmids with ≥80% similarity in their RFLP patterns were grouped into clusters. Twenty-seven of 51 plasmids belonged to 10 clusters (C1–C10).

Plasmids in strains 24 and 137 with nontypable replicons could be classified into cluster CIV by BamHI restriction (Fig. 1 and Table 1). Although plasmids with nontypable replicons could not be classified into clusters by PstI, the relationship was demonstrated by RFLP by BamHI analysis. This indicated that plasmid RFLPs can be helpful in understanding plasmid relatedness when incompatibility groups are not identifiable by PCR replicon typing.

Comparing the RFLP profiles generated by BamHI to those generated by PstI, clusters CI, CIII, CV, CVI, and CVII identified by BamHI corresponded to the C9, C10, C7, C4, and C8 identified by PstI, respectively, even though the number of plasmids in the clusters was not all the same (Table 1).

PFGE analysis

PFGE groups of the ESBL-KP strains were determined in the Lin study.¹⁰ The PFGE analysis revealed that 19 of the 51 isolates (37.3%) belonging to 5 groups (A, D, E, F, and G) indicated clonal relatedness (Table 1). The remaining 32 isolates had distinct PFGE patterns and were clonally unrelated. Groups A, D, E, F, and G comprised two, three, four, two, and eight isolates, respectively. Isolates in the same group were from the same hospital, indicating some ESBL-KP isolates were clonally spread in some hospitals. Among 27 cluster-classified plasmids by PstI restriction, 17 (63% [17/27]) plasmids were present in clonally unrelated isolates (Table 1). Moreover, only 6 clonally related isolates carried RFLP-clustered plasmids typed by BamHI and PstI (105, 107, 122, 100, 112, and 118), while the other 13 clonally related isolates carried plasmids of different clusters or cluster-unclassified plasmids (Table 1).

Discussion

In the current study, the most prevalent replicon type in ESBL-KP isolates in Taiwan was IncA/C (60.8%), which is similar to that reported in Paraguay (48% of K. pneumoniae isolates with IncA/C plasmids)¹⁸ but different from Tunisia (93.8% of K. pneumoniae isolates with IncFIIA),¹⁹ 9 Asian countries (81.8% with IncFIIA),²⁰ China (66.7% with IncFII),²¹ Bulgaria (58.3% with IncL/M),²² Scotland (IncN),²³ Pakistan (IncF),²⁴ and the Central African Republic (IncF).²⁵ Thus, the prevalent plasmid replicon types are diverse in ESBL-KP worldwide.

Very few studies reported the replicon types of plasmids in clinical isolates from Taiwan, including K. pneumoniae. Lee et al.¹² reported an IncFIIA plasmid among the four self-transferable plasmids in ESBL-KP (the other three were nontypable) from a medical center in southern Taiwan between 2004 and 2005. Wang et al.¹¹ found that the IncF replicon type was harbored in 7 of the 16 KPC-2-positive, carbapenem-nonsusceptible K. pneumoniae isolates from four hospitals between 2010 and 2012. These results of K. pneumoniae in Taiwan are different from those of the current study.

Some studies have reported a strong link between ESBL genes and particular plasmid types in K. pneumoniae, that is, the association of bla_CTX-M-15 with IncF-type plasmid,^19,22,25,26 IncN,²³ or IncR,²⁷ the association of bla_CTX-M-3 with IncL/M,²² the association of bla_CTX-M-14 with IncF,²⁸ and the association of bla_SHV-12 with IncR or IncHI2.²⁷ In contrast, the current study revealed a predominant linkage of IncA/C plasmids with ESBL genes (bla_CTX-M alone or bla_CTX-M coresident with bla_SHV genes), indicating IncA/C plasmids play an important role in the dissemination of ESBL genes in clinical strains of K. pneumoniae in Taiwan. Furthermore, resistances to non-β-lactams were cotransferred with the ESBL genes, that is, gentamicin, trimethoprim–sulfamethoxazole, or amikacin in the current study. Replicon A/C has also been associated with TEM β-lactamase genes,^26,29–31 AmpC β-lactamases CMY (mainly CMY-2),^29,32 and more recently associated with VIM carbapenemases.²⁹ The plasmids were a platform for distribution of the genes of resistance determinant enzymes.

Studies have described that the IncF plasmid family (with different subgroups) is the most prevalent in ESBL-carrying K. pneumoniae worldwide and associated with bla_CTX-M genes, specifically bla_CTX-M-15.^{19,20,25,26,29,33–35} Lee et al.²⁰ reported an IncFIIA-type plasmid predominantly associated with bla_CTX-M-15 in nine Asian countries. Zhuo et al.³⁵ found an epidemic IncFII plasmid carrying bla_CTX-M-15 in southern China. These results of Asian areas are different from those of the current study; however, international traffic and geographic proximity may facilitate the dissemination of plasmids and influence the trend of dominant plasmid replicon types and the linkage with ESBL genes. Moreover, clonal spread also occurred in the Asian studies, including the current study.^20,35 Therefore, continuous monitoring and characterization of plasmids together with monitoring of clones are necessary.

Plasmid RFLP typing demonstrated that >50% of plasmids fell into clusters (55.3% typed by BamHI and 52.9% by PstI), and plasmids in the same cluster belonged to the same incompatibility group, with the exception of the cluster C6 typed by PstI restriction. Furthermore, 63% of the 27 cluster-classified plasmids by PstI restriction were present in clonally unrelated isolates. This highlights that the circulation of similar plasmids in K. pneumoniae strains is mediated by horizontal transfer. Various RFLP patterns were identified even in the same incompatibility group. This may be partially attributed to differences in the targets of replicon typing and RFLP analysis. The targets for PBRT are those genes associated with replication, including replication genes, ori sites, iteron sequences, and plasmid-partitioning genes^6,14; however, the targets for RFLP are whole plasmid sequences recognized by restriction enzymes. RFLP is more discriminative than PBRT. RFLP with PstI possessed better discrimination power than RFLP with BamHI. Furthermore, plasmids in the same incompatibility group with distinct RFLP patterns suggest that the composition of plasmids with similar backbones is continuously evolving.

Replicon typing showed that ESBL genes (bla_CTX-M and bla_SHV) are carried by plasmids of different incompatibility groups other than the predominant IncA/C. Cluster analysis of plasmid RFLPs using PstI revealed that ESBL genes were distributed over cluster- and noncluster-classified plasmids. This indicated that ESBL genes can be found on similar or different plasmid platforms. Genetic elements, such as insertion sequences associated with ESBL genes, have been described. For example, studies have found that the insertion sequence, ISEcp1, is frequently located upstream of bla_CTX-M-3, bla_CTX-M-14, and bla_CTX-M-15^{28,33,34,36,37}; bla_CTX-M-14 is flanked by ISEcp1 and IS903,³⁶ and bla_SHV-5 and bla_SHV-12 are linked to IS26.^36,38 These genetic elements may be involved in the mobilization of ESBL genes and contribute to dissemination among plasmids.^36,37 The genetic environment of ESBL genes within the plasmids warrants further investigation.

In summary, this study demonstrated that IncA/C that carried bla_CTX-M and bla_SHV genes was the predominant replicon type in clinical ESBL-KP isolates from Taiwan. Plasmid RFLP typing can help determine the relatedness of plasmids when incompatibility groups cannot be determined. PstI possessed better discriminatory power than BamHI for the plasmids included in the current study. Greater than 50% of plasmids were divided into clusters, and >60% of cluster-classified plasmids by PstI restriction were present in clonally unrelated isolates, indicating that horizontal transfer of plasmids plays an important role in the wide distribution of ESBL genes.

Footnotes

Acknowledgments

This work was supported by the Department of Health, Taiwan (DOH102-DC-1508), National Health Research Institutes, and Kaohsiung Medical University Hospital, Taiwan.

Disclosure Statement

No competing financial interests exist.

References

W.C.

, Paterson

D.L.

, Sagnimeni

A.J.

, Hansen

D.S.

, Von Gottberg

, Mohapatra

, Casellas

J.M.

, Goossens

, Mulazimoglu

, Trenholme

, Klugman

K.P.

, McCormack

J.G.

, and Yu

V.L.

2002. Community-acquired Klebsiella pneumoniae bacteremia: global differences in clinical patterns. Emerg. Infect. Dis., 8:160–166.

Podschun

, and Ullmann

1998. Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors. Clin. Microbiol. Rev., 11:589–603.

Rawat

, and Nair

2010. Extended-spectrum β-lactamases in Gram Negative Bacteria. J. Glob. Infect. Dis., 2:263–274.

Doi

, Adams-Haduch

J.M.

, Peleg

A.Y.

, and D'Agata

E.M.

2012. The role of horizontal gene transfer in the dissemination of extended-spectrum beta-lactamase-producing Escherichia coli and Klebsiella pneumoniae isolates in an endemic setting. Diagn. Microbiol. Infect. Dis., 74:34–38.

W.L.

, Chuang

Y.C.

, and Walther-Rasmussen

2006. Extended-spectrum beta-lactamases in Taiwan: epidemiology, detection, treatment and infection control. J. Microbiol. Immunol. Infect., 39:264–277.

Carattoli

, Bertini

, Villa

, Falbo

, Hopkins

K.L.

, and Threlfall

E.J.

2005. Identification of plasmids by PCR-based replicon typing. J. Microbiol. Methods, 63:219–228.

Francia

M.V.

, Varsaki

, Garcillan-Barcia

M.P.

, Latorre

, Drainas

, and de la Cruz

2004. A classification scheme for mobilization regions of bacterial plasmids. FEMS Microbiol. Rev., 28:79–100.

Garcia-Fernandez

, and Carattoli

2010. Plasmid double locus sequence typing for IncHI2 plasmids, a subtyping scheme for the characterization of IncHI2 plasmids carrying extended-spectrum β-lactamase and quinolone resistance genes. J. Antimicrob. Chemother., 65:1155–1161.

Hopkins

K.L.

, Liebana

, Villa

, Batchelor

, Threlfall

E.J.

, and Carattoli

2006. Replicon typing of plasmids carrying CTX-M or CMY β-lactamases circulating among Salmonella and Escherichia coli isolates. Antimicrob. Agents Chemother., 50:3203–3206.

10.

Lin

C.J.

, Siu

L.K.

, Ma

, Chang

Y.T.

, and Lu

P.L.

2012. Molecular epidemiology of ciprofloxacin-resistant extended-spectrum β-lactamase-producing Klebsiella pneumoniae in Taiwan. Microb. Drug Resist., 18:52–58.

11.

Wang

J.T.

, Wu

U.I.

, Lauderdale

T.L.

, Chen

M.C.

, Li

S.Y.

, Hsu

L.Y.

, and Chang

S.C.

2015. Carbapenem-nonsusceptible Enterobacteriaceae in Taiwan. PLoS One, 10:e0121668.

12.

Lee

C.H.

, Liu

J.W.

, Li

C.C.

, Chien

C.C.

, Tang

Y.F.

, and Su

L.H.

2011. Spread of ISCR1 elements containing bla_DHA-1 and multiple antimicrobial resistance genes leading to increase of flomoxef resistance in extended-spectrum-β-lactamase-producing Klebsiella pneumoniae. Antimicrob. Agents Chemother., 55:4058–4063.

13.

Clinical and Laboratory Standards Institute. 2007. Performance Standards for Antimicrobial Susceptibility Testing; Seventeenth Informational Supplement M100-S17. CLSI, Wayne, PA.

14.

Johnson

T.J.

, and Nolan

L.K.

2009. Plasmid replicon typing. Methods Mol. Biol., 551:27–35.

15.

Villa

, Garcia-Fernandez

, Fortini

, and Carattoli

2010. Replicon sequence typing of IncF plasmids carrying virulence and resistance determinants. J. Antimicrob. Chemother., 65:2518–2529.

16.

Barton

B.M.

, Harding

G.P.

, and Zuccarelli

A.J.

1995. A general method for detecting and sizing large plasmids. Anal. Biochem., 226:235–240.

17.

, Lu

P.L.

, Siu

L.K.

, and Hsieh

M.H.

2013. Molecular typing and resistance mechanisms of imipenem-non-susceptible Klebsiella pneumoniae in Taiwan: results from the Taiwan Surveillance of Antibiotic Resistance (TSAR) study, 2002–2009. J. Med. Microbiol., 62:101–107.

18.

Khan

M.A.

, Lemmens

, Riera

, Blonk

, Goedhart

, Van Belkum

, Goessens

, Hays

J.P.

, and Van Westreenen

2009. Dominance of CTX-M-2 and CTX-M-56 among extended-spectrum β-lactamases produced by Klebsiella pneumoniae and Escherichia coli isolated in hospitals in Paraguay. J. Antimicrob. Chemother., 64:1330–1332.

19.

Lahlaoui

, De Luca

, Maradel

, Ben-Haj-Khalifa

, Ben Hamouda

, Kheder

, Ben Moussa

, Rossillini

G.M.

, and Docquier

J.D.

2015. Occurrence of conjugative IncF-type plasmids harboring the bla_CTX-M-15 gene in Enterobacteriaceae isolates from newborns in Tunisia. Pediatr. Res., 77:107–110.

20.

Lee

M.Y.

, Ko

K.S.

, Kang

C.I.

, Chung

D.R.

, Peck

K.R.

, and Song

J.H.

2011. High prevalence of CTX-M-15-producing Klebsiella pneumoniae isolates in Asian countries: diverse clones and clonal dissemination. Int. J. Antimicrob. Agents, 38:160–163.

21.

Cao

, Xu

, Zhang

, Shen

, Chen

, and Zhang

2014. Molecular characterization of clinical multidrug-resistant Klebsiella pneumoniae isolates. Ann. Clin. Microbiol. Antimicrob., 13:16.

22.

Markovska

, Schneider

, Ivanova

, Mitov

, and Bauernfeind

2014. Predominance of IncL/M and IncF plasmid types among CTX-M-ESBL-producing Escherichia coli and Klebsiella pneumoniae in Bulgarian hospitals. APMIS, 122:608–615.

23.

Younes

, Hamouda

, Dave

, and Amyes

S.G.

2011. Prevalence of transferable bla_CTX-M-15 from hospital- and community-acquired Klebsiella pneumoniae isolates in Scotland. J. Antimicrob. Chemother., 66:313–318.

24.

Habeeb

M.A.

, Haque

, Nematzadeh

, Iversen

, and Giske

C.G.

2013. High prevalence of 16S rRNA methylase RmtB among CTX-M extended-spectrum β-lactamase-producing Klebsiella pneumoniae from Islamabad, Pakistan. Int. J. Antimicrob. Agents, 41:524–526.

25.

Rafai

, Frank

, Manirakiza

, Gaudeuille

, Mbecko

J.R.

, Nghario

, Serdouma

, Tekpa

, Garin

, and Breurec

2015. Dissemination of IncF-type plasmids in multiresistant CTX-M-15-producing Enterobacteriaceae isolates from surgical-site infections in Bangui, Central African Republic. BMC Microbiol., 15:15.

26.

Carattoli

2009. Resistance plasmid families in Enterobacteriaceae. Antimicrob. Agents Chemother., 53:2227–2238.

27.

Rodrigues

, Machado

, Ramos

, Peixe

, and Novais

2014. Expansion of ESBL-producing Klebsiella pneumoniae in hospitalized patients: a successful story of international clones (ST15, ST147, ST336) and epidemic plasmids (IncR, IncFII_K). Int. J. Med. Microbiol., 304:1100–1108.

28.

Wang

X.R.

, Chen

J.C.

, Kang

, Jiang

, An

S.C.

, and Gao

Z.C.

2012. Prevalence and characterization of plasmid-mediated bla_ESBL with their genetic environment in Escherichia coli and Klebsiella pneumoniae in patients with pneumonia. Chin. Med. J. (Engl.), 125:894–900.

29.

Carattoli

2011. Plasmids in Gram negatives: molecular typing of resistance plasmids. Int. J. Med. Microbiol., 301:654–658.

30.

Marcade

, Deschamps

, Boyd

, Gautier

, Picard

, Branger

, Denamur

, and Arlet

2009. Replicon typing of plasmids in Escherichia coli producing extended-spectrum β-lactamases. J. Antimicrob. Chemother., 63:67–71.

31.

Novais

, Baquero

, Machado

, Canton

, Peixe

, and Coque

T.M.

2010. International spread and persistence of TEM-24 is caused by the confluence of highly penetrating Enterobacteriaceae clones and an IncA/C₂ plasmid containing Tn1696::Tn1 and IS5075-Tn21. Antimicrob. Agents Chemother., 54:825–834.

32.

Carattoli

, Villa

, Poirel

, Bonnin

R.A.

, and Nordmann

2012. Evolution of IncA/C bla_CMY-2-carrying plasmids by acquisition of the bla_NDM-1 carbapenemase gene. Antimicrob. Agents Chemother., 56:783–786.

33.

Coelho

, Gonzalez-Lopez

J.J.

, Miro

, Alonso-Tarres

, Mirelis

, Larrosa

M.N.

, Bartolome

R.M.

, Andreu

, Navarro

, Johnson

J.R.

, and Prats

2010. Characterisation of the CTX-M-15-encoding gene in Klebsiella pneumoniae strains from the Barcelona metropolitan area: plasmid diversity and chromosomal integration. Int. J. Antimicrob. Agents, 36:73–78.

34.

Novais

, Canton

, Moreira

, Peixe

, Baquero

, and Coque

T.M.

2007. Emergence and dissemination of Enterobacteriaceae isolates producing CTX-M-1-like enzymes in Spain are associated with IncFII (CTX-M-15) and broad-host-range (CTX-M-1, −3, and −32) plasmids. Antimicrob. Agents Chemother., 51:796–799.

35.

Zhuo

, Li

X.Q.

, Zong

Z.Y.

, and Zhong

N.S.

2013. Epidemic plasmid carrying bla_CTX-M-15 in Klebsiella penumoniae in China. PLoS One, 8:e52222.

36.

Diestra

, Juan

, Curiao

, Moya

, Miro

, Oteo

, Coque

T.M.

, Perez-Vazquez

, Campos

, Canton

, Oliver

, and Navarro

2009. Characterization of plasmids encoding bla_ESBL and surrounding genes in Spanish clinical isolates of Escherichia coli and Klebsiella pneumoniae. J. Antimicrob. Chemother., 63:60–66.

37.

Lartigue

M.F.

, Poirel

, and Nordmann

2004. Diversity of genetic environment of bla_CTX-M genes. FEMS Microbiol. Lett., 234:201–207.

38.

W.L.

, Chen

S.C.

, Hung

S.W.

, Chuang

Y.C.

, Chung

J.G.

, Chen

I.C.

, and Wu

L.T.

2006. Genetic association of bla_SHV-5 with transposable elements IS26 and IS5 in Klebsiella pneumoniae from Taiwan. Clin. Microbiol. Infect., 12:806–809.