Increased high-level gentamicin resistance in invasive Enterococcus faecium is associated with aac(6′)Ie-aph(2″)Ia -encoding transferable megaplasmids hosted by major hospital-adapted

R E S EA RCH AR T I C L E

Increased high-level gentamicin resistance in invasiveEnterococcus faecium is associated with

aac(6′)Ie-aph(2″)Ia-encoding transferable megaplasmids hostedby major hospital-adapted lineages

Torill C.S. Rosvoll1, Belinda L. Lindstad1, Tracy M. Lunde2, Kristin Hegstad1,2, Bettina Aasnæs2,Anette M. Hammerum3, Camilla H. Lester3, Gunnar S. Simonsen1,2, Arnfinn Sundsfjord1,2 & TorunnPedersen2

1Research Group for Host-Microbe Interactions, Department of Medical Biology, Faculty of Health Sciences, University of Tromsø, Tromsø,

Norway; 2Department of Microbiology and Infection Control, University Hospital of North Norway, Tromsø, Norway; and 3Statens Serum Institut,

Copenhagen, Denmark

Correspondence: Torunn Pedersen,

Department of Microbiology and Infection

Control, University Hospital of North Norway,

PO Box 56, Breivika, N-9038 Tromsø,

Norway. Tel.: +47 77 65 58 65; fax:

+47 77 62 70 15; e-mail:

[email protected]

Received 6 January 2012; revised 23 April

2012; accepted 24 May 2012.

Final version published online 27 June 2012.

DOI: 10.1111/j.1574-695X.2012.00997.x

Editor: Jacques Schrenzel

Keywords

high-level gentamicin resistance; aac(6′)Ie-aph

(2″)Ia; pLG1 replicon type; megaplasmids;

hospital-adapted lineages; virulence genes.

Abstract

Gentamicin is important in synergistic bactericidal therapy with cell wall agents

for severe enterococcal infections. During 2003–2008, a 10-fold increase in the

prevalence of high-level gentamicin resistance (HLGR), to above 50%, in blood

culture isolates of Enterococcus faecium, was reported by the Norwegian Surveil-

lance System for Antimicrobial Resistance. A representative national collection

of invasive E. faecium isolates (n = 99) from 2008 was examined by a multilevel

approach. Genotyping revealed a polyclonal population dominated by major

hospital-associated lineages (mainly ST203, ST17, ST18, ST202 and ST192).

The presence of aac(6′)-Ie-aph(2″)-Ia, encoding the bi-functional amino-

glycoside-modifying enzyme, was found in 98% of HLGR isolates (56/57).

Furthermore, a significantly higher prevalence of potential virulence genes,

toxin-antitoxin loci as well as pRE25 and pRUM type replicons was demon-

strated in isolates belonging to major hospital-associated lineages compared to

other sequence types. Megaplasmids of pLG1 replicon type (200–330 kb) were

present in 90% of the isolates. Co-hybridization analyses revealed genetic link-

age of aac(6′)-Ie-aph(2″)-Ia to this replicon type. Transfer of HLGR-encoding

plasmids was restricted to E. faecium. In conclusion, the increased prevalence

of HLGR in invasive E. faecium in Norway is associated with hospital-adapted

genetic lineages carrying aac(6′)-Ie-aph(2″)-Ia-encoding transferable megaplas-

mids of the pLG1 replicon type.

Introduction

The worldwide increase in healthcare-acquired enterococ-

cal infections is caused by distinct genetic lineages

adapted to the hospital environment (Leavis et al., 2006;

Willems & van Schaik, 2009; Willems et al., 2011). A sig-

nificant rise in invasive infections caused by Enterococcus

faecium, now approaching the frequency of Enterococcus

faecalis, has been accounted for by the expansion of a

polyclonal genetic subcluster previously known as clonal

complex 17 (Willems et al., 2005, 2011; Leavis et al.,

2006; Top et al., 2007). The subcluster is characterized by

high-level ampicillin resistance, presence of an enterococ-

cal surface protein (Esp) encoding pathogenicity island

and an enrichment of other potential virulence genes

(Willems et al., 2005; Willems & van Schaik, 2009).

Most presumed virulence factors in E. faecium are

involved in adherence to host tissue and biofilm forma-

tion (Hendrickx et al., 2009a; Sillanpaa et al., 2009; Sava

et al., 2010a). For Esp, involvement in initial adherence

and biofilm formation (Heikens et al., 2007), contribu-

tion to heart valve colonization (Heikens et al., 2011),

pathogenesis of urinary tract infection (Leendertse et al.,

2009) and bacteraemia (Sava et al., 2010a, b) in mice

ª 2012 Federation of European Microbiological Societies FEMS Immunol Med Microbiol 66 (2012) 166–176Published by Blackwell Publishing Ltd. All rights reserved

IMM

UN

OLO

GY

& M

EDIC

AL

MIC

ROBI

OLO

GY

have been shown. Serine-glutamate repeat containing pro-

tein A (SgrA) is also a surface adhesion protein implicated

in biofilm formation (Hendrickx et al., 2009a, b). Microbial

surface components recognizing adhesive matrix molecules

(MSCRAMMs) are proteins that interact with specific

extra-cellular matrix proteins and are assumed to be

important in host colonization. Among these, second colla-

gen adhesion of E. faecium (Scm) has been shown to bind

to collagen (types V and I) and fibrinogen (Sillanpaa et al.,

2008) whereas adhesion of collagen from E. faecium

(Acm) binds to collagen (types I and IV) and contri-

butes to the pathogenesis of experimental endocarditis

(Nallapareddy et al., 2003). The presence of scm and acm

genes is not specifically linked to hospital-adapted

strains, although acm appears to be more frequently

expressed in such strains (Nallapareddy et al., 2008).

Several MSCRAMM-encoding genes are enriched in the

hospital-adapted population: E. faecium collagen-binding

protein A (ecbA) (Hendrickx et al., 2009a, b); pilA (encod-

ing pili-like proteins, identified on plasmids) (Hendrickx

et al., 2008; Kim et al., 2010; Panesso et al., 2010); pilB

(encoding pili, implicated in biofilm formation and initial

adherens, contributes to UTI in mouse) (Hendrickx et al.,

2008; Sillanpaa et al., 2010); fms11, fms14 and fms15

(encoding cell wall-anchored E. faecium surface proteins)

(Hendrickx et al., 2007; Sillanpaa et al., 2009).

Enterococci express intrinsic low-level resistance to

aminoglycosides because of impaired uptake (Moellering,

1991). However, aminoglycosides are clinically important

in the treatment of severe enterococcal infections owing

to their synergistic and bactericidal effect in combination

with cell wall synthesis inhibitors such as penicillins and

glycopeptides (Arias et al., 2010). Enterococcus faecium

strains produce a chromosomally encoded aminoglycoside

acetyltransferase, AAC(6′)-Ii, which eliminates bactericidal

synergism between cell wall–active antimicrobials and

most clinically applicable aminoglycosides except strepto-

mycin and gentamicin (Costa et al., 1993). High-level

gentamicin resistance (HLGR) eliminates the synergistic

bactericidal effect and causes a major reduction in efficient

therapeutic options. Enzymatic modification of aminogly-

cosides is by far the major aminoglycoside resistance

mechanism in clinical isolates of both Gram-negative and

Gram-positive bacteria. The bi-functional aminoglycoside-

modifying enzyme (AME) AAC(6′)-Ie-APH(2″)-Ia found

in enterococcal, streptococcal and staphylococcal isolates

renders them high-level resistant to virtually all clinically

available aminoglycoside antibiotics, except streptomycin

and to some extent, arbekacin (Ferretti et al., 1986; Chow,

2000). The bi-functional gene aac(6′)-Ie-aph(2″)-Ia is

linked to Tn5281 (also known as Tn4001) and often plasmid

located, permitting cell-to-cell dissemination (Hodel-Christian

&Murray, 1991; Simjee et al., 1999; Watanabe et al., 2009).

Plasmids are widespread in enterococcal populations

and represent an important reservoir for accessory genes.

In murine models, megaplasmids encoding hylefm have

been shown to increase E. faecium colonization of the

gastrointestinal tract (Rice et al., 2009) and to enhance

the virulence in experimental peritonitis (Arias et al.,

2009). However, a recent study showed that the hylefmgene itself was not the main mediator of this enhanced

virulence (Panesso et al., 2011). Megaplasmids (with or

without hylefm) have recently been described worldwide in

clinical E. faecium isolates (Freitas et al., 2010, 2011;

Panesso et al., 2010; Laverde Gomez et al., 2011). The

pLG1 megaplasmid (280 kb) belonging to the RepA_N

family of replicons, carrying multiple resistance and puta-

tive virulence determinants, has been sequenced (Laverde

Gomez et al., 2011).

A high prevalence of HLGR has been observed in

European, Asian and South American countries (Lester

et al., 2008; Sangsuk et al., 2009; Dorobat et al., 2010;

Panesso et al., 2010; Araoka et al., 2011). During 2003–2008, a 10-fold increase in the prevalence of HLGR, to

above 50%, in blood culture isolates of E. faecium, was

reported by the Norwegian Surveillance System for Anti-

microbial Resistance (NORM) (NORM/NORM-VET,

2008), which seems to be part of an international trend.

Here, we present a detailed epidemiological study of a

representative national collection of invasive E. faecium

isolates during 2008, targeting molecular mechanisms

involved in the increased prevalence of HLGR.

Materials and methods

Bacterial isolates and antibiotic susceptibility

testing

A total of 166 E. faecium isolates were identified in

blood cultures in Norway during 2008 as part of the

NORM surveillance programme (http://www.unn.no/norm/

category8926.html). The isolates were confirmed to the

species level by molecular identification (Dutka-Malen

et al., 1995) and 99 isolates were randomly selected to be

included in the present study. The isolates were collected

from 20 hospitals (geographically spread and representing

the majority of Norwegian hospitals). Resistance to ampi-

cillin (84%), gentamicin (HLGR; 57%), streptomycin

(82%), linezolid (0%) and vancomycin (0%) was deter-

mined for all isolates as previously described (NORM/

NORM-VET, 2008).

PCR analyses and DNA sequencing

Total DNA was extracted using the BioRobot M48

work station (Qiagen, Oslo, Norway) together with the

FEMS Immunol Med Microbiol 66 (2012) 166–176 ª 2012 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

HLGR-encoding megaplasmids in invasive E. faecium 167

MagAttract® DNA Mini M48 kit (Qiagen). DNA extracts

positive for 16S rRNA gene PCR (modified primers from

(Weisburg et al., 1991)) were screened for the presence of

selected gene loci by specific PCRs (Table 1). The rep PCRs

targeted replication initiation genes of defined plasmid

groups previously detected in enterococci (Freitas et al.,

2010, Jensen et al., 2010; Rosvoll et al., 2010) in addition to

pLG1 (reppLG1) (Laverde Gomez et al., 2011). PCR primers,

control strains and references are given in Supporting

Information, Table S1. PCRs were carried out as simplex

reactions and at least five of each amplicon were confirmed

by DNA sequencing as described (Rosvoll et al., 2010).

Genotyping by multilocus sequence typing and

pulsed-field gel electrophoresis

Multilocus sequence typing (MLST) of E. faecium isolates

was performed as previously described (Homan et al.,

2002). PCR products were sequenced by Macrogen (Seoul,

Korea). The BioNumerics Software (Applied Maths Ver-

sions 6.01/6.5; Sint-Martens-Latem, Belgium) was used for

all DNA sequence analyses. New sequence types (STs) were

deposited in the database (http://efaecium.mlst.net). A min-

imum spanning tree was constructed based on categorical

coefficient of similarity and the priority rule that the highest

number of single-locus variants would be linked first.

Pulsed-field gel electrophoresis (PFGE) typing was per-

formed on whole-cell DNA, prepared and analysed as

described with pulse time 3.5–25 s for 12 h and 1–5 s for

8 h (Turabelidze et al., 2000). Similarity analysis of PFGE

patterns was performed with the Dice coefficient (tolerance

1%, optimization 0.5%) and clustering performed by

UPGMA. Isolates having a similarity of more than 84% were

considered to be related and were assigned to the same

cluster (Morrison et al., 1999).

Macrorestriction and co-hybridization analysis

Plasmid DNA content was examined by S1 nuclease

assays as described with pulse time 5–35 s for 17 h (Ros-

voll et al., 2010). Chromosomal localization of selected

genes was assessed by I-CeuI macrorestriction followed by

16S rDNA co-hybridization analysis. Agarose slices were

incubated for 4 h at 37 °C with 10 U I-CeuI endonucle-

ase in 100 lL of the supplied buffer (New England Biol-

abs Inc.) and DNA was separated by PFGE as described

using pulse time 60–90 s for 22 h. DNA was blotted onto

positively charged nylon membranes (Roche Applied Sci-

ence, Penzberg, Germany). Digoxigenin-labelled DNA

probes for 16S rDNA, aac(6′)-Ie-aph(2′)-Ia, hylEfm, axe-txe,x-e-ζ, reppRE25, reppRUM and reppLG1 were generated, and

high-stringency hybridization and detection performed as

described (Rosvoll et al., 2010). The analyses were repeated

several times employing the probes in various orders.

Primers and template strains are listed in Table S1.

Transfer of HLGR determinants

Transfer studies were performed by filter mating as

described with 24-h incubation on selective agar (Werner

et al., 2011). Ten HLGR isolates were selected as donors

and E. faecium BM4105-RF and 64/3 (for three donors)

as well as E. faecalis OG1RF and JH2-2 as recipients.

Transconjugants (TCs) were selected on BHI agar (Difco

Labs., Detroit) containing 150 lg mL�1 gentamicin and

25 lg mL�1 rifampicin and 25 lg mL�1 fusidic acid. The

experiments were repeated at least three times for each

donor. The presence of aac(6′)-Ie-aph(2″)-Ia and reppLG1was confirmed by PCR and Southern hybridization.

Transfer was verified by SmaI analyses. Retransfer was

conducted for TCs (resulting from six different donors)

using BM4105-Str as recipient and selected on BHI

(150 lg mL�1 gentamicin and 1000 lg mL�1 streptomy-

cin) and confirmed as described above.

Statistical analyses

Differences among groups were analysed by Fisher’s exact

test, or unpaired Student’s t-test. P < 0.05 was considered

statistically significant.

Results

Genotypic characterization of invasive

E. faecium isolates by MLST and PFGE

MLST and PFGE revealed a polyclonal population. In total,

26 STs and 37 PFGE clusters were detected. The most pre-

valent STs were ST203 (n = 28), ST17 (n = 18), ST18

(n = 10), ST202 (n = 8) and ST192 (n = 7) (Table 1). A

total of 15 STs contained only one isolate. Eight new STs

(574–581) were identified. The different STs were dispersedamong the 20 hospitals. In each hospital, one to six STs

were found. When more than one isolate were detected in

the same hospital, they represented at least two STs. ST203

was found in 12 different hospitals with a wide geographi-

cal distribution. PFGE typing resolved the isolates into 13

clusters with more than one isolate and 24 unique isolates

as shown in Table 1 and Fig. S1. All PFGE clusters were

found in more than one hospital. The dominant PFGE

cluster 6 (n = 23) was found in 11 different hospitals and

included mainly isolates of ST203. However, two single-

locus variants (ST577 and ST78) and one double-locus

variant (ST17) of ST203 also grouped into PFGE cluster 6.

The presence of two isolates with different STs was also

evident in PFGE cluster 7 and 11. Interestingly, the

ª 2012 Federation of European Microbiological Societies FEMS Immunol Med Microbiol 66 (2012) 166–176Published by Blackwell Publishing Ltd. All rights reserved

168 T.C.S. Rosvoll et al.

diversity of PFGE clusters differed among the STs; for

ST203 (n = 28), four different clusters or unique types

were found, while ST18 (n = 10) included eight different

clusters or unique types.

Presence and distribution of HLGR

The distribution of HLGR among STs is illustrated by the

minimum spanning tree as shown in Fig. 1. HLGR was

detected in isolates belonging to 11 different STs. All iso-

lates were divided into three groups as shown in Table 1:

STs that include both HLGR-positive and HLGR-negative

isolates (n = 76), STs including only HLGR isolates

(n = 5) and STs including only non-HLGR isolates

(n = 18). Most STs included in the mixed group (ST17,

ST18, ST78, ST192, ST202 and ST203) are known as

major hospital-associated genetic lineages. PCR screening

for AME-encoding genetic determinants revealed that all

Table 1. Genotypes and genetic characteristics of different STs of Enterococcus faecium

MLST PFGEHLGR isolates Non-HLGR isolates

ST No.* Hospitals† cluster No. Hospitals No. Vir‡ Rep§ TA¶ pLG1** No. vir rep TA pLG1

STs including HLGR and non-HLGR isolates

ST203 28 12 7 1 1 20 9.8 3.9 2.0 1.0 8 9.9 3.8 1.5 1.0

6 18 10

5 8 6

U†† 1 1

ST17 18 8 12 2 1 15 8.6 3.7 1.9 1.0 3 7.3 3.3 1.3 1.0

6 3 2

4 12 7

U 1 1

ST18 10 5 13 2 1 1 8.0 4.0 2.0 1.0 9 7.6 2.2 0.3 1.0

3 2 2

U 6 4

ST202 8 5 10 8 5 7 7.7 3.3 0.3 1.0 1 7.0 2.0 1.0 0

ST192 7 5 8 7 5 6 8.7 4.2 2.0 1.0 1 8.0 3.0 2.0 0

ST78 3 3 9 2 2 2 10.0 3.5 2.0 1.0 1 10.0 3.0 2.0 1.0

6 1 1

ST132 2 1 1 2 1 1 9.0 3.0 1.0 1.0 1 6.0 3.0 2.0 1.0

Number of isolates 76 52 24

Average per isolate 9.0 3.8 1.7 1.0 8.3 3.0 1.1 0.9

STs including only HLGR isolates

ST440 2 2 2 2 2 2 6.5 3.5 0.5 1.0 0 - - - -

STs 279, 575, 577 3 3 U 3 3 3 7.7 3.7 1.3 1.0 0 - - - -

Number of isolates 5 5 0 - - - -

Average per isolate 7.2 3.6 1.0 1.0

STs including only non-HLGR isolates

ST22 2 2 11 1 1 0 - - - 2 5.0 0.0 0.0 0.0

U 1 1

ST32 2 2 11 1 1 0 - - - 2 4.0 0.5 0.5 0.5

U 1 1

ST94 2 2 U 2 2 0 - - - 2 2.5 2.5 0.0 1.0

STs 19, 38, 52, 282, 296,

533, 574, 576, 578–581

12 10 U 12 12 0 - - - 12 5.3 2.3 0.8 0.7

Number of isolates 18 0 - - - 18

Average per isolate 4.8 1.9 0.6 0.6

Total numbers 99 57 8.8 4.4 1.6 1.0 42 6.8 2.5 0.9 0.8

*Number of isolates.†Number of different hospitals (from a total of 20) where the specific ST/PFGE cluster was found.‡Average number of putative virulence-encoding genes.§Average number of plasmid replication initiation genes.¶Average number of Toxin-Antitoxin loci.

**Average number replication initiation gene of pLG1 type.††A unique PFGE cluster, with similarity below 84% compared to any other PFGE clusters in the collection.

FEMS Immunol Med Microbiol 66 (2012) 166–176 ª 2012 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

HLGR-encoding megaplasmids in invasive E. faecium 169

HLGR isolates were positive for the aac(6′)-Ie-aph(2″)-Ia,except for one aph(2″)-Ib-positive isolate of ST132. One

isolate (ST192) characterized as wild-type gentamicin sus-

ceptible was positive for aac(6′)-Ie-aph(2″)-Ia. Owing to

its HLGR genotype and genetic resemblance to the other

HLGR isolates, it was included as a HLGR positive

(n = 57). The presence of aac(6′)-Ie-aph(2″)-Ia was con-

firmed by Southern hybridization for all PCR-positive

isolates.

Presence and distribution of putative

virulence-encoding genes, plasmid replicon

groups and TA-loci

The isolates were screened for the presence of 11 genetic

determinants implicated in virulence of E. faecium. The

occurrence of individual virulence genes was high (fms15:

93%; acm: 90%; pilA: 88%; srgA: 84%; scm: 83%; pilB:

82%; esp: 77%; fms11: 72%; fms14: 71%) except for hylEfm(13%) and ecbA (37%). Interestingly, pilA and fms15 were

present in all HLGR isolates. Moreover, all ST192 isolates

lacked the fms11 gene while all ST202 isolates lacked the

fms14 gene. There was an average of eight virulence genes

per isolate. In ST203 isolates, the average of virulence

genes was 9.8 compared to 7.3 in the non-ST203 isolates,

P < 0.0001. The ecbA gene was present in 100% of the

ST203 isolates compared to only 12.7% in the remaining

isolates.

The presence of 12 specific plasmid groups was investi-

gated by PCRs. The most prevalent were reppLG1 (90%),

reppRE25 (73%), reppRUM (66%) and reppRI1 (62%). Less

frequently, reppEF1071 (11%), reppEF18 (7%), reppIP501 (5%)

and reppMG1 (5%) were detected, while reppMBB1, reppS86,

reppAM373 and reppCF10 were not observed. There was an

average of 3.2 rep genes per isolate. A total of 95 isolates

were PCR positive for one to five rep genes. The pIP501

replicon type was only detected in the non-HLGR group

(5/18). Notably, most of the reppLG1-negative isolates (7/9)

were ampicillin and gentamicin susceptible as well as esp

negative and belonged to the non-HLGR group.

DNA sequencing of amplicons (n = 18) from the

pRUM replicon revealed that most of them (17/18) con-

stituted a new subtype with distinct base pair changes

compared to the published replication initiation gene of

pRUM (GenBank: AF507977.1). The new subtype has

been deposited in GenBank (GenBank: JX177613).

The plasmid content of each isolate was further analy-

sed by S1 nuclease assays as shown for selected isolates in

Fig. 2a. The 99 isolates harboured none to six plasmids

ranging in size from < 10 kb to > 400 kb. The plasmid

population could be divided into two large subgroups.

Plasmids below 100 kb constituted approximately 60% of

the total estimated number of plasmids. The second

group comprised megaplasmids from 200 kb to 330 kb.

Only a few plasmid bands were between 100 kb and

200 kb or > 400 kb. The total number of plasmids

accounted for by the S1 nuclease method was 311,

whereas 319 rep-amplicons were detected by PCR.

Toxin-antitoxin (TA) systems are believed to stabilize

and promote plasmid persistence. Screening by PCR for

axe-txe and x-e-ζ TA-encoding loci revealed a prevalence

of 66% and 65%, respectively. Taken together, 76% of

the isolates had one or both TA-loci.

Comparison of isolates when assembled into ST groups

(shown in Table 1) revealed that the mixed group had a

higher percentage of individual putative virulence genes

(significantly higher for all genes except acm, hyl and

fms14), TA-loci (axe-txe P < 0.0001 and x-e-ζ P < 0.0048)

and rep genes (reppRE25 P < 0.0001, reppRUM P < 0.0001)

compared with the non-HLGR group. Noteworthy, esp was

present in 98% of the isolates in the mixed group com-

pared to only 17% of the isolates in the non-HLGR

group (P < 0.0001). A comparison of HLGR-positive and

HLGR-negative isolates within the mixed group showed a

significant difference in the average number of viru-

lence genes as well as rep genes and TA-loci: 9.0 vs.

8.3 (P < 0.032); 3.8 vs. 3.0 (P < 0.0001); 1.7 vs. 1.1

(P < 0.0005), respectively. For individual genes, the pres-

78

17203

192

578

202

577279 282

18

132

574

440

575

32

22

533

19 581

52

38

94

579

296

576

580

Fig. 1. Minimum spanning tree for distribution of HLGR in different

Enterococcus faecium STs. Red indicates high-level resistance to

gentamicin and yellow wild-type gentamicin susceptibility. Every circle

represents an ST and the size corresponds to the number of isolates.

Short thick lines connect single-locus variants; thin lines connect

double-locus variants and broken lines triple-locus variants.

ª 2012 Federation of European Microbiological Societies FEMS Immunol Med Microbiol 66 (2012) 166–176Published by Blackwell Publishing Ltd. All rights reserved

170 T.C.S. Rosvoll et al.

ence of esp (P < 0.032) and reppRE25 (P < 0.0001) was sig-

nificantly higher in the HLGR-positive isolates. In addition,

81% of the HLGR-positive isolates harboured both TA-loci

compared to only 38% of the HLGR-negative isolates,

P < 0.0009.

Characterization of HLGR-encoding plasmids

Plasmid co-localization between HLGR-encoding genes,

TA-loci and replicon types was examined by sequential

hybridization analyses of S1 digested total DNA separated

by PFGE. Probes specific for aac(6′)-Ia-aph(2″)-Ie,axe-txe, x-e-ζ and the most common reps in the HLGR

isolates (reppRE25, reppRUM and reppLG1) were used. All

isolates (n = 99) were examined using reppRE25, reppRUMand TA probes (data not shown), while only the HLGR

isolates (n = 57) were included in aac(6′)-Ia-aph(2″)-Ieand reppLG1 hybridization (data shown for 13 representa-

tive isolates in Fig. 2b and c). The total number of iso-

lates with positive hybridization signals for the individual

probes corresponded to the PCR results and the positive

and negative control strains scored as expected. One

dominating hybridizing plasmid band was detected for

each probe for most of the isolates. However, for six iso-

lates, plasmid linkage could not be verified as the hybrid-

ization signals repeatedly corresponded to the location of

the agarose wells. A possible chromosomal localization

was explored for these isolates by I-CeuI macrorestriction

analyses. For one single isolate (ST203), co-hybridization

between the aac(6′)-Ia-aph(2″)-Ie and the 16S rDNA

probes was detected.

A total of 68 isolates had reppRE25 hybridizing plasmids,

ranging between 20 and 120 kb with the majority below

50 kb. Co-hybridization between reppRE25 and the x-e-ζprobe was observed in 75% of these isolates. In three

isolates, co-hybridization between reppRE25, axe-txe and

reppRUM probes was detected. Co-hybridization between

reppRE25 and aac(6′)-Ia-aph(2″)-Ia was not observed. The

reppRUM probe hybridized to plasmid bands, predominantly

between 70 and 100 kb and not above 120 kb and was

detected in 65 of the isolates. Co-hybridization between

reppRUM and axe-txe was only present in 38% of these.

Interestingly, the majority of the remaining axe-txe-positive

plasmids belonged to megaplasmids above 220 kb and

co-hybridized with the reppLG1 probe. Co-hybridization

between reppRUM and aac(6′)-Ie-aph(2″)-Ia was not

observed.

All HLGR isolates were PCR positive for reppLG1.

Hybridization analyses revealed reppLG1-positive plasmids

ranging from 200 kb to > 400 kb in size. Importantly, for

all aac(6′)-Ie-aph(2″)-Ia-positive plasmids, co-hybridiza-

tion to reppLG1 was detected as illustrated in Fig. 2. For

HLGR isolates of ST192 (lane 9) and ST202 (lanes 3 and 6),

M C1 C2 1 2 3 4 5 6 7 8 9 10 11 12 13M

339.5

Kb

291.0242.5194.0

145.5

97.0

48.523.1

9.4

6.6

MM C1 C2 1 2 3 4 5 6 7 8 9 10 11 12 13

339.5

Kb

291.0242.5194.0

145.5

97.0

48.523.1

9.4

6.6

MM C1 C2 1 2 3 4 5 6 7 8 9 10 11 12 13

339.5

Kb

291.0242.5194.0

145.5

97.0

48.523.1

9.4

6.6

(a)

(b)

(c)

Fig. 2. Co-hybridization analyses of HLGR encoding plasmids. S1

nuclease-digested total DNA of HLGR isolates, representing STs 17,

78, 192, 202, 203, 279; 404 and 575 (lanes 1–13) and control strains

(lane C1: positive control for aac(6′)-Ia-aph(2′’)-Ie, Enterococcus

faecium K60-39; lane C2: positive control for reppLGI, E. faecium

TX0016) were subjected to PFGE followed by Southern blotting. The

plasmid content of each isolate was visualized by ethidium bromide

staining of the resulting gel (a) and hybridization to aac(6′)-Ia-aph(2′’)-

Ie (b) and reppLG1 (c) specific probes shown in the corresponding

autoradiographs. The molecular sizes of the PFG marker (lane M) are

shown in the left panel.

FEMS Immunol Med Microbiol 66 (2012) 166–176 ª 2012 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

HLGR-encoding megaplasmids in invasive E. faecium 171

these plasmids were found to be approximately 200 kb

(6/6) and 240 kb (7/7), respectively, while plasmids of

about 300 kb were detected in HLGR isolates of ST17

(12/15; lanes 5, 8 and 10) and ST203 (8/20; lanes 1 and 2).

Furthermore, 57% of reppLG1 hybridizing plasmids

co-hybridized with axe-txe.

The location of hylEfm was investigated by hybridization

of the hylEfm PCR-positive isolates (n = 13). The gene was

detected in plasmids ranging from 70 to 280 kb, pre-

dominantly megaplasmids (data not shown). Moreover,

co-hybridization to the hylEfm probe was detected for the

200-kb plasmids found in HLGR-positive isolates of ST192

(5/6). One hylEfm-positive plasmid (70 kb) co-hybridized

with reppRUM.

Conjugative transfer of HLGR-encoding

plasmids

In vitro conjugative transfer of HLGR was examined by

filter mating. The donors carried the aac(6′)-Ie-aph(2′)-Iaon the chromosome (n = 1; ST203) or on plasmids

(n = 9; of ST17, ST78, ST192, ST202, ST203, ST440 or

ST575) of various sizes (200–300 kb). All donors were

able to transfer the HLGR determinant into E. faecium

(BM4105-RF) with reproducible transfer frequencies

ranging from 4 9 10�2 to 6 9 10�7 TC per recipient cell.

Similar transfer frequencies were observed when using

E. faecium 64/3 as a recipient. Transfer to E. faecalis

OG1-RF or JH2-2 was not observed.

The conjugative properties of the HLGR-encoding plas-

mids were further examined using selected TCs as donors

and E. faecium BM4105-Str as a recipient in a second

filter mating experiment. Re-transfer of the aac(6′)-Ie-aph(2′)-Ia-encoding plasmids was observed from all

BM4105-RF (n = 3) and 64/3 (n = 3) TCs tested. The

re-transfer frequencies were comparable to the first conju-

gative transfer for corresponding plasmids.

Discussion

We have performed a detailed epidemiological study of a

representative national collection of invasive E. faecium

isolates from 2008, targeting molecular mechanisms

involved in the 10-fold increased prevalence of HLGR

E. faecium since 2003. The increased level of HLGR

E. faecium seems to be part of a general trend in Europe,

although some differences between countries are observed

through the EARS-net (http://ecdc.europa.eu/en/activities/

surveillance/EARSNet/database/Pages/map_reports.aspx).

The relative increase in hospital-adapted E. faecium

infections has been associated with a polyclonal subcluster

with defined lineages (Willems et al., 2005, 2011). Our

results are in line with these observations. The dominant

STs found in this study (ST203, ST202, ST192, ST78,

ST18 and ST17) are considered major hospital-associated

lineages and represent 75% of the isolates. The HLGR

phenotype was mostly associated with these clones. The

most prevalent ST (ST203) was recently described as a

successful clone outcompeting ST17 and causing a sus-

tained hospital outbreak of VRE in an Australian health

service unit (Johnson et al., 2010). Surprisingly, ST18 dif-

fered from the other hospital-associated STs with a very

low prevalence of HLGR isolates, together with a high

diversity of different PFGE clusters. In total, the genotypic

characterization revealed genetic heterogeneity among the

invasive E. faecium isolates. Although the dominant

ST203 and the major PFGE cluster 6 was observed in 12

and 11 hospitals, respectively, the wide dispersion of STs

and PFGE clusters does not support any major national

clonal outbreak that could account for the rapid annual

increase in HLGR.

Rather we hypothesized that transferable plasmids were

involved in the dissemination of HLGR. The aac(6′)-Ie-aph(2′′)-Ia gene is associated with Tn5281 which is not

known to be conjugative (Hegstad et al., 2010). However,

the transposon has previously been reported on plasmids

below 100 kb (Simjee et al., 1999, 2000; Abbassi et al.,

2007). In this study, we detected the aac(6′)-Ie-aph(2′′)-Iagene on transferable megaplasmids of variable size (200–330 kb). This size variation could be explained by the

movement of the transposon itself to different plasmids,

or by genetic rearrangements of the plasmids harbouring

the transposon. Our data indicate the latter, as all

plasmids belong to the same replicon type and transfer of

the HLGR determinants always results in transfer of the

aac(6′)-Ie-aph(2′′)-Ia-carrying megaplasmid. Moreover, we

have observed changes in the plasmid size in the resulting

TCs after conjugative transfer indicating plasmid rear-

rangements. Some STs appeared to have HLGR-encoding

plasmids of distinct sizes which may reflect clonal dissem-

ination of STs harbouring these plasmids.

A high prevalence of plasmids of pRE25, pRUM and

pLG1 replicon types was detected. We observed a strong

correlation between the number of plasmids found by the

rep-PCR-based detection system (Jensen et al., 2010) and

the S1 nuclease-based PFGE separation of linearized plas-

mids. Based on the assumption of one rep gene per plas-

mid, 98% of the total plasmid content was detected with

the rep PCRs used in this study. This is in contrast to our

previous study, which did not include the pLG1 replicon

type, otherwise the prevalence of the most frequent

replicon types (reppRE25 and reppRUM) was comparable

(Rosvoll et al., 2010). We expect neither of the two

employed methods to give exact information about the

plasmid number in the strain collection. However, the

observed overall agreement in plasmid content between

ª 2012 Federation of European Microbiological Societies FEMS Immunol Med Microbiol 66 (2012) 166–176Published by Blackwell Publishing Ltd. All rights reserved

172 T.C.S. Rosvoll et al.

both methods assuming one rep gene per plasmid as a

main rule indicates that the dominant rep types in the

main contemporary hospital-adapted E. faecium lineages

are identified.

In general, a high prevalence of putative virulence

genes was found in Norwegian invasive E. faecium iso-

lates. The significantly higher prevalence of these genes

in major hospital-adapted STs compared to other STs is

consistent with the dominance of hospital-adapted lin-

eages with enhanced colonization and biofilm formation

abilities (Hendrickx et al., 2007; Nallapareddy et al.,

2008; Sillanpaa et al., 2009). We screened the isolates for

the presence of 11 known or potential virulence deter-

minants and observed a higher number in ST203 com-

pared to non-ST203 and a 100% linkage to ecbA,

encoding a novel E. faecium MSCRAMM that binds to

collagen type V and fibrinogen (Hendrickx et al.,

2009a, b), in ST203.

PCR-based detection of genes encoding AMEs revealed

a near 100% association between the expression of a

HLGR phenotype and the presence of aac(6′)-Ie-aph(2″)-Ia.A single HLGR isolate contained the aph(2″)-Ib. This

HLGR determinant has previously been sporadically

observed in E. faecium (Kao et al., 2000; Hammerum et al.,

2012). Co-hybridization analyses disclosed a near 100%

linkage of the aac(6′)-Ie-aph(2″)-Ia gene to megaplasmids

of the reppLG1 type. pLG1 is a newly sequenced 280-kb con-

jugative plasmid encoding VanA-type glycopeptide resis-

tance, macrolide resistance, carbon uptake-utilization

genes and putative virulence genes including hylefm and a

pilin gene cluster (Laverde Gomez et al., 2011). pLG1

belongs to the RepA_N family that comprises other entero-

coccal plasmids like pRUM and pAD1. A highly conserved

domain in the N-terminal end defines the family, and a C-

terminal domain is conserved within in each genus, but dif-

fers between genera. The central region of the protein

shows great heterogeneity and this sequence variability is

believed to provide compatibility for plasmids maintained

in the same host (Weaver et al., 2009). Curiously, in our

study, the axe-txe TA-locus was solely linked to plasmids of

pRUM or pLG1 replicon types both belonging to the

RepA_N family. High to medium transfer rates of HLGR-

encoding plasmids of pLG1 replicon type were observed for

all donor strains, but restricted to E. faecium. These obser-

vations are compatible with previous reports on the origi-

nal pLG1 plasmid as well as the narrow host range profile

of RepA_N family plasmids (Weaver et al., 2009; Laverde

Gomez et al., 2011).

Recent studies have shown that megaplasmids both

enhance colonization and virulence in E. faecium (Arias

et al., 2009; Rice et al., 2009). The surprisingly high prev-

alence of reppLG1-type plasmids observed in our collection

of invasive isolates supports the hypothesis that large

plasmids, encoding antimicrobial resistance and pathoge-

nicity factors, contribute to the success of the hospital-

associated subpopulation of E. faecium (Freitas et al.,

2010; Laverde Gomez et al., 2011). Moreover, most of the

reppLG1-negative isolates belong to the non-HLGR group,

which do not contain known hospital-adapted STs. The

success of hospital-adapted lineages has been contributed

to their capacity to acquire adaptive mechanisms provid-

ing selective advantages (Leavis et al., 2006; van Schaik

et al., 2010). The rapid increase in HLGR and the signifi-

cantly higher number of plasmids and virulence genes in

HLGR-positive vs. HLGR-negative isolates among the

hospital-associated STs suggest that HLGR-containing

isolates have an even better capacity to acquire and stabi-

lize accessory DNA that make them yet more successful

in the hospital environment a concept known as genetic

capitalism.

In summary, we have provided epidemiological and

molecular evidence for transferable aac(6′)-Ie-aph(2″)-Ialinked to reppLG1-type megaplasmids as a mechanism for

the significant increase of HLGR in Norwegian inva-

sive E. faecium. The efficient conjugative properties of

reppLG1-type megaplasmids within different E. faecium

populations are consistent with the observed polyclonal

population of HLGR E. faecium strains in Norway as well

as the observed increase of HLGR in invasive E. faecium

strains in several European countries. Moreover, these

megaplasmids appear to be frequently present in the

hospital-adapted population of E. faecium and may con-

tribute both to resistance and pathogenicity.

Acknowledgements

Karin Sixhøj Pedersen, Stefan S. Olsen and Frank Hansen

at SSI are thanked for excellent technical assistance. The

study was performed in collaboration with the diagnostic

laboratories in Norway forming the Norwegian Study

Group on Invasive Enterococci. Representatives of this

study group included Gorm Hansen (Oslo University

Hospital, Aker), Merete R. Ueland (Oslo University Hos-

pital, Radiumhospitalet), Reidar Hjetland (Førde Hospi-

tal), Haima Mylvaganam and Dag H. Skutlaberg

(Haukeland University Hospital, Bergen), Pirrko-Liisa

Kellokumpu (Haugesund Hospital), Einar Vik (Molde

Hospital), Hege E. Larsen (Nordland Hospital, Bodø),

Truls M. Leegaard (Oslo University Hospital, Rikshospita-

let), Jan E. Afset (St Olav University Hospital, Trondheim),

Paul Naaber (Stavanger University Hospital), Annette

Onken (Bærum Hospital, Bærum), Ellen Grimstad

(Drammen Hospital, Drammen), Viggo Hasseltvedt (Inn-

landet Hospital, Lillehammer), Dagfinn Skaare (Vestfold

Hospital, Tønsberg), Eivind Ragnhildstveit (Østfold Hospi-

tal, Fredrikstad), Stale Tofteland (Sørlandet Hospital,

FEMS Immunol Med Microbiol 66 (2012) 166–176 ª 2012 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

HLGR-encoding megaplasmids in invasive E. faecium 173

Kristiansand), Anne R. Oseid and Nils Grude (Unilabs Tel-

elab A/S, Skien), Gaute Syversen (Oslo University Hospital,

Ulleval) and Siv-Heidi Barkhald (University Hospital of

North Norway, Tromsø). This work was supported by

research grants from the Norwegian Research Council

(NFR, project no: 183653/S10, 2008-11), the Northern

Norway Regional Health Authority Medical Research Pro-

gram, the Danish Ministry of the Interior and Health as

part of the Danish Integrated Antimicrobial Resistance and

Research Programme (DANMAP) and the Scandinavian

Society for Antimicrobial Chemotherapy.

Conflicts of interest

None to declare.

References

Abbassi MS, Achour W & Hassen AB (2007) High-level

gentamicin-resistant Enterococcus faecium strains isolated

from bone marrow transplant patients: accumulation of

antibiotic resistance genes, large plasmids and clonal strain

dissemination. Int J Antimicrob Agents 29: 658–664.Araoka H, Kimura M & Yoneyama A (2011) A surveillance of

high-level gentamicin-resistant enterococcal bacteremia.

J Infect Chemother 17: 433–434.Arias CA, Panesso D, Singh KV, Rice LB & Murray BE (2009)

Cotransfer of antibiotic resistance genes and a hylEfm-

containing virulence plasmid in Enterococcus faecium.

Antimicrob Agents Chemother 53: 4240–4246.Arias CA, Contreras GA & Murray BE (2010) Management of

multidrug-resistant enterococcal infections. Clin Microbiol

Infect 16: 555–562.Chow JW (2000) Aminoglycoside resistance in enterococci.

Clin Infect Dis 31: 586–589.Costa Y, Galimand M, Leclercq R, Duval J & Courvalin P

(1993) Characterization of the chromosomal aac(6′)-Ii genespecific for Enterococcus faecium. Antimicrob Agents

Chemother 37: 1896–1903.Dorobat OM, Badicut I, Talapan D, Tenea C & Rafila A

(2010) Antibiotic resistance of Gram-positive cocci

isolated in 2008. Bacteriol Virusol Parazitol Epidemiol 55:

83–92.Dutka-Malen S, Evers S & Courvalin P (1995) Detection of

glycopeptide resistance genotypes and identification to the

species level of clinically relevant enterococci by PCR. J Clin

Microbiol 33: 24–27.Ferretti JJ, Gilmore KS & Courvalin P (1986) Nucleotide

sequence analysis of the gene specifying the bifunctional

6′-aminoglycoside acetyltransferase 2″-aminoglycoside

phosphotransferase enzyme in Streptococcus faecalis and

identification and cloning of gene regions specifying the two

activities. J Bacteriol 167: 631–638.Freitas AR, Tedim AP, Novais C, Ruiz-Garbajosa P, Werner G,

Laverde-Gomez JA, Canton R, Peixe L, Baquero F & Coque

TM (2010) Global spread of the hyl(Efm) colonization-

virulence gene in megaplasmids of the Enterococcus faecium

CC17 polyclonal subcluster. Antimicrob Agents Chemother

54: 2660–2665.Freitas AR, Coque TM, Novais C et al. (2011) Human and

swine hosts share vancomycin-resistant Enterococcus faecium

CC17 and CC5 and Enterococcus faecalis CC2 clonal clusters

harboring Tn1546 on indistinguishable plasmids. J Clin

Microbiol 49: 925–931.Hammerum AM, Lester CH, Olsen SS, Holm A, Hansen DS &

Schonheyder HC (2012) Molecular characterisation of high-

level gentamicin-resistant enterococci from bloodstream

infections in Denmark: first description of clonal spread of

aph(2″)-Ib. Int J Antimicrob Agents 39: 266–268.Hegstad K, Mikalsen T, Coque TM, Werner G & Sundsfjord A

(2010) Mobile genetic elements and their contribution to

the emergence of antimicrobial resistant Enterococcus faecalis

and Enterococcus faecium. Clin Microbiol Infect 16: 541–554.Heikens E, Bonten MJ & Willems RJ (2007) Enterococcal

surface protein Esp is important for biofilm formation of

Enterococcus faecium E1162. J Bacteriol 189: 8233–8240.Heikens E, Singh KV, Jacques-Palaz KD, van Luit-Asbroek M,

Oostdijk EA, Bonten MJ, Murray BE & Willems RJ (2011)

Contribution of the enterococcal surface protein Esp to

pathogenesis of Enterococcus faecium endocarditis. Microbes

Infect 13: 1185–1190.Hendrickx AP, van Wamel WJ, Posthuma G, Bonten MJ &

Willems RJ (2007) Five genes encoding surface-exposed

LPXTG proteins are enriched in hospital-adapted

Enterococcus faecium clonal complex 17 isolates. J Bacteriol

189: 8321–8332.Hendrickx AP, Bonten MJ, van Luit-Asbroek M, Schapendonk

CM, Kragten AH & Willems RJ (2008) Expression of two

distinct types of pili by a hospital-acquired Enterococcus

faecium isolate. Microbiology 154: 3212–3223.Hendrickx AP, Willems RJ, Bonten MJ & van Schaik W

(2009a) LPxTG surface proteins of enterococci. Trends

Microbiol 17: 423–430.Hendrickx AP, van Luit-Asbroek M, Schapendonk CM et al.

(2009b) SgrA, a nidogen-binding LPXTG surface adhesin

implicated in biofilm formation, and EcbA, a collagen

binding MSCRAMM, are two novel adhesins of hospital-

acquired Enterococcus faecium. Infect Immun 77: 5097–5106.Hodel-Christian SL & Murray BE (1991) Characterization of

the gentamicin resistance transposon Tn5281 from

Enterococcus faecalis and comparison to staphylococcal

transposons Tn4001 and Tn4031. Antimicrob Agents

Chemother 35: 1147–1152.Homan WL, Tribe D, Poznanski S, Li M, Hogg G, Spalburg E,

Van Embden JD & Willems RJ (2002) Multilocus sequence

typing scheme for Enterococcus faecium. J Clin Microbiol 40:

1963–1971.Jensen LB, Garcia-Migura L, Valenzuela AJ, Lohr M, Hasman

H & Aarestrup FM (2010) A classification system for

plasmids from enterococci and other Gram-positive

bacteria. J Microbiol Methods 80: 25–43.

ª 2012 Federation of European Microbiological Societies FEMS Immunol Med Microbiol 66 (2012) 166–176Published by Blackwell Publishing Ltd. All rights reserved

174 T.C.S. Rosvoll et al.

Johnson PD, Ballard SA, Grabsch EA, Stinear TP, Seemann T,

Young HL, Grayson ML & Howden BP (2010) A sustained

hospital outbreak of vancomycin-resistant Enterococcus

faecium bacteremia due to emergence of vanB E. faecium

sequence type 203. J Infect Dis 202: 1278–1286.Kao SJ, You I, Clewell DB, Donabedian SM, Zervos MJ, Petrin

J, Shaw KJ & Chow JW (2000) Detection of the high-level

aminoglycoside resistance gene aph(2″)-Ib in Enterococcus

faecium. Antimicrob Agents Chemother 44: 2876–2879.Kim DS, Singh KV, Nallapareddy SR, Qin X, Panesso D, Arias

CA & Murray BE (2010) The fms21(pilA)-fms20 locus

encoding one of four distinct pili of Enterococcus faecium is

harboured on a large transferable plasmid associated with gut

colonization and virulence. J Med Microbiol 59: 505–507.Laverde Gomez JA, van Schaik W, Freitas AR, Coque TM,

Weaver KE, Francia MV, Witte W & Werner G (2011) A

multiresistance megaplasmid pLG1 bearing a hylEfm

genomic island in hospital Enterococcus faecium isolates. Int

J Med Microbiol 301: 165–175.Leavis HL, Bonten MJ & Willems RJ (2006) Identification of

high-risk enterococcal clonal complexes: global dispersion

and antibiotic resistance. Curr Opin Microbiol 9: 454–460.Leendertse M, Heikens E, Wijnands LM et al. (2009)

Enterococcal surface protein transiently aggravates

Enterococcus faecium-induced urinary tract infection in mice.

J Infect Dis 200: 1162–1165.Lester CH, Sandvang D, Olsen SS, Schonheyder HC, Jarlov JO,

Bangsborg J, Hansen DS, Jensen TG, Frimodt-Moller N &

Hammerum AM (2008) Emergence of ampicillin-resistant

Enterococcus faecium in Danish hospitals. J Antimicrob

Chemother 62: 1203–1206.Moellering RC Jr (1991) The Garrod Lecture. The

enterococcus: a classic example of the impact of

antimicrobial resistance on therapeutic options. J Antimicrob

Chemother 28: 1–12.Morrison D, Woodford N, Barrett SP, Sisson P & Cookson

BD (1999) DNA banding pattern polymorphism in

vancomycin-resistant Enterococcus faecium and criteria for

defining strains. J Clin Microbiol 37: 1084–1091.Nallapareddy SR, Weinstock GM & Murray BE (2003) Clinical

isolates of Enterococcus faecium exhibit strain-specific

collagen binding mediated by Acm, a new member of the

MSCRAMM family. Mol Microbiol 47: 1733–1747.Nallapareddy SR, Singh KV, Okhuysen PC & Murray BE

(2008) A functional collagen adhesin gene, acm, in clinical

isolates of Enterococcus faecium correlates with the recent

success of this emerging nosocomial pathogen. Infect Immun

76: 4110–4119.NORM/NORM-VET (2008) NORM/NORM-VET 2008. Usage

of Antimicrobial Agents and Occurrence of Antimicrobial

Resistance in Norway. Tromsø/Oslo 2009. NORM/NORM-

VET, Tromsø/Oslo.

Panesso D, Reyes J, Rincon S et al. (2010) Molecular

epidemiology of vancomycin-resistant Enterococcus faecium:

a prospective, multicenter study in South American

hospitals. J Clin Microbiol 48: 1562–1569.

Panesso D, Montealegre MC, Rincon S, Mojica MF, Rice LB,

Singh KV, Murray BE & Arias CA (2011) The hylEfm gene

in pHylEfm of Enterococcus faecium is not required in

pathogenesis of murine peritonitis. BMC Microbiol 11: 20.

Rice LB, Lakticova V, Carias LL, Rudin S, Hutton R &

Marshall SH (2009) Transferable capacity for

gastrointestinal colonization in Enterococcus faecium in a

mouse model. J Infect Dis 199: 342–349.Rosvoll TC, Pedersen T, Sletvold H, Johnsen PJ, Sollid JE,

Simonsen GS, Jensen LB, Nielsen KM & Sundsfjord A

(2010) PCR-based plasmid typing in Enterococcus faecium

strains reveals widely distributed pRE25-, pRUM-, pIP501-

and pHTbeta-related replicons associated with glycopeptide

resistance and stabilizing toxin-antitoxin systems. FEMS

Immunol Med Microbiol 58: 254–268.Sangsuk L, Dejsirilert S, Sawanpanyalert P, Aswapokee N &

Jariyasethpong T (2009) Antimicrobial susceptibility of

enterococci in Thailand from 2000 to 2005. J Med Assoc

Thai 92(suppl 4): S1–S7.Sava IG, Heikens E & Huebner J (2010a) Pathogenesis and

immunity in enterococcal infections. Clin Microbiol Infect

16: 533–540.Sava IG, Heikens E, Kropec A, Theilacker C, Willems R &

Huebner J (2010b) Enterococcal surface protein contributes

to persistence in the host but is not a target of opsonic and

protective antibodies in Enterococcus faecium infection.

J Med Microbiol 59: 1001–1004.Sillanpaa J, Nallapareddy SR, Prakash VP, Qin X, Hook M,

Weinstock GM & Murray BE (2008) Identification and

phenotypic characterization of a second collagen

adhesin, Scm, and genome-based identification and

analysis of 13 other predicted MSCRAMMs, including

four distinct pilus loci, in Enterococcus faecium.

Microbiology 154: 3199–3211.Sillanpaa J, Prakash VP, Nallapareddy SR & Murray BE (2009)

Distribution of genes encoding MSCRAMMs and Pili in

clinical and natural populations of Enterococcus faecium.

J Clin Microbiol 47: 896–901.Sillanpaa J, Nallapareddy SR, Singh KV, Prakash VP, Fothergill

T, Ton-That H & Murray BE (2010) Characterization of the

ebp(fm) pilus-encoding operon of Enterococcus faecium and

its role in biofilm formation and virulence in a murine

model of urinary tract infection. Virulence 1: 236–246.Simjee S, Fraise AP & Gill MJ (1999) Plasmid heterogeneity

and identification of a Tn5281-like element in clinical isolates

of high-level gentamicin-resistant Enterococcus faecium

isolated in the UK. J Antimicrob Chemother 43: 625–635.Simjee S, Manzoor SE, Fraise AP & Gill MJ (2000) Nature of

transposon-mediated high-level gentamicin resistance in

Enterococcus faecalis isolated in the United Kingdom.

J Antimicrob Chemother 45: 565–575.Top J, Willems R, Blok H et al. (2007) Ecological replacement

of Enterococcus faecalis by multiresistant clonal complex 17

Enterococcus faecium. Clin Microbiol Infect 13: 316–319.Turabelidze D, Kotetishvili M, Kreger A, Morris JG Jr &

Sulakvelidze A (2000) Improved pulsed-field gel

FEMS Immunol Med Microbiol 66 (2012) 166–176 ª 2012 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

HLGR-encoding megaplasmids in invasive E. faecium 175

electrophoresis for typing vancomycin-resistant enterococci.

J Clin Microbiol 38: 4242–4245.van Schaik W, Top J, Riley DR et al. (2010) Pyrosequencing-

based comparative genome analysis of the nosocomial

pathogen Enterococcus faecium and identification of a large

transferable pathogenicity island. BMC Genomics 11: 239.

Watanabe S, Kobayashi N, Quinones D, Nagashima S, Uehara

N & Watanabe N (2009) Genetic diversity of enterococci

harboring the high-level gentamicin resistance gene aac(6′)-Ie-aph(2″)-Ia or aph(2″)-Ie in a Japanese hospital. Microb

Drug Resist 15: 185–194.Weaver KE, Kwong SM, Firth N & Francia MV (2009) The

RepA_N replicons of Gram-positive bacteria: a family of

broadly distributed but narrow host range plasmids. Plasmid

61: 94–109.Weisburg WG, Barns SM, Pelletier DA & Lane DJ (1991) 16S

ribosomal DNA amplification for phylogenetic study.

J Bacteriol 173: 697–703.Werner G, Freitas AR, Coque TM et al. (2011) Host range of

enterococcal vanA plasmids among Gram-positive intestinal

bacteria. J Antimicrob Chemother 66: 273–282.Willems RJ & van Schaik W (2009) Transition of Enterococcus

faecium from commensal organism to nosocomial pathogen.

Future Microbiol 4: 1125–1135.

Willems RJ, Top J, van Santen M et al. (2005) Global

spread of vancomycin-resistant Enterococcus faecium from

distinct nosocomial genetic complex. Emerg Infect Dis 11:

821–828.Willems RJ, Hanage WP, Bessen DE & Feil EJ (2011)

Population biology of Gram-positive pathogens: high-risk

clones for dissemination of antibiotic resistance. FEMS

Microbiol Rev 35: 872–900.

Supporting Information

Additional Supporting Information may be found in the

online version of this article:

Fig. S1. Dendrogram presenting PFGE patterns of the

E. faecium strain collection (n = 99).

Table S1. Primers and control strains used in this study.

Please note: Wiley-Blackwell is not responsible for the

content or functionality of any supporting materials sup-

plied by the authors. Any queries (other than missing

material) should be directed to the corresponding author

for the article.

ª 2012 Federation of European Microbiological Societies FEMS Immunol Med Microbiol 66 (2012) 166–176Published by Blackwell Publishing Ltd. All rights reserved

176 T.C.S. Rosvoll et al.