Figures
Abstract
The clonal dissemination of VanB-type vancomycin-resistant Enterococcus faecium (VREfm) strains in three Swedish hospitals between 2007 and 2011 prompted further analysis to reveal the possible origin and molecular characteristics of the outbreak strain. A representative subset of VREfm isolates (n = 18) and vancomycin-susceptible E. faecium (VSEfm, n = 2) reflecting the spread in time and location was approached by an array of methods including: selective whole genome sequencing (WGS; n = 3), multi locus sequence typing (MLST), antimicrobial susceptibility testing, virulence gene profiling, identification of mobile genetic elements conferring glycopeptide resistance and their ability to support glycopeptide resistance transfer. In addition, a single VREfm strain with an unrelated PFGE pattern collected prior to the outbreak was examined by WGS. MLST revealed a predominance of ST192, belonging to a hospital adapted high-risk lineage harbouring several known virulence determinants (n≥10). The VREfm outbreak strain was resistant to ampicillin, gentamicin, ciprofloxacin and vancomycin, and susceptible to teicoplanin. Consistently, a vanB2-subtype as part of Tn1549/Tn5382 with a unique genetic signature was identified in the VREfm outbreak strains. Moreover, Southern blot hybridisation analyses of PFGE separated S1 nuclease-restricted total DNAs and filter mating experiments showed that vanB2-Tn1549/Tn5382 was located in a 70-kb sized rep17/pRUM plasmid readily transferable between E. faecium. This plasmid contained an axe-txe toxin-antitoxin module associated with stable maintenance. The two clonally related VSEfm harboured a 40 kb rep17/pRUM plasmid absent of the 30 kb vanB2-Tn1549/Tn5382 gene complex. Otherwise, these two isolates were similar to the VREfm outbreak strain in virulence- and resistance profile. In conclusion, our observations support that the origin of the multicentre outbreak was caused by an introduction of vanB2-Tn1549/Tn5382 into a rep17/pRUM plasmid harboured in a pre-existing high-risk E. faecium ST192 clone. The subsequent dissemination of VREfm to other centres was primarily caused by clonal spread rather than plasmid transfer to pre-existing high-risk clones.
Citation: Sivertsen A, Billström H, Melefors Ö, Liljequist BO, Wisell KT, Ullberg M, et al. (2014) A Multicentre Hospital Outbreak in Sweden Caused by Introduction of a vanB2 Transposon into a Stably Maintained pRUM-Plasmid in an Enterococcus faecium ST192 Clone. PLoS ONE 9(8): e103274. https://doi.org/10.1371/journal.pone.0103274
Editor: Willem van Schaik, University Medical Center Utrecht, Netherlands
Received: February 13, 2014; Accepted: June 30, 2014; Published: August 25, 2014
Copyright: © 2014 Sivertsen et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by research grants from the Northern Norway Regional Health Authority Medical Research Program. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Enterococci, and Enterococcus faecium in particular, have undergone a transition from harmless gut commensals to be a leading cause of multidrug resistant hospital infections [1]. E. faecium is associated with urinary tract infections, endocarditis, infections in indwelling catheters and septicaemia in hospitalised patients [2], [3]. Notably, a pronounced increase in bacteraemias caused by E. faecium in Europe has been reported [4]. The ability of hospital adapted lineages of E. faecium to compile antibiotic resistance and virulence factors by horizontal gene transfer might be attributable to this observation [1].
The use of Multi Locus Sequence Typing (MLST) has been considered a standard method for global epidemiological surveillance [5], while pulsed-field gel electrophoresis (PFGE) is the preferred method for examination of E. faecium seen in local outbreaks. The clonal complex (CC) 17 has shown to pool hospital associated E. faecium strains characterised by a high rate of recombination, multidrug resistance as well as numerous virulence determinants [6]–[8]. Newer insights pertained with another type of population structure analysis, Bayesian Analysis of Population Structure (BAPS) of CC17 strains, show divergent origins of sequence type (ST) lineages within CC17. CC17 strains could largely be divided in two BAPS groups, 2-1 (lineage 78), and 3-3 (lineages 17 and 18), with the corresponding MLST ancestry nodes in parenthesis [9].
A worldwide increased prevalence of acquired resistance to commonly used antibiotics is observed in clinical isolates of E. faecium. A total of eight gene clusters, vanA,B,D,E,G,L,M,N, have been associated with acquired vancomycin resistance in enterococci (VRE) [10]–[13]. VanA VRE is most prevalent globally, but VanB-type VRE are predominant in Australia and on the rise in many European countries [14]–[19]. The vanB gene cluster include three subtypes, vanB1-3, conferring inducible low- to high-level resistance to vancomycin and susceptibility to teicoplanin [20]. The predominant subtype vanB2 is an integral part of an Integrative Conjugative Element transposon family Tn1549/Tn5382 supporting transfer of vanB2 [21]. Several plasmid replicon types encoding glycopeptide resistance as well as stabilising toxin-antitoxin systems have been linked to CC17 strains [22]. Moreover, a number of putative virulence genes have been associated with E. faecium and it is hypothesised that their phenotypes might work in concert to promote host colonisation and subsequent invasion [2].
The prevalence of vancomycin resistant E. faecium (VREfm) in Sweden remained low until 2007 when a large hospital associated outbreak occurred [15]. Three hospitals in separate counties were involved. The outbreak was not declared over until 2011.
The aim of the present study was to explore the origin of the outbreak strain by performing molecular characterisation of representative isolates of the outbreak strain and compare them with consecutive invasive E. faecium isolates from the same time period and location.
Materials and Methods
Bacterial isolates
All cases of vancomycin-resistant enterococci (VRE) are mandatorily reported to the Public Health Agency of Sweden and collected for resistance and epidemiological typing. In 2007, an increasing number of notified VRE-cases were seen in Stockholm County, related to a clonal VanB-type E. faecium strain. The strain was subsequently reported in two other geographically distant counties (Västmanland and Halland) [15]. During the autumn 2008, clonally related isolates of VanB-type Efm (n = 17) from separate infection/colonisation events in these three counties were selected for molecular studies. The selected isolates occurred early and late during the outbreak period and represented isolates with divergent resistance profiles and related PFGE subtypes. Four blood isolates from Collection B (see below) were also included in the molecular analyses, yielding a total of 21 isolates in Collection A.
In an attempt to reveal the origin of the E. faecium vanB outbreak strain, all consecutive E. faecium blood culture isolates from 1st of January 2006 to 31st of August 2009 (n = 191; Collection B) diagnosed at the Karolinska University Hospital Huddinge where the outbreak was first identified, were collected and analysed by PFGE; 2006 (n = 45), 2007 (n = 32), 2008 (n = 71) and 2009 (n = 43). Four of these isolates were selected for further analysis using whole genome sequencing (WGS) with the Roche 454 pyrosequencing platform. The selection criterion was based on PFGE patterns, where the first vancomycin susceptible (VSE1036), the first vancomycin resistant (VRE1044) and the most recent vancomycin resistant (VRE1261) isolate with indistinguishable or closely related PFGE-patterns to the outbreak strain, were chosen. For comparison, one vancomycin resistant isolate (VRE0576) from 2006 was chosen because of its divergent PFGE pattern. E. faecium 64/3 [23], BM4105RF and BM4105-Str [24] were used as recipients in filter mating experiments. Isolates from a polyclonal cluster of vanB2 positive E. faecium from 2002–2004 in the Swedish county Örebro [18] were included in the ICESluvan Q8 PCR to evaluate their vanB transposon signature.
Antimicrobial susceptibility testing
The minimum inhibitory concentration (MIC) was determined using Etest (BioMerieux) and interpreted according to the clinical breakpoints of the European committee on Antimicrobial Susceptibility Testing (EUCAST) (www.eucast.org). For ciprofloxacin, a tentative breakpoint was used classifying isolates with MIC >32 mg/L as high level resistant [25].
PFGE and MLST
For SmaI-digestion, the protocol adapted by Saeedi et al. [26] was used with 5 U/mL lysozyme added in the lysis buffer. The bands were separated with the following program: Block I switch time 3 to 26,5s for 14 hours and 50 minutes. Block II: switch time 0,5 to 8,5s for 6 hours and 25 minutes. Total run time 21 hours and 15 minutes at 6V with 120°. The PFGE patterns were analysed and compared using BioNumerics software (version 6.6, Applied Maths). The Dice coefficient was used for pair-wise comparison of patterns, and the un-weighted pair group method with arithmetic mean (UPGMA) for pattern grouping. Isolates clustering above 97% were considered identical and isolates with identity >90% closely related.
MLST was performed using the method adapted by Homan et al. [5] with the following primers: adk1n, adk2n, atp1n, atp2n, ddl1, ddl2, gdh1, gdh2, gyd1, gyd2, pstS1n, pstS2n, purK1n and purK2n.
Detection of genes by PCR and isolation of bacterial DNA
Extraction of DNA for all PCRs were performed by BioRobot M48 (Qiagen), according to the manufacturers manual. Primers and positive controls are described in Table S1.
The chosen virulence genes are associated with high-risk genotypes, and included esp [27], hyl [28], acm [29], efaAfm [30], sgrA, ecbA, scm, orf903/2010/2514 and pilA/B [31]–[33]. PCRs genotyping presence of vanB [20] and linkage to Tn5382 [34] was done. PCRs were conducted as stated in Table S1. PCR was performed using the JumpStart REDTaq Readymix PCR Reaction mix (Sigma), with a standard program of 1 min in 95°C followed by 30 cycles at 95°C for 30 sec, 30 sec of annealing in the temperature given in Table S1, 72°C for 1 min with a final elongation step at 72°C for 7 min. The presence of vanB in transconjugants was tested by PCR using 1 µl of bacterial culture in BHI broth and an additional initial denaturation step of 10 min at 95°C.
Southern blotting and hybridisation
PFGE analysis of S1-digested DNA was used to analyse the plasmid content. Plugs were made as for SmaI digestion, and the digestion was performed as described by Rosvoll et al. [22]. The Vacugene XL system (Amersham Biosciences) was used for Southern blotting. Consecutive hybridisation was performed using rep17/pRUM, vanB and axe-txe probes in the mentioned order. Characterisation of the reppLG1 [35] and rep2/pRE25 [36] plasmid determinants were also done by Southern blotting and hybridisation after S1-nucelase PFGE. Probes were made by amplification using positive controls (see Table S1), and labelled using the PCR DIG synthesis kit (Boehringer Mannheim). The same hybridisation protocol as in Rosvoll et al. [22] was used with the following modification: The DNA was purified after the first PCR using the Cycle Pure Kit (zDNA).
Conjugative transfer of vanB
Filter mating was performed according to Bjørkeng et al. [18] with some minor modifications, using the E. faecium 64/3 and E. faecium BM4105-RF as recipient strains for the first filter mating, and BM4015-Str in retransfer. Briefly, the isolates were grown together on MF-Millipore membrane filters for 24 h, spotted on selective BHI agar plates containing either vancomycin (8 mg/L), fusidic acid (10 mg/L) and rifampicin (20 mg/L), or all three antibiotics together. The bacterial suspension was serially diluted down to 10−9 and incubated at 37°C for 48 h. In the retransfer experiments, the recipients were selected on plates containing 1000 mg/L streptomycin.
Whole genome sequencing and analysis
Chromosomal DNA from the four isolates (VSE1036, VRE1044, VRE1261 and VRE0576) was prepared using the DNeasy Blood and Tissue Kit (Qiagen) with lysozyme (20 mg/mL) added to the lysis buffer and further treated with RNase. The protocol also allowed purification of plasmids. Libraries were prepared and used for whole genome shotgun sequencing on a 4-region picotiter plate with the Roche 454 FLX system according to standard protocols (www.454.com). Raw sequencing data were processed with standard filters using the GS Run Processor (v 2.6), generating between 246084 and 310421 reads for each of the strains with average lengths between approximately 307 and 320 nucleotides, corresponding to between 78750175 and 99124035 nucleotides. Reads were assembled de novo with the accompanying GS de novo assembler software (v 2.6) (454 Newbler algorithm) generating between 201 and 302 contigs with a length of more than 100 nucleotides. The GS Reference mapper software (v 2.6) was subsequently used for homology comparisons between the different strains and to find homologies to specific query gene sequences and also used to identify indels and point mutations that separated the different strains. Some small plasmids could be identified by screening for contigs where individual reads mapped to both ends of the contig. The contigs were tentatively linked to each other by comparison of molecular biology data, identification of individual reads mapping to two different contigs and comparisons with published genomes. The tentative gene content of the contigs was automatically analysed at NCBI with the Prokaryotic Genome Annotation Pipeline (PGAP).
In Figure S1 alignment of genome sequences against the reference genome of Aus0004 (NC_017022) was done by reordering the contigs of VRE1044 against the Aus0004 reference genome using Mauve 2.3.1. The reordered contigs of VRE1044 were then used as reference to reorder the contigs of the other isolates before the genomes were aligned using progressiveMauve [37].
The Whole Genome Shotgun projects have been deposited at DDBJ/EMBL/GenBank under the accessions JAAJ00000000 (VSE1036), JAAK00000000 (VRE0576), JAAL00000000 (VRE1044) and JAAM00000000 (VRE1261). The versions described in this paper are versions JAAJ01000000, JAAK01000000, JAAL01000000 and JAAM01000000.
Results
PFGE patterns and antimicrobial susceptibility of VREfm and VSEfm
In an attempt to identify a putative ancestor of the VREfm outbreak strain, we examined consecutive blood isolates (n = 191, Collection B) of E. faecium from Karolinska University Hospital, Huddinge, during a four-year period (2006–2009) covering the time of appearance of the presumably first clinical isolate of the outbreak VREfm strain. A dominant PFGE pattern (n = 37; 26 VSEfm and 11 VREfm) was indistinguishable or closely related to that of the VREfm outbreak strain (named SE-EfmB-0701). Representative isolates with this PFGE-pattern are shown in Figure 1. Among the blood isolates analysed in retrospect, this PFGE pattern was observed for the first time in a vancomycin susceptible isolate (VSEfm) from February 2007 and soon after in two more VSEfm. Three additional VREfm with identical PFGE pattern were detected during the autumn of 2007. The first known clinical isolate (VRE0651) was found in an abdominal infection in August 2007. In 2008 another eight VREfm and 14 VSEfm isolates with the same PFGE pattern were detected. During January 1st until August 31st 2009 no more VREfm but still 10 VSEfm of the SE-EfmB-0701 PFGE type were detected.
The dendogram shows that 20 of the isolates are clonal (lane 1–20) and one is divergent (lane 21). The symbol ♦ in the dendrogram indicates a similarity of 90.6%. The PFGE-type nomenclature is based on the following: SE stands for Sweden, EfmB stands for E. faecium with vanB, the number 07 represents year 2007 (the year the index was identified) and the last number is a serial number. The letter at the end describes which PFGE-subtype the isolate belongs to.
Antimicrobial susceptibility testing of the 37 blood culture isolates revealed high-level gentamicin resistance (n = 14; 38%) as well as ampicillin (100%) and ciprofloxacin (100%) resistance and teicoplanin susceptibility (100%). The vanB gene was detected in all VREfm isolates.
In collection A (n = 21), representing PFGE patterns as diverse as possible at the time of selection (autumn 2008), a total of 9 subgroups (a to i) of the pattern SE-EfmB-0701 were found (Table 1). All PFGE patterns displayed a similarity >90%, thereby fulfilling the suggested definition of relatedness (>81%) according to Morrison et al. [38]. The isolates did however not group together consistently in relation to their geographical origin (Figure 1).
The nineteen VREfm isolates in collection A had vancomycin MICs ranging from 8 to ≥256 mg/L and were susceptible to teicoplanin, consistent with the vanB2 genotype. All the isolates were resistant to ampicillin and ciprofloxacin, and seven displayed high-level resistance to gentamicin. The two VSE-isolates (VSE1036 and VSE1027) in collection A were susceptible to both vancomycin and teicoplanin, but expressed resistance to ampicillin and ciprofloxacin (Table 1).
MLST and WGS analyses of VREfm and VSEfm
Collection A (n = 21) were studied in greater detail by MLST and PCR for virulence genes (Table 1). Three isolates of the outbreak strain and one unrelated VREfm isolate were examined by WGS; VSE1036 (ST192), VRE1044 (ST192), VRE1261 (ST192) and VRE0576 (ST17). In collection A, all isolates including the two VSE belonged to ST192 except one single locus (VRE0673, ST78) and two double locus (VRE0881 and VRE0576, ST17) variants of ST192. All isolates except VRE0881 and VRE0576 belonged to the ST78 lineage. All isolates except VRE0576, a pre-outbreak vanB-positive ST17 isolate from 2006 had a related PFGE pattern as shown by SmaI PFGE (Figure 1). The MLST- and PFGE results concurred moderately in showing relation, since the isolates with the most divergent PFGE patterns (VRE0673, VRE0776 and VRE0881) in two of three cases had deviating MLST profiles.
The WGSs from the outbreak isolates VRE1044, VRE1261, and VSE1036 as well as the unrelated vanB-positive pre-outbreak isolate VRE0576 were aligned against the chromosome of an E. faecium isolate from Australia which contains a vanB2 transposon (Aus0004) (Figure S1, vanB2 transposons are indicated by red triangles). The WGS of the three ST192 outbreak isolates were naturally more homologous and shared several regions (highlighted by black triangles) that were not present in the genome sequences of either VRE0576 or Aus0004 (both ST17) although it should be noted that the plasmid sequences of Aus0004 were not included in this comparison. The VRE0576 and Aus0004 genomes showed many unique regions (white areas). Furthermore, VRE1044 and VRE1261 showed some unique regions (green triangles) that were not present in VSE1036. Further analyses of these regions suggested they belong to mobile genetic elements.VRE1044 and VRE1261 showed only minor differences.
Antimicrobial resistance determinants and virulence genes of VREfm and VSEfm
All VREfm isolates in collection A harboured the vanB2 gene as an integral part of Tn1549/Tn5382 demonstrated by the vanXB-ORFC-PCR (Table 1). This link was also confirmed by WGSs in three VREfm isolates (Table S2). WGS data of VRE576 revealed the same genetic organisation of the vanB2 transposon as well as 99% nucleotide (nt) identity to Tn1549. Interestingly, WGS data from VRE1044 and VRE1261 showed the same vanB2 transposon organisation and 99% nucleotide identity to Tn1549, but also an additional 2588 bp inserted between nt 5014 and 5015 of Tn1549. We performed ICESluvan Q8 PCR (Table S1) covering the putative insertion region in the remaining VRE isolates to disclose a potential unique insertion signature of the vanB transposon in SE-EfmB-0701 isolates. Presence of an approximately 2.6 kb insertion was confirmed in the vanB transposon of all the VREfm SE-EfmB-0701 isolates in collection A. The 2588 bp insert sequence is 89% identical to the region in Clostridium saccharolyticum-like K10 (GenBank Acc. No. FP929037) encoding a retron-type reverse transcriptase. In line with this, the 2588 bp sequence encodes a putative protein of 610 amino acids (aa) with 99% identity to a putative reverse transcriptase/maturase from Faecalibacterium prausnitzii A2-165 (GenBank Acc. No. EEU96266) and a putative group II intron-encoded protein LtrA (reverse transcriptase and RNA maturase) from Flavonifractor plautii ATCC 29863 (GenBank Acc. No. EHM54980). The putative protein further shows 43% identity to the group II intron 599 aa multifunctional protein LtrA in Lactococcus lactis (GenBank Acc. No. U50902). LtrA is known to have reverse transcriptase, RNA maturase and site-specific DNA endonuclease activity mediating intron splicing and mobility [39].
The gyrA and parC genes extracted from the WGSs revealed SNPs associated with ciprofloxacin resistance. Two mutation events in each gene were found (Table 1), and both aa combinations (GyrA Arg83, ParC Ile80 or GyrA Ile83, ParC Arg80) have been described previously in E. faecium isolates with ciprofloxacin MICs ≥16 mg/L [25], [40], [41]. Moreover, the tetracycline resistance determinant tetM and the macrolide resistance determinant ermB were also found in three and four WGS isolates, respectively (Table S2).
All isolates of the PFGE type SE-EfmB-0701 harboured esp, sgrA, acm, scm, pilB, efaAfm, orf2010 and orf 2514. Moreover, 17 of 20 isolates contained hyl. PCR data showed that the genes pilA, ecbA and orf903 occurred in six, one and one of 20 isolates, respectively. The ecbA and orf903 genes were found in isolates with unique PFGE subtypes in this collection (VRE0776 and VRE0673) (Table 1). WGS data revealed that pilA (VRE1044 and VRE1261) and ecbA (VSE1036, VRE1044 and VRE1261) were present with a nucleotide match of 1672/1976 (85%) for pilA and 2766/3173 (87%) for ecbA compared to reference sequences in isolates E1162 and TX16, respectively. In addition, VSE1036, VRE1044, and VRE1261 carried a truncated version of pilA (219 bp) 100% identical to the reference sequences. Blastn search of the pilA and ecbA sequences within the NCBI shotgun sequence database revealed that these two genes had a notable SNP variation between different strains (0–20%). The original pilA and ecbA primers used yielded no good matches in the WGS sequences which explain why we got few positive PCR products with these primers. All our collection A isolates were positive for these genes using new primers (pilA 2 and ecbA 2 Table S1) targeting conserved regions in the WGS and reference ecbA and pilA genes.
Plasmid localisation of Tn1549-type transposon
By BLAST alignment of the contigs from VRE1044 and VRE1261 containing the vanB2 transposon against the WGS data of the VSE1036 isolate, the exact AT-rich location of the vanB2 transposon insertion site could be identified in contig00062 of VSE1036 (Figure 2 and blue triangle in Figure S1). The transposon insertion site was identical for VRE1044 and VRE1261 and corresponded with 100% identity to sequence in contig00062 of VSE1036 (Figure 2).
The figure shows the transposon insertion regions of VRE0576 versus VRE1044 and VRE1261 and the corresponding region in VSE1036 (contig00062). Tn1549/Tn5382 left and right end imperfect inverted repeats are shown in bold capital letters. Vertical lines indicate identical nucleotides.
To get a better picture of the plasmid content we performed Southern blot hybridisation of S1 nuclease treated total DNAs separated by PFGE using specific replication (rep) gene probes (Table 1). WGS data gave additional information on plasmid sequences as well as resistance genes and Tn1549/Tn5382 (Table S2). Replication genes of rep-classes rep2/pRE25, rep11/pB82, rep17/pRUM and reppLG1 were present in all four WGS, with some nucleotide differences in the pre-outbreak isolate (VRE0576) compared to the other three isolates. VRE0576 also differed from the other isolates by lacking a rep14/pRI1-class gene present with identical nucleotide identity scores in the other WGSs, and by containing a putative repunique/pCIZ2-class gene absent in the other WGSs. Notably, VRE1044 and VRE1261 contained two rep2/pRE25-sequences with SNP differences on different contigs, consistent with the observed two rep2/pRE25 plasmids in the hybridisation analyses (Table 1 and S2). The VSE1036 isolate had a similar hybridisation pattern as VRE1044 and VRE1261, but did only have one 50-kb plasmid and one contig with rep2/pRE25.
Hybridisation results showed that all VREfm isolates harboured the vanB resistance gene on an approximately 70-kb rep17/pRUM replicon (shown for representative isolates VRE0726, VRE0734 and VRE0881 in Figure 3 lanes 5, 8 and 11, and VRE0690, VRE0653, VRE0776 in Figure 4 lanes 5, 7 and 9). Further hybridisation of selected isolates revealed a dominant common plasmid content pattern for VRE0726, VRE0651, VRE0678, VRE1044, VRE1261, and VRE0650, and some minor differences in the other 11 characterised isolates (Table 1).
These results illustrated transfer from donors VRE0726, VRE0734 and VRE0881 (lanes 5, 8 and 11) of a similar sized plasmid (approximately 140 kb) to 64/3 (lane 1) (1st generation transconjugants shown in lanes 6, 9 and 12) which was subsequently retransferred to BM4105Str (lane 2) (2nd generation transconjugants shown in lanes 7, 10 and 13) when using the 1st generation transconjugants as donors. Lane 3 vanB positive control V583, lane 4 rep17/pRUM, axe-txe and vanB positive control E. faecium U37, lanes M low-range PFGE marker.
Donors VRE0690, VRE0653 and VRE0776 (lanes 5, 7 and 9) and their respective transconjugants (lanes 4, 6 and 8) illustrate transfer of different sized plasmids co-hybridising to vanB and rep17/pRUM (circa 110–150 kb) into 64/3 (lane 1). Lane 2 vanB positive control V583, lane 3 rep17/pRUM and vanB positive control E. faecium U37, lane M low-range PFGE marker.
Interestingly, the two VSE isolates clonally related to the VREfm outbreak strain (VSE1027 and VSE1036) differed from the common pattern by having a rep17/pRUM replicon 30 kb smaller than that of the VREfm isolates (Table 1), corresponding to the size of the vanB2-Tn1549/Tn5382 transposon. Notably, the pRUM repA was found with 100% homology to the reference sequence in VRE0576 (GenBank Accession number JAAK00000000), and present with 97% homology in VSE1036 (JAAJ00000000), VRE1044 (JAAL00000000) and VRE1261 (JAAM00000000). A subsequent comparison of the rep17/pRUM replicon sequence in the three latter strains showed 100% homology. Thus, these plasmids share identical signature sequences for pRUM repA as well as for the Tn1549/Tn5382 transposon insertion site.
Tn1549/Tn5382-related genes, axe-txe and the vanB2 gene were found within the same contig in the VRE1044 and VRE1261 isolates, thus supporting the S1 nuclease PFGE hybridisation and PCR data showing linkage of vanB2-Tn1549/Tn5382 and axe-txe genes on the same replicon (representative isolates in Figure 3 lanes 5, 8 and 11). WGS analyses of VSE1036 showed that the sequence corresponding to the transposon insertion site of VRE1044 and VRE1261 (Figure 2) mapped to the same contig00062 as axe-txe (Table S2) typically found on rep17/pRUM replicons. Interestingly, the reference pRUM plasmid contains a putative relaxase gene and a mobilisation gene both consistently absent from all four WGSs.
Transferability studies of the VanB determinant
All but 2 (VRE0651 and VRE0673) out of 10 tested VREfm isolates produced transconjugants with a low frequency of 10−8 to 10−11 per donor, close to the detection limit, using E. faecium 64/3 as a recipient (Table 2). Transconjugants were confirmed by PFGE using SmaI-digestion (Figure S2 and S3). E. faecium recipient BM4105-RF did not support conjugation within the detection limit.
S1 nuclease PFGE and subsequent hybridisation of the 64/3 transconjugants revealed that all the transconjugants contained a vanB-rep17/pRUM plasmid of variable size (110–170 kb), larger than the original 70-kb donor plasmids (Figures 3, 4, S5 and S6). With the exception of VRE0690×64/3 (170 kb) (Figure 4 lane 4 and Figure S5 lane 7) and VRE0776×64/3 (110 kb) (Figure 4 lane 8 and Figure S5 lane 9), the transconjugants had a vanB-rep17/pRUM-rep2/pRE25 containing plasmid of about 140 kb (Figure 3 lanes 6, 9 and 12, Figure 4 lane 6, Figure S5 lanes 3, 5 and 11 and Figure S6 lanes 5, 8 and 11) which corresponds to the joint size of the 70-kb rep2/pRE25 and the 70-kb vanB-rep17/pRUM of their donors. Conversely, the two donors VRE0690 and VRE0776 that gave transconjugants with different sized plasmids both lacked a copy of the 70-kb rep2/pRE25 plasmid (Table 1). VRE0776 contain a 40-kb rep2/pRE25 plasmid which joint with the 70-kb vanB-rep17/pRUM could result in the 110-kb vanB-rep17/pRUM-rep2/pRE25 seen in Figure S5 lane 9. In VRE0690 the 170 kb plasmid of the transconjugant did not hybridise to rep2/pRE25 suggesting that the 170-kb plasmid result from joining of the 70-kb vanB-rep17/pRUM with the 100-kb unknown replicon (Figure S5 lane 7).
The transconjugants were tested for susceptibility to streptomycin in order to identify eligible donors for retransfer experiments. The first generation transconjugants originating from donors VRE0726 (ST192), VRE0734 (ST192) and VRE0881 (ST17) were susceptible. The streptomycin resistance in the other transconjugants probably originate from co-transfer of streptomycin resistance with the plasmids of their clinical isolate donors (VRE0653, VRE0683, VRE0690, VRE0688 and VRE0776) that all showed high-level resistance to streptomycin.
Retransfer and following S1 nuclease PFGE and Southern hybridisations demonstrated that the vanB-rep17/pRUM-rep2/pRE25 140-kb hybrid plasmids (Figure 3 and Figure S6) were stable in size and readily transferable with transfer rates of 10−3–10−5 transconjugants per donor (Table 2). Transconjugants were confirmed by PFGE using SmaI-digestion (Figure S4).
Discussion
A total of 872 VREfm cases were notified during the outbreak in 2007 to 2011, predominantly as faecal colonisation in elderly hospitalised patients with underlying diseases. Less than 10% of the VREfm isolates were recorded from blood, urine or wound samples [15].
Molecular characterisation of a representative subset of related outbreak isolates showed a strong predominance of ST192, which is a single-locus variant of ST78 and considered to be a high-risk genotype [42]. The MLST- and PFGE results were moderately congruent, as the isolates with less similar PFGE patterns and virulence profiles (VRE0673, VRE0776 and VRE0881) in two of three cases had a divergent MLST profile. Previous BAPS data have concluded that the ST78 and ST17 lineages are located in BAPS group 2-1 and 3-3, respectively, and not closely related. This contrasts our PFGE results which grouped VRE0881 (ST17 lineage) in the SE-EfmB-0701 PFGE type (ST78 lineage). This observation as well as the carriage of a unique virulence gene by VRE0673 (orf903) and VRE0776 (only ecbA gene detected by the original PCR primers), could be explained by the lean inclusion criteria of relatedness by PFGE, thereby accepting not genetically related pulsotypes as related. The long collection time (around a year) and evidence of increased DNA banding pattern polymorphism by E. faecium compared to other bacteria [38] was used as basis for our choice of inclusion criteria. However, the unique vanB transposon signature found in all the SE-EfmB-0701 isolates links VRE0881, VRE0776 and VRE0673 to the outbreak.
The examined isolates expressed multidrug-resistance and harboured several specific genes associated with increased virulence. The vancomycin susceptible isolates (VSE1027 and VSE1036) from the start of the outbreak period were clonally related and exerted the same co-resistance- and virulence profile as the VREfm isolates (Figure 1 and Table 1). Moreover, plasmid profiling and WGS data (Table 1 and S2, Figure S1) also indicated close relatedness between SE-EfmB-0701 PFGE type isolates with minor differences in plasmid profile as well as clear differences in gene content compared to the pre-outbreak isolate.
Our results support the notion that internalisation of the vanB transposon into the rep17/pRUM plasmid coincide with the successful spread of this high-risk strain. The rep17/pRUM replicon has previously been shown to harbour a segregation stability module encoded by a toxin-antitoxin cassette (axe-txe) which have been shown to support maintenance of linked antimicrobial resistance genes [43]. Notably, rep17/pRUM replicons with the axe-txe cassette have been shown to be present in a majority of CC17-like strains [22]. WGS analyses revealed that the SE-EfmB-0701 pRUM replicons contained an axe-txe module with a 100% identity to the original pRUM axe-txe sequence.
A rep17/pRUM-vanB2-Tn1549/Tn5382-axe-txe-plasmid of approximately 120–130 kb has previously been described in a polyclonal cluster of E. faecium from 2002–2004 in the Swedish county Örebro. This cluster originated from BAPS-group 3-3 (ST17, ST18 and single locus variants of these) [18]. The Örebro isolates displayed a different PFGE pattern from SE-EfmB-0701 [15] and had a vanB2 transposon without the unique signature found in SE-EfmB-0701.
The repeatedly observed fusion between the vanB2-containg rep17/pRUM plasmids and rep2/pRE25 replicons during conjugation experiments is an interesting feature. The WGS data did not support the presence of the putative relaxase and mobilisation protein associated with the reference pRUM plasmid. This could explain the need for the vanB2-containing rep17/pRUM plasmids to fuse with a conjugation system from other intracellular sources in order to be mobilised. Several studies have described mosaicism and/or recombination events between enterococcal plasmids [22], [44] which may support enhanced host range or other functional benefits associated with several replicons in one plasmid.
There is a noteworthy reservoir of vanB in intestinal anaerobes, and introduction of vanB2-Tn1549/Tn5382 in enterococci from other co-habitants (mainly Gram-positive anaerobes) in the intestinal environment has been experimentally observed [45]. Howden et al. [46] tested the ecological impact by phylogenetic analysis of the transposons and their insertion sites, and showed that a diversification was likely due to a higher grade of de novo VRE generation compared to cross-transmission between enterococcal strains than previously believed. They also observed an increasing incidence of nosocomial VRE infections despite engagement of control interventions to limit transmission between patients. Based on the extensive PFGE and selective MLST analyses in this study, it is highly probable that the closely related VSEfm ST192 strain was a successful hospital coloniser in Sweden already in 2007. Acquisition of the vanB2 transposon by the VSEfm ST192 outbreak strain is the most likely hypothesis on how vancomycin resistance appeared in this strain. The WGS data strongly support that a vanB2 transposon with unique signature was inserted within a rep17/pRUM-plasmid with an unique pRUM repA sequence signature in a strain already present, causing a parallel evolution between VSEfm clones without, and VREfm clones with the rep17/pRUM-vanB2-Tn1549/Tn5382-axe-txe arrangement.
Suggested clearance time for VRE faecal colonisation is estimated to be 4 years [47], but others have suggested that environmentally adapted VRE are capable of inhabiting the intestines in small numbers for even longer [48], [49].
The selective enriched broth used in some laboratories in Sweden before January 2009 with a vancomycin concentration of 32 mg/L was not suitable for vanB-type resistance screening. To address this problem, microbiological laboratories were then advised to reduce the concentration to 4 mg/L [15]. However, vanB-type VRE may have even lower MICs [17], [50]. Importantly, the EUCAST disk diffusion test used by most laboratories in Sweden as the phenotypic vancomycin-susceptibility test method rely on observing the zone edge quality for identification of low level vanB-type resistance. This introduces observer experience as a variable [51].
In conclusion, the molecular analyses revealed that the E. faecium outbreak strain belonged to the high-risk genetic lineage of ST192. The strain was resistant to several commonly used antibiotics and harboured several virulence genes. A successful rep17/pRUM-plasmid containing a vanB transposon with a unique genetic signature originating from other intestinal bacterial species was present in all the VREfm isolates related to the outbreak strain. The rep17/pRUM plasmid harboured a toxin-antitoxin module supporting plasmid maintenance. In addition the rep17/pRUM replicon can easily join with conjugative genetic elements supporting spread to other high-risk E. faecium clones. The current phenotypic screening methods might hamper efforts in limiting VREfm spread, as low-MIC vanB-type VRE might go undetected [51].
Supporting Information
Figure S1.
Alignment of genome sequences from VRE1044 (row 2), VRE1261 (row 3), VSE1036 (row 4) and VRE0576 (row 5) against the reference genome Aus0004 NC_017022 (row 1). The similarity plot indicates average similarity for each region. Coloured blocks indicate regions of sequence homology in the genomes and white areas indicate regions with low sequence homology. Red triangles indicate contigs containing Tn1549/Tn5382 (VRE1044 contigs 00036 and 00041, VRE1261 contigs 00049 and 00044/VRE0576 contig 00004). The blue triangle indicates the transposon Tn1549/Tn5382 insertion region in contig 00062 of VSE1036. This contig also contains the axe-txe genes typically found on rep17/pRUM replicons. Black (VRE1044, VRE1261 and VSE1036) and green (VRE1044 and VRE1261) triangles highlight regions with contigs or partial contigs found in the outbreak isolates but not in VRE576 or Aus0004.
https://doi.org/10.1371/journal.pone.0103274.s001
(PDF)
Figure S2.
SmaI PFGE of first generation transconjugants (TC) (lanes 5, 7, 9, 11) showing divergent band patterns compared with the clinical isolate donors (lanes 6, 8, 10 and 12) and similar pattern with recipient 64/3 (lane 2). Lanes 1 and 13 low-range PFGE marker, lane 3 vanB positive control E. faecalis V583, lane 4 rep17/pRUM positive control E. faecium U37, lanes 5 and 6 TC and donor VRE0726, lanes 7 and 8 TC and donor VRE0734, lanes 9 and 10 TC and donor VRE0683, lanes 11 and 12 TC and donor VRE0688.
https://doi.org/10.1371/journal.pone.0103274.s002
(PDF)
Figure S3.
SmaI PFGE of first generation transconjugants (TC) (lanes 5, 7, 9, 11) showing divergent band patterns compared with the clinical isolate donors (lanes 6, 8, 10 and 12) and similar pattern with recipient 64/3 (lane 2). Lanes 1 and 13 low-range PFGE marker, lane 3 vanB positive control E. faecalis V583, lane 4 rep17/pRUM positive control E. faecium U37, lanes 5 and 6 TC and donor VRE0690, lanes 7 and 8 TC and donor VRE0653, lanes 9 and 10 TC and donor VRE0776, lanes 11 and 12 TC and donor VRE0881.
https://doi.org/10.1371/journal.pone.0103274.s003
(PDF)
Figure S4.
SmaI PFGE of second generation transconjugants (TCs) (lanes 6–8, 11–13 and 15–17) showing divergent band patterns compared with the first generation transconjugant donors (lanes 5, 10 and 14) and similar pattern with recipient BM4105-Str (lane 4). Lanes 1, 9 and 18 low-range PFGE marker, lane 2 vanB positive control E. faecalis V583, lane 3 rep17/pRUM positive control E. faecium U37, lane 5 donor VRE0726×64/3, lanes 6–8 TCs VRE0726×64/3xBM4105-Str, lane 10 donor VRE0734×64/3, lanes 11–13 TCs VRE0734×64/3xBM4105-Str, lane 14 donor VRE0881×64/3, lanes 15–17 TCs VRE0881×64/3xBM4105-Str.
https://doi.org/10.1371/journal.pone.0103274.s004
(PDF)
Figure S5.
S1-nuclease PFGE and corresponding Southern hybridisations with rep2/pRE25 and rep17/pRUM probes showing co-hybridisation in first generation transconjugants (lanes 3, 5, 9 and 11). Lanes 1 and 12 low-range PFGE marker, lanes 2 and 3 donor and TC VRE0683, lanes 4 and 5 donor and TC VRE0688, lanes 6 and 7 donor and TC VRE0690, lanes 8 and 9 donor and TC VRE0776, lanes 10 and 11 donor and TC VRE0653.
https://doi.org/10.1371/journal.pone.0103274.s005
(PDF)
Figure S6.
S1-nuclease PFGE and corresponding Southern hybridisations with rep2/pRE25 and rep17/pRUM probes showing co-hybridisation in first (lane 5, 8 and 11) and second generation transconjugants (lane 6, 9 and 12). Lane M low-range PFGE marker, lane 1 rep17/pRUM and rep2/pRE25 positive control E. faecium U37, lane 2 recipient 64/3, lane 3 VRE1044, lane 4 VRE0726, lane 5 VRE0726×64/3, lane 6 VRE0726×64/3xBM4105-Str, lane 7 VRE0734, lane 8 VRE0734×64/3, lane 9 VRE0734×64/3xBM4105-Str, lane 10 VRE0881, lane 11 VRE0881×64/3, lane 12 VRE0881×64/3xBM4105-Str, lane 13 recipient BM4105-Str.
https://doi.org/10.1371/journal.pone.0103274.s006
(PDF)
Table S2.
Plasmid replication, resistance, toxin-antitoxin system and conjugative transposon genes found in the WGSs of the pre-outbreak isolate VRE576 and the three outbreak isolates VSE1036, VRE1044 and VRE1261. Gene identity refers to the reference sequence.
https://doi.org/10.1371/journal.pone.0103274.s008
(PDF)
Acknowledgments
We are grateful for good collaboration with the clinical microbiological laboratories in Stockholm, Halland and Västmanland County. We thank Bettina Aasnæs and Tracy Munthali Lunde for excellent technical assistance.
Author Contributions
Conceived and designed the experiments: HB ÖM BOL KTW A. Sundsfjord KH. Performed the experiments: A. Sivertsen HB ÖM. Analyzed the data: A. Sivertsen HB ÖM BOL KTW KH. Wrote the paper: A. Sivertsen HB ÖM BOL KTW MU VÖ A. Sundsfjord KH. Responsible for VRE identification in clinical samples at Karolinska University Hospital: VÖ MU.
References
- 1. Gilmore MS, Lebreton F, van Schaik W (2013) Genomic transition of enterococci from gut commensals to leading causes of multidrug-resistant hospital infection in the antibiotic era. Curr Opin Microbiol 16: 10–16.
- 2. Arias CA, Murray BE (2012) The rise of the Enterococcus: beyond vancomycin resistance. Nat Rev Microbiol 10: 266–278.
- 3. Hidron AI, Edwards JR, Patel J, Horan TC, Sievert DM, et al. (2008) NHSN annual update: antimicrobial-resistant pathogens associated with healthcare-associated infections: annual summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2006–2007. Infect Control Hosp Epidemiol 29: 996–1011.
- 4. de Kraker ME, Jarlier V, Monen JC, Heuer OE, van de Sande N, et al. (2013) The changing epidemiology of bacteraemias in Europe: trends from the European Antimicrobial Resistance Surveillance System. Clin Microbiol Infect 19: 860–868.
- 5. Homan WL, Tribe D, Poznanski S, Li M, Hogg G, et al. (2002) Multilocus sequence typing scheme for Enterococcus faecium. J Clin Microbiol 40: 1963–1971.
- 6. Willems RJ, Top J, van Santen M, Robinson DA, Coque TM, et al. (2005) Global spread of vancomycin-resistant Enterococcus faecium from distinct nosocomial genetic complex. Emerg Infect Dis 11: 821–828.
- 7. Willems RJL, van Schaik W (2009) Transition of Enterococcus faecium from commensal organism to nosocomial pathogen. Future Microbiol 4: 1125–1135.
- 8. Rathnayake IU, Hargreaves M, Huygens F (2012) Antibiotic resistance and virulence traits in clinical and environmental Enterococcus faecalis and Enterococcus faecium isolates. Syst Appl Microbiol 35: 326–333.
- 9. Willems RJ, Top J, van Schaik W, Leavis H, Bonten M, et al. (2012) Restricted gene flow among hospital subpopulations of Enterococcus faecium. MBio 3: e00151–00112.
- 10. Lebreton F, Depardieu F, Bourdon N, Fines-Guyon M, Berger P, et al. (2011) D-Ala-d-Ser VanN-type transferable vancomycin resistance in Enterococcus faecium. Antimicrob Agents Chemother 55: 4606–4612.
- 11. Boyd DA, Willey BM, Fawcett D, Gillani N, Mulvey MR (2008) Molecular characterization of Enterococcus faecalis N06-0364 with low-level vancomycin resistance harboring a novel D-Ala-D-Ser gene cluster, vanL. Antimicrob Agents Chemother 52: 2667–2672.
- 12. Courvalin P (2006) Vancomycin resistance in Gram-positive cocci. Clin Infect Dis 42 (suppl 1) S25–S34.
- 13. Xu X, Lin D, Yan G, Ye X, Wu S, et al. (2010) vanM, a new glycopeptide resistance gene cluster found in Enterococcus faecium. Antimicrob Agents Chemother 54: 4643–4647.
- 14. Werner G, Coque TM, Hammerum AM, Hope R, Hryniewicz W, et al. (2008) Emergence and spread of vancomycin resistance among enterococci in Europe. Euro Surveill 13: 1–11.
- 15. Söderblom T, Aspevall O, Erntell M, Hedin G, Heimer D, et al. (2010) Alarming spread of vancomycin resistant enterococci in Sweden since 2007. Euro Surveill 15: pii = 19620.
- 16. Granlund M, Carlsson C, Edebro H, Emanuelsson K, Lundholm R (2006) Nosocomial outbreak of vanB2 vancomycin-resistant Enterococcus faecium in Sweden. J Hosp Infect 62: 254–256.
- 17. Werner G, Klare I, Fleige C, Geringer U, Witte W, et al. (2012) Vancomycin-resistant vanB-type Enterococcus faecium isolates expressing varying levels of vancomycin resistance and being highly prevalent among neonatal patients in a single ICU. Antimicrob Resist Infect Control 1: 21.
- 18. Bjørkeng EK, Rasmussen G, Sundsfjord A, Sjöberg L, Hegstad K, et al. (2011) Clustering of polyclonal VanB-type vancomycin-resistant Enterococcus faecium in a low-endemic area was associated with CC17-genogroup strains harbouring transferable vanB2-Tn5382 and pRUM-like repA containing plasmids with axe-txe plasmid addiction systems. APMIS 119: 247–258.
- 19. Johnson PD, Ballard SA, Grabsch EA, Stinear TP, Seemann T, et al. (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.
- 20. Dahl KH, Simonsen GS, Olsvik Ø, Sundsfjord A (1999) Heterogeneity in the vanB gene cluster of genomically diverse clinical strains of vancomycin-resistant enterococci. Antimicrob Agents Chemother 43: 1105–1110.
- 21. Hegstad K, Mikalsen T, Coque TM, Jensen LB, Werner G, et al. (2010) Mobile genetic elements and their contribution to the emergence of antimicrobial resistant Enterococcus faecalis and E. faecium. Clin Microbiol Infect 16: 541–554.
- 22. Rosvoll TC, Pedersen T, Sletvold H, Johnsen PJ, Sollid JE, et al. (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.
- 23. Werner G, Willems RJ, Hildebrandt B, Klare I, Witte W (2003) Influence of transferable genetic determinants on the outcome of typing methods commonly used for Enterococcus faecium. J Clin Microbiol 41: 1499–1506.
- 24. Poyart C, Trieu-Cuot P (1994) Heterogeneric conjugal transfer of the pheromone-responsive plasmid pIP964 (IncHlyI) of Enterococcus faecalis in the apparent absence of pheromone induction. FEMS Microbiol Lett 122: 173–179.
- 25. Leavis HL, Willems RJ, Top J, Bonten MJ (2006) High-level ciprofloxacin resistance from point mutations in gyrA and parC confined to global hospital-adapted clonal lineage CC17 of Enterococcus faecium. J Clin Microbiol 44: 1059–1064.
- 26. Saeedi B, Hallgren A, Jonasson J, Nilsson LE, Hanberger H, et al. (2002) Modified pulsed-field gel electrophoresis protocol for typing of enterococci. APMIS 110: 869–874.
- 27. Leavis H, Top J, Shankar N, Borgen K, Bonten M, et al. (2004) A novel putative enterococcal pathogenicity island linked to the esp virulence gene of Enterococcus faecium and associated with epidemicity. J Bacteriol 186: 672–682.
- 28. Rice LB, Carias L, Rudin S, Vael C, Goossens H, et al. (2003) A potential virulence gene, hylEfm, predominates in Enterococcus faecium of clinical origin. J Infect Dis 187: 508–512.
- 29. 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.
- 30. Eaton TJ, Gasson MJ (2001) Molecular screening of Enterococcus virulence determinants and potential for genetic exchange between food and medical isolates. Appl Environ Microbiol 67: 1628–1635.
- 31. 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.
- 32. Hendrickx AP, van Luit-Asbroek M, Schapendonk CM, van Wamel WJ, Braat JC, et al. (2009) 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.
- 33. Hendrickx AP, Bonten MJ, van Luit-Asbroek M, Schapendonk CM, Kragten AH, et al. (2008) Expression of two distinct types of pili by a hospital-acquired Enterococcus faecium isolate. Microbiology 154: 3212–3223.
- 34. Dahl KH, Lundblad EW, Røkenes TP, Olsvik Ø, Sundsfjord A (2000) Genetic linkage of the vanB2 gene cluster to Tn5382 in vancomycin-resistant enterococci and characterization of two novel insertion sequences. Microbiology 146: 1469–1479.
- 35. Rosvoll TC, Lindstad BL, Lunde TM, Hegstad K, Aasnæs B, et al. (2012) 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 lineages. FEMS Immunol Med Microbiol 66: 166–176.
- 36. Jensen LB, Garcia-Migura L, Valenzuela AJ, Lohr M, Hasman H, et al. (2010) A classification system for plasmids from enterococci and other Gram-positive bacteria. J Microbiol Methods 80: 25–43.
- 37. Darling AE, Mau B, Perna NT (2010) progressiveMauve: multiple genome alignment with gene gain, loss and rearrangement. PLoS One 5: e11147.
- 38. 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.
- 39. Saldanha R, Chen B, Wank H, Matsuura M, Edwards J, et al. (1999) RNA and protein catalysis in group II intron splicing and mobility reactions using purified components. Biochemistry 38: 9069–9083.
- 40. el Amin NA, Jalal S, Wretlind B (1999) Alterations in GyrA and ParC associated with fluoroquinolone resistance in Enterococcus faecium. Antimicrob Agents Chemother 43: 947–949.
- 41. Werner G, Fleige C, Ewert B, Laverde-Gomez JA, Klare I, et al. (2010) High-level ciprofloxacin resistance among hospital-adapted Enterococcus faecium (CC17). Int J Antimicrob Agents 35: 119–125.
- 42. 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.
- 43. Grady R, Hayes F (2003) Axe-Txe, a broad-spectrum proteic toxin-antitoxin system specified by a multidrug-resistant, clinical isolate of Enterococcus faecium. Mol Microbiol 47: 1419–1432.
- 44. Freitas AR, Novais C, Tedim AP, Francia MV, Baquero F, et al. (2013) Microevolutionary events involving narrow host plasmids influences local fixation of vancomycin-resistance in Enterococcus populations. PLoS One 8: e60589.
- 45. Launay A, Ballard SA, Johnson PD, Grayson ML, Lambert T (2006) Transfer of vancomycin resistance transposon Tn1549 from Clostridium symbiosum to Enterococcus spp. in the gut of gnotobiotic mice. Antimicrob Agents Chemother 50: 1054–1062.
- 46. Howden BP, Holt KE, Lam MM, Seemann T, Ballard S, et al. (2013) Genomic insights to control the emergence of vancomycin-resistant enterococci. MBio 4: e00412–13.
- 47. Karki S, Land G, Aitchison S, Kennon J, Johnson PD, et al. (2013) Long term carriage of vancomycin-resistant enterococci in patients discharged from hospital: a 12-year retrospective cohort study. J Clin Microbiol 51: 3374–3379.
- 48. Johnsen PJ, Townsend JP, Bohn T, Simonsen GS, Sundsfjord A, et al. (2009) Factors affecting the reversal of antimicrobial-drug resistance. Lancet Infect Dis 9: 357–364.
- 49. Johnsen PJ, Østerhus JI, Sletvold H, Sørum M, Kruse H, et al. (2005) Persistence of animal and human glycopeptide-resistant enterococci on two Norwegian poultry farms formerly exposed to avoparcin is associated with a widespread plasmid-mediated vanA element within a polyclonal Enterococcus faecium population. Appl Environ Microbiol 71: 159–168.
- 50. Grabsch EA, Chua K, Xie S, Byrne J, Ballard SA, et al. (2008) Improved detection of vanB2-containing Enterococcus faecium with vancomycin susceptibility by Etest using oxgall supplementation. J Clin Microbiol 46: 1961–1964.
- 51. Hegstad K, Giske CG, Haldorsen B, Matuschek E, Schønning K, et al. (2014) Performance of the EUCAST disk diffusion method, the CLSI agar screen method, and the Vitek 2 automated antimicrobial susceptibility testing system for detection of clinical isolates of enterococci with low- and medium-level VanB-type vancomycin resistance: a multicenter study. J Clin Microbiol 52: 1582–1589.