SGM Journals
Research Article

ccrABEnt serine recombinase genes are widely distributed in the Enterococcus faecium and Enterococcus casseliflavus species groups and are expressed in E. faecium

Eva Katrin Bjørkeng, Girum Tadesse Tessema, Eirik Wasmuth Lundblad, Patrick Butaye, Rob Willems, Johanna Ericsson Sollid, Arnfinn Sundsfjord, Kristin Hegstad

  • 1Research Group for Host–Microbe Interactions, Department of Medical Biology, Faculty of Health Sciences, University of Tromsø, N-9037 Tromsø, Norway
  • 2Reference Centre for Detection of Antimicrobial Resistance (K-res), Department of Microbiology and Infection Control, University Hospital of North-Norway, N-9038 Tromsø, Norway
  • 3CODA-CERVA-VAR, Brussels, Belgium
  • 4Department of Pathology, Bacteriology and Poultry Diseases, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, 9820 Merelbeke, Belgium
  • 5Department of Medical Microbiology, University Medical Centre Utrecht, Heidelberglaan 100, Rm G04.614, 3584 CX, Utrecht, The Netherlands
  • Correspondence
    Kristin Hegstad
    Kristin.Hegstad{at}uit.no

    • Microbiology 2010; 156(12):3624–3634 · https://doi.org/10.1099/mic.0.041491-0

    View at publisher PubMed

    Abstract

    The presence, distribution and expression of cassette chromosome recombinase (ccr) genes, which are homologous to the staphylococcal ccrAB genes and are designated ccrABEnt genes, were examined in enterococcal isolates (n=421) representing 13 different species. A total of 118 (28 %) isolates were positive for ccrABEnt genes by PCR, and a number of these were confirmed by Southern hybridization with a ccrAEnt probe (n=76) and partial DNA sequencing of ccrAEnt and ccrBEnt genes (n=38). ccrABEnt genes were present in Enterococcus faecium (58/216, 27 %), Enterococcus durans (31/38, 82 %), Enterococcus hirae (27/52, 50 %), Enterococcus casseliflavus (1/4, 25 %) and Enterococcus gallinarum (1/2, 50 %). In the eight other species tested, including Enterococcus faecalis (n=94), ccrABEnt genes were not found. Thirty-eight sequenced ccrABEnt genes from five different enterococcal species showed 94–100 % nucleotide sequence identity and linkage PCRs showed heterogeneity in the ccrABEnt flanking chromosomal genes. Expression analysis of ccrABEnt genes from the E. faecium DO strain showed constitutive expression as a bicistronic mRNA. The ccrABEnt mRNA levels were lower during log phase than stationary phase in relation to total mRNA. Multilocus sequence typing was performed on 39 isolates. ccrABEnt genes were detected in both hospital-related (10/29, 34 %) and non-hospital (4/10, 40 %) strains of E. faecium. Various sequence types were represented by both ccrABEnt positive and negative isolates, suggesting acquisition or loss of ccrABEnt in E. faecium. In summary, ccrABEnt genes, potentially involved in genome plasticity, are expressed in E. faecium and are widely distributed in the E. faecium and E. casseliflavus species groups.

    • ↵†Present address: Nofima Mat AS, Osloveien 1, N-1430 Ås, Norway.

    • ↵‡Present address: Department of Chemistry, Biotechnology and Food Science, Norwegian University of Life Sciences, PO Box 5003, N-1432 Ås, Norway.

    • The GenBank/EMBL/DDBJ accession numbers for the ccrABEnt gene sequences from E. faecium (n=14), E. hirae (n=10 for ccrAEnt and 11 for ccrBEnt), E. durans (n=10), E. gallinarum (n=1) and E. casseliflavus (n=1) isolates are FJ572967–FJ573039.

    • A supplementary figure, showing representative real-time PCR results for the expression of five genes, and a supplementary table, showing the domains for CcrEnt proteins predicted in the Pfam database, are available with the online version of this paper.

    Edited by: K. E Weaver

    INTRODUCTION

    The emergence of multidrug-resistant hospital-acquired Enterococcus faecium as one of the most important pathogens in the developed world has been a remarkable development in the last two decades (Leavis et al., 2006; Werner et al., 2003). Molecular epidemiological studies and comparative genomic hybridization analyses of E. faecium (Leavis et al., 2007; Werner et al., 2003) have revealed genotypic differences between hospital and community isolates (Leavis et al., 2006). Mixed whole genome arrays demonstrated a distinct genetic make-up of hospital-associated E. faecium with more than 100 extra genes, possibly acquired by horizontal gene transfer (Leavis et al., 2007). The esp virulence gene, located on a putative pathogenicity island, is one of the determinants acquired by hospital-associated E. faecium. These observations, as well as current multilocus sequence typing (MLST) data, strongly indicate that gene flux and recombination contribute significantly to diversification and adaptation of E. faecium (Leavis et al., 2006; van Schaik et al., 2010).

    Recombinases facilitate the exchange of DNA fragments within and between bacteria and are thus pivotal in genome plasticity. Staphylococcal cassette chromosome (SCC) elements are vehicles for exchange of genetic information in staphylococci. These elements are characterized by the presence of terminal inverted repeats and unique recombinase genes, and are flanked by direct repeats (Ito et al., 2001, 2004; Katayama et al., 2003). So far, the major group of elements described are SCCmec I–VIII (International Working Group on the Classification of Staphylococcal Cassette Chromosome Elements, 2009) responsible for the spread of methicillin resistance between staphylococci. The movement of SCC elements is dependent on the gene products of the cassette chromosome recombinase genes (ccr), either the ccrA–ccrB complex or the single product of ccrC (Katayama et al., 2000; Noto & Archer, 2006). These proteins are serine recombinases of the resolvase/invertase family which integrate the SCC element in a site-specific manner (Ito et al., 1999). To our knowledge, ccr genes have only been reported in staphylococcal species.

    Here, we report for the first time to our knowledge, the presence of ccrAB genes in enterococci, hereby designated ccrABEnt, and show that they are expressed under standard in vitro growth conditions. Our analyses show that the ccrABEnt genes are widely distributed in Enterococcus species belonging to the E. faecium and Enterococcus casseliflavus species groups.

    METHODS

    Bacterial isolates.

    A total of 421 Enterococcus isolates of 13 species from three continents (Europe, USA and Australia) were included in the study; E. faecium (n=216), E. faecalis (n=94), E. durans (n=38), E. hirae (n=52), E. casseliflavus (n=4), E. avium (n=4), E. raffinosus (n=3), E. canintesti (n=2), E. canis (n=2), E. gallinarum (n=2), E. cecorum (n=2), E. asini (n=1) and E. dispar (n=1). Among the 216 E. faecium isolates, 72 were of human origin of which 58 were clinical isolates. Among the 94 E. faecalis isolates, 13 were of human origin of which eight were clinical isolates. Other enterococcal species included were exclusively of animal origin (poultry, dog, bovine and pig). Six ATCC strains were also included. Isolates used for phylogenetic analyses, MLST and/or PCRs to link ccrABEnt with surrounding genes are displayed in Table 1⇓.

    Table 1.

    Enterococcus isolates selected for MLST typing, phylogenetic analyses and/or PCRs to link ccrABEnt with surrounding genes

    Type of vancomycin resistance, ST or hospital-related ST (CC17 genogroup), and the presence of ccrABEnt genes is shown. nd, not determined; −, negative.

    The E. faecium ATCC 19434, E. faecalis ATCC 29212, E. gallinarum ATCC 49608, E. faecalis ATCC 19433 and E. faecalis ATCC 51575 were used as controls in species identification. All species were identified by ddl PCR (Dutka-Malen et al., 1995) or tRNA intergenic spacer PCR (Baele et al., 2000).

    Detection of ccrABEnt genes in the E. faecium DO genome by in silico analyses.

    Preliminary sequence data of the E. faecium DO strain were obtained from The Joint Genome Institute (JGI) website at (version 08.06.04). Searches for homologous proteins were performed using blast 2.0 (, on 8 February 2010) and fasta 33 (, on 8 February 2010). Translation of coding sequences (CDSs) into amino acid sequences was done using ExPASy proteomic tools (, on 8 February 2010).

    For prediction of CDSs we used ORF finder (, on 8 February 2010), Gene Mark (v2.4) (Besemer & Borodovsky, 1999), FgenesB (, on 2 August 2010) and artemis (Wellcome Trust Genome Campus, Hinxton, Cambridge, UK). Pairwise comparison and multiple sequence alignments were performed between the E. faecium CcrABEnt proteins and the previously identified four pairs of Staphylococcus aureus CcrABs (CcrAB1, CcrAB2, CcrAB3, CcrAB4; GenBank accession nos AB033763, D86934, AB037671 and AF411935) and CcrC (GenBank accession no. AB121219). Since ccrB1 and ccrB4 were truncated due to frame shift mutation, 1626 bp (ccrB1) and 1629 bp (ccrB4) ORFs were reconstituted by adding adenine to deleted positions in order to make the alignment better with the Staphylococcus hominis ccr sequence (GenBank accession no. AB063171) which has been fully sequenced (Ito et al., 2001). The comparison of DNA sequences was performed in BioEdit v.7.0.5.3 (), while multiple alignments were done using clustal w () or T-Coffee ().

    The evolutionary relationships of CcrABEnt, Ccr of staphylococci (deduced from ccrA1, ccrA2, ccrA3, ccrA4, ccrB1, ccrB3, ccrB4 and ccrC), and three other site-specific recombinases (site-specific integrase of bacteriophage φ-FC1 found in E. faecalis and two site-specific recombinases from Clostridium acetobutylicum ATCC824) were further investigated. These were included because they have been part of previous similar analyses (Ito et al., 2004) and because the ccrA and ccrB, as well as one of the recombinases from C. acetobutylicum (AE001437; locus tag no. CAC 2247), have been annotated as if they were DNA invertase Pin homologue proteins. The full-length ccrB1 of NCTC10442 and ccrB4 of HDE288 were reconstituted as described earlier (Ito et al., 2004). A neighbour-joining tree was constructed using mega3 (Kumar et al., 2004) by creating 2000 bootstrap replicates. Sites with gaps/missing data were excluded during analyses. Recombination within the sequenced regions of ccrAEnt and ccrBEnt was determined by phi test (Bruen et al., 2006).

    Protein structures were predicted using pstpred v2.4 (, on 17 December 2008) and the determinations of protein superfamilies were done using the HMM library, Genome assignment v1.65 (, on 8 February 2010), InterProscan (, on 8 February 2010) and Pfam (, on 8 February 2010). The programs EditSeq and SeqMan (dnastar) were used for sequence analysis. To detect repeat sequences, Nucleic Acid Dot Plot (, on 17 June 2010) and the Dotlet database (, on 8 February 2010) were used.

    DNA extraction, PCR amplification and DNA sequencing.

    Bacterial DNA extraction for PCR analyses was performed manually by using the InstaGene matrix kit (Bio-Rad) or the GenoM-48 robotic workstation using GenoPrep DNA from blood, standard kit (Genovision). DNA for hybridization purposes was isolated using guanidium isothiocyanate (Dahl & Sundsfjord, 2003).

    For long range PCR, 2 U DNA polymerase enzyme rTth XL (Perkin Elmer) was used per reaction and 1.4 mM Mg(OAc)2 in a standard XL PCR mix, or a 0.7× Pfu Ultra mix (Stratagene) with 2.5 U Pfu Ultra polymerase per reaction. DNA sequencing was performed using BigDye 3.1 technology (Applied Biosystems). Real-time PCR was performed using ABI Prism 7300 real-time PCR system (PE Biosystems) and TaqMan universal mastermix (Applied Biosystems).

    Detection of ccrABEnt genes and PCR linkage to surrounding genes.

    ccrABEnt genes were detected by PCR, using the primer pairs FA–RA and FB–RB, respectively (Table 2⇓), and genes in selected isolates were detected by Southern hybridization and DNA sequencing. PCRs were also performed on 13 of 14 ccrABEnt-positive E. faecium isolates selected for MLST as well as two ccrABEnt-positive E. faecium animal isolates from Norway, to search for the presence and conservation of gene synteny in the surrounding genes (Table 2⇓ and Fig. 1a⇓). Primers and probes were designed using E. faecium DO sequences as template.

    Figure image not available in archive
    Fig. 1.

    (a) Schematic presentation of the ccrABEnt region of E. faecium DO and the long-range PCRs used to link genes surrounding the ccrAEnt and ccrBEnt genes in E. faecium. (b) Schematic presentation of ccrABEnt, indicating the positions of the PCR primers used for mRNA linkage. Linkage of ccrAEnt and ccrBEnt mRNAs was performed using combinations of primers ccrAxF/ccrBRTR1, ccrAxF/CcrBxR, CcrARTF1/ccrBRTR1 and CcrARTF1/ CcrBxR.

    Table 2.

    Oligonucleotides used for expression analyses of ccrABEnt genes and for detection/characterization of the ccrABEnt region and detection of enterococcal virulence genes

    Expression analysis of ccrABEnt genes by real-time quantitative PCR.

    To analyse if ccrABEnt genes are expressed, E. faecium DO was grown aerobically in BHI broth at 37 °C for 18–24 h. Subsequently the culture was diluted 1 : 50 in BHI broth and grown with agitation to OD600 0.7 or to stationary phase (grown overnight). The cell suspension was centrifuged and the cells were immediately frozen on dry ice or liquid nitrogen before adding an RNA stabilizing solution, RNA later (Ambion). Alternatively, RNA later or RNA protect (Qiagen) was added directly to the inoculum, according to the manufacturer's instructions. RNA extraction was performed by using the RNeasy mini kit (Qiagen) using a prolonged lysis step of 45 min with 10 mg lysozyme and 10 U mutanolysin in a total volume of 100 μl. On-column DNase treatment was performed according to the manufacturer's instructions. A successive removal of DNA was performed using Turbo DNase (Ambion) according to the manufacturer's instructions. RNA integrity was determined by agarose gel electrophoresis. Reverse transcription of the total RNA was performed using the ABRTR1 primer and the High Capacity cDNA Reverse Transcription kit (Applied Biosystems) or Superscript III RNase H-reverse transcriptase (Invitrogen). Real-time PCR was performed on the cDNA using primers ccrAFre, ccrARre, ccrBFre, ccrBRre, recAFre, recARre, pbp5Fre, pbp5Rre, adkFre and adkRre, and probes ccrAEnt, ccrBEnt, recA, pbp5 and adk (Table 2⇑). Expression of ccrABEnt genes was compared with the expression of recA, pbp5 and adk. Ten-fold serial dilutions of E. faecium DO genomic DNA were used to make standard curves to determine PCR efficiency, using the equation: E=10(−1/slope)−1. The PCR efficiencies ranged from 88 to 104 % in one assay and 99 to 100 % in a second assay and were considered similar enough to be able to compare only Ct (threshold cycle) values for a semiquantitative relative measurement of expression. The expression experiments were performed in three triplicates; a no template control (NTC) and a minus reverse transcriptase control (−RT) was included in each assay. The −RT controls were in the range of an acceptable difference from the cDNA expression analysis (>5Ct difference).

    Analysis of ccrABEnt mRNA linkage by RT-PCR.

    RNA isolation was performed as described above. RNA was treated with the DNA-free kit (Ambion). Reverse transcription of total RNA was performed with SuperScript III reverse transcriptase (Invitrogen) using primers CcrBRTR1 or CcrBxR. RT-PCR without reverse transcriptase was performed on total RNA to check for DNA contamination. Linkage of ccrAEnt and ccrBEnt mRNAs as a bicistronic mRNA was analysed by PCRs on cDNAs using primers located in ccrAEnt (CcrARTR1 and CcrAxF) and ccrBEnt (CcrBRTR1 and CcrBxR) (Fig. 1b⇑ and Table 2⇑).

    Southern blot hybridization analyses.

    RFLP with XbaI (Promega) was performed on total genomic DNA for selected E. faecium isolates (DO, TUH7-55, E0470, E0745, E1304 and E1293). PFGE of SmaI-digested DNA from 76 E. faecium isolates was performed as described by Dahl et al. (1999). DNA fragments separated by gel electrophoresis were transferred to a positively charged nylon membrane (Boehringer) by vacuum blotting using a Vacugene XL system (Amersham Biosciences). Southern blot hybridization was performed with a DIG-labelled (Boehringer) ccrAEnt probe based on E. faecium DO.

    MLST.

    MLST was performed on a subset of isolates using the following primers: adk1n, adk2n, atpA1n, atpA2n, ddl1, ddl2, gdh1, gdh2, gyd-1, gyd2, pstS1n, pstS2, purK1n and purK2n (Homan et al., 2002).

    RESULTS AND DISCUSSION

    ccrABEnt sequences in the E. faecium DO genome

    Genes similar to the ccrA and ccrB genes of S. aureus (GenBank accession no. D86934) were identified in the draft sequence of the E. faecium DO genome. blast searches indicated two CDSs (locus tag nos 2319 and 2398) in E. faecium DO contig 655 (version 08.06.04) similar and in an identical order to the staphylococcal ccrA and ccrB. They were named ccrAEnt and ccrBEnt. No available reports have previously shown ccrAEnt and ccrBEnt genes in enterococci. The ccrAEnt and ccrBEnt CDSs are 1374 bp and 1638 bp in size, respectively. The two ccrABEnt genes in E. faecium DO were similar in length to the staphylococcal ccrAB2 (Katayama et al., 2000).

    The ccrABEnt gene synteny was confirmed to be the same as in staphylococci (Katayama et al., 2000) for 14 of 15 E. faecium isolates by linkage PCR (Table 3⇓). No available results have previously shown whether staphylococcal ccrA and ccrB genes are transcribed as separate units or as a bicistronic mRNA. RT-PCR analysis of total RNA from E. faecium DO revealed that the ccrABEnt genes were transcribed as a bicistronic mRNA, confirming the bioinformatics results. Knowing the function of ccrAB in staphylococci, we hypothesize that ccrABEnt genes in enterococci might be part of a larger integrative genetic element in E. faecium. The GC content of E. faecium DO contig 655 (35 %), the ccrABEnt CDSs (35 %) and the whole genome (38 %) is not substantially different. No putative termini (repeats) were identified in contig 655 by nucleic acid dot plot or DotLet analyses. Thus it was not possible to identify a putative integrative element. The genome sequence of contig 655 is limited to the tnp transposase determinant (Fig. 1a⇑) at the left side and it has not been possible to identify the continuation of this sequence in another DO contig. The sequence at the other side of ccrABEnt also contains a lot of putative transposases (belonging to several insertion sequence families) in addition to hypothetical proteins (, on 21 June 2010) which may well be part of an integrative element.

    Table 3.

    Long-range linkage PCR results for the ccrAEnt and ccrBEnt chromosomal region among 15 ccrABEnt-positive E. faecium isolates

    +, Positive; −, negative; na, not applicable (one of the genes/CDSs not present); nd, not determined.

    Pairwise comparison and multiple sequence alignments were performed between the E. faecium CcrABEnt proteins and the Ccr proteins of S. aureus. The similarities of CcrA and CcrB between E. faecium and S. aureus N315 were 55 and 69 %, respectively. The N-terminal resolvase and recombinase domains, as well as the predicted catalytic serine residue of the recombinase active site were highly conserved between the Staphylococcus and Enterococcus CcrAB proteins. Moreover, the Enterococcus CcrBEnt was predicted to include an Ogr/delta-like domain (a phage transcription activator). Two algorithms, Pfam and ProScan, predicted both the resolvase and recombinase domains in the examined Ccr protein sequences (Supplementary Table S1, available with the online version of this paper).

    The evolutionary relationships of CcrABEnt, Ccr of staphylococci and three other site-specific recombinases were further investigated. The phylogenetic analyses revealed an evolutionary relationship between CcrAEnt and CcrBEnt from enterococci and the staphylococcal CcrAB cluster (Fig. 2⇓). However, the low identity score between the enterococcal and staphylococcal proteins does not support a recent horizontal transfer of the ccr genes between these species.

    Figure image not available in archive
    Fig. 2.

    Phylogram for CcrAEnt, CcrBEnt, other Ccrs, and three site-specific recombinase proteins. The deduced amino acid sequences of the following genes were used: ccrA1 and ccrB1* (from NCTC10442, GenBank accession no. AB033763); ccrA2 and ccrB2 (from N315, GenBank accession no. D86934); ccrA3 and ccrB3 (from 85/2082, GenBank accession no. AB037671); ccrA4 and ccrB4* (from HDE288, GenBank accession no. AF411935); ccrC [from JSCC 3624 (WIS), GenBank accession no. AB121219]; site-specific integrase (from phi-FC1, GenBank accession no. AF124258); and two site-specific recombinases (from C. acetobutylicum ATCC824, GenBank accession no, AE001437, locus tag nos CAC 1228 and CAC 2247). The scale bar indicates genetic distance in substitutions per site. The Ccr clusters are circled. The amino acid sequences were aligned using T-Coffee. The neighbour-joining phylogenetic tree was constructed with mega3 from 2000 bootstrap replicates using the P-distance model. The dataset consisted of 447 amino acids with 37 parsimony-informative sites for CcrA and 547 amino acids with 50 parsimony-informative sites for CcrB.

    ccrABEnt genes are expressed in E. faecium

    Analyses of ccrABEnt gene expression were performed during both the exponential and stationary phase of E. faecium DO grown in rich medium. Both genes were expressed in approximately the same amounts in exponential phase. ccrABEnt genes were expressed >70-fold lower than the pbp5, recA and adk genes (Supplementary Fig. S1). The mRNA abundance of ccrABEnt was lower in stationary phase than in exponential phase.

    ccrABEnt genes are dispersed among Enterococcus species belonging to the E. faecium and E. casseliflavus species groups

    Of a total of 421 enterococcal isolates, 118 (28 %) were positive for ccrABEnt genes in five species by PCR; E. faecium (58/216, 27 %), E. durans (31/38, 82 %), E. hirae (27/52, 50 %), E. casseliflavus (1/4, 25 %) and E. gallinarum (1/2, 50 %) (Table 1⇑). One E. hirae isolate was positive by PCR for ccrBEnt only. Eight other species including E. faecalis were negative for ccrABEnt (data not shown).

    A blast search for the ccrABEnt genes and the surrounding regions against Enterococcus strains revealed the presence of ccrABEnt in E. faecium E1071, 1,231,408 and C68 (, on 21 June 2010) and no ccrABEnt sequence or protein matches with high identity scores in other available Enterococcus genomes (, on 21 June 2010). E. faecium E1071 and E. faecium 1,231,408 showed sequence similarity with the DO sequence in parts of the hypothetical protein, ccrBEnt and parts of ccrAEnt. E. faecium C68 showed similarity with DO in parts of the hypothetical protein, both ccrAEnt and ccrBEnt and parts of the replication initiation factor (REP factor).

    ccrABEnt gene sequences (GenBank accession nos FJ572967–FJ573039) from E. faecium (n=14), E. hirae (n=10 for ccrAEnt and 11 for ccrBEnt), E. durans (n=10), E. gallinarum (n=1) and E. casseliflavus (n=1) isolates were aligned and a neighbour-joining phylogenetic tree was made with 2000 bootstrap replicates using the P-distance model (Fig. 3⇓). The ccrABEnt genes both clustered into two major clades represented by the majority of E. faecium (clade I) and E. hirae (clade II) isolates, respectively. With 7 of 10 isolates clustering in clade II, E. hirae appears to be slightly more dispersed between the two ccrAEnt clades. ccrABEnt from the E. gallinarum and E. casseliflavus isolates clustered in clade II with the majority of ccrABEnt from the E. hirae isolates. In E. durans, 6 of 10 ccrAEnt genes clustered in clade I, while 7 of 10 ccrBEnt clusters were in clade II. Except for ccrAEnt from E. faecium E1304, the ccrABEnt genes of the human isolates clustered in clade I whereas the animal isolates were found in both clades. Incongruence between ccrAEnt and ccrBEnt phylogenies within an isolate was noted for 11 isolates, all of animal origin (Fig. 3⇓, isolates marked with asterisks). Phi tests revealed no statistically significant evidence for recombination within the sequenced regions of the ccrAEnt and ccrBEnt genes. However, the incongruence suggests recombination of the ccrEnt genes outside the sequenced regions of the two genes. Incongruence between these genes has also been seen for S. aureus (Ito et al., 2004).

    Figure image not available in archive
    Fig. 3.

    Phylogram for ccrAEnt and ccrBEnt genes. The nucleotide sequences for ccrAEnt and ccrBEnt genes from E. faecium (n=15), E. hirae (n=10/11), E. durans (n=10), E. casseliflavus (n=1) and E. gallinarum (n=1) were used (GenBank accession nos FJ572967–FJ573039). Upper case letters in parentheses represent the origin of the isolate: (C), chicken; (H), human; (D), dog; (P), pig; and (A), unknown animal origin. Outgroups are represented by S. aureus N315 ccrA and ccrB (GenBank accession no. D86934). The asterisks indicate isolates in which ccrAEnt and ccrBEnt belong to different clades. All sequences were aligned using clustal w. The neighbour-joining phylogenetic tree was made with mega4.0 using 2000 bootstrap replicates and the P-distance model. Bootstrap values higher than 80 % are shown at the branches. The scale bar indicates genetic distance in substitutions per site. The two main clades of ccrAEnt and ccrBEnt are indicated. The dataset consisted of 547 nt with 494 parsimony-informative sites for ccrAEnt and 513 nt with 227 parsimony-informative sites for ccrBEnt.

    ccrABEnt genes were only found in isolates belonging to the E. faecium and E. casseliflavus species groups that belong to the same tree branch in phylogenetic trees based on enterococcal 16S and sodA gene diversity (Devriese et al., 1993; Poyart et al., 2000). The absence of ccrABEnt in the other species could be explained by the low number of isolates tested, except for E. faecalis, or by a lack of integration sites recognized by ccrABEnt in the strains not belonging to the E. faecium or E. casseliflavus groups. Alternatively, their ccrABEnt genes may exhibit such a low sequence identity to the ccrABEnt genes identified in this study that they are missed using the PCR and hybridization conditions used in the present study.

    Variations of the ccrABEnt genes and the surrounding region between selected E. faecium isolates

    PFGE analysis and Southern hybridization of 76 E. faecium isolates with the ccrAEnt probe confirmed the PCR results. One ccrAEnt PCR-negative strain (399/F98/A1) was ccrAEnt-positive by Southern blot hybridization (data not shown) indicating that sequence diversity affects PCR amplification. Also, XbaI analyses of ccrAEnt and ccrBEnt genomic regions revealed heterogeneity and only one copy of ccrAEnt. The ccrAEnt probe hybridized to an approximately 10 kb fragment in DO, TUH7-55, E1304 and E1293 isolates; however, the ccrAEnt-positive fragment of E0470 and E0745 was approximately 24 kb (data not shown). To investigate this in more detail, the presence of ccrABEnt flanking genomic genes identified in the DO genome was determined by multiple PCRs in 15 ccrABEnt-positive and 16 ccrABEnt-negative isolates (Fig. 1a⇑). Examinations of the ccrABEnt surrounding region in several isolates showed a variable pattern of the ccrABEnt flanking sequences with hospital-associated isolates showing most sequence similarity with the DO sequence (Table 3⇑). All 31 isolates were positive for the tnp gene-specific PCR (tnp belongs to the IS30 family) as well as for orf1 PCR and three ccrABEnt-positive isolates of different sequence types (STs) were also positive for the REP factor gene PCR. This REP factor gene harbours a REP_trans domain belonging to superfamily pfam02486. This family represents probable topoisomerases that makes a sequence-specific single stranded nick in the origin of replication. Plasmid REPs, phage REPs (RstAs) and transposon REPs (Cro/CI transcriptional regulators) belong to this family. Long-range PCRs confirmed linkage of these genes with ccrABEnt and conservation of gene synteny surrounding ccrABEnt with the exception of isolates 64/F99/H6, 399/F99/A10, 399/F99/H8, and S399/F99/A14, for which linkage of tnp–orf1 and orf1–ccrBEnt was not confirmed. Furthermore, a ccrBEnt–ccrAEnt linkage was not shown in 64/F99/H6 (Table 3⇑ and Fig. 1a⇑). The inability to link genes that were positive on gene-specific PCRs may indicate that the region between these genes is larger than expected or that the specific genes are located at other regions in the genome. The transposase of the IS30 family is, for instance, located at more than one site in E. faecium DO. Annotation of contig 655 (, on 21 June 2010) also indicates that the ccrABEnt genes are located in a region containing several transposases. The regions surrounding ccrAB in staphylococci contain highly variable genes encoding ORFs of unknown functions. These variable regions are called J1 and J2, and variations in these regions are used to define the SCCmec subtypes (International Working Group on the Classification of Staphylococcal Cassette Chromosome Elements, 2009), and so our results from enterococci are in line with these observations of highly variable regions surrounding ccrAB in staphylococci. CcrA and CcrB have roles in the excision and integration of SCCmec in staphylococci (Wang & Archer, 2010) and we have showed that the ccrABEnt genes are expressed in E. faecium DO. It has been postulated that SCC may carry the genes conferring methicillin resistance but may also enable genetic exchange of other genes among staphylococcal species (Katayama et al., 2003). However, to our knowledge, no studies have provided direct experimental evidence for intercellular transfer of SCC between staphylococci.

    DNA sequencing of the ccrABEnt, tnp and orf1 of the 15 ccrABEnt-positive isolates showed 94–100 % and 96–100 % sequence identity in ccrAEnt and ccrBEnt genes (GenBank accession nos FJ572978–FJ572981, FJ572997–FJ573001, FJ573014–FJ573018, FJ573032–FJ573036), respectively, while sequences of orf1 and tnp were 100 % identical in all isolates (data not shown). According to Hanssen et al. (2004), up to 4 % variation within the ccrAB genes has been observed for a given staphylococcal species. The ccrAB genes found in SCCmec types II and IV can vary up to 5 % at the nucleotide level (Noto & Archer, 2006). Since both ccrABEnt genes and the staphylococcal ccrAB genes show sequence variations within the recombinase genes, which have the same gene synteny and variable surrounding regions, we hypothesize that they may have similar functions in contributing to excision and integration of surrounding genes within the genome and possibly also mobilization of surrounding genes between cells.

    Investigation of possible association between ccrABEnt and ST within E. faecium of human origin

    MLST analyses of E. faecium isolates (n=39) revealed that the ccrABEnt genes are dispersed among different STs (Table 1⇑). Ten of 29 (34 %) hospital-related E. faecium isolates were ccrABEnt-positive, while 4 of 10 (40 %) non-hospital-related isolates were positive. Furthermore, specific STs within hospital-related strains were represented by both ccrABEnt-positive and -negative isolates (Table 1⇑), suggesting that ccrABEnt genes are acquired and not a part of the core genome.

    Concluding remarks

    Cassette chromosome recombinases may be important in recombination and genome plasticity in enterococci. Expression analyses indicate that the recombinase genes are active in E. faecium DO and thus, may play a role in the recombination or movement of genetic elements. Further investigation of the ccrAEnt and ccrBEnt will be essential to reveal the contribution of these genes for recombination and mobilization events in enterococci.

    Acknowledgments

    This work was supported by research grants from the Norwegian Research Council (projects no. 165997 and 183653/S19), Northern Norway regional Health Authority Medical Research Program and the European Commission (LSHE-CT-2007-03410 ‘ACE’). The DNA sequence of E. faecium DO was supported at the JGI by the US Department of Energy's Office of Science, Biological and Environmental Research Program, and by the University of California, Lawrence Berkeley National Laboratory under contract no. DE-AC02-05CH11231, Lawrence Livermore National Laboratory under contract no. DE-AC52-07NA27344 and Los Alamos National Laboratory under contract no. DE-AC02-06NA25396. We would like to thank Professor Barbara E. Murray, University of Texas Medical School at Houston, USA, for kindly providing bacterial isolates. We also thank Dr Stein-Christian Mohn at the University of Bergen for his help and advice, and Trine Tessem at the University of Tromsø for her excellent technical assistance.

    References

    1. Arduino, R. C., Murray, B. E. & Rakita, R. M. (1994). Roles of antibodies and complement in phagocytic killing of enterococci. Infect Immun 62, 987–993.
    2. Baele, M., Baele, P., Vaneechoutte, M., Storms, V., Butaye, P., Devriese, L. A., Verschraegen, G., Gillis, M. & Haesebrouck, F. (2000). Application of tRNA intergenic spacer PCR for identification of Enterococcus species. J Clin Microbiol 38, 4201–4207.
    3. Besemer, J. & Borodovsky, M. (1999). Heuristic approach to deriving models for gene finding. Nucleic Acids Res 27, 3911–3920.
    4. Bruen, T. C., Philippe, H. & Bryant, D. (2006). A simple and robust statistical test for detecting the presence of recombination. Genetics 172, 2665–2681.
    5. Carias, L. L., Rudin, S. D., Donskey, C. J. & Rice, L. B. (1998). Genetic linkage and cotransfer of a novel, vanB-containing transposon (Tn5382) and a low-affinity penicillin-binding protein 5 gene in a clinical vancomycin-resistant Enterococcus faecium isolate. J Bacteriol 180, 4426–4434.
    6. Dahl, K. H. & Sundsfjord, A. (2003). Transferable vanB2 Tn5382-containing elements in fecal streptococcal strains from veal calves. Antimicrob Agents Chemother 47, 2579–2583.
    7. Dahl, K. H., Simonsen, G. S., 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.
    8. Devriese, L. A., Pot, B. & Collins, M. D. (1993). Phenotypic identification of the genus Enterococcus and differentiation of phylogenetically distinct enterococcal species and species groups. J Appl Bacteriol 75, 399–408.
    9. 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.
    10. Hanssen, A. M., Kjeldsen, G. & Sollid, J. U. E. (2004). Local variants of staphylococcal cassette chromosome mec in sporadic methicillin-resistant Staphylococcus aureus and methicillin-resistant coagulase-negative staphylococci: evidence of horizontal gene transfer? Antimicrob Agents Chemother 48, 285–296.
    11. Homan, W. L., Tribe, D., Poznanski, S., Li, M., Hogg, G., Spalburg, E., van Embden, J. & Willems, R. (2002). Multilocus sequence typing scheme for Enterococcus faecium. J Clin Microbiol 40, 1963–1971.
    12. International Working Group on the Classification of Staphylococcal Cassette Chromosome Elements (2009). Classification of staphylococcal cassette chromosome mec (SCCmec): guidelines for reporting novel SCCmec elements. Antimicrob Agents Chemother 53, 4961–4967.
    13. Ito, T., Katayama, Y. & Hiramatsu, K. (1999). Cloning and nucleotide sequence determination of the entire mec DNA of pre-methicillin-resistant Staphylococcus aureus N315. Antimicrob Agents Chemother 43, 1449–1458.
    14. Ito, T., Katayama, Y., Asada, K., Mori, N., Tsutsumimoto, K., Tiensasitorn, C. & Hiramatsu, K. (2001). Structural comparison of three types of staphylococcal cassette chromosome mec integrated in the chromosome in methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother 45, 1323–1336.
    15. Ito, T., Ma, X. X., Takeuchi, F., Okuma, K., Yuzawa, H. & Hiramatsu, K. (2004). Novel type V staphylococcal cassette chromosome mec driven by a novel cassette chromosome recombinase, ccrC. Antimicrob Agents Chemother 48, 2637–2651.
    16. Johnsen, P. J., Østerhus, J. I., Sletvold, H., Sørum, M., Kruse, H., Nielsen, K., Simonsen, G. S. & Sundsfjord, A. (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.
    17. Katayama, Y., Ito, T. & Hiramatsu, K. (2000). A new class of genetic element, Staphylococcus cassette chromosome mec, encodes methicillin resistance in Staphylococcus aureus. Antimicrob Agents Chemother 44, 1549–1555.
    18. Katayama, Y., Takeuchi, F., Ito, T., Ma, X. X., Ui-Mizutani, Y., Kobayashi, I. & Hiramatsu, K. (2003). Identification in methicillin-susceptible Staphylococcus hominis of an active primordial mobile genetic element for the staphylococcal cassette chromosome mec of methicillin-resistant Staphylococcus aureus. J Bacteriol 185, 2711–2722.
    19. Kumar, S., Tamura, K. & Nei, M. (2004). mega3: integrated software for Molecular Evolutionary Genetics Analysis and sequence alignment. Brief Bioinform 5, 150–163.
    20. Leavis, H. L., Bonten, M. J. & Willems, R. J. (2006). Identification of high-risk enterococcal clonal complexes: global dispersion and antibiotic resistance. Curr Opin Microbiol 9, 454–460.
    21. Leavis, H. L., Willems, R. J., van Wamel, W. J., Schuren, F. H., Caspers, M. P. & Bonten, M. J. (2007). Insertion sequence-driven diversification creates a globally dispersed emerging multiresistant subspecies of E. faecium. PLoS Pathog 3, e7.
    22. Noto, M. J. & Archer, G. L. (2006). A subset of Staphylococcus aureus strains harboring staphylococcal cassette chromosome mec (SCCmec) type IV is deficient in CcrAB-mediated SCCmec excision. Antimicrob Agents Chemother 50, 2782–2788.
    23. 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.
    24. Poyart, C., Quesnes, G. & Trieu-Cuot, P. (2000). Sequencing the gene encoding manganese-dependent superoxide dismutase for rapid species identification of enterococci. J Clin Microbiol 38, 415–418.
    25. van Schaik, W., Top, J., Riley, D., Boekhorst, J., Vrijenhoek, J., Schapendonk, C., Hendrickx, A., Nijman, I., Bonten, M. & other authors (2010). Pyrosequencing-based comparative genome analysis of the nosocomial pathogen Enterococcus faecium and identification of a large transferable pathogenicity island. BMC Genomics 11, 239.
    26. Wang, L. & Archer, G. L. (2010). Roles of CcrA and CcrB in excision and integration of staphylococcal cassette chromosome mec, a Staphylococcus aureus genomic island. J Bacteriol 192, 3204–3212.
    27. Werner, G., Willems, R. J. L., 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.
    28. Willems, R. J., Top, J., van Santen, M., Robinson, D. A., Coque, T. M., Baquero, F., Grundmann, H. & Bonten, M. J. (2005). Global spread of vancomycin-resistant Enterococcus faecium from distinct nosocomial genetic complex. Emerg Infect Dis 11, 821–828.