Genetic linkage of the vanB2 gene cluster to Tn5382 in vancomycin-resistant enterococci and characterization of two novel insertion sequences

Kristin H. Dahl1, Eirik W. Lundblad1, Torunn P. Røkenes1, Ørjan Olsvik1,2 and Arnfinn Sundsfjord1,2

Department of Medical Microbiology, University and University Hospital of Tromsø, N-9037 Tromsø, Norway1
Norwegian Institute for Gene Ecology, N-9037 Tromsø, Norway2

Author for correspondence: Kristin H. Dahl. Tel: +47 77 64 57 62. Fax: +47 77 64 53 50. e-mail: kristind{at}fagmed.uit.no


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
VanB-type vancomycin resistance is encoded by the vanB gene cluster, which disseminates by horizontal gene transfer and clonal spread of vancomycin-resistant enterococci (VRE). Genetic linkage of the vanB gene cluster to transposon Tn5382 and the insertion sequences IS16 and IS256-like has previously been shown. In this study linkage of defined vanB gene cluster subtypes to these elements was examined. All the vanB2 subtype strains studied (n=14) revealed co-hybridization of vanB and Tn5382, whereas the strains of vanB1 (n=8) and vanB3 (n=1) subtypes were Tn5382 negative. Conjugative cotransfer of the vanB2 gene cluster and Tn5382 was demonstrated for two strains. DNA sequencing of the vanXB–ORFC region in vanB2 strains confirmed that the vanB2 gene cluster is an integral part of Tn5382. No general pattern of linkage was observed with regard to IS16 and IS256-like. Two novel insertion sequences were identified in specific vanB2 subtype strains. (i) A 1611 bp element (ISEnfa110) was detected in the left flank of Tn5382. Its insertion site, lack of terminal inverted and direct repeats, and two conserved motifs in its putative transposase all conform to the conventions of the IS110 family. (ii) A 787 bp element (ISEnfa200) was detected in the vanSBvanYB intergenic region. Its ORF encoded a putative protein with 60–70% identity to transposases of the IS200 family. No further copies of ISEnfa110 were found by colony hybridization of 181 enterococcal isolates, whereas ISEnfa200 was found in four additional vanB2 strains from the USA. The five strains had identical ISEnfa200 element insertion sites, and Tn5382 was located downstream from a pbp5 gene conferring high-level ampicillin resistance. These isolates showed related PFGE patterns, suggesting possible clonal spread of a VRE strain harbouring a Tn5382vanB2–ISEnfa200 element linked to a pbp5 gene conferring ampicillin resistance.


View this table:
[in this window]
[in a new window]
 
Table 3. GenBank and GenPept sequence accession numbers

 
Keywords: Enterococcus, insertion element, Tn5382, vanB

Abbreviations: IS, insertion sequence; VRE, vancomycin-resistant enterococci

GenBank and GenPept accession numbers are given in Table 3.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Enterococci are part of the normal faecal flora of humans and animals and have emerged as important nosocomial pathogens. In addition to their many inherent resistance traits, enterococci are able to obtain resistance by acquisition of new DNA (Leclercq, 1997 ). Insertion sequences (ISs) are small mobile genetic elements that can be involved in acquisition and dissemination of virulence or antimicrobial-resistance genes, as well as genes involved in catabolic pathways. Transposition of ISs has been shown to generate silent mutations, gene inactivation and activation of downstream genes. To limit the potentially detrimental effects of excessive genome rearrangements, transposition is generally kept at low levels (Galas & Chandler, 1989 ). ISs are grouped into families on the basis of similarities in genetic organization, conservation of domains or motifs in their presumptive transposase sequences, their terminal nucleotide sequences and generation of target duplications (Mahillon & Chandler, 1998 ). Homologous ISs can mobilize the intervening DNA, creating structures termed composite or compound transposons (Galas & Chandler, 1989 ; Quintiliani & Courvalin, 1996 ). The importance of composite transposons in the spread of antimicrobial resistance determinants in enterococci has been emphasized by the description of a large composite chromosomal element Tn5385 in Enterococcus faecalis conferring multiple antimicrobial resistance (Rice & Carias, 1998 ).

Acquired glycopeptide resistance in enterococci is phenotypically and genotypically heterogeneous, with VanA and VanB phenotypes being the most commonly encountered forms (Arthur et al., 1993 ; Quintiliani & Courvalin, 1996 ). Their mechanism of resistance involves the synthesis of peptidoglycan precursors that terminate in D-lactate instead of D-alanine, thus reducing the affinity for glycopeptide antibiotics (Evers & Courvalin, 1996 ). The vanA gene cluster is located on the non-conjugative transposon Tn1546 or related elements, which can be part of large mobile chromosomal elements as well as non-conjugative and conjugative plasmids (Arthur et al., 1993 ; Handwerger & Skoble, 1995 ; Handwerger et al., 1995 ). Dissemination of the vanB gene cluster appears to result from conjugation of plasmids (Woodford et al., 1995 ) or intercellular transfer of chromosomal elements (Carias et al., 1998 ; Quintiliani & Courvalin, 1996 ) as well as clonal spread. The vanB gene cluster has not been universally linked to a specific mobile genetic element. Recently an approximately 27 kb putative conjugative transposon Tn5382 containing the vanB gene cluster was described. Intercellular transfer of Tn5382 as an integral part of a larger chromosomal element also conferring ampicillin resistance was documented (Carias et al., 1998 ), but transposition of Tn5382 itself has not been shown. Intracellular transposition of the vanB gene cluster as part of the composite transposon Tn1547 flanked by insertion sequence elements IS16 and IS256-like to a conjugative plasmid, and interspecies transfer of plasmids or large chromosomal elements containing Tn1547, have also been described (Quintiliani & Courvalin, 1996 ). The epidemic potential of conjugative plasmids containing vanB is indicated by the presence of transferable vanB-encoding plasmids in unrelated clinical Enterococcus faecium and E. faecalis strains (Woodford et al., 1995 ).

Based on sequence differences the vanB gene cluster can be divided into at least three distinct subtypes: vanB1, vanB2 and vanB3 (Dahl et al., 1999 ; Patel et al., 1998 ). In this study we report that the vanB2 subtype gene cluster is an integral part of the putative conjugative transposon Tn5382. Two novel ISs, ISEnfa110 and ISEnfa200, were found in specific vanB2 subtype strains of vancomycin-resistant enterococci (VRE). These IS elements were characterized with respect to their DNA sequence and distribution in enterococci.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Bacterial strains.
The 23 VanB-type strains (Dahl et al., 1999 ) included in the study are listed in Table 1. E. faecalis BM4281 containing Tn1547 (Quintiliani & Courvalin, 1996 ) was vanB subtyped but not included in the other analyses. When screening for ISEnfa200- and ISEnfa110-like elements, the following strains were included: (i) 63 Enterococcus strains from the European VRE study (Schouten et al., 1999 ) – 29 E. faecium, 33 E. faecalis and one Enterococcus gallinarum, from England, Germany, Hungary, Italy, Poland, Portugal and Spain; (ii) 81 VanA-type VRE strains (Simonsen et al., 2000 ) from diverse ecological and geographical sources – 75 E. faecium, three Enterococcus hirae, two Enterococcus durans and one E. faecalis, from England, France, Germany, Ireland, the Netherlands, Norway and the USA; and (iii) 14 reference strains – E. faecalis ATCC 51575, ATCC 49757, ATTC 29212 and ATCC 19433, E. faecium ATCC 51559 and ATCC 19434, E. gallinarum ATCC 49608, E. casseliflavus ATCC 25788, E. saccharolyticus ATCC 43076, E. avium ATCC 14025, E. durans ATCC 19432, E. flavescens ATCC 49996, E. hirae ATCC 8043 and E. raffinosus ATCC 49427.


View this table:
[in this window]
[in a new window]
 
Table 1. VanB-type Enterococcus strains used in this study and results from vanB gene cluster subtyping and Tn5382 hybridization

 
Conjugation experiments.
Conjugation was done by filter mating (Trieu-Cuot et al., 1988 ) with minor modifications. A 100 µl sample of mating mixture containing donor strains TUH2-18 or TUH7-15 and the recipient strain E. faecium BM4105-RF (Poyart & Trieu-Cuot, 1994 ) was spread on a 45 µm nitrocellulose membrane filter (Millipore) placed on top of brain heart infusion (BHI) agar. After 18 h incubation at 37 °C, cells were resuspended in 1 ml BHI broth and spread on BHI agar containing 8 µg vancomycin ml-1, 20 µg rifampicin ml-1 and 10 µg fucidic acid ml-1.

Preparation of genomic DNA.
DNA for genomic digestion was isolated using the QIAGEN Blood & Cell Culture DNA kit. Bacterial DNA for PCR was prepared with a Dynabeads DNA DIRECT Kit (Dynal) as described by Dahl et al. (1999) or by isolation of total DNA using the GenomicPrep Cells and Tissue DNA Isolation Kit (Pharmacia Biotech) with modification of the lysis step. The bacterial cells were incubated in fresh lysis solution (10 mM Tris/HCl pH 8·0, 1 mM EDTA, 25%, w/v, sucrose, 10 mg lysozyme ml-1 and 0·1 mg RNase A ml-1) at 37 °C for 35 min to achieve optimal lysis.

DNA transfer and hybridization.
Southern transfer of digested genomic DNA to a positively charged nylon membrane (Boehringer Mannheim) was carried out by vacuum blotting (Vacugene XL system, Pharmacia Biotech) according to the manufacturer’s instructions. Colony blotting was performed as described by Sambrook et al. (1989) with the following modification: the nylon membrane was incubated on a 3 MM paper saturated with fresh lysis solution (10 mM Tris/HCl, 1 mM EDTA, 25%, w/v, sucrose, 10 mg lysozyme ml-1) for 60 min at 37 °C to ensure complete lysis of the enterococci. DNA was fixed to the nylon membrane by UV cross-linking. Probes were labelled using the PCR DIG Probe Synthesis Kit (Boehringer Mannheim) and purified by agarose gel electrophoresis followed by extraction by QIAquick Gel Extraction Kit (QIAGEN). Hybridization was carried out at 68 °C and detection performed using the DIG Luminescent Detection Kit (Boehringer Mannheim). All protocols were performed according to the manufacturers’ instructions.

Total DNA from the following bacteria was used as templates for probe synthesis: the vanB probe came from E. faecalis V583 (Evers et al., 1994 ), the Tn5382 and pbp5 probes from E. faecium C68 (Carias et al., 1998 ), the IS16 and IS256 probes from E. faecalis BM4281 (Quintiliani & Courvalin, 1996 ), the ISEnfa110 probe from TUH4-67, the ISEnfa200 probe from TUH7-15, and the 16S rDNA probe from E. faecalis DS16C2 (Franke & Clewell, 1981 ). The 16S rRNA genes in E. faecium and E. faecalis have at least 98% homology.

PCR.
PCRs were performed in a GeneAmp PCR System 2400 (Perkin-Elmer) using Perkin-Elmer Standard PCR reaction mix with GeneAmp PCR buffer and Taq DNA polymerase. PCR elongation times were adjusted according to the expected size of PCR amplicons and the alignment temperatures were adjusted according to the specific nucleotide sequence of the primers and hence their melting temperatures.

PCR amplicons and primer sequences used in PCR identification, DNA sequencing and probe syntheses are listed in Table 2. Additional primer sequences and amplicons used in the characterization of ISEnfa110 and ISEnfa200 are described in Fig. 1(a) and (b), respectively.


View this table:
[in this window]
[in a new window]
 
Table 2. PCR primers used in this study

 


View larger version (14K):
[in this window]
[in a new window]
 
Fig. 1. Schematic representation of (a) Tn5382 left end of strain TUH4-67 with the ISEnfa110 element inserted and (b) the vanSBvanYB intergenic region of strain TUH7-15 with insertion ISEnfa200. Positions of primers and PCR amplicons used for DNA sequencing of the novel IS elements are indicated in the figure. The open boxes mark ORFs and the arrows indicate the coding strand and direction of transcription. IS ends and terminal inverted repeats of Tn5382 are indicated. The sequences of the primers are as follows: (a) 9-3650, GTTCTTATTCCGCAGGTGGTGATT; 4-67R3, ACTGGTACACCGGCTCGTCTA; 4-67F1, CTGCAATGTGCTCCAATAGAC; 4-67R8, CACTTCTGGTATTGGGTAGTAAAG; 4-67F8, AGTGGGTGGACTTTGTCGCTA; 4-67R12, ATTGCATTGACTACCGAGACA; 4-67F5, ATCGATGTTTCCAAGGGTAAG; 9-3362, ACGCCATGCTATTTACTTCCGGC; (b) vanSbF1, ATATGCGCTGGAAAACACCTC; 7-15F11, GAAATTCCGCCAAGCATGAGT; 7-15R7, CGCTCGAATATCATTAGTGTACT; vanYbR2, CCCCAGATTGTTTCATATGCC.

 
DNA sequencing and PFGE.
PCR amplicons were directly sequenced on both DNA strands by ABI Prism 377 (Perkin-Elmer), as described by Dahl et al. (1999) . Agarose plugs containing enterococcal DNA were digested with SmaI and analysed by PFGE as described by Dahl et al. (1999) with minor modifications. The plugs were incubated with 40 U mutanolysin ml-1 in addition to lysozyme to ensure proper lysis of the cells before treatment with 50 µg proteinase K ml-1.

Computer analysis and sequence accession numbers.
Editing, initial analysis of the DNA sequences and alignments were performed using the Sequence Navigator (Perkin Elmer) software package. Nucleotide sequences were compared to sequences in the GenBank, EMBL, DDBJ and PDB databases and protein sequences to non-redundant GenBank CDS translations, PDB, SWISS-PROT, SPupdate and PIR by using the BLASTN, BLASTP and BLASTX local alignment search tools (Altschul et al., 1990 ). Possible ORFs were found using the ORF finder located at the National Center for Biotechnology Information website. Searches for repeats and palindromes were performed using the GCG package. RNAdraw V1.0 was used to determine potential secondary structures. Novel nucleotide accession numbers and previously published GenBank and GenPept sequences mentioned in the text are shown in Table 3.


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Characterization of the VanB strain collection and vanB typing
The 23 VanB-type strains (Table 1) were characterized by PFGE and DNA sequencing of the vanB gene and the vanSBvanYB intergenic region as described by Dahl et al. (1999) . The vanB gene cluster of strains C68, D366, 3174, 3332 and 3568 was subtyped as vanB2 (Table 1). Subtyping of the remaining 18 PFGE unrelated strains (Table 1) has been published previously (Dahl et al., 1999 ). Altogether, eight strains were of the vanB1 subtype, 14 of the vanB2 subtype and one of the vanB3 subtype. PFGE analyses revealed that TUH7-15, C68, 3174, 3332 and 3568 might be related (see below), whereas the other strains were unrelated according to the criteria for PFGE typing by Tenover et al. (1995) .

The vanB2 gene cluster is an integral part of Tn5382
A vanB-containing 27 kb putative conjugative transposon, Tn5382, was recently described in E. faecium C68. The presence of Tn5382 was also documented in 15 strains from the USA as well as in one strain from France (Carias et al., 1998 ). In this study, genetic linkage of the specific vanB subtypes to Tn5382 was examined. The VanB strain collection was analysed by a Tn5382-specific PCR internal to the nonintegrase (left) end of Tn5382 (Carias et al., 1998 ), Southern hybridization using Tn5382 and vanB amplicons as probes (Table 2), and DNA sequencing of the vanXB–ORFC intergenic region and flanking coding sequences in Tn5382 (Fig. 2). Fourteen of the 23 strains (61%) were Tn5382 PCR positive (Table 1), which is comparable to the result obtained by Carias et al. (1998) . All Tn5382 PCR positive strains were of the vanB2 subtype (Dahl et al., 1999 ). SmaI-digested total DNA was subjected to PFGE (Fig. 3a) and Southern hybridization (Fig. 3b, c). Co-hybridization of the vanB and Tn5382 probes to the same fragments was detected for all strains (n=14) of the vanB2 subtype, as illustrated in Fig. 3 for strains TUH2-18 (lane 4) and TUH7-15 (lane 7). Strains of the vanB1 or vanB3 subtypes were negative for Tn5382 by PCR and hybridization (Fig. 3, lanes 2, 3 and 9).



View larger version (6K):
[in this window]
[in a new window]
 
Fig. 2. Map of Tn5382 showing the localization of the vanB2 gene cluster, ISEnfa110 and ISEnfa200. The open boxes mark ORFs and the arrows indicate direction of transcription. Terminal inverted repeats of Tn5382 are indicated.

 


View larger version (32K):
[in this window]
[in a new window]
 
Fig. 3. (a) PFGE of SmaI-digested total DNA of VanB-type strains, transconjugants and the recipient strain used in conjugation experiments, and (b, c) corresponding Southern hybridization with a vanB probe (b) and Tn5382 probe (c). Lanes 1 and 11, low-range PFG marker (New England Biolabs); lane 2, TUH4-64 (vanB1); lane 3, TUH4-65 (vanB1); lane 4, TUH2-18 (vanB2) donor; lane 5, BM2-18 transconjugant; lane 6, E. faecium BM4105-RF recipient; lane 7, TUH7-15 (vanB2) donor; lane 8, BM7-15 transconjugant; lane 9, VRE45 (vanB3); lane 10, C68 (positive control vanB and Tn5382).

 
Corresponding analyses were performed on five vancomycin-resistant transconjugants derived from each of TUH2-18 and TUH7-15. All transconjugants were Tn5382 PCR positive (data not shown). PFGE revealed minor genetic rearrangements in the transconjugants (Fig. 3a), and Southern blot analysis (Fig. 3b, c) showed co-hybridization of vanB and Tn5382 probes to novel fragments consistent with insertion events as illustrated by one transconjugant derived from TUH2-18 (lane 5) and one from TUH7-15 (lane 8). These results indicate cotransfer of the vanB2 gene cluster and Tn5382 during conjugation due to genetic linkage. The DNA sequence of the vanXB–ORFC intergenic region and flanking sequences in the vanB2 strains (Table 3) revealed a 550 bp vanXB–ORFC sequence identical to the corresponding part of the published Tn5382 sequence (Table 3) except for a 2 bp insertion and a 1 bp substitution. The 301 bp vanXB sequence upstream of this region revealed sequences almost identical to the corresponding region of the published vanB1 subtype gene cluster (Table 3) except for 16–18 bp substitutions which probably reflect different vanXB sequence in vanB1 and vanB2 subtypes. Thus, these results confirm that the vanB2 subtype gene cluster of these strains is an integral part of Tn5382 as shown in Fig. 2.

The DNA sequence of the vanB2 ligase gene seems to be less conserved than the vanB1 ligase gene (Dahl et al., 1999 ; Patel et al., 1998 ), indicating that the vanB2 gene cluster might be older in evolutionary terms or evolve more rapidly due to more efficient mechanisms for horizontal spread. As part of Tn5382, the vanB2 gene cluster may be transferable between cells and within the cell to conjugative chromosomal elements or plasmids, thereby promoting a more efficient spread as indicated by the finding of a wide geographical distribution of the vanB2 gene cluster. However, molecular epidemiological studies with regard to the relative distribution of vanB subtypes have not been performed to test this assumption.

Presence of IS16 and IS256-like in VanB-type VRE strains
The vanB gene cluster in E. faecalis BM4281 has been reported to be linked to the mobile genetic elements IS16 and IS256-like as a functional composite transposon, Tn1547. Hybridization studies of clinical VanB-type VRE isolates revealed that the vanB gene cluster was not necessarily linked to IS16 or IS256-like elements (Quintiliani & Courvalin, 1996 ). The vanB gene cluster of Tn1547 (Quintiliani & Courvalin, 1996 ) belongs to the vanB1 subtype (this study, Table 3). In this study the VanB strain collection was examined for prevalence of IS16-like and IS256-like elements and their genetic linkage to vanB gene cluster subtypes.

PCR analyses identified IS16-like sequences in 15 of 23 strains (65%) and IS256-like sequences in 19 of 23 strains (83%) (data not shown). These observations are consistent with the finding of IS16-like sequences in 21 of 32 (66%) and IS256-like sequences in 31 of 32 (97%) clinical VanB-type E. faecium and E. faecalis strains (Quintiliani & Courvalin, 1996 ), and IS256-like sequences in 88 out of 103 (85%) clinical E. faecium and E. faecalis strains (Rice & Thorisdottir, 1994 ). PFGE followed by Southern hybridization did not show any universal genetic linkage or co-hybridization of the vanB subtypes to IS16- and IS256-like sequences. Co-hybridization of vanB and both IS16-like and IS256-like sequences was found in only two vanB2 subtype strains, TUH7-15 and 3174. Co-hybridization of vanB and IS16-like sequences was found in two different fragments in TUH1-79, and in one fragment in TUH2-18 and 3568. Co-hybridization of vanB and IS256-like sequences was found in one fragment in V583, TUH7-54 and C68 (data not shown).

Multiple copies of both IS16- and IS256-like sequences were detected in several strains (data not shown). Our results confirm that IS16- and IS256-like sequences are common in enterococci, but that these elements are not generally linked to the vanB1 gene cluster as observed for the composite transposon Tn1547 in E. faecalis BM4281. Multiple copies of these IS elements raise possibilities for IS-based composite transposons and genomic rearrangements. IS256 is capable of forming novel mobile elements in enterococci as illustrated by the detection of Tn5281 (Hodel-Christian & Murray, 1991 ), conferring high level resistance to aminoglycosides, and Tn5384, encoding macrolide resistance and high level gentamicin resistance (Rice et al., 1995 ).

Characterization of ISEnfa110, a novel member of the IS110 family
The prevalence of Tn5382 in the VanB strain collection was determined by a Tn5382-specific PCR and hybridization using a 311 bp sequence starting 237 bp downstream of the left inverted repeat of Tn5382 in a nonintegrase sequence (Carias et al., 1998 ). Tn5382 PCR of TUH4-67 DNA showed an enlargement of this region. To identify the exact sequence of this enlargement, the complementary strand of the fragment internal to the nonintegrase (left) end of Tn5382 (Carias et al., 1998 ) was sequenced in strain C68 (Table 3). Sequence analysis of the same region in TUH4-67 revealed a 1611 bp insertion with a putative ORF of 1209 bp transcribed in the opposite direction to the integrase gene of Tn5382 (Figs 1a and 2, Table 3).

No extensive DNA homology was found with known insertion sequences by the BLAST program. However, the deduced 402 amino acid sequence of the ORF within this insert showed low levels (21–24%) of identity to about 380 amino acids from putative transposases of IS110, IS2112, IS1110, IS902 and IS901 (Table 3). An alignment of the amino acid sequences of this ORF to transposases of IS elements belonging to the IS110 family (Mahillon & Chandler, 1998 ), clearly demonstrated the relatedness between this IS element, hereafter designated ISEnfa110, and IS110 family members. The transposases of the IS110 family show some regions of identity within the N- and C-terminals. Two regions highly conserved in transposases of the IS110 family (Hernandez Perez et al., 1994 ) were present in the ISEnfa110 transposase.

Like most of the elements in this family, ISEnfa110 had no terminal inverted repeats. Target sequences of IS110 elements exhibit similarities to the circle junction of the element. Recent in vivo studies of IS110 elements have shown DNA fragments with abutted IS ends only when the transposase gene is intact. These observations are consistent with the model for transposition of IS110 elements by circularization and site-specific recombination (Mahillon & Chandler, 1998 ). A 2 bp sequence (5'-CT-3') was directly repeated at each end of ISEnfa110 after insertion into the integration site. According to the IS110 family criteria these elements do not create target duplications, but a match often occurs between one end of the IS and the target site sequence (Hernandez Perez et al., 1994 ). Thus, the CT sequence is probably present both in the attachment site of the circular form of the element and in the target site. ISEnfa110 showed additional sequence homology between the putative circle junction and the site of integration (Fig. 4).



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 4. Sequence similarity between the region surrounding the crossover point of the potential ISEnfa110 circular form and the integration site in Tn5382 of strain TUH4-67. Homologies are indicated by black shading. The ISEnfa110 circle and target sequence have been aligned by the addition of pads (-) to the sequences. Potential promoters (-35 and -10 boxes) upstream of the ISEnfa110 transposase gene in both the putative circular and the integrated form of the IS element are indicated. Nucleotide numbers corresponding to the linear form of ISEnfa110 are indicated. The crossover point (CT) is marked by **.

 
The ISEnfa110 ORF was preceded by a putative RBS (GTATTGGAGGN7ATG) that displays some complementarity (underlined) to the 3' extremity of the Bacillus subtilis 16S rRNA. The upstream region also revealed two putative promoter regions with some homology to the -35 (TTGACA) and -10 sequences (TATAAT) observed for Escherichia coli and B. subtilis promoters (Moran et al., 1982 ). A -35 region (CTGCCA) located in the target sequence was separated by 19 bp from a -10 sequence (TACAAT) starting at nucleotide 5 of the IS element (Fig. 4). The 19 bp spacer region is not optimal for transcription initiation. A stronger promoter might assemble upon circularization of ISEnfa110 with an optimal spacer region of 17 bp and a putative -35 region with the DNA sequence TTGACT (Fig. 4). The second putative promoter had a -35 (TTCCCT) and a -10 sequence (TACAAT) separated by a spacer region of 17 bp (Fig. 4). This promoter was located between nucleotides 47 to 75 of the IS element.

Characterization of ISEnfa200: a novel member of the IS200 family
The vanB subtyping strategy revealed an enlargement of the vanSBvanYB intergenic region in strain TUH7-15. Sequence analysis identified a 789 bp insert region (Table 3) with a 471 bp ORF oriented in the same direction as the vanB gene cluster (Figs 1b and 2). Searching with the BLAST program in different protein databases revealed sequence similarity to the transposases of the IS200 family, which was strengthened by the preserved structural features of this family (see below). The IS element was designated ISEnfa200.

ISEnfa200 encoded a putative 156 amino acid protein with 70% identity to the first 147 amino acids of the putative 151 amino acid transposase of IS1469 from Clostridium perfringens (Brynestad et al., 1997 ; Table 3). Amino acid identities above 60% were also detected with putative IS200 transposases of Salmonella typhimurium, S. typhi, Yersinia pestis, Escherichia coli and Actinobacillus actinomycetemcomitans (Table 3). The IS1469 ORF had sequence homology greater than 60% to ORFs in other IS200-like elements (Brynestad et al., 1997 ). In contrast, the ISEnfa200 DNA sequence only showed a significant degree of homology to other IS200 sequences in smaller regions. Thus, ISEnfa200 had a higher level of conservation at the amino acid level compared to the DNA level, suggesting that the coding capacity of the ORF was selectively maintained.

ISEnfa200 possessed general features of IS200 elements. The ISEnfa200 element (787 bp) was comparable in size to IS1469 (789 bp) and Salmonella spp. IS200 (708 bp). The element lacked terminal inverted repeats and had one ORF preceded by several sequences capable of forming hairpin structures (see below). The sequence data indicated that the ISEnfa200 insertion had generated a 2 bp target duplication (5'-TG-3'). However, it is also possible that these 2 bp were part of the ISEnfa200 element or sequences flanking the donor element at the previous target site. ISEnfa200 was inserted in an A+T-rich sequence (68% A+T bp in the neighbouring 22 bp) in a non-coding region upstream of the promoter region in the vanSBvanYB intergenic sequence. Thus, the insertion does not seem to affect expression of the proteins encoded by the vanB gene cluster.

Sequence analysis showed that the primary transcript may form a potential RNA factor independent transcription terminator at the left end of ISEnfa200 [nucleotides 2–16, {Delta}G0 -10·9 kcal mol-1 (-45·6kJ mol-1), followed by 4 U residues]. A second potential RNA stem–loop structure [nucleotides 119–174, {Delta}G0 -22.0 kcal mol-1 (-92·0 kJ mol-1)] that might occlude a putative RBS was also present. IS200 in S. typhimurium, S. abortusovis and Vibrio cholerae showed high sequence conservation in the stem–loop regions (Beuzón & Casadesús, 1997 ). These features still seemed to be preserved in ISEnfa200, although no significant DNA sequence homology was found to other IS200 elements in these regions. The RBS (ACAAAGGAGGN7ATG) upstream of the ISEnfa200 ORF was located at a standard distance from the start codon and displays a strong complementarity (underlined) to the 3' extremity of the B. subtilis 16S rRNA (UCUUUCCUCC) (Moran et al., 1982 ). A putative promoter with -35 (CTTACA) and -10 (TATAAT) regions separated by a spacer region of 16 bp was also located upstream of the putative start codon.

Rarity of ISEnfa110 and ISEnfa200 in enterococci
The presence of ISEnfa110- and ISEnfa200-like elements was examined by colony hybridization of 181 Enterococcus strains of geographically diverse origins using internal PCR fragments (Table 2) of the IS elements as probes. All colonies hybridized with the positive control 16S rDNA probe (data not shown). Only TUH4-67 was positive for ISEnfa110, whereas five strains (TUH7-15, C68, 3174, 3332 and 3568), including the original ISEnfa200 strain, were positive for ISEnfa200 (data not shown).

Southern hybridization of total DNA showed hybridization of the ISEnfa110 probe to an approximately 10 kb DraI restriction fragment of strain TUH4-67 (data not shown), indicating the presence of a single copy of ISEnfa110. ISEnfa110 has a G+C content of 51 mol%, which is higher than the G+C range of enterococci (34–42 mol%) but in the same range as the target site in Tn5382 (55 mol%). ISEnfa200 hybridization with total DNA revealed a positive 1·5 kb DraI restriction fragment (data not shown) in the five ISEnfa200-positive strains, indicating the presence of a single ISEnfa200 element. The low prevalence of these novel IS elements might reflect recent acquisition or low transposition frequencies of these elements in enterococci. ISEnfa200 has a G+C content of 41 mol%, which is typical of that of Enterococcus species (34–42 mol%). The low G+C range of ISEnfa200 reflects to a large extent the preferential use of A and T residues at the third position of codons in the putative ORF.

IS200 elements are mainly found in Gram-negative bacteria, except for IS1469 from C. perfringens and ISEnfa200 in E. faecium. ISEnfa200 had the highest degree of homology to IS1469 (see above). The presence of closely related IS elements in E. faecium and C. perfringens might indicate interspecies gene flux in a common environmental niche. The ISEnfa200 element was inserted in the vanB gene cluster, which has been shown to be transferable among enterococcal strains (Carias et al., 1998 ; Quintiliani & Courvalin, 1996 ; Woodford et al., 1995 ). Spread of ISEnfa200 will probably be a result of vanB gene cluster transfer rather than transposition of the IS element itself since IS200 elements have extremely low transposition frequencies.

A widespread USA vanB E. faecium strain with a chromosomal Tn5382vanB2–ISEnfa200 element genetically linked to pbp5?
Recent reports have shown that the putative conjugative transposon Tn5382 can be located downstream of a pbp5 gene encoding ampicillin resistance in clonally distinct E. faecium strains from different parts of the USA (Carias et al., 1998 ; Hanrahan et al., 1998 ). The two resistance determinants were co-transferred during conjugation as parts of a larger genetic element. In this study strains of different geographical origins were analysed for the occurrence of Tn5382 downstream of pbp5, to examine if these resistance elements are linked. The vanB2 positive strains (n=14) were examined for pbp5–Tn5382 linkage by PCR-amplifying a 1079 bp region between pbp5 and Tn5382 (Table 3). Five USA strains (TUH7-15, C68, 3174, 3332 and 3568) revealed a pbp5–Tn5382 amplicon of the expected size (data not shown). PFGE analysis of SmaI-digested DNA (Fig. 5) suggests genetic relatedness between these strains. The ISEnfa200 integration site in the vanSBvanYB intergenic region (Table 3) was identical in all five strains. Southern blot analyses showed co-hybridization of the pbp5 probe, Tn5382 probe and ISEnfa200 probe (data not shown), confirming the linkage between pbp5 and the Tn5382vanB2–ISEnfa200 element.



View larger version (100K):
[in this window]
[in a new window]
 
Fig. 5. PFGE of SmaI-digested total DNA of a vanB2 subtype strain from France (D366) and five possibly related vanB2 subtype strains from the USA. Lanes 1 and 8, low-range PFG marker (New England Biolabs); lane 2, D366; lane 3, TUH7-15; lane 4, C68; lane 5, 3174; lane 6, 3332; lane 7, 3568.

 
PFGE has so far been the most accurate method in molecular epidemiological studies of enterococci. Additional molecular markers might be useful. Transposition of IS200 occurs at extremely low frequency (Casadesus & Roth, 1989 ), which would make ISEnfa200 a very stable epidemiological marker. The finding of ISEnfa200 in the vanB2 gene cluster of Tn5382 in these geographically unrelated USA strains could be due to horizontal transfer of a large chromosomal element among enterococcal strains as suggested by Carias et al. (1998) . However, the genetic relatedness between strains as shown by PFGE analysis suggests a possible clonal spread in the USA of a VRE strain harbouring a Tn5382vanB2–ISEnfa200 element linked to a pbp5 gene conferring ampicillin resistance.


   ACKNOWLEDGEMENTS
 
Kristin H. Dahl and Eirik W. Lundblad contributed equally to this work.

This work was supported by grants from the Norwegian Research Council, Scandinavian Society for Antimicrobial Chemotherapy, and the Odd Berg Foundation. We thank the following people for providing strains: P. R. Chadwick of North Manchester General Hospital, Manchester, UK; P. Courvalin, of the Institut Pasteur, Paris, France; S. Harthug and A. Digranes of Haukeland University Hospital, Bergen, Norway; I. Klare of Robert Koch Institute, Wernigerode, Germany; E. B. Myhre of Lund University Hospital, Lund, Sweden; R. Patel of Mayo Clinic and Foundation, MN, USA; M. A. Pfaller and S. A. Marshall of University of Iowa College of Medicine, Iowa City, IA, USA; L. B. Rice, Case Western Reserve University School of Medicine, Cleveland, OH, USA; and A. Voss, University Hospital St Radboud, Nijmegen, the Netherlands. We also thank Marit R. Mikalsen and Mette S. Wesmajervi for excellent technical assistance, and Patrice Courvalin, Gunnar S. Simonsen and Johanna E. Sollid for helpful discussions and critical reading of the manuscript.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). Basic local alignment search tool.J Mol Biol215, 403-410.[Medline]

Arthur, M., Molinas, C., Depardieu, F. & Courvalin, P. (1993). Characterization of Tn1546, a Tn3-related transposon conferring glycopeptide resistance by synthesis of depsipeptide peptidoglycan precursors in Enterococcus faecium BM4147.J Bacteriol175, 117-127.[Abstract]

Beuzón, C. R. & Casadesús, J. (1997). Conserved structure of IS200 elements in Salmonella.Nucleic Acids Res25, 1355-1361.[Abstract/Free Full Text]

Brynestad, S., Synstad, B. & Granum, P. E. (1997). The Clostridium perfringens enterotoxin gene is on a transposable element in type A human food poisoning strains.Microbiology143, 2109-2115.[Abstract]

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 Bacteriol180, 4426-4434.[Abstract/Free Full Text]

Casadesus, J. & Roth, J. R. (1989). Absence of insertions among spontaneous mutants of Salmonella typhimurium.Mol Gen Genet216, 210-216.[Medline]

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 Chemother43, 1105-1110.[Abstract/Free Full Text]

Evers, S. & Courvalin, P. (1996). Regulation of VanB-type vancomycin resistance gene expression by the VanSB–VanRB two-component regulatory system in Enterococcus faecalis V583.J Bacteriol178, 1302-1309.[Abstract]

Evers, S., Reynolds, P. E. & Courvalin, P. (1994). Sequence of the vanB and ddl genes encoding D-alanine:D-lactate and D-alanine:D-alanine ligases in vancomycin-resistant Enterococcus faecalis V583.Gene140, 97-102.[Medline]

Franke, A. E. & Clewell, D. B. (1981). Evidence for a chromosome-borne resistance transposon (Tn916) in Streptococcus faecalis that is capable of ‘conjugal’ transfer in the absence of a conjugative plasmid.J Bacteriol145, 494-502.[Medline]

Galas, D. J. & Chandler, M. (1989). Bacterial insertion sequences. In Mobile DNA, pp. 109-162. Edited by D. E. Berg & M. M. Howe. Washington, DC: American Society for Microbiology.

Handwerger, S. & Skoble, J. (1995). Identification of chromosomal mobile element conferring high-level vancomycin resistance in Enterococcus faecium. Antimicrob Agents Chemother39, 2446-2453.[Abstract]

Handwerger, S., Skoble, J., Discotto, L. F. & Pucci, M. J. (1995). Heterogeneity of the vanA gene cluster in clinical isolates of enterococci from the Northeastern United States.Antimicrob Agents Chemother39, 362-368.[Abstract]

Hanrahan, J. A., Hoyen, C. & Rice, L. B. (1998). Evidence for the geographic dispersion of a transferable mobile element conferring resistance to ampicillin and vancomycin in VanB E. faecium. In Abstracts of the 38th Interscience Conference on Antimicrobial Agents and Chemotherapy, p. 95. Washington, DC: American Society for Microbiology.

Hernandez Perez, M., Fomukong, N. G., Hellyer, T., Brown, I. N. & Dale, J. W. (1994). Characterization of IS110, a highly mobile genetic element from Mycobacterium avium. Mol Microbiol12, 717-724.[Medline]

Hodel-Christian, S. L. & Murray, B. E. (1991). Characterization of the gentamicin resistance transposon Tn5281 from Enterococcus faecalis and comparison to staphylococcal transposons Tn4001 and Tn4031.Antimicrob Agents Chemother35, 1147-1152.[Medline]

Leclercq, R. (1997). Enterococci acquire new kinds of resistance.Clin Infect Dis24, S80-S84.[Medline]

Mahillon, J. & Chandler, M. (1998). Insertion sequences.Microbiol Mol Biol Rev62, 725-774.[Abstract/Free Full Text]

Moran, C. P.Jr, Lang, N., LeGrice, S. F. J., Lee, G., Stephens, M., Sonenshein, A. L., Pero, J. & Losick, R. (1982). Nucleotide sequences that signal the initiation of transcription and translation in Bacillus subtilis. Mol Gen Genet186, 339-346.[Medline]

Patel, R., Uhl, J. R., Kohner, P., Hopkins, M. K., Steckelberg, J. M., Kline, B. & Cockerill, F. R.III (1998). DNA sequence variation within vanA, vanB, vanC-1, and vanC-2/3 genes of clinical Enterococcus isolates.Antimicrob Agents Chemother42, 202-205.[Abstract/Free Full Text]

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 Lett122, 173-180.[Medline]

Quintiliani, R.Jr & Courvalin, P. (1996). Characterization of Tn1547, a composite transposon flanked by the IS16 and IS256-like elements, that confers vancomycin resistance in Enterococcus faecalis BM4281.Gene172, 1-8.[Medline]

Rice, L. B. & Carias, L. L. (1998). Transfer of Tn5385, a composite, multiresistance chromosomal element from Enterococcus faecalis.J Bacteriol180, 714-721.[Abstract/Free Full Text]

Rice, L. B. & Thorisdottir, A. S. (1994). The prevalence of sequences homologous to IS256 in clinical enterococcal isolates.Plasmid32, 344-349.[Medline]

Rice, L. B., Carias, L. L. & Marshall, S. H. (1995). Tn5384, an IS256-based composite mobile element in Enterococcus faecalis. In Genetics of Streptococci, Enterococci and Lactococci, pp. 71-75. Edited by J. J. Ferretti, M. S. Gilmore, T. R. Klaenhammer & F. Brown. Basel: S. Karger.

Sahm, D. F., Kissinger, J., Gilmore, M. S., Murray, P. R., Mulder, R., Solliday, J. & Clarke, B. (1989). In vitro susceptibility studies of vancomycin-resistant Enterococcus faecalis.Antimicrob Agents Chemother33, 1588-1591.[Medline]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Schouten, M. A., Hoogkamp-Korstanje, J. A. A., Kraak, W. A. G., Bartels, C. J. M., Roelofs-Willemse, H. J. G. R. & Voss, A. (1999). Vancomycin-resistant enterococci (VRE) prevalence in Europe.Clin Microbiol Infect5, 121.

Simonsen, G. S., Myhre, M. R. M., Dahl, K. H., Olsvik, Ø. & Sundsfjord, A. (2000). Typability of Tn1546-like elements in vancomycin resistant enterococci using long-range PCRs and specific analysis of polymorpic regions. Microb Drug Resist (in press).

Tenover, F. C., Arbeit, R. D., Goering, R. V., Mickelsen, P. A., Murray, B. E., Persing, D. H. & Swaminathan, B. (1995). Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing.J Clin Microbiol33, 2233-2239.[Free Full Text]

Trieu-Cuot, P., Carlier, C. & Courvalin, P. (1988). Conjugative plasmid transfer from Enterococcus faecalis to Escherichia coli.J Bacteriol170, 4388-4391.[Medline]

Williamson, R., Al-Obeid, S., Shlaes, J. H., Goldstein, F. W. & Shlaes, D. M. (1989). Inducible resistance to vancomycin in Enterococcus faecalis D366.J Infect Dis159, 1095-1104.[Medline]

Woodford, N., Morrison, D., Johnson, A. P., Bateman, A. C., Hastings, J. G. M., Elliott, T. S. J. & Cookson, B. (1995). Plasmid-mediated vanB glycopeptide resistance in enterococci.Microb Drug Resist1, 235-240.[Medline]

Received 6 December 1999; revised 18 February 2000; accepted 8 March 2000.