Laboratoire de Génétique et Microbiologie (UMR INRA-UHP no. 1128, IFR no. 110), Faculté des Sciences, Université Henri Poincaré (Nancy 1), BP239, 54506 Vanduvre-lès-Nancy, France
Correspondence
Gérard Guédon
guedon{at}nancy.inra.fr
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The GenBank accession numbers for the sequences reported in this paper are AJ586568, AJ586569, AJ586570 and AJ586571.
Present address: Department of Molecular Biology and Microbiology, Howard Hughes Medical Institute/Tufts University, Boston, MA 02111, USA.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Various elements that excise by site-specific recombination, self-transfer the resulting circular form by conjugation and integrate by recombination between a specific site of this circular form and a site in the genome of their host have been found in different bacteria (Burrus et al., 2002a, b
; Churchward, 2002
). Various groupings of large subsets of these elements were recently proposed, including novel classes like CONSTINs (Hochhut & Waldor, 1999
), conjugative genomic islands (Osborn & Boltner, 2002
) or extension of a more ancient class, the conjugative transposons (Merlin et al., 2000
). All these elements were recently proposed to be grouped as ICEs, irrespective of the mechanisms of conjugative transfer (cell-to-cell contact using pili or cell aggregation, transfer as single- or double-stranded DNA) and integration (low or high specificity, serine or tyrosine recombinase) (Burrus et al., 2002a
). Numerous putative ICEs were found by sequence analyses in genomes of various low G+C Gram-positive bacteria (Burrus et al., 2002b
; Paulsen et al., 2003
), various
- and
-Proteobacteria (Toussaint et al., 2003
; van der Meer & Sentchilo, 2003
), Bacteroides thetaiotaomicron (Xu et al., 2003
) and the high-G+C Gram-positive bacterium Streptomyces coelicolor (Bentley et al., 2002
). Therefore, ICEs are likely to be widespread among bacteria.
Comparisons between a large array of proven and putative ICEs from low-G+C Gram-positive bacteria showed that these elements evolve by acquisition or exchanges of modules with various transferable elements, including at least ICEs and plasmids (Burrus et al., 2002b; Garnier et al., 2000
; Roberts et al., 2001
). The insertion of an ICE into an unrelated ICE (Clewell & Flannagan, 1993
; Scott & Churchward, 1995
), of an ICE into a plasmid (Christie & Dunny, 1986
; Garnier et al., 2000
; Salyers et al., 1995a
, b
), of a plasmid into an ICE (Gasson et al., 1992
), of a mobilizable transposon into an ICE (Whittle et al., 2001
), of a type I transposon into an ICE (Hochhut et al., 2001b
) or of a type II transposon into an ICE (McDougal et al., 1998
) have been reported or deduced from sequence analyses. Such events followed by rearrangements (e.g. deletions) could explain at least some of the module exchanges and/or acquisitions between different ICEs or between ICEs and plasmids. These events can also lead to mobilization in cis of one element by another.
The 34 734-bp element ICESt1 is integrated in the 3' end of fda, an ORF encoding a putative fructose-1,6-diphosphate aldolase, from the lactic acid bacterium Streptococcus thermophilus CNRZ368 (Burrus et al., 2000). This element excises by a tyrosine-integrase-mediated site-specific recombination event between two 27-bp direct repeats included in its two attachment sites, attL and attR. ICESt1 carries a putative conjugative module distantly related to that of Tn916, indicating that ICESt1 is an ICE or derived from an ICE (Burrus et al., 2002b
). In this paper, we examine the characteristics of elements flanked by attachment sites closely related to those of ICESt1 and integrated in the same site in seven S. thermophilus strains. However, most of these elements (cis-mobilizable elements; CIMEs) do not carry recombination and conjugation modules and have probably evolved from ICEs by loss of these modules. Internal active or truncated attachment sites were found within ICESt1 and two other elements. Comparison of the CIMEs, ICESt1 and another related putative ICE suggests that these ICEs result from site-specific integration of an ICE into the attR attachment site of a CIME and subsequent partial deletion of the attI site created by the integration. A model of site-specific accretion of CIMEs on ICEs and cis-mobilization of CIMEs by ICEs is proposed.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Probes and hybridization.
Genomic DNA fragments generated by digestion with restriction endonucleases were separated by agarose electrophoresis and transferred onto Nitran membranes N (Amersham Pharmacia Biotech). Probes were obtained by digesting DNA of recombinant plasmids and then purified by elution from agarose. The probes were then digoxigenin-labelled by random priming using the Dig DNA Labelling and Detection Kit (Roche Biochemical). Hybridization, high-stringency washing (60 °C) and detection were performed using manufacturer's recommendations. Hybridization signals were detected using a Fluo-S Multimager (Bio-Rad) or on X-OMAT LS film (Kodak).
PCR.
The Expand Long Template PCR System Kit (Roche Biochemical) was used to amplify the elements integrated in fda and adjacent sequences on a GeneAMP PCR System 2400 thermal cycler (Applied Biosystems). These PCRs (long-range PCRs) were carried out in a 50 µl reaction volume, including manufacturer's buffers, 100 pmol of primers O132.3 and O131.2 (Table 1), 0·35 or 0·5 mM each dNTP, 0·75 µl manufacturer's enzyme mix and 25250 ng genomic DNA. After an initial denaturation step (2 min at 92 °C), a 35-cycle PCR was performed (denaturation 10 s at 92 °C, annealing 30 s at 46·5 °C, extension at 68 °C for 1545 min depending on the length of the PCR product), followed by a final extension step (7 min at 68 °C).
|
Real-time PCRs were performed with 60 pg200 ng DNA, 12·5 µl qPCR Mastermix (Eurogentec), 0·75 µl SYBR Green I (x10 000 dilution; Eurogentec) and 40 pmol primers ex4 and ex5 (Table 1) in a final volume of 25 µl. After the calibration of the thermocycler (2 min at 50 °C) and the activation of the hot start polymerase (10 min at 95 °C), 50 cycles (denaturation 15 s at 95 °C, annealing and extension 1 min at 55 °C) were performed to amplify the attB locus resulting from excision of ICESt1 or ICESt3. Then the melting curve of the PCR product was analysed with iCycler 2.3 software (Bio-Rad) to verify PCR specificity. It was acquired by heating the PCR product to 95 °C over 3 min, cooling it to 55 °C over 10 s and slowly heating it at 0·05 °C s1 to 95 °C under continuous fluorescence monitoring. For each run, a standard dilution of the 362-bp DNA fragment containing attB (preliminary obtained by PCR) was used to check the relative efficiency and quality of primers. A negative control (distilled water) was included in all real-time PCR assays. Each dilution was performed at least in duplicate. Real-time PCRs were carried out on an iCycler iQ real-time PCR detection system (Bio-Rad). The amounts of attB were normalized to the amount of chromosomal DNA deduced from A260 measurements.
DNA cloning, DNA sequencing and sequence analysis.
Prior to cloning, PCR products were purified using the High Pure PCR Product Purification Kit (Roche Biochemical). EcoRI or HindIII fragments of the long-range PCR products were ligated into the EcoRI or HindIII site of pBC KS+ (Stratagene). DNA fragments obtained by partial AluI digestion of long-range PCR products were cloned into pPCR-Script Amp KS+ (Stratagene). The ligation mixture was used to transform E. coli SURE by electroporation (Dower et al., 1988). Plasmids were extracted by the alkaline lysis method (Hopwood et al., 1985
).
DNA sequencing was performed on recombinant plasmids, PCR products or long-range PCR products. Sequencing reactions were performed on a GeneAMP PCR System 2400 thermal cycler using the ABI Prism BigDye Terminator Cycle Sequencing Ready Reaction Kit (PE Applied Biosystems). Sequencing products were analysed with an ABI PRISM 310 Genetic Analyser (PE Applied Biosystems). Searches for ORFs were performed with Heuristic GeneMark (Besemer & Borodovsky, 1999) (http://opal.biology.gatech.edu/GeneMark/heuristic_hmm2.cgi). BLASTN, BLASTX, BLASTP and PSI-BLAST (Altschul et al., 1997
) were used to search for similarities to sequences in the GenBank database. Sequences were aligned by the CLUSTALW program (Thompson et al., 1994
). Topology predictions for proteins were made using the HMMTOP program (http://www.enzim.hu/hmmtop) (Tusnady & Simon, 1998
). Signal peptides were searched using SignalP V2.0 (http://www.cbs.dtu.dk/services/SignalP/) (Nielsen & Krogh, 1998
; Nielsen et al., 1997
). Helixturnhelix DNA-binding motifs were predicted using HTH (http://npsa-pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSA/npsa_hth.html) (Dodd & Egan, 1990
).
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Moreover, 15 probes obtained from ICESt1 and covering the whole element, except the region of ICESt1 containing ISs, were also hybridized on EcoRI or ScaI genomic DNA patterns of the 18 tested strains to search for elements related to ICESt1. No elements or even regions closely related to ICESt1 were detected in the 12 strains that do not contain any element integrated in the 3' end of fda and in ATCC 19258 (data not shown). For all the strains carrying an element integrated into fda, except ATCC 19258, some of the probes obtained from ICESt1 hybridized with EcoRI or ScaI DNA fragments from the tested strains. Comparison of the hybridization patterns suggests that all these fragments are included in an element integrated in the 3' end of fda (data not shown) and therefore that all the elements related to ICESt1 found in the tested strains are integrated in the 3' end of fda.
Long-range PCR experiments using primers specific for sequences adjacent to ICESt1 (Table 1) were performed to amplify the putative integrated elements and their adjacent sequences (Fig. 1a
). Large PCR products (13·126·4 kb) were obtained in each of the seven strains (Fig. 1b
). The PCR products obtained from IP6756, CNRZ302 and ATCC 19987 have an identical size and identical EcoRI and HindIII digestion patterns, while the PCR product obtained from CNRZ455 only differs by a 1·3 kb additional sequence (data not shown). Therefore, the elements present in these four strains are very closely related. The EcoRI and HindIII digestion patterns of the PCR products obtained from CNRZ385, CNRZ308, ATCC 19258 and CNRZ302 were very different from each other and from the size of the fragments deduced from ICESt1 sequence, suggesting the presence of four different elements. The long-range PCR products obtained for CNRZ385, CNRZ308, ATCC 19258 and CNRZ302 were sequenced and found to contain elements which were named ICESt3,
CIME308, CIME19258 and CIME302, respectively. All these elements were found to be inserted in the 3' end of fda at the same location as ICESt1 (Fig. 2
and Fig. 3b
).
|
|
|
The putative proteins encoded by the cluster orf385C-arp1-orfQ-orf385B-orf385A-arp2 of ICESt3, except that encoded by orf385B, are related to transcriptional regulators and/or to proteins encoded by the operon orf14-orf13-orf12-orf11 of Tn5252 from Streptococcus pneumoniae (Table 2
, Fig. 2
). Therefore, this cluster may constitute a regulation module. The ORFs arp1, orfQ and arp2 of ICESt3 are closely related (8286 % identity) to ORFs from the putative regulation module of ICESt1 orfR-arp1-orfQ-orfP-arp2. However,
orf385C, orf385B and orf385A of this cluster are replaced by two unrelated ORFs, orfR and orfP in ICESt1 (Fig. 2
). Therefore, the excision, integration and/or conjugation of ICESt3 and ICESt1 may be differentially regulated. attB sites that resulted from the excision of these two elements were quantified by real-time PCR. After 16 h of cell growth (stationary phase), 0·9 % of the ICESt3 copies were found to be excised, whereas about 0·0001 % of the ICESt1 copies were excised.
The other ORFs of ICESt3 encode proteins that are not related or are distantly related to those of ICESt1 (Fig. 2, Table 2
). The orf385F and orf385G ORFs encode putative proteins related to methyltransferases belonging to type II restriction-modification (RM) systems (Table 2
). This suggests that ICESt3, like the cluster sth368IR-sth368IM of ICESt1 (Burrus et al., 2001
), may encode an RM system. None of the ICESt3 ORFs was found to encode a protein related to known restriction endonucleases. However, the sequence comparison generally failed to detect significant similarities between restriction endonucleases belonging to type II RM systems even if they cut identical sites or have similar tertiary structure (Pingoud & Jeltsch, 2001
).
The defective element CIME302 is likely to derive from an ICE by deletion of its recombination and conjugation modules while retaining putative attL and attR sites
The 13 147-bp element CIME302 is flanked by 27-bp direct repeats identical to those bordering ICESt1 and ICESt3 (data not shown). The left 4407-bp sequence and the right 190-bp sequence of CIME302 are almost identical (99·3 % and 99·5 %, respectively) to the left and right ends of ICESt1 (Fig. 2). Therefore, CIME302 is delimited by potential attL and attR attachment sites. A
int sequence almost identical to the 96-bp 3' end of the gene int of ICESt1 and ICESt3 (98·8 and 97·8 % identity, respectively) was found to the left of attR. However, CIME302 does not contain any sequence related to xis and to the other parts of int, suggesting that this element cannot mediate its own integration/excision. Moreover, the hybridization of the probe I131.1, which contains most of the gene int of ICESt1, reveals only the EcoRI fragment that includes the
int sequence from CIME302 (data not shown). Thus, the genome of strain CNRZ302 does not contain an integrase gene closely related to those of ICESt1 or ICESt3. Furthermore, nested PCR amplification failed to detect the attachment site attI that would result from the excision of the element (data not shown), suggesting that CIME302 does not excise in CNRZ302.
CIME302 does not carry any sequence related to the putative conjugation modules of ICESt1 and ICESt3, suggesting that it is neither mobilizable in trans nor conjugative. The 3' end of the truncated integrase gene is joined to a sequence closely related (91·4 %) to the left part of the putative regulation module of ICESt1 including orfR and the 75-bp 3' end of arp1 (Fig. 2, Table 2
). The 4407-bp left region of CIME302 shares 99·3 % identity with the left region of ICESt1, which includes attL, orfY, gst, orfX and a truncated IS981 copy. The 33-bp end of this truncated IS981 copy located between orfX and IS1191 of ICESt1 is lacking in CIME302. The left end of the orfR1-
arp1 region of CIME302 is closely related to ICESt1 and includes only the right part of the putative
-independent terminator found between sth368IM and orfR of ICESt1, i.e. the right copy of its inverted repeat but not the left copy. The sequences of ICESt1 located between this truncated IS981 and this putative terminator are replaced by unrelated or distantly related sequences in CIME302. Therefore, CIME302 probably derives from an ICE related to ICESt1 by deletion of the arp1-int region, i.e. a region which includes the conjugation module and most of the regulation and recombination modules, and by replacement of a region related to the IS1191-sth368IM region from ICESt1 by the region orf302J-orf302A of CIME302.
The region acquired by CIME302 encodes a putative RM system. orf302A encodes a putative methyltransferase related to those of various type II RM systems and orf302C encodes a putative endonuclease distantly related to R.MvaI and R.BcnI (Table 2). The element found in CNRZ455 (CIME455) only differs from CIME302 by the insertion of an IS1191 copy within the putative promoter of gst (Fig. 2
).
The defective element CIME19258 is flanked by sequences related to attL and attR of ICESt1 and includes a truncated ICE related to Tn1549 and ICELm1
The 14 999-bp element CIME19258 is flanked by sequences very closely related to the 27-bp repeats flanking ICESt1, ICESt3 and CIME302 (two differences in the left copy and four in the right copy) generating 15-bp perfect direct repeats (Fig. 3). The right copy of this repeat includes the 3' end of the ORF fda. The 102-bp right end of CIME19258 (including the 15-bp sequence) is 75·5 % identical to the 103-bp attR site of ICESt1 (Fig. 3b
). Moreover, perfect and imperfect repeats are found at positions similar to those of related repeats found in ICESt1, ICESt3 and CIME302. Similarly, the 202-bp left end of CIME19258 (including the 15-bp sequence) shows 59·9 % identity with the attL site of ICESt1 (Fig. 3a
). This suggests that CIME19258 is bordered by attL and attR attachment sites.
Neither CIME19258 (Fig. 2, Table 2
) nor the genome of strain ATCC 19258 (hybridization data not shown), which harbours CIME19258, has sequences closely related to the xis and int genes and to the putative conjugation modules of ICESt1 and ICESt3. Furthermore, nested PCR amplification failed to detect the attI site that would result from the excision of the element (data not shown). This suggests that CIME19258, like CIME302, is neither integrative nor conjugative. However, like the complete or truncated regulation modules of ICESt1 (orfR and arp1), ICESt3 (arp1) and CIME302 (orfR1), the right end of CIME19258 includes ORFs or pseudogenes (orfR, orf19258B and
orf19258A) which encode putative proteins related to those encoded by the operon orf14-orf13-orf12-orf11 from the ICE Tn5252 from S. pneumoniae (Table 2
, Fig. 2
). The right ORF of this CIME19258 cluster,
orf19258A, is truncated and disrupted by an IS1193 copy that is not flanked by direct repeats resulting from target duplication. Since ISs can cause deletion of adjacent sequences by various mechanisms (Galas & Chandler, 1989
; Kleckner et al., 1996
), this suggests that the sequences adjacent to the left end of IS1193 were probably deleted after the insertion of the IS. Therefore, CIME19258 could derive from an ICE related to ICESt1 and ICESt3 by the IS1193-mediated deletion of the conjugation module and most of the regulation and recombination modules.
The left end of CIME19258 encodes a putative excisionase, Xis, and a putative integrase, Int, closely related to those of Tn1549 and Tn916 from Enterococcus faecalis and Tn5382 from S. pneumoniae (Table 2). The putative excisionase is unrelated to that of ICESt1 while the integrase is very distantly related (about 21 % amino acid identity for the carboxyl moiety of the predicted protein). The 3' end of int is flanked by a 193-bp sequence showing 67 % identity with the putative attR site of Tn5382 (Carias et al., 1998
). orf19258G and a pseudogene,
orf19258F, located at the right of the putative recombination module, are distantly related to regulation genes of the ICEs Tn1549, Tn5382 and Tn916, and closely related to two ORFs of the putative ICE ICELm1 of Listeria monocytogenes (Burrus et al., 2002b
) (Table 2
). The three adjacent ORFs (cadA-orf19258E-cadC) encoding a putative Cd2+ resistance system are closely related to ORFs of ICELm1 (83 % identity). The region int-cadC has an organization identical to that of the right of ICELm1. This suggests that an ICE has integrated within CIME19258 (Fig. 3
). However, CIME19258 does not possess any conjugation module nor sequences related to attL from Tn1549, Tn916, Tn5382 or ICELm1, suggesting that the conjugation module and the attL site of the ICE have been deleted after integration.
Exchange of large sequences between the defective and truncated element CIME308 and Lactococcus lactis plasmids
The right end of the 16 841-bp element CIME308 is flanked by a 27-bp sequence which includes the 3' end of fda and is identical to those flanking ICESt1, ICESt3 and CIME302 (data not shown). The 924-bp right end of
CIME308 is almost identical (99·4 %) to that of ICESt1, including the 3' end of int and attR. However,
CIME308 does not carry any sequence related to the 503-bp left end of int, to xis or to the conjugation and regulation modules of ICESt1. The truncated int sequence is joined to an IS1191 copy that is not flanked by direct repeats resulting from target duplication. Therefore, this structure could result from the IS1191-mediated deletion of the regulation, conjugation and recombination modules of an ICE.
The left end of CIME308 is unrelated to those of ICESt1, ICESt3, CIME302 and CIME19258 (Fig. 2
) and corresponds to a copy of IS1193. This copy is not flanked by direct repeats resulting from the target duplication. Furthermore, the 14-bp right end of IS1195L and the 39-bp sequence located between the left end of the ICEs (or CIMEs) and IS1195L are missing. This suggests that IS1193 was involved in the deletion of the left end of
CIME308. Therefore,
CIME308 could derive from an ICE by the IS1191-mediated deletion of the regulation, conjugation and recombination modules and the IS1193-mediated deletion of its left end.
The comparison of the nucleotide sequence of CIME308 by BLASTN searches with GenBank sequences revealed an element inserted in the 3' end of fda of S. thermophilus 4134 (GenBank accession no. AJ315964). The sequence of
CIME4134 is almost identical (99·3 % identity) to the central and right regions of
CIME308 (Fig. 2
). cadA and cadC of
CIME4134 ORFs also found in
CIME308 encode a P-type Cd2+ efflux ATPase and its negative regulator, respectively, proteins which have been shown to be involved in Cd2+ and Zn2+ resistance (Schirawski et al., 2002
). The hsdRMS loci of
CIME4134 and
CIME308 are very closely related to various loci encoding type I RM systems, although the ORFs hsdS of the two elements encode distantly related specificity subunits (31 % aa identity) (Schirawski et al., 2002
). However, these RM systems are likely not to be functional since hsdR of
CIME308 has a nonsense mutation and hsdR of
CIME4134 has a frameshift mutation. The 3534-bp left end of
CIME308 is absent in
CIME4134 (Fig. 2
). It includes amiA, which encodes a putative binding subunit of an oligopeptide ABC carrier, and a complete IS1193 copy.
A very large part of CIME4134, and therefore
CIME308 (i.e. orf308A,
hsdR, hsdM, cadA, cadC, orf308B and
orf308C), is almost identical to sequences of L. lactis plasmids (Fig. 2
), suggesting that this region has recently been transferred between L. lactis and S. thermophilus (Schirawski et al., 2002
) and between CIMEs (or ICEs) and plasmids. Furthermore, the IS1191 copy found in
CIME308 is almost identical to IS905 from L. lactis and it was previously proposed that IS1191/IS905 might be transferred from S. thermophilus to L. lactis (Guédon et al., 1995
). Complete or truncated copies of ISs which were transferred between these two species (Bourgoin et al., 1998
, 1996
; Guédon et al., 1995
) are also found in ICESt1 (IS981 and IS1191), ICESt3 (IS1191), CIME19258 (ISS1 and IS1194) and CIME455 (IS1191) (Fig. 2
).
ICESt1, ICESt3 and CIME302 have a composite tandem structure
Comparison of internal ICESt1 sequences with sequences of its borders, i.e. attL and attR sites, revealed a 22-bp sequence which is identical to the left part of the 27-bp direct repeat defining ICESt1 limits (Fig. 3a). This sequence has the same orientation as the two 27-bp sequences belonging to attL and attR and is located at 7529 bp to the left of ICESt1. Moreover, the 339-bp sequence located on the right of the 22-bp direct repeat is 59 and 66 % identical to the attachment site attL of ICESt1 and ICESt3, respectively. The first 202 bp are also almost identical (99·0 %) to the left end of CIME19258. Therefore, this sequence could be an internal recombination site, attL', suggesting that ICESt1 is a composite element and that its central and right regions might be an ICE, ICESt2.
A 449-bp fragment showing a sequence identical to the attI' site and a 317-bp fragment containing the chromosomal counterpart attB', both resulting from the site-specific recombination between the 22-bp sequences from attL' and attR, were amplified by nested PCR experiments (Fig. 4). None of these PCR products was obtained in strain NST1008, a strain derived from CNRZ368 where the int gene of ICESt1 is disrupted (Burrus et al., 2000
). Therefore, the entire 27 bp of the direct repeats flanking ICESt1 are not absolutely required to promote site-specific recombination. However, the frequency of excision was too low to be measured by real-time quantitative PCR, whereas about 0·0001 % of ICESt1 copies were found to be excised in the stationary phase (data not shown). The sequence differences between attL' and attL probably account for these differences in excision frequency. While the recombination between attL' and attR leads to excision of a 27 205-bp circular form harbouring the entire regulation, integration and conjugation modules, a 7557-bp element flanked by the ICESt1 attL site and the attB' site remained integrated into the 3' end of the fda. The attempts to amplify junction fragments resulting from a site-specific recombination between attL and attL' failed (data not shown). Therefore, ICESt1 probably results from the accretion of two elements: (i) ICESt2, an autonomous site-specific ICE, and (ii) CIME368 which could be specifically mobilized in cis by ICESt2.
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In S. thermophilus, CIMEs and CIMEs related to ICESt1 seem to be common since 6 of the 19 tested strains harbour a CIME integrated in the fda locus, whereas only two harbour a putative ICE. Furthermore, these strains have not been found to be closely related by comparison of the hybridization pattern of probes specific to rRNA loci or ISs (Charron-Bourgoin et al., 2001
). Elements which are related to proven or putative ICEs and lack conjugation and integration modules while retaining attL and attR sites, were not previously reported. However, they would escape detection in sequenced genomes since they would only be detected by comparison with the ends of a related ICE or by the presence of direct repeats flanking the element. Direct repeats (22600 bp) flanking 150 kb regions which can be acquired by horizontal transfers are abundant within B. subtilis genome and at least in nine other completely sequenced genomes of prokaryotes (Nobusato et al., 2000
; Rocha et al., 1999
). Furthermore, some of these elements have been found to carry putative restriction-modification modules (Nobusato et al., 2000
) like the CIMEs of S. thermophilus. Like the CIMEs, some of these elements flanked by direct repeats might derive from ICEs.
Site-specific integrative elements are very common in bacterial genomes. Some of them, like prophages and ICEs, encode their own transfer mechanisms. Other elements, such as various pathogenicity islands (PAIs) (Hacker et al., 1997), possess integrative functions and have undergone horizontal transfer, but the mechanisms mediating these transfers remain unknown. However, the ICE CTnscr-94 and the PAI SPI-7 from Salmonella enterica were recently found to share closely related recombination and conjugation modules, suggesting that this PAI may be an ICE (Pembroke et al., 2002
; Pickard et al., 2003
). Although other PAIs were never proposed as ICEs, a region of the genome of the Enterococcus faecalis V583 genome, EfaD2, was recently found to be a part of a putative ICE (Burrus et al., 2002a
, b
) and was independently found to be included within the left end of a large PAI (Shankar et al., 2002
). Furthermore, a PAI of Legionella pneumophila harbours an integrase gene and a module encoding a type IV secretion system closely related to that encoded by the conjugation module of plasmid F of E. coli (Brassinga et al., 2003
). Therefore, it is likely that at least some PAIs are ICEs.
Numerous PAIs which are flanked by direct repeats and are associated with tRNA genes do not possess functional recombination modules or have lost the integrative functions by inactivation of the integrase by point mutations or deletions (Hacker & Kaper, 2000). One of these PAIs found in Bartonella tribocorum harbours a truncated integrase gene and a module encoding a type IV secretion system closely related to that encoded by the conjugation module of plasmid R388 (Seubert et al., 2003
), suggesting that it might derive from an ICE. Therefore, some PAIs might derive from ICE by deletion of their conjugation and recombination modules and therefore might be CIMEs.
ICESt1 was found to be a composite element that probably resulted from the accretion of a putative ICE, ICESt2, and of a CIME. Sequence comparison also showed that three ICESt1-related elements, ICESt2, ICESt3 and CIME302, had a composite structure. These elements probably evolved by accretion resulting from site-specific recombinations and deletions (Fig. 5). An ICE integrated into the 3' end of fda would have lost regulation, conjugation and integration modules while retaining the attL and attR attachment sites, leading to a CIME (step 1). Then, a related ICE acquired by conjugation (step 2) integrates by site-specific recombination between its attI site and between the attR site of the CIME (step 3), leading to the structure attL-CIME-attI'-ICE-attR'. Internal deletion would have truncated the internal attI' site, generating a functional attL' site or a
att site (step 4). The entire attI site might also be deleted, leading to a structure that does not seem to be composite. The whole structure would be a novel ICE and could be excised by recombination between attL and attR (step 5) and transferred to a new host bacterium (step 6). This accretion event would lead to the acquisition of novel modules by the ICE and to mobilization of the CIME in cis by the ICE.
|
It should also be emphasized that the attL-CIME-att-ICE-attR or attL-CIME-attL'-ICE-attR structures found in ICESt1 or ICESt3 could also arise by integration in a tandem fashion of two ICEs sharing the same integration site, and subsequent deletion of the regulation, conjugation and recombination modules of the left ICE (including a partial deletion of attI'). The formation of a tandem array of related ICEs has been reported for clc (Ravatn et al., 1998
), pSAM2 (Possoz et al., 2001
), and for the two related ICEs R391 and SXT (Hochhut et al., 2001a
). These tandem arrays might result from the site-specific integration of an incoming element acquired by conjugation into the attL or attR sites flanking a resident element. Whereas these structures were reported to be unstable (Hochhut et al., 2001a
; Possoz et al., 2001
; Ravatn et al., 1998
), deletion of the internal recombination module, including attI', would stabilize the composite element.
Evolution of ICEs or other types of integrative elements by site-specific accretion and deletion has not been proposed previously. However, it should be emphasized that very few ICEs were well characterized and completely sequenced. Numerous putative ICEs were recently found in completely sequenced genomes (Bentley et al., 2002; Burrus et al., 2002b
; Paulsen et al., 2003
; Toussaint et al., 2003
; van der Meer & Sentchilo, 2003
; Xu et al., 2003
). However, in most cases, their limits are not precisely known. Nevertheless, a rapid survey of published sequences shows that some ICEs or putative ICEs have a composite structure, suggesting that they have evolved by site-specific accretion.
The Tn5276-related ICE Tn5481 from the lactic acid bacterium L. lactis harbours an internal 187-bp sequence sharing 88 % identity with the left end of the element, i.e. attL (Immonen et al., 1998). The two attL-type sequences border a 13·3-kb region involved in nisin biosynthesis and resistance. This structure was compared previously to those of composite transposons flanked by IS elements. Such a structure would act rather like an internal attL' site as it occurs for ICESt1, and Tn5481 could be composed of a CIME and an ICE. The conjugative transfer of a Tn5276-like element lacking nisin biosynthesis has been reported (Kelly et al., 2000
). The size difference between the ICEs with and without the nisin module agrees with the size quoted for the complete nisin module flanked by the two attL-type attachment sites. This suggests that Tn5481 was formed by accretion and mobilization of a CIME encoding nisin synthesis by an ICE encoding sucrose metabolism.
The ICE pSAM2 from Streptomyces ambofaciens ATCC 23877 is flanked by two 58-bp direct repeats included in its attachment sites, attL and attR. A third 58-bp direct repeat (attX) is located 42 kb to the left of pSAM2 (Boccard et al., 1989). The whole attX-42 kb-attL-pSAM2-attR region is replaced by an attB site in the related strain DSM49697. This suggests that pSAM2 belongs to a composite element, flanked by attX and attR sites, which has probably arisen through site-specific accretion.
The 133-kb PAI SPI-7 from Salmonella enterica is a putative ICE (Pembroke et al., 2002) and is integrated into the 3' part of pheU which encodes a tRNAphe (Pickard et al., 2003
). An integrase gene (sty4680) is located at its right end near pheU. A truncated ORF closely related to sty4680 (93 % identity), sty4678, is located 1·6 kb to the left of sty4680 (GenBank accession no. NC_003198). A 611-kb ICE, the symbiosis island, is integrated into a tRNAPhe gene of Mesorhizobium loti MAFF303099 (Sullivan et al., 2002
). mll6432, located at the left end of the element near the tRNA gene, encodes the integrase of the symbiosis island. A truncated ORF closely related to mll6432 (81 % identity), msl6419, is located 12·1 kb to the right of mll6432 (GenBank accession no. NC_002678). Such structures may have arisen by tandem accretion of ICEs and CIMEs harbouring related complete or truncated integrase genes and deletion of the att site separating the elements.
The composite structure of various ICEs suggests that site-specific accretion of ICEs and CIMEs and subsequent deletions of attI could be a major mechanism of acquisition of novel modules by these elements. This mechanism would also lead to site-specific mobilization of CIMEs by ICEs in cis.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bentley, S. D., Chater, K. F., Cerdeno-Tarraga, A. M. & 40 other authors (2002). Complete genome sequence of the model actinomycete Streptomyces coelicolor A3(2). Nature 417, 141147.[CrossRef][Medline]
Besemer, J. & Borodovsky, M. (1999). Heuristic approach to deriving models for gene finding. Nucleic Acids Res 27, 39113920.
Boccard, F., Smokvina, T., Pernodet, J. L., Friedmann, A. & Guérineau, M. (1989). The integrated conjugative plasmid pSAM2 of Streptomyces ambofaciens is related to temperate bacteriophages. EMBO J 8, 973980.[Abstract]
Bourgoin, F., Guédon, G., Pébay, M., Roussel, Y., Panis, C. & Decaris, B. (1996). Characterization of a mosaic ISS1 element and evidence for the recent horizontal transfer of two different types of ISS1 between Streptococcus thermophilus and Lactococcus lactis. Gene 178, 1523.[CrossRef][Medline]
Bourgoin, F., Guédon, G., Gintz, B. & Decaris, B. (1998). Characterization of a novel insertion sequence, IS1194, in Streptococcus thermophilus. Plasmid 40, 4449.[CrossRef][Medline]
Bourgoin, F., Pluvinet, A., Gintz, B., Decaris, B. & Guédon, G. (1999). Are horizontal transfers involved in the evolution of the Streptococcus thermophilus exopolysaccharide synthesis loci? Gene 233, 151161.[CrossRef][Medline]
Brassinga, A. K., Hiltz, M. F., Sisson, G. R., Morash, M. G., Hill, N., Garduno, E., Edelstein, P. H., Garduno, R. A. & Hoffman, P. S. (2003). A 65-kilobase pathogenicity island is unique to Philadelphia-1 strains of Legionella pneumophila. J Bacteriol 185, 46304637.
Burrus, V., Roussel, Y., Decaris, B. & Guédon, G. (2000). Characterization of a novel integrative element, ICESt1, in the lactic acid bacterium Streptococcus thermophilus. Appl Environ Microbiol 66, 17491753.
Burrus, V., Bontemps, C., Decaris, B. & Guédon, G. (2001). Characterization of a novel type II restriction-modification system, Sth368I, encoded by the integrative element ICESt1 of Streptococcus thermophilus CNRZ368. Appl Environ Microbiol 67, 15221528.
Burrus, V., Pavlovic, G., Decaris, B. & Guédon, G. (2002a). Conjugative transposons: the tip of the iceberg. Mol Microbiol 46, 601610.[CrossRef][Medline]
Burrus, V., Pavlovic, G., Decaris, B. & Guédon, G. (2002b). The ICESt1 element of Streptococcus thermophilus belongs to a large family of integrative and conjugative elements that exchange modules and change their specificity of integration. Plasmid 48, 77.[CrossRef][Medline]
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, 44264434.
Charron-Bourgoin, F., Pluvinet, A., Morel, C., Guédon, G. & Decaris, B. (2001). Polymorphism of eps loci involved in exopolysaccharide synthesis of Streptococcus thermophilus. Lait 81, 281288.[CrossRef]
Christie, P. J. & Dunny, G. M. (1986). Identification of regions of the Streptococcus faecalis plasmid pCF-10 that encode antibiotic resistance and pheromone response functions. Plasmid 15, 230241.[Medline]
Churchward, G. (2002). Conjugative transposons and related mobile elements. In Mobile DNA II, pp. 177191. Edited by N. L. Craig, R. Craigie, M. Gellert & M. L. Lambowitz. Washington DC: American Society for Microbiology.
Clewell, D. B. & Flannagan, F. E. (1993). The conjugative transposons of Gram-positive bacteria. In Bacterial Conjugation, pp. 369388. Edited by D. B. Clewell. New York: Plenum.
Dodd, I. B. & Egan, J. B. (1990). Improved detection of helix-turn-helix DNA-binding motifs in protein sequences. Nucleic Acids Res 18, 50195026.[Abstract]
Dower, W. J., Miller, J. F. & Ragsdale, C. W. (1988). High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res 16, 61276145.[Abstract]
Galas, D. J. & Chandler, M. (1989). Bacterial insertion sequences. In Mobile DNA, pp. 109162. Edited by D. E. Berg & M. M. Howe. Washington DC: American Society for Microbiology.
Garnier, F., Taourit, S., Glaser, P., Courvalin, P. & Galimand, M. (2000). Characterization of transposon Tn1549, conferring VanB-type resistance in Enterococcus spp. Microbiology 146, 14811489.
Gasson, M. J., Swindell, S., Maeda, S. & Dodd, H. M. (1992). Molecular rearrangement of lactose plasmid DNA associated with high-frequency transfer and cell aggregation in Lactococcus lactis 712. Mol Microbiol 6, 32133223.[Medline]
Guédon, G., Bourgoin, F., Pébay, M., Roussel, Y., Colmin, C., Simonet, J. M. & Decaris, B. (1995). Characterization and distribution of two insertion sequences, IS1191 and iso-IS981, in Streptococcus thermophilus: does intergeneric transfer of insertion sequences occur in lactic acid bacteria co-cultures? Mol Microbiol 16, 6978.[Medline]
Hacker, J. & Kaper, J. B. (2000). Pathogenicity islands and the evolution of microbes. Annu Rev Microbiol 54, 641679.[CrossRef][Medline]
Hacker, J., Blum-Oehler, G., Muhldorfer, I. & Tschape, H. (1997). Pathogenicity islands of virulent bacteria: structure, function and impact on microbial evolution. Mol Microbiol 23, 10891097.[Medline]
Hochhut, B. & Waldor, M. K. (1999). Site-specific integration of the conjugal Vibrio cholerae SXT element into prfC. Mol Microbiol 32, 99110.[CrossRef][Medline]
Hochhut, B., Marrero, J. & Waldor, M. K. (2000). Mobilization of plasmids and chromosomal DNA mediated by the SXT element, a constin found in Vibrio cholerae O139. J Bacteriol 182, 20432047.
Hochhut, B., Beaber, J. W., Woodgate, R. & Waldor, M. K. (2001a). Formation of chromosomal tandem arrays of the SXT element and R391, two conjugative chromosomally integrating elements that share an attachment site. J Bacteriol 183, 11241132.
Hochhut, B., Lotfi, Y., Mazel, D., Faruque, S. M., Woodgate, R. & Waldor, M. K. (2001b). Molecular analysis of antibiotic resistance gene clusters in Vibrio cholerae O139 and O1 SXT constins. Antimicrob Agents Chemother 45, 29913000.
Hopwood, D. A., Bibb, M. J., Chater, K. F. & 7 other authors (1985). Genetic Manipulation of Streptomyces: a Laboratory Manual. Norwich: The John Innes Institute.
Immonen, T., Wahlstrom, G., Takala, T. & Saris, P. E. (1998). Evidence for a mosaic structure of the Tn5481 in Lactococcus lactis N8. DNA Seq 9, 245261.[Medline]
Kelly, W. J., Davey, G. P. & Ward, L. J. (2000). Novel sucrose transposons from plant strains of Lactococcus lactis. FEMS Microbiol Lett 190, 237240.[CrossRef][Medline]
Kleckner, N., Chalmers, R. M., Kwon, D., Sakai, J. & Bolland, S. (1996). Tn10 and IS10 transposition and chromosome rearrangements: mechanism and regulation in vivo and in vitro. Curr Top Microbiol Immunol 204, 4982.[Medline]
Manganelli, R., Romano, L., Ricci, S., Zazzi, M. & Pozzi, G. (1995). Dosage of Tn916 circular intermediates in Enterococcus faecalis. Plasmid 34, 4857.[CrossRef][Medline]
McDougal, L. K., Tenover, F. C., Lee, L. N., Rasheed, J. K., Patterson, J. E., Jorgensen, J. H. & LeBlanc, D. J. (1998). Detection of Tn917-like sequences within a Tn916-like conjugative transposon (Tn3872) in erythromycin-resistant isolates of Streptococcus pneumoniae. Antimicrob Agents Chemother 42, 23122318.
Merlin, C., Mahillon, J., Nesvera, J. & Toussaint, A. (2000). Gene recruiters and transporters: the modular structure of bacterial elements. In The Horizontal Gene Pool: Bacterial Plasmids and Gene Spread, pp. 363408. Edited by C. M. Thomas. Newark: Hardwood Academic Publishers.
Nielsen, H. & Krogh, A. (1998). Prediction of signal peptides and signal anchors by a hidden Markov model. Proc Int Conf Intell Syst Mol Biol 6, 122130.[Medline]
Nielsen, H., Engelbrecht, J., Brunak, S. & von Heijne, G. (1997). Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng 10, 16.[CrossRef]
Nobusato, A., Uchiyama, I., Ohashi, S. & Kobayashi, I. (2000). Insertion with long target duplication: a mechanism for gene mobility suggested from comparison of two related bacterial genomes. Gene 259, 99108.[CrossRef][Medline]
Osborn, M. A. & Boltner, D. (2002). When phage, plasmids, and transposons collide: genomic islands, and conjugative- and mobilizable-transposons as a mosaic continuum. Plasmid 48, 202212.[CrossRef][Medline]
Paulsen, I. T., Banerjei, L., Myers, G. S. & 27 other authors (2003). Role of mobile DNA in the evolution of vancomycin-resistant Enterococcus faecalis. Science 299, 20712074.
Pembroke, J. T., MacMahon, C. & McGrath, B. (2002). The role of conjugative transposons in the Enterobacteriaceae. Cell Mol Life Sci 59, 20552064.[CrossRef][Medline]
Pickard, D., Wain, J., Baker, S. & 12 other authors (2003). Composition, acquisition, and distribution of the Vi exopolysaccharide-encoding Salmonella enterica pathogenicity island SPI-7. J Bacteriol 185, 50555065.
Pingoud, A. & Jeltsch, A. (2001). Structure and function of type II restriction endonucleases. Nucleic Acids Res 29, 37053727.
Possoz, C., Ribard, C., Gagnat, J., Pernodet, J. L. & Guérineau, M. (2001). The integrative element pSAM2 from Streptomyces: kinetics and mode of conjugal transfer. Mol Microbiol 42, 159166.[CrossRef][Medline]
Ravatn, R., Studer, S., Springael, D., Zehnder, A. J. & van der Meer, J. R. (1998). Chromosomal integration, tandem amplification, and deamplification in Pseudomonas putida F1 of a 105-kilobase genetic element containing the chlorocatechol degradative genes from Pseudomonas sp. Strain B13. J Bacteriol 180, 43604369.
Reimmann, C. & Haas, D. (1993). Mobilization of chromosomes and nonconjugative plasmids by cointegrative mechanisms. In Bacterial Conjugation, pp. 137188. Edited by D. B. Clewell. New York: Plenum.
Roberts, A. P., Johanesen, P. A., Lyras, D., Mullany, P. & Rood, J. I. (2001). Comparison of Tn5397 from Clostridium difficile, Tn916 from Enterococcus faecalis and the CW459tet(M) element from Clostridium perfringens shows that they have similar conjugation regions but different insertion and excision modules. Microbiology 147, 12431251.
Rocha, E. P., Danchin, A. & Viari, A. (1999). Functional and evolutionary roles of long repeats in prokaryotes. Res Microbiol 150, 725733.[CrossRef][Medline]
Salyers, A. A., Shoemaker, N. B. & Li, L. Y. (1995a). In the driver's seat: the Bacteroides conjugative transposons and the elements they mobilize. J Bacteriol 177, 57275731.
Salyers, A. A., Shoemaker, N. B., Stevens, A. M. & Li, L. Y. (1995b). Conjugative transposons: an unusual and diverse set of integrated gene transfer elements. Microbiol Rev 59, 579590.[Medline]
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Schirawski, J., Hagens, W., Fitzgerald, G. F. & Van Sinderen, D. (2002). Molecular characterization of cadmium resistance in Streptococcus thermophilus strain 4134: an example of lateral gene transfer. Appl Environ Microbiol 68, 55085516.
Scott, J. R. & Churchward, G. G. (1995). Conjugative transposition. Annu Rev Microbiol 49, 367397.[CrossRef][Medline]
Seubert, A., Hiestand, R., De La Cruz, F. & Dehio, C. (2003). A bacterial conjugation machinery recruited for pathogenesis. Mol Microbiol 49, 12531266.[CrossRef][Medline]
Shankar, N., Baghdayan, A. S. & Gilmore, M. S. (2002). Modulation of virulence within a pathogenicity island in vancomycin-resistant Enterococcus faecalis. Nature 417, 746750.[CrossRef][Medline]
Smith, C. J., Tribble, G. D. & Bayley, D. P. (1998). Genetic elements of Bacteroides species: a moving story. Plasmid 40, 1229.[CrossRef][Medline]
Sullivan, J. T., Trzebiatowski, J. R., Cruickshank, R. W. & 11 other authors (2002). Comparative sequence analysis of the symbiosis island of Mesorhizobium loti strain R7A. J Bacteriol 184, 30863095.
Terzaghi, B. E. & Sandine, W. E. (1975). Improved medium for lactic streptococci and their bacteriophages. Appl Microbiol Lett 50, 807813.
Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22, 46734680.[Abstract]
Toussaint, A. & Merlin, C. (2002). Mobile elements as a combination of functional modules. Plasmid 47, 2635.[CrossRef][Medline]
Toussaint, A., Merlin, C., Monchy, S., Benotmane, M. A., Leplae, R., Mergeay, M. & Springael, D. (2003). The biphenyl- and 4-chlorobiphenyl-catabolic transposon Tn4371, a member of a new family of genomic islands related to IncP and Ti plasmids. Appl Environ Microbiol 69, 48374845.
Tusnady, G. E. & Simon, I. (1998). Principles governing amino acid composition of integral membrane proteins: application to topology prediction. J Mol Biol 283, 489506.[CrossRef][Medline]
van der Meer, J. R. & Sentchilo, V. (2003). Genomic islands and the evolution of catabolic pathways in bacteria. Curr Opin Biotechnol 14, 17.[CrossRef]
Whittle, G., Hund, B. D., Shoemaker, N. B. & Salyers, A. A. (2001). Characterization of the 13-kilobase ermF region of the Bacteroides conjugative transposon CTnDOT. Appl Environ Microbiol 67, 34883495.
Xu, J., Bjursell, M. K., Himrod, J., Deng, S., Carmichael, L. K., Chiang, H. C., Hooper, L. V. & Gordon, J. I. (2003). A genomic view of the human-Bacteroides thetaiotaomicron symbiosis. Science 299, 20742076.
Received 3 November 2003;
revised 16 January 2004;
accepted 23 January 2004.
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
J MED MICROBIOL | ALL SGM JOURNALS |