Elements in the co-evolution of relaxases and their origins of transfer.

Abstract

The central elements in the conjugative mobilization of most plasmids are the relaxase and its cognate origin of transfer (oriT). The relaxase of the plasmid R1162, together with its oriT, belong to a large and widely distributed family of related relaxase/oriT pairs. Several of the properties of these elements are considered for R1162 and for other members of this family with a view to understanding how systems for mobilization might have evolved.

Full text

Page 1

Review
Elements in the co-evolution of relaxases and their origins
of transfer
Christopher Parker, Eric Becker, Xiaolin Zhang, Sarah Jandle, Richard Meyer*
Section of Molecular Genetics and Microbiology, School of Biology, University of Texas at Austin, Austin, TX 7871, USA
Received 20 December 2004
Available online 1 February 2005
Communicated by Dr. Dhruba K. Chattoraj
Abstract
The central elements in the conjugative mobilization of most plasmids are the relaxase and its cognate origin of
transfer (oriT). The relaxase of the plasmid R1162, together with its oriT, belong to a large and widely distributed fam-
ily of related relaxase/oriT pairs. Several of the properties of these elements are considered for R1162 and for other
members of this family with a view to understanding how systems for mobilization might have evolved.
? 2005 Elsevier Inc. All rights reserved.
Keywords: Conjugation; Mobilization; Plasmid; Origin of transfer; Relaxase
The mechanism of conjugative DNA transfer
appears to be remarkably conserved across a
broad spectrum of plasmids in both Gram-nega-
tive and Gram-positive bacteria. Proteins assemble
at a site on the plasmid DNA, the origin of trans-
fer (oriT), to form a complex called the relaxo-
some. The centrally important component of this
complex is the relaxase. This protein cleaves one
of the oriT DNA strands by a reversible transeste-
rification to generate a protein–DNA covalent
intermediate (Matson et al., 1993; Pansegrau
et al., 1990; Scherzinger et al., 1993). For the relax-
ases examined to date, the entering nucleophile is a
tyrosine, and the protein becomes joined to the
DNA by a 50tyrosyl phosphodiester linkage. In vi-
tro, the protein–DNA intermediate is long-lived
and readily isolated (Becker and Meyer, 2002;
Pansegrau et al., 1993). Incubation of the interme-
diate with the other cleavage product regenerates
full-length oriT DNA accompanied by the release
of free protein. In a similar fashion, the plasmid
DNA is presumably recircularized after transfer
following displacement of the linked tyrosine by
the 30OH at the end of the transferred strand. In
this paper we will briefly consider several of the
0147-619X/$ - see front matter ? 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.plasmid.2004.12.007
*Corresponding author. Fax: +1 512 471 7088.
E-mail address: rmeyer@mail.utexas.edu (R. Meyer).
Plasmid 53 (2005) 113–118
www.elsevier.com/locate/yplas

Page 2

properties of the relaxases and their associated
oriTs, and what these properties suggest about
how systems for mobilization have evolved into
efficient mechanisms for the transfer of DNA.
The relaxosome of the IncQ plasmid R1162
(similar to RSF1010) reflects some of the general
features of relaxosomes while retaining overall a
simple structure. The origin of transfer contains
an inverted repeat separated from the nick site
by an AT-rich region (Fig. 1, top). The outer
arm of the inverted repeat is not required for a
functional relaxosome since in its absence transfer
is initiated normally (Kim and Meyer, 1989). The
adjacent DNA, containing the AT-rich region
and the nick site, is also contacted by the relaxase,
but here binding and subsequent nicking require
disruption of the duplex DNA (Zhang and Meyer,
1995). The requirement for at least partial single-
strandedness at the nick site appears to be a gen-
eral feature of relaxases, but the details of how this
is achieved varies from plasmid to plasmid. Many
plasmids use one or more accessory proteins to as-
sist in strand separation. For example, binding
sites for both the plasmid-encoded protein, TraY,
as well as host-encoded IHF are required for opti-
mal nicking by the F factor relaxase (Fu et al.,
1991; Nelson et al., 1995), and it is likely that both
these proteins restructure the DNA to assist in
strand separation.
Assembly of the R1162 relaxosome results in an
overall increase in the superhelical density of the
plasmid DNA, and the most highly underwound
topoisomers are preferentially cleaved (Zhang
et al., 2003). Presumably, increased torsional strain
results in a more severe disruption of the duplex
DNA at oriT. MobC, one of the accessory proteins
of the R1162 relaxosome, can induce increased
supercoiling in the cell in the absence of the other
relaxosome proteins (in preparation). This protein
is relatively nonspecific, and the topoisomer profile
of even plasmids such as pBR322 is changed in sta-
tionary phase cells when MobC is present. How-
ever, in log-phase cells, plasmids containing oriT
are preferentially targeted by the protein; with
the DNA near the nick site, but not the inverted
repeat, required for activity. MobC probably
binds to regions of supercoiled DNA containing
a localized disruption of the duplex (the protein
is inactive on nonsupercoiled DNA) and assembles
to form a complex (Scherzinger et al., 1992) which
then locally increases the superhelical density. In
the cell, this is sensed as a deficiency in supercoil-
ing elsewhere in the molecule, which is then acted
upon by the cellular gyrase.
It is noteworthy that the action of MobC is at
least partially independent of the relaxase. This
suggests that in the evolution of efficient transfer
systems, relaxases are opportunistic in recruiting
auxiliary proteins with the properties required to
increase the efficiency of nicking. For this reason,
even closely related relaxases can have different
requirements for helper proteins. The plasmid
pSC101 encodes a relaxase very similar to that of
R1162, but does not encode a protein correspond-
ing to MobC (Meyer, 2000). The pSC101 relaxase
might have evolved along with its oriT to separate
the DNA strands at the nick site, without requir-
ing a helper protein, although we have not yet ex-
cluded definitively an unidentified host protein in
the relaxosome. Recently we have isolated muta-
tions, mapping to the inner arm of the inverted re-
peat of the pSC101 oriT that decrease the
Fig. 1. The origin of transfer of plasmid R1162 (top). Shown
below are several other oriTs, putative or verified, belonging to
the same family, to indicate the conservation of the core region
(shaded), and the variability in the sequence and size of the
inverted repeats.
114
C. Parker et al. / Plasmid 53 (2005) 113–118

Page 3

frequency of transfer. These mutations can be par-
tially suppressed by MobC. Our interpretation is
that the mutations either lower the affinity of the
relaxase for the DNA, or change the orientation
of the protein in the relaxosome. This makes the
relaxase less able to initiate strand separation
around the nick site. MobC, because it indepen-
dently binds to oriT, can promote strand separa-
tion and increase the frequency of transfer.
Another way that relaxases frequently become
optimally adapted to their role in transfer is by
having a second, transfer-related function, con-
tained within the C-terminal region of the protein.
Thus, the TraI protein the F factor is both a relax-
ase and helicase, and both functions are required
for transfer (Matson et al., 2001). The C-terminal
region of the R1162 relaxase is a primase. In this
case, the primase is also made as a separate protein
(Scholz et al., 1989), and the linkage between the
relaxase and primase proteins probably evolved
to increase the overall efficiency of the Mob sys-
tem. The R1162 primase is required for replication
of the plasmid, and is site-specific, acting on two
single-strand initiation sites located in the origin
of vegetative replication (oriV) (Honda et al.,
1989; Lin and Meyer, 1987). It could also be in-
volved in strand replacement synthesis in both do-
nor and recipient, following a round transfer.
However, the primase is not absolutely required
for transfer: when the rest of the R1162 Mob sys-
tem is cloned into a vector such as pBR322, the
resulting molecule is mobilized at high frequency
by the self-transmissible plasmid R751 (Brasch
and Meyer, 1986). When a less efficient mobilizing
vector is used, then the priming system increases
the transfer frequency (Henderson and Meyer,
1996), showing that the system can be active dur-
ing mating. Under normal circumstances, then,
does the fusion provide a benefit during transfer?
To address this question, we needed to look at
R1162 primase activity during transfer under con-
ditions where potential priming systems are not
being provided on the plasmid by foreign DNA
(for example, by cloning vector DNA), and where
there is no ongoing vegetative replication. We
developed a procedure in which R1162 derivatives,
lacking the genes for replication, are introduced
into potential conjugative donors, and these cells
then immediately mated (Parker and Meyer,
2002). Because the donors lack one or more of
the plasmid-specificreplication
incoming molecules are unable to replicate, but
can be transferred into recipient cells and rescued
by integration into the chromosome by means of
the k integrase. Thus, there is no plasmid vegeta-
tive replication in either donor or recipient. In
addition, the method allows us to modify the
DNA physically and these alterations will then en-
ter the transfer cycle.
Using the electroporation and transfer method,
we introduced into donor cells test plasmid DNA
containing a base-pair mismatch, so that there
was a XhoI site on the strand that is transferred,
and a SapI site on the recipient. After a round of
transfer, DNA entering the recipient and captured
by integration at attB should contain the XhoI site,
whereas subsequent rounds of transfer, following
strand replacement in the donor, would result in
integrated DNA having a SapI site. We measured
the relative amounts of XhoI and SapI DNA (by
PCR amplification of integrated DNA and diges-
tion of this DNA with each restriction enzyme) un-
der two conditions. In one case, the test plasmid
DNA contained one of the initiation sites for the
R1162 primase, oriented for replacement synthesis
in the donor, and in the second case, this site was
absent. The primase was provided by the R1162
relaxase in the donor cells. We found that the ratio
of SapI to XhoI transfers was much greater when
the priming site was present. This indicated that
in the donor, the R1162 priming system could re-
place strands lost by transfer, and these could then
be used in subsequent rounds of mating.
Our observation that one of the initiation sites in
the replicative origin is responsible for much of the
replacement strand synthesis in the donor led us to
expect that the other, oppositely oriented site
would be utilized for replacement synthesis in the
recipient. However, when we tested for transfer
plasmids lacking this site, they transferred almost
as well as those with the complete origin of replica-
tion (submitted). Since R751, the mobilizing vector
in these experiments, encodes a primase of low
specificity (Miele et al., 1991), it was possible that
this protein was initiating strand replacement syn-
thesis. However, deletion of the primase gene traC
proteins,the
C. Parker et al. / Plasmid 53 (2005) 113–118
115

Page 4

in fact reduced the frequency of transfer. We con-
cluded that there must be a robust, endogenous
system in Escherichia coli for priming on single-
stranded DNA introduced by conjugation. The
presence of such a system had been suggested some
time ago, by Lanka and Barth (1981), who also
proposed that Salmonella enterica sv. typhimurium
might be less able to recover the entering DNA.
We therefore repeated our experiments, but used
Salmonella donors and recipients. We found that
in this species test plasmids lacking oriV were very
poorly transferred, but that the frequency was in-
creased significantly when the plasmid contained
an R1162 primase initiation site, oriented for
replacement synthesis in the recipient. Further-
more, the R1162 primase was also required, con-
firming that it was the plasmid priming system
that was initiating the replacement strands. To
show this, we took advantage of the fact that both
the R1162 and pSC101 relaxases are active on the
pSC101 oriT, although only the R1162 protein
has a linked primase. Test plasmids that contained
the pSC101 oriT were mobilized efficiently by both
the pSC101 and R1162 relaxases (in the presence of
their associated accessory proteins) in E. coli · E.
coli matings, whether or not oriV was present on
the transferring molecule. In contrast, in matings
involving Salmonella, the R1162 relaxase, having
its associated primase, and a test plasmid with the
origin of replication were both required for efficient
transfer. Our data suggest that the plasmid priming
system can substitute for the host system in the pro-
cess of plasmid recovery following transfer. This
back-up capability could well contribute to the
broad host-range of plasmids such as R1162.
The R1162 relaxase is a member of a large and
widely distributed family of Mob proteins (Fran-
cia et al., 2004). The corresponding real or putative
oriTs, a few of which are shown in Fig. 1, are
structurally similar, consisting of a highly con-
served region, which we term the core, and an in-
verted repeat that differs in size and sequence.
Single-stranded DNA consisting of only the core
is correctly cleaved by the relaxase (Scherzinger
et al., 1993), although the protein binds poorly
to this DNA in a gel-shift assay (unpublished).
The outer arm of the inverted repeat is not re-
quired for strand cleavage in the relaxosome but
is required for termination (Kim and Meyer,
1989). The clear implication, supported by genetic
and DNA–protein binding assays (Bhattacharjee
et al., 1992), is that the inverted repeat forms a
hairpin loop when the DNA is single-stranded
and that this structure substitutes for the inner
arm in duplex DNA. The variability of the in-
verted repeats suggests that they arose indepen-
dently severaltimes
precursors. A clue to their origin is revealed in
the sequence of the pSC101 oriT (Fig. 2): part of
the core is present, in inverted orientation, adja-
cent to the outside arm of the inverted repeat. Sev-
eral other oriTs in the R1162 Mob family also
show remnants of the core (Becker and Meyer,
2003). This feature indicates that the oriT arose
from a progenitor element, consisting of a con-
served core and adjacent DNA, which was then
duplicated and inverted. The second core region
was subsequently inactivated by deletion. Possibly
this was necessary to prevent simultaneous cleav-
age of both strands, which could interfere with a
step in transfer. Alternatively, an intact, inverted
core could form an elongated hairpin after trans-
fer, blocking access to the 30end of the strand.
In vitro, such hairpins interfere with the cleavage
reaction at the nick site. In any case, the duplica-
tion is clearly a modification to allow the recircu-
larization of transferred DNA by recreating a
relaxase binding site at the trailing 30end of the
DNA. In agreement with this, the linear chromo-
some of Agrobacterium tumefaciens C58 contains
a gene encoding a protein highly homologous to
the R1162 relaxase, and an adjacent core sequence,
but the inverted repeat is lacking. In this case,
recircularization after transfer would not be
andfrom different
Fig. 2. The nicked strand of the pSC101 oriT, folded to show
the conservation of part of the core site in the inverted
orientation. The shaded region is required for transfer.
116
C. Parker et al. / Plasmid 53 (2005) 113–118

Page 5

required, and presumably any transferred DNA
would be rescued by recombination.
If oriTs in plasmids arise by duplication and
inversion, what are the elements from which they
are derived? One possibility, for the IncQ relaxase
family at least, is that these elements are part of a
cyclic process in which mobilization modules are
slowly but constantly being destroyed and then
re-created. Plasmids occasionally become inte-
grated into the chromosome and then decompose,
losing their distinct identity, with their remains
appearing as genomic islands. These events would
be part of the larger process that results in the con-
tinual fusion and resegregation taking place in the
universe of mobile elements, accounting for their
mosaic structure as suggested by Osborn and Bolt-
ner (2002). During this process, which can involve
additional transfer events, nonessential genes are
lost or are redirected to new functions, and rear-
rangements occur with neighboring DNA. How-
ever, the remaining genes can serve as a source
for the evolution of new plasmids, and in particu-
lar the acquisition of a relaxase gene would result
in a nascent system for mobilization. This system
would evolve toward efficient transfer by muta-
tions that define the relaxase–oriT pair, by con-
scription of other proteins to the developing
relaxosome, and by duplication and inversion of
the core, as described above.
The model assumes first that the IncQ relaxases
can be relatively nonspecific in the selection of nas-
cent oriTs. The R1162 relaxase can recognize and
cleave a set of DNAs varying in sequence, both
in the inverted repeat and in the core (Becker
and Meyer, 2000, 2003). Cleavage of single-
stranded DNA can be particularly nonspecific
(Scherzinger et al., 1992) and supercoiled DNA
might have sufficient single-strandedness to be an
occasional target for nicking. Second, there must
be a supply of relaxase genes on the chromosomes
of bacteria that can be recruited for new mobiliza-
tion systems. The genomes of Xanthomonas axo-
nopodis pv. citri and Erwinia carotovora subsp.
atroseptica each contain two genes encoding pro-
teins highly similar to the R1162 relaxase, and
there is one in Photorhabdus luminescens and
E. coli CFT073 (Bell et al., 2004; da Silva et al.,
2002; Duchaud et al., 2003; Welch et al., 2002).
In each case, the relaxase is not simply part of a
complete, integrated plasmid that has formed an
Hfr-like strain, since there is no obvious oriT or
core associated with the gene. In Erwinia, each
relaxase gene is located in a putative genomic
island derived from an integrated plasmid. It is
striking that the relaxase is conserved within the is-
land, without any trace of an oriT, which in plas-
mids is normally closely linked to the relaxase
gene. The relaxases might be continuing to serve
in gene transfer on a remote origin, but in any case
they remain available to participate in the evolu-
tion of new, active oriTs.
Acknowledgment
Our work was supported by a grant from the
National Institutes of Health (GM37462).
References
Becker, E., Meyer, R., 2000. Recognition of oriT for DNA
processing at termination of a round of conjugal transfer. J.
Mol. Biol. 300, 1067–1077.
Becker, E.C., Meyer, R., 2002. MobA, the DNA strand
transferase of plasmid R1162. The minimal domain required
for DNA processing at the origin of transfer. J. Biol. Chem.
277, 14575–14580.
Becker, E., Meyer, R., 2003. Relaxed specificity of the R1162
nickase: a model for evolution of a system for conjugative
mobilization of plasmids. J. Bacteriol. 185, 3538–3546.
Bell, K., Sebaihia, M., Pritchard, L., Holden, M., Hyman, L.,
Holeva, M., Thomsom, N., Bentley, S., Churcher, L.,
Mungall, K., Atkin, R., Bason, N., Brooks, K., Chilling-
worth, T., Clark, K., Doggett, J., Fraser, A., Hance, Z.,
Hauser, H., Jagels, K., Moule, S., Norbertczak, H.,
Ormond, D., Price, C., Quail, M., Sanders, M., Walker,
S., Whitehead, S., Salmond, G., Birch, P., Parkkhill, J.,
Toth, I., 2004. Genome sequence of the enterobacterial
phytopathogen Erwinia carotovora subsp. atroseptica and
characterization of virulence factors. Proc. Natl. Acad. Sci.
USA 101, 11105–11110.
Bhattacharjee, M., Rao, X.M., Meyer, R.J., 1992. Role of the
origin of transfer in termination of strand transfer during
bacterial conjugation. J. Bacteriol. 174, 6659–6665.
Brasch, M.A., Meyer, R.J., 1986. Genetic organization of
plasmid R1162 DNA involved in conjugative mobilization.
J. Bacteriol. 167, 703–710.
da Silva, A., Ferro, J., Reinach, F., Farah, C., Furlan, L.,
Quaggio, R., Monteiro-Vitorello, C., Van Sluys, M.,
Almeida, N., Alves, L., do Amaral, A., Bertolini, M.,
C. Parker et al. / Plasmid 53 (2005) 113–118
117

Page 6

Camargo, R.J., Camarotte, G., Cannavan, F., Cardozo, J.,
Chambergo, F., Ciapina, L., Cicarelli, R., Coutinho, L.,
Cursino-Santos, J., El-Dorry, H., Faria, J., Ferreira, A.,
Ferreira, R., Ferro, M., Formighieri, E., Franco, M.,
Greggio, C., Gruber, A., Katsuyama, A., Kishi, L., Leite,
E., Lemos, E., Lemos, M., Locali, E., Machado, M.,
Madeira, A., Martinez-Rossi, N., Martins, E., Meidanis, J.,
Menck, C., Miyaki, C., Moon, D., Moreira, L., Novo, M.,
Okura, V., Oliveira, M., Oliveira, V., Pereira, H., Rossi, A.,
Sena, J., Silva, C., de Souza, R., LA, S., Takita, M., Tamura,
E., Teixeira, E., Tezza, R., Trindade dos Santos, M., Truffi,
S., Tsai, S., White, F., Setubal, J., Kitajima, J., . Comparison
of the genomes of two Xanthomonas pathogens with differing
host specificities. Nature 417, 459–463.
Duchaud, E., Rusniok, C., Frangeuil, L., Buchreiser, C.,
Givaudan, A., Taourit, S., Bocs, S., Boursauz-Eude, C.,
Chandler, M., Charles, J.-F., Dassa, E., Derose, R.,
Derzelle, S., Freyssinet, G., Gaudriault, S., Medigue, C.,
Lanois, A., Powell, K., Siguier, P., Vincent, R., Wingate, V.,
Zouine, M., Glaser, P., Boemare, N., Danchin, A., Kunst,
F., 2003. The genome sequence of the entomopathogenic
bacterium Photorhabdus luminescens. Nat. Biotechnol. 21,
1307–1313.
Francia, M., Varsaki, A., Garcillan-Barcia, M., Latorre, A.,
Drainas, C., de la Cruz, F., 2004. A classification scheme for
mobilization regions of bacterial plasmids. FEMS Micro-
biol. Rev. 28, 79–100.
Fu, Y., Tsai, M.-M., Luo, Y., Deonier, R., 1991. Deletion
analysis of the F plasmid oriT locus. J. Bacteriol. 173, 1012–
1020.
Henderson, D., Meyer, R., 1996. The primase of broad host-
range plasmid R1162 is active in conjugal transfer. J.
Bacteriol. 178, 6888–6894.
Honda, Y., Sakai, H., Komano, T., Bagdasarian, M., 1989.
RepB0is required in trans for the two single-strand DNA
initiation signals in oriV of plasmid RSF1010. Gene 80,
155–159.
Kim, K., Meyer, R.J., 1989. Unidirectional transfer of broad
host-range plasmid R1162 during conjugative mobilization.
Evidence for genetically distinct events at oriT. J. Mol. Biol.
208, 501–505.
Lanka, E., Barth, P.T., 1981. Plasmid RP4 specifies a deoxy-
ribonucleic acid primase involved in its conjugal transfer
and maintenance. J. Bacteriol. 148, 769–781.
Lin, L.S., Meyer, R.J., 1987. DNA synthesis is initiated at two
positions within the origin of replication of plasmid R1162.
Nucleic Acids Res. 15, 8319–8331.
Matson, S., Nelson, W., Morton, B., 1993. Characterization of
the reaction product of the oriT nicking reaction catalyzed
by Escherichia coli DNA helicase I. J. Bacteriol. 175, 2599–
2606.
Meyer, R., 2000. Identification of the mob genes of plasmid
pSC101 and characterization of a hybrid pSC101-R1162
system for conjugal mobilization. J. Bacteriol. 182, 4875–
4881.
Matson, S., Sampson, J., Byrd, D., 2001. F plasmid conjugative
DNA transfer. The TraI helicase is essential for DNA
strand transfer. J. Biol. Chem. 276, 2372–2379.
Miele, L., Strack, B., Kruft, V., Lanka, E., 1991. Gene
organization and nucleotide sequence of the primase region
of IncP plasmids RP4 and R751. DNA Seq. 2, 145–162.
Nelson, W.C., Howard, M.T., Sherman, J.A., Matson, S.W.,
1995. The traY gene product and integration host factor
stimulate
Escherichiacoli
nicking at the F plasmid oriT. J. Biol. Chem. 270,
28374–28380.
Osborn, A., Boltner, D., 2002. When phage, plasmids, and
transposons collide: genomic islands, and conjugative- and
mobilizable-transposons as a mosaic continuum. Plasmid
48, 202–212.
Pansegrau, W., Ziegelin, G., Lanka, E., 1990. Covalent
association of the traI gene product of plasmid RP4 with
the 50-terminal nucleotide at the relaxation nick site. J. Biol.
Chem. 265, 10637–10644.
Pansegrau, W., Schroder, W., Lanka, E., 1993. Relaxase (TraI)
of the IncPa plasmid RP4 catalyzes a site-specific cleaving-
joining reaction of single-stranded DNA. Proc. Natl. Acad.
Sci. USA 90, 2925–2929.
Parker, C., Meyer, R., 2002. Selection of plasmid molecules for
conjugative transfer and replacement strand synthesis in the
donor. Mol. Microbiol. 46, 761–768.
Scherzinger, E., Lurz, R., Otto, S., Dobrinski, B., 1992. In vitro
cleavage of double- and single-stranded DNA by plasmid
RSF1010-encoded mobilization proteins. Nucleic Acids
Res. 20, 41–48.
Scherzinger, E., Kruft, V., Otto, S., 1993. Purification of the
large mobilization protein of plasmid RSF1010 and char-
acterization of its site-specific DNA-cleaving/DNA-joining
activity. Eur. J. Biochem. 217, 929–938.
Scholz, P., Haring, V., Wittmann-Liebold, B., Ashman, K.,
Bagdasarian, M., Scherzinger, E., 1989. Complete nucleo-
tide sequence and gene organization of the broad-host-
range plasmid RSF1010. Gene 75, 271–288.
Welch, R., Burland, V., Plunkett, G.R., Redford, P., Roesch,
D., Rasko, D., Buckles, E., Liou, S., Boutin, A., Hackett, J.,
Stroud, D., Mayhew, G., Rose, D., Zhou, S., Schwartz, D.,
Perna, N., Mobley, H., Donnenberg, M., Blattner, F., 2002.
Extensive mosaic structure revealed by the complete genome
sequence of uropathogenic Escherichia coli. Proc. Natl.
Acad. Sci. USA 99, 1720–1724.
Zhang, S., Meyer, R.J., 1995. Localized denaturation of oriT
DNA within relaxosomes of the broad host-range plasmid
R1162. Mol. Microbiol. 17, 727–735.
Zhang, X., Zhang, S., Meyer, R., 2003. Molecular handcuffing
of the relaxosome at the origin of conjugative transfer of the
plasmid R1162. Nucleic Acids Res. 31, 4762–4768.
DNA helicase I-catalyzed
118
C. Parker et al. / Plasmid 53 (2005) 113–118