Specific mutations within the AT-rich region of a plasmid replication origin affect either origin opening or helicase loading.

Abstract

Prokaryotic and eukaryotic replicons possess a distinctive region containing a higher than average number of adenine and thymine residues (the AT-rich region) where, during the process of replication initiation, the initial destabilization (opening) of the double helix takes place. In many prokaryotic origins, this region consists of repeated 13-mer motifs whose function has not yet been specified. Here we identify specific mutations within the 13-mer sequences of the broad-host-range plasmid RK2 that can result in defective origin opening or that do not affect opening but induce defects in helicase loading. We also show that after the initial recruitment of helicase at the DnaA-box sequences of the plasmid origin, the helicase is translocated to the AT-rich region in a reaction requiring specific sequence of the 13-mers and appropriate facing of the origin motifs. Our results demonstrate that specific sequences within the AT-rich region of a replication origin are required for either origin opening or helicase loading.

Full text

Page 1

Specific mutations within the AT-rich region
of a plasmid replication origin affect either
origin opening or helicase loading
Magdalena Rajewska†, Lukasz Kowalczyk†, Grazyna Konopa‡, and Igor Konieczny†§
†Department of Molecular and Cellular Biology, Intercollegiate Faculty of Biotechnology and‡Department of Molecular Biology, Faculty of Biology,
University of Gdansk, 24 Kladki, 80-822, Gdansk, Poland
Communicated by Donald R. Helinski, University of California at San Diego, La Jolla, CA, June 17, 2008 (received for review October 15, 2007)
Prokaryotic and eukaryotic replicons possess a distinctive region
containing a higher than average number of adenine and thymine
residues (the AT-rich region) where, during the process of replica-
tion initiation, the initial destabilization (opening) of the double
helix takes place. In many prokaryotic origins, this region consists
of repeated 13-mer motifs whose function has not yet been
specified. Here we identify specific mutations within the 13-mer
sequences of the broad-host-range plasmid RK2 that can result in
defective origin opening or that do not affect opening but induce
defects in helicase loading. We also show that after the initial
recruitment of helicase at the DnaA-box sequences of the plasmid
origin, the helicase is translocated to the AT-rich region in a
reaction requiring specific sequence of the 13-mers and appropri-
ate facing of the origin motifs. Our results demonstrate that
specific sequences within the AT-rich region of a replication origin
are required for either origin opening or helicase loading.
13-mer sequences ? DNA replication ? DnaB helicase ? plasmid RK2
T
specific helicase loading. Binding of the initiator (Rep), a single
protein, or a complex of several proteins to the replication origin
results in the melting of the double-stranded DNA helix in a
region of the origin with low internal thermodynamic stability
becauseofahigherthanaveragecontentofadenineandthymine
residues (AT-rich region). In plasmid replicons, the involvement
of both the host-encoded replication initiator DnaA and the
plasmid encoded Rep protein is essential for origin melting,
helicaserecruitment,andhelicaseloadingontothessDNAinthe
opened AT-rich region.
DNA replication of the broad-host-range plasmid RK2 is
initiated by the binding of the plasmid-encoded protein TrfA to
direct repeats (iterons) present at its replication origin (oriV), a
process similar to that of other plasmid systems (1, 2). The
formation of an open complex at oriV by the TrfA initiation
protein requires HU and is stabilized by the host DnaA protein
(3). In contrast, the initiation of replication at the Escherichia
coli chromosomal origin, oriC, involves a single initiator, DnaA,
binding to DnaA-box sequences and resulting in duplex DNA
melting (4–7). During replication initiation at RK2 oriV, helix
destabilization occurs within the AT-rich region, which includes
four 13-mer repeated sequences (L, M1, M2, R) that exhibit a
DNA consensus sequence of TAAACnTTnTTTT (3). Similar
motifs have been identified in a majority of theta-replicating
plasmids and bacterial chromosomes including E. coli, where
three 13-mers (L, M and R) have been identified in oriC (4–6).
The position of the 13-mer clusters and their specific sequences
are important for proper functioning of both oriV (8, 9) and oriC
(5, 10, 11). Although the importance of 13-mers is well estab-
lished, their exact role is not completely understood.
Certain E. coli proteins, including SeqA (12, 13), IciA (11),
and DnaA (14, 15), specifically interact with the oriC AT-rich
region. In addition, DnaB helicase and DnaC interact with
he initiation of DNA synthesis requires a mechanism to
coordinate DNA unwinding, helicase recruitment, and site-
ssDNA in the open complex (16, 17). Helicase loading at the
opened oriC (18) requires the interaction of DnaA with DnaB
(19). The nature of the interaction of DnaA with the AT-rich
region is unclear. Earlier evidence indicated that DnaA, when
bound with ATP, interacts with six-nucleotide sequences,
termed ATP-DnaA boxes, within the AT-rich region (15, 20).
More recent data suggest that the ATP-DnaA boxes are not the
sites for DnaA interaction within the 13-mer region (21). When
RK2 DNA replication is initiated in E. coli, DnaB helicase is
recruited by DnaA to dsDNA containing oriV DnaA-box se-
quenceslocatedupstreamofthe13-mercluster(22).ATP-DnaA
boxes are not present in oriV (3, 9). It has been shown that the
TrfA protein is essential for the helicase complex activity (23);
however, the mechanism of this activation has not yet been
elucidated.
In this work, mutations in the 13-mers that result in defects in
helicase loading, but not in origin opening, have been charac-
terized and used for the analysis of helicase complex formation
at the replication origin of the broad-host-range plasmid RK2.
Results and Discussion
DNA Unwinding Activity of E. coli DnaB Helicase Is Affected by
Mutations in the AT-Rich Region. In our previous study, 26 plasmid
mutants with an altered sequence, position, or spacing of the
13-mer motifs of the RK2 replication origin were constructed
and analyzed for replication activity (9). The data showed that
both the sequence and the position of 13-mers are critical for
oriV functionality. To determine what particular stage during the
initiation of plasmid replication requires 13-mer sequence spec-
ificity, we analyzed in vitro the individual steps of replication
initiation in these oriV mutants.
E. coli DnaB helicase complex activity on miniRK2 plasmid
templates with either wild-type or altered 13-mer sequences was
identified electrophoretically by the formation of a substantially
unwound form of the supercoiled plasmid DNA, designated FI*
(23). The formation of the FI* form is a result of the combined
activities of plasmid and bacterial replication initiation proteins,
including DnaA, DnaC, gyrase, SSB, HU, and TrfA. We opti-
mized the reaction conditions so that the helicase activity was
?50% on the wild-type template, as measured by densitometry
scan of the gel (Fig. 1 lane 1). This allowed us to observe either
anincreaseorareductionofhelicaseactivityoftheoriVmutants.
Thirteen mutants exhibiting no or substantially reduced helicase
activity were selected (Fig. 1) (Table 1). No mutants with
increased helicase activity in comparison to the wild-type tem-
plate were identified. Because the TrfA protein exists in two
forms: 33 kDa, allowing for replication in E. coli and 44 kDa,
Author contributions: M.R. and I.K. designed research; M.R., L.K., and G.K. performed
research; M.R., L.K., G.K., and I.K. analyzed data; and M.R. and I.K. wrote the paper.
The authors declare no conflict of interest.
§To whom correspondence should be addressed. E-mail: igor@biotech.ug.gda.pl.
© 2008 by The National Academy of Sciences of the USA
11134–11139 ?
PNAS ?
August 12, 2008 ?
vol. 105 ?
no. 32www.pnas.org?cgi?doi?10.1073?pnas.0805662105

Page 2

essential for replication in Pseudomonas, we performed FI*
experiments by using either the short or long version of the
protein. The effects were similar, regardless of the version of
TrfA protein present in the reaction (data not shown); therefore,
in subsequent experiments only the TrfA 33kDa form of the
protein was used. The substitution of 13-mer M2 with M1
(mutant M2-M1) resulted in low helicase activity (Fig. 1, lane 3),
whereas no activity was detected with mutant R-M2 (Fig. 1, lane
2). Deletion (Fig. 1, lanes 4 and 5) and insertion (Fig. 1 lanes
9–11) mutants exhibited only faint FI* activity. Helicase was also
inactive on DNA templates where the sequence, but not AT-
richness (the AT/GC ratio), was changed (Fig. 1, lanes 6–8). The
substitution of the right (R) or the middle (M2) 13-mer with
chromosomal E. coli oriC 13-mer (mutants R-C and M2-C,
respectively) reduced the activity of DNA helicase, whereas no
activity was detected if the left (L) 13-mer was substituted (Fig.
1, lanes 12–14).
Our results demonstrated that the lack of replication activity
of the analyzed oriV minireplicons with altered 13-mer se-
quences could be a result of disturbances in helicase-dependent
template unwinding. Because the formation of the FI* is a result
of origin opening, helicase recruitment, and finally helicase
complex activity, we could not exclude the possibility that the
observed disturbances in helicase activity were due to failures at
earlier steps of replication initiation, including open complex
formation.
Open Complex Formation Is Affected by Sequence Changes Within the
AT-Rich Region. Full RK2 origin opening occurs in the presence
of HU, TrfA, and DnaA proteins (3). To explore the possibility
that mutations within the AT-rich region reduce DNA helix
destabilization, we performed potassium permanganate
(KMnO4) footprinting and compared the extent of origin open-
ing for the wild-type and mutated oriV templates. KMnO4reacts
preferentially with pyrimidines in single-stranded DNA, modi-
fying primarily thymines. Specific modification of plasmid su-
percoiled DNA can be detected by the termination of a primer
extension reaction from a32P-labeled primer at the modified
residues. Because the majority of thymine residues present
within the AT-rich region of RK2 origin are located on the top
DNA strand, the primer extension analysis was carried out only
on that strand. The pattern of KMnO4modifications of the top
strandoftheplasmidRK2wild-typeoriginisshowninFig.2.The
footprinting experiments were carried out in the presence of
eitherTrfAandDnaAproteinoronlyTrfA.Foreachseries(Fig.
2 A–D), wild-type template analysis was performed as a control.
The experiments show that for six mutants (L-C, TAA1–
3ATT, A2T, ?3, ?6, ?11), the extent of opening was signifi-
cantly reduced (Fig. 2 A, C, and D). For all these mutants, the
DNA origin destabilization occurred only at the right 13-mer
(R). Interestingly, the origin opening was reduced even in
mutants with unchanged internal thermodynamic stability
(TAA1–3ATT and A2T), indicating that this is not the only
criterion for the functionality of AT-rich region. Reduction of
the extent of origin opening was observed for two deletion
mutants, ?R and ?M2 (Fig. 2B). The formation of full-length
open complex that covers the whole AT-rich region (all four
TTT11-13AAA
125463789 10 11 12 13 14
60
40
20
0
FI* (%)
wt
R-M2
M2-M1
∆R
∆M2
A2T
TAA1-3ATT
+3
+6
+11
R-C
M2-C
L-C
FI*
FI
Fig. 1.
activity on oriV mutant templates was determined by using the FI* assay. The
position of the FI* DNA band, the extensively unwound covalently closed
circular DNA generated by helicase activity on the supercoiled (FI) template
DNA, is indicated by an arrow. The reactions were performed as described in
Materials and Methods. DnaB at 600 ng and DnaC at 20 ng were used to
optimizedthereactionsothatthehelicaseactivitywas?50%onthewild-type
template.BlackcolumnsindicatetheconversiontoFI*form,calculatedonthe
basis of densitometry analysis, as the percentage of total DNA in the reaction.
DnaB helicase activity at oriV mutant templates. E. coli DnaB helicase
Table 1. Effect of mutations within 13-mer motifs and oriV insertion mutants on origin replication activity
oriV mutantSequence of mutated AT-rich region or insertion
In vivoIn vitro
TFFII FI*KMnO4
EM
wt
R-C
M2-C
L-C
R-M2
M2-M1
?R
?M2
A2T
TAA1–3ATT
TTT11–13AAA
?3
?6
?11
oriV3–6
oriV3–11
TAACCTGTCTTTTAACCTGCTTTTAAACCAATATTTATAAACCTTGTTTT
TAACCTGTCTTTTAACCTGCTTTTAAACCAATATTTAGATATCTTATTAG
TAACCTGTCTTTTAACCTGCTTTTGATATCTTATTAGTAAACCTTGTTTT
GATATCTTATTAGAACCTGCTTTTAAACCAATATTTATAAACCTTGTTTT
TAACCTGTCTTTTAACCTGCTTTTAAACCAATATTTATAAATATTGGTTT
TAACCTGTCTTTTAACCTGCTTTTAAAAAGCAGGTTATAAACCTTGTTTT
TAACCTGTCTTTTAACCTGCTTTTAAACCAATATTTATAA. . . . . . . . . .
TAACCTGTCTTTTAACCTGCTTTTA. . . . . . . . . . . . TAAACCTTGTTTT
TTACCTGTCTTTTTACCTGCTTTTAAACCAATATTAATTAACCTTGTTTT
ATTCCTGTCTTTTTTCCTGCTTTTAAACCAATATTTAATATTCTTGTTTT
TAACCTGTCTAATAACCTGCTTAAATTCCAATATTTATAAACCTTGTAAA
AGA
GAATTC
GTAGATACAAG
GAATTC
GAATTCGTCAG
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?/?
?
nt
?
?
nt
nt
nt
nt
?
nt
nt
nt
?
?
?/?
?
?
?/?
?
?
?
?
?
?
?
?
?
?
?/?
?
?
?
?
?
?
?
?
?
?
?/?
?/?
?
?
?
?
?
?
?
nt
All mutants were constructed and isolated previously (8, 9). In the mutants shown, DNA sequence letters in bold indicate changes from the wild-type,
underlinedlettersindicatepermanganate-modifiablethyminebases,anddottedlinescorrespondtodeletedregions.Inmutants?3,?6,and?11theindicated
sequence was inserted between the 13-mers. In mutants oriV3–6 and oriV3–11, the indicated sequence was inserted between the DnaA-box region and the
AT-rich region (8). TF and FII results were presented previously (8, 9). FI*, KMnO4and EM results are from this work except KMnO4 with oriV3–6 (8). TF,
transformation frequency; FII, replication activity in E. coli crude extract; FI*, DnaB helicase-mediated template unwinding; KMnO4, local strand opening; EM,
helicase loading to the AT-rich region; ?, activity of the altered origin similar in comparison to the wild-type origin activity; ?/?, lower activity; ?, no (or
substantially reduced) activity; nt, not tested.
Rajewska et al.PNAS ?
August 12, 2008 ?
vol. 105 ?
no. 32 ?
11135
BIOCHEMISTRY

Page 3

13-mers) was observed for five mutants. Two of them, R-M2 and
M2-M1, displayed an unusual feature. Partial origin destabili-
zation appeared even in the absence of initiation proteins (Fig.
2B). For three other mutants R-C, M2-C, and TTT11–13AAA
(Fig. 2 A and C), the extent of open complex formation was
similar to that of the wild-type template. Similar to wild-type, in
these mutants the open complex was formed also in the absence
of DnaA protein, but it occurred at the left part of the AT-rich
region(Fig.2AandC).Eventhoughtheextentoforiginopening
is the same for all these five mutants, a difference was observed
in the pattern of permanganate modifications. This difference,
however, can be attributed to alterations in nucleotide sequence,
especially the thymine residues, of the analyzed mutants.
Our results demonstrate that the majority of the analyzed
mutations within the AT-rich region have a significantly reduced
plasmid open complex formation. This is similar to the situation
in which introduced mutations within the E. coli oriC AT-rich
region decreased or prevented the origin opening (11). Surpris-
ingly, however, our analysis of oriV mutants (R-C, M2-C, R-M2,
M2-M1 and TTT11–13AAA) revealed that the extent of origin
opening was the same as for the wild-type template. This raised
the possibility that low or absence of helicase activity of these
mutants could be the result of disturbances of helicase complex
loading at the open AT-rich region.
Mutations Within the 13-mers Affect Helicase Loading at the Open
AT-Rich Region. For the analysis of helicase complex formation at
the RK2 origin, we used in vitro cross-linking followed by
electron microscopy (EM) analysis (see Materials and Methods).
This approach allowed identification of the position of the
helicase complex on the replication origin. To identify the
optimal helicase concentration for these experiments, we used
various DnaBC concentrations to analyze the unwinding of the
wild-type oriV template (Fig. 3 A and B). We determined a
saturating helicase concentration (600 ng per reaction), in which
complete conversion to the FI* form was observed. The com-
plete conversion indicates that a vast majority of plasmid par-
ticlesweresuccessfullyloadedbythehelicasecomplex.Whenthe
oriV mutants with an unchanged origin opening but reduced
helicase activity (R-C, M2-C, R-M2, M2-M1, and TTT11–
13AAA) (see above) were analyzed, the highest level of FI*
conversion was ?50% (Fig. 3 A and B). Only very limited or no
template unwinding activity was detected on the TTT11–
13AAA and R-M2 mutant templates (Fig. 3 A and B). Despite
a high molar excess (100-fold) of DnaBC in comparison to the
supercoiled DNA template, no helicase activity was detected on
the R-M2 template (Fig. 3A, lane 7).
The above experiments allowed identification of the optimal
DnaBC concentration (gray rectangle on Fig. 3B) to use for
cross-linking and EM analysis (see below). Also, these results
demonstrated that the inhibition of helicase unwinding activity
as a result of alterations of 13-mers could not, or only to a limited
extent, be compensated for by increasing the concentration of
the DnaBC complex.
It has been shown that TrfA interacts with DnaB helicase
recruited via interaction with DnaA protein at DnaA-boxes in
oriV, located upstream of the AT-rich sequence (22). Therefore
TrfA could possibly play a role in the translocation of the
helicase complex from the DnaA-boxes to the AT-rich region.
To explore this possibility, we assembled the reactions by using
saturating concentrations of DnaBC as described above for FI*
analysis.TheutilizationoftheDnaBK236A(18)mutantprotein,
which forms nucleoprotein complexes but does not unwind DNA
template, allowed for the analysis of the position of the helicase
complex in the plasmid origin. After cross-linking with formal-
dehyde, the localization of the helicase complex on the 792 bp
wild-type oriV fragment (Fig. 4A) was determined (see Materials
and Methods). In a control reaction, the formation of the
nucleoprotein complex containing DnaA, DnaB, and DnaC
proteins, but not TrfA, was analyzed. Statistical analysis con-
firmed our previously published observation (22) that the ma-
jorityofDnaBhelicase(95%)wasfoundattheDnaA-boxes(Fig.
4 B(i), C, and D). Only in 5% of the analyzed DNA particles was
DnaB detected within the AT-rich region. When TrfA protein
was added to the reaction mixture, the proportion changed
significantly. In 58% of the plasmid particles, helicase was
detected in the oriV fragment corresponding to the AT-rich
region (Fig. 4 B(ii), C, and D). This result demonstrates trans-
location of the helicase complex from the DnaA-boxes to the
AT-richregion.BecausetheadditionofTrfAresultsintheorigin
opening (ref. 3 and above) and the release of helicase activity
(ref. 23 and above), we concluded that the observed DnaB
complex was most probably formed at the open origin structure.
After the addition of TrfA, the number of helicase complexes at
the DnaA-box region decreased substantially (Fig. 4). Because
TrfA interaction with DnaB helicase has been reported (22) and
wt
- +
+
+3
- +
+
+6
- +
+
+11
- +
+
-
+
--
+
--
+
- -
+
-
1243
R-M2
- +
+
∆ ∆ ∆ ∆R
- +
+
∆ ∆ ∆ ∆M2
- +
+
wt
- +
+
M2-M1
- +
+
-
+
- -
+
- -
+
- -
+
- -
+
-
45321
ABCD
wt
- +
+
L-C
- +
+
R-C
- +
+
M2-C
- +
+
L
M1
M2
R
DnaA
TrfA
-
+
--
+
--
+
- -
+
-
1234
wt
- +
A2T
- +
+
TAA1-
3ATT
- +
+
TTT11-
13AAA
- +
+
-
+
+
- -
+
--
+
--
+
-
3412
Fig.2.
at 100 ng were present in the reactions as indicated. Dotted lines indicate the region where the DNA helix destabilization occurs. Position and orientation of
the 13-mers for wild-type oriV is depicted on the left.
StrandopeningofmutantoriVtemplates.PotassiumpermanganatefootprintingwascarriedoutasdescribedinMaterialsandMethods.TrfAandDnaA
11136 ?
www.pnas.org?cgi?doi?10.1073?pnas.0805662105Rajewska et al.

Page 4

it also has been shown that TrfA interacts with the helicase
complex formed on the DnaA-boxes (22), we propose direct
involvement of TrfA in the observed helicase complex translo-
cation. The interaction of E. coli DnaB helicase with other
plasmid initiating proteins has been demonstrated for plasmids
R6K (24) and pSC101 (25).
To complete our analysis, we tested the possibility that the
alterations within the oriV AT-rich region reduced the efficiency
of helicase loading at the open origin structure. We tested
assembly of the helicase on oriV templates with 13-mer alter-
ations (R-C, M2-C, R-M2, M2-M1, and TTT11–13AAA) that,
asdemonstratedabove,donotaffectoriginopeningbutdiminish
helicase activity. In comparison to the wild-type template,
assembly of the helicase complex at the AT-rich region was
substantially reduced in all analyzed mutants (Fig. 4 C and D).
ThereactionswerecarriedoutinthepresenceofTrfAandunder
the same experimental conditions as for the wild-type plasmid.
The lowest percentage (15%) of DnaB molecules in the AT-rich
region was observed for mutant R-M2. Although we did not
observe any specific pattern of 13-mers alterations affecting
helicase loading, the reduced helicase complex formation was
observed on templates, with changes introduced into the 13-mer
M2 or R located in the right part of the oriV AT-rich region
(Table 1). The reduction of helicase complex formation at the
AT-rich region of the mutant origins generally corresponded to
thereductionofhelicaseactivityobservedonthesametemplates
(Figs. 3 and 4D); however, for two mutants (R-M2 and TTT11–
13AAA) the result was not so striking. Cross-linking and utili-
zation of the DnaB K236A mutant could account for the
stabilization of helicase complexes at those templates.
Helicase Translocation Requires Appropriate Facing of the Origin
Motifs. Our EM studies showed that the position of the helicase
ontheDNAtemplatechangedasaresultoftheadditionofTrfA,
a process we defined as translocation. However, it hadn’t been
directly shown whether it was an actual change in the helicase
position on the DNA or a change in the specific site on the DNA
template that was in direct contact with the helicase. Several
possible mechanisms behind the observed process could be
considered. To address this issue, we conducted experiments by
using oriV mutants that contained insertions between the DnaA-
box region and the AT-rich region. The insertion of 6 bp
(oriV3–6 mutant) knocked out template activity in vivo (8),
which we identified as not due to an inhibition of origin opening
(8), but as the result of the failure of the helicase to translocate
and a consequent loss of helicase activity (Fig. 5 A and B). In
contrast to the 6-bp insertion, the introduction of an approxi-
mately full helical turn (oriV3–11 mutant) between the DnaA-
box region and the AT-rich region allowed helicase translocation
and activity in vitro (Fig. 5 A and B) and replication activity in
vivo (8). These results demonstrate that translocation of helicase
from the DnaA-boxes to the 13-mers requires appropriate facing
oftheoriginmotifs.Thisimpliesthathelicasetranslocationisnot
a spontaneous dissociation from its interaction with DnaA and
DnaA-boxes and establishment of contact with ssDNA within
the 13-mers. Also, it is very unlikely that the helicase slides from
one position, the DnaA-boxes, to another and thereby estab-
lishes contact with the opened AT-rich region. Obviously, heli-
case translocation requires some form of face-to-face contact
which most probably could be promoted by a specific change of
the origin DNA structure. Possibly, TrfA together with HU
changes the oriV nucleoprotein structure in such a way that
helicase contact with the ssDNA within the 13-mers is favored,
resulting in the release of helicase from the DnaA-boxes. This
mechanism would require TrfA interaction with DnaB (22) and
TrfAcooperationwiththehostDnaAprotein,whichrecruitsthe
helicase to DnaA-boxes (23) and is involved in stabilization of
origin melting by TrfA (3). The involvement of histone-like
proteins in this process is essential , as HU is required for origin
opening by TrfA (3). Also, TrfA, in an iteron- and/or 13-mer-
dependent manner, could possibly form filaments that restruc-
ture the DNA at the replication origin, promoting both origin
opening and helicase translocation. It was recently proposed for
plasmid pPS10 RepA protein that defined DNA sequences
promote the assembly of the protein into a distinct amyloid
nanostructure (26). TrfA, similar to pPS10 RepA, contains
regions that form filaments through ?-sheets, which we identi-
fied by using the TANGO algorithm (http://tango.crg.es).
The requirement for a specific sequence for helicase activity
within the AT-rich region could be a general feature of different
replicons. A selective activation by stretches of thymine residues
of the Mcm4/6/7 helicase complex has been reported and
proposed to be a determinant for the selection of DNA repli-
cation initiation sites in mammalian genomes (27). We demon-
strate that loading of the DnaB helicase, primarily recruited by
hostDnaAproteinatDnaA-boxes,dependsontheactivityofthe
RK2 plasmid replication initiator to facilitate helicase translo-
cation to the AT-rich opened region, in a reaction requiring
specific sequence of the 13-mers and appropriate facing of the
origin motifs.
wt
R-C
M2-C
R-M2
TTT11-
13AAA
M2-M1
DnaB(C)
1567243
-
conversion (%)
100
50
0
100
100
50
0
50
0
100
50
0
100
50
0
100
DnaB(C) (µg)
01 0.50.75 0.25
50
0
AB
Fig. 3.
The reactions were performed in the presence of E. coli DnaA and TrfA
proteins as described in Materials and Methods. (A) (lanes 1–7): Reactions
containing the increasing amounts of DnaBC helicase complex (0, 25, 50, 100,
300, 600, 900 ng). (B) Efficiency of conversion of supercoiled (FI) (opened
circles) to the extensively unwound FI* DNA (filled squares), calculated on the
basis of densitometry analysis. The gray bar indicates the saturating amount
of DnaBC helicase complex in the reaction.
Efficiency of unwinding by DnaB helicase at oriV mutant templates.
Rajewska et al. PNAS ?
August 12, 2008 ?
vol. 105 ?
no. 32 ?
11137
BIOCHEMISTRY

Page 5

Materials and Methods
Plasmids, Proteins, and Reagents. The plasmids containing mutations within
the AT-rich region were constructed as described previously (5). Plasmids
oriV3–6 and oriV3–11 are described in Doran et al. (8). For TrfA, preparations
ofthehistidine-taggedversionofthelargelymonomericTrfAmutantprotein
His-6-TrfA G254D/S267L, which exhibits a high specific activity for iteron
binding (28), was used. E. coli DnaA, wild-type DnaB, the K236A ATP-non
hydrolyzing mutant DnaB (18), and E. coli DnaC proteins were purified
according to published procedures (29–31). Commercially available proteins
and chemicals were as follows: SSB and T4 polynucleotide kinase from Pro-
mega; BSA (fraction V), creatine phosphate, creatine kinase, rNTPs, KMnO4,
and Protein A 10-nm colloidal gold labeled for EM analysis from Sigma;
Sepharose CL-4B from Amersham Pharmacia; and the MiniPrep DNA purifi-
cation kit from A&A Biotechnology. AmpliTaq Gold polymerase and [32P]ATP
were from Perkin-Elmer. A 24-mer (GCTTGCATGCCTGCAGGTCGACTC) used
for primer extension was from Thermohybaid.
Permanganate Footprinting and Primer Extension. Permanganate footprinting
was carried out as described (3), with modifications. The standard reaction
mixture (25 ?l) contained 8 ng of HU, 100 ng of E. coli DnaA, 300 ng of oriV
DNA,and100ngofHis-6-TrfAG254D/S267L.Eachsamplewaspurifiedfurther
by using a Gene Clean III kit (BIO101). The analysis was performed on the top
DNA strand with the 24-nucleotide primer, as described previously (3).
Helicase Activity Assay (FI*). The reactions were performed essentially as
described(23).Wild-typeandmutatedRK2supercoiledminireplicons(300ng)
were used as the DNA template. Unless indicated otherwise, the mixture
contained DnaA (10 ng), DnaB (600 ng), DnaC (120 ng), His-6-TrfA G254D/
S267L protein (100 ng), HU (5 ng), SSB (230 ng), and gyrase (120 ng). The
samples were electrophoresed at 25 V for 22 h, and the gel was stained with
ethidiumbromide.Thehelicaseactivitywascalculatedbydensitometry,using
BioRad Quantity One software. pUC19A supercoiled DNA (1000 ng) contain-
templates; HU; DnaA; DnaB; DnaC; and, if indicated, TrfA. (A) Schematic
depiction of an oriV SnaBI, 792-bp fragment with marked DnaA-boxes and
AT-rich region. (B) Typical EM images of DnaB helicase (indicated with an
arrow) at the DnaA-box sequences of oriV (i) or at the AT-rich region (ii). (C)
Histograms present measurements of the helicase complex position on the
analyzed DNA fragments. The black line indicates the midpoint between the
DnaA-boxes and the AT-rich region. The helicase complexes at these regions
areindicatedbylightgrayanddarkgray,respectively.Nindicatesthenumber
of particles taken for the EM analysis. Statistical analysis of the obtained
results is given in the table (D).
DnaA-boxA+T
SnaBI SnaBl
(0 bp)
(792 bp)
0 100%
% of DNA fragment length
wt (ABC)
wt (ABCT)
R-C (ABCT)
M2-C (ABCT)
R-M2 (ABCT)
M2-M1 (ABCT)
TTT11-13AAA (ABCT)
oriV mutant
95
42
67
63
85
73
74
in DnaAbox
5
58
33
37
15
27
26
in A+T
Helicase complex (%)
C
D
B
A
wt
(ABC)
N=139
0 2550
0
20
40
0 2550
0
20
40
wt
(ABCT)
N=125
R-C (ABCT)
N=193
0 2550
0
20
40
M2-C
(ABCT)
N=126
0 2550
0
20
40
R-M2
(ABCT)
N=129
M2-M1
(ABCT)
N=114
0 2550
0
20
40
TTT11-13AAA
(ABCT)
N=139
Number of DNA-helicase complexes
0 25 50
0
20
40
025 50
0
20
40
iii
Fig. 4.
AT-rich region. To analyze the position of the helicase complex at the oriV
origin, we performed cross-linking experiments and EM analysis as described
in Materials and Methods. Reactions contained wild-type or mutant oriV
Effect of 13-mers mutations on DnaB complex formation at the oriV
DnaB(C)---
12546378910 11 12
FI*
FI
wtoriV3-6 oriV3-11
B
A
oriV3-6 (ABC)
oriV3-6 (ABCT)
oriV3-11 (ABC)
oriV3-11 (ABCT)
oriV mutant
94
93
90
49
in DnaAbox
6
7
10
51
in A+T
Helicase complex (%)
Fig. 5.
motifs. (A) FI* reactions containing increasing amounts of the DnaBC helicase
complex (at 0, 100, 300, 600 ng) on wild-type or mutant templates containing
insertionsof6bp(oriV3–6)or11bp(oriV3–11)betweentheDnaA-boxregion
and the AT-rich region containing 13-mers. (B) Results obtained in EM exper-
iments performed for those mutants as described in Materials and Methods.
Helicase translocation requires appropriate facing of the origin
11138 ?
www.pnas.org?cgi?doi?10.1073?pnas.0805662105Rajewska et al.