A novel zinc-binding fold in the helicase interaction domain of the Bacillus subtilis DnaI helicase loader
Nucleic Acids Research
A novel zinc-binding fold in the helicase interaction domain of the Bacillus subtilis DnaI helicase loader
Karin V. Loscha 2
Kristaps Jaudzems 1
Charikleia Ioannou 0
Xun-Cheng Su 2
Flynn R. Hill 0
Gottfried Otting 2
Nicholas E. Dixon 0 2
Edvards Liepinsh 1
0 School of Chemistry, University of Wollongong , NSW 2522 , Australia
1 Institute of Organic Synthesis , Aizkraukles iela 21, Riga, LV 1006 , Latvia
2 Research School of Chemistry, Australian National University , Canberra ACT 0200 , Australia
The helicase loader protein DnaI (the Bacillus subtilis homologue of Escherichia coli DnaC) is required to load the hexameric helicase DnaC (the B. subtilis homologue of E. coli DnaB) onto DNA at the start of replication. While the C-terminal domain of DnaI belongs to the structurally wellcharacterized AAA+ family of ATPases, the structure of the N-terminal domain, DnaI-N, has no homology to a known structure. Three-dimensional structure determination by nuclear magnetic resonance (NMR) spectroscopy shows that DnaI presents a novel fold containing a structurally important zinc ion. Surface plasmon resonance experiments indicate that DnaI-N is largely responsible for binding of DnaI to the hexameric helicase from B. stearothermophilus, which is a close homologue of the corresponding much less stable B. subtilis helicase.
DNA replication in all organisms is carried out by
replisomes, multiprotein machines that contain a ring-shaped
DNA helicase for separation of the strands of the DNA
duplex (1). In bacteria, the helicase forms a hexameric
structure that encircles single-stranded (ss) DNA. Several
other replication proteins bind to the helicase and its
loading onto DNA is one of the earliest steps in replisome
assembly (2). In Escherichia coli, loading of the DnaB
helicase hexamer onto ssDNA is assisted by DnaC, a
loader protein that associates with it to form a ring of
six DnaC molecules on one face of the helicase ring (3).
In Bacillus subtilis and some other Gram positive bacteria
such as Staphylococcus aureus, the corresponding
hexameric helicase is named DnaC and the helicase loader protein
is DnaI. In contrast to E. coli DnaB, which is loaded onto
ssDNA by ring opening (4,5), the B. subtilis helicase
dissociates more readily into monomers and the role of
DnaI is to assist its assembly on the ssDNA (6). This
process is assisted in B. subtilis by a pair of DNA-remodelling
co-loader proteins (called DnaB and DnaD), which guide
the DnaC/DnaI complex to specific sites in the DNA
(7–13). However, in vitro and in the presence of ATP,
DnaI alone is sufficient for loading of the helicase onto
The B. subtilis helicase and its loader form a complex
of six helicase and six DnaI molecules in analogy to
the E. coli DnaB/DnaC complex (6). The amino-acid
sequences of the helicase loaders from the two species
indicate the presence of nucleotide-binding sites, in
agreement with the requirement of ATP for helicase loading
(1,15–17). In addition, experiments with DnaI (14)
showed that it consists of two structured domains. The
larger C-terminal AAA+ domain (1,2,16) contains
the nucleotide-binding site and a cryptic site for ssDNA
binding, whereas the N-terminal domain (here termed
DnaI-N) seems to be primarily responsible for helicase
binding and acts as a molecular switch that regulates the
accessibility of the ssDNA-binding site in the C-terminal
domain (14). DnaI is a larger protein than E. coli DnaC
(36 versus 28 kDa) and DnaI-N is unrelated in sequence
to the N-terminal portion of the E. coli homologue (15).
As the B. subtilis DnaC helicase is a less stable protein that
can exist as a mixture of oligomeric forms, its interactions
with DnaI have mostly been inferred using the
wellbehaved stable hexameric helicase from B.
stearothermophilus (in the following referred to as Bst DnaB), which
shares 82% sequence identity and 92% similarity with the
B. subtilis helicase.
The interaction between DnaI and Bst DnaB depends
on the ability of DnaI to bind zinc (14). The zinc-binding
site is located in DnaI-N and is lost in the mutants C67A,
C70A and H84A. Remarkably, single mutations of the
other two cysteine residues, C76 and C101, did not
completely abolish zinc binding or helicase interaction,
suggesting that side chains of these two residues might
substitute for each other as zinc ligands.
To date, very little is known of the structural basis of
interactions of helicase loaders with their helicase
partners, and this has limited understanding of the
mechanisms of helicase loading. Although no complete
structure of any helicase loader has been determined to
date at atomic resolution, the structure of the AAA+
domain of Aquifex aeolicus DnaC, missing the N-terminal
helicase-binding domain, has recently been reported (2).
Its structure shows association of domains as a helical
filament that suggests a mechanism for association of
the loader proteins with the DnaA replication initiator
protein at chromosomal origins of replication to
accomplish recruitment of the helicase (2).
The structures of several related hexameric
DnaBfamily helicases reported recently (18–21) all reveal
unusual domain arrangements where the N-terminal
primase-interacting domains form a trimer of dimers
stacked on top of a hexamer of C-terminal RecA-like
helicase domains. The only reported structural
information on a helicase/loader protein complex comes from
low-resolution single particle analysis of negative stain
(3) and cryo-electron (22) microscopic images of the
E. coli DnaB/DnaC complex, which revealed a ring
of DnaC molecules stacked on the C-terminal face of
DnaB, which in the presence of ATP results in closing
of the central ssDNA-binding channel (22).
Here we present the first report of the 3D structure of
the helicase-binding domain of a replicative helicase
loader, the N-terminal domain of B. subtilis DnaI, and
identify its zinc-coordinating residues. In addition,
experiments have been conducted to delineate the boundary
between the N- and C-terminal domains of DnaI to
define the minimal domain required for binding to
the Bst helicase.
MATERIALS AND METHODS
Expression and purification of DnaI and Bst DnaB
The production of full-length DnaI has been described
earlier (14). The procedure used in the present work is
provided in the Supplementary Data. The DnaI-C76A
mutant protein (14) and the B. stearothermophilus DnaB
helicase were prepared as described (18), using an
expression strain (15) provided by Dr Panos Soultanas
(University of Nottingham, UK). Concentrations of all
proteins were determined spectrophotometrically at
280 nm, using calculated values of e280 (23).
Limited proteolysis and characterization of
Limited proteolysis experiments were done for 30 min
at 308C with solutions of 126 mg of full-length DnaI in
60 ml of storage buffer (50 mM Tris–HCl, pH 7.6, 10%
(w/v) glycerol, 2 mM dithiothreitol, 1 mM EDTA) using
a DnaI:trypsin ratio of 125:1. Additional experiments
were conducted in the presence of 3 mM ATP, 12 mM
MgCl2, or both. The products were analyzed by 12%
SDS–PAGE. The gels were washed in water before
selected bands were excised. N-terminal sequences of
fragments were determined by Edman degradation
(Biomolecular Resource Facility, Australian National
Construction of the expression system for DnaI-N106
Plasmid pND876, which contains the dnaI gene from
pET-15b-dnaI (7) under transcriptional control of strong
bacteriophage promoters in plasmid vector pND706
(24), was used as template for the PCR amplification of
the dnaI deletion mutant with primer 50-GAGATATA
CATATGGAACCAATC-30 incorporating an ATG
initiation codon (italicized) as part of an NdeI site
(underlined). Primer 50-AAAAACGCGTTACTTCCGTTTGA
CTGGACATTCG-30 was complementary to the 30 end
of the dnaI fragment (MluI sites underlined; complement
of stop codon italicized). The PCR product was digested
with NdeI and MluI and inserted into vector pND706
between the same restriction sites to yield pKL1272.
The gene encoding the dnaI mutant [coding for the
N-terminal 106 residues of DnaI, DnaI-N106 (Figure 1)]
is under strict control of tandem heat-inducible phage
promoters pR and pL. The plasmid pKL1272 was
transformed into the E. coli strains AN1459 (25) and
BL21( DE3)recA (26) for protein expression in LB or
15N-enriched media, respectively.
Expression and purification of DnaI-N106
Cells from the E. coli strain AN1459/pKL1272 were
grown in 4 l of LB medium containing thymine (25 mg/l)
and ampicillin (50 mg/l) at 308C to A595 = 0.61. A rapid
temperature shift to 428C induced the overproduction
of DnaI-N106, and growth was continued for a further
3 h (final A595 = 1.18). The cells (5.05 g) were harvested
at 11 000 g for 6 min, frozen in liquid N2 and stored
at –708C. All purification steps following cell growth
were carried out at 0–48C. Thawed cells were resuspended
in 76 ml of lysis buffer (50 mM Tris–HCl, pH 7.6, 10% w/v
sucrose, 100 mM NaCl, 2 mM dithiothreitol and 10 mM
spermidine) and lysed using a French press at 12 000 psi.
Proteins in the soluble fraction after centrifugation
(39 000 g, 1 h) were precipitated by addition of
ammonium sulfate (0.4 g/ml). After stirring for 1 h, the
suspension was centrifuged (39 000 g, 45 min); the pellet was
dissolved in 30 ml of buffer A (50 mM Tris–HCl, pH 7.6,
10% w/v glycerol, 2 mM dithiothreitol, 40 mM NaCl)
and dialysed against two changes (1 l each) of buffer A.
The protein was applied to a Toyopearl DEAE-TSK650M
column (2.5 14 cm) pre-equilibrated in buffer A. The
column was washed with 100 ml of buffer A at a flow
rate of 1 ml/min. DnaI-N106 eluted in the flow-through
and solid ammonium sulfate (8.1 g) was added to the
combined fractions (18 ml). The suspension was stirred for 1 h,
and the pellet after centrifugation (39 000 g, 45 min)
was dissolved in 4 ml of buffer B (50 mM Tris–HCl,
pH 7.6, 10% w/v glycerol, 2 mM dithiothreitol, 100 mM
NaCl). The solution was applied to a column (2.5 62 cm)
of Sephadex G50 gel filtration resin pre-equilibrated
in buffer B. The protein was eluted at a flow rate of 0.5
ml/min and the pooled fractions (21 ml) yielded 15 mg
of purified DnaI-N106. The purification was monitored
by 15% SDS–PAGE. The identity of the protein was
confirmed by ESI-MS (VG Quattro II triple quadrupole
mass spectrometer) using a 0.1% formic acid solution
with 1 mM 2-mercaptoethanol. The mass observed
for unlabeled DnaI-N106 was 12 524 1 Da (predicted
12 525 Da), indicating presence of the N-terminal
15N-labeled DnaI-N106 was produced using the
E. coli strain BL21( DE3)recA/pKL1272 in 15N-enriched
medium (Silantes, Munich, Germany), yielding 2.74 g
of cells from a 1 l culture. The purification process was
as described above, except that the gel filtration step was
replaced by chromatography over an 8 ml Mono-Q
column (GE Healthcare; flow rate 1.5 ml/min) after
dialysis against buffer C (50 mM Tris–HCl, pH 7.6,
2 mM dithiothreitol) containing 70 mM NaCl. 15N-labeled
DnaI-N106 eluted at about 110 mM NaCl in a linear
gradient (240 ml) of 70–150 mM NaCl in buffer C. The
yield of purified 15N-labeled protein was 13 mg l–1 of
Expression and purification of DnaI-N123 and DnaI-N132
C-terminally His6-tagged DnaI-N132 was purified as
described previously (14). The DnaI-N123 construct was
made using the QuikChange site-directed mutagenesis
kit (Stratagene) by introduction of a TAA stop codon
after that encoding Gln123 in the pET22b-dnaI plasmid
described by Soultanas (14). The primers used were 50-GC
ATGTATATCCAGTAAGATCTTCTTGGAG-30 and its
complement (stop codon italicized). This strategy resulted
in an untagged protein that was purified in the same way
NMR sample preparation
NMR samples were prepared by dialysis of DnaI-N
proteins against NMR buffer (10 mM sodium phosphate
pH 7.0, 0.1% (w/v) sodium azide, 100 mM NaCl and
1 mM dithiothreitol). Following concentration to 0.5 ml
by ultrafiltration (Amicon Ultrafree-4 centrifugal filter
device with a 5000 molecular weight cutoff), D2O was
added to 10% (v/v). The final NMR samples contained
1.1 mM unlabeled and 0.5 mM 15N-labeled DnaI-N106.
All NMR experiments were carried out at 258C
using Bruker Avance 600 and 800 MHz NMR
spectrometers equipped with TCI cryoprobes. The backbone
resonance assignment was obtained from the analysis of a 3D
NOESY-15N-HSQC spectrum [60 ms mixing time,
t1max(1H) = 20 ms, t2max(15N) = 13 ms, t3max = 106 ms]
together with 3D HNHA and 2D NOESY spectra. The
side-chain resonances were assigned using NOESY,
TOCSY and DQF-COSY spectra. A 15N-HSQC spectrum
with the INEPT delays set to 20 ms yielded correlations
via 2JHN couplings of the side chain of His84. NOE
restraints were collected from the 3D
NOESY-15NHSQC spectrum, 2D NOESY spectra of the 15N-labeled
and unlabeled samples in 90% H2O/10% D2O solution
(60 and 40 ms mixing time, respectively) and a 2D
NOESY spectrum in D2O solution (40 ms mixing time).
Resonance assignments of DnaI-N123 were obtained
from 3D HNCA, HN(CO)CA and CBCA(CO)NH
spectra using 15N/13C doubly labeled protein, prepared
as above following growth of cells in 15N/13C-enriched
The cross-peaks in the NOESY spectra of DnaI
were assigned and integrated using the program XEASY
(27). Stereospecific resonance assignments were obtained
for 21 CbH2 groups, 10 CgH2 groups, 1 CdH2 group,
5 isopropyl groups and 14 side-chain amides. The
stereospecific resonance assignments were determined by using
the HABAS and GLOMSA routines implemented in
the program CYANA (28), including J-coupling
information from HNHA, COSY and TOCSY spectra to support
the stereospecific assignment of CbH2 groups. The NMR
structure calculations using CYANA (28) involved eight
iterations of automatic NOE assignment using the routine
CANDID (29) followed by a simulated annealing
procedure starting from 100 random conformers that were
annealed in 20 000 steps using torsion-angle dynamics.
The 20 conformers with the lowest residual restraint
violations were energy minimized in a shell of water
using the protocol re_h2o.inp (30) and the program
CNS (31) with standard parameters.
Table 1 shows an overview of the restraints used and
structural statistics. Secondary structure elements and
root mean square deviation (r.m.s.d.) values were
calculated using the program Molmol (32). The chemical shifts
and coordinates of the structure have been deposited in
the BioMagResBank (accession code 15926) and PDB
(accession code 2K7R).
Thiol titrations with DTNB
Samples of full-length DnaI, DnaI-C76A and DnaI-N123
were dialyzed extensively under N2 at 48C against
50 mM Tris–HCl pH 8.0, 0.1 M NaCl. Reactions with
5,50-dithiobis(2-nitrobenzoic acid) (DTNB, Sigma) were
followed in matched quartz cuvettes in a UV-1700
UV-visible spectrophotometer (Shimadzu) at 238C. The
concentration of protein ( 5 mM, in 580 ml of buffer)
was first determined from a UV spectrum before a
10 mM solution of DTNB (15–20 ml) was added to initiate
the reaction. The release of 5-thio-2-nitrobenzoate
anion was monitored spectrophotometrically at 412 nm
[e412 = 14 150 M 1 cm 1; (33)] until no further change
occurred (30–40 min).
DnaI-N/Bst DnaB interaction analysis
Surface plasmon resonance (SPR) experiments were
conducted in a BIAcore T100 instrument operating
at 208C at a flow rate of 5 ml/min, using streptavidin
Number of uniquely assigned NOE cross-peaks
Number of non-redundant NOE upper-distance limits
Number of scalar coupling constants (HN-Ha)a
Number of dihedral-angle restraintsb
Number of distance restraints for zinc ionc
Intra-protein energy (kcal/mol)d
Maximum NOE-restraint violations (A˚ )
Maximum dihedral-angle restraint violations (o)
r.m.s.d. to the mean for N, Ca and C0 (A˚ )e
r.m.s.d. to the mean for all heavy atoms (A˚ )e
Ramachandran plot appearancef
Most favoured regions (%)
Additionally allowed regions (%)
Generously allowed regions (%)
Disallowed regions (%)
aFrom the 3D HNHA experiment.
bEstimated from chemical shift data using the program TALOS (43)
(105 restraints), and using the grid search algorithm HABAS (44) (419
c3 Zn – Cys Sg (2.4 A˚), 1 Zn – His Ne2 (2.05 A˚), 3 Zn – Cb (3.5 A˚), 3
Cys Sg – Cys Sg (3.85 A˚), 3 Cys Sg – His Ne2 (3.85 A˚). No angle
restraints were used.
dPARALLHDG force field (45).
eFor residues 20–104.
fFrom PROCHECK-NMR (46).
(SA) chips. Two different surfaces were used. For one,
30-biotinylated oligonucleotide dT35, synthesized
by GeneWorks (Adelaide, Australia), was immobilized
on one flow cell by injection of a 20 nM solution for
19 min, which resulted in an increase of 600 response
units (RU). For the other, a 100 nM solution of
N-terminally biotinylated DnaI-N-peptide (DnaI residues
1–18, preceded by biotin-Ser-Gly, synthesized at the
Biomolecular Resource Facility, Australian National
University) was injected for 90 s, producing a 730 RU
response. In each case, another flow cell remained
unmodified and served as a reference. The buffer used
for all experiments was 10 mM HEPES pH 7.5, 200 mM
NaCl, 3 mM EDTA, 0.05% surfactant P20, 10 mM MgCl2
and 2.7 mM ATP; protein samples were prepared
by direct dilution into this buffer. The interaction
of Bst DnaB with the three DnaI-N domains and an
unmodified peptide corresponding to residues 1–18 of
DnaI (MEPIGRSLQGVTGRPDFQ) was probed by
immobilization of Bst DnaB (365 RU) on dT35 and
subsequent injection (120 s) of increasing concentrations
of DnaI-N. The DnaI-N domains were left to dissociate
in buffer for 180 s. When necessary, the flow cells were
regenerated by three consecutive injections of 1 M NaCl,
50 mM NaOH (1 min) followed by one injection (1 min)
of 1 M MgCl2.
The interaction of Bst DnaB (concentrations up to
200 nM, as monomer) with immobilized biotinylated
DnaI-N-peptide was tested under the same conditions.
The involvement of this peptide in interaction with Bst
DnaB was also investigated by a competition assay
where the unmodified peptide (DnaI residues 1–18)
was mixed (at concentrations of 0, 20 or 50 mM) with
DnaI-N106 (10 mM), DnaI-N123 (10 mM) or DnaI-N132
(5 mM) prior to injection over the dT35-Bst DnaB surface.
The equilibrium dissociation constants (KD) of all
complexes, and standard errors, were derived from
steady state data by least squares fits to a 1:1 interaction
model. Before fitting, binding data were corrected by
subtraction of the reference response due to simple
refractive index changes and the small changes in equilibrium
responses due to dissociation of Bst DnaB during data
collection. No attempt was made to normalize data from
separate injections to correct for dissociation of <5% of
Bst DnaB over the course of each experiment.
Limited proteolysis of DnaI with trypsin under native
conditions showed a small number of distinct fragments
corresponding to cleavage after Arg105, Lys111, Lys112
and Lys118 (Supplementary Figures S1 and S2). In the
presence of ATP and MgCl2, the preferred cleavage
site shifted from Lys118 to Lys111, confirming that
ATP binds to DnaI and may cause conformational
changes. To probe the C-terminal boundary of the
N-terminal domain of DnaI, the constructs comprising
residues 1–106, 1–123 and 1–132 (referred to as
DnaIN106, DnaI-N123 and DnaI-N132, respectively) were
prepared, with a His6 tag at the C-terminal end of the
DnaI-N132 domain. [Limited proteolysis with subtilisin
indicated a cleavage site after residue 132 (14).] TOCSY
spectra showed the best spectral quality for DnaI-N106,
whereas the longer domains gave broader resonances
indicative of aggregation. Subsequently, 15N-labeled
samples were prepared of DnaI-N106 and DnaI-N123,
and a 13C/15N-labeled sample of DnaI-N123.
NMR resonance assignments
The 15N-HSQC spectrum of DnaI-N106 presented
welldispersed resonances characteristic of a monomeric folded
protein (Figure 2). Virtually complete resonance
assignments were obtained, except for the backbone amide of
Glu2, the CeH3 group of Met1, the g–, d–, and e– protons
of Lys65, the CeH2 group of Lys106 and Hg of Leu87.
In the course of the 3D structure determination,
stereospecific resonance assignments were obtained for 21 pairs
of CbH2, five pairs of CgH2 and one pair of CdH2 protons
as well as five pairs of g-and d-methyl groups of Leu
and Val, five pairs of g1 methylene protons of Ile, and
all side-chain NH2 protons of Gln and Asn.
A TOCSY spectrum of DnaI-N132 indicated narrow
resonances for about ten of the C-terminal residues in
addition to the His6 tag, indicating increased mobility
and, therefore, no participation in the structured part of
the N-terminal domain (data not shown). We therefore
constructed the DnaI-N123 domain, expecting it to be
more fully structured.
However, resonance assignments of DnaI-N123 were
much more difficult than for DnaI-N106 because the
signals were broad. Only backbone resonance assignments
were attempted using a 0.9 mM 13C/15N-labeled sample.
Three-fold dilution did not improve the line widths of the
NMR resonances, indicating that complete dissociation of
the aggregates was not possible at the concentrations
required for NMR. Most of the amide chemical shifts
observed for this sample were the same as for
DnaIN106. For the segment with residues 107–123, tentative
resonance assignments were obtained for only nine
NMR spectroscopy cannot be used to determine directly
the location of the zinc ion, so we confirmed the identity
of zinc ligands by independent methods. DnaI-N106
contains one histidine, four cysteines (Figure 1) and one
zinc ion (14), and all of the cysteine residues of
fulllength DnaI are in the DnaI-N106 domain. The 15N
chemical shift of the Ne2 resonance of His84 (221 ppm;
Supplementary Figure S3) was characteristic of a
zincbound imidazole nitrogen (34), confirming that the
side chain of His84 is one of the zinc ligands (14). In
addition, there was no sign of structural heterogeneity,
indicating that the zinc ion was present in all protein
molecules and in the same coordination environment.
The affinity of DnaI-N for zinc was high, as treatment
of a sample with 4 mM EDTA did not alter the
appearance of the 15N-HSQC spectra. Thiol titrations of full
length DnaI in triplicate with DTNB (0.32 mM, at pH
8.0) gave biphasic pseudo first-order kinetics, with one
exposed thiol group titrating quickly (kobs = 0.09 0.16
s 1), and three others about 50-fold more slowly;
kobs = (1.9 – 2.7) 10 3 s 1. Near identical
stoichiometries and rate constants were observed for titration of
DnaI-N123. The slowly reacting cysteines are presumed
to be protected from reaction by being coordinated to
zinc, and to titrate simultaneously as the metal-binding
site is destroyed. In contrast, titration of the DnaI-C76A
mutant protein with 0.25 mM DTNB under similar
conditions showed only single-phase kinetics, with three
thiols titrating slowly; kobs = (0.5–0.7) 10 3 s 1. These
data thus clearly identify His84 as a zinc ligand and Cys76
as the exposed faster-reacting cysteine residue, and
suggest that the other three cysteine thiols are coordinated
to the zinc ion. These conclusions were confirmed by 3D
structure determination, as described below.
Structure of DnaI-N106
The structure of DnaI-N106 comprises four a helices and
two b strands. In addition, a short 310 helix near the
C-terminus was found in most of the NMR conformers
(Figure 3A). The structure presents a novel fold, as a
search of the protein data bank using the program Dali
(35) failed to reveal a domain of similar structure. The
zinc ion is coordinated by Cys67, Cys70, His84 and
Cys101 (Figure 3A), whereas the side chain of Cys76 is
located in the disordered loop between helix 4 and the
first b strand (Figure 3 and Supplementary Figure S4).
The zinc ion probably plays an important structural
role, as it ties together three sequentially distant
segments of the polypeptide chain in a region where the
structure has few hydrophobic residues (Figure 3D).
In agreement with the zinc-coordinating residues found
in the present NMR study, Cys76 is not conserved in
B. amyloliquefaciens and B. pumilus, while all the other
zinc-coordinating residues are (Figure 1). Nonetheless,
some of the DnaI homologues from other bacteria
show no evidence of a zinc-binding site. In those proteins,
the polypeptide segment between helix 4 and the first b
strand is shorter which probably stabilizes the structure.
Where this segment is longer, as in the homologue
from S. aureus, it cannot be excluded that zinc binds at
a similar site using a different set of cysteine and histidine
residues (Figure 1).
The N-terminal 14 residues were disordered. As
expected for a mobile polypeptide segment, the NMR
signals of the N-terminal residues were also significantly
narrower than those of the structurally defined part of
the protein. Similarly, the 1H NMR signals of residues
63–82 in the loop between helix 4 and the first b strand
showed a reduced line width.
Conceivably, the loop residues following helix 4 could
assume a single, rigid conformation in the presence of the
C-terminal ATP-binding domain. Although a detailed
analysis of the longer construct DnaI-N123 was hampered
by broad line widths, significant chemical shift changes
observed for the loop residues suggests an interaction
with the segment following the C-terminus of
DnaIN106 (Figure 3B). It cannot be excluded, however,
that the chemical shift change is non-specific, reflecting
spatial proximity of two or several DnaI-N domains
induced by self-association of the C-terminal segment
of DnaI-N123. This interpretation would be consistent
with the observation that the NMR signals of residues
107–123 were particularly broad yet were at chemical
shifts indicative of random coil structures.
As expected for a conserved fold, the side chains
of many of the conserved hydrophobic residues have low
solvent accessibility because they contribute to the
hydrophobic core of the protein (Figure 4). Conversely,
many of the residues with low side-chain solvent
accessibility are hydrophobic. For example, the completely
conserved residues Leu57, Gly82 and Pro85 are buried
and the side chains of Phe37 and Tyr98 have low solvent
accessibilities (Figure 4).Remarkably, the positions of
Ile4, Leu8 and Val11 in the unstructured N-terminus
of B. subtilis DnaI-N are also occupied by hydrophobic
residues in all DnaI-N domains shown in Figure 1,
although those residues are highly solvent accessible
(Figure 4). Phe17 is also highly conserved despite high
solvent accessibility of its side chain (>75%). Their
conservation suggests a functional role of these flexible
residues, presumably in protein–protein interactions.
Interactions with Bst DnaB
The interaction of Bst DnaB with all three DnaI-N
constructs was probed by SPR, using a biotinylated
oligonucleotide (dT35) immobilized on a streptavidin-coated
surface. Bst DnaB was first injected over the
oligonucleotide to a consistent binding level. In the presence of ATP
under the conditions used, Bst DnaB binds to dT35 and
dissociates very slowly; less than 9% dissociation was
observed 30 min after injection. Subsequently, increasing
concentrations of DnaI-N domains were injected. This
experimental design allows for the real-time observation
of the interaction of ssDNA-bound Bst DnaB with each
of the DnaI-N constructs and direct comparison of the
All three DnaI-N domains (DnaI-N106, 123 and 132)
interacted with Bst DnaB with association and
dissociation rates that were fast, in contrast to the slower rates
that full-length DnaI exhibited under similar conditions
(14). Full-length DnaI is presumed to form a much more
stable complex with the helicase because it oligomerizes
cooperatively on the helicase surface via interactions
among its C-terminal domains (2,14). This cooperativity
in association and multiphasic dissociation made it
impossible to obtain reliable thermodynamic data with
Provided that the sole contact between Bst DnaB
and DnaI is in the zinc-binding region of DnaI-N106 as
suggested in previous work (14), all three DnaI-N domains
were anticipated to interact with similar strength.
However, the experimentally determined KD values
(DnaI-N106, 18.5 0.9 mM; DnaI-N123, 7.3 0.1 mM;
DnaI-N132, 0.65 0.02 mM; Figure 5) revealed significant
differences, with the KD values decreasing with each longer
construct, especially with DnaI-N123 in comparison with
DnaI-N132. This indicates that the interaction of DnaI
with Bst DnaB may extend beyond the folded core of
the DnaI-N domain to the neighboring segments.
Although the NMR experiments showed that this region
is most likely unstructured in DnaI-N132, it appears that
some residues therein may be capable of assuming a
structure that contributes to interaction with the helicase.
We also examined the possible contribution of the
flexible, conserved N-terminal segment of DnaI to the
interaction with Bst DnaB. An N-terminally biotinylated
peptide comprising DnaI residues 1–18 was immobilized
on the surface of a streptavidin chip and Bst DnaB
was injected over it. We were unable to detect significant
interaction at the helicase concentrations used; KD
was estimated to be at least 50 mM. We also looked for
evidence of direct interaction of a corresponding
unmodified peptide with immobilized Bst DnaB. Under
conditions where formation of a 1:1 complex should have
given a response of 17.4 RU, we detected responses
of 2.2 and 3.4 RU with peptide at 20 and 50 mM,
respectively. These data indicate a very weak interaction,
with KD > 0.14 mM. In competition experiments, the
presence of peptide at 20 or 50 mM modestly reduced
the response obtained when 10 mM DnaI-N123 or 5 mM
DnaI-N134 were injected over immobilized Bst DnaB
(responses were reduced by <10%), which again indicates
a very weak interaction of the N-terminal region of DnaI
with Bst DnaB. The function of the conserved, but
unstructured, N-terminal segment of DnaI thus remains
uncertain, though we cannot rule out that it might
contribute modestly to helicase interaction.
3D structure determination of DnaI-N106, the folded core
of the N-terminal domain of B. subtilis DnaI, revealed a
zinc-binding module with a novel fold. High sequence
conservation of DnaI-N was found only among proteins
with putative primosomal functions. The zinc-binding
motif of DnaI differs from conventional zinc fingers, in
particular as the zinc-coordinating residues are separated
in the amino-acid sequence by two relatively long
polypeptide segments. The zinc-binding motif may play a
purely structural role. It has been shown not to be
involved in DNA binding, and study of alanine mutants
of zinc-coordinating residues showed that impairment
of zinc binding correlated with impaired binding to
Bst DnaB (14). Although this is consistent with the
zincbinding region being involved in interaction with the
helicase, it would also be readily explained if the primary
role of the metal ion is to prevent unfolding of the DnaI-N
By study of the binding of three different DnaI-N
domain constructs to ssDNA-bound Bst DnaB helicase,
we showed that the structured core domain DnaI-N106
bears most of the interacting residues, and that progressive
C-terminal extension of this domain to 123 and 132
residues results in somewhat stronger binding (Figure 5).
With reliable binding assays in hand, we are now in a good
position to explore this interaction further by mutagenesis.
Having defined the structure of DnaI-N may also facilitate
structural study of the helicase/helicase loader complex by
X-ray crystallography. The structural basis of this
interaction in any organism is still poorly defined (2).
The stability of the DnaI-N fold clearly does not
depend on conservation of hydrophobic side chains at
positions 4, 8 and 11 in the mobile N-terminal segment
of DnaI-N106. We have also observed these residues to
be flexible in full length DnaI, indicating that they do not
fold back to contact the C-terminal AAA+ domain (data
not shown). Therefore we speculate that these residues,
together with the highly conserved, solvent exposed
hydrophobic residue at position 17, are required for
binding to another protein. Specific protein–protein
interactions, where a mobile terminal polypeptide segment of
one protein binds to a well-structured domain of its
binding partner, is a recurrent motif in bacterial replisomes,
governing for example the interactions between the c and
g (36), t and a (37,38), and e and a (39) subunits of the
E. coli polymerase III complex, as well as the
interactions of polymerases and many other proteins with the
b sliding clamp (40), and of ssDNA-binding protein with
its many binding partners (41). We were unable, however,
to detect a strong interaction with Bst DnaB of a peptide
corresponding to the N-terminal 18 residues of DnaI, and
note that a version of DnaI lacking the N-terminal seven
residues still bound its natural binding partner, the
B. subtilis DnaC helicase, very strongly as detected by
yeast two hybrid experiments (7).
The N-terminal domain of DnaI acts as a molecular
switch that regulates the accessibility of the
ssDNAbinding site of the C-terminal domain (14). In agreement
with this notion, the fragmentation pattern observed
after limited tryptic digestion of DnaI varied depending
on the presence of ATP and Mg2+, which presumably
bind to the Walker A and B motifs in the C-terminal
domain (Supplementary Data). The change in
fragmentation pattern may reflect a conformational change affecting
the accessibility of the linker segment between the N- and
C-terminal domains, although it cannot be excluded that
the ATP-Mg2+ complex alone is sufficient to hinder and
redirect the approach of trypsin.
B. subtilis DnaI and E. coli DnaC share significant
sequence homology in their C-terminal domains, but
there is no similarity between DnaI-N and the N-terminal
segment of E. coli DnaC (15). E. coli DnaC cannot assume
the DnaI fold determined here because it does not have
enough residues in the N-terminal polypeptide segment,
though this segment has also been shown to comprise at
least part of the helicase binding domain of DnaC (42).
In view of the homology between the DnaI-N domains
of B. subtilis and S. aureus (Figure 1), breaking the
interaction of DnaI-N with the helicase would be of
pharmaceutical interest. The 3D structure of DnaI-N
determined in this work sets the stage for directed
mutagenesis experiments and further structural studies to
identify the interaction surface between the two proteins.
Supplementary Data are available at NAR Online.
We thank Drs Shigeki Moriya and Panos Soultanas for
plasmids, Dr Kiyoshi Ozawa for recording preliminary
NMR data and Dr Patrick Schaeffer for help with the
limited proteolysis experiments. N-terminal amino-acid
sequencing and peptide synthesis were carried out by the
Biomolecular Research Facility at the Australian National
Financial support from the Latvian Scientific Council for
K.J. and E.L. is greatly acknowledged. The project and
the purchase of the NMR equipment were supported by
grants and Fellowships from the Australian Research
Council [to G.O. and N.E.D.]. Funding for open access
charge: University of Wollongong.
Conflict of interest statement. None declared.
1. Schaeffer , P.M. , Headlam , M.J. and Dixon , N.E. ( 2005 ) Proteinprotein interactions in the eubacterial replisome . IUBMB Life , 57 , 5 - 12 .
2. Mott , M.L. , Erzberger , J.P. , Coons , M.M. and Berger , J.M. ( 2008 ) Structural synergy and molecular crosstalk between bacterial helicase loaders and replication initiators . Cell , 135 , 623 - 634 .
3. San Martin,C., Radermacher , M. , Wolpensinger , B. , Engel , A. , Miles , C.S. , Dixon , N.E. and Carazo , J.-M. ( 1998 ) Three-dimensional reconstructions from cryoelectron microscopy images reveal an intimate complex between helicase DnaB and its loading partner DnaC . Structure , 6 , 501 - 509 .
4. Yuzhakov , A. , Turner , J. and O'Donnell , M. ( 1996 ) Replisome assembly reveals the basis for asymmetric function in leading and lagging strand replication . Cell , 86 , 877 - 886 .
5. Davey , M.J. and O'Donnell , M. ( 2003 ) Replicative helicase loaders: ring breakers and ring makers . Curr. Biol ., 13 , R594 - R596 .
6. Velten , M. , McGovern , S. , Marsin , S. , Ehrlich , S.D. , Noirot , P. and Polard , P. ( 2003 ) A two-protein strategy for the functional loading of a cellular replicative DNA helicase . Mol. Cell , 11 , 1009 - 1020 .
7. Imai , Y. , Ogasawara , N. , Ishigo-oka , D. , Kadoya , R. , Daito , T. and Moriya , S. ( 2000 ) Subcellular localization of Dna-initiation proteins of Bacillus subtilis: evidence that chromosome replication begins at either edge of the nucleoids . Mol. Microbiol ., 36 , 1037 - 1048 .
8. Bruand , C. , Farache , M. , McGovern , S. , Ehrlich , S.D. and Polard , P. ( 2001 ) DnaB, DnaD and DnaI proteins are components of the Bacillus subtilis replication restart primosome . Mol. Microbiol ., 42 , 245 - 255 .
9. Marsin , S. , McGovern , S. , Ehrlich , S.D. , Bruand , C. and Polard , P. ( 2001 ) Early steps of Bacillus subtilis primosome assembly . J. Biol. Chem. , 276 , 45818 - 45825 .
10. Rokop ,M.E, Auchtung, J.M. and Grossman , A.D. ( 2004 ) Control of DNA replication initiation by recruitment of an essential initiation protein to the membrane of Bacillus subtilis . Mol. Microbiol ., 52 , 1757 - 1767 .
11. Turner , I.J. , Scott , D.J. , Allen , S. , Roberts , C.J. and Soultanas , P. ( 2004 ) The Bacillus subtilis DnaD protein: a putative link between DNA remodeling and initiation of DNA replication . FEBS Lett. , 577 , 460 - 464 .
12. Bruand , C. , Velten , M. , McGovern , S. , Marsin , S. , S e´re` na,C., Ehrlich , D. and Polard , P. ( 2005 ) Functional interplay between the Bacillus subtilis DnaD and DnaB proteins essential for initiation and re-initiation of DNA replication . Mol. Microbiol ., 55 , 1138 - 1150 .
13. Zhang ,W., Carneiro , M.J.V.M. , Turner , I.J. , Allen , S. , Roberts , C.J. and Soultanas , P. ( 2005 ) The Bacillus subtilis DnaD and DnaB proteins exhibit different DNA remodelling activities . J. Mol. Biol ., 351 , 66 - 75 .
14. Ioannou , C. , Schaeffer , P.M. , Dixon , N.E. and Soultanas , P. ( 2006 ) Helicase binding to DnaI exposes a cryptic DNA-binding site during helicase loading in Bacillus subtilis . Nucleic Acids Res ., 34 , 5247 - 5258 .
15. Soultanas , P. ( 2002 ) A functional interaction between the putative primosomal protein DnaI and the main replicative DNA helicase DnaB in Bacillus . Nucleic Acids Res ., 30 , 966 - 974 .
16. Davey , M.J. , Jeruzalmi , D. , Kuriyan , J. and O'Donnell , M. ( 2002 ) Motors and switches: AAA+ machines within the replisome . Nat. Rev. Mol. Cell Biol ., 3 , 826 - 835 .
17. Davey , M.J. , Fang , L. , McInerney , P. , Georgescu , R.E. and O'Donnell , M. ( 2002 ) The DnaC helicase loader is a dual ATP/ADP switch protein . EMBO J ., 21 , 3148 - 3159 .
18. Bailey , S. , Eliason , W.K. and Steitz , T.A. ( 2007 ) Structure of hexameric DnaB helicase and its complex with a domain of DnaG primase . Science , 318 , 459 - 463 .
19. Wang , G. , Klein , M.G. , Tokonzaba , E. , Zhang , Y. , Holden , L.G. and Chen , X.S. ( 2008 ) The structure of a DnaB-family replicative helicase and its interactions with primase . Nat. Struct. Mol. Biol ., 15 , 94 - 100 .
20. Biswas , T. and Tsodikov , O.V. ( 2008 ) Hexameric ring structure of the N-terminal domain of Mycobacterium tuberculosis DnaB helicase . FEBS J ., 275 , 3064 - 3071 .
21. Lo , Y.-H. , Tsai , K.-L. , Sun , Y.-J. , Chen , W.-T. , Huang , C.-Y. and Hsiao , C.-D. ( 2009 ) The crystal structure of a replicative hexameric helicase DnaC and its complex with single-stranded DNA . Nucleic Acids Res ., 37 , 804 - 814 .
22. Ba´ rcena , M. , Ruiz , T. , Donate , L.E. , Brown , S.E. , Dixon , N.E. , Radermacher , M. and Carazo , J.M. ( 2001 ) The DnaB DnaC complex: a structure based on dimers assembled around an occluded channel . EMBO J ., 20 , 1462 - 1468 .
23. Gill , S.C. and von Hippel , P.H. ( 1989 ) Calculation of protein extinction coefficients from amino acid sequence data . Anal. Biochem. , 182 , 319 - 326 .
24. Love ,C.A., Lilley , P.E. and Dixon , N.E. ( 1996 ) Stable highcopy-number bacteriophage promoter vectors for overproduction of proteins in Escherichia coli . Gene , 176 , 49 - 53 .
25. Vasudevan , S.G. , Armarego , W.L.F. , Shaw , D.C. , Lilley , P.E. , Dixon , N.E. and Poole , R.K. ( 1991 ) Isolation and nucleotide sequence of the hmp gene that encodes a haemoglobin-like protein in Escherichia coli K-12 . Mol. Gen. Genet., 226 , 49 - 58 .
26. Williams , N.K. , Prosselkov , P. , Liepinsh , E. , Line , I. , Sharipo , A. , Littler , D.R. , Curmi , P.M.G. , Otting , G. and Dixon , N.E. ( 2002 ) In vivo protein cyclization promoted by a circularly permuted Synechocystis sp. PCC6803 DnaB mini-intein . J. Biol. Chem. , 277 , 7790 - 7798 .
27. Bartels , C. , Xia , T. , Billeter , M. , Gu¨ ntert , P. and Wu¨ thrich , K. ( 1995 ) The program XEASY for computer-supported NMR spectral analysis of biological macromolecules . J. Biomol. NMR , 6 , 1 - 10 .
28. Gu¨ ntert , P. , Mumenthaler , C. and Wu¨ thrich , K. ( 1997 ) Torsion angle dynamics for NMR structure calculation with the new program DYANA . J. Mol. Biol ., 273 , 283 - 298 .
29. Herrmann , T. , Gu¨ ntert , P. and Wu¨ thrich , K. ( 2002 ) Protein NMR structure determination with automated NOE assignment using the new software CANDID and the torsion angle dynamics algorithm DYANA . J. Mol. Biol ., 319 , 209 - 227 .
30. Linge , J.P. , Williams , M.A. , Spronk ,C.A.E.M., Bonvin , A.M.J.J. and Nilges , M. ( 2003 ) Refinement of protein structures in explicit solvent . Proteins , 50 , 496 - 506 .
31. Brunger , A.T. , Adams , P.D. , Clore , G.M. , Delano , W.L. , Gros , P. , Grosse-Kunstleve , R.W. , Jiang , J. , Kuszewski , J. , Nilges , M. , Pannu , N.S. et al. ( 1998 ) Crystallography and NMR System: a new software suite for macromolecular structure determination . Acta Crystallogr. D , 54 , 905 - 921 .
32. Koradi , R. , Billeter , M. and Wu¨ thrich , K. ( 1996 ) MOLMOL: a program for display and analysis of macromolecular structures . J. Mol. Graph ., 14 , 51 - 55 .
33. Riddles , P.W. , Blakeley , R.L. and Zerner , B ( 1979 ) Ellman's reagent: 5,50-dithiobis(2-nitrobenzoic acid) - a reexamination . Anal. Biochem. , 94 , 75 - 81 .
34. Chen , Y.L. , Park , S. , Thomburg , R.W. , Tabatabai , L.B. and Kintanar , A. ( 1995 ) Structural characterization of the active site of Brucella abortus Cu-Zn superoxide dismutase: a 15N and 1H NMR investigation . Biochemistry , 34 , 12265 - 12275 .
35. Holm , L. and Sander , C. ( 1996 ) Mapping the protein universe . Science , 273 , 595 - 603 .
36. Ozawa , K. , Jergic , S. , Crowther , J.A. , Thompson , P.R. , Wijffels , G. , Otting , G. and Dixon , N.E. ( 2005 ) Cell-free in vitro protein synthesis in an autoinduction system for NMR studies of protein-protein interactions . J. Biomol. NMR , 32 , 235 - 241 .
37. Jergic , S. , Ozawa , K. , Williams , N.K. , Su , X.-C. , Scott , D.D. , Hamdan , S.M. , Crowther , J.A. , Otting , G. and Dixon , N.E. ( 2007 ) The unstructured C-terminus of the t subunit of Escherichia coli DNA polymerase III holoenzyme is the site of interaction with the a subunit . Nucleic Acids Res ., 35 , 2813 - 2824 .
38. Su ,X.-C., Jergic , S. , Keniry , M.A. , Dixon , N.E. and Otting , G. ( 2007 ) Solution structure of domains IVa and V of the t subunit of Escherichia coli DNA polymerase III and interaction with the a subunit . Nucleic Acids Res ., 35 , 2825 - 2832 .
39. Ozawa , K. , Jergic , S. , Park, A.Y. , Dixon , N.E. and Otting , G. ( 2008 ) The proofreading exonuclease subunit e of Escherichia coli DNA polymerase III is tethered to the polymerase subunit a via a flexible linker . Nucleic Acids Res ., 36 , 5074 - 5082 .
40. Dalrymple , B.P. , Kongsuwan , K. , Wijffels , G. , Dixon , N.E. and Jennings , P.A. ( 2001 ) A universal protein-protein interaction motif in the eubacterial DNA replication and repair systems . Proc. Natl Acad. Sci. USA , 98 , 11627 - 11632 .
41. Shereda , R.D. , Kozlov , A.G. , Lohman , T.M. , Cox , M.M. and Keck , J.L. ( 2008 ) SSB as an organizer/mobilizer of genome maintenance complexes . Crit. Rev. Biochem. Mol. Biol ., 43 , 289 - 318 .
42. Ludlam , A.V. , McNatt , M.W. , Carr , K.M. and Kaguni , J.M. ( 2001 ) Essential amino acids of Escherichia coli DnaC protein in an N-terminal domain interact with DnaB helicase . J. Biol. Chem. , 276 , 27345 - 27353 .
43. Cornilescu , G. , Delaglio , F. and Bax , A. ( 1999 ) Protein backbone angle restraints from searching a database for chemical shift and sequence homology . J. Biomol. NMR , 13 , 289 - 302 .
44. Gu¨ ntert , P. , Braun , W. , Billeter , M. and Wu¨ thrich , K. ( 1989 ) Automated stereospecific proton NMR assignments and their impact on the precision of protein structure determinations in solution . J. Am. Chem. Soc. , 111 , 3997 - 4004 .
45. Linge , J.P. and Nilges , M. ( 1999 ) Influence of non-bonded parameters on the quality of NMR structures: a new force field for NMR structure calculation . J. Biomol. NMR , 13 , 51 - 59 .
46. Laskowski , R.A. , Rullmann ,J.A.C., MacArthur , M.W. , Kaptein , R. and Thornton , J.M. ( 1996 ) AQUA and PROCHECK-NMR: programs for checking the quality of protein structures solved by NMR . J. Biomol. NMR, 8 , 477 - 486 .
47. Altschul , S.F. , Madden , T.L. , Scha¨ ffer,A.A., Zhang , J. , Zhang , Z. , Miller , W. and Lipman , D.J. ( 1997 ) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs . Nucleic Acids Res ., 25 , 3389 - 3402 .
48. Koradi , R. , Billeter , M. and Wu¨ thrich , K. ( 1996 ) MOLMOL: A program for display and analysis of macromolecular structures . J. Mol. Graph ., 14 , 51 - 55 .
49. Sevilla-Sierra , P. , Otting , G. and Wu¨ thrich , K. ( 1994 ) Determination of the nuclear magnetic resonance structure of the DNA-binding domain of the P22 c2 repressor (1-76) in solution and comparison with the DNA-binding domain of the 434 repressor . J. Mol. Biol ., 235 , 1003 - 1020 .