doi:10.1016/j.jmb.2006.01.091
J. Mol. Biol. (2006) 357, 1063–1076
Quaternary Polymorphism of Replicative Helicase
G40P: Structural Mapping and Domain Rearrangement
Rafael Núñez-Ramı́rez1, Yolanda Robledo1, Pablo Mesa2, Silvia Ayora2
Juan Carlos Alonso2, José Marı́a Carazo1* and Luis Enrique Donate1
1
Department of Macromolecular Structure, Centro
Nacional de Biotecnologı́a
CSIC, Campus Universidad
Autónoma de Madrid, 28049
Cantoblanco, Madrid, Spain
2
Department of Microbial
Biotechnology, Centro Nacional
de Biotecnologı́a, CSIC, Campus
Universidad Autónoma de
Madrid, 28049 Cantoblanco
Madrid, Spain
Quaternary polymorphism is a distinctive structural feature of the DnaB
family of replicative DNA hexameric helicases. The Bacillus subtilis
bacteriophage SPP1 gene 40 product (G40P) belongs to this family. Three
different quaternary states have been described for G40P homohexamers,
two of them with C3 symmetry, and the other with C6 symmetry. We
present three-dimensional reconstructions of the different architectures of
G40P hexamers and a variant lacking the N-terminal domain. Comparison
of the G40P and the deletion mutant structures sheds new light on the
functional roles of the N and C-terminal domains, at the same time that it
allows the direct structural mapping of these domains. Based on this new
information, hybrid EM/X-ray models are presented for all the different
symmetries. These results suggest that quaternary polymorphism of
hexameric helicases may be implicated in the translocation along the DNA.
q 2006 Elsevier Ltd. All rights reserved.
*Corresponding author
Keywords: ATPase; DNA replication; DNA translocase; helicase; 3D-EM
Introduction
DNA helicases are molecular motors that unwind
double-stranded DNA fuelled by the energy
produced by nucleoside triphosphate hydrolysis.1
Helicases are involved in all processes of DNA
metabolism that need single-stranded DNA as a
substrate. Sequence analysis of this group of
enzymes has led to their classification in different
superfamilies (SF1 to SF3) and several smaller
families (e.g. DnaB family), which have in common
a set of conserved helicase motifs.2
Helicases assemble into different oligomeric
states. Structural studies of monomeric,3,4 dimeric5
and hexameric helicases6,7 have shown that conserved sequence motifs of these enzymes are
situated in the nucleotide-binding core which was
first described for the RecA protein.8 Moreover, the
most conserved residues within the characteristic
motifs of helicases are implicated in nucleotide
binding and NTPase activity, all of them around
the so-called P-loop. A conserved arginine finger
Abbreviations used: 2D, two-dimensional; 3D, threedimensional; C3, 3-fold symmetry; C6, 6-fold symmetry;
C3C6, intermediate symmetry; C7, 7-fold symmetry; CTD,
C-terminal domain; EM, electron microscopy; GXP,
geneX product; NTD, N-terminal domain; wt, wild-type.
E-mail address of the corresponding author:
[email protected]
(Arg-finger) is also essential for NTPase activity.9,10
This Arg-finger belongs to another subunit in the
case of oligomeric helicases or to another domain in
monomeric helicases.11
The helicases of the DnaB family are intimately
associated with the DNA primase and participate in
the replicative process assembled as hexamers. The
main representative of this family is Escherichia coli
DnaB (EcoDnaB). Other members are gp4 of E. coli
T7 bacteriophage (T7gp4) and G40P of Bacillus
subtilis SPP1 bacteriophage.12 Both electron
microscopy (EM) and X-ray diffraction studies12
have shown that hexameric helicases assemble as
ring-shaped structures with a central channel.
Interaction with DNA seems to be through their
inner channel,13–15 and one of the conserved motifs
(motif H4) has been described to be responsible for
this interaction.16 Some helicases of the DnaB family
have previously been shown to assemble in several
states with different point symmetry. Projection
images of EcoDnaB17 and G40P18 revealed the
existence of particles with 3-fold (C3) and 6-fold
rotational symmetry (C6), as well as an intermediate
state. Replicative helicases of other families, for
example RepA of plasmid RSF101019 and papilloma
E1 helicase,20 may display a similar polymorphism,
although the intermediate state was not identified.
The RecA-like helicase domains are usually
joined to other domains. For example, limited
trypsinolysis studies have demonstrated that
0022-2836/$ - see front matter q 2006 Elsevier Ltd. All rights reserved.
1064
Eco-DnaB protein (52 kDa) could be cleaved into
two major discrete polypeptides, to render a 33 kDa
C-terminal domain (CTD) and a 12 kDa N-terminal
domain (NTD), linked by a small hinge region.21,22
Whereas the helicase domain laid in the CTD, the
NTD has an undetermined associated function. In a
similar way, the bacteriophage T7 gp4 contains a
helicase domain physically linked to a primase
domain.23 This modular organization may be
common in helicases.
The G40P helicase is required for the initiation of
B. subtilis bacteriophage SPP1 DNA replication.24
G40P hexamers show all the activities associated
with DnaB-like enzymes, including DNA unwinding with a 5 0 to 3 0 polarity.25 Analysis of the G40P
primary structure indicates that this protein presents a modular organization similar to EcoDnaB,
with a NTD and a CTD linked by a hinge region.
A mutant lacking the NTD (G40PDN109)26 retains
the DNA binding and the DNA helicase activity,27
although without the characteristic functional
polarity of the wild-type (wt) enzyme. In a pHdependent manner, G40PDN109 displays helicase
activity in both polarities, suggesting that the NTD
main role is the regulation of the enzyme helicase
action.
We have generated three-dimensional (3D) reconstructions for the different states of G40P helicase,
as well as for the G40PDN109 mutant. Our study
shows that G40PDN109 primarily oligomerizes as a
hexamer with preferential 6-fold rotational sym-
G40P Helicase Quaternary Polymorphism
metry, although it is able to undergo the structural
reorganization required for the transition between
the different symmetry states. We present the first
3D data for a hexameric helicase of Gram-positive
bacterial origin, allowing the experimental localization of the NTD within the whole structure of a
DnaB-like replicative helicase. The NTD plays an
active role in the stability of the complex and in the
symmetry transition, but is not essential for enzyme
activity. A model for the rearrangement of the CTD
and NTD in the symmetry transition, as well as
their functional role, is proposed.
Results
Three-dimensional structure of G40P
3D reconstructions were obtained for the three
different architectures of G40P. The Random
Conical Tilt approach28 was followed to obtain
non-biased 3D initial references. Figure 1(a) and (b)
shows a representative tilting pair for a G40P
preparation in buffer A. G40P oligomers show a
preferred view in which the protein appears as a
globular-shaped particle with a central stainpenetrating region. Non-overlapping pairs of particles were manually selected and extracted from
the micrographs to a total of 13,743. The set of
untilted images were used for the initial 2D
alignment and classification, whereas the corres-
Figure 1. EM and 2D analysis of G40P hexamers. (a) Untilted and (b) 508 tilted electron micrographs of negatively
stained G40P oligomers in buffer A. Arrowheads point to some representative particles; numbers indicate
correspondence between partners in the tilting pair. (c), (e) and (g) Rotational power spectra, calculated between
radii 5–8 nm, of C3, C3C6, and C6 average images, respectively. (d), (f) and (h) 2D average images of the re-aligned C3,
C3C6 and C6 groups, respectively. The bar represents 50 nm in (a) and 5 nm in (d). Arrowheads in (f) indicate the main
differences between the C3 and C3C6 average images.
G40P Helicase Quaternary Polymorphism
ponding tilted images were used as input for the 3D
reconstruction algorithms.
The untilted images were aligned and classified
as described in Materials and Methods. Classification was performed using the rotational power
spectra rather than the images themselves. Classes
with rotational symmetry similar to those identified
in previous studies18 were selected. Rotational
spectra representative of each of these groups are
presented in Figure 1(c), (e) and (g). The C3 group is
characterized by a rotational power spectrum that
shows a maximum value for the 3-fold component
followed by a noticeable contribution of the 6-fold
one (Figure 1(c)). The rotational power spectrum of
the C6 group only shows a peak for the 6-fold
harmonic (Figure 1(g)). The rotational spectrum of
the third group shows the 6-fold harmonic as the
strongest one, and an important contribution from
the 3-fold component (Figure 1(e)), as was
described for the intermediate state.18 This group
will be referred to as the C3C6 symmetry group.
Only particles clearly corresponding to the
described symmetry patterns were selected. The
three groups were roughly populated with 700
particles each. Images associated to each group
were independently realigned. The average images
calculated for each group are shown in Figure 1(d),
(f) and (h) next to their corresponding rotational
power spectra. As previously reported,18 the
average images from groups C3 (Figure 1(d)) and
C3C6 (Figure 1(f)) present six central masses and
three smaller blobs in the vertexes, while the
average image from group C6 (Figure 1(h)) assembles as six masses with the same shape and
orientation. The main difference observed between
the average images of the C3 and C3C6 classes
concerns the presence of a weak density protruding
out of the three central masses (arrowheads in
Figure 1(f)). This restrictive selection is due to the
fact that the Random Conical Tilt approach strictly
requires that all the untilted images, partners of the
tilted images used for the 3D reconstruction, were
structurally homogeneous and in exactly the same
orientation with respect to the supporting film.
Therefore, only images that clearly belonged to one
of the symmetry groups and did not present any
sign of possible rocking were further considered for
3D reconstruction.
The angular information derived from the
rotational alignment of the untilted images,
together with the data from the tilting process,
were used to define the conical geometry in order to
obtain an initial 3D reconstruction for each group.
Each initial reconstruction was then used as the
starting point in an iterative process of angular
refinement performed only within their associated
data set of images, until convergence was reached.
Figure 2 presents the isosurface rendering of
the three G40P reconstructed volumes. The three
reconstructions are of similar height (5.5 nm) and
possess a passing channel that runs open from
top to bottom with an even diameter of roughly
4 nm. The C3 and C3C6 reconstructions share a
1065
very similar architecture. The maps present three
main units that can be subdivided into two
masses. In other words, they are consistent with
a trimer of dimers. Rotational analysis of sections
taken perpendicularly to the rotational axis of the
volumes (data not shown) revealed that the C3
and C3C6 volumes possess two different faces,
one of predominant 3-fold symmetry (bottom
face) and an opposite face with 6-fold symmetry
(top face). Similar results were observed with the
homologous EcoDnaB helicase.29 Despite the
overall similarity between the C3 and C3C6
volumes, it is worth noting that there are some
minor but very reproducible differences. The
main one lies in a small protrusion sticking out
in three of the six monomers (arrowhead in
Figure 2), that probably corresponds to the small
protrusion mentioned in the 2D analysis (arrowhead in Figure 1(f)). Conversely, the C6 reconstruction (Figure 2, column C6) shows a hexamer
whose architecture is based on six identical
masses rendering a volume with 6-fold symmetry
on both faces. Nevertheless, the top and bottom
faces are easily distinguished, since the top face
shows a wider diameter with the six monomers,
of isolated appearance, reaching out towards the
exterior. The diameter of the bottom face is
narrower and the neighboring monomers make
closer contacts.
Three-dimensional structure of G40PDN109
Previously it was shown that the G40PDN109
enzyme, which contains the hinge region and the
CTD, but lacks the NTD, is able to unwind
DNA.27 We used here G40PDN109 to localize the
NTD in wt G40P structure by 3D-EM studies.
Figure 3(a) and (b) presents a representative
tilting pair of negatively stained G40PDN109 in
buffer
A.
Globular-shaped
particles
of
G40PDN109 show morphology and dimensions
similar to the wild-type particles. The quality of
the G40PDN109 images (in terms of contrast and
sharpness) was systematically lower than those
of G40P. Also, the number of particles was
always lower in the G40PDN109 samples,
although both samples were diluted to the
same concentration. A total of 4195 non-overlapping pairs of particles were manually selected
and processed as described for G40P.
Unlike G40P, classification showed no groups
of predominant 3-fold symmetry. A group of 1500
particles was selected, whose spectra had a single
peak of energy at the 6-fold component
(Figure 3(c)). The average image for this C6
group (Figure 3(d)) shows a very similar appearance to the average image of the C6 group of
G40P (Figure 1(h)), indicating that G40PDN109
oligomerizes as a hexamer. This result also
suggests that the CTD and NTD are on top of
each other along the rotational axis, because there
is no clear reduction in the dimension of the
particle.
1066
G40P Helicase Quaternary Polymorphism
Figure 2. 3D structures of the three G40P architectures. Isosurface renderings at threshold values accounting for 100%
of the mass are shown for the C3 (left column), C3C6 (middle column) and C6 (right column) groups. Reconstructions are
shown from different viewing points (top, capsized and side on top, middle and bottom rows, respectively). Top and
capsized views are related by a 458 rotation, and side and top views by a 908 rotation. Arrowheads mark the main
differences between the very similar C3 and C3C6 structures. The maximum diameter is 14 nm and the height is roughly
5.5 nm for all structures. The estimated resolution is 24 Å in all instances.
The Random Conical Tilt approach was used to
calculate an initial volume, using only the set of
tilted images assessed to be C6. This volume was
refined through an iterative projection matching
algorithm. A final volume was then obtained
using all the images collected for this mutant,
because G40PDN109 oligomers do not seem to
exhibit quaternary polymorphism under the
buffer A conditions. This procedure ensures a
more complete coverage of the projection space
than the one typically obtained by using only the
tilted images.
Figure 3. EM and 2D analysis of G40PDN109 hexamers. (a) Untilted and (b) 508 tilted electron micrographs of
negatively stained G40PDN109 oligomers in buffer A. Arrowheads point to some representative particles; numbers
indicate correspondence between partners in the tilting pair. (c) Rotational power spectrum, calculated between radii 5–
8 nm of the particle, representative of images classified as C6, the only group found. (d) 2D average of the re-aligned C6
group. The bar represents 50 nm in (a) and 5 nm in (d).
1067
G40P Helicase Quaternary Polymorphism
Figure 4(a) shows isosurface renderings of the
final reconstruction obtained for G40PDN109. The
reconstructed volume (4.5 nm in height) consists of
six identical masses that, at threshold values
accounting for 100% of the mass, do not connect
with each other. Weak connections were observed at
threshold values accounting for 150% of the
expected mass. As opposed to the C6 architecture
of G40P, the widths of the top and bottom faces of
the G40PDN109 hexamer are roughly the same. The
six masses are arranged around an open channel
with dimensions (4 nm in diameter) and general
morphology very similar to those of the G40P
oligomers.
The C6 architecture of G40P and G40PDN109
oligomers was compared by direct superposition of
the reconstructed volumes. Best fitting of the
monomers of G40PDN109 onto the monomer of
G40P requires a slight upwards movement indicated by an arrow in Figure 4(b). After applying this
movement to all the monomers, the G40PDN109
volume superimposed well, and completely, within
the region of the wider face of the G40P C6 structure
(Figure 4(c)). Conversely, there was not a region
equivalent to the narrower face of G40P in the
G40PDN109 volume (arrowhead in Figure 4(c)).
From this comparison we conclude that the G40P
CTD corresponds to the wider face of the C6
volume, therefore the NTD (absent in
G40PDN109) is located at the narrower face. It is
then clear that, indeed, the CTD and NTD are one
on top of each other along the rotational axis.
Quaternary polymorphism of G40PDN109
Figure 4. 3D structure of G40PDN109. (a) Isosurface
rendering at a threshold value accounting for 100% of the
mass of G40PDN109 shown from different points of view,
as indicated. The structure has a maximum diameter of
about 14 nm and a height of 4.5 nm. A passing channel
runs open from top to bottom. The estimated resolution is
22 Å. (b) Close-up of the comparison between monomers
from C6 G40P (translucient silver) and G40PDN109 (filled
green) showing the place the monomer occupies in the
derivative hexamer (left). A slight upward movement of
the mutant monomer (curved black arrow) results in a
more accurate fitting (right). (c) The G40PDN109 volume
and the C6 structure of G40P are compared after fitting
each monomer. It is noticeable the accurate coincidence
between the two structures at the wider face of the G40P
C6 volume, while the narrower face (indicated by an
arrowhead) remains empty after superposition.
In the structural studies of G40PDN109 in buffer
A presented in the previous section we did not
detect any polymorphism, with all the oligomers
belonging to the C6 group. However, and because
polymorphism is typical in many replicative
helicases,17,18,20 we wanted to explore if other buffer
conditions could determine the appearance of
oligomeric forms other than the C6. This task was
further motivated by a recent work27 in which it is
described that G40PDN109 indeed has a quite
different activity depending on the buffer composition. All these aspects considered, we analyzed
the structural organization of G40PDN109 by 2DEM in exactly the same buffers used in the assays,
the helicase activity27, that is buffers B and E. The
main difference between buffers B and E is the
pH value. In buffer B (pH 7.5) G40PDN109 had
almost no helicase activity, while in buffer E
(pH 6.5) the mutant shows the highest helicase
activity.27.
The wt G40P helicase at pH 6.5 showed a
quaternary polymorphism similar to the one
described here (Figure 5). However, G40PDN109
under this incubation condition displays a different
behavior; it assembles into the three main different
architectures described for the wt enzyme, although
the C6 architecture remains as the most populated
group. The C 6 average images obtained for
G40PDN109 and for wt G40P at pH 6.5 share a
similar organization as stated in the previous
section. Similarly, the general features of the C3
average image of G40PDN109 are comparable in
shape and size to those obtained for the wt G40P.
This similarity decreases when comparing the C3C6
state of both proteins. In fact, whilst in G40P
alternate monomers are clearly seen, in
G40PDN109 the six monomers are quite similar.
This fact is clearly reflected in the rotational power
spectra of both average images. Thus, the C3C6 state
of G40PDN109 presents a rotational power spec-
1068
G40P Helicase Quaternary Polymorphism
features are similar to those found for the hexameric
population. In addition, within the hexameric
population of G40PDN109 no quaternary polymorphism is found because all the hexameric
particles present 6-fold rotational symmetry, and
no particles with 3-fold rotational symmetry were
found.
Modelling of the NTD and CTD of G40P
Figure 5. Quaternary polymorphism of G40PDN109.
2D-average images of G40P and G40PDN109 oligomers
after incubation in buffer E (pH 6.5) and B (pH 7.5). The
pH value of these two buffers is indicated on the left-hand
side of the panel. The three main different architectures
(indicated at the top of the panel) were found for G40P at
both pHs and for G40PDN109 at pH 6.5. G40PDN109
oligomers incubated at pH 7.5 were found as heptamers
(C7) and as C6 group hexamers. The contour map of the
symmetrized average image and the rotational power
spectrum of each average image are shown. The
proportion of each group (%) from the total number of
frontal views is indicated for each 2D analysis (rows).
Bars represent 5 nm.
trum in which the 3-fold component is highly
reduced while the 6-fold component gets up to 90%
of the energy. On the contrary, the rotational power
spectrum of G40P intermediate state shows that,
although the 6-fold component is the most powerful
harmonic (70% energy), there is also a 3-fold
component that gets up to 30% of the energy.
Thus it seems than the C3C6 state found for
G40PDN109 at pH 6.5 is more closely related to
the 6-fold symmetry that in the case of wt G40P.
At pH 7.5, wt G40P presents a quaternary
polymorphism identical to the one described at
pH 6.5, with little modifications in the percentages
of particles assigned to each class. Conversely,
incubation of G40PDN109 at pH 7.5 renders
oligomers with a different quaternary polymorphism than the one found at pH 6.5. In fact, at
pH 7.5, G40PDN109 oligomerizes not only as a
hexamer (60% of the frontal views), but also as a
heptamer (Figure 5, C7) (40%) whose structural
Since there is no atomic resolution information
about G40P helicase, homology models were built
for G40P NTD and CTD. The sequences of G40P
NTD and CTD were aligned against the sequences
of EcoDnaB NTD and T7gp4 CTD, respectively.
Figure 6(a) shows the results of these sequence
alignments, together with a secondary structure
prediction for the G40P sequence. The percentages
of identities and conservative substitutions between
the sequences are 27% and 45% for the NTDs, and
24% and 41% for the CTDs, respectively. Moreover,
there is a good agreement between the position of
the predicted secondary structure elements for the
sequences of the NTD and CTD of G40P and the
position of the secondary structure elements in their
respective homologous structures.30,31 The
sequence identity and the similar distribution of
the secondary structure elements strongly suggest
that G40P domains share a similar fold with their
homologues.
The models of G40P NTD and CTD are shown in
Figure 6(b) and (c) together with the structures of
their corresponding homologues. Three relevant
areas (motif H4, the P-loop and an Arg-finger,
Arg522 in T7gp4 sequence) are highlighted. The
motif H4 (green-filled circle) has been described to
interact with DNA,16 and in the X-ray structure of
T7gp4 is located on the wall of the central channel.
The P-loop (grey-filled circle) forms part of the
NTP-binding pocket. The Arg-finger (pink-filled
circle), contributed by a neighboring monomer in
the case of oligomeric helicases, is also a constituent
of the NTP-binding site.10,31
Fitting of the G40P NTD and CTD into the C6
architecture
The models of the G40P NTD and CTD were
filtered to the resolution obtained for the 3D
reconstructions. Figure 7(a) presents two views of
the 3D model of the CTD within its corresponding
low-resolution envelope (shown in yellow). Motif
H4 (green), the P-loop and surrounding (grey), the
Arg-finger (pink), and the N-terminal residues are
indicated. The corresponding low-resolution envelope containing the 3D model of the NTD of G40P is
shown in Figure 7(b).
The low-resolution models of the NTDs and
CTDs were fitted into the reconstructed volume of
G40P C6 architecture. The model of the CTD was
manually placed within the larger mass of G40P,
previously identified as the CTD. The models were
positioned so that they adopt an orientation very
1069
G40P Helicase Quaternary Polymorphism
Figure 6. Modelling of the NTD and CTD of G40P. (a) Sequence alignment of the NTD and CTD of G40P against
EcoDnaB and T7gp4, respectively. The sequence of the hinge region of G40P is also shown. Secondary structure elements,
predicted in G40P and observed in the homologues, are boxed in different colors (helices in blue and strands in red).
Filled circles highlight areas of known relevance in G40P homologues (e.g. the P-loop in grey, the Arg-finger in pink and
motif H4 in green). (b) Homology model of the G40P NTD (left) and structure of EcoDnaB NTD (right). (c) Homology
model of the G40P CTD (top) and structure of T7gp4 CTD (bottom). Relevant areas are shown in the two structures using
the same coloring scheme as in (a).
similar to that of the T7gp4 CTD in the crystal
structure.32 Afterwards, the smaller and globular
NTD was manually fitted into the remaining
density. Under this disposition, each NTD bridges
two adjacent CTDs (Figure 7(c)). The N-terminal
residues of the CTD are positioned in the external
face of the ring, so the more feasible monomer
consists of a CTD and the NTD just below. The six
CTDs orient their motif H4 towards the inside of the
channel (green spheres in Figure 7(d)), the place
where DNA interaction during helicase activity is
assumed to take place.14 The region equivalent to
that of Arg522 in T7gp4 (pink dots) protrudes out
towards the P-loop (black dot) of a neighboring
monomer, in the same way reported for the T7gp4
structure.
Fitting the G40P NTD and CTD into the 3-fold
architectures
Maintaining the overall domain organization of
the G40P C6 architecture (Figure 8(a)), the NTD and
CTD homology models were fitted in the map of the
3-fold architectures (C3 and C3C6). In the proposed
domain arrangement of the 3-fold architectures
(Figure 8(b)), three of the monomers remain
unmodified with respect to their organization in
the C6 architecture (e.g. monomer 1 in Figure 8). The
NTDs of the other three monomers (e.g. monomer 2
in Figure 8) move from their position in the C6
architecture towards the periphery of the hexamer
(Figure 8(c), black arrows). This results in a new
interaction between the displaced NTDs and their
neighboring CTDs (Figure 8(d)). The density that
connects these NTDs with their corresponding
CTDs (asterisk in Figure 8(b)) would account for
the hinge region, which has not been included in the
hybrid model because there is not structural
information (asterisk in Figure 8(d)).
The three CTDs, associated with the NTDs
whose position changes, swivel upwards
(Figure 8(c), red arrow) with respect to their
position in the C6 architecture (Figure 8(a)). This
movement results in (i) a downward movement
in the inner channel of the motif H4, and (ii) the
formation of two kinds of NTP-binding sites,
termed type-S (short) and type-L (large). These
NTP-binding sites differ by the distance between
the P-loop in one monomer and the Arg-finger in
the next. This separation is longest in type-L sites
and shortest in type-S sites.
Discussion
The ability of G40P to adopt different quaternary
organizations is shared by other members of the
DnaB family (EcoDnaB),17 and possibly by members
of different families (RepA of plasmid RSF1010 and
papilloma virus helicase E1).18,20 This suggests that
the helicase quaternary polymorphism may be
relevant to the activity of this type of enzymes,
1070
G40P Helicase Quaternary Polymorphism
Figure 7. Domain localizations in
the C6 structure of the G40P hexamer. (a) Two views, related by a
908 rotation, of G40P CTD model
and its low-resolution envelope
(yellow). Colored spheres indicate
the P-loop (grey), Arg-finger (pink)
and motif H4 (green). The positions
of the CTD N-terminal residues are
indicated (NTaa). (b) G40P NTD
model and its low-resolution envelope (blue). (c) Close-up of side and
top views of the NTD and CTD
models fitted into the G40P C6
reconstruction (silver). The CTD
locates on the wider face and the
NTD on the narrower one of the C6
structure. For simplicity, only a
portion of the oligomer is shown.
The N-terminal residues of the CTD
are indicated. Letters N and C label
the NTD and CTD of the same
monomer. (d) Complete overview
of the NTDs and CTDs fitted into
the C6 hexamer. The 3D-EM map is
shown for only half of the ring.
Filled semi-spheres mark the
locations in the C6 hexameric structure of the P-loop (black) and Argfinger (pink), which form the NTP
binding sites. Motif H4 (green) lines
the wall of the inner channel of the
hexamer where the DNA binds.
although no firm link has been established between
polymorphism and its functionality
Here, we have addressed the 3D characterization
of G40P quaternary polymorphism. This enzyme is
an adequate model to understand the initial steps of
the replication process in Gram-positive bacteria.
Independent 3D reconstructions of the three architectures of G40P18 have been described. Inspection
of the three reconstructed volumes for G40P shows
that the C3 and C3C6 architectures share a very
similar organization. The principal difference consists of the alternate protrusion observed in three of
the six monomers, which corresponds with the
principal difference observed for the average
images of top views of groups C3 and C3C6. In the
EM 2D image processing, we could only detect
these differences after paying specific attention to
the outermost corona of the projection images. The
origin of this local heterogeneity could reside in a
higher flexibility on this area. Slight movements
might easily produce a gradient of architectures
with different values for the 3-fold and the 6-fold
components in the rotational power spectra of the
top views. So, the quaternary polymorphism of
G40P could be viewed as an architecture where six
monomers arrange with the same orientation (C6
architecture) plus several 3-fold symmetry architectures where three out of the six monomers would
benefit from higher mobility than the other three (C3
and C3C6 architectures).
The characterization of G40P 3D structures in their
different conformational states is certainly a first step
towards a better structure–function understanding of
this particular system. Additionally, the emerging
view is that we are facing a situation that seems to be
general to DNA replicative helicases. In this context it
is natural to make comparison with previous
structural data on other replicative helicases and, in
particular, with EcoDnaB. Indeed, we find that this
transition between the 3-fold and 6-fold states has
been also observed in EcoDnaB in 2D and 3D
studies17,33–35 including a model on domain assignment that is similar to the one proposed here.35
However, we provide for the first time a direct
experimental evidence for the domain mapping as
well as studies on the implication of some experimental conditions in the quaternary polymorphism.
These findings provide a level of detail and precision
that could shed light on understanding the dynamics
of these ubiquitous systems.
Besides this general agreement there are, however,
discrepancies among the structural studies, in
G40P Helicase Quaternary Polymorphism
1071
Figure 8. Domain localization in
the 3-fold architectures and the
transition between symmetry
states. Side and top views of the
distribution of the NTD, in blue,
and the CTD, in yellow, for the
reconstructed C6 (a) and C 3C6
(b) architectures of G40P. Numbers
from 1 to 6 indicate the different
monomers; N and C indicate the
NTD or CTD for each monomer.
The distribution proposed for the
C3C6 architecture is also applicable
to the C3 one. Close-up of side and
top views of the domain arrangements in the C6 architecture (c) and
in the 3-fold architectures (d). For
clarity only three monomers are
shown. The P-loop (filled black
semi-sphere) together with the
Arg-finger of a neighboring monomer (filled pink semi-sphere) make
up an NTP-binding site, indicated
by an ellipse. The motif H4 (filled
green semi-sphere) is indicated. All
six NTP binding sites in the C6
architecture are identical. However,
two kinds of NTP-binding sites,
type-L and type-S, exist in the 3fold structures. Three alternate H4
motifs change their location on the
wall of the channel in the 3-fold
architectures. The asterisks in
(b) and (d) indicate density linking
the NTD to the CTD, proposed to
account for the hinge region.
particular concerning the dimension of the reconstructions in the direction of the symmetry axis.
Although this Z-compression/extension might represent a genuine structural difference between these
two systems, we cannot rule out that this discrepancy
is simply due to the different methodologies used for
each reconstruction. The Random Conical Tilt scheme
used here is a completely reference-free reconstruction method, as opposed to other approaches in
which the use of a particular reference may introduce
a model bias. Also, it is acknowledged that during the
negative staining process in the specimen preparation
for electron microscopy certain compression in the
direction perpendicular to the grid may occur. If this
were the case, all the images used for Random Conical
Tilt reconstruction would correspond to the same
structure, while reconstruction using other
approaches may merge images from structures
compressed in different directions.
Function of the G40P NTD
In an attempt to clarify the involvement of the
NTD of DnaB-like replicative helicases in their
quaternary polymorphism and biological activities,
a mutant lacking this domain (G40PDN109) was
analyzed. Previous studies have shown that
G40PDN109 has helicase, ATPase and DNA
binding activities.27 The results presented here
show that G40PDN109 is able to assemble as a
hexamer, although the low percentages of particles, in comparison with the wt enzyme, suggests
that the oligomers of the mutant protein are more
unstable. Indeed, analysis by gel filtration chromatography showed that, in conditions where
G40P elutes as a hexamer, G40PDN109 elutes as
hexamer and dimer. Thus, G40PDN109, at high
protein concentration (1.3 mM), elutes as a hexamer
in the presence of low salt (50 mM NaCl) and as a
mixture of hexamers and dimers at high salt
(300 mM NaCl). At low protein concentration
(0.4 mM) and high salt only dimers of
G40PDN109 are found (our unpublished results).
Under all conditions tested G40P elutes as a
hexamer. Thus, it seems that G40P NTD is not
strictly necessary for the enzymatic activities of the
enzyme, although this domain could contribute to
the stability of the hexamer.
1072
The Random Conical Tilt approach was used to
calculate the non-biased initial reference volume for
the refinement of the G40PDN109 reconstruction. At
threshold values accounting for 100% of the mass, no
connections could be observed between the six
identical masses that made up the C6 architecture of
G40PDN109, although they must exist since
functional hexamers are observed. Connectivity
among these regions showed up only at low
thresholds (150% of the expected mass). As stated
before, G40PDN109 hexamers seem to be structurally
less stable than G40P ones. So, the regions of contact
between neighboring monomers in G40PDN109
could be less structured and more flexible. Consequently, the averaging process intrinsic to the 3D
reconstruction might have reduced the averaged
density in these regions.
We have shown that G40PDN109 is able to
oligomerize into the three architectures described
for the wt enzyme, although the C6 architecture was
always the most predominant one. Therefore, the
CTD plus the linker region of G40P are enough to
display the quaternary polymorphism of the wt
enzyme. A similar situation is found for the RepA
helicase of plasmid RSF1010, which constitutively
lacks an NTD equivalent to those of DnaB or G40P.
Instead of an NTD, RepA possesses a short track of
four amino acid residues (the N-terminal hook),
revealed to be critical in the formation and
maintenance of the hexameric structure. 9
Under certain experimental conditions, RepA oligomers were found to exist in 6-fold and 3-fold
organizations.19 This seems to indicate that the
helicase (C-terminal) domain plus the linker or
the N-terminal hook might be sufficient to initiate
the structural reorganization leading to the transition between the C6 and C3 symmetry states.
In contrast to the observed difference in wild-type
G40P, the top and bottom faces of the reconstructed
G40PDN109 volume are very similar. Interestingly,
the mutant protein is able to unwind dsDNA with
both 5 0 to 3 0 and with 3 0 to 5 0 polarity.27 It is
tempting to correlate the lack of functional polarity
in G40PDN109 with a decreased structural polarity
of the reconstructed volume, in contrast with the
known functional polarity of the wild-type enzyme.
Summarizing, we find that the presence of the
G40P NTD confers stability to the hexamer and
structural and functional polarity to the enzyme, and
facilitates the transition between the different architectures.
The symmetry transition
Comparison of the C6 volumes of G40P and
G40PDN109 allowed us to locate the NTD and CTD
within the C6 architecture. This information was
used to understand the domain movements that
could explain the organization of the 3-fold
architectures (C3 and C3C6), which share a very
similar 3D arrangement.
In the transition from the C6 architecture to the
3-fold architectures, three of the monomers suffer
G40P Helicase Quaternary Polymorphism
minor modifications whereas the other three
undergo significant changes. The NTDs of the
later monomers move away to the periphery of
the structure. This results in a modification in the
structure of the hinge region, which would then
exist in two different conformations, retracted and
extended. Three of the hinge regions would adopt
an extended conformation and the other three
would be in a retracted one, similar to the one
adopted by all hinges in the C6 architecture.
The six CTDs in the C6 architecture arrange in an
identical orientation, similar to the heptameric form
of T7gp4.32 Consequently, all the intermonomeric
interfaces are equivalent. However, in the 3-fold
architectures, three of the CTDs undergo a swiveling
within the plane of the ring similar to the one
described for the hexameric and helical form of
T7gp4.10,31 As a result, the interfaces between
monomers change and two different types of NTPbinding sites appear: type-S and type-L. In the type-S
site a shorter distance between the P-loop and the
Arg-finger is observed than in the type-L site. This
correlates with biophysical studies36 that showed that
EcoDnaB possesses two types of NTP-binding sites,
with low and high affinity, and performs ATP
hydrolysis in a bimodal fashion. These studies were
carried out at pH 8.0, for which only the C3
architecture has been observed.34
The rotation between neighboring monomers
observed in T7gp4 has been proposed to be the
motion responsible for the DNA translocation.10 In
helicases with quaternary polymorphism, the alternation between the different architectures would
produce similar conformational changes needed for
DNA translocation. The movements of the CTDs
that swivel upwards, in the transition from the C6
architecture to the 3-fold architectures, could be the
mechanism to pull the DNA (Figure 9). A motif H4
in one monomer, interacting with DNA, descends in
the transition, making possible a new interaction of
the DNA with a neighboring motif H4. A new
transition from the 3-fold architecture to the 6-fold
(C6) architecture returns the structure to the initial
step. The DNA–protein interaction would involve
one or two subunits, as was described for T7gp4.37
Conclusion
Here we present for the first time 3D reconstruction of the three different conformational states of
the G40P helicase. Also, an experimentally based
domain assignment is performed. From these
results, it is proposed that the symmetry transition
is directly linked to DNA translocation. Indeed, it is
important to note that Gram-positive G40P helicase
would undergo similar rearrangement to transit
between the symmetry states to Gram-negative
EcoDnaB helicase.35 Although the exact mechanism
for how the NTP hydrolysis energy is coupling to
the DNA unwinding remains unknown, direct
experimental data suggest that the simultaneous
presence of the different architectures is needed for
1073
G40P Helicase Quaternary Polymorphism
Figure 9. Mechanism for DNA translocation. The conformational changes in the transition from the C6 architecture
(a) to the 3-fold architectures (b) would allow the translocation of DNA. A new transition to the C6 state (c) returns to the
initial state but with a movement of the DNA from one subunit to another. The different states interacting with the DNA
are shown from the channel. The CTDs and NTDs are represented in light and dark grey, respectively. The DNA is
represented as a tube. Interactions between DNA and the motif H4 of the subunits are depicted with black asterisks.
the enzymatic activity. Thus, G40PDN109 shows
helicase activity only under conditions where
quaternary polymorphism is feasible (this work).
In addition, the biochemically inactive helicaseloader complex of E. coli (DnaB$DnaC)38 was
observed to exist only in the C3 architecture.34
Consequently, the existence of several architectures
seems to be important in order to orchestrate the
different protein–protein and protein–DNA interactions required to accomplish DNA recognition
and loading onto the origin of replication. The
alternation between these architectures could help
in the transduction of the signal generated in the
NTP-binding pocket and in the translocation and
unwinding along the DNA.
Materials and Methods
Electron microscopy
G40P and G40PDN109 were purified as described.24,26
Samples were dialyzed against buffer A (50 mM
Hepes (pH 7.0), 50 mM NaCl, 10 mM MgCl2, 1 mM
dithiothreitol, 1 mM ATP) for Random Conical Tilt
experiments, and against buffer B (50 mM Hepes–NaOH
(pH 7.5), 1 mM DTT, 2 mM ATP, 10 mM MgCl2, 15 mM
NaCl, 5% (v/v) glycerol, 50 mg/ml BSA) or E (50 mM
Pipes–NaOH (pH 6.5), 1 mM DTT, 2 mM ATP, 10 mM
MgCl2, 15 mM NaCl, 5% glycerol, 50 mg/ml BSA) for the
analysis of the relationship between structure and helicase activity. Subsequently, diluted solutions of G40P and
G40PDN109 (corresponding to 0.7 mM when referred to
the hexamers) were adsorbed onto glow-discharged
collodion/carbon coated grids, stained with 2% (w/v)
uranyl acetate, and visualized in a Jeol 1200 EX-II
transmission electron microscope at 60,000! magnification and 80 kV accelerating voltage. Micrographs were
taken on Kodak SO-163 plates. Tilting pairs were
recorded first at 508 and then at 08, using a eucentric
goniometer and low-dose conditions.
Image processing
Micrographs were digitized in a Zeiss-Integraph
scanner (Photoscan TD model) at a pixel size of 4.1 Å.
The software packages Xmipp39,40 and SPIDER41 were
used for image processing and classification. A total
number of 13,743 particles for G40P, and 4195 for
G40PDN109 were selected (80!80 pixels) for the Random
Conical Tilt experiment. In the analysis of the relationship
between structure and helicase activity 7832 and 18,013
particles were selected for G40P at pH 6.5 and 7.5,
respectively, and 7499 and 6433 particles for G40PDN109
at pH 6.5 and 7.5, respectively. Particles were centered by
cross-correlation against a radially symmetrised average
image of the unaligned population. Rotational power
spectra42 were then calculated between radii of 2–8 nm,
covering the whole particle. A self-organizing map
algorithm (KerDenSOM)43 was employed to classify the
initial population, based on variability among the
rotational power spectra. Groups having the 3 or 6
rotational harmonic as the highest one were selected.
Each subset was translationally and rotationally aligned
by cross-correlation methods and the so-called pyramidal
system for pre-alignment construction.44 At this stage, the
spectra were calculated only on pixels between radii of
5–8 nm. Final assessments of the homogeneity of the
image subsets were carried out by KerDenSOM analysis
of both, the rotational power spectra of the images, and
the images themselves.
3D reconstruction
3D reconstructions were performed following the
Random Conical Tilt scheme.28 Projection matching 3D
orientation searches were carried out with SPIDER.45 For
G40P, showing quaternary polymorphism, only the tilting
counterpart images of the top views selected after
classification were used in the reconstruction of each
class. For G40PDN109, which did not show polymorphism, all the untilted and tilted images were used in the
angular refinement. Resolution was assessed by the
Fourier Shell Correlation criterion.46 The final 3D
reconstructions were calculated with the ART algorithm.47 The resulting volumes were filtered to their
estimated resolution. Isosurface renderings of the recon-
1074
structed electron densities were displayed using
AMIRA†, at a threshold value accounting for 100% of
the expected mass assuming a mean protein density of
1.33 g/cm3.48
Homology modeling
The G40P sequence (Q38152)49 was obtained from the
ExPASy Molecular Biology Server‡.50 Secondary structure predictions were calculated for the whole amino acid
sequence of the G40P polypeptide using the PSIPRED
Protein Structures Prediction Server§.51 Homology
models were built with the SWISS-MODEL Protein
Modelling Servers.52–54 The G40P NTD and CTD models
were built, using as templates the NTD of EcoDnaB (pdb
accession code 1b79),30 and the CTD of T7gp4 (pdb
accession code 1cr2),31 respectively. Homology models
were filtered to 24 Å and manually fitted to the G40P
reconstructions using AMIRA.
Electron Microscopy Data Bank accession numbers
The maps have been deposited at the EMBL-EBI
Electron Microscopy Data Bank (EMDB). The identifier
codes are 1135 for the C6 architecture, 1146 for the C3C6,
1147 for the C3, and 1148 for the G40PDN109 mutant map.
Acknowledgements
We are very grateful to Dr C. San Martin and Dr S.
H. W. Scheres for their expert advice. We thank C.
Patiño, R. Arranz and C. Rollo for their help at the
Electron Microscopy Facility of the Centro Nacional
de Biotecnologı́a. This work was partially supported by grants BFU2004-00217 from Dirección
General de Investigación, Ministerio de Educación
y Ciencia (MEC), 1R01H67465-8 from National
Institutes of Health, FP6-502828 (3D-EM Network
of Excellence) from European Union, G03-185 (IM3
Network) from Fondo de Investigaciones Sanitarias
to J.M.C.; BIO 2001-4339 from MEC to J.M.C. and
J.C.A.; BMC2003-00150 from MEC and S-0505/
MAT/0283 from Comunidad Autónoma de Madrid
to J.C.A.; BMC 2003-01969 from MEC to S.A. R.N. is
recipient of a Pre-doctoral Fellowship from the
CSIC (Programa I3P), supported by the European
Social Fund. A Postdoctoral Fellowship from CAM
partially supported Y.R. and P.M. L.E.D. holds a
contract funded by Programa Ramón y Cajal.
References
1. Lohman, T. M. & Bjornson, K. P. (1996). Mechanisms
of helicase-catalyzed DNA unwinding. Annu. Rev.
Biochem. 65, 169–214.
† http://amira.zib.de
‡ http://us.expasy.org
§ http://bioinf.cs.ucl.ac.uk/psipred/psiform.html
s http://swissmodel.expasy.org//SWISS-MODEL.
html
G40P Helicase Quaternary Polymorphism
2. Gorbalenya, A. E. & Koonin, E. V. (1993). Helicases:
amino acid sequence comparisons and structurefunction relantionships. Curr. Opin. Struct. Biol. 3,
419–429.
3. Yao, N., Hesson, T., Cable, M., Hong, Z., Kwong,
A. D., Le, H. V. & Weber, P. C. (1997). Structure of the
hepatitis C virus RNA helicase domain. Nature Struct.
Biol. 4, 463–467.
4. Korolev, S., Hsieh, J., Gauss, G. H., Lohman, T. M. &
Waksman, G. (1997). Major domain swiveling
revealed by the crystal structures of complexes of
E. coli Rep helicase bound to single-stranded DNA
and ADP. Cell, 90, 635–647.
5. Subramanya, H. S., Bird, L. E., Brannigan, J. A. &
Wigley, D. B. (1996). Crystal structure of a DExx box
DNA helicase. Nature, 384, 379–383.
6. Roleke, D., Hoier, H., Bartsch, C., Umbach, P.,
Scherzinger, E., Lurz, R. & Saenger, W. (1997).
Crystallization and preliminary X-ray crystallographic and electron microscopic study of a bacterial
DNA helicase (RSF1010 RepA). Acta Crystallog. sect.
D, 53, 213–216.
7. Bird, L. E., Hakansson, K., Pan, H. & Wigley, D. B.
(1997). Characterisation and crystallisation of the
helicase domain of bacteriophage T7 gene 4 protein.
Nucl. Acids Res. 25, 2620–2626.
8. Story, R. M. & Steitz, T. A. (1992). Structure of the recA
protein-ADP complex. Nature, 355, 374–376.
9. Niedenzu, T., Roleke, D., Bains, G., Scherzinger, E. &
Saenger, W. (2001). Crystal structure of the hexameric
replicative helicase RepA of plasmid RSF1010. J. Mol.
Biol. 306, 479–487.
10. Singleton, M. R., Sawaya, M. R., Ellenberger, T. &
Wigley, D. B. (2000). Crystal structure of T7 gene 4
ring helicase indicates a mechanism for sequential
hydrolysis of nucleotides. Cell, 101, 589–600.
11. Caruthers, J. M. & McKay, D. B. (2002). Helicase
structure and mechanism. Curr. Opin. Struct. Biol. 12,
123–133.
12. Patel, S. S. & Picha, K. M. (2000). Structure and
function of hexameric helicases. Annu. Rev. Biochem.
69, 651–697.
13. Bujalowski, W. & Jezewska, M. J. (1995). Interactions
of Escherichia coli primary replicative helicase DnaB
protein with single-stranded DNA. The nucleic acid
does not wrap around the protein hexamer. Biochemistry, 34, 8513–8519.
14. Ayora, S., Weise, F., Mesa, P., Stasiak, A. & Alonso,
J. C. (2002). Bacillus subtilis bacteriophage SPP1
hexameric DNA helicase, G40P, interacts with forked
DNA. Nucl. Acids Res. 30, 2280–2289.
15. Egelman, E. H., Yu, X., Wild, R., Hingorani, M. M. &
Patel, S. S. (1995). Bacteriophage T7 helicase/primase
proteins form rings around single-stranded DNA that
suggest a general structure for hexameric helicases.
Proc. Natl Acad. Sci. USA, 92, 3869–3873.
16. Washington, M. T., Rosenberg, A. H., Griffin, K.,
Studier, F. W. & Patel, S. S. (1996). Biochemical
analysis of mutant T7 primase/helicase proteins
defective in DNA binding, nucleotide hydrolysis,
and the coupling of hydrolysis with DNA unwinding.
J. Biol. Chem. 271, 26825–26834.
17. Yu, X., Jezewska, M. J., Bujalowski, W. & Egelman,
E. H. (1996). The hexameric E. coli DnaB helicase can
exist in different quaternary states. J. Mol. Biol. 259,
7–14.
18. Bárcena, M., Martı́n, C. S., Weise, F., Ayora, S., Alonso,
J. C. & Carazo, J. M. (1998). Polymorphic quaternary
1075
G40P Helicase Quaternary Polymorphism
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
organization of the Bacillus subtilis bacteriophage
SPP1 replicative helicase (G40 P). J. Mol. Biol. 283,
809–819.
Bárcena, M. (2000). Structural analysis of the quaternary polymorphism in replicative helicases and of
the DnaB$DnaC complex of Escherichia coli. PhD
Thesis, Universidad Autonoma de Madrid.
Fouts, E. T., Yu, X., Egelman, E. H. & Botchan, M. R.
(1999). Biochemical and electron microscopic image
analysis of the hexameric E1 helicase. J. Biol. Chem.
274, 4447–4458.
Biswas, S. B., Chen, P. H. & Biswas, E. E. (1994).
Structure and function of Escherichia coli DnaB
protein: role of the N-terminal domain in helicase
activity. Biochemistry, 33, 11307–11314.
Nakayama, N., Arai, N., Kaziro, Y. & Arai, K. (1984).
Structural and functional studies of the dnaB protein
using limited proteolysis. Characterization of
domains for DNA-dependent ATP hydrolysis and
for protein association in the primosome. J. Biol. Chem.
259, 88–96.
Frick, D. N., Baradaran, K. & Richardson, C. C. (1998).
An N-terminal fragment of the gene 4 helicase/
primase of bacteriophage T7 retains primase activity
in the absence of helicase activity. Proc. Natl Acad. Sci.
USA, 95, 7957–7962.
Pedré, X., Weise, F., Chai, S., Luder, G. & Alonso, J. C.
(1994). Analysis of cis and trans acting elements
required for the initiation of DNA replication in the
Bacillus subtilis bacteriophage SPP1. J. Mol. Biol. 236,
1324–1340.
Ayora, S., Stasiak, A. & Alonso, J. C. (1999). The
Bacillus subtilis bacteriophage SPP1 G39P delivers and
activates the G40P DNA helicase upon interacting
with the G38P-bound replication origin. J. Mol. Biol.
288, 71–85.
Martı́nez-Jiménez, M. I., Mesa, P. & Alonso, J. C.
(2002). Bacillus subtilis tau subunit of DNA polymerase III interacts with bacteriophage SPP1 replicative
DNA helicase G40P. Nucl. Acids Res. 30, 5056–5064.
Mesa, P., Ayora, S. & Alonso, J. C. (2005). Bacillus
subtilis bacteriophague SPP1 G40P helicase lacking
the N-terminal domain translocate bi-directionally.
J. Mol. Biol.. In the press.
Radermacher, M., Wagenknecht, T., Verschoor, A. &
Frank, J. (1987). Three-dimensional reconstruction
from a single-exposure, random conical tilt series
applied to the 50S ribosomal subunit of Escherichia
coli. J. Microsc. 146, 113–136.
San Martı́n, C., Radermacher, M., Wolpensinger, B.,
Engel, A., Miles, C. S., Dixon, N. E. & 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.
Fass, D., Bogden, C. E. & Berger, J. M. (1999). Crystal
structure of the N-terminal domain of the DnaB
hexameric helicase. Struct. Fold. Des. 7, 691–698.
Sawaya, M. R., Guo, S., Tabor, S., Richardson, C. C. &
Ellenberger, T. (1999). Crystal structure of the helicase
domain from the replicative helicase-primase of
bacteriophage T7. Cell, 99, 167–177.
Toth, E. A., Li, Y., Sawaya, M. R., Cheng, Y. &
Ellenberger, T. (2003). The crystal structure of the
bifunctional primase-helicase of bacteriophage T7.
Mol. Cell. 12, 1113–1123.
San Martı́n, M. C., Stamford, N. P., Dammerova, N.,
Dixon, N. E. & Carazo, J. M. (1995). A structural
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
model for the Escherichia coli DnaB helicase based on
electron microscopy data. J. Struct. Biol. 114, 167–176.
Donate, L. E., Llorca, O., Bárcena, M., Brown, S. E.,
Dixon, N. E. & Carazo, J. M. (2000). pH-controlled
quaternary states of hexameric DnaB helicase. J. Mol.
Biol. 303, 383–393.
Yang, S., Yu, X., VanLoock, M. S., Jezewska, M. J.,
Bujalowski, W. & Egelman, E. H. (2002). Flexibility of
the rings: structural asymmetry in the DnaB hexameric helicase. J. Mol. Biol. 321, 839–849.
Bujalowski, W. & Klonowska, M. M. (1993). Negative
cooperativity in the binding of nucleotides to
Escherichia coli replicative helicase DnaB protein.
Interactions with fluorescent nucleotide analogs.
Biochemistry, 32, 5888–5900.
Yu, X., Hingorani, M. M., Patel, S. S. & Egelman, E. H.
(1996). DNA is bound within the central hole to one or
two of the six subunits of the T7 DNA helicase. Nature
Struct. Biol. 3, 740–743.
Wahle, E., Lasken, R. S. & Kornberg, A. (1989). The
dnaB-dnaC replication protein complex of Escherichia
coli. II. Role of the complex in mobilizing dnaB
functions. J. Biol. Chem. 264, 2469–2475.
Sorzano, C. O., Marabini, R., Velazquez-Muriel, J.,
Bilbao-Castro, J. R., Scheres, S. H., Carazo, J. M.
& Pascual-Montano, A. (2004). XMIPP: a new
generation of an open-source image processing
package for electron microscopy. J. Struct. Biol.
148, 194–204.
Marabini, R., Masegosa, I. M., San Martı́n,
M. C., Marco, S., Fernandez, J. J., de la Fraga,
L. G. et al. (1996). Xmipp: an image processing
package for electron microscopy. J. Struct. Biol.
116, 237–240.
Frank, J. S. B. & Dowse, H. (1981). SPIDER—a
modular software system for electron image processing. Science, 214, 1353–1355.
Crowther, R. A. & Amos, L. A. (1971). Harmonic
analysis of electron microscope images with
rotational symmetry. J. Mol. Biol. 60, 123–130.
Pascual-Montano, A., Donate, L. E., Valle, M.,
Bárcena, M., Pascual-Marqui, R. D. & Carazo, J. M.
(2001). A novel neural network technique for analysis
and classification of EM single-particle images.
J. Struct. Biol. 133, 233–245.
Marco, S. C. M., de la Fraga, L. G., Carazo, J. M. &
Carrascosa, J. L. (1996). A variant to the “random
approximation” of the reference-free alignment algorithm. Ultramicroscopy, 66, 5–10.
Radermacher, M. (1994). Three-dimensional reconstruction from random projections: orientational
alignment via Radon transforms. Ultramicroscopy, 53,
121–136.
Saxton, W. O. & Baumeister, W. (1982). The correlation
averaging of a regularly arranged bacterial cell
envelope protein. J. Microsc. 127, 127–138.
Marabini, R., Herman, G. T. & Carazo, J. M. (1998). 3D
reconstruction in electron microscopy using ART with
smooth spherically symmetric volume elements
(blobs). Ultramicroscopy, 72, 53–65.
Squire, P. G. & Himmel, M. E. (1979). Hydrodynamics
and protein hydration. Arch. Biochem. Biophys. 196,
165–177.
Alonso, J. C., Luder, G., Stiege, A. C., Chai, S., Weise, F.
& Trautner, T. A. (1997). The complete nucleotide
sequence and functional organization of Bacillus
subtilis bacteriophage SPP1. Gene, 204, 201–212.
Gasteiger, E., Gattiker, A., Hoogland, C., Ivanyi, I.,
Appel, R. D. & Bairoch, A. (2003). ExPASy: the
1076
G40P Helicase Quaternary Polymorphism
proteomics server for in-depth protein knowledge
and analysis. Nucl. Acids Res. 31, 3784–3788.
51. McGuffin, L. J., Bryson, K. & Jones, D. T. (2000). The
PSIPRED protein structure prediction server. Bioinformatics, 16, 404–405.
52. Peitsch, M. C. (1995). Protein modeling by E-mail. Bio/
Technology, 13, 658–660.
53. Schwede, T., Kopp, J., Guex, N. & Peitsch, M. C.
(2003). SWISS-MODEL: an automated protein homology-modeling server. Nucl. Acids Res. 31,
3381–3385.
54. Guex, N. & Peitsch, M. C. (1997). SWISS-MODEL and
the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis, 18, 2714–2723.
Edited by J. Karn
(Received 23 September 2005; received in revised form 26 January 2006; accepted 27 January 2006)
Available online 9 February 2006