Molecular Cell, Vol. 11, 1349–1360, May, 2003, Copyright 2003 by Cell Press
Modular Architecture of the Bacteriophage T7 Primase
Couples RNA Primer Synthesis to DNA Synthesis
and the eukaryotic/archaeal primases. Their sequences
are unrelated (Griep, 1995; Ilyina et al., 1992) and they
have different protein folds (Augustin et al., 2001; Keck
and Berger, 2001; Keck et al., 2000; Podobnik et al.,
2000). The marked divergence of bacterial and eukary-
otic primases makes the bacterial primases attractive
primases have six conserved sequence motifs (Ilyina et
al., 1992). Motif I residues ligate a zinc metal in the
conserved motifs (II–VI) are located in the C-terminal
determinant for specific binding to a DNA template (Ku-
sakabe and Richardson, 1996), and it is essential for the
synthesis of RNA and the priming of DNA synthesis
(Bernstein and Richardson, 1988; Kusakabe et al., 1999;
Powers and Griep, 1999). A crystal structure of the ZBD
from the Bacillus stearothermophilus primase revealed
it is a zinc ribbon, a motif present in a variety of DNA
and RNA binding proteins (Pan and Wigley, 2000).
The topoisomerase-primase (TOPRIM) fold is a con-
served structural motif within the active site regions of
the bacterial primases, type IA and type II topoisomer-
ases, and several other phosphoryl transfer enzymes
(Aravind et al., 1998). The weakly conserved signature
sequence of the TOPRIM fold is present in a number
of enzymes catalyzing the metal-assisted formation or
cleavage of phosphodiester bonds in DNA (http://pfam.
acidic residues of the TOPRIM motif were suggested as
metal chelating residues in the active sites of these
enzymes. Two crystal structures of the RPD of E. coli
in a shallow cleft adjacent to the TOPRIM fold and
nik et al., 2000). This cluster of conserved residues,
which bound to a Yttrium (Y2?) heavy metal in the crystal
structure, was suggested as the location of the primase
active site. The TOPRIM folds of DnaG and the topo-
isomerases are similar, but they are embedded in differ-
ent structural contexts (Berger et al., 1996; Lima et al.,
reflecting the different substrate binding requirements
and catalytic mechanisms of these enzymes.
Phage T7 encodes DNA polymerase, primase, heli-
case, and single-stranded DNA binding activities. To-
for T7 DNA polymerase, these proteins catalyze the rep-
mase and helicase activities reside in a bifunctional pri-
mase-helicase protein that assembles into ring-shaped
hexamers having distinctive primase and helicase do-
mains (Egelman et al., 1995; VanLoock et al., 2001) that
Rosenberg et al., 1996). The C-terminal helicase frag-
ment assembles into a hexameric ring-shaped structure
(Guo et al., 1999; Singletonet al., 2000) that translocates
5? to 3? along one DNA strand, displacing the comple-
mentary strand in a reaction powered by nucleotide hy-
drolysis (Bird et al., 1997; Matson et al., 1983; Patel et
Masato Kato, Takuhiro Ito, Gerhard Wagner,
Charles C. Richardson, and Tom Ellenberger*
Department of Biological Chemistry and Molecular
Harvard Medical School
Boston, Massachusetts 02115
DNA primases are template-dependent RNA polymer-
ases that synthesize oligoribonucleotide primers that
can be extended by DNA polymerase. The bacterial
primases consist of zinc binding and RNA polymerase
domains that polymerize ribonucleotides at templat-
ing sequences of single-stranded DNA. We report a
crystal structure of bacteriophage T7 primase that re-
veals its two domains and the presence of two Mg2?
ions bound to the active site. NMR and biochemical
data show that the two domains remain separated
until the primase binds to DNA and nucleotide. The
zinc bindingdomain alonecan stimulateprimer exten-
sion by T7 DNA polymerase. These findings suggest
that the zinc binding domain couples primer synthesis
with primerutilization bysecuring theDNA templatein
the primase active site and then delivering the primed
DNA template to DNA polymerase. The modular archi-
tecture of the primase and a similar mechanism of
priming DNA synthesis are likely to apply broadly to
DNA polymerases cannot begin DNA synthesis on sin-
gle-stranded templateswithout aprimer strandcomple-
mentary to the template. Conserved residues of DNA
polymerase contact the primer strand to align its
3?-hydroxyl for the condensation of nucleotides (Brauti-
gam and Steitz, 1998; Doublie and Ellenberger, 1998).
DNA primases synthesize short oligoribonucleotides
that are recognized and extended by DNA polymerase
(Frick and Richardson, 2001). This priming reaction is
fork to produce Okazaki fragments during semiconser-
vative replication. During DNA replication, the primase
gains access to the lagging strand template through the
action of a DNA helicase. In several replication systems,
the primase and helicase physically interact (Bird et al.,
2000; Chang and Marians, 2000; Marintcheva and Wel-
ler, 2001; Valentine et al., 2001). Primer synthesis is also
kov et al., 1999), and these interactions compensate for
the weak template binding of primases and regulate
The amino acid sequences of primases can be
grouped into two families—the prokaryotic primases
al., 1992). A plausible mechanism of DNA unwinding has
been proposed on the basis of crystal structures of
helicase fragments (Sawaya et al., 1999; Singleton et
The N-terminal primase of the T7 primase-helicase
initiates primer synthesis at the sequence 5?-GTC-3?,
templating the synthesis of pppAC. The conserved 3? C
of the priming site is required for RNA synthesis, but it is
1981). Functional tetraribonucleotide primers are syn-
thesized at the extended sequence 5?-(G/T)(G/T)GTC-
3?, which directs the synthesis of pppACCC, pppACCA,
and pppACAC (Frick and Richardson, 1999; Fujiyama
et al., 1981; Mendelman and Richardson, 1991). Other
bacterial primasesrecognize differentpriming sites(Ku-
sakabe et al., 1999; Kusakabe and Richardson, 1996).
T7 primase has two functions during replication. The
primase synthesizes oligoribonucleotides for extension
by DNA polymerase (Frick and Richardson, 2001). How-
ever, the short RNA primers synthesized by T7 primase
are not effectively utilized by T7 DNA polymerase unless
the primase is present (Kusakabe and Richardson,
tion of primers by DNA polymerase. The primase could
stimulate primer utilization by preventing the dissocia-
tion of short RNA primers from the DNA template (Kusa-
kabe and Richardson, 1997) or by directly interacting
with T7 DNA polymerase to secure the primed DNA
template in the polymerase active site (Chowdhury et
al., 2000; Kato et al., 2001).
Here, we report crystallographic and NMR studies of
an N-terminal fragment of the T7 primase-helicase that
encompasses the functional primase. This is the first
complete structure of a DnaG-type primase that is fully
active in primer synthesis. Together with NMR and bio-
chemical data, the structure reveals that the ZBD and
RPD of the primase are brought together during RNA
synthesis, and then the ZBD alone delivers the primed
DNA template to DNA polymerase.
Bipartite Structure of T7 Primase
A crystal structure of T7 primase was determined at
2.9 A˚ resolution (Table 1). The experimental electron
density revealed most residues of two primase mole-
cules in an asymmetric unit, with the exception of nine
N-terminal residues and a connecting loop between ?5
and ?1 which are disordered (Figures 1B and 2). Al-
though the crystals are grown in the presence of 2.5
is no electron density for a bound ATP. However, the
nated by the conserved cysteines of the ZBD.
T7 primase consists of two domains (Figure 1A), a
small N-terminal ZBD (residues 10–56) and a larger
C-terminal RPD (residues 68–255). An extended poly-
peptide linker (residues 57–67) connects the ZBD and
RPD. Its flexibility is evident from their different orienta-
tions in the two primase molecules that were crystal-
lized. After superimposing their RPDs, a rotation of
about 30? aroundan axis of helix ?1 isrequired to super-
impose the ZBDs. This flexible connection is notable
because it could provide a conditional switch for primer
synthesis, bringing together both domains once the pri-
mase engages a DNA template. In the hexameric pri-
actions between adjacent primase domains (Lee and
Richardson, 2002). During T7 phage growth, an N-ter-
minally truncated protein lacking the ZBD and the linker
of the primase-helicase is expressed by translation ini-
tiating from codon 64 (M64) (Dunn and Studier, 1983).
This truncated protein has the normal helicase activity,
but it is defective in primer synthesis (Bernstein and
Richardson, 1988) because it is precisely deleted for the
ZBD and the interdomain linker (Figure 4).
The Zinc-Binding Domain
The ZBD of T7 primase is a four-stranded, antiparallel
? sheet flanked by a C-terminal ? helix (Figure 1A). Its
architecture is characteristic of the zinc-ribbon family
of metal binding motifs (Qian et al., 1993), and it closely
resembles the zinc-ribbon motif from the ZBD of the
B. stearothermophilus (Bst) primase (Pan and Wigley,
2000) (Figure 3; rmsd of 1.1 A˚for 23 C? atoms of these
ZBDs). The zinc ribbon of the Bst primase has an addi-
tional ? strand at its N terminus, corresponding to the
disordered residues 1–9 of T7 primase. The ZBD inter-
acts with the DNA template during primer synthesis
and this interaction could organize the N-terminal resi-
dues to form the missing ? strand. The Bst ZBD also
has several additional ? helices appended to the N- and
C-terminal ends, creating a significantly larger structure
than the T7 ZBD (Figure 3). An alignment of the se-
quences of prokaryotic primases reveals gaps at the
ure 4), suggesting that the bacterial primases resemble
the larger Bst ZBD, whereas the smaller T7 ZBD is em-
blematic of phage-type primases.
T7 primase binds one atom of zinc (Mendelman et al.,
1994), and a bound zinc is evident in the ZBD of both of
the primase molecules, based on an anomalous Fourier
Results and Discussion
A Fragment of T7 Primase-Helicase
Primes DNA Synthesis
An N-terminal fragment (residues 1–272) of T7 primase-
helicase that encompasses all six conserved primase
motifs synthesizes oligoribonucleotides as efficiently as
the intact primase-helicase (Frick et al., 1998). However,
the resulting oligonucleotides are not extended by T7
DNA polymerase in the presence of this primase frag-
ment. Since the fragment, lacking a helicase domain,
binds to DNA weakly (Frick and Richardson, 1999), we
repeated this experiment with a higher concentration
of a fragment that was shorter (residues 1–255) and
amenable to crystallization. At high concentrations, the
fragment canprime DNA synthesis catalyzedby T7 DNA
polymerase, albeit less efficiently than the wild-type pri-
mase helicase (see Supplemental Data at http://www.
the helicase domain, the primase fragment must locate
priming sites by random collisions. Nonetheless, all of
the essential activities for primer synthesis and primer
utilization are present in the 1–255 primase fragment.
Structure and Dynamics of T7 Primase
Table 1. X-Ray Data and Refinement Statistics
X-Ray Intensity DataNativea
Figure of meritc(15-2.7 A˚)
Bond length (A˚)
Bond angles (?)
Torsion angles (?)
Overall mean B value (A˚2)
Number of atoms
Most favored (%)
Generously allowed (%)
aCollected at the station F1 of the Cornell High Energy Synchrotron Source (Ithaca, NY).
bCollected at the beamline X8C of the National Synchrotron Light Source (Upton, NY).
cThe numbers in brackets represent the values for the last resolution shells.
dRsym??hkl?i| I(hkl)I? ?I(hkl)? | / ?hkl?i?I(hkl)i?.
ea and c indicate acentric and centric reflections, respectively.
fano, based on anomalous diffraction differences.
gRfreewas calculated using 5% of the reflection data.
map calculated with X-ray data collected at the K ab-
sorption edge for zinc (? ? 1.284 A˚). The zinc atom is
coordinated in a tetrahedral geometry by two pairs of
cysteine residues located in the ?2/?3 loops and the
nation was reported for the Bst ZBD, except that a histi-
dine replaces one of the cysteines (Pan and Wigley,
2000). The zinc binding site of T7 primase is partially
exposed on the surface of the ZBD facing away from
tics of the surrounding residues suggest that the zinc
atom does not directly participate in interactions with
DNA or with nucleotides. The bound metal is instead
likely to support primer synthesis by stabilizing the pro-
tein fold of the ZBD. The surface of the ZBD away from
the zinc binding site and facing the RPD is likely to
interact with the DNA template. A hinge-like motion of
the linker could bring the ZBD and RPD together and
allow both domains to contact the template during RNA
synthesis. The proposed DNA binding surface of the
ZBD is decorated with a central patch of aromatic resi-
dues (F11, F29, Y37, Y35, and W42) that are surrounded
by a rim of polar and charged residues (H14, D24, D31,
H33, E40, and K41). Aspartate 31 and H33 are important
determinants of binding to the 5?-GTC-3? priming sites
(Kusakabe et al., 1999). In the crystal, two neighboring
ZBDs pack against one another and bury the aromatic
residues that we propose could stack against the bases
of a DNA template.
The RNA Polymerase Domain
The RPD is composed of two substructures, a C-ter-
minal TOPRIM fold that resembles the corresponding
region of the E. coli DnaG primase (Keck et al., 2000;
Podobnik et al., 2000) and a structurally divergent
N-terminal subdomain with a mixed ?/? fold (Figures 1
the residues of conserved motifs II–VI (Figure 4) are
clustered near the cleft. The overall shape of the RPD
is unlike the cashew-like silhouette of the DnaG RPD
(Figure 3A) because of the divergent N-terminal subdo-
extension that mediates interactions with the DnaB heli-
case. T7 primase lacks this structure and is covalently
joined tothe helicase in abifunctional primase-helicase.
The N-terminal subdomain of the T7 RPD consists of
an antiparallel, four-stranded ? sheet packed against
an ? helix and two short 310helices. Primase motifs II
and III are located in the connecting loops between a
pair of helices (?2/?2) and a pair of ? strands (?7/?8)
within this subdomain. These conserved residues face
the shallow cleft separating the N-terminal subdomain
from the TOPRIM fold (Figures 1A and 4). DnaG has
an analogous four-stranded ? sheet but is otherwise
3). An insertion between strands ?1 and ?4 of this ?
sheet significantly increases the overall size of DnaG’s
RPD (Figure 4). An ? helix located at the N terminus of
the DnaG RPD (helix ?1) replaces part of the extended
Figure 1. Crystal Structure of T7 Primase
(A) A stereo diagram of T7 primase is shown as a ribbon model with the secondary elements colored; ? helix (blue), ? strand (green), and 310
helix (magenta). Dotted lines denote disordered regions of the protein (residues 1–9 and 45–48 of the molecule shown here). A zinc atom
(yellow) and two magnesium atoms (pink) are bound to the primase.
(B) Two primase molecules that occupy the asymmetric unit of the crystals are related by noncrystallographic 2-fold symmetry, with the ZBD
of each molecule “swapped” onto the RPD of the neighboring molecule.
linker that connects the RPD of T7 primase to the ZBD
(Figure 3), possibly making this connection more rigid
Mutational studies of prokaryote primases have iden-
tified catalytically important residues (Lee and Richard-
son, 2001; Sun et al., 1999; Ziegelin et al., 1995) located
around the shallow cleft between the TOPRIM and
N-terminal subdomains of the RPD (Figures 1A and 4).
Five invariant acidic residues (E157, D161, D207, D209,
and D237) from primase motifs IV–VI cluster on the
TOPRIM side of the cleft, creating an acidic patch at
the center of the active site (Figures 2 and 5D). These
residues bind two metal ions in the crystal structure
(Figures 1A and 2). We presume these are Mg2?ions
because the electron density corresponding to the met-
als increases after soaking crystals in magnesium con-
centrations higher than 2.5 mM MgCl2. The side chains
of E157 and D207 coordinate one of the metals (Figure
2). The side chain of D209 also points toward the bound
The TOPRIM Fold and the Primase Active Site
The C-terminal TOPRIM fold of T7 primase consists of
ces and a 310helix (Figures 3 and 4). It contains the
residues from conserved motifs IV–VI, and it resembles
the TOPRIM fold of DnaG (Figure 3). A search of the
Protein Data Bank (Holm and Sander, 1993) revealed
to the TOPRIM folds from a variety of DNA topoisomer-
ases (Rmsd ? 3 A˚; see Supplemental Data at http://
despite the weakly conserved amino acid sequences.
Structure and Dynamics of T7 Primase
Figure 2. Metal Binding in the Active Site of T7 Primase
A stereo view of the primase active site is shown with the experimentally phased electron density superimposed, contoured at 1 ?. The
conserved acidic residues of primase motifs V and VI are labeled and they chelate two Mg2?ions (gold spheres). A nearby water molecule
(red sphere) is coordinated by one of the Mg2?ions.
metal, but the distance is too far for a direct contact
with the metal. These three acidic residues are perfectly
conserved in proteins containing the TOPRIM fold, and
their spatial arrangement resembles that of a conserved
triad of acidic residues within the active site of RNA
polymerases (Gnatt et al., 2001; Vassylyev et al., 2002).
This cluster of acidic residues coordinates a Mg2?ion
in a crystal structure of topoisomerase IV (Nichols et al.,
1999) and a Y2?ion in the structure of the RPD of DnaG
(Keck et al., 2000), although this metal binding site in
DnaG was not detected by Fe2?affinity cleavage (God-
son et al., 2000). A second Mg2?ion is coordinated by
D237 of T7 primase (Figure 2), and it corresponds to the
metal binding site of DnaG that was detected by Fe2?
affinity cleavage (Godson et al., 2000). However, this
tate 237 is a conserved residue of primase motif VI, but
it is not conserved in all TOPRIM folds. We are less
certain about the biological significance of the second
bound Mg2?, but biochemical evidence suggests that
DnaG binds at least two Mg2?ions (Urlacher and Griep,
1995). Prima facie, the two Mg2?ions are reminiscent
ofa twometal-ion mechanismof nucleotidepolymeriza-
tion (Steitz, 1998).
A group of basic residues in the N-terminal half of the
RPD faces the shallow cleft between the N-terminal and
TOPRIM subdomains of T7 primase. Lysine 122, K128,
K131, and K137 all contribute to RNA synthesis activity
(Lee and Richardson, 2001). They are in a region corre-
sponding to the ATP binding site of DnaG, which was
identified by crosslinking of ATP analogs to K241 (Mus-
taev and Godson, 1995; Sun et al., 1999), corresponding
to K137 of T7 primase. These data implicate the shallow
cleft as the binding site for nucleotide.
Interaction with DNA and Nucleotide Substrates
The T7 primase crystallized in an open conformation
with its ZBD and RPD some distance apart (Figure 1).
These domains must come together to engage nucleo-
in binding to DNA (Kusakabe et al., 1999; Kusakabe and
Richardson, 1996; Mendelman et al., 1994), especially
the residues D31 and H33, which recognize the cryptic
3? C of the priming site. The bacterial primases have an
analogous pair of charged residues in the ?3/4 loop
(E55 and K56 of the Bst ZBD) (Pan and Wigley, 2000),
suggestive of a similar mode of interaction with their
ble linker between the ZBD and RPD could bring D31
the DNA template for interactions with nucleotides (Fig-
ure 5D). The ZBDs of two primase molecules are inter-
twined so that the ZBD of each molecule packs against
the adjacent RPD in a domain swapping arrangement
(Figure 2B) that we believe has no biological signifi-
to prevent movement of the ZBD in response to sub-
Figure 3. Modular Architecture of T7 and DnaG Primases
(A) The structure of T7 primase has features resembling the ZBD of Bst primase (Pan and Wigley, 2000) and the RPD from E. coli DnaG (Keck
et al., 2000; Podobnik et al., 2000). The structurally conserved regions are colored as follows: the zinc ribbon motif (red), the N-terminal
subdomain of the RPD (yellow), and the C-terminal TOPRIM fold (blue). The bound metal ions in T7 primase are depicted as silver spheres.
Bst ZBD has additional ? helices (gray) flanking the conserved zinc ribbon motif. A unique subdomain at the C terminus of DnaG (gray)
supports its interactions with the DnaB helicase.
(B) A topological comparison of the primases. The subdomains are colored as (A), and the secondary structures are depicted as arrows for
? strands and cylinders for ? helices and 310helices. The broken lines indicate disordered regions of T7 primase. The dotted line between
Bst ZBD and the RPD of E. coli DnaG indicates the predicted connection between these domains.
strates. We therefore examined the interaction of T7
primer using nuclear magnetic resonance.
In the absence of ligands, the1H-15N HSQC spectrum
of T7 primase shows a number of well-resolved amide
N-H cross peaks characteristic of a folded protein along
Structure and Dynamics of T7 Primase
Figure 4. Structure-Based Sequence Alignment of T7 Primase with Bacterial Primases
Structures of the ZBD of the Bst primase (Pan and Wigley, 2000) and the RPD of E. coli DnaG (Keck et al., 2000) were superimposed on T7
primase, and the structurally aligned amino acids are shown here. The secondary structures of these proteins are indicated above and below
each sequence as cylinders (? helices), arrows (? strands), or cylinders labeled “?” (310helices). The wavy lines show residues that are
disordered in the crystal structures. The conserved primase motifs defined by Ilyina et al. (1992; colored gray) include five highly conserved
acidic residues (red) and two arginines (blue). The black arrow indicates the starting residue of a truncated form of T7 primase-helicase that
is expressed with the full-length protein during phage growth.
with unresolved peaks at the crowded center region of
the spectrum (Figure 5A). A closer inspection of the
HSQC spectrum reveals a distinct class of well-dis-
persed proton resonances with higher-than-average in-
tensities. These stronger peaks probably correspond to
residues from the smaller ZBD, moving independently
and with a shorter correlation time than the larger RPD
and thus generating narrower and taller signals. The1H-
and it confirms that the well-dispersed peaks arise from
amide protons of residues within the ZBD (compare
Figures 5A and 5C). The conservation of the peak posi-
tions of the ZBD in full-length primase indicates that the
ZBD doesn’t tightly associate with the RPD. Additional
relaxation experiments.The T1andT2relaxationtimes of
residues from the ZBD and RPD of the primase fall into
two distinctlydifferent groups thatindicate thatthe ZBD
exhibits significantly faster rotatory diffusive motion
than the RPD (see Supplemental Data at http://www.
molecule.org/cgi/content/full/11/5/1349/DC1). We con-
clude that in the absence of substrates the ZBD and
RPD move independently and there is little interaction
between these domains.
Upon addition of a DNA template and a complemen-
tary oligoribonucleotide primer (5?-ACCC-3?), the1H-15N
HSQC spectrum of T7 primase changes dramatically
with the well-dispersed resonances from the ZBD shift-
peaks in Figures 5A and 5B). A number of unassigned
peaks also change in response to substrate binding,
including many residues that are most likely from the
RPD (peaks not circled in Figures 5A and 5B). The ob-
served spectral changes arising from residues in both
the ZBD and RPD indicate that both domains interact
with the primed DNA template, consistent with the
model shown in Figure 5D. However, the spectrum of
Figure 5B has less than the expected number of peaks,
indicating that part of the system is not in a unique state
but fluctuates between different conformations leading
to broadening of peaks. The disappearance of signals
could also be due to aggregation. However, a large frac-
tion of the signals originating from the RPD remains
unchanged (Figure 5B versus 5A), indicating there is
no major problem with aggregation of the system. In
addition, Figures 5A and 5B show a region of poorly
structured signals in the center of the spectrum. This is
often observed in proteins that are partially aggregated
or contain unstructured segments. Because this feature
of the spectrum is similar before and after the addition
of the primed template, we conclude that addition of
substrate does not cause additional aggregation of the
primase. In contrast to the spectral changes seen in
Figure 5B, the HSQC spectrum of the primase does
not change upon addition of the DNA template alone.
However, the subsequent addition of ATP and CTP to
activate RNA synthesis triggers the same spectral
changes as those described above. The involvement of
both the ZBD and RPD in binding to a primed DNA
template could be explained either by closure of the
primase in cis (Figure 5D) or by the interactions of the
ZBD and RPD from two different primase molecules in
trans. Althoughthe NMRexperiments donot distinguish
between these possibilities, trans interactions between
neighboring primase domains of the ring-shaped T7 pri-
mase-helicase appearto be unfavored(E. Tothand T.E.,
We propose that the ZBD secures the DNA template
in cis within the active site of the RPD, allowing nucleo-
tides to base pair with the template and contact the
metals ligated by residues of primase motifs V and VI
(Figure 5D). We have modeled the interaction of the
Figure 5. The Conformation of T7 Primase Changes Upon Binding to a Primed DNA Template
(A) A two-dimensional1H-15N HSQC spectrum of T7 primase is shown. Many of the well-dispersed peaks are from residues within the ZBD
(circled; cf. [C]).
(B) The HSQC spectrum of the primase changes dramatically upon binding to a DNA template annealed with the tetraribonucleotide ACCC.
The well-dispersed peaks arising from the ZBD (circled, compare with [A]) disappear in complex with the primer-template pair. Many of the
unassigned peaks (uncircled) arising from the RPD also change in the complex, indicating that both domains engage the primed template.
(C) The two-dimensional1H-15N HSQC spectrum of the ZBD alone (residues 1–59) shows the locations of peaks arising from the residues of
(D) The NMR data of (A) through (C) can be explained by a substrate-induced change in the conformation of T7 primase from that seen in
the crystal structure of the unliganded primase (see Experimental Procedures for the modeling exercise). In the closed conformation modeled
here, the primase engages the primer 3? end (blue strand) in the RPD active site while the ZBD contacts the priming site of the template (red
strand). The solvent accessible surface of the primase is shown, colored according to electrostatic potential (blue ??12 kT; red ??12 kT;
calculated with GRASP [Nicholls et al., 1991]).
ZBD domain with the bases of the primase recognition
sequence 5?-GTC-3? to include contacts with D31 and
H33, important determinants of DNA binding specificity
(Kusakabe et al., 1999; Kusakabe and Richardson,
1996). This model places the most highly conserved and
and/or the bound nucleotides, and the 3? end of the
RNA next to the metal binding sites of the active center.
The model also suggests several mechanisms for lim-
iting the length of oligoribonucleotides synthesized by
the primase. If the ZBD remains bound to the priming
of the DNA template might be restricted by the length
of the ZBD tether. Alternatively, the ZBD might remain
during RNA synthesis, causing a progressive loss of
sequence-specific DNA interactions that limits product
extension. The data presented below suggest that the
Structure and Dynamics of T7 Primase
Figure 6. The ZBD of T7 Primase Stimulates Primer Utilization by T7 DNA Polymerase
(A) The efficient extension of oligoribonucleotides by T7 DNA polymerase requires the addition of T7 primase (see Experimental Procedures).
Lane 1, no proteins added; lane 2, T7 DNA polymerase (1 ?M) only; lanes 3–5, addition of T7 DNA polymerase (1 ?M) plus T7 primase at the
indicated concentrations (?M). The fully extended primer strand (black arrow) and one of the abortive products (white arrow) are indicated.
(B) Similar reactions were carried out with the RPD or ZBD of the primase added separately. The RPD has no effect on primer utilization,
whereas the ZBD stimulates primer extension by T7 DNA polymerase to the same extent as the intact primase. Lanes 1 and 2, T7 DNA
polymerase (1 ?M) plus the RPD at indicated concentrations (?M); lanes 3–10, the DNA polymerase (1 ?M) plus the ZBD at the indicated
(C) The cryptic 3? cytidine of the priming site (underlined) is required for stimulation of DNA synthesis by the primase. Lanes 1–5, reactions
templated by the correct 3?-CTGGG-5? priming site using the combinations of proteins indicated at the top of the figure. Lanes 6–10, similar
reactions using the modified priming site, 3?-ATGGG-5?, which anneals to the pACCC primer but lacks the cryptic 3? C. On the modified
template, T7 primase (lane 8) and the ZBD (lane 10) are unable to stimulate primer extension.
ZBD remains bound to DNA during RNA synthesis and
in the subsequent step when the primed DNA template
is passed to T7 DNA polymerase.
beit less efficiently than the intact primase. Thus, the
ZBD is mainly responsible for enhancing primer utiliza-
tionand theRPD isnot requiredduring primerextension
by T7 DNA polymerase. The cryptic 3? C of the 5?-GTC-
3? recognition sequence is required not only for RNA
synthesis (Frick and Richardson, 1999) but also for the
utilization of the primer by DNA polymerase (Figure 6C).
Since the ZBD specifically interacts with the 3? C of
the priming site (Kusakabe and Richardson, 1996), this
result implies that the ZBD remains bound to the primed
DNA template as DNA polymerase engages the primer.
DNA template (Figure 5B) is suggestive of a conforma-
tional equilibrium—the peaks are significantly broad-
ened when the primase is complexed to RNA and DNA
(compare Figures 5A and 5B). We suggest this equilib-
rium is populated by the closed primase complex de-
picted in Figure 5D and an open conformation in which
the ZBD remains bound to the primed template. We
conclude that the ZBD remains bound to the primed
DNA template while RPD is released, exposing the
primed DNA template and ZBD of the primase for inter-
sistent with modeling studies showing that the RPD is
too large to fit into the DNA binding groove of T7 DNA
dated in complex with a nascent RNA primer that is
being extended by the polymerase. A unique loop (resi-
dues 401–404) located at the base of the thumb of T7
DNA polymerase is also required for primer utilization
(Chowdhury et al., 2000), and this loop might interact
with the ZBD at the onset of primer extension.
The ZBD Stimulates Primer Utilization
by T7 DNA Polymerase
In addition to synthesizing primers, T7 primase also en-
hances the efficiency with which the primers are ex-
tended byT7 DNA polymerase.Given the shortlength of
the primer,we might expectthe primase tobe intimately
associated with T7 DNA polymerase to promote exten-
sion of the primer by the polymerase (Chowdhury et al.,
2000; Kato et al., 2001). However, our attempts to create
a docking model of the primase bound to DNA polymer-
ase were frustrated by the large size of the RPD in
comparison to the narrow DNA binding groove of the
polymerase. We therefore considered the possibility
that the ZBD alone could deliver the primed DNA tem-
plate to T7 DNA polymerase.
T7 DNA polymerase initiates DNA synthesis efficiently
only when the primase is present (Figure 6A). The stimu-
lation of primer utilization by T7 DNA polymerase is
maximal at micromolar concentrations of T7 primase
(see Supplemental Data at http://www.molecule.org/
lower DNA binding affinity compared to the full-length
primase-helicase (Kato et al., 2001; Kusakabe et al.,
1998). The RPD alone (50 ?M) does not stimulate the
extension of primers by T7 DNA polymerase, whereas
the ZBD does support primer extension (Figure 6B), al-
M sodium formate, and 5.5 mM DTT. Crystals grew at 22?C over the
course of 1 week. The crystals belong to space group P3121 with
unit cell dimensions of a ? 138.6 A˚, b ? 138.6 A˚, and c ? 85.4 A˚.
There were two molecules of the primase in the asymmetric unit
with a solvent content of approximately 67%. The selenomethionine
protein crystallized under the same conditions as the wild-type
The structure was determined by a multiwavelength anomalous
diffraction experiment using crystals of the selenomethionine-sub-
stituted protein (Table 1). The crystals diffract X-rays to a resolution
of 2.3 A˚ in some directions, but the crystallographic model was
refined against the isotropic X-ray data extending to a resolution
limit of 2.9 A˚. The crystals were transferred in the harvest solutions
containing 50 mM MES (pH 6.3), 4 M sodium formate, 5 mM DTT,
and 5 mM MgCl2and incubated for 1 hr at 22?C. The concentration
tration of 20 mM. The crystals were then transferred in a cryoprotec-
tant solution containing 50 mM MES (pH 6.3), 4 M sodium formate,
20 mM MgCl2, 3 mM DTT, and 10% glycerol for about 1 s and then
flash-frozen in liquid nitrogen. Native X-ray diffraction data were
collected at station F1 of the Cornell High Energy Synchrotron
Source (Ithaca, NY). X-ray data were collected at three different
wavelengths at beamline X-8C of the National Synchrotron Light
tory X-ray source (Cu K? radiation). The data were processed with
DENZO/SCALEPACK (Otwinowski and Minor, 1997). Out of 14 po-
tential selenium sites in two protein molecules, 11 were located by
SOLVE (Terwilliger and Berendzen, 1999), and one more selenium
site and two zinc sites were found by difference Fourier methods
using the experimental phases. Two remaining selenium sites are
located in the disordered N-terminal segment of the primase. Heavy
atom refinement was performed with MLPHARE (Bailey, 1994), fol-
lowed by density modification using DM (Cowtan and Main, 1996).
The resulting electron density was of sufficient quality to model
nearly all the primase residues by using O (Jones et al., 1991). Model
refinement was carried out with CNS (Brunger et al., 1998), followed
by the additional refinement cycles with REFMAC5 (Murshudov et
al., 1997). The figures of the crystallographic model were prepared
using MOLSCRIPT (Kraulis, 1991), Ribbons (Carson, 1997), and
GRASP (Nicholls et al., 1991). The coordinates of T7 primase have
been deposited in the Protein Data Bank (accession code 1NUI).
The crystal structure and NMR studies of T7 primase
reveal a flexible linkage between the ZBD and RPD that
allows the ZBD to perform two essential functions. The
ZBD recognizes the temple sequence for primer synthe-
sis and assists this reaction by securing the template
into the active site of the primase. After the primer is
synthesized, the ZBD materially participates in primer
ture of the primase allows the ZBD to freely dissociate
from the RPD to deliver the primed template to the DNA
polymerase. The related DnaG primase from E. coli
transfers its primers to the processivity clamp of DNA
polymerase instead (Yuzhakov et al., 1999). Given their
functional and structural similarities to T7 primase, it
seems likely that other prokaryotic DNA primases could
similarly chaperone primed templates to DNA polymer-
ase or its associated factors.
Expression and Purification of T7 Primase Proteins
The coding sequence for the T7 primase fragment (residues 1–255)
was amplified by polymerase chain reaction and cloned into the
expression plasmid pET17b (Novagen). A bacterial expression plas-
Proteins were expressed in E. coli BL21(DE3) cells grown in Luria-
Bertani medium at 37?C until the culture reached an OD600? 1.5.
The culture was then chilled to 25?C, and protein expression was
induced for an additional 18 hr. The primase was purified as de-
scribed (Frick et al., 1998), except that the HiTrap blue affinity col-
umn was substituted with heparin sepharose. The purified primase
was concentrated to 50 mg ml?1by ultrafiltration (Centricon 10;
Amicon Inc) and then diluted with an equal volume of glycerol and
stored at ?20?C. The selenomethionine-substituted primase was
overexpressed in M9 minimal media containing 60 mg L?1seleno-
methionine and 10 ?M ZnSO4as described (Van Duyne et al., 1993)
and purified as described above. Perdeuterated-15N-labeled pro-
teins were expressed in M9 minimal media prepared with
15N-(NH4)2SO4and 10 ?M ZnSO4in D2O. The purification of the ZBD
was carried out as described above except that the heparin sepha-
rose chromatography step was omitted. The purified RPD was pro-
vided by Luis Brieba (Harvard Medical School).
Two dimensional1H-15N HSQC spectra were collected with a Varian
INOVA (500 MHz) spectrometer for samples containing 75 ?M per-
deuterated-15N-labeled T7 primase in 200 mM HEPES (pH 7.5), 100
mM NaCl, 20 mM DTT, 1 mM EDTA, 0.1 mM PMSF, and 20 mM
MgCl2in 90% H2O/10% D2O at 25?C. The addition of 0.2 mM DNA
template (5?-GGGTCAA-3?) and each 1 mM ATP and CTP to the
above solution eliminated or changed the positions of many peaks
in the 2D HSQC spectrum, suggesting that a complex with the
tides with a preformed RNA primer (0.4 mM each of 5?-ACCC-3?
and DNA template). The spectra of the15N-labeled ZBD were mea-
sured using 0.5 mM protein in the same buffer described above,
but without the template and primer. All spectra were processed
using NMRPipe (Delaglio et al., 1995) and analyzed with NMRView
(Johnson and Blevins, 1994).15N T1and T2values of T7 primase were
the delay T: for T1, T ? 10, 40 120, 240, 480, 720, 960, and 1200 ms;
for T2, T ? 10, 30, 50, 70, 90, 110, 130, and 150 ms (Farrow et
al., 1994). Analyses of T1and T2values were carried out with the
automated function of NMRView. Well-dispersed peaks corre-
sponding to the ZBD and RPD were randomly picked, and the aver-
aged values were calculated for the ZBD and RPD separately (see
Supplemental Data at http://www.molecule.org/cgi/content/full/11/
RNA-Primed DNA Synthesis Assay
The RNA-primed DNA synthesis assay was described previously
(Kato et al., 2001), with the following modifications. The reaction
mixture (10 ?l) contained 40 mM Tris-HCl (pH 7.5), 10 mM MgCl2,
10 mM DTT, 100 ?g ml?1BSA, 50 mM potassium glutamate, 25 ng
?l?1M13 ssDNA, 1 mM each of ATP and CTP, 0.3 mM each of dATP,
dCTP, and dGTP, 0.3 mM [3H]-dTTP, 100 nM T7 DNA polymerase,
and the indicated concentrations of gp4 protein, T7 primase, or a
monomeric mutant of the gene 4 protein, gp4?D2D3 (Kato et al.,
2001). The reactions were incubated for 30 min at 37?C, and DNA
synthesis was measured using a previously described filter binding
assay (Lee et al., 1998). DNA synthesis assays using a synthetic
ribotetranucleotide (ACCC) were carried out in reactions containing
100 ?M 26-mer DNA template (5?-CAGTGACGGGTCGTTTATCGTC
GGCA-3?) and 100 ?M [32P]-ACCC instead of M13 ssDNA, ATP, and
CTP, and the indicated concentrations of T7 primase, the ZBD, or
were quenched with an equal volume 98% formamide containing
bromphenol blue, followed by separation of the products on a 25%
acrylamide gel containing 3M urea and autoradiography.
X-Ray Structure Determination of T7 Primase
T7 primase was crystallized by the hanging drop vapor diffusion
method. The protein drop initially contained 8 mg ml?1primase in
50 mM MES (pH 6.3), 2 M sodium formate, 5.5 mM DTT, and 2.5
mM ATP. The reservoir solution contained 100 mM MES (pH 6.3), 4
Modeling the Primase-DNA Complex
The structure of the RNA polymerase II complexed with DNA:RNA
hybrid strands (PDB code 1T6H) (Gnatt et al., 2001) was superim-
posed on the crystallographic model of T7 primase. The highly con-
Structure and Dynamics of T7 Primase
served residues E157, D207, and D209 of T7 primase were aligned
with the corresponding residues D485, D481, and D483 of RNA
polymerase II (chain A; rmsd ? 1.9 A˚for these side chains). The
superimposed DNA:RNA heteroduplex from RNA Pol II was dis-
played on the structure of T7 primase, and the ZBD was then manu-
ally adjusted to interact with the priming site of the DNA template
so that D31 and H33 close to 3? C of the 5?-GTC-3? priming site.
loop in the DNA-binding crevice of bacteriophage T7 DNA polymer-
ase influences primer utilization. Proc. Natl. Acad. Sci. USA 97,
Cowtan, K., and Main, P. (1996). Phase combination and cross vali-
dation in iterated density-modification calculation. Acta Crystallogr.
Delaglio, F., Grzesiek, S., Vuister, G.W., Zhu, G., Pfeifer, J., and Bax,
A. (1995). NMRPipe: a multidimensional spectral processing system
based on UNIX pipes. J. Biomol. NMR 6, 277–293.
Doublie, S., and Ellenberger, T. (1998). The mechanism of action of
T7 DNA polymerase. Curr. Opin. Struct. Biol. 8, 704–712.
Dunn, J.J., and Studier, F.W. (1983). Complete nucleotide sequence
of bacteriophage T7 DNA and the locations of T7 genetic elements.
J. Mol. Biol. 166, 477–535.
Egelman, H.H., Yu, X., Wild, R., Hingorani, M.M., and 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.
Farrow, N.A., Muhandiram, R., Singer, A.U., Pascal, S.M., Kay, C.M.,
Gish, G., Shoelson, S.E., Pawson, T., Forman-Kay, J.D., and Kay,
L.E. (1994). Backbone dynamics of a free and phosphopeptide-
complexed Src homology 2 domain studied by 15N NMR relaxation.
Biochemistry 33, 5984–6003.
Frick, D.N., and Richardson, C.C. (1999). Interaction of bacterio-
phage T7 gene 4 primase with its template recognition site. J. Biol.
Chem. 274, 35889–35898.
Frick, D.N., and Richardson, C.C. (2001). DNA primases. Annu. Rev.
Biochem. 70, 39–80.
Frick, D.N., Baradaran, K., and Richardson, C.C. (1998). An N-ter-
minal 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.
Fujiyama, A., Kohara, Y., and Okazaki, T. (1981). Initiation sites for
discontinuousDNAsynthesis ofbacteriophageT7.Proc. Natl.Acad.
Sci. USA 78, 903–907.
Gnatt, A.L., Cramer, P., Fu, J., Bushnell, D.A., and Kornberg, R.D.
(2001). Structural basis of transcription: an RNA polymerase II elon-
gation complex at 3.3 A˚resolution. Science 292, 1876–1882.
Godson, G.N., Schoenich, J., Sun, W., and Mustaev, A.A. (2000).
Identification of the magnesium ion binding site in the catalytic
center of Escherichia coli primase by iron cleavage. Biochemistry
Griep, M.A. (1995). Primase structure and function. Indian J. Bio-
chem. Biophys. 32, 171–178.
Guo, S., Tabor, S., and Richardson, C.C. (1999). The linker region
between the helicase and primase domains of the bacteriophage
T7 gene 4 protein is critical for hexamer formation. J. Biol. Chem.
Holm, L., and Sander, C. (1993). Protein structure comparison by
alignment of distance matrices. J. Mol. Biol. 233, 123–138.
Ilyina, T.V., Gorbalenya, A.E., and Koonin, E.V. (1992). Organization
and evolution of bacterial and bacteriophage primase-helicase sys-
tems. J. Mol. Evol. 34, 351–357.
Johnson, B.A., and Blevins, R.A. (1994). NMRView: a computer pro-
gram for the visualization and analysis of NMR data. J. Biomol. NMR
Jones, T.A., Zou, J.Y., Cowan, S.W., and Kjeldgaard (1991). Im-
proved methods for building protein models in electron density
maps and the location of errors in these models. Acta Crystallogr.
A 47 (Pt 2), 110–119.
Kato, M., Frick, D.N., Lee, J., Tabor, S., Richardson, C.C., and Ellen-
berger, T. (2001). A complex of the bacteriophage T7 primase-heli-
case and DNA polymerase directs primer utilization. J. Biol. Chem.
Keck, J.L., and Berger, J.M. (2001). Primus inter pares (first among
equals). Nat. Struct. Biol. 8, 2–4.
Keck, J.L., Roche, D.D., Lynch, A.S., and Berger, J.M. (2000). Struc-
ture of the RNA polymerase domain of E. coli primase. Science 287,
We thank Eric Toth for assistance with X-ray data collection and
discussions, Luis Brieba for providing the purified RPD, JoonSoo
Lee, Seung-Joo Lee, and David Frick for advice on primase activity
assays, and Patrick O’Brien and Ying Li for critically reading the
manuscript. We also thank the beamline staff of the National Syn-
chrotron Light Source (Upton, NY) and the Cornell High Energy
Synchrotron Source (Ithaca, NY) for assistance with X-ray data col-
lection. This work was supported by grants from the National Insti-
tutes of Health (R01 GM55390 to T.E. and R01 GM54397 to C.C.R.)
and the National Science Foundation (NSF MCB9816072 to G.W.)
and by the resources of the Harvard/Armenise Structural Biology
Center at Harvard Medical School. M.K. was supported by a fellow-
ship from the Uehara Memorial Foundation. T.E.E. gratefully ac-
knowledges support from the Hsien Wu and Daisy Yen Wu Profes-
sorship at Harvard Medical School.
Received: January 23, 2003
Revised: April 15, 2003
Accepted: May 8, 2003
Published: May 22, 2003
Aravind, L., Leipe, D.D., and Koonin, E.V. (1998). Toprim—a con-
served catalytic domain in type IA and II topoisomerases, DnaG-
type primases, OLD family nucleases and RecR proteins. Nucleic
Acids Res. 26, 4205–4213.
Augustin, M.A., Huber, R., and Kaiser, J.T. (2001). Crystal structure
of a DNA-dependent RNA polymerase (DNA primase). Nat. Struct.
Biol. 8, 57–61.
Bailey, S. (1994). The CCP4 suite-programs for protein crystallogra-
phy. Acta Crystallogr. D50, 760–763.
Benkovic, S.J., Valentine, A.M., and Salinas, F. (2001). Replisome-
mediated DNA replication. Annu. Rev. Biochem. 70, 181–208.
Berger, J.M., Gamblin, S.J., Harrison, S.C., and Wang, J.C. (1996).
Structure and mechanism of DNA topoisomerase II. Nature 379,
Bernstein, J.A., and Richardson, C.C. (1988). A 7-kDa region of the
bacteriophage T7 gene 4 protein is required for primase but not for
helicase activity. Proc. Natl. Acad. Sci. USA 85, 396–400.
Bird, L.E., Hakansson, K., Pan, H., and Wigley, D.B. (1997). Charac-
terization and crystallization of the helicase domain of bacterio-
phage T7 gene 4 protein. Nucleic Acids Res. 25, 2620–2626.
Bird, L.E., Pan, H., Soultanas, P., and Wigley, D.B. (2000). Mapping
protein-protein interactions within a stable complex of DNA primase
and DnaB helicase from Bacillus stearothermophilus. Biochemistry
Brautigam, C.A., and Steitz, T.A. (1998). Structural and functional
insights provided by crystal structures of DNA polymerases and
their substrate complexes. Curr. Opin. Struct. Biol. 8, 54–63.
Brunger, A.T., Adams, P.D., Clore, G.M., DeLano, W.L., Gros, P.,
Grosse-Kunstleve, R.W., Jiang, J.S., Kuszewski, J., Nilges, M.,
Pannu, N.S., et al. (1998). Crystallography & NMR system: a new
software suite for macromolecular structure determination. Acta
Crystallogr. D Biol. Crystallogr. 54, 905–921.
Carson, M.(1997). Ribbons. InMethods in Enzymology,C.W. Carter,
Jr., and R.M. Sweet, eds. (Orlando: Academic Press), pp. 493–505.
Chang, P., and Marians, K.J. (2000). Identification of a region of
Escherichia coli DnaB required for functional interaction with DnaG
at the replication fork. J. Biol. Chem. 275, 26187–26195.
Chowdhury, K., Tabor, S., and Richardson, C.C. (2000). A unique
Kraulis, P.J. (1991). MOLSCRIPT: a program to produce both de-
tailed andschematic plots ofprotein structures. J.Appl. Crystallogr.
Kusakabe, T., and Richardson, C.C. (1996). The role of the zinc
motif in sequence recognition by DNA primases. J. Biol. Chem. 271,
Kusakabe, T., and Richardson, C.C. (1997). Gene 4 DNA primase of
bacteriophage T7 mediates the annealing and extension of ribo-
oligonucleotides at primase recognition sites. J. Biol. Chem. 272,
Kusakabe, T., Baradaran, K., Lee, J., and Richardson, C.C. (1998).
Roles of the helicase and primase domain of the gene 4 protein of
bacteriophage T7 in accessing the primase recognition site. EMBO
J. 17, 1542–1552.
Kusakabe, T.,Hine, A.V.,Hyberts, S.G.,and Richardson,C.C. (1999).
The Cys4 zinc finger of bacteriophage T7 primase in sequence-
Lee, J., Chastain, P.D., Kusakabe, T., Griffith, J.D., and Richardson,
C.C. (1998). Coordinated leading and lagging strand DNA synthesis
on a minicircular template. Mol. Cell 1, 1001–1010.
Lee, S.J., and Richardson, C.C. (2001). Essential lysine residues in
the RNA polymerase domain of the gene 4 primase-helicase of
bacteriophage T7. J. Biol. Chem. 276, 49419–49426.
Lee, S.J., and Richardson, C.C. (2002). Interaction of adjacent pri-
mase domains within the hexameric gene 4 helicase-primase of
bacteriophage T7. Proc. Natl. Acad. Sci. USA 99, 12703–12708.
Lima, C.D., Wang, J.C., and Mondragon, A. (1994). Three-dimen-
sional structure of the 67K N-terminal fragment of E. coli DNA topo-
isomerase I. Nature 367, 138–146.
Marintcheva, B., and Weller, S.K. (2001). A tale of two HSV-1 heli-
cases: roles of phage and animal virus helicases in DNA replication
and recombination. Prog. Nucleic Acid Res. Mol. Biol. 70, 77–118.
Matson, S.W., Tabor, S., and Richardson, C.C. (1983). The gene 4
protein of bacteriophage T7. Characterization of helicase activity.
J. Biol. Chem. 258, 14017–14024.
Mendelman, L.V., and Richardson, C.C. (1991). Requirements for
primer synthesis by bacteriophage T7 63-kDa gene 4 protein. Roles
of template sequence and T7 56-kDa gene 4 protein. J. Biol. Chem.
Mendelman, L.V., Beauchamp, B.B., and Richardson, C.C. (1994).
Requirement for a zinc motif for template recognition by the bacte-
riophage T7 primase. EMBO J. 13, 3909–3916.
Mondragon, A., and DiGate, R. (1999). The structure of Escherichia
coli DNA topoisomerase III. Struct. Fold. Des. 7, 1373–1383.
Murshudov, G.N., Vagin, A.A., and Dodson, E.J. (1997). Refinement
of macromolecular structures by the maximum-likelihood method.
Acta Crystallogr. D53, 240–255.
Mustaev, A.A., and Godson, G.N. (1995). Studies of the functional
topography of the catalytic center of Escherichia coli primase. J.
Biol. Chem. 270, 15711–15718.
Nicholls, A., Sharp, K.A., and Honig, B. (1991). Protein folding and
association: insights from the interfacial and thermodynamic prop-
erties of hydrocarbons. Proteins 11, 281–296.
Nichols, M.D., DeAngelis, K., Keck, J.L., and Berger, J.M. (1999).
Structure and function of an archaeal topoisomerase VI subunit with
homology to the meiotic recombination factor Spo11. EMBO J. 18,
Otwinowski,Z., andMinor, W.(1997). Processingof X-raydiffraction
data collected in oscillation mode. In Methods in Enzymology, C.W.
Carter, Jr., and R.M. Sweet, eds. (Orlando: Academic Press), pp.
Pan, H., and Wigley, D.B. (2000). Structure of the zinc-binding do-
Large scale purification and biochemical characterization of T7 pri-
mase/helicase proteins. Evidence for homodimer and heterodimer
formation. J. Biol. Chem. 267, 15013–15021.
Podobnik, M., McInerney, P., O’Donnell, M., and Kuriyan, J. (2000).
A TOPRIM domain in the crystal structure of the catalytic core of
erases. J. Mol. Biol. 300, 353–362.
Powers, L., and Griep, M.A. (1999). Escherichia coli primase zinc is
sensitive to substrate and cofactor binding. Biochemistry 38, 7413–
Qian, X., Jeon, C., Yoon, H., Agarwal, K., and Weiss, M.A. (1993).
Structure of a new nucleic-acid-binding motif in eukaryotic tran-
scriptional elongation factor TFIIS. Nature 365, 277–279.
Richardson, C.C. (1983). Bacteriophage T7: minimal requirements
for the replication of a duplex DNA molecule. Cell 33, 315–317.
Rosenberg, A.H., Griffin, K., Washington, M.T., Patel, S.S., and
Studier, F.W. (1996). Selection, identification, and genetic analysis
of random mutants in the cloned primase/helicase gene of bacterio-
phage T7. J. Biol. Chem. 271, 26819–26824.
Sawaya,M.R.,Guo, S.,Tabor,S.,Richardson, C.C.,andEllenberger,
T. (1999). Crystal structure of the helicase domain from the replica-
tive helicase-primase of bacteriophage T7. Cell 99, 167–177.
Singleton, M.R., Sawaya, M.R., Ellenberger, T., and Wigley, D.B.
(2000).Crystal structureof T7gene 4ring helicaseindicates amech-
anism for sequential hydrolysis of nucleotides. Cell 101, 589–600.
Steitz, T.A. (1998). A mechanism for all polymerases. Nature 391,
Sun, W., Schoneich, J., and Godson, G.N. (1999). A mutant Esche-
richia coli primase defective in elongation of primer RNA chains. J.
Bacteriol. 181, 3761–3767.
Tabor, S., and Richardson, C.C. (1981). Template recognition se-
T7. Proc. Natl. Acad. Sci. USA 78, 205–209.
Terwilliger,T.C.,and Berendzen,J.(1999).Automated MADandMIR
structure solution. Acta Crystallogr. D Biol. Crystallogr. 55, 849–861.
Urlacher, T.M., and Griep, M.A. (1995). Magnesium acetate induces
a conformational change in Escherichia coli primase. Biochemistry
Valentine, A.M., Ishmael, F.T., Shier, V.K., and Benkovic, S.J. (2001).
A zinc ribbon protein in DNA replication: primer synthesis and mac-
romolecular interactions by the bacteriophage T4 primase. Bio-
chemistry 40, 15074–15085.
Van Duyne, G.D., Standaert, R.F., Karplus, P.A., Schreiber, S.L.,
and Clardy, J. (1993). Atomic structures of the human immunophilin
FKBP-12 complexes with FK506 and rapamycin. J. Mol. Biol. 229,
VanLoock, M.S., Chen, Y.J., Yu, X., Patel, S.S., and Egelman, E.H.
(2001). The primase active site is on the outside of the hexameric
bacteriophage T7 gene 4 helicase-primase ring. J. Mol. Biol. 311,
Vassylyev, D.G., Sekine, S., Laptenko, O., Lee, J., Vassylyeva, M.N.,
Borukhov, S., and Yokoyama, S. (2002). Crystal structure of a bacte-
rial RNA polymerase holoenzyme at 2.6 A˚resolution. Nature 417,
in the trombone orchestra. Science 287, 2435–2436.
Yuzhakov, A., Kelman, Z., and O’Donnell, M. (1999). Trading places
on DNA–a three-point switch underlies primer handoff from primase
to the replicative DNA polymerase. Cell 96, 153–163.
Ziegelin, G., Linderoth, N.A., Calendar, R., and Lanka, E. (1995).
Domain structure of phage P4 alpha protein deduced by mutational
analysis. J. Bacteriol. 177, 4333–4341.
The model coordinates and structure factors for the T7 primase are