Download Crystal Structure of the DNA Polymerase Processivity Factor of T4

doi:10.1006/jmbi.1999.3511 available online at http://www.idealibrary.com on
J. Mol. Biol. (2000) 296, 1215±1223
Crystal Structure of the DNA Polymerase Processivity
Factor of T4 Bacteriophage
Ismail Moarefi, David Jeruzalmi, Jennifer Turner, Mike O'Donnell
and John Kuriyan*
Howard Hughes Medical
Institute, The Rockefeller
University, 1230 York Avenue
New York, NY, 10021, USA
The protein encoded by gene 45 of T4 bacteriophage (gene 45 protein or
gp45), is responsible for tethering the catalytic subunit of T4 DNA Polymerase to DNA during high-speed replication. Also referred to as a sliding DNA clamp, gp45 is similar in its function to the processivity factors
of bacterial and eukaryotic DNA polymerases, the b-clamp and PCNA,
respectively. Crystallographic analysis has shown that the b-clamp and
PCNA form highly symmetrical ring-shaped structures through which
duplex DNA can be threaded. Gp45 shares no sequence similarity with
b-clamp or PCNA, and sequence comparisons have not been able to
establish whether it adopts a similar structure. We have determined the
Ê resolution, using
crystal structure of gp45 from T4 bacteriophage at 2.4 A
multiple isomorphous replacement. The protein forms a trimeric ringshaped assembly with overall dimensions that are similar to those of the
bacterial and eukaryotic processivity factors. Each monomer of gp45 contains two domains that are very similar in chain fold to those of b-clamp
and PCNA. Despite an overall negative charge, the inner surface of the
ring is in a region of positive electrostatic potential, consistent with a
mechanism in which DNA is threaded through the ring.
# 2000 Academic Press
*Corresponding author
Introduction
Highly processive DNA polymerases in bacteria
and eukaryotes achieve remarkable speed in DNA
replication by tethering their catalytic subunits to
ring-shaped proteins (e.g. see Huang et al., 1981;
Kuriyan & O'Donnell, 1993; Stillman, 1994;
Stukenberg et al., 1991). These proteins, known as
sliding clamps, encircle DNA and allow the polymerases to move rapidly without detachment from
the template. This principle was ®rst established
for the b-clamp of Escherichia coli DNA polymerase
III complex, which allows the polymerase to move
for thousands of nucleotides along the template
without dissociation (reviewed by Kelman &
O'Donnell, 1995).
The crystal structure of the b-clamp showed that
two molecules of b form a ring-shaped dimer that
can girdle duplex DNA without steric hindrance
(Kong et al., 1992). Although the overall charge of
the protein is negative, the distribution of charged
residues within the protein is such that the space
E-mail address of the corresponding author:
[email protected]
0022-2836/00/051215±9 $35.00/0
inside the ring is a region of positive electrostatic
potential. The structure provided an immediate
explanation for how the protein can move rapidly
along DNA without dissociation, since it revealed
the basis for a non-speci®c topological interaction
with DNA.
The connection between structure and function
that is so apparent in the b-clamp naturally raised
the question as to whether the processivity factors
for other DNA polymerases adopt a similar ringshaped structure. The bacterial b-clamp is a dimer
of 40 kDa subunits, whereas the eukaryotic DNA
polymerase processivity factor PCNA and gp45
(protein encoded by gene 45) are trimers of
25 kDa subunits. The internal symmetry of the bring suggested that PCNA and gp45 could form
similar rings if each monomer of these proteins
contained two of the three domains of the b-clamp
(Kong et al., 1992). While this proposal has an
attractive simplicity to it, the amino acid sequences
of these proteins are completely unrelated and the
truth of this proposal can be established only by
crystallographic analysis. The crystal structures of
yeast and human PCNA have been determined
and have con®rmed that the principles established
for the bacterial processivity factor are readily
# 2000 Academic Press
1216
transferred to the eukaryotic system (Gulbis et al.,
1996; Krishna et al., 1994).
The bacteriophage T4 replication apparatus has
served as an important model system for understanding the principles underlying high-speed
DNA replication. Replication of T4 DNA is performed by four proteins. These are the DNA polymerase (gp43), the processivity clamp (gp45) and
two proteins (gp44 and gp62) that make up an
ATP-dependent clamp-loader assembly that loads
the clamp onto DNA. The gene 45 protein of
phage T4 is the original ``sliding clamp'', identi®ed
by Alberts and colleagues as one of the proteins
responsible for the remarkably processive replication of DNA by T4 DNA polymerase and for its
ability to pass through DNA hairpins (Huang et al.,
1981). gp45 acts also as a transcriptional activator
for late genes, by interacting with RNA polymerase
(Herendeen et al., 1989).
The gp45 protein has been shown to form trimers in the absence of DNA (Jarvis et al., 1989),
and cryo-electron microscopic images suggest that
it is a disk-shaped protein that encircles DNA
(Gogol et al., 1992). Although we have speculated
that its architecture would resemble that of E. coli
b subunit and eukaryotic PCNA (Kong et al., 1992),
the lack of any signi®cant sequence similarity
between these proteins requires that the crystal
structure of gp45 be determined. This is particularly the case, since the sequence alignment on
which we based the proposal that the b-clamp,
PCNA and gp45 would resemble each other (Kong
et al., 1992) was shown by our subsequent crystallographic analysis of PCNA to be almost completely in error. The structure of PCNA and the
b-clamp do, however, resemble each other closely
(Krishna et al., 1994).
Here, we present the crystal structure of the
gene 45 protein of T4 bacteriophage, gp45, which
forms a ring-shaped trimer in the crystal. A struc-
T4 DNA Polymerase Processivity Factor
ture-based sequence alignment shows that the
sequence identity between gp45 and yeast PCNA
is only 10 %. Despite this lack of similarity in their
sequences, the structures of the two proteins are
striking similar and it seems probable that they
interact with DNA in a similar manner. Here, we
focus on a discussion of the architectural elements
of gp45 and a comparison of this structure to that
of PCNA. We have provided preliminary versions
of the coordinates of gp45 to other researchers in
advance of publication. Several other groups have
commented on the implications of our structure for
gp45 function in DNA replication and RNA transcription (Alley et al., 1999; Pietroni et al., 1997;
Soumillion et al., 1998; Wong & Geiduschek, 1998).
Results
The overall structure of gp45 resembles that
of PCNA
The structure of gp45 was determined by multiple isomorphous replacement (MIR), using two
mercury derivatives that were obtained by mutating two residues in the protein separately to
cysteine. The ®nal model for the protein was
Ê (see Materials
re®ned against X-ray data to 2.4 A
and Methods). The crystals contain a trimer of
gp45 in the asymmetric unit, which forms a ring
(Figure 1). The longest dimensions across the outer
Ê , and the ring has
surface of the ring are 85 A
Ê
an internal diameter of 35 A and a thickness of
Ê (these distances are between pairs of
25-26 A
Ca atoms).
The ring formed by gp45 is very similar to
that formed by PCNA and the b-clamp
(Figure 1). As in PCNA and b-clamp, the topology of the chain fold of gp45 is highly symmetrical, despite the lack of apparent internal
symmetry in the amino acid sequence (Figure 2).
Each monomer of gp45 consists of two domains,
Figure 1. Comparison of the structures of gp45 (left) and PCNA (right). Each monomer in the trimer is colored differently. The two domains in each monomer are referred to as domain 1 and domain 2. The structure of PCNA
depicted here is that of the protein from the yeast Saccharomyces cerevisiae (Krishna et al., 1994), RCSB Protein Databank code 1PLQ.
1217
T4 DNA Polymerase Processivity Factor
each of which is composed of a twofold repeat
of a b-a-b-b-b structural motif. The entire ring is
made up of this motif repeated 12 times around
a circle. The 12 a-helices of the motif are
arranged adjacent to each other and they line
the inner surface of the ring. The b-strands form
six b-sheets, which are on the outer surface, of
the ring and give it integrity by straddling the
inter-domain and inter-molecular interfaces.
The fold of the polypeptide chain is very similar to that seen in PCNA and the E. coli b
clamp subunit (Figure 3). Each monomer of
gp45 consists of two domains of 100 residues
each. Each domain is wedge-shaped, with two
a-helices arranged in an anti-parallel arrangement along the broad face of the wedge. Two bsheets that are almost orthogonal to each other
form the two outer surfaces of the wedge, and
these sheets are continued across the interdomain and inter-molecular boundaries.
A striking feature of the b and the PCNA structures is the nature of the connecting loops that link
different domains. Adjacent domains are packed in
a head-to-tail arrangement in these proteins, which
necessitates a long inter-domain linker that passes
over the central b-sheet in a characteristic manner
that is conserved in the bacterial and eukaryotic
clamps. The structure of gp45 shows that it has a
similar inter-domain connector (Figure 4(b)). The
similarity in the inter-domain connectors between
b-clamp, PCNA and gp45 suggests that it might
correlate with a conserved functional requirement.
In eukaryotes, the cell-cycle inhibitor p21/waf-1/
cip1 has been shown to directly inhibit DNA repli-
cation by binding to the inter-domain connector of
PCNA (Figure 4(b)) (Flores-Rozas et al., 1994;
Gulbis et al., 1996; Waga et al., 1994; Warbrick et al.,
1995). A crucial role for the linker in the attachment of the DNA polymerase to the ring has been
demonstrated recently (Shamoo & Steitz, 1999).
Using our structure of gp45 as a molecular replacement search model, Steitz and co-workers have
determined the structure of the processivity factor
for the DNA polymerase of bacteriophage RB69, a
close relative of bacteriophage T4. Their structure,
which is very similar to that of T4 gp45, shows
that the interaction between the RB69 processivity
factor and the DNA polymerase mimics the interaction of p21/waf-1/cip1 with PCNA (Shamoo &
Steitz, 1999).
The structures of gp45 and PCNA diverge
in detail
While the similarities between gp45 and PCNA
are striking, there is considerable divergence
between the two structures when they are examined closely. Much of the differences result from a
loss of regularity in the formation of the secondary
structural elements in gp45. The two domains in
gp45 are less similar to each other than those of
PCNA, which gives the gp45 ring a triangular
appearance that contrasts to the clearly hexagonal
shape of PCNA (see ®gure 1). For example, one of
the a-helices in the second domain of gp45 is
reduced to just one helical turn (Figure 2). Several
of the b-strands in gp45 are shorter than the corresponding elements in PCNA (the N-terminal
Figure 2. Structure-based sequence alignment of gp45 and yeast PCNA. The crystal structures of gp45 and PCNA
were aligned using DALI (http://www2.embl-ebi.ac.uk/dali/fssp/) (Holm & Sander, 1993), and the regions considered by DALI to be structurally overlapped are boxed. Secondary structural elements are depicted as red rectangles (a-helices) and blue arrows (b-strands). Sequence similarity is indicated in color, with green for identity and
shades of green to yellow for decreasing degrees of similarity. The alignment has been organized to highlight the
internal symmetry of the molecules. Secondary structural elements have been sequentially labeled (e.g. aA1, bB1 for
domain 1 and aA2. bB2 for domain 2. This numbering follows that of PCNA (Krishna et al., 1994).
1218
T4 DNA Polymerase Processivity Factor
Figure 3. Structural superposition of domains from gp45 (red) and yeast PCNA (yellow). Ca traces of domain 1 of
the two proteins are shown in stereo, as superimposed by DALI (Holm & Sander, 1993).
strand of PCNA is missing altogether in gp45), and
the hydrogen bonding networks are less regular.
The functional signi®cance of the structural irregularities in gp45 is unclear, but they may be
related to the fact that the gp45 ring is less stable
than that of PCNA and b-clamp. Unlike b and
PCNA, which do not readily fall off DNA once
they are loaded, gp45 attains stability on DNA
only in the presence of the polymerase (Yao et al.,
1994). It appears that the gp45 ring readily adopts
an open con®guration in solution (Alley et al.,
1999). Consistent with this, the inter-domain interfaces in gp45 are less extensive than in PCNA or in
the b-clamp (Figure 4). The number of hydrogen
bonds at the interface is reduced from eight in
PCNA to four in gp45, and the total surface area
that is buried in gp45 upon trimer formation
Ê 2) is 52 % of the area buried in PCNA
(2600 A
Ê
(5000 A2).
Alignment of the structures of gp45 and PCNA
using the DALI server (Holm & Sander, 1993)
results in the superposition of 78 residues in the Nterminal domains of gp45 and PCNA, with an rms
Ê over these resideviation in Ca positions of 2.7 A
dues (Figure 3). The overall sequence identity
between gp45 and PCNA, after structural alignment of both domains, is 10 % (Figure 2). For comparison, the two domains of yeast PCNA share
15 % sequence identity, and the sequence identity
between various domains of the b-clamp, or
between domains of the b-clamp and PCNA,
ranges from 6 % to 15 %. In terms of rms deviations, the structure of gp45 is not strikingly different from that of the other clamp proteins, but the
increased irregularity in secondary structural
elements gives gp45 a distinctive appearance when
compared to the other two.
The electrostatic properties of the gp45 ring
are similar to that of PCNA
Despite these differences in the details of the
architecture, the most striking result obtained on
comparing the structure of gp45 with that of
PCNA and the b-clamp is the conservation of the
nature of the electrostatic potential generated by
these proteins (Figure 5). All three proteins have
net negative charge: the b-clamp dimer has a net
charge of ÿ22, and the yeast PCNA and gp45
trimers have net charges of ÿ60 and ÿ21, respectively, assuming histidine to be neutral. As in
PCNA and b, the distribution of charged sidechains in gp45 is such that the center of the ring
is a region of positive electrostatic potential
(Figure 5).
In addition to the conservation of positive electrostatic potential in the center of the rings, the
internal diameter of the hole in the center is similar
Ê ). Duplex Bin all three processivity clamps (35 A
form or A-form DNA have similar cross-sectional
Ê (B), 25.5 (A)), and both can
widths (23.7 A
readily pass through the centers of each of the
rings (Figure 6).
Conclusions
DNA polymerases involved in chromosomal
replication are notable in being very poorly processive in the absence of their cognate processivity
factors. The coupling of the polymerase to a separate DNA-bound protein in order to achieve high
processivity allows for proper regulation of the
polymerase, as well as facilitating the rapid cycling
of the polymerase from one site of discontinuous
DNA replication to another. The striking similarity
T4 DNA Polymerase Processivity Factor
1219
Figure 4. Comparison of the inter-domain regions of gp45 (left) and human PCNA Protein Databank entry 1AXC)
(Gulbis et al., 1996) (right). (a) Domains 1 and 2 are stitched together by an eight-stranded b-sheet in both cases. The
PCNA dimer is held together by eight hydrogen bonds, while gp45 has only four. (b) The secondary structural
elements in gp45 are more irregular, however. For example, b-sheet continuity is disrupted by an increase in separation of two adjacent strands in gp45 (indicated by a red arrow). The inter-domain connector, which provides a binding site for the p2l/waf1 inhibitor protein (green) in human PCNA, is shown in blue.
between the structures of the processivity factors
from bacteria, eukaryotes and now from T4 bacteriophage suggests strongly that the mechanism
of achieving processivity by hooking the polymerase to a ring-shaped sliding clamp was arrived at
early in evolution and has been elaborated on ever
since.
A common origin for these replication
machines is also suggested by consideration of
the ATP-dependent proteins that load these
clamps onto DNA. These proteins, the gp44/62
complex in T4 bacteriophage, the g-complex in
eubacteria, and the replication factor-C (RF-C)
complex in eukaryotes, share recognizable
sequence similarity and are likely to be related
structurally (Guenther et al., 1997; O'Donnell
et al., 1993). Now that the structural principles
underlying processivity are established for all the
well-studied replication systems, it is anticipated
that the focus of structural analysis will turn to
the elucidation of the clamp-loader complexes
and intact replisomes.
1220
T4 DNA Polymerase Processivity Factor
Figure 5. Electrostatic potential of gp45 and yeast PCNA. (a) The electrostatic potential for gp45 is shown, as calculated by GRASP (Nicholls et al., 1991). The left panel shows a map of the electrostatic potential in a plane that passes
through the center of the gp45 ring. The colors range from red (corresponding to an electrostatic potential energy of
ÿ2.5 kbT, where kb is Boltzmann's constant and T is the temperature) to blue (‡2.5 kbT). The ionic strength of the
solvent was set to be equivalent to 150 mM NaCl. The gp45 ring is a symmetric trimer, but the conformation of the
inter-domain connector is slightly different in each of the three molecules. This accounts for the asymmetric appearance of regions of positive potential (blue) along the outer surface of the ring, which is caused by different orientations of lysine residues in this linker. The molecular surface of the gp45 ring, in the same orientation as that used to
calculate the electrostatic potential map, is shown at the right. (b) The electrostatic potential map for yeast PCNA,
displayed as for gp45.
Materials and Methods
Recombinant gp45 protein was expressed in E. coli
(strain DH5-a) from plasmid pTL151-W (Rush et al.,
1989). Cultures were grown at 32 C to mid-logarithmic
phase and protein production was induced by shifting
the temperature to 42 C for three hours. The protein
was puri®ed by anion-exchange chromatography followed by af®nity chromatography on heparin-Sepharose.
Single-site cysteine mutants were generated using mutagenic primers and the polymerase chain reaction (PCR).
Ampli®ed fragments containing the appropriate restriction sites were cloned into the NdeI and BamHI sites of
the pET26b vector and expression of the mutant proteins
was carried out in E. coli strain BL21/pLysS according to
standard procedures. For crystallization, wild-type and
mutant gp45 proteins were concentrated to 15 mg/ml in
10 mM Tris (pH 8.0) by ultra®ltration (Amicon). The
crystallization of gp45 has been reported (Thaller et al.,
1984) but, since the earlier analysis did not yield highquality crystals, we screened for new conditions using a
sparse matrix screen (Jancaric & Kim, 1991). After optim-
ization of growth conditions, diffraction quality crystals
were obtained overnight by vapor diffusion at 4 C over
a reservoir consisting of 60 mM Pipes (pH 6.6), 200 mM
CaSO4. 0.1 % (v/v) 1,4-dioxane, 15 % (v/v) glycerol, 15 %
(v/v) PEG-MME 5 K (polyethylene glycol mono-methyl
ether, average 5000 Da). The mutant gp45 proteins were
derivatized with mercury by dialysis against 0.1 mM
PHMB (para-hydroxymercuribenzoic acid) in 10 mM Tris
(pH 8.0) for 16 hours at 4 C. Unreacted PHMB was
removed by dialysis against a great excess of 10 mM Tris
(pH 8.0), and the derivatized proteins were crystallized
under conditions identical with those used for the wildtype protein.
For data collection, crystals were ®rst transferred into
a solution containing 60 mM Pipes (pH 6.6), 500 mM
CaSO4, 0.1 % 1,4-dioxane, 20 % glycerol, 15 % PEG-MME
5K. After incubation for 15 minutes the crystals were.
¯ash-frozen in liquid propane at 100 K. The crystals are
in the primitive orthorhombic space group (P212121,
Ê , b ˆ 93.9 A
Ê , c ˆ 141.9 A
Ê ), with three molecules
a ˆ 66.2 A
Ê resolin the asymmetric unit. A native data set to 2.4 A
ution was collected at beamline X25 (National Synchro-
1221
T4 DNA Polymerase Processivity Factor
Figure 6. DNA can pass through the rings of b-clamp, PCNA and gp45. The molecular surface of B-form DNA is
shown in the center. The rings of gp45 (left), the b-clamp of E. coli DNA polymerase III (Kong et al., 1992) (center)
and human PCNA (Gulbis et al., 1996) (right) are shown in cross-section.
Table 1. Crystallographic statistics
Ê , b ˆ 93.9 A
Ê , c ˆ 141.8 A
Ê , a ˆ 90 , b ˆ 90 , g ˆ 90
Space group P212121; cell parameters a ˆ 66.2 A
A. Multiple isomorphous replacement analysis
Dataset
Ê)
Resolution (A
Native
20-2.7
(X25-NSLS)
Thr8Cys(Hg)
20-3.3
(RU200)
Lys8Cys(Hg)
20-3.3
(RU200)
Overall figure of merit 0.53
B. Molecular model refinement
Dataset
Native
X25/NSLS
Reflections
jFj > 2s atoms
30,890
a
Reflections (measured/unique)
299,147
25,239
64,123
13,981
142,449
13,865
Completeness
2 s cutoff
Rsym a
(overall/outer (overall/outer
shell)
shell)
Riso
b
Phasing
powerc
96.3/90.3
8.6/14.6
85.9/76.1
10.2/18.8
20.3
2.2
97.2/98.8
11.7/25.2
19.8
2.1
Completeness
Rsym (overall/
outer shell)
9.6/29.0
Resolution
Reflections (measured/unique)
(overall/outer shell)
500-2.45
30,934
282,837
30,937
92.5/95.4
Number of
atoms
5,250
Rfree e
(Rworking f)
25.91
23.22
rmsd bonds
Ê)
(A
0.007
rmsd B-values
rmsd angles for main chain
Ê 2)
(deg.)
(A
1.29
1.5
Rsym ˆ 100 jI ÿ hIij/I, where I is the integrated intensity of a given re¯ection.
Riso ˆ 100 jFPH ÿ FPj/FP, where FPH and FP are the derivative and native structure factor amplitudes, respectively.
Phasing power ˆjFHj/jjFPHcalcjj, where FH is the calculated heavy-atom structure factor amplitude.
d
Figure of merit ˆhjP(a)eia/jP(a)ji, where a is the phase and P(a) is the phase probability distribution.
e
R free ˆ jFobs ÿ Fcalcj/Fobs, calculated using 10 % of the data.
f
R working ˆ jFobs ÿ Fcalcj/Fobs,
b
c
1222
tron Light Source, Brookhaven) from a single frozen
Ê and
native crystal, using an X-ray wavelength of 1.1 A
an MAR image-plate area detector. All data were processed using programs DENZO and SCALEPACK.
Attempts to solve the structure by molecular replacement, using the structures of PCNA as search models,
were unsuccessful. An exhaustive screen for heavy-metal
derivatives failed to identify useful compounds. Sitespeci®c mutagenesis was therefore used to replace polar
and charged amino acid residues in the N-terminal
region (residues 5 to 20) by cysteine. The resulting proteins were tested for expression, were derivatized with
mercury using the procedure described above, and crystallization was attempted. Successful mercury derivatives were obtained for cysteine substitutions at two
positions: Thr8 and Lys12. The structure was solved by
multiple isomorphous replacement (MIR). Heavy-atom
positions were located using SHELX-90 and re®ned
using HEAVY. The resulting electron density map was
interpretable, and was signi®cantly improved by threefold averaging over the non-crystallographic symmetry.
The structure was re®ned using strict non-crystallographic symmetry (NCS) until the very ®nal stages. The
inter-domain linker regions in the three molecules in the
gp45 trimer adopt somewhat distinct conformations, and
these had to be modeled separately. NCS restraints were
used in the ®nal stages of the re®nement, with the linker
region excluded from the NCS restraints (Table 1).
After a preliminary model had been built and re®ned
we were noti®ed by G. J. Latham that the sequence of
gp45 in GenBank (accession number 215899; (Spicer et al.,
1982)) had three errors in it. The correct sequence (GenBank accession number M10160; (Latham, 1998; Yeh
et al., 1998)) includes three changes (R92P, D99V and
L128V) and one insertion (Asp after Asp197). The
revised sequence was more consistent with the electron
density. In the D99V case, for example, an acidic sidechain that was in the interior of the protein is replaced
with a hydrophobic one. The ®nal model was re®ned
using the revised sequence.
RCSB Protein Data Bank accession code
Coordinates for the gp45 protein and associated structure factors have been deposited with the Protein Data
Bank at the Research Collaboratory for Structural Bioinformatics under the accession code 1CZD.
Acknowledgments
We thank Xiadong Wu, Olga Livanis and Ramakoti
Suresh for help with crystallization, and Ramakoti Suresh and Lonny Berman for help with X-ray data collection. This work was supported, in part, by a grant from
the NIH to J.K. (45547) and to M.O.D. (38839). The
National Synchrotron Light Source is supported by the
United States Department of Energy Of®ces of Health
and Environmental Research and of Basic Energy
Sciences under prime contract DE-AC02-98CH10886, by
the National Science Foundation, and by National Institutes of Health grant 1P41 RR12408-01A1.
References
Alley, S. C., Shier, V. K., Abel-Santos, E., Sexton, D. J.,
Soumillion, P. & Benkovic, S. J. (1999). Sliding
T4 DNA Polymerase Processivity Factor
clamp of the bacteriophage T4 polymerase has open
and closed subunit interfaces in solution. Biochemistry, 38, 7696-7709.
Flores-Rozas, H., Kelman, Z., Dean, F. B., Pan, Z.-Q.,
Harper, J. W., Elledge, S. J., O'Donnell, M. &
Hurwitz, J. (1994). Cdk-interacting protein 1 directly
binds with proliferating cell nuclear antigen and
inhibits DNA replication catalyzed by the DNA
polymerase d holoenzyme. Proc. Natl Acad. Sci.
USA, 91, 8655-8659.
Gogol, E. P., Young, M. C., Kubasek, W. L., Jarvis, T. C.
& von Hippel, P. H. (1992). Cryoelectron microscopic visualization of functional subassemblies of
the bacteriophage T4 DNA replication complex.
J. Mol. Biol. 224, 395-412.
Guenther, B. D., Onrust, R., Sali, A., O'Donnell, M. &
Kuriyan, J. (1997). Crystal structure of the d0 subunit
of the clamp-loader complex of E. coli DNA polymerase III. Cell, 91, 335-345.
Gulbis, J. M., Kelman, Z., Hurwitz, J., O'Donnell, M. &
Kuriyan, J. (1996). Structure of the C-terminal
region of p21waf1/cip1 complexed with human
PCNA. Cell, 87, 297-306.
Herendeen, D. R., Kassavetis, G. A., Barry, J., Alberts,
B. M. & Geiduschek, E. P. (1989). Enhancement of
bacteriophage T4 late transcription by components
of the T4 DNA replication apparatus. Science, 245,
952-958.
Holm, L. & Sander, C. (1993). Protein structure comparison by alignment of distance matrices. J. Mol. Biol.
233, 123-138.
Huang, C.-C., Hearst, J. E. & Alberts, B. M. (1981). Two
types of replication proteins increase the rate at
which T4 DNA polymerase traverses the helical
regions in a single stranded DNA template. J. Biol.
Chem. 256, 4087-4094.
Jancaric, J. & Kim, S.-H. (1991). Sparse matrix sampling:
a screening method for crystallization of proteins.
J. Appl. Crystallog. 24, 409-411.
Jarvis, T. C., Paul, L. S. & von Hippel, P. H. (1989).
Structural and enzymatic studies of the T4 DNA
replication system. I. Physical characterization of
the polymerase accessory protein complex. J. Biol.
Chem. 264, 12709-12716.
Kelman, Z. & O'Donnell, M. (1995). DNA polymerase
III holoenzyme: structure and function of a chromosomal replicating machine. Annu. Rev. Biochem. 64,
171-200.
Kong, X. P., Onrust, R., O'Donnell, M. & Kuriyan, J.
(1992). Three-dimensional structure of the beta subunit of E. coli DNA polymerase III holoenzyme: a
sliding DNA clamp. Cell, 69, 425-437.
Krishna, T. S. R., Kong, X.-P., Gary, S., Burgers, P. &
Kuriyan, J. (1994). Crystal structure of the eukaryotic DNA polymerase processivity factor PCNA.
Cell, 79, 1233-1243.
Kuriyan, J. & O'Donnell, M. (1993). Sliding clamps of
DNA polymerases. J. Mol. Biol. 234, 915-925.
Latham, G. J. (1998). Title. Res. Commmum Biochem. Cell.
Mol. Biol. 2, 139-147.
Nicholls, A., Sharp, K. A. & Honig, B. (1991). Protein
folding and association: insights from the interfacial
and thermodynamic properties of hydrocarbons.
Proteins: Struct. Funct. Genet. 11, 281-296.
O'Donnell, M., Onrust, R., Dean, F. B., Chen, M. &
Hurwitz, J. (1993). Homology in accessory proteins
of replicative polymerases - E. coli to humans. Nucl.
Acids Res. 21, 1-3.
T4 DNA Polymerase Processivity Factor
Pietroni, P., Young, M. C., Latham, G. J. & von Hippel,
P. H. (1997). Structural analyses of gp45 sliding
clamp interactions during assembly of the bacteriophage T4 DNA polymerase holoenzyme. I. Conformational changes within the gp44/62-gp45-ATP
complex during clamp loading. J. Biol. Chem. 272,
31666-31676.
Rush, J., Lin, T. C., Quinones, M., Spicer, E. K., Douglas,
I., Williams, K. R. & Konigsberg, W. H. (1989). The
44P subunit of the T4 DNA polymerase accessory
protein complex catalyzes ATP hydrolysis. J. Biol.
Chem. 264, 10943-10953.
Shamoo, Y. & Steitz, T. A. (1999). Building a replisome
from interacting pieces: sliding clamp complexed to
a peptide from DNA polymerase and a polymerase
editing complex. Cell, 99, 155-166.
Soumillion, P., Sexton, D. J. & Benkovic, S. J. (1998).
Clamp subunit dissociation dictates bacteriophage
T4 DNA polymerase holoenzyme disassembly. Biochemistry, 37, 1819-1827.
Spicer, E. K., Noble, J. A., Nossal, N. G., Konigsberg,
W. H. & Williams, K. R. (1982). Bacteriophage T4
gene 45. Sequences of the structural gene and its
protein product. J. Biol. Chem. 257, 8972-8979.
Stillman, B. (1994). Smart machines at the DNA replication fork. Cell, 78, 725-728.
Stukenberg, P. T., Studwell-Vaughan, P. S. & O'Donnell,
M. (1991). Mechanism of the sliding b-clamp of
1223
DNA polymerase III holoenzyme. J. Biol. Chem. 266,
11328-11334.
Thaller, C., Alberts, B., Goldsmith, E., Sprang, S. &
Fletterick, R. (1984). Crystallization of the gene 45
protein from the DNA replication fork of the bacteriophage T4. J. Biol. Chem. 259, 2568-2569.
Waga, S., Hannon, G. J., Beach, D. & Stillman, B. (1994).
The p2l inhibitor of cyclin-dependent kinases controls DNA replication by interaction with PCNA.
Nature, 369, 574-578.
Warbrick, E., Lane, D. P., Glover, D. M. & Cox, L. S.
(1995). A small peptide inhibitor of DNA replication
de®nes the site of interaction between the cyclindependent kinase inhibitor p21 waf1 and the proliferating cell nuclear antigen. Curr. Biol. 5, 275-282.
Wong, K. & Geiduschek, E. P. (1998). Activator-sigma
interaction: a hydrophobic segment mediates the
interaction of a sigma family promoter recognition
protein with a sliding clamp transcription activator.
J. Mol. Biol. 284, 195-203.
Yao, N., Turner, J., Kelman, Z., Stukenberg, P. T., Dean,
F., Schechter, D., Pan, Z.-Q., Hurwitz, J. &
O'Donnell, M. (1994). Clamp loading, unloading
and intrinsic stability of the PCNA, beta and gp45
sliding clamps of human, E. coli and T4 replicases.
Genes Cells, 1, 101-113.
Yeh, L.-S., Hsu, T. & Karam, J. D. (1998). Divergence of
a DNA replication gene cluster in the T4-related
bacteriophage RB69. J. Bacteriol. 180, 2005-2013.
Edited by I. A. Wilson
(Received 12 October 1999; received in revised form 9 December 1999; accepted 10 December 1999)