annual reviews - replication
TRANSCRIPT
-
8/4/2019 Annual Reviews - Replication
1/36
Annu. Rev. Biochem. 2005. 74:283315doi: 10.1146/annurev.biochem.73.011303.073859
Copyright c 2005 by Annual Reviews. All rights reserved
CELLULAR DNA REPLICASES: Componentsand Dynamics at the Replication Fork
Aaron Johnson2 and Mike ODonnell1,21Howard Hughes Medical Institute, 2The Rockefeller University,
New York City, New York 10021-6399; email: [email protected],
Key Words DNA replication, DNA sliding clamps, DNA polymerase, processivityclamp loader, protein-DNA interactions
Abstract Chromosomal DNA replicases are multicomponent machines that haveevolved clever strategies to perform their function. Although the structure of DNAis elegant in its simplicity, the job of duplicating it is far from simple. At the heartof the replicase machinery is a heteropentameric AAA+ clamp-loading machine thatcouples ATP hydrolysis to load circular clamp proteins onto DNA. The clamps encircleDNA and hold polymerases to the template for processive action. Clamp-loader and
sliding clamp structures have been solved in both prokaryotic and eukaryotic systems.The heteropentameric clamp loaders are circular oligomers, reflecting the circularshape of their respective clamp substrates. Clamps and clamp loaders also functionin other DNA metabolic processes, including repair, checkpoint mechanisms, and cellcycle progression. Twin polymerases and clamps coordinate their actions with a clamploader and yet other proteins to form a replisome machine that advances the replicationfork.
CONTENTS
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284THE E. coli REPLICASE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
Pol III Core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
The Sliding Clamp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
The Complex Clamp Loader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
The Holoenzyme Particle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Replisome Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Protein Trafficking on DNA Sliding Clamps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
THE EUKARYOTIC REPLICASE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
Proliferating Cell Nuclear Antigen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
The RFC Clamp Loader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300Eukaryotic DNA Polymerases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
The Eukaryotic Replisome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
Alternate RFC Complexes and Other Roles for Clamp Loaders . . . . . . . . . . . . . . . . 307
CONCLUDING REMARKS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
0066 4154/05/0707 0283$20 00 283
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
2/36
284 JOHNSON ODONNELL
INTRODUCTION
Chromosomal DNAs are exceedingly large molecules, and the vast repository of
information they hold must be duplicated with precision [see (1) for overview].The sheer size of these molecules may explain why all cells utilize a clamp loader
to position a circular sliding clamp on DNA that tethers the DNA polymerases to
their long substrates for highly processive synthesis (see Figure 1). The eukaryotic
proliferating cell nuclear antigen (PCNA) and prokaryotic () clamp proteins have
unrelated sequences, yet they have strikingly similar structures and thus share a
common ancestor (2, 3). The clamps must be opened and closed around DNA, and
this job is performed by a multiprotein clamp-loading ATPase [reviewed in (4)].
Figure 1 Components of cellular replicases. The table lists the three replicase componentsin the well-defined systems of Escherichia coli, eukaryotes, Archaea, and T4 phage. The
scheme below is a generalized mechanism of replicase action. A multiprotein clamp-loader
couples ATP binding and hydrolysis to loading of a ring-shaped processivity clamp that is
then used by the replicative polymerase as a tether to the DNA template.
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
3/36
CELLULAR REPLICASES 285
The eukaryotic replication factor C (RFC) and the prokaryotic ( complex) clamp-
loader subunits are arranged in a circle and are each members of the AAA+ protein
family [briefly reviewed in (4)].
The replicases of all cells function within a greater context, involving many otherproteins. For example, two basic activities at a replication fork include a helicase
to separate the duplex DNA and a primase that produces short RNA primers,
which are required to initiate DNA synthesis. This larger machinery, frequently
termed the replisome, is fairly well defined in prokaryotes (1, 58). In E. coli,
and presumably other bacteria, the leading- and lagging-strand polymerases are
connected to the clamp loader, which also binds a homohexameric helicase (DnaB
in E. coli). The helicase acts on the lagging strand and activates the RNA primase
(DnaG in E. coli). The replicase also makes a specific functional connection to
single-stranded DNA-binding protein (SSB), which is present at the fork to protectsingle-stranded DNA (ssDNA) and melt hairpins.
The eukaryotic replisome, in contrast, involves many more proteins and is less
defined at the current time [reviewed in (911)]. The eukaryotic helicase, primase,
and SSB are all heterooligomers. The helicase is thought to be the heterohexam-
eric MCM complex (12), and the primase is the four-subunit DNA polymerase
/primase that makes a hybrid RNA/DNA primer (13). The heterotrimeric repli-
cation protein A (RPA) functions as the SSB (14). Several additional eukaryotic
proteins without prokaryotic counterparts are thought to act during replication ini-
tiation and fork progression as well (e.g., Cdc45, GINS complex, Dpb11, Sld2,and Sld3) (1517). The functions of most of these components are presently un-
known. In addition, the leading- and lagging-strand polymerases are thought to be
different enzymes, Pol (34 subunits) and Pol (4 subunits). It is still unclear
which strands these polymerases act upon. Finally, many of these proteins are
regulated by posttranslational modification, including the PCNA clamp which is
ubiquitinated and sumoylated (1820).
The clamp and clamp loaders of cellular replicases, prokaryote and eukaryote
alike, also function in several other processes besides replication. For example,
PCNA and bind DNA ligase as well as mismatch and excision repair proteins,although the exact role of these interactions is still not entirely clear (2123).
A weak consensus sequence for proteins that bind PCNA reveals a broad array
of additional proteins that bind this clamp and are generally involved in repair,
chromatin structure, or cell cycle control (21). The clamps also function with
other DNA polymerases (23, 24), probably helping to target them to sites where
their action is required. Eukaryotes even utilize alternative RFC clamp loaders in
which one subunit is replaced by a unique protein that presumably specializes the
complex for the alternative function (2527, 234243). These alternative clamp
loaders load PCNA clamps at specific target sites or, in one case, load an alternatePCNA-like clamp.
Because of space limitations, this review focuses on prokaryotic and eukaryotic
replicases and forgoes a description of archaebacterial, bacteriophage, and viral
replicases. However, archaebacteria also utilize a similar clamp and clamp-loader
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
4/36
286 JOHNSON ODONNELL
strategy, as does the T4 bacteriophage (see the table in Figure 1) [reviewed in
(7, 28), respectively]. We regret that space limitations prevent discussion of the
phage T4 replisome in a similar manner; it is an enormously valuable system from
which many basic early concepts of replisome action emerged, and it continuesto provide fresh insights. A recent review in this series (7) expertly distills the T4
contributions to date and how they relate to the E. coli system. The replicative
polymerases of most bacteriophage and viruses do not utilize a clamp and clamp
loader, but these polymerases generally require one or two accessory factors that
confer processivity [see T7 phage (7, 29, 30), vaccinia virus (31, 32), herpes virus
(33, 34), and Pol (35)]. The authors have also refrained from discussing DNA
replication initiation due to space limitation. Many useful reviews exist on this
topic [see (1, 28, 36, 37)].
THE E. coliREPLICASE
The first cellular replicase to be studied, E. coli DNA polymerase III holoenzyme,
was isolated as an intact particle from E. coli extracts in the early 1980s, and its
structure and function serve as a suitable paradigm for its eukaryotic counterpart
(1). We, therefore, begin this review with the prokaryotic replicase machinery.
Pol III Core
DNA polymerase III (Pol III) was originally identified in a mutant strain ofE. coli,
polA (38). This strain lacked the comparatively strong DNA polymerase I activity,
unmasking the replicative polymerase. The identity of this activity was likely the
core polymerase subcomplex. The specific activity of Pol III core is similar to Pol I,
but as will be described below, Pol III core functions with accessory proteins that
convert it to an exceedingly efficient enzyme having the highest specific activity
of any E. coli DNA polymerase (1).
Pol III core is a 1:1:1 heterotrimer of the polymerase, 3-5 proofreading ex-onuclease, and subunits (39, 40) (see Table 1). The subunit (encoded by dnaE)
contains the DNA polymerase activity, incorporating 8 nucleotides/second (ntd/s),
similar to Pol III core (20 ntd/s) (41). The (dnaQ) subunit is the proofreading
3-5 exonuclease. It is interesting to note that without the processivity of holoen-
zyme is markedly reduced from >50 kb to about 1.5 kb (42), thereby ensuring that
the proofreading subunit is present during genome duplication. In contrast, the
small subunit (holE) has no known function besides a slight stimulation of (43)
and the holEgene can be deleted with little consequence (44).
Little structural information is available for Pol III core. The subunit is amember of the C family of DNA polymerases (45). Members of this family are
present only in bacteria, and it remains the only major family for which no crystal
structure has been solved. One may presume that it will contain the characteristic
palm, thumb, and fingers domains and two-metal mechanism of catalysis present
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
5/36
CELLULAR REPLICASES 287
TABLE 1 Escherichia coli replisome components and associated functionsa
in all other polymerases (46). Deletion mutagenesis suggests that the active site islocated in the N-terminal two thirds of, whereas contacts to other holoenzyme
components are made through the C-terminal region (47, 48).
Atomic resolution structures of portions of and are available (4951). The
subunit is composed of two domains (52). The N-terminal domain (186 residues
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
6/36
288 JOHNSON ODONNELL
out of 243) contains the exonuclease active site and the binding site. The crystal
structure of this active fragment has been solved (49), and it is structurally similar
to other polymerase-associated exonucleases and the exonuclease domain of DNA
polymerase I. The preferred DNA substrate for is a single-stranded 3
terminus,and a fully base-paired recessed 3 terminus is a poor substrate for alone (53, 54).
However, tightly associates with the C terminus of and stimulates activity on
a recessed 3 terminus, probably by bringing to the primer site. preferentially
degrades a primer/template with a mismatched recessed 3 terminus. In addition,
polymerization is inhibited by a mismatch, which provides more time to complete
the excision.
The structure of bears a resemblance to a DNA-interacting domain of eukary-
otic DNA polymerase (50, 51). This structural similarity may underlie the slight
stimulation of by and increased mutator phenotype of a holE-null mutant in an mutant background (44).
The Sliding Clamp
Pol III core by itself is slow, incorporating 20 ntd/s, and weakly processive, ex-
tending only 110 bases per binding event (41). In fact, no matter how much time
or enzyme is available, the Pol III core cannot extend a unique primer full circle
around an M13 ssDNA genome (55, 56).
To becomean efficient replicase, the core polymerase requires the clamp. Cou-pled to , core becomes exceedingly fast (750 ntd/s) and processive (>50 kb).
Biochemical studies initially revealed that binds DNA topologically (57), im-
plying it has a ring shape and encircles the duplex, whereupon it freely slides along
it. This hypothesis was quickly proven by structure analysis (3) (Figure 2). Core
polymerase directly associates with (5759), initially occupying about 22 base
pairs (bp) of primer (60, 61). As core extends DNA, it pulls the clamp along
behind it.
The crystal structure of the dimer (3) shows that the two crescent-shaped
protomers form a ring with a large central channel of35 A in diameter that mayeasily accommodate double-stranded (ds) DNA modeled inside (Figure 2a,b).
In fact, room exists for one or two layers of water between the DNA and ,
suggesting may ice skate along the duplex. The center of the ring is lined with
12 -helices, pairs of which are supported by an outer -sheet. This helix pair and
sheet motif forms one globular domain, which is repeated six times around the ring,
creating a sixfold pseudosymmetry. Each monomer consists of three domains
and dimerizes head-to-tail with another to produce two structurally distinct faces.
One face has several loops and protruding C termini. This is the face of the ring
that interacts with other proteins as discussed below (62, 63).The clamp is a tight dimer, and its half-life on DNA is over 1 h at 37C
(64, 65). The subunit of complex can open by itself, as determined by its
ability to rapidly remove the clamp from DNA (64, 66). It appears that only one
binds 2, and it does not dissociate the dimer into monomers, even though it
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
7/36
CELLULAR REPLICASES 289
Figure 2 Structure and dynamics of the sliding clamp. (a,b) Crystal structure of the
dimer with modeled double-stranded DNA through the central channel (courtesy ofD. Jeruzalmi and J. Kuriyan). Panel (b) highlights the two faces of the ring. The C-terminal
face (right) is implicated in many of the interactions of with other proteins. (c) Superposi-
tion of one subunit from the dimer structure (purple) and the monomer from the 11crystal structure (yellow), using domain II as a reference. (d) Model of an open clamp made
by arranging two monomer structures from 11 to create one dimer interface [described in
(67)]. Panels (c) and (d) adapted from (67), copyright 2001, with permission from Elsevier.
must destabilize one interface (62, 65, 66). This idea is consistent with the abilityof complex to load that is cross-linked across one dimer interface (66). A
cocrystal structure of bound to a monomeric mutant of has been solved (67).
This structure shows two distinct points of contact between and . The contact
points are located on the opposite ends of the same -helix, which is located in the
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
8/36
290 JOHNSON ODONNELL
N-terminal domain of. One end of the helix binds a hydrophobic pocket between
domains II and III of . The other end appears to push on a loop at the dimer
interface, leading to a distorted interface that can no longer close. It is proposed
that getsagripon via the hydrophobic pocket and pushes on the loop to crack theinterface or to hold the interface open. Mutational analysis of these contact points
has confirmed their roles in clamp-loader interaction and clamp opening (68).
The - structure explains how the interface is cracked but not how the ring
opens to accommodate DNA in the central channel. Comparison of the dimer
and monomer (from the - structure) shows significant rigid body motions
between the domains, leading to a shallower crescent shape in the monomer (67)
(seeFigure2c).Thisimpliesthatthe ringisunderspringtensionintheclosedstate
until disrupts one interface, allowing tension between the domains to relax and
producing a gap for DNA-strand passage (see Figure 2d). The strong interactionsat the dimer interfaces likely maintain this tension.
The Complex Clamp Loader
The E. coli clamp loader is a complex composed of five different subunits in a
defined stoichiometry: 31
1 1 1 (69, 70). The complex harnesses the energy
of ATP binding and hydrolysis to topologically link to a primed DNA, then it
ejects from DNA, leaving the closed clamp behind (57, 66, 71). It is convenient to
dissect the clamp loader by the primary function of each subunit (4). The three subunits are the only subunits that bind ATP (72, 73) and have thus been termed
the motor of the complex. The subunit is called the wrench because it is the
main clamp-interacting subunit, and it can open the dimer interface by itself.
The subunit modulates - contact (66). appears to be a rigid protein (74).
Unlike and ,thedomainsof have more intramolecular interactions and assume
the same orientation in the structure and within 3. This feature has earned
the term stator, the stationary part of a machine upon which other parts move.
In contrast, the three domains of assume different orientations in the trimer.
The domains of also assume different orientations in - compared to 3
.The 311 complex is termed the minimal clamp loader, as it is sufficient to
place on a DNA template (75). The and subunits are not essential for the
clamp-loading mechanism (76), but links the clamp loader to SSB and primase
(77, 78), which will be discussed below. serves as a connector to (76) and
also strengthens the 3 complex (79).
A broad outline of the clamp-loading mechanism has been determined from
biochemical studies (4, 66). The nucleotide-free clamp loader has a low affinity
for the clamp (62). The subunit, which binds tightly to , is likely sequestered
by the other subunits without ATP present. Upon ATP binding, the complexundergoes a conformational change (62, 71) that allows tight binding to the clamp,
whereupon opens one dimer interface of the clamp (65). The clamp-clamp loader
complex has a strong affinity for DNA, particularly a primed template (71, 80).
The DNA, presumably threaded through the clamp, stimulates ATPase activity in
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
9/36
CELLULAR REPLICASES 291
complex, and this allows the ring to close around DNA and the clamp loader to
eject, thereby recycling it for repeated rounds of clamp loading (66, 81). Hydrolysis
of ATP leaves the clamp loader in a low-affinity DNA-binding state, presumably
until ADP dissociates and the complex can be recharged with triphosphatenucleotides (80, 8284).
The crystal structure of the 3 complex provided deeper insight into clamp-
loader action (70). This accomplishment was followed two years later by the crystal
structure of the eukaryotic clamp loader bound to its clamp (85), which has filled
in many details that will be discussed later. The five subunits ofE. coli 3 are
arranged in a circular fashion (70) (Figure 3b). The three subunits are adjacent
to one another, flanked on one side by and the other by , which in turn interact
with each other to complete the ring. Each subunit consists of three domains, and
oligomerization is mediated mainly by the C-terminal domain III (Figure 3a,b).The N-terminal domains I and II of all five subunits adopt the chain fold of the
AAA+ (ATPases associated with a variety of cellular activities) family (Figure 3a)
[reviewed in (86, 87)]. Although only binds ATP (72), has conserved elements
of the AAA+ family (88). In contrast, the sequence has diverged, making the
discovery of its AAA+ fold surprising.
The ATP sites of complex are located at subunit interfaces and are supple-
mented with residues from the adjacent subunit (see Figure 3c). The interfacial
location of ATP sites is typical of AAA+ oligomers and is presumed to promote
communication among the subunits.
contributes a key arginine to ATP Site 1of the complex, as seen in the left panel of Figure 3c (red residue) with an
ATP molecule modeled along the phosphate-binding loop (blue). This arginine is
located in a conserved serine-arginine-cysteine (SRC) motif that is present in all
known clamp loaders. Mutation of the Arg, and the homologous residue in ,
causes a severe defect in clamp-loading and ATPase activity and also disrupts in-
teraction with DNA ( mutants) or ( mutants) (89, 90). Mutation of the P-loop
of has more severe consequences that result in complete loss of activity (73).
These observations highlight the coordination of the ATP cycle with the clamp-
loading mechanism, suggesting regulation of substrate binding by conformationalchanges in distinct ATP sites during nucleotide binding and hydrolysis.
The C termini (domains III) of the five subunits of the minimal complex form
a tight ring, or collar (91), but the N-terminal AAA+ domains (I and II) are more
loosely associated, with a total lack of contact between and in these domains,
creating a gap in the N-terminal portion of the ring (70) (Figure 3b). This gap may
function to allow DNA to pass into positioned under the clamp-loader complex,
as discussed in the RFC section of this review. Both ATP- and -clamp binding oc-
cur in these N-terminal domains (67, 92). The loose connections of these domains
has led to the suggestion that some conformational freedom of the N termini is es-sential for function (93). The structure of3 lacks ATP and thus is in the inactive
state. Consistent with this, a dimer cannot be docked onto the complex (replac-
ing with -) without significant clashes, even with the substantial gap between
and (70). Biophysical experiments in solution indicate that the distance between
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
10/36
292 JOHNSON ODONNELL
Figure 3 Crystal structure of the E. coli 311 complex. (a) Isolated subunits from
the structure of the complex, domains IIII noted [adapted from (87), copyright 2002
Nature Publishing Group (http://www.nature.com)]. (b) Side view of3 (left). The
-interacting helix of is marked in yellow. View from the C termini of3 (right)
[adapted from (70), copyright 2001, with permission from Elsevier]. (c) The ATP sitesof the ring-shaped complex are located at subunit interfaces, as illustrated in the
cartoon (right) and taken from the structure of ATP Site 1 (left) with ATP (purple)
modeled against the P-loop (blue). The Ser-Arg-Cys (SRC) motif of , conserved in
all clamp loaders, points its arginine (red) toward the phosphate of ATP [adapted
with permission from (89)].
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
11/36
CELLULAR REPLICASES 293
and does not change dramatically during clamp association (93). Hence, some
other conformational change may occur that does not necessitate increasing the
gap between and . For example, may be pushed downward relative to , thus
exposing the -binding element of without changing the distance between and. In addition, the subunit also binds (94). Taken together, these observations
imply that an extensive surface of interaction is formed between the clamp and
clamp loader. How the clamp loader binds the clamp and DNA is illuminated by
the RFC-PCNA structure discussed later in this review.
The Holoenzyme Particle
On the basis of intracellular Pol III subunit concentrations, a portion of clamp
loader is associated with the Pol III holoenzyme, but a majority of the clamp
loader is free in solution (64, 95). The clamp loader associated with the holoenzyme
contains a different form of the dnaX gene. The subunit is produced through a
ribosomal frameshift in the dnaXgene, which causes almost immediate termination
of translation to produce a 47.5-kDa protein (96, 97). The full-length product of
dnaX is the subunit (71.1 kDa), which contains the sequence plus a unique
23.6 kDa C-terminal region (c) (Figure 4a). The 23.6 kDa c is comprised of two
domains, IV and V, that bind DnaB and Pol III core (through ), respectively (98,
99). The c region is not required for clamp loading but is essential for cell viability
(100), probably owing to its ability to organize the replisome as discussed below.
The DnaX protein is present in three copies in the clamp-loader complex, and
thus various species may assemble in stoichiometries of 3, 21, 12, and 3(69). Each C terminus will recruit one polymerase, and therefore the more
present in the clamp loader, the more core polymerase molecules it may bind.
At least two polymerases are required for concurrent synthesis of leading and
lagging strands (Figure 4b). Because of this requirement, it is thought that the
E. coli replicase contains two Pol III cores attached to a 21 clamp loader
and that a specific order of assembly leads to this particle (69, 101, 102), which
is termed Pol III (or Pol III star). The clamp associates with Pol III in an
ATP-dependent manner to form the Pol III holoenzyme. The single-copy subunits
of the clamp loader define an inherent asymmetry, and thus by definition, the
two cores attached to the two subunits are in somewhat different environments
(8). The consequence of this asymmetric structure might be minimal because a
proline-rich segment of separates the clamp-loader and polymerase-interacting
domains, suggesting a flexible connection. However, DnaB or other holoenzyme
subunits may hold the cores in defined asymmetric positions (103). It has been
proposed that the asymmetric structure imposes distinct properties onto the two
polymerases, modeling their behavior to fit the different needs of replicating the
leading and lagging strands (8, 102, 104109).
Replisome Dynamics
The subunits of Pol III holoenzyme not only connect two core polymerases
to the central clamp loader, but also connect the replicase to the DnaB helicase
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
12/36
294 JOHNSON ODONNELL
Figure 4 Organization and dynamics of the E. coli replisome. (a) The subunit of DNA
polymerase III holoenzyme is comprised of the clamp-loader domains IIII ( sequence)
and the replisome organization domains IV and V that bind the DnaB helicase and Pol III
core, respectively. (b) Polymerase cycling at the replication fork. As the replisome advances,the clamp loader loads a clamp on an RNA primer (pink) synthesized by DnaG (upper
right). When the lagging-strand polymerase replicates to a nick, it dissociates from DNA and
(lower right) and cycles to the newly loaded clamp (lower left).
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
13/36
CELLULAR REPLICASES 295
(see Figure 4b). The homohexameric DnaB encircles the lagging strand, and when
coupled to DNA synthesis through , its unwinding rate increases from 35 bp/s
to near holoenzyme speed (98, 99, 103, 110112).
As the replisome advances, the polymerase on the leading strand simply extendsDNA in a continuous fashion. Presumed impediments to leading strand extension
include sites of DNA damage and the consequent collapse of the fork. Repair and
restart of synthesis is an elaborate process detailed in recent reviews (113, 114). Pol
III holoenzyme has been implicated in mediating the restart of DNA replication
after fork stalling (115).
Lagging-strand replication is a discontinuous process of fits and starts that re-
peats in a cycle time of 13 s. An overview of the lagging-strand cycle is illustrated
in Figure 4b. Each Okazaki fragment is initiated by primase, which synthesizes
an RNA primer of about 1012 nucleotides (116, 117). Primase action requiresinteraction with DnaB, which involves a C-terminal region of primase (118, 119).
Primase extends the RNA in the opposite direction of helicase unwinding and is
presumed to separate from DnaB, which may account for its observed distributive
action (120). Primase remains attached to the RNA primed site through its inter-
action with SSB (121123). Although primase eventually dissociates, release of
primase is accelerated by the subunit of the clamp loader, which binds SSB in
a competitive fashion, recruiting the clamp loader to the DNA template to com-
pete with primase (124). The clamp loader then places onto the primer for the
lagging-strand polymerase.As the lagging polymerase extends a fragment, a loop is generated because it
is connected to the leading polymerase (via the clamp loader), yet extends DNA
in the opposite direction (see top left diagram), as originally proposed (125) and
recently confirmed (126) in the T4 system. The 13 kb Okazaki fragment will be
completed within a few seconds (Figure 4b, top right diagram), and at this point,
the core must rapidly release from DNA to start the next fragment (bottom right
diagram). The highly processive Pol III requires a specific mechanism for this
release step, which disengages core from , leaving the clamp behind on the
finished fragment. The release step occurs only at a nick, thus ensuring completionof the fragment, and requires the subunit (127130). The lagging-strand core
is now free to bind a new clamp placed on the next RNA primer by the clamp
loader (bottom left diagram).
Replication fork progression has been studied in E. coli, using rolling-circle
DNA templates (103, 107, 116, 120, 131133). That clamps accumulate on
the lagging strand and that the single clamp loader can rapidly load clamps on
DNA repeatedly during fork progression have been demonstrated using this system
(103). The accumulated clamps can be recycled by the unloading action of
complex and also by the fivefold excess of free subunit in the E. coli cell (64).
Protein Trafficking on DNA Sliding Clamps
Many different proteins function with sliding clamps. An understanding of the
way proteins coordinate their traffic flow on clamps is now emerging. To illustrate
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
14/36
296 JOHNSON ODONNELL
how proteins may switch positions to utilize the clamp, we briefly describe three
important switches on that occur during the progression of theE. coli replication
fork.
The / clamp loader and Pol III core are known to bind the same face of inessentially the same spot, and therefore these two factors compete for the clamp
(63, 134). Yet both the core and / complex must function with at the start of
each DNA chain. This protein trafficking event is facilitated by the fact that ATP
hydrolysis ejects the clamp loader from (Figure 5a) (71). In addition, the clamp
loader remains in a reduced activity state for a short interval, presumably owing to
slow ADP release (80, 83). The core polymerase is then free to bind the abandoned
clamp and, in fact, binds tighter on DNA than off (63).
Switching on is also thought to occur when the leading polymerase stalls.
In this case it is thought that a bypass polymerase, either Pol IV or Pol V, takespossession of the clamp to extend DNA through the stall site. Recent structural
analysis suggests that Pol IV can associate in two ways with (135, 136). Pol IV
binds the edge of the ring and one hydrophobic pocket, but Pol IV is angled
off the DNA (135). It is possible that in this binding mode Pol IV may bind to
simultaneously with Pol III (see Figure 5b). Upon stalling of Pol III, Pol IV
must break its interaction with the side of the ring and swing down to the DNA,
presumably maintaining its hold on the hydrophobic pocket of . This action
would displace Pol III from DNA and perhaps disrupts the Pol III- contact as
well. Pol IV is distributive, even with , allowing Pol III to regain the clampafter the lesion is bypassed. Recent evidence with Pol III, Pol V, and suggests
that a similar trade-off may occur (137). At the end of an Okazaki fragment, the
normally highly processive Pol III rapidly dissociates from (127130), freeing
the polymerase to extend the next Okazaki fragment. This trafficking event is
mediated by the c portion of.The subunit binds via the extreme C-terminal
residues (134). c also binds the C terminus and disrupts core- interaction.
However, c binds ssDNA, andthis prevents from binding the C-terminal residues
of . Hence, so long as there is ssDNA template, c is turned off, and core
functions with . But when all available ssDNA is converted to duplex, c turnson and separates core from (see Figure 5c).These switch processes explain how
interactions with the clamp are modulated by ATP or DNA structure to promote
the trafficking of different proteins on sliding clamps to ensure progression of
replication.
THE EUKARYOTIC REPLICASE
Although the complexity of architecture, interaction, and regulation of DNA repli-cation is far greater in eukaryotes than in bacteria, the core replicase components
are structurally and functionally more similar than different. However, beyond this
basic machinery lies a much larger network of proteins required for propagation
of the replication fork and regulation of its advance as well as coordinating its
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
15/36
CELLULAR REPLICASES 297
Figure 5 Three examples of protein trafficking on sliding clamps. Interactions with
the clamp fromE. coli. (a)The complex clamploader associates tightly with when
bound to ATP. DNA triggers ATP hydrolysis, resulting in low affinity for and DNA.
(b) When Pol III, the replicative polymerase, encounters a lesion in the DNA template,
it stalls, unable to overcome its inherent fidelity to incorporate opposite a damaged base.
Stalling allows an error-prone polymerase, such as Pol IV (red) passively traveling on
, an opportunity to trade places with Pol III on to replicate past the lesion. [Adapted
with permission from (135).] (c) Pol III maintains a tight grip on via the polymerase
C terminus. However, when it completely replicates its substrate DNA, the polymerase
must release from to recycle to the next primed site. The subunit modulates this
interaction, binding the polymerase C tail only when no more single-stranded template
is present. This severs the connection between the polymerase and the clamp [adapted
with permission from (134), copyright 2003, National Academy of Sciences, U.S.A.].
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
16/36
298 JOHNSON ODONNELL
activity with other DNA metabolic machineries. Many of these details are only
now coming into focus.
Proliferating Cell Nuclear AntigenPCNA derives its name from the early finding that the protein is abundant in pro-
liferating cells (138). PCNA was shown to be directly involved in DNA replication
through its ability to stimulate DNA polymerase in replicating long stretches of
primed DNA (139141). This characteristic suggested it may be analogous to the
E. coli subunit, although neither were known to be sliding clamps at the time.
Subsequent work showed the importance of PCNA for simian virus 40 (SV40)
DNA replication and its enhanced activity in the presence of RFC and ATP (142).
As a monomeric unit, PCNA is about two-thirds the size of, and accordingly,
each protomer contains two structurally similar domains instead of the three do-
mains found in (2). The chain folds of the two PCNA domains are the same
as those found in the domains of . PCNA and form very similar ring-shaped
structures, except PCNA must trimerize to form a six-domain ring (Figure 6a).
Like , the PCNA ring is quite stably attached to DNA (t1/2 = 24 min) (143).
The PCNA protomers are also arranged head-to-tail to create two distinct faces
of the ring, mirroring . As in the E. coli system, the eukaryotic clamp loader and
polymerase compete for binding the same face of the PCNA ring, the face from
which the C termini project (144, 145).
A variety of proteins involved in DNA repair and cell cycle control interact
with PCNA, but this topic is covered elsewhere [reviewed in (21, 23)]. Relevant
to the current review, a weak PCNA-binding consensus sequence has emerged:
Q-x-x-h-X-X-a-a, where x = any residue; h = L,I,M; and a = F,Y (21, 23). The
crystal structure of human PCNA in complex with a peptide derived from the cell
cycle regulator p21WAF1/Cip1 was the first to demonstrate that these clamp-binding
Figure 6 Structures of the eukaryotic clamp and clamp loader from Saccharomycescerevisiae. (a) View of the C-terminal face of PCNA. The ring-shaped PCNA is a head-
to-tail trimer of a two-domain monomer. The sixfold pseudosymmetry of the clamp
is evident in PCNA as well. (b) The structure of replication factor C (RFC) bound to
PCNA reveals the structural similarity between RFC and complex. RFC binds to the
C-terminal face of PCNA. (ce) The RFC subunits are arranged in a helix that tracks
the minor groove of B-form DNA modeled through the PCNA ring. (d) In this cartoon,
the 5 terminus of a recessed primer template is positioned to exit the central channel of
the clamp and clamp loader through the gap between RFC1 and RFC5. ( e) N-terminal
regions of the five RFC subunits and the PCNA ring from the RFC-PCNA structure.Two conserved helices in each RFC subunit (yellow) are in position to interact with
DNA (orange/green) that passes through the central channel of PCNA (gray)withthe5
terminus (green spheres) exiting between RFC1 and RFC5. [Adapted with permission
from (85), copyright 2004 Nature Publishing Group (http://www.nature.com).]
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
17/36
CELLULAR REPLICASES 299
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
18/36
300 JOHNSON ODONNELL
proteins interact with PCNA at a hydrophobic pocket located between the two
domains of a protomer (146). The finding generalizes to the position at which
DNA polymerase binds the T4 gp45 clamp (147149). Also, the subunit of Pol
III core and subunit of complex bind E. coli at a spot between domains IIand III (24, 67, 150).
The RFC Clamp Loader
The eukaryotic clamp loader was first isolated as a required component of the
in vitro SV40 genome replication system (151) [reviewed in (152)]. The activity
was originally named replication factor C (RFC) (51) or activator-1 (153). RFC
is a DNA-dependent ATPase that functions with PCNA to confer processivity on
DNA polymerase (142) [reviewed in (154, 155)]. The similarity between RFCandE. coli complex was apparent early on. RFC consists of five different proteins
that are homologous to each other and to and ofE. coli complex (156, 157).
Common sequence motifs among these clamp-loader AAA+ proteins have been
termed RFC boxes.
In overview, RFC acts similarly to complex. In an ATP-dependent reaction,
RFC loads PCNA onto a recessed 3 primer/template junction and then dissociates,
allowing PCNA to function with Pol (158, 159). The RFC subunits in S. cere-
visiae are referred to as RFC15 (157) (see Table 2 for human RFC nomenclature).
RFC25 share a similar molecular weight and three-domain architecture charac-teristic of the E. coli subunit (85, 160). RFC1 also contains these three domains,
along with sizable N- and C-terminal extensions (161). The N-terminal region
(residues 1275 in S. cerevisiae) has clear homology to DNA ligases, although
there is no evidence of ligase activity. The removal of the RFC1 N terminus re-
sults in sensitivity to DNA-damaging agents (162), but this region is not necessary
for in vitro clamp loading (163) or cell viability (162). The C-terminal domain
(residues 660861 in S. cerevisiae) has not yet been characterized genetically or
biochemically.
The five-subunit composition of RFC bears a close similarity to the minimal E. coli 3
complex (85). On the basis of subunit interactions and sequence
similarity to complex subunits, the subunit arrangement of the RFC pentamer
was proposed (4, 164) and has since been proven by structure analysis (85). RFC1
shares characteristics of the subunit in its conserved clamp-interacting residues
and position in the pentamer (163, 165). RFC24 are considered -like in their
ability to form a trimeric ATPase subassembly (166). RFC5 is in the position of,
and like it contains an SRC motif but lacks a consensus phosphate-binding loop
(P-loop). Like the 3 complex, the ATP sites of RFC are located at subunit
interfaces. However, unlike 3 , RFC contains four competent ATPase sitesbecause RFC1 also binds ATP where does not. In fact, RFC5 also binds a
nucleotide in the crystal structure.
During the RFC ATPase cycle in S. cerevisiae, the complex initially binds
two ATP, then a third upon PCNA binding, and a fourth when it locates the
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
19/36
CELLULAR REPLICASES 301
TABLE 2 Eukaryotic replisome components
Replisome
component
Saccharomycescerevisiae
(kDa) Human (kDa)
Function and remarks[Schizosaccharomyces pombe
name (S.p.)]
RFCa RFC (277.7)a RFC (314.9)a Pentameric clamp loadera
RFC1 (94.9) p140 (128.2) Binds ATP; phosphorylated
RFC2 (39.7) p37 (39.2) Binds ATP
RFC3 (38.2) p36 (40.6) Binds ATP
RFC4 (36.1) p40 (39.7) Binds ATP
RFC5 (39.9) p38 (38.5) Binds ATP or ADP
PCNA PCNA (28.9) PCNA (28.7) 87 kDa homotrimeric processivity
sliding clampa
Pol a Pol (220.2)a Pol (238.7)a Replicative DNA polymerasea
Pol3 (124.6) p125 (123.6) DNA polymerase, 3-5 exonuclease,
binds PCNA; subunit A (S.p. Pol3)
Pol31 (55.3) p50 (51.3) Structural subunit; subunit B (S.p. Cdc1)
Pol32 (40.3) p66 (51.4) Binds PCNA; subunit C (S.p. Cdc27);
binds Pol large subunit
p12 (12.4) Structural, stimulates processivity; subunit
D (S.p. Cdm1)
Pol a Pol (378.7)a Pol (350.3)a Replicative DNA polymerasea
Pol2 (255.7) p261 (261.5) DNA polymerase, 3
-5
exonuclease(S.p. Pol2/cdc20)
Dpb2 (78.3) p59 (59.5) Binds polymerase subunit (S.p. Dpb2)
Dpb3 (22.7) p17 (17.0) Binds Dpb4
Dpb4 (22.0) p12 (12.3) Present in ISW2/yCHRAC chromatin-
remodeling complex (S.p. Dpb4)
Pol a Pol (355.6)a Pol (340.6)a DNA polymerase/primasea
Pol1 (166.8) p180 (165.9) DNA polymerase
Pol12 (78.8) p68 (66.0) Structural subunit
Pri2 (62.3) p55 (58.8) Interacts tightly with p48
Pri1 (47.7) p48 (49.9) RNA primase catalytic subunit
MCMa MCM (605.6)a MCM (535)a Putative 3-5 replicative helicasea
Mcm2 (98.8) Mcm2 (91.5) Phosphorylated by Dbf4-dependent kinase
Mcm3 (107.5) Mcm3 (91.0) Ubiquitinated, acetylated
Mcm4 (105.0) Mcm4 (96.6) Helicase with MCM6,7; phosphorylated
by CDK; aka Cdc54
Mcm5 (86.4) Mcm5 (82.3) Aka Cdc46; Bob1 is a mutant form
Mcm6 (113.0) Mcm6 (92.3) Helicase with MCM4,7
Mcm7 (94.9) Mcm7 (81.3) Helicase with MCM4,6; ubiquitinated
RPAa
RPA (114)a
RPA (100.5)a
Single-stranded DNA-binding proteina
RPA70 (70.3) RPA70 (70.3) Binds DNA, stimulates Pol
RPA30 (29.9) RPA30 (29) Binds RPA70 and 14, phosphorylated
RPA14 (13.8) RPA14 (13.5) Binds RPA30
aInformation concerns a protein complex.
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
20/36
302 JOHNSON ODONNELL
primer/template DNA, which triggers ATP hydrolysis, causing RFC to eject and
leave the closed PCNA ring on DNA (167). This mechanism excludes the route
of RFC first encountering DNA and subsequently recruiting PCNA. ATP site mu-
tational studies have demonstrated that, in general, disrupting any one of the fourconsensus ATP sites has a significant effect on the activity of RFC (168170),
although ATP binding may suffice to rescue this defect in at least one ATP site (in
RFC4 for S. cerevisiae). These single P-loop mutants still bind PCNA readily but
are deficient in DNA binding (170).
A crystal structure of an RFC-PCNA complex has been solved (85). To promote
stability of this nucleotide-dependent interaction by preventing hydrolysis, ATPS
was used, and RFC was produced with R to Q mutations in the SRC motifs. The
RFC-ATPS-PCNA crystal structure has provided a detailed view of how a clamp
loader interacts with its cognate clamp (see Figure 6b). In addition, the structure hasrevealed how DNA binds to a clamp loader. The main PCNA contacts occur through
RFC1 and RFC3, which bind to the hydrophobic pocket between the domains of
two different PCNA protomers. RFC1 interacts extensively with PCNA, whereas
RFC3 seems to be only partially engaged with PCNA and the ring is closed. In
the RFC-ATPS-PCNA structure, RFC2 and RFC5 do not bind PCNA at all. This
suboptimal interaction of RFC with PCNA may explain why the clamp remains
closed and only slightly perturbed from its unbound structure. Alternatively, the
closed PCNA may be due to crystal packing forces, instability of an open-ring
complex, or the arginine to glutamine mutation in the four SRC motifs. Althoughthis mutation should ensure that no ATPS becomes hydrolyzed, it may have
prevented or perturbed some necessary RFC subunit-PCNA interaction needed for
clamp opening.
The way RFC binds DNA is suggested by the helical arrangement of RFC sub-
units in the structure, with the helical axis passing through the central channel of
the closed PCNA (see Figure 6c,e). The helix begins at RFC1, the subunit that is
fully engaged with the hydrophobic pocket on one PCNA protomer. Adjacent to
RFC1 is RFC4, displaced by the helical operator of 61 rotation and 5.5 A trans-
lation. Overall, the helical operations that relate all five subunits have a pitch of5.6 A per 60 rotation, ending with RFC5, which lies 25 A above PCNA and is
significantly separated from the ATPase domain of RFC1. This right-handed helix
and pitch mimics that of duplex B-form DNA. Furthermore, there is sufficient
space in the center of RFC to model a DNA duplex, which passes right through
the center of PCNA. Each RFC subunit has two -helices that are oriented so their
positive dipole tracks the minor groove phosphate backbone of DNA modeled into
the structure (Figure 6e). Several basic residues on these helices are conserved in
E. coli and eukaryotic clamp loaders, consistent with the idea that DNA binds
within this central area of RFC. Modeling a 3
-recessed, primed template from an-other crystal structure into the center of the ring shows that the specific recognition
of a primer/template by RFC might occur by a simple clash of the primed-template
junction with the domain III cap of the RFC subunits. A stiff duplex DNA could
not proceed through the RFC structure because it would hit the cap and could
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
21/36
CELLULAR REPLICASES 303
not bend to exit out the side of RFC. The flexible ssDNA of a primed junction
may bend to exit through the gap between RFC1 and 5, perhaps assisted by a
positive patch on the RFC1 surface (see Figure 6e). This hypothesis of specific
primer/template recognition is in agreement with recent biophysical evidence ofRFC/DNA association (171, 172).
Eukaryotic DNA Polymerases
Many DNA polymerases are now known to function in the eukaryotic cell. These
polymerases all share a common catalytic mechanism [reviewed in (11)], but most
serve a specific function outside of basic genome duplication. DNA polymerases
, , and are the established replicative polymerases and are thought to function
together at the replication fork to copy genomic DNA in a semi-discontinuousmanner. Pol , , and are all members of the B-family of DNA polymerases (45).
DNA POLYMERASE DNA polymerase (Pol ) was initially thought to be the
main replicase before Pol was discovered. Pol is unique in its ability to initiate
DNA synthesis by first synthesizing its own12-nucleotide RNA primer and then
extending it with about 20 bases of DNA (173, 174) [reviewed in (13)]. Pol is
now thought to be the eukaryotic primase, making a hybrid RNA/DNA primer,
followed by a polymerase switch allowing the replicase to take over elongation
(175). The switch is mediated by RFC, which displaces Pol from the primer viacompetition for RPA (176, 177).
Pol consists of four subunits (13). The DNA polymerase activity is found in
the largest subunit (p180), and primase activity is located in the smallest subunit
(p48). The exact functions of the middle two subunits are not clear, but all four
subunits are present in Pol isolated from yeast, human,Xenopus,andDrosophila.
DNA POLYMERASE DNA polymerase (Pol ) in fission yeast, human, and other
eukaryotic organisms is composed of four essential subunits (178, 179). Interest-
ingly, in S. cerevisiae, Pol has only three subunits, and no apparent homologueexists for the fourth subunit (180). Furthermore, the third subunit can be deleted
from budding yeast, although cell growth is compromised (181).
A unified subunit nomenclature for Pol has recently been proposed (182). On
the basis ofSchizosaccharomyces pombe Pol subunits and their homologues in
other eukaryotes, the subunits have been renamed AD for the Pol3(A), Cdc1(B),
Cdc27(C), andCdm1(D) polypeptides. Much debate has centered on the possibility
of the Pol complex self-associating into a dimer. The most recent evidence in
multiple systems indicates that Pol complex contains only one copy of each
subunit across a wide range of concentrations and has an elongated shape thatresulted in the earlier confusion over whether it was a dimeric polymerase particle
(183, 184). Pol subcomplexes can also be isolated as a core A/B dimer (180,
185). A zinc-finger module in subunit A interacts with subunit B, which acts as a
bridge to subunits C and D (when present) (183, 186, 187). The polymerase and
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
22/36
304 JOHNSON ODONNELL
3-5 exonuclease activities of Pol are present in the large A subunit. The Pol
polymerase extends DNA with high fidelity, but the exonuclease seems surprisingly
ineffective in vitro in S. pombe (188) and may be most important when Pol is
clamped onto DNA by PCNA (189, 190).Pol associates with PCNA via interactions with at least two of its subunits.
The clamp is positioned behind the polymerase (191), preventing polymerase
dissociation as it extends a primer (192). The strongest interaction is between
PCNA and subunit C (193). Subunit C contains the consensus PCNA interaction
motif, and studies in yeast show that this PCNA-interacting subunit stimulates Pol
(183, 194). Surprisingly, stimulation is not dependent upon the PCNA-interacting
motif but mainly on the domain involved in connection to the A/B complex. The
A/B heterodimer alone can also associate with PCNA, presumably through subunit
A (180). The stimulation of subunit C on Pol activity might be due to proteincomplex stabilization rather than catalytic enhancement. Even the small D subunit
from the human four-subunit Pol complex significantly stimulated the A/B/C
subcomplex in assays with RFC and PCNA (179).
DNA POLYMERASE (POL ) Studies showing Pol is essential in yeast placed it at
the replication fork (195). Indeed, chromatin immunoprecipitation (ChIP) assays
in yeast indicate that Pol is located at origins prior to S phase (196) and moves
away from origins upon releasing an S phase block (197), consistent with a role
in chromosome replication. However, there are some conflicting genetic studieson whether the intrinsic DNA polymerase activity is required for replication or
whether the protein may instead serve another role, either as a DNA sensor or
checkpoint protein (198200), or perhaps it holds together other proteins that are
essential in the replisomal particle (201). Although further work will be required to
fully understand the exact role of Pol , it is widely believed to be directly involved
in DNA synthesis at the replication fork.
A recent report on Pol concludes that it is a heterotetramer with a stoichiometry
of 1:1:1:1, presumably in all eukaryotes (202). The DNA polymerase and 3 -5
exonuclease of Pol reside in the largest subunit and appear to have a higherfidelity than Pol /PCNA (203). The third-largest subunit, Dpb4, is also a member
of a complex that appears to be involved in chromatin remodeling (204) (A. Tackett
and B. Chait, personal communication). Pol activity does not absolutely require
PCNA, but PCNA stimulation increases as the ionic strength is raised (205207).
The PCNA interaction motif on the large subunit of Pol is not essential for cell
viability, but mutational analysis of this sequence suggests a role in DNA repair
(208).
The Eukaryotic Replisome
The essential nature of both Pol and Pol and the fact that they both function
with PCNA as monomeric polymerase particles have led to renewed proposals
that they function together to replicate the leading and lagging strands. Xenopus
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
23/36
CELLULAR REPLICASES 305
extracts have shown that depletion of either polymerase results in a dramatic drop
in replication (209, 210). Depletion of Pol gave rise to many short nascent
strands that may have been improperly elongated Okazaki fragments, whereas
Pol depletion merely slowed down the replication machinery. Viability of Pol-deleted strains of yeast suggests that Pol is not essential for cell viability (198,
201, 211). However, an inactive point mutant of Pol resulted in cell death (212).
Onepossible explanation for this apparent contradiction is that another polymerase,
presumably Pol , can carry out DNA replication in the complete absence of Pol
but that the Pol point mutant acted as a dominant negative to stop the fork.
Intensive studies of in vitro replication from the SV40 origin have provided
great insight into eukaryotic fork function, but the SV40 T antigen fills many roles
ordinarily performed by host proteins, including helicase function (213). Study
of polymerases in this system show that Pol and Pol , and even Pol aloneunder some conditions, are sufficient for replication. However, the small size of
the genome and use of T antigen may minimize the requirements for replication
[reviewed in (152)]. Reconstitution of a eukaryotic replisome on a rolling-circle
template may greatly facilitate our understanding of the roles of Pol and Pol
and which strand(s) they operate on.
The 10-fold-smaller size of eukaryotic lagging-strand fragments (200 bp)
compared to E. coli is counterbalanced by the 10-fold-slower rate of fork move-
ment. Stoichiometric use of PCNA clamps during lagging-strand replication has
not been demonstrated but is presumed to occur as it does with the E. coli clamp. That PCNA clamps are left behind on lagging-strand fragments is implied
by the fact that PCNA interacts with and, in some cases, stimulates the factors
necessary for Okazaki fragment maturation (214216). There is also an interest-
ing observation that PCNA-DNA complexes may persist through mitosis, marking
chromosomes for epigenetic inheritance (217, 218).
The eukaryotic replisome factors that contain helicase, primase, and SSB are
each composed of multi-protein assemblies in eukaryotes [reviewed in (9)]. This
complexity is in contrast to the single-subunit factors in E. coli. For example, even
the SSB in eukaryotes, termed RPA, is composed of three subunits in a 1:1:1 het-erotrimer [reviewed in (14)]. It is widely believed that the eukaryotic helicase is the
heterohexameric MCM2-7 complex [reviewed in (12)], although this conclusion
is not yet firm. Like E. coli DnaB, the hexameric MCM complex is ring shaped,
but each MCM subunit is a different polypeptide. Subunit arrangements for the
MCM2-7 complex have been proposed (219, 220). Helicase activity has been ob-
served only for the MCM4/6/7 subcomplex (221223), yet all six MCM genes are
essential in a variety of systems (224) [also, see references within (12)]. These
findings have led to the suggestion that one or more of the MCM2,3,5 subunits
act as regulators (219, 222, 225). Consistent with this view, MCM2 inhibits theMCM4/6/7 helicase. The MCM complex is also a target of phosphorylation and
ubiquitination and is thought to require activation for helicase action after assembly
on DNA. MCM subunits are AAA+ proteins and thus are thought to have an evo-
lutionary origin distinct from DnaB, which is constructed from the RecA module.
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
24/36
306 JOHNSON ODONNELL
Figure 7 Hypothetical arrangement of proteins at the eukaryotic replication fork. The
hexameric MCM complex encircles the leading strand. In this cartoon, Pol is placed
on the leading strand and Pol on the lagging, with RFC bridging the two polymerases
and helicase. Pol /primase action places it on the lagging strand along with RPA bound
to the looping single-stranded DNA. Other factors involved in replication and known
to bind certain proteins at the replication fork include Cdc45, Sld2, Sld3, Dpb11, and
the heterotetrameric GINS complex.
Furthermore, these helicases translocate on DNA with opposite polarities (221
223), thereby placing the MCMs on the leading strand (see Figure 7). However,
like DnaB, the MCMs have been shown to be capable of encircling two DNA
strands (223, 226), and evidence that they may form a double hexamer exists in
both Archaea (227, 228) and S. pombe (229) systems. One line of evidence for
this comes from an archaeal single-gene MCM that produces a circular double
hexamer with helicase activity (227).MCM helicase activity is rather weak, somewhat reminiscent of the relatively
weak helicase activity ofE. coli DnaB. As described earlier, DnaB becomes highly
active when coupled to Pol III holoenzyme, and this coupling occurs through the
subunit of the clamp loader (111). It seems likely that a similar arrangement may
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
25/36
CELLULAR REPLICASES 307
exist in eukaryotes. In the scheme of Figure 7, RFC is proposed to act as a scaffold
similar to the / complex in E. coli, bridging the two polymerases and coupling
them to the helicase. The proposed connections are simply inferred because only
scant evidence exists for an RFC-Pol interaction (176, 177), and none as yetexists for RFC binding either Pol or the MCM complex. However, evidence
exists for the presence of many other protein actors required for DNA synthesis
and presumed to act at the eukaryotic replication fork. Although the individual
functions of these factors are largely unknown, protein interaction studies indicate
a network as illustrated schematically in Figure 7. Cdc45 has been known for some
time to be required for replication, and interaction between Cdc45 and MCMs has
been documented (230, 231). Newer actors include Sld3, which binds Cdc45 (17),
and the heterotetramer GINS complex, which appears to have a ring shape (15,
232). The GINS complex also appears to bind Pol , and similar to Sld3, assemblesat origins just prior to DNA synthesis (15). Dpb11 is thought to bind both Pol
and Pol (233), and it also forms a complex with Sld2 (16). Biochemical study
of these various factors, alone and in combinations, will be required to understand
their individual roles in chromosome replication and to determine whether they all
function together at each replication fork.
Alternate RFC Complexes and Other Roles for Clamp Loaders
Sliding clamps are used in a variety of DNA metabolic processes (21, 23), and one
may presume that their respective clamp loader is also involved in most of these
processes. In addition, the subunit composition of the eukaryotic clamp loader is
altered to perform novel functions in DNA metabolism. This alteration involves
the use of an alternate clamp-loader subunit, in place of RFC1. In fission yeast
and human, the Rad17 subunit (Rad24 in S. cerevisiae) replaces p140 (RFC1) in
complex with RFC25 (25, 234). The Rad17-RFC complex is involved in the DNA
damage checkpoint response, along with a novel PCNA-like sliding clamp formed
from the trimer Rad9/Rad1/Hus1 (Ddc1/Rad17/Mec3 in S. cerevisiae) [reviewed
in (235)]. This clamp loader loads the 911 clamp onto primed DNA in an RPA-
stimulated reaction (236239) and does so with opposite polarity to RFC (238).
The function of the 911 clamp is not clear, but it presumably recruits other factors
when loaded on DNA. There may be a 3-5 exonuclease activity in Rad9 or Hus1,
thus providing a biochemical activity for the ring itself (240). RFC1 can also be
replaced by Ctf18/Chl12 (26, 241) or Elg1 (27, 242, 243), which are involved in
cohesion and genome stability, respectively, although the specific roles of these
complexes are not understood.
CONCLUDING REMARKS
Each passing year brings significant advances in our understanding of replica-
tion fork mechanisms. However, for each question answered, 10 more crop up.
Even in the relatively well-defined and intensively studied prokaryotic system
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
26/36
308 JOHNSON ODONNELL
there remain numerous questions. We still do not know how the proteins are truly
arranged at the replication fork. How are the two polymerases oriented, do they
face the same direction? Does the helicase encircle one or more strands? Does
it act as a double hexamer? How is the helicase positioned on the replicase?How is the length of Okazaki fragments measured and regulated? What hap-
pens when the replisome encounters a lesion on the leading or lagging strand?
Does the helicase continue to unwind DNA, and if so, how far does it go be-
fore it stops? To advance past a lesion, E. coli mounts the SOS response and
gets past the lesion by recombinative repair processes, collectively termed repli-
cation restart [reviewed in (113, 114)]. How does the replication machinery
coordinate its action with those of repair and recombination? How does the repli-
some deal with transcribing RNA polymerase and other proteins tightly bound to
DNA?The sliding clamps in both prokaryotes and eukaryotes interact with many dif-
ferent DNA polymerases and repair proteins. For example, and PCNA both
interact with mismatch repair and excision repair proteins. What is the role of the
clamps in these processes? The clamps also bind a variety of specialized DNA
polymerases such as those of the Y-family, which have low fidelity but can by-
pass certain lesions. How do these polymerases trade places with the replicase at
the right time and place to bypass lesions? How is the use of the clamp by these
low-fidelity enzymes restricted and handed back to the high-fidelity replicase af-
ter a lesion is bypassed? PCNA appears to be modified by ubiquitination and/orsumoylation to assist this process. How do these modifications control the traf-
ficking of different polymerases on PCNA? Eukaryotes use a variety of alternate
clamp loaders in which RFC1 is replaced by another protein. These RFC1 sub-
stitutes appear to be involved in the DNA damage response, chromatin cohesion,
and genome stability. How do these alternate clamp loaders function and do they
interface with the normal replication machinery? Clearly much work remains to
be done to address these important issues.
Knowledge of the eukaryotic replisome composition is rapidly becoming a
complex challenge. How many more factors are there? How do they connect andwhat are their individual functions? Do all these proteins function within the same
replisome or are there various types of eukaryotic replisomes, specialized for
different regions of the chromosome or for different times in S phase? How is
replisome assembly and function regulated at the start of S phase? How does this
tie in with growth factors that work at the cell surface, and the complex kinase
and cell cycle control networks that reach to the nucleus? How do DNA damage
checkpoint and other checkpoint mechanisms exert their influence during ongoing
S-phase events?
The authors hope this review highlighted several important advances in ourknowledge of replicase structure and function. However, these concluding remarks
underscore how much is yet to be learned and perhaps provide some sense of how
far we have yet to go.
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
27/36
CELLULAR REPLICASES 309
ACKNOWLEDGMENTS
The authors are grateful to the following individuals for helpful comments and
discussions: Greg Bowman, Brian Chait, Megan Davey, David Jeruzalmi, John
Kuriyan, and Alan Tackett. We are also grateful to Nina Yao for providing helpwith artwork. This work was supported by a grant from the National Institutes of
Health (GM38839) and by the Howard Hughes Medical Institute.
The Annual Review of Biochemistry is online at
http://biochem.annualreviews.org
LITERATURE CITED
1. Kornberg A, Baker T. 1992. DNA Repli-
cation. New York: Freeman. 931 pp. 2nd
ed.
2. Krishna TS, Kong XP, Gary S, Burg-
ers PM, Kuriyan J. 1994. Cell 79:1233
43
3. Kong XP, Onrust R, ODonnell M,
Kuriyan J. 1992. Cell 69:42537
4. ODonnell M, Jeruzalmi D, Kuriyan J.
2001. Curr. Biol. 11:R93546
5. Marians KJ. 1992. Annu. Rev. Biochem.
61:673719
6. Kelman Z, ODonnell M. 1995. Annu.
Rev. Biochem. 64:171200
7. Benkovic SJ, Valentine AM, Salinas F.
2001. Annu. Rev. Biochem. 70:181208
8. McHenry CS. 2003. Mol. Microbiol. 49:
115765
9. Waga S, Stillman B. 1998. Annu. Rev.
Biochem. 67:7215110. Burgers PM. 1998. Chromosoma 107:
21827
11. Hubscher U, Maga G, Spadari S. 2002.
Annu. Rev. Biochem. 71:13363
12. Forsburg SL. 2004. Microbiol. Mol. Biol.
Rev. 68:10931
13. Lehman IR, Kaguni LS. 1989. J. Biol.
Chem. 264:426568
14. Wold MS. 1997. Annu. Rev. Biochem. 66:
619215. Takayama Y, Kamimura Y, Okawa M,
Muramatsu S, Sugino A, Araki H. 2003.
Genes Dev. 17:115365
16. Kamimura Y, Masumoto H, Sugino A,
Araki H. 1998. Mol. Cell. Biol. 18:6102
9
17. Kamimura Y, Tak YS, Sugino A, Araki
H. 2001. EMBO J. 20:2097107
18. Hoege C, Pfander B, Moldovan GL, Py-
rowolakis G, Jentsch S. 2002. Nature
419:13541
19. Haracska L, Torres-Ramos CA, Johnson
RE, Prakash S, Prakash L. 2004. Mol.
Cell. Biol. 24:426774
20. Kannouche PL, Wing J, Lehmann AR.
2004. Mol. Cell 14:491500
21. Warbrick E. 2000. BioEssays 22:997
1006
22. Lopez de Saro FJ, ODonnell M. 2001.
Proc. Natl. Acad. Sci. USA 98:837680
23. Maga G, Hubscher U. 2003. J. Cell Sci.
116:305160
24. Lopez de Saro FJ, Georgescu RE, Good-
man MF, ODonnell M. 2003. EMBO J.22:640818
25. Lindsey-Boltz LA, Bermudez VP, Hur-
witz J, Sancar A. 2001. Proc. Natl. Acad.
Sci. USA 98:1123641
26. Naiki T, Kondo T, Nakada D, Matsumoto
K, Sugimoto K. 2001. Mol. Cell. Biol. 21:
583845
27. Bellaoui M, Chang M, Ou J, Xu H, Boone
C, Brown GW. 2003. EMBO J. 22:4304
1328. Grabowski B, Kelman Z. 2003.Annu. Rev.
Microbiol. 57:487516
29. Tabor S, Huber HE, Richardson CC.1987.
J. Biol. Chem. 262:1621223
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
28/36
310 JOHNSON ODONNELL
30. Huber HE,Tabor S, Richardson CC. 1987.
J. Biol. Chem. 262:1622432
31. Klemperer N, McDonald W, Boyle K,
Unger B, Traktman P. 2001. J. Virol. 75:
12298307
32. McDonald WF, Klemperer N, Traktman
P. 1997. Virology 234:16875
33. Hernandez TR, Lehman IR. 1990. J. Biol.
Chem. 265:1122732
34. Gottlieb J, Marcy AI, Coen DM, Chall-
berg MD. 1990. J. Virol. 64:597687
35. Kaguni LS. 2004. Annu. Rev. Biochem.
73:293320
36. Kelly TJ, Brown GW. 2000. Annu. Rev.Biochem. 69:82980
37. Bell SP, Dutta A. 2002. Annu. Rev.
Biochem. 71:33374
38. Kornberg T, Gefter ML. 1971. Proc. Natl.
Acad. Sci. USA 68:76164
39. McHenry CS, Crow W. 1979. J. Biol.
Chem. 254:174853
40. Studwell-Vaughan PS, ODonnell M.
1993. J. Biol. Chem. 268:1178591
41. Maki H, Kornberg A. 1985.J. Biol. Chem.260:1298792
42. Studwell PS, ODonnell M. 1990. J. Biol.
Chem. 265:117178
43. Taft-Benz SA, Schaaper RM. 2004. J.
Bacteriol. 186:277480
44. Slater SC, Lifsics MR, ODonnell M,
Maurer R. 1994. J. Bacteriol. 176:815
21
45. Braithwaite DK,Ito J. 1993.Nucleic Acids
Res. 21:78780246. Steitz TA. 1999. J. Biol. Chem. 274:
1739598
47. Kim DR, McHenry CS. 1996. J. Biol.
Chem. 271:20699704
48. Kim DR, McHenry CS. 1996. J. Biol.
Chem. 271:2069098
49. Hamdan S, Carr PD, Brown SE, Ollis DL,
Dixon NE. 2002. Structure 10:53546
50. Keniry MA, Berthon HA, Yang JY, Miles
CS, Dixon NE. 2000. Protein Sci. 9:72133
51. DeRose EF, Darden T, Harvey S, Gabel
S, Perrino FW, et al. 2003. Biochemistry
42:363544
52. Perrino FW, Harvey S, McNeill SM.
1999. Biochemistry 38:160019
53. Brenowitz S, Kwack S, Goodman MF,
ODonnell M, Echols H. 1991. J. Biol.
Chem. 266:788892
54. Miller H, Perrino FW. 1996.Biochemistry
35:1291925
55. Fay PJ, Johanson KO, McHenry CS,Bam-
bara RA. 1981. J. Biol. Chem. 256:976
83
56. LaDuca RJ, Fay PJ, Chuang C, McHenry
CS, Bambara RA. 1983.Biochemistry 22:
517788
57. Stukenberg PT, Studwell-Vaughan PS,ODonnell M. 1991. J. Biol. Chem. 266:
1132834
58. Kuwabara N, Uchida H. 1981. Proc. Natl.
Acad. Sci. USA 78:576467
59. LaDuca RJ, Crute JJ, McHenry CS, Bam-
bara RA. 1986. J. Biol. Chem. 261:7550
57
60. Reems JA, Wood S, McHenry CS. 1995.
J. Biol. Chem. 270:560613
61. Yao N, Leu FP, Anjelkovic J, TurnerJ, ODonnell M. 2000. J. Biol. Chem.
275:1144050
62. Naktinis V, Onrust R, Fang L, ODonnell
M. 1995. J. Biol. Chem. 270:1335865
63. Naktinis V, Turner J, ODonnell M. 1996.
Cell 84:13745
64. Leu FP, Hingorani MM, Turner J,
ODonnell M. 2000. J. Biol. Chem. 275:
3460918
65. Stewart J, Hingorani MM, Kelman Z,ODonnell M. 2001. J. Biol. Chem. 276:
1918289
66. Turner J, Hingorani MM, Kelman Z,
ODonnell M. 1999. EMBO J. 18:771
83
67. Jeruzalmi D, Yurieva O, Zhao Y, Young
M, Stewart J, et al. 2001. Cell 106:41728
68. Indiani C, ODonnell M. 2003. J. Biol.
Chem. 278:4027281
69. Pritchard AE, Dallmann HG, Glover BP,McHenry CS. 2000. EMBO J. 19:6536
45
70. Jeruzalmi D, ODonnell M, Kuriyan J.
2001. Cell 106:42941
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
29/36
CELLULAR REPLICASES 311
71. Hingorani MM, ODonnell M. 1998. J.
Biol. Chem. 273:2455063
72. Tsuchihashi Z, Kornberg A. 1989.J. Biol.
Chem. 264:1779095
73. Xiao H, Naktinis V, ODonnell M. 1995.
J. Biol. Chem. 270:1337883
74. Guenther B, Onrust R, Sali A, ODonnell
M, Kuriyan J. 1997. Cell 91:33545
75. Onrust R, ODonnell M. 1993. J. Biol.
Chem. 268:1176672
76. Xiao H, Dong Z, ODonnell M. 1993. J.
Biol. Chem. 268:1177984
77. Kelman Z, Yuzhakov A, Andjelkovic J,
ODonnell M. 1998. EMBO J. 17:243649
78. Glover BP, McHenry CS. 1998. J. Biol.
Chem. 273:2347684
79. Olson MW, Dallmann HG, McHenry CS.
1995. J. Biol. Chem. 270:2957077
80. Ason B, Handayani R, Williams CR,
Bertram JG, Hingorani MM, et al. 2003.
J. Biol. Chem. 278:1003340
81. Bloom LB, Turner J, Kelman Z, Beechem
JM, ODonnell M, Goodman MF. 1996.J. Biol. Chem. 271:30699708
82. Ason B, Bertram JG, Hingorani MM,
Beechem JM, ODonnell M, et al. 2000.
J. Biol. Chem. 275:300615
83. Bertram JG, Bloom LB, Hingorani MM,
Beechem JM, ODonnell M, Goodman
MF. 2000. J. Biol. Chem. 275:28413
20
84. Williams CR, Snyder AK, Kuzmic P,
ODonnell M, Bloom LB. 2004. J. Biol.Chem. 279:437685
85. Bowman GD, ODonnell M, Kuriyan J.
2004. Nature 429:72430
86. Neuwald AF, Aravind L, Spouge JL,
Koonin EV. 1999. Genome Res. 9:2743
87. Davey MJ, Jeruzalmi D, Kuriyan J,
ODonnell M. 2002. Nat. Rev. Mol. Cell
Biol. 3:82635
88. Dong Z, Onrust R, Skangalis M,
ODonnell M. 1993. J. Biol. Chem. 268:1175865
89. Johnson A, ODonnell M. 2003. J. Biol.
Chem. 278:1440613
90. Snyder AK, Williams CR, Johnson A,
ODonnell M, Bloom LB. 2004. J. Biol.
Chem. 279:438693
91. Song MS, McHenry CS. 2001. J. Biol.
Chem. 276:4870915
92. Podobnik M, Weitze TF, ODonnell M,
Kuriyan J. 2003. Structure 11:25363
93. Goedken ER, Levitus M, Johnson A,
Bustamante C, ODonnell M, Kuriyan J.
2004. J. Mol. Biol. 336:104759
94. Leu FP, ODonnell M. 2001. J. Biol.
Chem. 276:4718594
95. Wu YH, Franden MA, Hawker JR Jr,
McHenry CS. 1984. J. Biol. Chem. 259:
121172296. Flower AM, McHenry CS. 1986. Nucleic
Acids Res. 14:8091101
97. Tsuchihashi Z, Kornberg A. 1990. Proc.
Natl. Acad. Sci. USA 87:251620
98. Gao D, McHenry CS. 2001.J. Biol. Chem.
276:444146
99. Gao D, McHenry CS. 2001.J. Biol. Chem.
276:443340
100. Blinkova A, Hervas C, Stukenberg PT,
Onrust R, ODonnell ME, Walker JR.1993. J. Bacteriol. 175:601827
101. Onrust R, Finkelstein J, Turner J, Nakti-
nis V, ODonnell M. 1995. J. Biol. Chem.
270:1336677
102. Glover BP, McHenry CS. 2000. J. Biol.
Chem. 275:301720
103. Yuzhakov A, Turner J, ODonnell M.
1996. Cell 86:87786
104. McHenry CS, Johanson KO. 1984. Adv.
Exp. Med. Biol. 179:31519105. Johanson KO, McHenry CS. 1984.J. Biol.
Chem. 259:458995
106. McHenry CS. 1988. Biochim. Biophys.
Acta 951:24048
107. Wu CA, Zechner EL, Hughes AJ Jr, Fran-
den MA, McHenry CS, Marians KJ. 1992.
J. Biol. Chem. 267:406473
108. Dallmann HG, Thimmig RL, McHenry
CS. 1995. J. Biol. Chem. 270:2955562
109. Glover BP, McHenry CS. 2001. Cell105:92534
110. Kim S, Dallmann HG, McHenry CS, Mar-
ians KJ. 1996. J. Biol. Chem. 271:21406
12
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
30/36
312 JOHNSON ODONNELL
111. Kim S, Dallmann HG, McHenry
CS, Marians KJ. 1996. Cell 84:643
50
112. Dallmann HG, Kim S, Pritchard AE, Mar-
ians KJ, McHenry CS. 2000. J. Biol.
Chem. 275:1551219
113. Kowalczykowski SC. 2000. Trends Bio-
chem. Sci. 25:15665
114. Marians KJ. 2004. Philos. Trans. R. Soc.
London Ser. B 359:7177
115. Grompone G, Seigneur M, Ehrlich SD,
Michel B. 2002. Mol. Microbiol. 44:
133139
116. Zechner EL, Wu CA, Marians KJ. 1992.J. Biol. Chem. 267:404553
117. Zechner EL, Wu CA, Marians KJ. 1992.
J. Biol. Chem. 267:405463
118. Tougu K, Marians KJ. 1996. J. Biol.
Chem. 271:21398405
119. Tougu K, Marians KJ. 1996. J. Biol.
Chem. 271:2139197
120. Wu CA,Zechner EL, Reems JA, McHenry
CS, Marians KJ. 1992. J. Biol. Chem.
267:407483121. Stayton MM, Kornberg A. 1983. J. Biol.
Chem. 258:1320512
122. Sun W, Godson GN. 1996. J. Bacteriol.
178:67015
123. Sun W, Godson GN. 1998. J. Mol. Biol.
276:689703
124. Yuzhakov A, Kelman Z, ODonnell M.
1999. Cell 96:15363
125. Sinha NK, Morris CF, Alberts BM. 1980.
J. Biol. Chem. 255:429093126. Chastain PD 2nd, Makhov AM, Nos-
sal NG, Griffith J. 2003. J. Biol. Chem.
278:2127685
127. ODonnell ME. 1987. J. Biol. Chem.
262:1655865
128. Stukenberg PT, Turner J, ODonnell M.
1994. Cell 78:87787
129. Li X, Marians KJ. 2000. J. Biol. Chem.
275:3475765
130. Leu FP, Georgescu R, ODonnell M.2003. Mol. Cell 11:31527
131. Wu CA, Zechner EL, Marians KJ. 1992.
J. Biol. Chem. 267:403044
132. Higuchi K, Katayama T, Iwai S, Hidaka
M,HoriuchiT,MakiH.2003. Genes Cells
8:43749
133. McInerney P, ODonnell M. 2004.J. Biol.
Chem. 279:2154351
134. Lopez de Saro FJ, Georgescu RE,
ODonnell M. 2003. Proc. Natl. Acad. Sci.
USA 100:1468994
135. Bunting KA, Roe SM, Pearl LH. 2003.
EMBO J. 22:588392
136. Burnouf DY, Olieric V, Wagner J, Fujii
S, Reinbolt J, et al. 2004. J. Mol. Biol.
335:118797
137. Duzen JM, Walker GC, Sutton MD. 2004.
DNA Repair3:30112138. Miyachi K, Fritzler MJ, Tan EM. 1978. J.
Immunol. 121:222834
139. Tan CK, Castillo C, So AG, Downey
KM. 1986. J. Biol. Chem. 261:12310
16
140. Prelich G, Kostura M, Marshak DR,
Mathews MB, Stillman B. 1987. Nature
326:47175
141. Prelich G, Tan CK, Kostura M, Mathews
MB, So AG, et al. 1987. Nature 326:51720
142. Tsurimoto T, Stillman B. 1990. Proc.
Natl. Acad. Sci. USA 87:102327
143. Yao N, Turner J, Kelman Z, Stukenberg
PT, Dean F, et al. 1996. Genes Cells 1:
10113
144. Mossi R, Jonsson ZO, Allen BL, Hardin
SH, Hubscher U. 1997. J. Biol. Chem.
272:176976
145. Oku T, Ikeda S, Sasaki H, Fukuda K,Morioka H, et al. 1998. Genes Cells 3:
35769
146. Gulbis JM, Kelman Z, Hurwitz J,
ODonnell M, Kuriyan J. 1996. Cell 87:
297306
147. Shamoo Y, Steitz TA. 1999. Cell 99:155
66
148. Franklin MC, Wang J, Steitz TA. 2001.
Cell 105:65767
149. Hingorani MM, ODonnell M. 2000.Curr. Biol. 10:R2529
150. Dalrymple BP, Kongsuwan K, Wijffels G,
Dixon NE, Jennings PA. 2001. Proc. Natl.
Acad. Sci. USA 98:1162732
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
31/36
CELLULAR REPLICASES 313
151. Fairman M, Prelich G, Tsurimoto T, Still-
man B. 1988. Biochim. Biophys. Acta
951:38287
152. Waga S, Stillman B. 1994. Nature
369:20712
153. Lee SH, Eki T, Hurwitz J. 1989. Proc.
Natl. Acad. Sci. USA 86:736165
154. Mossi R, Hubscher U. 1998. Eur. J.
Biochem. 254:20916
155. Majka J, Burgers PM. 2004. Prog. Nucleic
Acid Res. Mol. Biol. 78:22760
156. ODonnell M, Onrust R, Dean FB, Chen
M, Hurwitz J. 1993. Nucleic Acids Res.
21:13157. Cullmann G, Fien K, Kobayashi R, Still-
man B. 1995.Mol. Cell. Biol. 15:466171
158. Tsurimoto T, Stillman B. 1991. J. Biol.
Chem. 266:195060
159. Podust VN, Tiwari N, Stephan S, Fanning
E. 1998. J. Biol. Chem. 273:3199299
160. Uhlmann F, Gibbs E, Cai J, ODonnell
M, Hurwitz J. 1997. J. Biol. Chem.
272:1006571
161. Bunz F, Kobayashi R, Stillman B. 1993.Proc. Natl. Acad. Sci. USA 90:1101418
162. Gomes XV, Gary SL, Burgers PM. 2000.
J. Biol. Chem. 275:1454149
163. Uhlmann F, Cai J, Gibbs E, ODonnell
M, Hurwitz J. 1997. J. Biol. Chem.
272:1005864
164. Yao N, Coryell L, Zhang D, Georgescu
RE, Finkelstein J, et al. 2003. J. Biol.
Chem. 278:5074453
165. Fotedar R, Mossi R, Fitzgerald P, Rous-selle T, Maga G, et al. 1996. EMBO J.
15:442333
166. Cai J, Gibbs E, Uhlmann F, Phillips B, Yao
N, et al. 1997. J. Biol. Chem. 272:18974
81
167. Gomes XV, Schmidt SL, Burgers PM.
2001. J. Biol. Chem. 276:3477683
168. Podust VN, Tiwari N, Ott R, Fanning E.
1998. J. Biol. Chem. 273:1293542
169. Cai J, Yao N, Gibbs E, Finkelstein J,Phillips B, et al. 1998. Proc. Natl. Acad.
Sci. USA 95:1160712
170. Schmidt SL, Gomes XV, Burgers PM.
2001. J. Biol. Chem. 276:3478491
171. Gomes XV, Burgers PM. 2001. J. Biol.
Chem. 276:3476875
172. Hingorani MM, Coman MM. 2002. J.
Biol. Chem. 277:4721324
173. Conaway RC, Lehman IR. 1982. Proc.
Natl. Acad. Sci. USA 79:252327
174. Conaway RC, Lehman IR. 1982. Proc.
Natl. Acad. Sci. USA 79:458588
175. Tsurimoto T, Stillman B. 1991. J. Biol.
Chem. 266:196168
176. Maga G, Hubscher U. 1996.Biochemistry
35:576477
177. Yuzhakov A, Kelman Z, Hurwitz J,
ODonnell M. 1999. EMBO J. 18:618999
178. Zuo S, Bermudez V, Zhang G, Kelman Z,
HurwitzJ.2000.J. Biol. Chem. 275:5153
62
179. Podust VN, Chang LS, Ott R, Dianov
GL, Fanning E. 2002. J. Biol. Chem.
277:3894901
180. Burgers PM, Gerik KJ. 1998. J. Biol.
Chem. 273:1975662
181. Gerik KJ, Li X, Pautz A, BurgersPM. 1998. J. Biol. Chem. 273:19747
55
182. MacNeill SA, Baldacci G, Burgers PM,
Hubscher U. 2001. Trends Biochem. Sci.
26:1617
183. Bermudez VP, MacNeill SA, Tappin I,
Hurwitz J. 2002. J. Biol. Chem. 277:
3685362
184. Johansson E, Majka J, Burgers PM. 2001.
J. Biol. Chem. 276:4382428185. Lee MY, Tan CK, Downey KM, So AG.
1984. Biochemistry 23:190613
186. MacNeill SA, Moreno S, Reynolds N,
Nurse P, Fantes PA. 1996. EMBO J.
15:461328
187. Sanchez Garcia J, Ciufo LF, Yang X,
Kearsey SE, MacNeill SA. 2004. Nucleic
Acids Res. 32:300516
188. Chen X, Zuo S, Kelman Z, ODonnell M,
Hurwitz J, Goodman MF. 2000. J. Biol.Chem. 275:1767782
189. Mozzherin DJ, McConnell M, Jasko MV,
Krayevsky AA, Tan CK, et al. 1996. J.
Biol. Chem. 271:3171117
byUniversidadeFederaldoRiodeJaneiroon08/16/11.Forpersonaluseonly.
-
8/4/2019 Annual Reviews - Replication
32/36
314 JOHNSON ODONNELL
190. Hashimoto K, Shimizu K, Nakashima N,
Sugino A. 2003. Biochemistry 42:14207
13
191. Mozzherin DJ, Tan CK, Downey KM,
Fisher PA. 1999. J. Biol. Chem. 274:
1986267
192. Einolf HJ, Guengerich FP. 2000. J. Biol.
Chem. 275:1631622
193. Reynolds N, Warbrick E, Fantes PA, Mac-
Neill SA. 2000. EMBO J. 19:110818
194. Johansson E, Garg P, Burgers PM. 2004.
J. Biol. Chem. 279:190715
195. Morrison A, Araki H, Clark AB,
Hamatake RK, Sugino A. 1990. Cell 62:114351
196. Feng W, Rodriguez-Menocal L, Tolun G,
DUrso G. 2003.Mol. Biol. Cell 14:3427
36
197. Aparicio OM, Weinstein DM, Bell SP.
1997. Cell 91:5969
198. Navas TA, Zhou Z, Elledge SJ. 1995. Cell
80:2939
199. Navas TA, Sanchez Y, Elledge SJ. 1996.
Genes Dev. 10:263243200. DUrso G, Nurse P. 1997. Proc. Natl.
Acad. Sci. USA 94:1249196
201. Kesti T, Flick K, Keranen S, Syvaoja JE,
W