CHAPTER SIX

The Eukaryotic Replication

Machine
D. Zhang*, M. O'Donnell*,†,1
*The Rockefeller University, New York, NY, United States
†
Howard Hughes Medical Institute, The Rockefeller University, New York, NY, United States
1
Corresponding author: e-mail address: [email protected]

Contents
1. Introduction 192
2. The CMG Helicase 194
3. The Polymerase Alpha-Primase 201
4. The Leading and Lagging Strand DNA Polymerases Epsilon and Delta 205
5. The PCNA Clamp and RFC Clamp Loader 208
6. The Eukaryotic Replisome Structure and Function 210
7. Comparison of Bacterial and Eukaryotic Replisomes 215
8. Future Perspectives 219
Acknowledgments 220
References 220

Abstract
The cellular replicating machine, or “replisome,” is composed of numerous different
proteins. The core replication proteins in all cell types include a helicase, primase,
DNA polymerases, sliding clamp, clamp loader, and single-strand binding (SSB) protein.
The core eukaryotic replisome proteins evolved independently from those of bacteria
and thus have distinct architectures and mechanisms of action. The core replisome pro-
teins of the eukaryote include: an 11-subunit CMG helicase, DNA polymerase alpha-
primase, leading strand DNA polymerase epsilon, lagging strand DNA polymerase delta,
PCNA clamp, RFC clamp loader, and the RPA SSB protein. There are numerous other
proteins that travel with eukaryotic replication forks, some of which are known to be
involved in checkpoint regulation or nucleosome handling, but most have unknown
functions and no bacterial analogue. Recent studies have revealed many structural
and functional insights into replisome action. Also, the first structure of a replisome from
any cell type has been elucidated for a eukaryote, consisting of 20 distinct proteins,
with quite unexpected results. This review summarizes the current state of knowledge
of the eukaryotic core replisome proteins, their structure, individual functions, and how
they are organized at the replication fork as a machine.

The Enzymes, Volume 39 # 2016 Elsevier Inc. 191

ISSN 1874-6047 All rights reserved.
http://dx.doi.org/10.1016/bs.enz.2016.03.004
192 D. Zhang and M. O'Donnell

1. INTRODUCTION
DNA replication is a once-in-a-lifetime event. It must go from start
to finish without failure, unlike the other nucleic acid-based informational
pathways of transcription and translation that can afford to sometimes fail
and then start over. Furthermore, eukaryotic genomes are very large, over
4 billion base pairs in the human. Replication of the genome must not only
finish the job, but must do so with exquisitely high fidelity, leaving no
mistakes, or exceedingly few, to preserve the species. The replication
process is initiated at origins by highly regulated proteins and cell cycle
kinases [1–3]. Once started, replication of both strands of duplex DNA
is carried out by many different proteins and enzymes that function in a
highly choreographed fashion [4,5]. Many of these proteins bind one
another relatively tightly, forming a replisome complex that acts somewhat
like the moving gears of a sewing machine. Some proteins that are central
to the DNA replication process act in a dynamic fashion, coming in and
out of the replisome at a moving replication fork. This review will focus
on the major set of proteins that function to propel the eukaryotic repli-
cation fork. The term “core replisome” will refer to those proteins that are
necessary to propagate the replication fork, regardless of their affinity to the
protein complex that moves with the fork. Numerous other proteins act
after the fork has passed, to fix mismatches and to repair and ligate
Okazaki fragments. These repair proteins will not be within the focus of
this review.
The eukaryotic replisome is much more complicated than bacterial and
viral replisomes, as one might expect from the greater genome size and
complexity of a eukaryotic cell. The eukaryotic replisome contains all the
core replisome proteins of the bacterial replisome, plus many others.
The core replisome proteins are the helicase, primase, DNA polymerases,
sliding clamp, clamp loader, and single-strand binding (SSB) protein. The
eukaryotic core components are multisubunit complexes, each with more
subunits than their bacterial counterparts. There are many other factors
that move with the eukaryotic replisome and have no analogous bacterial
protein. These additional factors, outside the core machinery, appear to
be involved in two major processes. One is regulation of the fork, as some
of the components of the eukaryotic replisome are involved in checkpoint
and repair pathways. Another is the need for the eukaryotic replisome to
handle nucleosomes, which bacteria do not have. Nucleosomes are at the
heart of animal development.
The Eukaryotic Replisome 193

This review will discuss each of the eukaryotic replisome core compo-
nents and how they fit and function together. We conclude with a brief
comparison of the eukaryotic replisome to the bacterial replisome. There
are a few additional facts to mention in this section before starting off
on these subjects. One is that eukaryotes have linear chromosomes and
use numerous origins, unlike most bacteria and archaea that typically use
one or a few origins within a circular genome [6]. Thus, eukaryotic origins
are often spaced only tens of kilobases apart, and eukaryotic replication forks
do not have to cover the megabase distances that bacterial forks must cover
[7]. This may be why the rate of eukaryotic replication forks average about
10–50 nucleotides compared to the 10- to 20-fold faster rate of bacterial and
phage replisomes [7–9]. The use of linear chromosomes requires that
eukaryotes have telomere ends, which are replicated by the telomerase ribo-
zyme. This review will not cover telomere biology, and the reader is referred
to review on this subject [10–12].
Genome sequencing of cells from the three domains of life, bacteria,
archaea, and eukaryotes, reveal that most of the core replisome components
evolved twice, independently [13,14]. Thus, the bacterial core replisome
enzymes do not share a common ancestor with the analogous components
in eukaryotes and archaea [13,14], while the archaea and eukaryotic core
replisome machinery share a common ancestor [15,16]. An exception to this
are the clamps and clamp loaders, which are homologous in all three
domains of life [14,17,18]. This is quite unlike the processes of transcription
and translation, the two other major nucleic acid informational pathways,
both of which have homologous core components and a universal genetic
code in bacteria, archaea, and eukaryotes. Why the core replisome machin-
ery evolved independently in bacteria and archaea/eukaryotes is unknown.
One possibility is that the last universal common ancestor (LUCA) cell rep-
licated its nucleic acid genome in a much more streamlined and simple fash-
ion compared to modern day cells, such as seen in some phage and viruses.
For example, many phage and viruses do not replicate both strands of
double-strand (ds) DNA at the same time. If the two strands of dsDNA
are not made simultaneously (ie, one strand is completed before the other
strand is started), replication simplifies tremendously. For example, priming
is only needed once or twice and can be substituted by a nick generated by an
endonuclease for a covalently closed genome, a tRNA primer as in retrovi-
ruses, terminal protein priming or by RNA polymerase [6]. Instead of using
a helicase to unwind DNA, the DNA polymerase may be capable of strand
displacement (eg, phi29 bacteriophage). Or if RNA is the genome, it may be
digested during replication as in retrovirus replication [6]. Interestingly, the
194 D. Zhang and M. O'Donnell

helicase and primase are the most different among the core replisome factors
of bacterial and eukaryotic replisomes. The ATPase motor of bacterial rep-
licative helicases (eg, E. coli DnaB) is based on a RecA fold, while the six
ATPase subunits of the eukaryotic Mcm helicase are based on the AAA+
fold [19,20]. The primase of bacteria is based on a Toprim fold, that has
an evolutionary relationship to topoisomerases [21–23], while the primase
of eukaryotes shares homology to the X-family of DNA polymerases [24].
There are six classes of DNA polymerases that are unrelated in sequence
[25,26]. The eukaryotic replicative DNA polymerases are in the B-family,
while the only domain of life that contains C-family DNA polymerases are
bacteria, which use them for genome replication [27]. Only the sliding
clamp processivity factor and the clamp loader of bacteria and eukaryotes
share a common ancestor, and one can question why this may be so. The
structure and function of the circular clamp and clamp loader were first dis-
covered in the context of E. coli replication [28,29], but later studies have
shown that the clamp and clamp loader are used by numerous different pro-
teins in various DNA repair, checkpoint, and cell cycle regulatory pathways
in all cell types [30,31]. Thus, the clamp/clamp loader may have evolved for
a nonreplicative use in LUCA and then was recruited for replication later in
cellular evolution after the split of bacteria from archaea/eukaryotes.
The core components of the eukaryotic replisome are listed in Table 1
for Saccharomyces cerevisiae (budding yeast). Other important systems for
replisome studies include: Schizosaccharomyces pombe, Drosophila melanogaster,
Xenopus laevis, and human. The composition of the full eukaryotic replisome
is still an active area of investigation, and new factors that move with the
replisome continue to be found. Despite the many proteins that move with
eukaryotic replisomes, biochemical reconstitution studies confirm that an
active moving replisome that replicates both leading and lagging strands
assembles from the components listed in Table 1 [32,33]. This does not lessen
the importance of the many additional factors that move with replisomes,
because the ability to regulate the replisome, to adhere the sister chromatids,
and to handle nucleosomes are all central to genomic integrity. On that note,
we begin an exploration of the individual components of the replisome.

2. THE CMG HELICASE

The eukaryotic replicative helicase is an 11-subunit complex referred
to as CMG [34,35]. CMG is an acronym of the three components required
for helicase activity. The “C” refers to Cdc45, a protein that shares
The Eukaryotic Replisome 195

Table 1 Core Replication Proteins of the Saccharomyces cerevisiae Replisome

Subunit (MW in kDa) Complex (MW in Da) Complex Function
Mcm2 98,779 Mcm2–7 605,627 CMG 786,038 Helicase
Mcm3 107,516
Mcm4 105,002
Mcm5 86,410
Mcm6 112,977
Mcm7 94,943
Sld5 33,947 GINS 105,163
Psf1 24,203
Psf2 25,061
Psf3 21,952
Cdc45 74,248
Pol1 166,808 Pol alpha 355,534 Primase
Pol12 78,774
Pri1 47,690
Pri2 62,262
Pol2 255,669 Pol epsilon 378,670 Leading polymerase
Dpb2 78,229
Dpb3 22,665
Dpb4 21,997
Pol 3 124,618 Pol delta 220,222 Lagging polymerase
Pol31 55,295
Pol32 40,309
Rfa1 79,347 RPA 114,099 Single-strand binding
protein
Rfa2 29,936
Rfa3 13, 816
196 D. Zhang and M. O'Donnell

homology to the RecJ family of nucleases but is not known to contain

nuclease activity [36,37]. The “M” refers to the six Mcm motor subunits
that form the ring shaped heterohexamer Mcm2–7 complex [38,39]. The
“G” stands for the heterotetramer GINS complex, which is an acronym
formed from the first letter of the Japanese spelling of the numbers
5,1,2,3 (go-ichi-ni-san) and their individual subunit names are Sld5, Psf1,
Psf2, and Psf3 [40]. Together, the 11-subunit CMG assembly is an active
helicase [34,35]. Archaea use a single Mcm subunit that hexamerizes to form
an active Mcm helicase [41,42], although protein homologs of Cdc45 and
GINS are present in archaea [36]. The Mcm2–7 of budding yeast displays
some helicase activity [38], but human and Drosophila Mcm2–7 display
no detectable helicase activity without the Cdc45, GINS accessory factors
[34,43].
CMG is formed during activation of an origin, a fascinating multistep
reaction (reviewed in Ref. [1–3,5]). Briefly, origin activation is separated
into two distinct phases of the cell cycle, G1 and S. In G1, an origin is
“licensed” in a process that requires the six-subunit origin recognition com-
plex, along with Cdc6, Cdt1, and the Mcm2–7 complex. The licensing
process results in two Mcm2–7 hexamers that encircle dsDNA and are posi-
tioned head-to-head (ie, the two hexamers bind via their N-terminal
regions). In proceeding to S phase, several origin activation factors work
together, along with two cell cycle kinases, to form the CMG complex
in which Cdc45 and GINS become tightly associated with Mcm2–7, and
each CMG complex encircles only one strand of DNA. The CMG helicases
proceed to unwind DNA, forming the ssDNA needed for priming and
replisome assembly. The detailed mechanism of the origin activation process
is still an active area of research, and the complete reconstitution of this
process from pure proteins has recently been achieved in the S. cerevisiae sys-
tem [44]. It is interesting to note that a mixture of Mcm2–7, Cdc45, and
GINS does not result in formation of CMG [34,35]. Thus, CMG requires
some or all of the origin activation factors for its formation. Presumably some
type of energetic barrier must be surpassed to put CMG together from its
component parts.
Helicases have been subdivided into superfamilies (SF), monomeric and
dimeric helicases in SFI and SFII, hexameric AAA+ helicases of SF3 and SF6
which track 30 -50 along ssDNA, and RecA-based helicases of SF4 and SF5
that track 50 -30 along ssDNA [20]. The eukaryotic Mcm2–7 and archaeal
Mcm are SF6 helicases, the eukaryotic SV40 (T-antigen (T-Ag)) and pap-
illoma virus (E1 protein) are SF3 helicases, and the bacterial cellular and
The Eukaryotic Replisome 197

animal mitochondrial helicases, and phage T4 and T7 helicases are in the

SF4 family. All the replicative helicases function as circular homohexamers,
except CMG which contains the circular Mcm2–7 heterohexamer motor
plus five accessory factors. A feature that is shared by the motor subunits
of all replicative helicases is their construction from two major domains,
the N- and C-terminal domains (NTDs and CTDs) (Fig. 1A). The
CTDs contain the ATP sites, and thus are the motor domains. The ATP
sites are located at subunit interfaces which may facilitate intersubunit com-
munication driven by ATP hydrolysis. The NTDs of the different helicases
are quite distinct but appear to form a scaffold on which the motor domains
are attached. The NTDs of the Mcm subunits contain an OB fold
(oligonucleotide/oligosaccharide-binding fold), which often bind ssDNA,
and are demonstrated to do so in the NTD hexameric ring of an archaeal
Mcm [45].
The ring-shaped helicases of all cell types are currently thought to act by
the steric exclusion mechanism of DNA unwinding (Fig. 1B). An alternative
mechanism is the side channel extrusion model. The distinguishing charac-
teristic of these models is whether one or both strands of DNA enter the
central pore [46–51]. In the steric exclusion mechanism only one DNA
strand enters the helicase, while the other strand is excluded, and as the heli-
case tracks along the ssDNA it acts as a wedge to split the parental duplex.
The side channel extrusion mechanism requires both strands to enter the
central pore, and DNA is split inside the helicase, followed by one strand

Fig. 1 Replicative helicases are hexamers that have the appearance of two stacked rings.
(A) All replicative helicases are composed of a hexameric motor. The individual subunits
consist of two large globular domains, a N-terminal domain (NTD) and C-terminal domain
(CTD). Thus, helicase hexamers have the appearance of two rings stacked on one another.
(B) Steric exclusion mechanism of unwinding. In the steric exclusion mechanism, one
strand is excluded, while the tracking strand threads through the central pore of the
helicase. The eukaryotic helicases encircle the leading strand, as illustrated.
198 D. Zhang and M. O'Donnell

being directed out a side channel between the CTD and NTD. Interest-
ingly, the bacterial helicases and CMG have been shown capable of tracking
over either ssDNA or dsDNA, and thus they could possibly function by
either mechanism [43,48,49,52,53]. But the available experimental evidence
is that all replicative helicases, from bacteria to eukaryotes, function by the
steric exclusion mechanism [47,52–56]. It is not entirely clear whether
nucleotide hydrolysis is required to actively melt the duplex DNA, or
whether the duplex frays by thermal melting, and that helicase tracking
along ssDNA prevents reannealing. These modes are referred to as
“active” and “passive” unwinding, respectively. The rate of thermal fraying
has been measured to be rapid and possibly sufficient to account for the rate
of the fast bacterial helicases [57–59]. Current estimates of ATP per base pair
(bp) unwound are one ATP every 1–2 bp, and given the 12 kcal/ATP
compared to 3.6 kcal for 2 bp of dsDNA, the energetics are compatible
with either case [60–62]. Considering the 10- to 20-fold slower rate of
eukaryotic fork movement, active unwinding would not seem necessary,
but displacement of nucleosomes may become a major energetic barrier.
The main evidence for CMG acting by the steric exclusion mechanism
derives from studies in the Xenopus system, using DNA substrates that con-
tain a streptavidin–biotin block on either the leading or lagging strand [54].
In the steric exclusion mechanism, a bulky group attached to the excluded
strand that lies on the outside of the central channel should not slow the heli-
case, but a bulky block on the leading strand should restrict DNA entry
into the central channel and thus prevent unwinding. This is, in fact, the
observed result for E. coli DnaB [48]. Xenopus egg extracts replicate exoge-
nous DNA in a manner that depends on all the known components of the
replisome, and therefore this system is believed to reflect the behavior of a
complete replisome. In the Xenopus study, the “block” was formed by two
adjacent biotinylated nucleotides, each bound to streptavidin. When pres-
ented with a leading strand block, the replisome is halted, as expected of
a helicase that tracks along this strand. But the lagging strand block did
not stop the replisome, indicating that the lagging strand stays outside the
helicase and supporting a steric exclusion mechanism [54]. Unlike studies
of E. coli DnaB, the lagging strand block resulted in replisome pausing in
the Xenopus study. The nature of this pause is not understood at the current
time but suggests that the lagging strand may have extensive interaction with
the outside surface of CMG. Indeed, studies with archaeal Mcms, and cross-
linking studies of Drosophila CMG with DNA, support the proposal that the
lagging strand wraps around the outside of the Mcm ring [63,64].
The Eukaryotic Replisome 199

Unlike other hexameric helicases, the Mcm2–7 has an opening between

two of its subunits, sometimes referred to as the Mcm2/5 gate. This is
observed by EM 3D reconstruction studies of Mcm2–7 from two different
organisms and shown in Fig. 2 for D. melanogaster Mcm2–7 [65,66]. A gap
between the Mcm2/5 subunits was predicted from earlier biochemical
studies and is required during the initiation process [39,67]. At an origin,
Mcm2–7 must first be loaded onto dsDNA to form a double hexamer.
Although EM studies of the double hexamer do not visualize a gap between
subunits, biochemical studies demonstrate that the Mcm2/5 interface must

Fig. 2 EM structures of the Mcm2–7 helicase motors. (A) CryoEM atomic structure of
Saccharomyces cerevisiae Mcm2–7 double hexamer. Left, surface view; each subunit is
a different color (gray shades in the print version). Right, cut away illustration of the cen-
tral channel of the double hexamer, with an inset that illustrates a spiral arrangement of
loops that may bind dsDNA. (B) EM structure of Mcm2–7 from Drosophila melanogaster
adopts two conformations; one (right) has a prominent gap between the AAA + CTD
domains of Mcm2 and Mcm5. (C) EM structure of Drosophila melanogaster CMG with
and without ADP-BeF3. Panel (A) adapted by permission from Macmillan Publishers Ltd.
N. Li, Y. Zhai, Y. Zhang, W. Li, M. Yang, J. Lei, B.K. Tye, N. Gao, Structure of the eukaryotic
MCM complex at 3.8 A, Nature 524 (2015) 186–191. Panels (B) and (C) are adapted by per-
mission from Macmillan Publishers Ltd. A. Costa, I. Ilves, N. Tamberg, T. Petojevic, E. Nogales,
M.R. Botchan, J.M. Berger, The structural basis for MCM2-7 helicase activation by GINS and
Cdc45, Nat. Struct. Mol. Biol. 18 (2011) 471–477.
200 D. Zhang and M. O'Donnell

be allowed to open during Mcm2–7 double hexamer assembly at an ori-

gin [68]. During formation and activation of CMG, an opening between
Mcm subunits is needed a second time to exclude one of the DNA strands
from the central channel for the steric exclusion mechanism of DNA
unwinding. Although it is not known which subunit interface opens for
the transition to encircling ssDNA, it could be the Mcm2/5 gate. In a tour
de force, the >1.2 MDa yeast Mcm2–7 double hexamer structure has been
solved at atomic resolution by cryoEM [69] (Fig. 2). The Mcm2–7 hexamers
are closed in the double hexamer, but interestingly the two hexamers are
rotated relative to one another such that if the Mcm2/5 subunits were
to open, the two gaps would not be aligned, and the double hexamer
would remain topologically linked to DNA. The exact mechanism by
which dsDNA at the origin is melted is not yet known. The double hexamer
structure shows the two hexamers are askew, indicating that the DNA
between them would be kinked and distorted. It is proposed that ATP
may fuel a twisting motion that could lead to DNA unwinding [69].
The atomic model of the Mcm2–7 double hexamer reveals a wide cen-
tral channel, 30–40 Å in diameter, consistent with ability of Mcm2–7 to
encircle dsDNA. Interestingly, two constriction points were observed in
each hexamer, one in the NTD and one in the CTD. The constriction point
in the CTD motor domains corresponds to loops that form a right hand spi-
ral and are predicted to track the grooves of DNA. These loops may corre-
spond to the loops in E1 helicase that appear to move with nucleotide
hydrolysis and are proposed to “escort” ssDNA through the hexamer during
translocation, one nucleotide per ATP hydrolyzed [70].
The EM structure of Drosophila CMG, and more recently the yeast
CMG, has been determined (Fig. 2) [65,71]. The accessory factors bind
the outside surface of the Mcm2–7 ring and span the Mcm2/5 interface,
effectively trapping DNA inside even if the Mcm2/5 interface were to open.
Furthermore, the accessory factors form a second, smaller channel lined
by the outside surfaces of Mcm2/5/3 and the GINS and Cdc45 subunits.
It was originally suggested that this second channel may hold the lagging
strand [65], and this could possibly explain the observed replisome pausing
by lagging strand-specific blocks in the Xenopus system [54].
Association of several other proteins with CMG has been identified in
pull-down experiments of nuclease-treated cell extracts using epitopes
directed against the GINS complex [72]. The resulting “replisome progres-
sion complex,” or RPC, contained, in addition to CMG, the following pro-
teins: Ctf4, Pol alpha, Mcm10, Topo I, the checkpoint response factors
The Eukaryotic Replisome 201

Mrc1, Tof1, Csm3, and the nucleosome mobility factor, FACT [72].
A criteria for RPC isolation were the ability to withstand two washes of
the immunoprecipitation beads. Hence, one can anticipate other factors
bind the replisome but associate too weakly to survive these conditions.
We will revisit some of the factors of the RPC in sections to follow.

3. THE POLYMERASE ALPHA-PRIMASE

DNA polymerases cannot initiate DNA synthesis de novo; they
require a preexisting primed site (ie, a 30 ss/ds junction). Thus, initiation
of replication on both the leading and the lagging strands requires a primase
enzyme that synthesizes an RNA primer to initiate DNA synthesis. Since the
leading strand is extended in the direction of fork movement, it may only
need to be primed one time. But due to the antiparallel structure of
DNA the lagging strand is extended in the opposite direction of fork
unwinding, and this requires formation of multiple Okazaki fragments that
each initiated by a primer synthesized by primase. Okazaki fragments are
only 100–200 nucleotides in eukaryotes, and therefore priming events must
be frequent [6].
Primases synthesize a short section of RNA of a dozen nucleotides or
less. Why is RNA used for the primer, and not DNA? The answer to this
question is not firm, but there are some reasonable possibilities. One is that
rNTPs are in 10- to 100-fold higher concentration in cells compared to
dNTPs [73]. The initial condensation of two nucleotides during de novo
synthesis of a primer requires the protein to bind two nucleoside triphos-
phates at the same time. Thus, the higher concentration of rNTPs over
dNTPs may have favored the use of RNA over DNA. Another possible
reason is that the de novo synthesis of a primer is an inherently low-fidelity
process [74–76]. Thus, use of RNA for primer formation could serve as a
convenient “marker” for removal of the low-fidelity primer, and its ultimate
replacement by DNA by a high-fidelity polymerase. Interestingly, archaeal
primase can use dNTPs [77,78], but they bind rNTPs tighter [79] and
Okazaki fragments are tipped with 50 RNA in the archaeal cell [80].
The eukaryotic primase is a four-subunit enzyme that contains both
RNA primase and DNA polymerase activities in separate subunits, unlike
the single subunit bacterial primase [81–83]. This enzyme is called DNA
polymerase alpha-primase, referred to in this review as Pol alpha. Pol alpha
was the first eukaryotic polymerase discovered. After finding that Pol alpha is
both a DNA polymerase and contains an inherent primase activity [84], it
202 D. Zhang and M. O'Donnell

was thought that this one enzyme complex could perform both leading
and lagging strand synthesis. Indeed, for many years Pol alpha was thought
to be the only eukaryotic replicative polymerase. The primase activity was
found to be associated with the two smallest subunits, Pri1 (large) and Pri2
(small) [85]. Subsequent studies have shown that the catalytic site of primase
is in the smallest subunit, Pri1 [21,77,86], but the larger of the two primase
subunits is also required for activity, and its conserved C-terminal region
contains an iron–sulfur cluster that is essential for the initiation of primer
synthesis [87–89]. After synthesis of 7–9 rNMPs, the enzyme switches to
the DNA polymerase site, located in the largest subunit, Pol1. This transfer
occurs internally, without enzyme dissociation from DNA [90–92].
Interestingly, DNA synthesis terminates within about 20 nucleotides of
extension [93,94], but Pol alpha will rebind and continue DNA synthesis,
and by repeated cycles of extension can produce DNA chains of many kilo-
bases. In the presence of replicative polymerases, Pol alpha acts as a primase
by synthesis of a hybrid RNA/DNA primer of 20–30 nucleotides [94] that
is then captured and extended by processive replicative DNA polymer-
ases [95]. Structural studies of the polymerase subunit imply that it distin-
guishes A form RNA–DNA from B form DNA–DNA, and that the
enzyme may not bind well to B form dsDNA, possibly explaining why
the enzyme dissociates after a few tens of dNMPs are added to the RNA
site [93].
Crystal structure analysis show the polymerase and primase active site
regions contain three conserved acidic residues that bind two metal ions,
common to DNA polymerase active sites [96–98]. While the Pol1 polymer-
ase subunit is a B family polymerase, the primase shares homology to
the X-family of DNA polymerases [99]. Low-resolution EM structures
(eg, 25 Å) of the entire Pol alpha shows a CTD that is connected to the
B family polymerase structure by a flexible linker and that the CTD binds
to the B subunit (Fig. 3) [97,100,101]. The leading and lagging strand DNA
polymerases (Pol epsilon and Pol delta, respectively) also contain a B subunit
of similar size that is essential to cell viability and contain an oligosaccharide-
binding domain and phosphodiesterase-like region [99,102]. We know very
little about the function of the B subunit in any of these DNA polymerases.
The Pol1 CTD also binds the primase subunits, and thus the Pol alpha holo-
enzyme consists of two enzymatic components separated by >80 Å and
connected by a flexible tether (Fig. 3). The functional consequences of
this architecture and how it relates to the intramolecular hand-off of the
RNA primer to the DNA polymerase are not certain.
The Eukaryotic Replisome 203

Fig. 3 Polymerase and primase are separated by a flexible tether in Pol alpha. (A) EM
reconstruction of Pol alpha from budding yeast. (B) Cartoon illustration summarizing
the EM structure. Reproduced from Fig. 1e of R. Nunez-Ramirez, S. Klinge, L. Sauguet,
R. Melero, M.A. Recuero-Checa, M. Kilkenny, R.L. Perera, B. Garcia-Alvarez, R.J. Hall, E. Nogales,
L. Pellegrini, O. Llorca, Flexible tethering of primase and DNA Pol alpha in the eukaryotic
primosome, Nucleic Acids Res. 39 (2011), 8187–8199 by permission of Oxford University Press.

Pol alpha lacks 30 -50 proofreading exonuclease activity and this was puz-
zling during the years that Pol alpha was thought to be the replicative DNA
polymerase. Upon discovery of the high-fidelity replicative DNA polymer-
ases, Pol delta, and Pol epsilon (which have proofreaders), Pol alpha was
recognized to function primarily as a primase that generates a hybrid
RNA/DNA primer for the replicative polymerases [95,103]. Since DNA
polymerase alpha has a lower fidelity than the replicative polymerases delta
and epsilon, the DNA portion of the primer must either be removed or effi-
ciently proofread by mismatch repair and ExoI [104,105]. Removal of the
RNA is performed by either the Fen1 flap endonuclease or Dna2 nuclease,
and replaced with DNA by Pol delta [106,107]. Why Pol alpha has a DNA
polymerase is not understood. In fact, the archaeal primase consists only of
the two small RNA priming subunits, homologous to those of Pol alpha,
and does not contain the DNA polymerase or B subunits [98]. It is possible
that the DNA polymerase activity of Pol alpha serves a role beyond priming.
For example, telomeres are extended by telomerase that generates a ssDNA
50 tail. The 50 ssDNA of a telomere binds the shelterin complex and makes a
DNA loop [108], but some of the telomeric ssDNA must be converted to
dsDNA and perhaps Pol alpha serves this role.
Pol alpha has long been known to bind a protein called Ctf4 [109]. Ctf4 is
a component of the RPC and helps to stabilize the association of Pol alpha
with the replisome [72,110]. EM studies indicate that full-length Ctf4 con-
tains a central flexible hinge, complicating crystal structure analysis of the
full-length protein [111]. However, the crystal structure of the C-half of
204 D. Zhang and M. O'Donnell

Ctf4 has been determined (Fig. 4A and B), and it binds a peptide sequence
within the N-terminal region of the polymerase subunit of Pol alpha [111].
Ctf4 is a homotrimer, with a beta-propeller forming the trimer interface, and
C-terminal alpha helices that bind the Pol alpha peptide sequence (Fig. 4C).
Ctf4 also binds a peptide of the Sld5 subunit of the GINS complex in the
same position as the Pol alpha peptide [110,112]. Hence, the Ctf4 trimer
bridges the leading strand CMG helicase to the lagging strand Pol alpha
(Fig. 4D). The third protomer of Ctf4 is available to bind a third protein.

Fig. 4 Ctf4 is a homotrimer that binds peptide sequences within Pol alpha and GINS.
(A) and (B) Crystal structure of the C-half of Ctf4. The trimer is formed by a beta-propeller
motif (blue (gray in the print version)) that supports an alpha helical structure (yellow
(light gray in the print version)). (A) Top view, (B) side view. (C) Close up of a Pol1 peptide
(green (gray in the print version)) bound to the alpha helical region of Ctf4. (D) Proposed
interactions of the Ctf4 trimer with Sld5 in the GINS component of CMG, and Pol alpha.
The yellow (light gray in the print version) sphere suggests that an unidentified
protein(s) may bind Ctf4. Adapted by permission from Macmillan Publishers Ltd.
A.C. Simon, J.C. Zhou, R.L. Perera, F. van Deursen, C. Evrin, M.E. Ivanova, M.L. Kilkenny,
L. Renault, S. Kjaer, D. Matak-Vinkovic, K. Labib, A. Costa, L. Pellegrini, A Ctf4 trimer couples
the CMG helicase to DNA polymerase alpha in the eukaryotic replisome, Nature 510 (2014)
293–297.
The Eukaryotic Replisome 205

The finding that Ctf4 binds a consensus peptide is similar to studies of

PCNA, which binds a short consensus peptide found in many proteins
and acts as a platform for dynamic interactions with its many protein part-
ners [30]. If Ctf4 acts in a similar fashion, it may act as a “hub” for protein
trafficking within the replisome.

4. THE LEADING AND LAGGING STRAND DNA

POLYMERASES EPSILON AND DELTA
Eukaryotes use two different B-family DNA polymerases for
leading and lagging strand synthesis, Pol epsilon and Pol delta, respectively
[113–115]. Both polymerases contain a 30 -50 exonuclease proofreader, and a
mutation in the exonuclease site of either enzyme is associated with human
cancers [116]. Pol delta was first identified as the replicative polymerase from
biochemical studies of simian virus 40 (SV40) replication [103]. This small
virus only encodes one protein for its own replication, the SV40 large T-Ag.
T-Ag is both an origin binding protein and a helicase. Hence, all the
remaining enzymes for replication are derived from the host. The in vitro
SV40 DNA replication system initially required cell extracts and pure
T-Ag [117], but years of intensive study and fractionation eventually iden-
tified each of the required host proteins. The studies showed that Pol alpha
synthesized the primed sites for both strands, and that bulk replication was
performed by a second DNA polymerase, Pol delta, along with its required
accessory factors, PCNA and replication factor C (RFC) [103]. PCNA and
RFC were later shown to be the clamp and clamp loader, upon discovery of
the function of the analogous accessory factors in E. coli, discussed in
Section 5 [18,28,29].
Although SV40 DNA replication only requires Pols alpha and delta,
genetic studies identified Pol epsilon as a third DNA polymerase that is
essential for cellular replication [118]. Assignment of Pol epsilon to the lead-
ing strand and Pol delta to the lagging strand has followed from several
experiments by the Kunkel lab and his collaborators [113–115,118,119].
Initial studies used a subtle mutation in the active site of each polymerase
that generates an asymmetric mutation; AT is converted to GC only when
the polymerase replicates one strand but not the other. The mutant Pol delta
or Pol epsilon retains wild-type catalytic and proofreading activity, and the
mutator phenotype is quite mild. Strains having either polymerase mutant
were constructed, and a Ura marker gene was placed in one direction or
the other near the strong, early firing origin ARS306 [114,115]. Mutations
206 D. Zhang and M. O'Donnell

in the URA3 gene were sequenced from strains containing either the Pol
delta or the Pol epsilon active site mutant. These studies strongly supported
the use of Pol epsilon on the leading strand and Pol delta on the lagging
strand. Further studies were performed with polymerase mutants that
misincorporate ribonucleotides at relatively high frequency, enabling mis-
incorporation events to be examined genome wide by deep sequencing
technology, and the results supported Pol epsilon as the major leading strand
polymerase and Pol delta on the lagging strand [119]. These studies were
extended to S. pombe with similar conclusions [120]. Furthermore, DNA
cross-linking studies demonstrated that Pol epsilon cross-linked to the lead-
ing strand and Pol delta cross-linked to the lagging strand [121]. Biochemical
study of the polymerases showed that Pol delta functions with Fen1 nuclease
to efficiently process RNA from Okazaki fragments and provides a ligatable
nick, but Pol epsilon is not capable of this reaction, supporting a role of Pol
delta on the lagging strand [122,123]. While these studies support the assign-
ment of Pol epsilon on the leading strand, there is one study that arrived at a
different conclusion in which Pol delta replicates both strands [124]. The
different conclusions may possibly be explained by the need to mutate cer-
tain repair pathways to perform studies of this sort [124,125]. For example,
use of the mutant polymerases converting AT to GC required the mismatch
repair pathway to be knocked out, and use of rNTP misincorporation
mutants of the DNA polymerases required the RNaseH repair pathway
to be knocked out. Hence, the asymmetric distribution of mutations might
possibly originate from differential pathways for repair of the leading or lag-
ging strands. Given the wealth of data on Pol epsilon as the leading polymer-
ase, this review will assume that Pol epsilon operates on the leading strand.
There is no disagreement about use of Pol delta on the lagging strand. The
reader should keep in mind that a conclusive assignment of Pol epsilon as
the leading strand polymerase awaits further studies.
The active polymerase region of Pol epsilon, encompassing the first half
of the POL2 gene, can be deleted without killing the cell, although progres-
sion through S phase is severely compromised [126,127]. It was proposed
that a back-up DNA polymerase, probably Pol delta, takes over leading
strand synthesis under this circumstance [126,127]. This observation is rem-
iniscent of the case in E. coli in which the dnaE gene encoding the replicative
polymerase (alpha subunit of Pol III) can be deleted but slow growing
mutant cells still survive through use of Pol I [128]. Interestingly, the essen-
tial part of Pol2 is the C-terminal half, having the sequence of an inactive
B-family polymerase, and it is thought to serve an essential structural role.
The Eukaryotic Replisome 207

Pol epsilon is composed of four subunits, the large polymerase subunit,

the B subunit called Dpb2 (DNA polymerase-binding protein 2), and two
small histone-like subunits, Dpb3 and Dpb4 [129]. Of these, only Dpb2 is
essential in budding yeast, although mutants in Dpb3 or Dpb4 disrupt the
ability to maintain heterochromatin. All three accessory subunits bind the
C-half of Pol2, and considering that Dpb2 is essential, this may possibly
explain the essential role of the C-half of Pol2. Pol epsilon is stimulated
in synthesis by the PCNA clamp, but there are no obvious PCNA-binding
motifs in the Pol epsilon subunits. Thus, the points of contact between Pol
epsilon and PCNA are not yet identified.
Pol delta consists of three subunits in yeast: Pol3 (the polymerase), Pol31
(B subunit), and Pol32 [130]. Pol delta in higher organisms contains a small
fourth subunit, p12 [131]. Pol delta requires the PCNA clamp for detectable
activity, and the Pol32 subunit contains a PCNA-binding motif. However,
the POL32 gene is not essential in yeast, and therefore other, less obvious
points of contact must link Pol delta to PCNA. Pol delta, like Pols epsilon
and alpha, is a B family DNA polymerase. Eukaryotes contain a fourth
B family polymerase, Pol zeta, which acts as a translesion bypass polymerase
capable of extending DNA over a template lesion [132]. Interestingly, Pol
zeta contains the Pol31 and Pol32 subunits [133,134]. For information about
translesion polymerases, the reader is referred to reviews on this fascinating
topic [26,132].
Pols delta, epsilon, and alpha are B family polymerases, but they all share a
feature that does not generalize to other B family polymerases. Specifically,
these polymerases have a cysteine-rich region in their CTDs that contain
two metal-binding sites, CysA and CysB [135]. These sites have been
characterized in the Pol3 subunit of Pol delta, and the findings are likely
to generalize. The CysB site binds an iron–sulfur cluster and is required
for Pol3 to bind the B subunit [135]. The CysA site binds zinc, and this site
is needed for Pol delta to function with PCNA [135]. The fact that iron–
sulfur clusters have different oxidation states suggest that it may sense the
oxidation–reduction state of the cell, but further studies are needed to
explore the functions of these metal-binding sites.
The crystal structure of the catalytic regions of DNA polymerases alpha,
delta, and epsilon shows the expected structure of B family polymerases
[97,136,137]. As with DNA polymerases of all six families (A, B, C, X,
Y, RT), the B family polymerases are composed of three subdomains
(fingers, palm, and thumb) that form the shape of a right hand, and the palm
contains the active site, in which three conserved acidic residues bind two
208 D. Zhang and M. O'Donnell

metal ions that catalyze DNA polymerization [25,26]. Pol epsilon contains
a distinct feature not present in Pols delta and alpha, referred to as a
“processivity domain” [136]. The processivity domain closes the gap
between the fingers and thumb, thereby completely encircling the DNA
substrate.

5. THE PCNA CLAMP AND RFC CLAMP LOADER

The first discovery of a circular protein that surrounds DNA was the
E. coli beta clamp [28,29]. Since that time numerous proteins have been
shown to encircle DNA for function. The initial biochemical study identi-
fying the ring shape demonstrated that beta binds DNA topologically and
slides along dsDNA and can slide off the end of a linear DNA [138]. Proof
that beta was a circular protein soon followed, with the crystal structure
showing a homodimer that forms a ring with a hole of sufficient diameter
to encircle dsDNA [28]. The fact that each beta protomer is constructed
of three globular subdomains enabled the prediction that eukaryotic PCNA,
which is 2/3 the size of beta, would be a homotrimer of subunits composed
of two globular subdomains [28]. This prediction was proven correct with
the crystal structure of PCNA (Fig. 5A) [139,140]. These clamps bind to
DNA polymerases and hold them to DNA for high processivity.
Many subsequent studies of PCNA showed that it binds numerous pro-
teins involved in maintaining genomic integrity, including mismatch repair
proteins, nucleosome assembly factors, nucleases, ligase, lesion bypass DNA
polymerases, and cell cycle kinases. In fact, the E. coli beta clamp has also
been shown to bind and function with numerous other proteins involved
in maintaining genomic integrity. For more details, the reader is referred
to a recent review on this topic [30] and to chapter “The many roles of
PCNA in eukaryotic DNA replication” by Boehm et al.
Sliding clamps require a five-subunit clamp loader that uses ATP to
assemble the clamp at a primed site [18]. In eukaryotes, the clamp loader
is referred to as RFC that was discovered as a polymerase accessory protein
in the SV40 DNA replication system [103]. The subunits of bacterial and
eukaryotic clamp loaders are homologous to one another, and each are in
the AAA+ family of ATPases (ATPases associated with diverse cellular
activities) [141,142]. The ATP-binding region of AAA+ proteins is an
ancient fold found in different types of proteins, including proteases, vesicle
fusion factors, helicases, origin-binding proteins, and many others [19].
AAA + proteins generally function as oligomers that remodel other proteins.
The Eukaryotic Replisome 209

Fig. 5 Structure and function of the PCNA clamp and RFC clamp loader. (A) Crystal
structure of human PCNA (1AXC) [139]. The subunits of the homotrimer are shown in
different colors (gray shades in the print version). (B) Structure of yeast RFC bound to
PCNA (1SXJ). RFC is shown in the space filling representation. PCNA is in the ribbon
representation. DNA (green (dark gray in the print version)/orange (gray in the print
version)) is modeled into the structure. (C) Scheme of RFC clamp loader function in
assembly of PCNA onto a primed site. Panel (B) adapted with permission of Nature
Publishing Group from Fig. 6b of G.D. Bowman, M. O'Donnell, J. Kuriyan, Structural analysis
of a eukaryotic sliding DNA clamp-clamp loader complex, Nature 429 (2004) 724–730.

The first AAA+ protein to be solved structurally was the delta prime subunit
of the E. coli clamp loader [142]. Subsequently, the structures of clamp
loaders from E. coli, yeast, and phage T4 were solved [143–145]. These
clamp loaders from diverse organisms have a similar structure and mecha-
nism of action. The ATP sites are located at the interface between subunits,
requiring residues from both protomers for hydrolysis, and this strategic
architecture organizes the structural changes in a multiprotein complex that
are needed to open a clamp, position it on a primed site, and to close the
clamp around DNA.
The structure of yeast RFC bound to PCNA is shown in Fig. 5B. The
five-clamp loading subunits are arranged in a right hand spiral with the AAA
+ domains held together by strong interactions between their CTDs that
form a tight pentameric collar. The AAA+ domains contain a gap between
two of the subunits; the gap enables DNA to pass into a central chamber
210 D. Zhang and M. O'Donnell

formed by the AAA + domains of the five subunits. ATP is required for RFC
to open the PCNA clamp, but PCNA is not open in the crystal structure
despite the presence of ATP. This is due to use of ATP site mutations in
RFC to prevent hydrolysis during crystal formation. However, by analogy
to other clamp–clamp loader structures [145,146], the PCNA interface that
opens is directly below the gap between RFC1 and RFC5. An outline of the
mechanism by which clamp loaders function is illustrated in Fig. 5C and is an
amalgam of structural and biochemical information on clamp loaders from
E. coli, phage T4, yeast, and human. Three ATP sites are located at subunit
interfaces, and residues from both subunits at each ATP site are essential to
nucleotide hydrolysis. Upon PCNA ring opening, a primed site can pass
through the gap in PCNA and between the AAA+ domains of RFC1
and RFC5 to fit into the inner chamber of RFC. The DNA-binding
site of RFC (and the bacterial clamp loader) contains conserved DNA-
binding residues located on loops in the five AAA+ domains [143,147].
The sequence independent, primed site structure-dependent specificity is
facilitated by the pentameric collar that has no opening for DNA, and thus
the DNA must sharply bend out the side of the clamp loader. The ss/ds
DNA junction of a primed site contains the flexibility needed for this sharp
bend, but dsDNA is too rigid. Binding of DNA induces the five subunits to
adopt a pitch that closely follows the DNA duplex, and this organizes the
ATP site residues into an active conformation [145,148]. ATP hydrolysis
enables the clamp loader to eject, and the PCNA ring to close around
the primed site for use by DNA polymerase (or other enzymes).

6. THE EUKARYOTIC REPLISOME STRUCTURE AND

FUNCTION
Recent biochemical studies have reconstituted an active eukaryotic
replication fork from purified proteins (Fig. 6A) [32,33]. These protein
factors also enabled reconstitution of replisome assemblies that have been
studied by single-particle EM 3D reconstruction techniques, revealing a
surprising architecture of the core eukaryotic replisome [71]. The factors
required to reconstitute a functional eukaryotic replication fork include:
Pol alpha, Pol delta, Pol epsilon, the RFC clamp loader, PCNA clamp,
RPA SSB protein, and the 11-protein CMG helicase complex, for a total
of 31 different subunits. Replication studies utilized a linear synthetic
3 kb forked DNA that lack dC on one strand and dG on the other, enabling
specific monitoring of either the leading or lagging strand depending on the
radionucleotide added to the assay. The CMG helicase functioned with Pol
The Eukaryotic Replisome 211

Fig. 6 Reconstitution of eukaryotic replication forks reveals the mechanism of assem-

bly. (A) SDS polyacrylamide gels of proteins used in replication fork reconstitution stud-
ies. (B) Scheme of replication fork assembly having distinct DNA polymerases on the two
strands. Left, Pol alpha interacts with CMG to prime both strands. Middle, Pol epsilon
functions with CMG on the leading strand (top), while Pol delta does not (bottom). Right,
Pol delta functions on the lagging strand, while Pol epsilon does not. Adapted with per-
mission from Fig. 8 of R.E. Georgescu, G.D. Schauer, N.Y. Yao, L.D. Langston, O. Yurieva, D.
Zhang, J. Finkelstein, M.E. O'Donnell, Reconstitution of a eukaryotic replisome reveals sup-
pression mechanisms that define leading/lagging strand operation, Elife 4 (2015) e04988.

epsilon, RFC, PCNA, and RPA to synthesize the leading strand, while
reactions using Pol delta in place of Pol epsilon were much less efficient,
indicating that CMG specifically stabilized Pol epsilon on the leading
strand [32]. This result is consistent with genetic studies identifying Pol epsi-
lon as the leading strand polymerase. A subsequent study revealed the
212 D. Zhang and M. O'Donnell

underlying basis by which Pol epsilon is stabilized on the leading strand

by CMG. Specifically, CMG directly binds to Pol epsilon, forming a
15-subunit CMG–Pol epsilon (CMGE) complex that can be isolated on
a glycerol gradient [149].
Pol alpha functions with CMG to synthesize primed sites and can even
function distributively to extend the leading and lagging strands in the
absence of Pol epsilon [33]. Interestingly, Pol alpha required CMG for prim-
ing activity in the presence of the RPA SSB protein, indicating a direct
interaction of Pol alpha with CMG [33]. Indeed, previous studies found that
the primase subunits of Pol alpha bind Mcm4 and that the Mcm4,6,7 com-
plex stimulates priming activity [150]. When Pol epsilon is present with Pol
alpha and CMG, Pol epsilon switches with Pol alpha after priming [33].
Unexpectedly, Pol epsilon was incapable of producing lagging stand prod-
ucts with CMG/Pol alpha and even inhibited Pol alpha in the extension of
lagging strand primers [33]. This result was unanticipated, especially given
that Pol epsilon enhanced leading strand synthesis in the very same reactions
that it decreased lagging strand synthesis. These opposite results on the two
strands suggest that in the presence of CMG, Pol epsilon displays an inherent
bias for activity on the leading strand. Addition of Pol delta to reactions con-
taining Pols alpha and epsilon resulted in efficient Okazaki fragment synthe-
sis, with similar levels of both leading and lagging strand products [33]. In
overview, the biochemical studies of the eukaryotic replication enzymes
indicate that the asymmetric replisome architecture, with distinct polymer-
ases on the leading and lagging strands, is inherent to the 31 core replisome
proteins, and the many additional proteins known to travel with the
replisome are not required for this aspect of fork architecture and function.
Asymmetry is based in the recruitment of Pol epsilon to the leading strand by
CMG, the inability of Pol epsilon to function on the lagging strand, the high
activity of Pol delta on the lagging strand, and the low activity of Pol delta
with CMG on the leading strand. These findings imply that the eukaryotic
replication fork uses distinct polymerases for the leading and lagging strands
because it is the only active solution for fork progression with CMG, as illus-
trated in Fig. 6B.
As described earlier, a 15-protein leading strand CMGE complex can be
reconstituted from CMG and Pol epsilon [149]. In addition, a mixture of
Pol epsilon, CMG, Ctf4, and Pol alpha yielded a 20-protein particle
(CMG–PolE–Ctf4–Pol alpha) that could be imaged in the EM [71]. To
image the replisome and help identify each component, the replisome
was built up in stages. First, the 3D structure of the CMGE was determined
The Eukaryotic Replisome 213

Fig. 7 EM 3D reconstruction of the CMGE leading strand replisome. Single-particle EM

of the 15-subunit CMGE complex from Saccharomyces cerevisiae resulted in a 15 Å
resolution structure with protein assignments as shown. Adapted with permission from
Nature Publishing Group (NSMB). J. Sun, Y. Shi, R.E. Georgescu, Z. Yuan, B.T. Chait, H. Li, M.E.
O'Donnell, The architecture of a eukaryotic replisome, Nat. Struct. Mol. Biol. 22 (2015)
976–982.

at 15 Å resolution (Fig. 7). Then additional protein complexes were imaged;

the final particle contains a helicase, primase, a leading strand polymerase,
and a lagging strand polymerase (Fig. 8A and B). Surprisingly, the two
DNA polymerases are located on the two opposite sides of the helicase. Prior
to this, DNA polymerases were thought to act only behind the helicase.
DNA is demonstrated to enter CMG via the CTDs of the Mcm2–7 ring
[151], and given this DNA orientation, the polymerase on the C-side of
CMG must ride ahead of the replication fork. Application of protein cross-
linking with mass spectrometry readout identified Pol epsilon as the poly-
merase that occupies the position at the C-side of CMG, and thus Pol epsi-
lon rides on top of the helicase, adjacent to the unwinding point where the
parental DNA enters into CMG (Fig. 8B). CMG functions by the steric
exclusion model of helicase unwinding [54], and therefore the lagging strand
is excluded to the outside of CMG, and the leading strand ssDNA threads
though the Mcm2–7 central channel. To accommodate this unwinding
mode, the leading strand must make a U-turn when it comes out the
NTD side of Mcm2–7 in order to reach Pol epsilon at the C-side of
CMG. This DNA path requires about 40 nucleotides of ssDNA, unless there
is a shortcut for the leading ssDNA to reach Pol epsilon through the middle
of CMG (eg, through the Mcm2/5 gate). Interestingly, there is evidence for
substantial ssDNA on the leading strand in the Xenopus system in which use
214 D. Zhang and M. O'Donnell

Fig. 8 Architecture of the eukaryotic replisome. (A) 2D class averages replisome sub-
complexes reconstituted using proteins of the Saccharomyces cerevisiae replisome.
(B) Protein arrangement within the eukaryotic replisome as inferred from the 3D
structure of CMGE along with the 2D class averages of subcomplexes [71], and the
experimentally determined direction of DNA threading through CMG [151]. Panel
(A) adapted with permission from Fig. 2a and Fig. 5 of J. Sun, Y. Shi, R.E. Georgescu, Z. Yuan,
B.T. Chait, H. Li, M.E. O'Donnell, The architecture of a eukaryotic replisome, Nat. Struct. Mol.
Biol. 22 (2015) 976–982. Panel (B) adapted with permission from Nature Publishing Group
(NSMB). J. Sun, Y. Shi, R.E. Georgescu, Z. Yuan, Chait, H. Li, M.E. O'Donnell, The architecture
of a eukaryotic replisome, Nat. Struct. Mol. Biol. 22 (2015) 976–982.
The Eukaryotic Replisome 215

of DNA containing a site-specific intrastrand cross-link showed that the

replisome stops on the leading strand 20–40 nucleotides before the cross-
link [54].
It should be noted that it is still possible that the leading ssDNA could
enter CMG through the N-terminal face, which would place Pol epsilon
under CMG, and Pol alpha-primase would ride on top of CMG adjacent
to the DNA split point where it could prime the leading strand. For exam-
ple, the many proteins that assemble CMG on DNA at an origin may be
different than the biochemical in vitro studies that identified the DNA direc-
tion through CMG. If upon further study the replisome architecture
depicted in Fig. 8 stands, one may ask why cells would have evolved a
replisome architecture with a polymerase riding ahead of the helicase? Pol
epsilon has been documented to bind histone octamers [152,153]. Hence,
one possible reason for Pol epsilon at the top of the fork is that this would
enable Pol epsilon to encounter nucleosomes on the parental DNA.
Furthermore, genetic studies show that the Dpb3/4 subunits of Pol epsilon
are required for heterochromatin maintenance during replication, consistent
with Pol epsilon interaction with nucleosomes [153–155]. The placement of
Pol epsilon on top of CMG would bring the leading strand duplex product
DNA above the helicase and close to the parental DNA, and this may facil-
itate nucleosome transfer from the parental duplex to the leading strand
daughter duplex. Nucleosomes are heavily modified by posttranslational
modifications, providing epigenetic information that underlies the differen-
tiated state of the different cells of a multicellular organism. Hence, cells must
divide asymmetrically during embryogenesis, enabling daughter cells to take
distinct developmental paths for the different tissues of the body. The
replisome is the only cellular machine that must deal with every nucleosome
in the cell, and its asymmetric architecture could play a role in the produc-
tion of two different daughter cells that have unique epigenetic nucleosome
signatures. Specifically, if asymmetric distribution of nucleosomes occurs
upon passage of the replication fork, even if this happens at only a few stra-
tegic sites in the genome, it could underlie asymmetric cell division during
development of a multicellular animal.

7. COMPARISON OF BACTERIAL AND EUKARYOTIC

REPLISOMES
Most of the core replisome proteins of bacteria and eukaryotes are
unrelated in both sequence and structure [13–15]. Thus, the organization
216 D. Zhang and M. O'Donnell

Fig. 9 Comparison of bacterial and eukaryotic replisomes. (A) The bacterial replisome.
(B) The eukaryotic replisome. See text for details.

and internal workings of bacterial replisomes may be quite different from

eukaryotic replisomes. The current view of the organization of the bacterial
replisome is illustrated in Fig. 9A. The bacterial replisome is organized by the
clamp loader, which contains three copies of the tau subunit [156]. The tau
subunit has a C-terminal extension that is not required for clamp loading and
binds the DnaB helicase and the DNA Pol III [157]. Thus, the bacterial
replisome has three identical copies of Pol III linked to one central clamp
loader that also adheres to the hexameric helicase encircling the lagging
strand. Three molecules of Pol III in the bacterial replisome were first
observed in vitro [156], and then confirmed in vivo [158]. Use of three
Pol IIIs at one replication fork was unexpected given there are only two
strands of DNA to replicate. Both in vitro and in vivo single molecule studies
indicate that two of the Pol IIIs function on the lagging strand [158–160].
Considering that the lagging strand has numerous primed sites, use of
multiple polymerases may be advantageous.
Bacterial primase interacts with DnaB helicase for priming activity, but
primase does not form an integral part of the moving replisome; E. coli
primase repeatedly binds and releases from DnaB during priming activity
and fork movement [161]. As the bacterial fork progresses, the lagging strand
ssDNA is coated with SSB protein that protects and organizes the ssDNA [6].
The Eukaryotic Replisome 217

As RNA primers are formed, the clamp loader loads new clamps onto RNA
primers for function with Pol III. Single molecule studies show that the beta
clamp can tether Pol III to DNA for many tens of kilobases [162], but bac-
terial Okazaki fragments are only 1–2 kb in length [6]. Therefore, some
mechanism must limit the processivity of Pol III-beta and loosen the grip
of the lagging Pol III from its beta clamp, freeing Pol III to recycle from
a completed Okazaki fragment to a new primed site. There are at least
two mechanisms that perform this polymerase recycling function. One is
called “collision release” and occurs when Pol III-beta completes a section
of DNA and bumps into a 50 terminus (ie, of the previous Okazaki frag-
ment). When this occurs, Pol III loosens its grip on beta and ejects from
DNA, leaving the clamp behind [138]. Collision release is also observed
in the phage T4 system [163]. One study indicates that the rate of collision
release may not be fast enough for a moving fork [164], but experiments
with reconstituted E. coli replisomes demonstrate that collision release occurs
about 50% of the time [165]. There is a second mechanism, referred to as
“signal release” in which the lagging polymerase receives a signal to release
from its clamp before an Okazaki fragment is complete, first observed in the
T4 system [166]. Signal release has since been observed in E. coli and phage
T7 systems [165,167]. The signal in phage replisomes may be either primase,
the primer generated by primase, or the clamp assembled on a new primed
site [166,167]. Study of the signal in the E. coli system rules out primase and
indicates instead that the signal is the torsional strain incurred by a replisome
having connected polymerases, each of which travel in spirals to make helical
dsDNA products [165]. Independent (nonconnected) leading and lagging
DNA polymerases would be able to freely rotate during synthesis of their
respective strands. Torsional strain incurred by the rotation of connected
leading/lagging strand polymerases generate the high amount of force
required to separate the tight Pol III-beta connection, and experimental
evidence supports the role of torsional strain in signal release in the E. coli
system [165]. For further information on the bacterial replisome, see chapter
“The E. coli DNA replication fork” by Lewis et al.
The eukaryotic replisome has certain features in common with the bac-
terial replisome, but also many important differences. We will review some
of these distinctions later, but it is important to keep in mind that replisome
studies in the eukaryotic system are not as advanced as the study of bac-
terial systems, and thus there is still much to be learned about eukaryotic
replisomes before a true comparison can be considered firm or complete.
One of the obvious differences between eukaryotic and bacterial replisomes
218 D. Zhang and M. O'Donnell

is that the RFC clamp loader is not reported to stably associate with any
component of the eukaryotic replisome, nor is it identified as a component
of the RPC. Thus, at the current time the clamp loader is not thought to be
an integral part of the eukaryotic replisome. This contrasts with the bacterial
replisome where the clamp loader is the central organizer of the replisome.
However, both replisomes utilize sliding clamps that provide processivity to
their respective polymerases.
Another difference between bacterial and eukaryotic replisomes is the
location of the helicase. The CMG helicase tracks along the leading strand,
opposite the tracking direction of bacterial replicative helicases. However,
like bacterial systems [168–171], the eukaryotic leading polymerase (Pol
epsilon) interacts with the helicase for function [32,149]. Thus, a functional
polymerase–helicase connection generalizes from bacteria to eukaryotes.
Another similarity is that Pol alpha priming activity requires CMG
during fork progression in the presence of the RPA SSB protein [33].
Hence, the property of a primase–helicase interaction required for priming
also appears to generalize from bacteria to eukaryotes. The use of a Pol
alpha complex that generates a RNA–DNA primer appears unique to
eukaryotes, but the reader is reminded that the exact function of the
DNA polymerase component of Pol alpha is probably yet to be discovered
(discussed earlier). Distinctive to the eukaryotic replisome is the fact that Pol
alpha-primase is an integral part of the replisome, while the primase of
bacteria is not [72].
The EM 3D reconstruction of the core eukaryotic replisome places the
leading polymerase above the helicase [71]. If the function of this architec-
tural facet relates to nucleosomes, this feature may not generalize to bacteria.
The use of a Ctf4 trimer “hub” in the eukaryotic replisome might be anal-
ogous to the trimeric tau CTD in bacteria. In both cases, the trimer mediates
a polymerase-to-helicase interaction, but Ctf4 is not essential in budding
yeast and Pol alpha still functions with CMG without Ctf4 in vitro [33],
while the CTD of E. coli tau is essential both in vivo and for in vitro
replisome activity [170,172]. It is too early to tell if Ctf4 should be compared
to the CTD of bacterial tau.
Another distinction of eukaryotic replisomes, compared to the E. coli
replisome, is the use of two different DNA polymerases for the leading
and lagging strands in eukaryotes. Interestingly, some bacteria have two
essential C-family polymerases [173]. But there is conflicting evidence
whether they both function at the fork or whether one may be specific
to repair [174,175]. For example, the second C-family polymerase in the
The Eukaryotic Replisome 219

Gram-positive system is essential, but its synthetic rate with beta is 10 times
slower than fork movement, and it has low fidelity because it lacks a
proofreading exonuclease [174]. Further study is required to resolve this
interesting issue.
The eukaryotic lagging strand Pol delta is not known to travel with the
replisome, and this is in sharp contrast to E. coli, phage T4, and phage T7
systems in which the lagging polymerase travels with the fork. When a lag-
ging polymerase is part of the replisome and travels with the fork, it forms
DNA loops, one for each Okazaki fragment [176]. These DNA loops have
been observed in the EM and by single molecule experiments [176]. If
eukaryotic Pol delta does not travel with the replisome, there would be
no Okazaki fragment loops at eukaryotic forks. Further studies are required
to address this issue.
Eukaryotic replisomes are highly regulated by posttranslational modifica-
tions, unlike bacteria. Eukaryotic replisomes are also known to carry with
them several factors involved in the DNA damage checkpoint response that
controls fork movement or stability at times of DNA damage [72,177–179].
The CMG helicase is phosphorylated during DNA damage, and treatment
with checkpoint kinases downregulates CMG helicase activity in vitro
[180]. DNA damage also results in ubiquitination of the PCNA clamp,
and this modification correlates with enhanced translesion synthesis
[181,182]. Several other eukaryotic replisome proteins are known to be pho-
sophorylated or ubiqutinated in a cell cycle-specific fashion, but the function
of most of these modifications is unclear and will require further study.

8. FUTURE PERSPECTIVES
We have learned numerous important insights about how replisomes
function from studies of bacterial cells and their bacteriophages. However,
the different evolutionary lineage of the eukaryotic core replication proteins,
and the many additional factors they contain that have no known bacterial
counterpart make future studies of the eukaryotic replisome very important.
This is especially true given the central importance of DNA replication to
genomic integrity, and thus to human health and disease. The increased
complexity of eukaryotic replisomes relative to bacterial replisomes delayed
the development of a fully reconstituted in vitro eukaryotic system needed
to understand the internal workings of the replisome machinery. Only
recently have studies accomplished the reconstitution of the eukaryotic
replisome from pure recombinant proteins, and these systems will provide
220 D. Zhang and M. O'Donnell

a basis for further mechanistic investigations on the structure and function

of the eukaryotic replisome. Important questions include: Why does the
CMG helicase have 11 subunits, while all other replicative helicases are
homohexamers? What is the function of the numerous additional proteins
that travel with the eukaryotic replication fork? How do various posttrans-
lational modifications of replisome proteins alter their activity? How does
the replisome deal with nucleosomes? Do nucleosome chaperones cooper-
ate with the replisome during fork progression? How does the eukaryotic
replisome deal with encounters with RNA polymerase and how does it cope
with various types of DNA damage? The many new and recent findings
about eukaryotic replisomes detailed in this review have provided a rich
foundation upon which further advances can be made.

ACKNOWLEDGMENTS
The authors are grateful for funding from the National Institutes of Health, US
1R01GM115809 and the Howard Hughes Medical Institute. We also appreciate the
assistance of Dr. Nina Yao for help in preparing figures.

REFERENCES
[1] A. Costa, I.V. Hood, J.M. Berger, Mechanisms for initiating cellular DNA replication,
Annu. Rev. Biochem. 82 (2013) 25–54.
[2] S. Tanaka, H. Araki, Helicase activation and establishment of replication forks at chro-
mosomal origins of replication, Cold Spring Harb. Perspect. Biol. 5 (2013) 511–541.
[3] S. Tognetti, A. Riera, C. Speck, Switch on the engine: how the eukaryotic replicative
helicase MCM2-7 becomes activated, Chromosoma 124 (2015) 13–26.
[4] K.E. Duderstadt, R. Reyes-Lamothe, A.M. van Oijen, D.J. Sherratt, Replication-fork
dynamics, Cold Spring Harb. Perspect. Biol. 6 (2014).
[5] M. O’Donnell, L.D. Langston, B. Stillman, Principles and Concepts of DNA
Replication in Bacteria, Archaea, and Eukarya, first ed., Cold Spring Harbor
Laboratory Press, New York, 2013.
[6] A. Kornberg, T.A. Baker, DNA Replication, second ed., W.H. Freeman, New York,
1992.
[7] G. Guilbaud, A. Rappailles, A. Baker, C.L. Chen, A. Arneodo, A. Goldar,
Y. d’Aubenton-Carafa, C. Thermes, B. Audit, O. Hyrien, Evidence for sequential
and increasing activation of replication origins along replication timing gradients in
the human genome, PLoS Comput. Biol. 7 (2011) e1002322.
[8] M.K. Raghuraman, E.A. Winzeler, D. Collingwood, S. Hunt, L. Wodicka,
A. Conway, D.J. Lockhart, R.W. Davis, B.J. Brewer, W.L. Fangman, Replication
dynamics of the yeast genome, Science 294 (2001) 115–121.
[9] M.D. Sekedat, D. Fenyo, R.S. Rogers, A.J. Tackett, J.D. Aitchison, B.T. Chait, GINS
motion reveals replication fork progression is remarkably uniform throughout the yeast
genome, Mol. Syst. Biol. 6 (2010) 353.
[10] M.A. Giardini, M. Segatto, M.S. da Silva, V.S. Nunes, M.I. Cano, Telomere and
telomerase biology, Prog. Mol. Biol. Transl. Sci. 125 (2014) 1–40.
[11] P. Martinez, M.A. Blasco, Replicating through telomeres: a means to an end, Trends
Biochem. Sci. 40 (2015) 504–515.
The Eukaryotic Replisome 221

[12] V. Pfeiffer, J. Lingner, Replication of telomeres and the regulation of telomerase,

in: S.D. Bell, M. Mechali, M.L. DePhamphilis (Eds.), DNA Replication, first ed.,
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 2013, p. 576.
[13] P. Forterre, J. Filee, H. Myllykallio, Origin and Evolution of DNA and DNA
Replication Machineries, Landes Bioscience, Cold Spring Harbor, NY, 2004, p. 24.
[14] D.D. Leipe, L. Aravind, E.V. Koonin, Did DNA replication evolve twice indepen-
dently? Nucleic Acids Res. 27 (1999) 3389–3401.
[15] P. Forterre, The common ancestor of archaea and eukarya was not an archaeon,
Archaea 2013 (2013) 372396.
[16] K.S. Makarova, E.V. Koonin, Archaeology of eukaryotic DNA replication, Cold
Spring Harb. Perspect. Med. 3 (2013) a012963.
[17] M. Hedglin, R. Kumar, S.J. Benkovic, Replication clamps and clamp loaders, Cold
Spring Harb. Perspect. Biol. 5 (2013) a010165.
[18] B.A. Kelch, D.L. Makino, M. O’Donnell, J. Kuriyan, Clamp loader ATPases and the
evolution of DNA replication machinery, BMC Biol. 10 (2012) 34.
[19] J.P. Erzberger, J.M. Berger, Evolutionary relationships and structural mechanisms of
AAA + proteins, Annu. Rev. Biophys. Biomol. Struct. 35 (2006) 93–114.
[20] M.R. Singleton, M.S. Dillingham, D.B. Wigley, Structure and mechanism of helicases
and nucleic acid translocases, Annu. Rev. Biochem. 76 (2007) 23–50.
[21] D.N. Frick, C.C. Richardson, DNA primases, Annu. Rev. Biochem. 70 (2001)
39–80.
[22] J.L. Keck, D.D. Roche, A.S. Lynch, J.M. Berger, Structure of the RNA polymerase
domain of E. coli primase, Science 287 (2000) 2482–2486.
[23] M. Podobnik, P. McInerney, M. O’Donnell, J. Kuriyan, A TOPRIM domain in the
crystal structure of the catalytic core of Escherichia coli primase confirms a structural
link to DNA topoisomerases, J. Mol. Biol. 300 (2000) 353–362.
[24] B.W. Kirk, R.D. Kuchta, Arg304 of human DNA primase is a key contributor to
catalysis and NTP binding: primase and the family X polymerases share significant
sequence homology, Biochemistry 38 (1999) 7727–7736.
[25] T.A. Steitz, DNA polymerases: structural diversity and common mechanisms, J. Biol.
Chem. 274 (1999) 17395–17398.
[26] W. Yang, An overview of Y-Family DNA polymerases and a case study of human
DNA polymerase eta, Biochemistry 53 (2014) 2793–2803.
[27] N.Y. Yao, M.E. O’Donnell, Comparison of bacterial and eukaryotic replisome com-
ponents, in: R.A. Bradshaw, P.D. Stahl (Eds.), Encyclopedia of Cell Biology,
Academic Press, Amsterdam, Netherlands, 2015, pp. 396–417.
[28] X.P. Kong, R. Onrust, M. O’Donnell, J. Kuriyan, Three-dimensional structure of the
beta subunit of E. coli DNA polymerase III holoenzyme: a sliding DNA clamp, Cell
69 (1992) 425–437.
[29] P.T. Stukenberg, P.S. Studwell-Vaughan, M. O’Donnell, Mechanism of the sliding
beta-clamp of DNA polymerase III holoenzyme, J. Biol. Chem. 266 (1991)
11328–11334.
[30] R. Georgescu, L. Langston, M. O’Donnell, A proposal: evolution of PCNA’s role as a
marker of newly replicated DNA, DNA Repair (Amst) 29 (2015) 4–15.
[31] G. Maga, U. Hubscher, Proliferating cell nuclear antigen (PCNA): a dancer with many
partners, J. Cell Sci. 116 (2003) 3051–3060.
[32] R.E. Georgescu, L.D. Langston, N.Y. Yao, O. Yurieva, D. Zhang, J. Finkelstein,
T. Agarwal, M.E. O’Donnell, Mechanism of asymmetric polymerase assembly at
the eukaryotic replication fork, Nat. Struct. Mol. Biol. 21 (2014) 664–670.
[33] R.E. Georgescu, G.D. Schauer, N.Y. Yao, L.D. Langston, O. Yurieva, D. Zhang,
J. Finkelstein, M.E. O’Donnell, Reconstitution of a eukaryotic replisome reveals sup-
pression mechanisms that define leading/lagging strand operation, Elife 4 (2015)
e04988.
222 D. Zhang and M. O'Donnell

[34] I. Ilves, T. Petojevic, J.J. Pesavento, M.R. Botchan, Activation of the MCM2-7
helicase by association with Cdc45 and GINS proteins, Mol. Cell 37 (2010) 247–258.
[35] S.E. Moyer, P.W. Lewis, M.R. Botchan, Isolation of the Cdc45/Mcm2-7/GINS
(CMG) complex, a candidate for the eukaryotic DNA replication fork helicase, Proc.
Natl. Acad. Sci. U.S.A. 103 (2006) 10236–10241.
[36] K.S. Makarova, E.V. Koonin, Z. Kelman, The CMG (CDC45/RecJ, MCM, GINS)
complex is a conserved component of the DNA replication system in all archaea and
eukaryotes, Biol. Direct 7 (2012) 7.
[37] L. Sanchez-Pulido, C.P. Ponting, Cdc45: the missing RecJ ortholog in eukaryotes?
Bioinformatics 27 (2011) 1885–1888.
[38] M.L. Bochman, A. Schwacha, The Mcm2-7 complex has in vitro helicase activity,
Mol. Cell 31 (2008) 287–293.
[39] M.J. Davey, C. Indiani, M. O’Donnell, Reconstitution of the Mcm2-7p
heterohexamer, subunit arrangement, and ATP site architecture, J. Biol. Chem.
278 (2003) 4491–4499.
[40] Y. Takayama, Y. Kamimura, M. Okawa, S. Muramatsu, A. Sugino, H. Araki, GINS, a
novel multiprotein complex required for chromosomal DNA replication in budding
yeast, Genes Dev. 17 (2003) 1153–1165.
[41] J.P. Chong, M.K. Hayashi, M.N. Simon, R.M. Xu, B. Stillman, A double-hexamer
archaeal minichromosome maintenance protein is an ATP-dependent DNA helicase,
Proc. Natl. Acad. Sci. U.S.A. 97 (2000) 1530–1535.
[42] Z. Kelman, J.K. Lee, J. Hurwitz, The single minichromosome maintenance protein of
Methanobacterium thermoautotrophicum DeltaH contains DNA helicase activity,
Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 14783–14788.
[43] Y.H. Kang, W.C. Galal, A. Farina, I. Tappin, J. Hurwitz, Properties of the human
Cdc45/Mcm2-7/GINS helicase complex and its action with DNA polymerase epsilon
in rolling circle DNA synthesis, Proc. Natl. Acad. Sci. U.S.A. 109 (2012) 6042–6047.
[44] J.T. Yeeles, T.D. Deegan, A. Janska, A. Early, J.F. Diffley, Regulated eukaryotic DNA
replication origin firing with purified proteins, Nature 519 (2015) 431–435.
[45] C.A. Froelich, S. Kang, L.B. Epling, S.P. Bell, E.J. Enemark, A conserved MCM
single-stranded DNA binding element is essential for replication initiation, Elife
3 (2014) e01993.
[46] A.S. Brewster, G. Wang, X. Yu, W.B. Greenleaf, J.M. Carazo, M. Tjajadi,
M.G. Klein, X.S. Chen, Crystal structure of a near-full-length archaeal MCM:
functional insights for an AAA+ hexameric helicase, Proc. Natl. Acad. Sci. U.S.A.
105 (2008) 20191–20196.
[47] K.J. Hacker, K.A. Johnson, A hexameric helicase encircles one DNA strand and
excludes the other during DNA unwinding, Biochemistry 36 (1997) 14080–14087.
[48] D.L. Kaplan, M.J. Davey, M. O’Donnell, Mcm4,6,7 uses a “pump in ring” mechanism
to unwind DNA by steric exclusion and actively translocate along a duplex, J. Biol.
Chem. 278 (2003) 49171–49182.
[49] D.L. Kaplan, M. O’Donnell, DnaB drives DNA branch migration and dislodges
proteins while encircling two DNA strands, Mol. Cell 10 (2002) 647–657.
[50] S.S. Patel, I. Donmez, Mechanisms of helicases, J. Biol. Chem. 281 (2006)
18265–18268.
[51] A.M. Pyle, Translocation and unwinding mechanisms of RNA and DNA helicases,
Annu. Rev. Biophys. 37 (2008) 317–336.
[52] Y.J. Jeong, V. Rajagopal, S.S. Patel, Switching from single-stranded to double-
stranded DNA limits the unwinding processivity of ring-shaped T7 DNA helicase,
Nucleic Acids Res. 41 (2013) 4219–4229.
[53] J.H. Shin, Y. Jiang, B. Grabowski, J. Hurwitz, Z. Kelman, Substrate requirements
for duplex DNA translocation by the eukaryal and archaeal minichromosome
maintenance helicases, J. Biol. Chem. 278 (2003) 49053–49062.
The Eukaryotic Replisome 223

[54] Y.V. Fu, H. Yardimci, D.T. Long, T.V. Ho, A. Guainazzi, V.P. Bermudez,
J. Hurwitz, A. van Oijen, O.D. Scharer, J.C. Walter, Selective bypass of a lagging
strand roadblock by the eukaryotic replicative DNA helicase, Cell 146 (2011)
931–941.
[55] D.L. Kaplan, The 3’-tail of a forked-duplex sterically determines whether one or two
DNA strands pass through the central channel of a replication-fork helicase, J. Mol.
Biol. 301 (2000) 285–299.
[56] H. Yardimci, X. Wang, A.B. Loveland, I. Tappin, D.Z. Rudner, J. Hurwitz, A.M. van
Oijen, J.C. Walter, Bypass of a protein barrier by a replicative DNA helicase, Nature
492 (2012) 205–209.
[57] Y.J. Jeong, M.K. Levin, S.S. Patel, The DNA-unwinding mechanism of the ring heli-
case of bacteriophage T7, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 7264–7269.
[58] D.S. Johnson, L. Bai, B.Y. Smith, S.S. Patel, M.D. Wang, Single-molecule studies
reveal dynamics of DNA unwinding by the ring-shaped T7 helicase, Cell
129 (2007) 1299–1309.
[59] D. Jose, S.E. Weitzel, P.H. von Hippel, Breathing fluctuations in position-specific
DNA base pairs are involved in regulating helicase movement into the replication fork,
Proc. Natl. Acad. Sci. U.S.A. 109 (2012) 14428–14433.
[60] E. Delagoutte, P.H. von Hippel, Molecular mechanisms of the functional coupling
of the helicase (gp41) and polymerase (gp43) of bacteriophage T4 within the DNA
replication fork, Biochemistry 40 (2001) 4459–4477.
[61] I. Donmez, S.S. Patel, Coupling of DNA unwinding to nucleotide hydrolysis in a
ring-shaped helicase, EMBO J. 27 (2008) 1718–1726.
[62] P.H. von Hippel, E. Delagoutte, A general model for nucleic acid helicases and their
"coupling" within macromolecular machines, Cell 104 (2001) 177–190.
[63] T. Petojevic, J.J. Pesavento, A. Costa, J. Liang, Z. Wang, J.M. Berger, M.R. Botchan,
Cdc45 (cell division cycle protein 45) guards the gate of the eukaryote replisome heli-
case stabilizing leading strand engagement, Proc. Natl. Acad. Sci. U.S.A. 112 (2015)
E249–E258.
[64] E. Rothenberg, M.A. Trakselis, S.D. Bell, T. Ha, MCM forked substrate specificity
involves dynamic interaction with the 5’-tail, J. Biol. Chem. 282 (2007) 34229–34234.
[65] A. Costa, I. Ilves, N. Tamberg, T. Petojevic, E. Nogales, M.R. Botchan, J.M. Berger,
The structural basis for MCM2-7 helicase activation by GINS and Cdc45, Nat. Struct.
Mol. Biol. 18 (2011) 471–477.
[66] A.Y. Lyubimov, A. Costa, F. Bleichert, M.R. Botchan, J.M. Berger, ATP-dependent
conformational dynamics underlie the functional asymmetry of the replicative helicase
from a minimalist eukaryote, Proc. Natl. Acad. Sci. U.S.A. 109 (2012) 11999–12004.
[67] M.L. Bochman, A. Schwacha, The Saccharomyces cerevisiae Mcm6/2 and Mcm5/3
ATPase active sites contribute to the function of the putative Mcm2-7 ’gate’, Nucleic
Acids Res. 38 (2010) 6078–6088.
[68] S.A. Samel, A. Fernandez-Cid, J. Sun, A. Riera, S. Tognetti, M.C. Herrera, H. Li,
C. Speck, A unique DNA entry gate serves for regulated loading of the eukaryotic
replicative helicase MCM2-7 onto DNA, Genes Dev. 28 (2014) 1653–1666.
[69] N. Li, Y. Zhai, Y. Zhang, W. Li, M. Yang, J. Lei, B.K. Tye, N. Gao, Structure of the
eukaryotic MCM complex at 3.8 A, Nature 524 (2015) 186–191.
[70] E.J. Enemark, L. Joshua-Tor, Mechanism of DNA translocation in a replicative
hexameric helicase, Nature 442 (2006) 270–275.
[71] J. Sun, Y. Shi, R.E. Georgescu, Z. Yuan, B.T. Chait, H. Li, M.E. O’Donnell, The
architecture of a eukaryotic replisome, Nat. Struct. Mol. Biol. 22 (2015) 976–982.
[72] A. Gambus, R.C. Jones, A. Sanchez-Diaz, M. Kanemaki, F. van Deursen,
R.D. Edmondson, K. Labib, GINS maintains association of Cdc45 with MCM in
replisome progression complexes at eukaryotic DNA replication forks, Nat. Cell Biol.
8 (2006) 358–366.
224 D. Zhang and M. O'Donnell

[73] P.V. Hauschka, Analysis of nucleotide pools in animal cells, Methods Cell Biol.
7 (1973) 361–462.
[74] S. Cotterill, G. Chui, I.R. Lehman, DNA polymerase-primase from embryos of Dro-
sophila melanogaster DNA primase subunits, J. Biol. Chem. 262 (1987) 16105–16108.
[75] R.J. Sheaff, R.D. Kuchta, Misincorporation of nucleotides by calf thymus DNA
primase and elongation of primers containing multiple noncognate nucleotides by
DNA polymerase alpha, J. Biol. Chem. 269 (1994) 19225–19231.
[76] S.S. Zhang, F. Grosse, Accuracy of DNA primase, J. Mol. Biol. 216 (1990) 475–479.
[77] R.D. Kuchta, G. Stengel, Mechanism and evolution of DNA primases, Biochim.
Biophys. Acta 1804 (2010) 1180–1189.
[78] L. Liu, K. Komori, S. Ishino, A.A. Bocquier, I.K. Cann, D. Kohda, Y. Ishino, The
archaeal DNA primase: biochemical characterization of the p41-p46 complex from
Pyrococcus furiosus, J. Biol. Chem. 276 (2001) 45484–45490.
[79] S.H. Lao-Sirieix, S.D. Bell, The heterodimeric primase of the hyperthermophilic
archaeon Sulfolobus solfataricus possesses DNA and RNA primase, polymerase and
3’-terminal nucleotidyl transferase activities, J. Mol. Biol. 344 (2004) 1251–1263.
[80] F. Matsunaga, C. Norais, P. Forterre, H. Myllykallio, Identification of short ‘eukary-
otic’ Okazaki fragments synthesized from a prokaryotic replication origin, EMBO
Rep. 4 (2003) 154–158.
[81] G.R. Banks, J.A. Boezi, I.R. Lehman, A high molecular weight DNA polymerase
from Drosophila melanogaster embryos. Purification, structure, and partial characteriza-
tion, J. Biol. Chem. 254 (1979) 9886–9892.
[82] L.S. Kaguni, J.M. Rossignol, R.C. Conaway, I.R. Lehman, Isolation of an intact
DNA polymerase-primase from embryos of Drosophila melanogaster, Proc. Natl. Acad.
Sci. U.S.A. 80 (1983) 2221–2225.
[83] B. Sauer, I.R. Lehman, Immunological comparison of purified DNA polymerase alpha
from embryos of Drosophila melanogaster with forms of the enzyme present in vivo,
J. Biol. Chem. 257 (1982) 12394–12398.
[84] R.C. Conaway, I.R. Lehman, A DNA primase activity associated with DNA poly-
merase alpha from Drosophila melanogaster embryos, Proc. Natl. Acad. Sci. U.S.A.
79 (1982) 2523–2527.
[85] L.S. Kaguni, J.M. Rossignol, R.C. Conaway, G.R. Banks, I.R. Lehman, Association
of DNA primase with the beta/gamma subunits of DNA polymerase alpha from
Drosophila melanogaster embryos, J. Biol. Chem. 258 (1983) 9037–9039.
[86] W.C. Copeland, X. Tan, Active site mapping of the catalytic mouse primase subunit
by alanine scanning mutagenesis, J. Biol. Chem. 270 (1995) 3905–3913.
[87] S. Klinge, J. Hirst, J.D. Maman, T. Krude, L. Pellegrini, An iron-sulfur domain of the
eukaryotic primase is essential for RNA primer synthesis, Nat. Struct. Mol. Biol.
14 (2007) 875–877.
[88] B.E. Weiner, H. Huang, B.M. Dattilo, M.J. Nilges, E. Fanning, W.J. Chazin, An iron-
sulfur cluster in the C-terminal domain of the p58 subunit of human DNA primase,
J. Biol. Chem. 282 (2007) 33444–33451.
[89] L.K. Zerbe, R.D. Kuchta, The p58 subunit of human DNA primase is important for
primer initiation, elongation, and counting, Biochemistry 41 (2002) 4891–4900.
[90] W.C. Copeland, T.S. Wang, Enzymatic characterization of the individual mammalian
primase subunits reveals a biphasic mechanism for initiation of DNA replication,
J. Biol. Chem. 268 (1993) 26179–26189.
[91] T. Eki, J. Hurwitz, Influence of poly(ADP-ribose) polymerase on the enzymatic
synthesis of SV40 DNA, J. Biol. Chem. 266 (1991) 3087–3100.
[92] R.J. Sheaff, R.D. Kuchta, D. Ilsley, Calf thymus DNA polymerase alpha-primase:
“communication” and primer-template movement between the two active sites,
Biochemistry 33 (1994) 2247–2254.
The Eukaryotic Replisome 225

[93] R.L. Perera, R. Torella, S. Klinge, M.L. Kilkenny, J.D. Maman, L. Pellegrini,
Mechanism for priming DNA synthesis by yeast DNA polymerase alpha, Elife
2 (2013) e00482.
[94] H. Singh, R.G. Brooke, M.H. Pausch, G.T. Williams, C. Trainor, L.B. Dumas, Yeast
DNA primase and DNA polymerase activities. An analysis of RNA priming and its
coupling to DNA synthesis, J. Biol. Chem. 261 (1986) 8564–8569.
[95] B. Stillman, DNA polymerases at the replication fork in eukaryotes, Mol. Cell
30 (2008) 259–260.
[96] M.A. Augustin, R. Huber, J.T. Kaiser, Crystal structure of a DNA-dependent RNA
polymerase (DNA primase), Nat. Struct. Biol. 8 (2001) 57–61.
[97] S. Klinge, R. Nunez-Ramirez, O. Llorca, L. Pellegrini, 3D architecture of DNA Pol
alpha reveals the functional core of multi-subunit replicative polymerases, EMBO J.
28 (2009) 1978–1987.
[98] S.H. Lao-Sirieix, L. Pellegrini, S.D. Bell, The promiscuous primase, Trends Genet.
21 (2005) 568–572.
[99] E. Johansson, S.A. Macneill, The eukaryotic replicative DNA polymerases take shape,
Trends Biochem. Sci. 35 (2010) 339–347.
[100] M.L. Kilkenny, M.A. Longo, R.L. Perera, L. Pellegrini, Structures of human primase
reveal design of nucleotide elongation site and mode of Pol alpha tethering, Proc. Natl.
Acad. Sci. U.S.A. 110 (2013) 15961–15966.
[101] R. Nunez-Ramirez, S. Klinge, L. Sauguet, R. Melero, M.A. Recuero-Checa,
M. Kilkenny, R.L. Perera, B. Garcia-Alvarez, R.J. Hall, E. Nogales, L. Pellegrini,
O. Llorca, Flexible tethering of primase and DNA Pol alpha in the eukaryotic
primosome, Nucleic Acids Res. 39 (2011) 8187–8199.
[102] M. Makiniemi, H. Pospiech, S. Kilpelainen, M. Jokela, M. Vihinen, J.E. Syvaoja,
A novel family of DNA-polymerase-associated B subunits, Trends Biochem. Sci.
24 (1999) 14–16.
[103] S. Waga, B. Stillman, The DNA replication fork in eukaryotic cells, Annu. Rev. Bio-
chem. 67 (1998) 721–751.
[104] H. Hombauer, C.S. Campbell, C.E. Smith, A. Desai, R.D. Kolodner, Visualization of
eukaryotic DNA mismatch repair reveals distinct recognition and repair intermediates,
Cell 147 (2011) 1040–1053.
[105] S.E. Liberti, A.A. Larrea, T.A. Kunkel, Exonuclease 1 preferentially repairs
mismatches generated by DNA polymerase alpha, DNA Repair (Amst) 12 (2013)
92–96.
[106] R. Ayyagari, X.V. Gomes, D.A. Gordenin, P.M. Burgers, Okazaki fragment matura-
tion in yeast. I. Distribution of functions between FEN1 AND DNA2, J. Biol. Chem.
278 (2003) 1618–1625.
[107] X. Li, J. Li, J. Harrington, M.R. Lieber, P.M. Burgers, Lagging strand DNA synthesis
at the eukaryotic replication fork involves binding and stimulation of FEN-1 by
proliferating cell nuclear antigen, J. Biol. Chem. 270 (1995) 22109–22112.
[108] T. de Lange, Shelterin: the protein complex that shapes and safeguards human telo-
meres, Genes Dev. 19 (2005) 2100–2110.
[109] J. Miles, T. Formosa, Protein affinity chromatography with purified yeast DNA
polymerase alpha detects proteins that bind to DNA polymerase, Proc. Natl. Acad.
Sci. U.S.A. 89 (1992) 1276–1280.
[110] A. Gambus, F. van Deursen, D. Polychronopoulos, M. Foltman, R.C. Jones,
R.D. Edmondson, A. Calzada, K. Labib, A key role for Ctf4 in coupling the
MCM2-7 helicase to DNA polymerase alpha within the eukaryotic replisome, EMBO
J. 28 (2009) 2992–3004.
[111] A.C. Simon, J.C. Zhou, R.L. Perera, F. van Deursen, C. Evrin, M.E. Ivanova,
M.L. Kilkenny, L. Renault, S. Kjaer, D. Matak-Vinkovic, K. Labib, A. Costa,
226 D. Zhang and M. O'Donnell

L. Pellegrini, A Ctf4 trimer couples the CMG helicase to DNA polymerase alpha in the
eukaryotic replisome, Nature 510 (2014) 293–297.
[112] H. Tanaka, Y. Katou, M. Yagura, K. Saitoh, T. Itoh, H. Araki, M. Bando,
K. Shirahige, Ctf4 coordinates the progression of helicase and DNA polymerase alpha,
Genes Cells 14 (2009) 807–820.
[113] T.A. Kunkel, P.M. Burgers, Dividing the workload at a eukaryotic replication fork,
Trends Cell Biol. 18 (2008) 521–527.
[114] S.A. Nick McElhinny, D.A. Gordenin, C.M. Stith, P.M. Burgers, T.A. Kunkel, Divi-
sion of labor at the eukaryotic replication fork, Mol. Cell 30 (2008) 137–144.
[115] Z.F. Pursell, I. Isoz, E.B. Lundstrom, E. Johansson, T.A. Kunkel, Yeast DNA poly-
merase epsilon participates in leading-strand DNA replication, Science 317 (2007)
127–130.
[116] C. Palles, J.B. Cazier, K.M. Howarth, E. Domingo, A.M. Jones, P. Broderick,
Z. Kemp, S.L. Spain, E. Guarino, I. Salguero, A. Sherborne, D. Chubb,
L.G. Carvajal-Carmona, Y. Ma, K. Kaur, S. Dobbins, E. Barclay, M. Gorman,
L. Martin, M.B. Kovac, S. Humphray, A. Lucassen, C.C. Holmes, D. Bentley,
P. Donnelly, J. Taylor, C. Petridis, R. Roylance, E.J. Sawyer, D.J. Kerr, S. Clark,
J. Grimes, S.E. Kearsey, H.J. Thomas, G. McVean, R.S. Houlston, I. Tomlinson,
Germline mutations affecting the proofreading domains of POLE and POLD1 predis-
pose to colorectal adenomas and carcinomas, Nat. Genet. 45 (2013) 136–144.
[117] J.J. Li, T.J. Kelly, Simian virus 40 DNA replication in vitro, Proc. Natl. Acad. Sci.
U.S.A. 81 (1984) 6973–6977.
[118] A. Morrison, H. Araki, A.B. Clark, R.K. Hamatake, A. Sugino, A third essential DNA
polymerase in S. cerevisiae, Cell 62 (1990) 1143–1151.
[119] A.R. Clausen, S.A. Lujan, A.B. Burkholder, C.D. Orebaugh, J.S. Williams,
M.F. Clausen, E.P. Malc, P.A. Mieczkowski, D.C. Fargo, D.J. Smith,
T.A. Kunkel, Tracking replication enzymology in vivo by genome-wide mapping
of ribonucleotide incorporation, Nat. Struct. Mol. Biol. 22 (2015) 185–191.
[120] I. Miyabe, T.A. Kunkel, A.M. Carr, The major roles of DNA polymerases epsilon and
delta at the eukaryotic replication fork are evolutionarily conserved, PLoS Genet.
7 (2011) e1002407.
[121] C. Yu, H. Gan, J. Han, Z.X. Zhou, S. Jia, A. Chabes, G. Farrugia, T. Ordog,
Z. Zhang, Strand-specific analysis shows protein binding at replication forks and
PCNA unloading from lagging strands when forks stall, Mol. Cell 56 (2014) 551–563.
[122] P. Garg, C.M. Stith, N. Sabouri, E. Johansson, P.M. Burgers, Idling by DNA poly-
merase delta maintains a ligatable nick during lagging-strand DNA replication, Genes
Dev. 18 (2004) 2764–2773.
[123] Y.H. Jin, R. Ayyagari, M.A. Resnick, D.A. Gordenin, P.M. Burgers, Okazaki
fragment maturation in yeast. II. Cooperation between the polymerase and 3’-5’-
exonuclease activities of Pol delta in the creation of a ligatable nick, J. Biol. Chem.
278 (2003) 1626–1633.
[124] R.E. Johnson, R. Klassen, L. Prakash, S. Prakash, A major role of DNA polymerase
delta in replication of both the leading and lagging DNA strands, Mol. Cell 59 (2015)
163–175.
[125] B. Stillman, Reconsidering DNA polymerases at the replication fork in eukaryotes,
Mol. Cell 59 (2015) 139–141.
[126] R. Dua, D.L. Levy, J.L. Campbell, Analysis of the essential functions of the C-terminal
protein/protein interaction domain of Saccharomyces cerevisiae pol epsilon and its unex-
pected ability to support growth in the absence of the DNA polymerase domain,
J. Biol. Chem. 274 (1999) 22283–22288.
[127] T. Kesti, K. Flick, S. Keranen, J.E. Syvaoja, C. Wittenberg, DNA polymerase epsilon
catalytic domains are dispensable for DNA replication, DNA repair, and cell viability,
Mol. Cell 3 (1999) 679–685.
The Eukaryotic Replisome 227

[128] O. Niwa, S.K. Bryan, R.E. Moses, Alternate pathways of DNA replication: DNA
polymerase I-dependent replication, Proc. Natl. Acad. Sci. U.S.A. 78 (1981)
7024–7027.
[129] Z.F. Pursell, T.A. Kunkel, DNA polymerase epsilon: a polymerase of unusual size (and
complexity), Prog. Nucleic Acid Res. Mol. Biol. 82 (2008) 101–145.
[130] P.M. Burgers, K.J. Gerik, Structure and processivity of two forms of Saccharomyces
cerevisiae DNA polymerase delta, J. Biol. Chem. 273 (1998) 19756–19762.
[131] L. Liu, J. Mo, E.M. Rodriguez-Belmonte, M.Y. Lee, Identification of a fourth subunit
of mammalian DNA polymerase delta, J. Biol. Chem. 275 (2000) 18739–18744.
[132] G.N. Gan, J.P. Wittschieben, B.O. Wittschieben, R.D. Wood, DNA polymerase zeta
(pol zeta) in higher eukaryotes, Cell Res. 18 (2008) 174–183.
[133] R.E. Johnson, L. Prakash, S. Prakash, Pol31 and Pol32 subunits of yeast DNA poly-
merase delta are also essential subunits of DNA polymerase zeta, Proc. Natl. Acad. Sci.
U.S.A. 109 (2012) 12455–12460.
[134] A.V. Makarova, J.L. Stodola, P.M. Burgers, A four-subunit DNA polymerase zeta
complex containing Pol delta accessory subunits is essential for PCNA-mediated muta-
genesis, Nucleic Acids Res. 40 (2012) 11618–11626.
[135] D.J. Netz, C.M. Stith, M. Stumpfig, G. Kopf, D. Vogel, H.M. Genau, J.L. Stodola,
R. Lill, P.M. Burgers, A.J. Pierik, Eukaryotic DNA polymerases require an iron-sulfur
cluster for the formation of active complexes, Nat. Chem. Biol. 8 (2011) 125–132.
[136] M. Hogg, P. Osterman, G.O. Bylund, R.A. Ganai, E.B. Lundstrom, A.E. Sauer-
Eriksson, E. Johansson, Structural basis for processive DNA synthesis by yeast DNA
polymerase epsilon, Nat. Struct. Mol. Biol. 21 (2014) 49–55.
[137] M.K. Swan, R.E. Johnson, L. Prakash, S. Prakash, A.K. Aggarwal, Structural basis of
high-fidelity DNA synthesis by yeast DNA polymerase delta, Nat. Struct. Mol. Biol.
16 (2009) 979–986.
[138] P.T. Stukenberg, J. Turner, M. O’Donnell, An explanation for lagging strand replica-
tion: polymerase hopping among DNA sliding clamps, Cell 78 (1994) 877–887.
[139] J.M. Gulbis, Z. Kelman, J. Hurwitz, M. O’Donnell, J. Kuriyan, Structure of the
C-terminal region of p21(WAF1/CIP1) complexed with human PCNA, Cell
87 (1996) 297–306.
[140] T.S. Krishna, X.P. Kong, S. Gary, P.M. Burgers, J. Kuriyan, Crystal structure of the
eukaryotic DNA polymerase processivity factor PCNA, Cell 79 (1994) 1233–1243.
[141] G. Cullmann, K. Fien, R. Kobayashi, B. Stillman, Characterization of the five repli-
cation factor C genes of Saccharomyces cerevisiae, Mol. Cell Biol. 15 (1995) 4661–4671.
[142] B. Guenther, R. Onrust, A. Sali, M. O’Donnell, J. Kuriyan, Crystal structure of the
delta’ subunit of the clamp-loader complex of E. coli DNA polymerase III, Cell
91 (1997) 335–345.
[143] G.D. Bowman, M. O’Donnell, J. Kuriyan, Structural analysis of a eukaryotic sliding
DNA clamp-clamp loader complex, Nature 429 (2004) 724–730.
[144] D. Jeruzalmi, M. O’Donnell, J. Kuriyan, Crystal structure of the processivity clamp
loader gamma (gamma) complex of E. coli DNA polymerase III, Cell 106 (2001)
429–441.
[145] B.A. Kelch, D.L. Makino, M. O’Donnell, J. Kuriyan, How a DNA polymerase clamp
loader opens a sliding clamp, Science 334 (2011) 1675–1680.
[146] T. Miyata, H. Suzuki, T. Oyama, K. Mayanagi, Y. Ishino, K. Morikawa, Open clamp
structure in the clamp-loading complex visualized by electron microscopic image anal-
ysis, Proc. Natl. Acad. Sci. U.S.A. 102 (2005) 13795–13800.
[147] E.R. Goedken, S.L. Kazmirski, G.D. Bowman, M. O’Donnell, J. Kuriyan, Mapping
the interaction of DNA with the Escherichia coli DNA polymerase clamp loader com-
plex, Nat. Struct. Mol. Biol. 12 (2005) 183–190.
[148] K.R. Simonetta, S.L. Kazmirski, E.R. Goedken, A.J. Cantor, B.A. Kelch,
R. McNally, S.N. Seyedin, D.L. Makino, M. O’Donnell, J. Kuriyan, The mechanism
228 D. Zhang and M. O'Donnell

of ATP-dependent primer-template recognition by a clamp loader complex, Cell

137 (2009) 659–671.
[149] L.D. Langston, D. Zhang, O. Yurieva, R.E. Georgescu, J. Finkelstein, N.Y. Yao,
C. Indiani, M.E. O’Donnell, CMG helicase and DNA polymerase epsilon form a
functional 15-subunit holoenzyme for eukaryotic leading-strand DNA replication,
Proc. Natl. Acad. Sci. U.S.A. 111 (2014) 15390–15395.
[150] Z. You, M. De Falco, K. Kamada, F.M. Pisani, H. Masai, The mini-chromosome
maintenance (Mcm) complexes interact with DNA polymerase alpha-primase and
stimulate its ability to synthesize RNA primers, PLoS One 8 (2013) e72408.
[151] A. Costa, L. Renault, P. Swuec, T. Petojevic, J. Pesavento, I. Ilves, K. MacLellan-
Gibson, R.A. Fleck, M.R. Botchan, J.M. Berger, DNA binding polarity, dimeriza-
tion, and ATPase ring remodeling in the CMG helicase of the eukaryotic replisome,
Elife 3 (2014) e03273.
[152] M. Foltman, C. Evrin, G. De Piccoli, R.C. Jones, R.D. Edmondson, Y. Katou,
R. Nakato, K. Shirahige, K. Labib, Eukaryotic replisome components cooperate to
process histones during chromosome replication, Cell Rep. 3 (2013) 892–904.
[153] A.J. Tackett, D.J. Dilworth, M.J. Davey, M. O’Donnell, J.D. Aitchison, M.P. Rout,
B.T. Chait, Proteomic and genomic characterization of chromatin complexes at a
boundary, J. Cell Biol. 169 (2005) 35–47.
[154] T. Iida, H. Araki, Noncompetitive counteractions of DNA polymerase epsilon and
ISW2/yCHRAC for epigenetic inheritance of telomere position effect in Saccharomy-
ces cerevisiae, Mol. Cell Biol. 24 (2004) 217–227.
[155] T. Tsubota, R. Tajima, K. Ode, H. Kubota, N. Fukuhara, T. Kawabata, S. Maki,
H. Maki, Double-stranded DNA binding, an unusual property of DNA polymerase
epsilon, promotes epigenetic silencing in Saccharomyces cerevisiae, J. Biol. Chem.
281 (2006) 32898–32908.
[156] P. McInerney, A. Johnson, F. Katz, M. O’Donnell, Characterization of a triple DNA
polymerase replisome, Mol. Cell 27 (2007) 527–538.
[157] D. Gao, C.S. McHenry, tau binds and organizes Escherichia coli replication proteins
through distinct domains. Domain IV, located within the unique C terminus of
tau, binds the replication fork, helicase, DnaB, J. Biol. Chem. 276 (2001) 4441–4446.
[158] R. Reyes-Lamothe, D.J. Sherratt, M.C. Leake, Stoichiometry and architecture of
active DNA replication machinery in Escherichia coli, Science 328 (2010) 498–501.
[159] R.E. Georgescu, I. Kurth, M.E. O’Donnell, Single-molecule studies reveal the func-
tion of a third polymerase in the replisome, Nat. Struct. Mol. Biol. 19 (2011) 113–116.
[160] G. Lia, B. Michel, J.F. Allemand, Polymerase exchange during Okazaki fragment syn-
thesis observed in living cells, Science 335 (2012) 328–331.
[161] K. Tougu, K.J. Marians, The interaction between helicase and primase sets the repli-
cation fork clock, J. Biol. Chem. 271 (1996) 21398–21405.
[162] N.Y. Yao, R.E. Georgescu, J. Finkelstein, M.E. O’Donnell, Single-molecule analysis
reveals that the lagging strand increases replisome processivity but slows replication
fork progression, Proc. Natl. Acad. Sci. U.S.A. 106 (2009) 13236–13241.
[163] K.J. Hacker, B.M. Alberts, The rapid dissociation of the T4 DNA polymerase holo-
enzyme when stopped by a DNA hairpin helix. A model for polymerase release fol-
lowing the termination of each Okazaki fragment, J. Biol. Chem. 269 (1994)
24221–24228.
[164] Q. Yuan, C.S. McHenry, Cycling of the E. coli lagging strand polymerase is triggered
exclusively by the availability of a new primer at the replication fork, Nucleic Acids
Res. 42 (2014) 1747–1756.
[165] I. Kurth, R.E. Georgescu, M.E. O’Donnell, A solution to release twisted DNA during
chromosome replication by coupled DNA polymerases, Nature 496 (2013) 119–122.
The Eukaryotic Replisome 229

[166] J. Yang, S.W. Nelson, S.J. Benkovic, The control mechanism for lagging strand poly-
merase recycling during bacteriophage T4 DNA replication, Mol. Cell 21 (2006)
153–164.
[167] S.M. Hamdan, J.J. Loparo, M. Takahashi, C.C. Richardson, A.M. van Oijen, Dynam-
ics of DNA replication loops reveal temporal control of lagging-strand synthesis,
Nature 457 (2009) 336–339.
[168] F. Dong, S.E. Weitzel, P.H. von Hippel, A coupled complex of T4 DNA replication
helicase (gp41) and polymerase (gp43) can perform rapid and processive DNA strand-
displacement synthesis, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 14456–14461.
[169] S.M. Hamdan, D.E. Johnson, N.A. Tanner, J.B. Lee, U. Qimron, S. Tabor, A.M. van
Oijen, C.C. Richardson, Dynamic DNA helicase-DNA polymerase interactions
assure processive replication fork movement, Mol. Cell 27 (2007) 539–549.
[170] S. Kim, H.G. Dallmann, C.S. McHenry, K.J. Marians, Coupling of a replicative poly-
merase and helicase: a tau-DnaB interaction mediates rapid replication fork movement,
Cell 84 (1996) 643–650.
[171] M. Manosas, M.M. Spiering, F. Ding, V. Croquette, S.J. Benkovic, Collaborative
coupling between polymerase and helicase for leading-strand synthesis, Nucleic Acids
Res. 40 (2012) 6187–6198.
[172] A. Blinkova, C. Hervas, P.T. Stukenberg, R. Onrust, M.E. O’Donnell, J.R. Walker,
The Escherichia coli DNA polymerase III holoenzyme contains both products of the
dnaX gene, tau and gamma, but only tau is essential, J. Bacteriol. 175 (1993)
6018–6027.
[173] E. Dervyn, C. Suski, R. Daniel, C. Bruand, J. Chapuis, J. Errington, L. Janniere,
S.D. Ehrlich, Two essential DNA polymerases at the bacterial replication fork,
Science 294 (2001) 1716–1719.
[174] I. Bruck, M.F. Goodman, M. O’Donnell, The essential C family DnaE polymerase
is error-prone and efficient at lesion bypass, J. Biol. Chem. 278 (2003) 44361–44368.
[175] C.S. McHenry, Chromosomal replicases as asymmetric dimers: studies of subunit
arrangement and functional consequences, Mol. Microbiol. 49 (2003) 1157–1165.
[176] S.J. Benkovic, A.M. Valentine, F. Salinas, Replisome-mediated DNA replication,
Annu. Rev. Biochem. 70 (2001) 181–208.
[177] A. Calzada, B. Hodgson, M. Kanemaki, A. Bueno, K. Labib, Molecular anatomy and
regulation of a stable replisome at a paused eukaryotic DNA replication fork, Genes
Dev. 19 (2005) 1905–1919.
[178] T. Maculins, P.J. Nkosi, H. Nishikawa, K. Labib, Tethering of SCF(Dia2) to the
replisome promotes efficient ubiquitylation and disassembly of the CMG helicase,
Curr. Biol. 25 (2015) 2254–2259.
[179] Y.J. Sheu, J.B. Kinney, A. Lengronne, P. Pasero, B. Stillman, Domain within the
helicase subunit Mcm4 integrates multiple kinase signals to control DNA replication
initiation and fork progression, Proc. Natl. Acad. Sci. U.S.A. 111 (2014)
E1899–E1908.
[180] I. Ilves, N. Tamberg, M.R. Botchan, Checkpoint kinase 2 (Chk2) inhibits the activity
of the Cdc45/MCM2-7/GINS (CMG) replicative helicase complex, Proc. Natl.
Acad. Sci. U.S.A. 109 (2012) 13163–13170.
[181] A.E. Elia, A.P. Boardman, D.C. Wang, E.L. Huttlin, R.A. Everley, N. Dephoure,
C. Zhou, I. Koren, S.P. Gygi, S.J. Elledge, Quantitative proteomic atlas of
ubiquitination and acetylation in the DNA damage response, Mol. Cell 59 (2015)
867–881.
[182] J. McIntyre, M.P. McLenigan, E.G. Frank, X. Dai, W. Yang, Y. Wang, R. Woodgate,
Posttranslational regulation of human DNA polymerase iota, J. Biol. Chem.
290 (2015) 27332–27344.