MOLECULAR AND CELLULAR BIOLOGY, Aug. 1997, p. 4312–4321
0270-7306/97/$04.0010
Copyright © 1997, American Society for Microbiology
Vol. 17, No. 8
The Replication Origin Decision Point Is a Mitogen-Independent,
2-Aminopurine-Sensitive, G1-Phase Event That
Precedes Restriction Point Control
JIA-RUI WU
AND
DAVID M. GILBERT*
Department of Biochemistry and Molecular Biology, SUNY Health
Science Center, Syracuse, New York 13210
Received 21 November 1996/Returned for modification 15 January 1997/Accepted 5 May 1997
At a distinct point during G1 phase (the origin decision point [ODP]), Chinese hamster ovary (CHO) cell
nuclei experience a transition (origin choice) that is required for specific recognition of the dihydrofolate
reductase (DHFR) origin locus by Xenopus egg extracts. We have investigated the relationship between the ODP
and progression of CHO cells through G1 phase. Selection of the DHFR origin at the ODP was rapidly
inhibited by treatment of early G1-phase cells with the protein kinase inhibitor 2-aminopurine (2-AP). Inhibition of the ODP required administration of 2-AP at least 3 h prior to phosphorylation of the retinoblastoma
tumor suppressor protein (Rb) and the restriction point (R point). Cells deprived of either serum or isoleucine
from metaphase throughout early G1 phase acquired the capacity to replicate in Xenopus egg extract (replication licensing) and subsequently passed through the ODP on the same schedule as cells cultured in complete
growth medium. After growth arrest at the R point with hypophosphorylated Rb protein, serum- or isoleucinedeprived cells experienced a gradual loss of replication licensing. However, recognition of the DHFR origin by
Xenopus egg cytosol remained stable in growth-arrested cells until the point at which all nuclei had lost the
capacity to initiate replication. These results provide evidence that the ODP requires a mitogen-independent
protein kinase that is activated after replication licensing and prior to R-point control.
a strict requirement for initiation of replication (5). A second
round of replication requires permeabilization of the nuclear
membrane, exposure to egg cytosolic factors, and subsequent
reassembly of the membrane (4, 11). Similarly, when preformed mammalian nuclei are introduced into Xenopus egg
extracts, replication of DNA within G2-phase nuclei requires
permeabilization of the nuclear membrane, whereas impermeable G1-phase nuclei replicate efficiently (11, 33). Based on
these results, it has been proposed that the nuclear membrane
restricts entry of an essential replication licensing factor (RLF)
to the short period during mitosis when the nuclear membrane
is absent (4, 11). Destruction of RLF during S phase would
ensure that replication cannot reinitiate until the subsequent
mitosis. Thus, studies using Xenopus egg extracts have provided evidence that precise duplication of the genome requires
the coordinated action of (i) a replication licensing signal that
acts when the nuclear envelope is absent and (ii) an initiation
signal that requires a completely assembled nuclear envelope
and destroys RLF. As long as the licensing and initiation signals act sequentially, DNA will be replicated once per cell
cycle.
Which of these steps might interact with specific DNA sequences? In S. cerevisiae, the in vivo ORC footprint displays an
extended area of protection at the replication origin that appears during mitosis and disappears at the onset of S phase,
consistent with the assembly of an RLF complex at replication
origins during mitosis and destruction of that complex by the
initiation signal (15). This modification is dependent on the
product of the cdc6 gene (9). Although a Xenopus cdc6 homolog and other potential components of Xenopus RLF have
been identified (8, 10, 30, 36), as have the Xenopus ORC
homologs (7, 40), it has been impossible to demonstrate assembly of these proteins at specific DNA sequences, since
Xenopus egg extracts can efficiently initiate replication within
any DNA sequence (11). This apparent lack of sequence spec-
The initiation of eukaryotic DNA replication represents the
culmination of a series of decisions made during G1 phase. To
understand the regulatory mechanisms that control entry into
S phase, it is necessary to examine precisely the cell cycle stages
at which the replication machinery assembles. In Saccharomyces cerevisiae, progress has been facilitated by the identification
of a complex of six proteins (the origin recognition complex
[ORC]) that is constitutively bound to replication origins and
interacts with other gene products in a cell cycle-regulated
fashion (9). In higher eukaryotes, proteins homologous to the
various ORC subunits have been identified (7, 18, 21, 40).
However, origin-specific binding has not been demonstrated,
in a large part due to the inability to conclusively identify
specific DNA sequences that can function as origins of replication in multicellular organisms (11). Although replication
initiates within defined chromosomal loci in cultured mammalian cells, functional assays have failed to identify specific DNA
sequences that serve as the sites of assembly of regulatory
proteins (13). Without the availability of such sequences, it has
not been possible to study directly the assembly of the replication initiation complex.
Xenopus egg extracts are the most widely used cell-free system for studying the initiation of DNA replication in eukaryotes. These extracts appear to be capable of replicating
any DNA-containing substrate (e.g., purified DNA, chromatin,
whole cell nuclei) under cell cycle-regulated control. When
purified DNA templates are introduced into this system, they
are first assembled into chromatin and organized into nuclear
structures, followed by exactly one round of semiconservative
DNA replication (4). Assembly of an intact nuclear envelope is
* Corresponding author. Mailing address: Department of Biochemistry and Molecular Biology, SUNY Health Science Center, 750 East
Adams St., Syracuse, NY 13210. Phone: (315) 464-8723. Fax: (315)
464-8750. E-mail: [email protected].
4312
VOL. 17, 1997
ificity is not an artifact of the cell-free nature of this system;
initiation of replication within Xenopus embryonic chromosomes also occurs at apparently random chromosomal sites
(28). In this system then, the once-per-cell-cycle regulation of
DNA replication does not require the assembly of initiation
factors on specific replication origin sequences. Thus, it remains an open question as to whether the ORC and RLF
proteins specify sites of initiation of replication in higher eukaryotes or whether their role is restricted to providing the
potential, or license, to replicate.
A significant breakthrough in our understanding of how
specific replication origins are established came with the demonstration that Xenopus egg cytosol can recognize the Chinese
hamster ovary (CHO) dihydrofolate reductase (DHFR) origin,
provided that the substrate DNA is presented in the form of an
intact, late-G1-phase nucleus (19, 45). With intact early-G1phase nuclei, Xenopus egg extracts initiate replication at apparently random sites (45). At a discrete point during G1 phase
(the origin decision point [ODP]), nuclei undergo a transition
that specifies the DHFR origin, allowing for recognition of this
origin by Xenopus egg cytosol. Since pre-ODP nuclei were
shown to have completed the licensing step (45), recognition of
specific replication origins requires an additional regulatory
event that takes place during G1 phase, after replication licensing and prior to initiation. Interestingly, Xenopus early cleavage cycles lack a G1 phase, oscillating between the licensing
and initiation steps. In fact, replication in Xenopus embryos
becomes localized to specific sites after the blastula stage,
coincident with the appearance of a G1 phase (27). Taken
together, these results suggest that the potential to replicate is
acquired during mitosis through the assembly of ORC and
RLF, whereas the selection of specific replication origins requires passage through G1 phase.
Normal progression through G1 phase involves a complex
series of mitogen-stimulated signalling events, many of which
are mediated by protein kinases, and a period of high metabolic activity, culminating in passage through the restriction
point (R point). Cells can be arrested at multiple points during
early G1 phase through the administration of protein kinase
inhibitors (17, 20, 37), removal of serum mitogens, or inhibition of protein synthesis (38, 49). However, once cells pass
through the R point, entry into S phase becomes independent
of these signals (38, 39, 41, 48, 49). One molecular target of
these early G1 signalling events is the retinoblastoma gene
product (Rb), likened to a gatekeeper for S-phase entry (43).
In its hypophosphorylated form early in G1 phase, Rb tightly
associates with several transcription factors that are required
for entry into S phase, preventing their function or causing
them to function as repressors (42). In late G1, Rb is phosphorylated by the cyclin-dependent kinases (cdks), dissociating
Rb and unleashing these activators to perform their function.
As part of our ongoing efforts to characterize the mechanism
of origin selection at the ODP, we sought to place the ODP
into the temporal scheme of previously identified G1-phase
regulatory events. Here we show that the protein kinase inhibitor 2-aminopurine (2-AP), which has been shown to block
serum-stimulated induction of the proto-oncogenes c-fos and
c-myc (50) and to inhibit phosphorylation of Rb (37), inhibits
selection of the CHO DHFR origin when administered just
prior to the ODP. However, both the R point and Rb phosphorylation occur several hours after the ODP. Furthermore,
we show that cells deprived of serum or isoleucine from metaphase throughout G1 phase undergo replication licensing, proceed through the ODP transition on schedule, and then arrest
at the R point with hypophosphorylated Rb protein. Thus, it
appears that the ODP is a response to a protein kinase signal-
REPLICATION ORIGIN DECISION POINT
4313
ling pathway that is activated early in G1 phase and proceeds
independent of serum growth factors or high levels of protein
synthesis.
MATERIALS AND METHODS
Cell culture and synchrony. CHOC 400 is a derivative of CHO cells that
contains ;1,000 tandemly integrated copies of a 243-kb segment of DNA surrounding the DHFR gene (23), significantly increasing the ability to detect
initiation of replication. The pattern of initiation within the amplified locus is
indistinguishable from that within the parental CHO cell line (16). CHOC 400
cells were cultured in Dulbecco modified Eagle medium supplemented with
nonessential amino acids and 5% fetal bovine serum. Synchronization of cells in
metaphase and selection by mitotic shake-off were performed as described previously (19). Synchronization of exponentially growing CHOC 400 cells by isoleucine deprivation was performed as described previously (6). For synchronization by serum starvation, exponentially growing cells were washed twice with
serum-free Dulbecco modified Eagle medium and incubated in serum-free medium for 4 days (96 h). For experiments with 2-AP, mitotic cells were first plated
to complete medium. At the indicated times, medium was replaced with medium
containing 2-AP (free base; Sigma A-3509) dissolved directly in growth medium
and adjusted to pH 7.2. Consistent with previously reported results (2), 2-AP
treatment of synchronized cells prior to anaphase resulted in the failure of
proper chromosome segregation and the formation of micronuclei. Thus, 2-AP
was added no earlier than 1.5 h after mitosis.
Western blotting. A total of 1.5 3 106 cells synchronized in metaphase were
plated to 60-mm-diameter culture dishes. Dishes were washed twice in ice-cold
phosphate-buffered saline and lysed for 5 min on ice in 100 ml of ice-cold lysis
buffer (50 mM Tris [pH 8.0], 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, 1
mM sodium vanadate, 50 mM NaF, 1 mM phenylmethylsulfonyl fluoride, 1 mg of
pepstatin per ml, 1 mg of aprotinin per ml, 2 mg of chymostatin per ml). The
lysate was centrifuged for 5 min at 16,000 3 g and 4°C, and the cleared supernatant (100 ml of cell lysate at 10 to 15 mg/ml of total protein) was frozen at
270°C. For Rb detection, the conditions were as follows. Sixty micrograms of
total protein from each sample was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with a 4% stacking gel and a 6.5%
resolving gel. Protein was transferred to an Immobilon-P transfer membrane as
instructed by the manufacturer (Millipore). Membranes were incubated in block
buffer (20 mM Tris [pH 7.5], 137 mM NaCl, 0.05% Tween 20, 5% dried milk, 10
mM NaN3) for 1 h at room temperature with gentle shaking and washed twice for
5 min in TBS-T (20 mM Tris [pH 7.5], 137 mM NaCl, 0.05% Tween 20).
Membranes were then incubated for 6 h at room temperature with 1.5 mg of
rabbit polyclonal anti-Rb (C-15; Santa Cruz Biotechnology) per ml in block
buffer and washed three times for 5 min each in TBS-T. Membranes were then
incubated for 1 h at room temperature with anti-rabbit horseradish peroxidaseconjugated immunoglobulin G (Sigma) in block buffer (1:2,000 dilution) and
washed four times for 5 min each in TBS-T, and chemiluminescence was detected according to the Amersham ECL protocol. Cyclin A was analyzed similarly except that SDS-PAGE was performed with a 5% stacking gel and 10%
resolving gel and cyclin A protein was detected with rabbit polyclonal anti-cyclin
A (C-19; Santa Cruz Biotechnology) at 2.0 mg/ml of block buffer.
Measuring the R point. CHOC 400 cells (105) synchronized in metaphase were
plated onto coverslips in 24-well dishes. At the indicated times, cells in individual
wells were washed twice with serum-free or isoleucine-free medium and then
further incubated in the same medium. At 13.5 h after mitosis, 30 mg of 5-bromodeoxyuridine (BrdU) per ml was added to the medium of all wells for 30 min,
the medium was removed, and cells were fixed by the addition of ice-cold 70%
ethanol. Cells were stained with anti-BrdU as described previously (19). For
experiments with isoleucine-free medium, cell growth arrest required a critical
density of cells (1.2 3 105/well), below which a significant fraction of cells were
capable of entering S phase. Cells for control experiments that determined the
time of entry of cells into S phase in complete medium were prepared similarly
except that cells were incubated in complete medium continuously and pulselabeled for 30 min with BrdU every hour.
Preparation of intact nuclei and analysis of DNA synthesis in Xenopus egg
extract. Cells were permeabilized for 5 min at 0°C at 5 million per ml of transport
buffer (19) in a final concentration of digitonin that varied from 20 to 80 mg/ml,
depending on the digitonin batch. Permeabilization was stopped with an equal
volume of transport buffer containing 3% bovine serum albumin, and the percentage of nuclei that excluded labeled immunoglobulin G (intact nuclei) was
evaluated as described previously (45). Permeable nuclei were prepared as for
intact nuclei except that both the concentration of digitonin and the time of
incubation at 0°C were doubled. Xenopus egg extracts were prepared and handled as described previously (3). Conditions for measuring DNA synthesis by
acid precipitation and incorporation of BrdU triphosphate (BrdUTP) were as
described previously (19).
ELFH assay after serum or isoleucine starvation. CHOC 400 cells (3 3 106)
synchronized in metaphase were washed twice with serum-free or isoleucine-free
medium in a 1.5-ml tube at 6,000 rpm for 10 s in a microcentrifuge. Cells were
plated to 100-mm-diameter dishes containing the appropriate medium and incubated for either 4 or 14 h. Cells were then collected with trypsin and perme-
4314
WU AND GILBERT
abilized with digitonin, and nuclei were incubated in Xenopus egg cytosol supplemented with aphidicolin (100 mg/ml) for 2 h at 21°C. Nuclei were washed
three times in hypotonic buffer, labeled for 10 min at 12°C, and analyzed by early
labeled fragment hybridization (ELFH) as previously described (19) and summarized in the legend to Fig. 1.
On occasion, cells do not completely arrest in G1 phase in either serum- or
isoleucine-free medium. This was true for both the short-term protocol (4 and
14 h after mitosis) and the long-term protocol (36 or 96 h from an exponentially
growing culture). This may have to do with intracellular pools of metabolites,
concentration of cells, or other unknown variables. Therefore, a coverslip was
included in all plates. Thirty minutes prior to collection, the coverslip with
attached cells was transferred to a 35-mm-diameter dish, and cells were pulselabeled with BrdU for 30 min, fixed, and stained as described above. Only
populations of cells that had less than 5% S-phase contamination were considered for results presented in this report.
RESULTS
2-AP inhibits origin choice. To examine the possibility that
a protein kinase might regulate passage through the ODP,
CHOC 400 cells were synchronized in metaphase by brief (4-h)
exposure to nocodazole and collected by mitotic shake-off
(.98% mitotic figures). Cells were then released into G1 phase
for 1.5 h and shifted to medium containing different concentrations of 2-AP. The ODP has been shown to consistently take
place between 3 and 4 h after mitosis in CHOC 400 cells (45).
Thus, passage through the ODP can be monitored by preparing intact nuclei from cells at 4 h after metaphase, stimulating
nuclei to enter S phase by incubation in Xenopus egg extract,
and mapping the sites of initiation of replication at the DHFR
locus (Fig. 1).
We first determined whether 2-AP treatment affected the
efficiency of DNA synthesis in Xenopus egg extracts. CHOC
400 cells were treated with 10 mM 2-AP at 1.5 h and collected
at 4 h (4-h12-AP nuclei). As a control, untreated cells were
collected at 1.5 h (1.5-h nuclei) and at 4 h (4-h nuclei). Intact
nuclei prepared from these cells were introduced into Xenopus
egg extract supplemented with [a-32P]dATP, and DNA synthesis was monitored by the accumulation of radiolabel into
acid-precipitable DNA (Fig. 1C). DNA synthesis with all three
nuclear preparations began after a 20-min lag period and proceeded at the same rate, demonstrating that the assembly of
initiation complexes in the extract was not impeded by the
prior treatment of cells with 2-AP. DNA synthesis was also
sensitive to the addition to the extracts of the protein kinase
inhibitor 6-dimethylaminopurine (6-DMAP), which specifically inhibits the initiation of replication (3, 19, 32, 45), indicating that DNA synthesis in all cases results from initiation of
replication and not the elongation of preexisting replication
forks.
The sites of initiation of replication within the DHFR locus
were then mapped by using the ELFH assay. The ELFH assay
quantifies the distribution of nascent DNA at various positions
within the DHFR locus shortly after the initiation of DNA
synthesis (19, 32, 45). The 1.5-h, 4-h, and 4-h12-AP nuclei
were introduced into Xenopus egg extract supplemented with
aphidicolin for 2 h. Aphidicolin is an inhibitor of replicative
polymerases that inhibits the elongation of replication forks
after the establishment of 200- to 500-bp nascent DNA chains
(reference 19 and references therein). Thus, this procedure
allows initiation of replication and the accumulation of newly
formed replication forks arrested close to their sites of initiation by aphidicolin. Nuclei were then washed free of aphidicolin, nascent DNA chains were labeled briefly with [a-32P]
dATP, and the resulting [32P]DNA chains were hybridized to
15 unique probes distributed over a 130-kb region that includes
the DHFR ori-b. To control for differences in probe size,
deoxyadenosine content of the segments analyzed (which
would influence specific activity), and the hybridization effi-
MOL. CELL. BIOL.
ciency of genomic DNA to each probe, replication intermediates were also labeled in nuclei from exponentially growing
CHOC 400 cells (which serve as a pool of replication forks
distributed randomly throughout the DHFR locus) and then
hybridized to these same probes (Fig. 1A). The counts per
minute hybridized to each probe was then corrected for this
variation by calculating the ratio of [32P]DNA per probe with
synchronized nuclei to the corresponding value for exponentially proliferating nuclei. The relative number of origin-proximal replication intermediates at each map position can then
be plotted as shown in Fig. 1B. Consistent with our previously
reported results (19, 32, 45), with intact 4-h nuclei as a substrate (post-ODP), Xenopus egg extracts initiated replication
within the ori-b initiation locus at sites distributed in a pattern
indistinguishable from that of cultured cells (cultured cell pattern shown in references 19, 32, and 45), whereas with 1.5-h
(pre-ODP) or 4-h12-AP nuclei, replication initiated at sites
dispersed throughout the 130-kb region, indicating that the
administration of 2-AP prevented passage through the ODP.
To construct a dose-response curve of the ODP to 2-AP, the
relative specificity of initiation within the DHFR locus was
defined as the average relative early DNA synthesis values for
probes B through E (highlighted by the shaded vertical line in
Fig. 1B), and a series of experiments, similar to the one shown
in Fig. 1A to C, were performed with various concentrations of
2-AP. Results revealed that at concentrations at or above 10
mM, 2-AP consistently prevented the selection of replication
origins at the ODP. This is in agreement with the concentrations of 2-AP previously reported to affect other aspects of
cellular metabolism in other cell lines (1, 37, 50).
To establish a time frame during which the administration of
2-AP is effective at inhibiting the ODP, 10 mM 2-AP was
added to cell cultures at various times after metaphase, and
intact nuclei were prepared from 2-AP-treated cells at 4 h and
introduced into Xenopus egg extract for evaluation of the specificity of initiation as in Fig. 1. Results revealed (Fig. 2) that
2-AP treatment exerted an inhibitory effect on the passage of
cells through the ODP within 30 min prior to the completion of
the ODP transition within cells. In fact, the pattern of initiation
of replication within nuclei treated with 2-AP at different times
during G1 phase closely resembles the pattern of initiation of
replication within untreated nuclei isolated at these same times
(45), suggesting that the inhibitory action of 2-AP is rapid and
is exerted very close to or coincident with the time of the ODP
transition. Similar results were obtained when cells were treated with 3 mM 6-DMAP (data not shown).
The ODP occurs prior to phosphorylation of Rb. Rb, a
putative participant in R-point control, is phosphorylated in
mid-G1 phase, and this phosphorylation event is believed to be
one of the most important signals that drives cells into S phase
(42, 43). To determine whether selection of the DHFR replication origin at the ODP is upstream or downstream of this
important cell cycle hallmark, the phosphorylation of Rb during G1 phase was examined by monitoring the shift in molecular mass of this protein from 110 to 116 kDa that occurs upon
phosphorylation (reference 43 and references therein). CHOC
400 cells were synchronized in metaphase and released into G1
phase, and samples were collected every hour for Western
blotting analysis. Results (Fig. 3A) showed that Rb is dephosphorylated within 1 h after metaphase, consistent with other
reports (35), and is phosphorylated between 6 and 8 h after
metaphase. Thus, the ODP (3 to 4 h after metaphase) occurs
at least 3 h prior to Rb phosphorylation.
Levels of cyclin A protein were also monitored in these same
cells. Transcription of the cyclin A gene is induced by cyclin
E/cdk2 at the onset of S phase, and protein levels rise during S
VOL. 17, 1997
REPLICATION ORIGIN DECISION POINT
4315
FIG. 1. The ODP is inhibited by 2-AP. (A) CHOC 400 cells were synchronized in metaphase and plated into complete medium. At 1.5 h after metaphase, one set
of cells was collected (1.5 Hr Nuclei), and a second set was transferred to medium containing 10 mM 2-AP and collected at 4 h (4 Hr 1 2-AP). Control cultures remained
in drug-free medium for 4 h (4 Hr Nuclei). Cells were permeabilized with digitonin, and the resulting intact nuclei were incubated for 2 h in a Xenopus egg extract
supplemented with aphidicolin. Nuclei were then washed free of aphidicolin, and the earliest-replicating nascent DNA chains were labeled briefly with [a-32P]dATP.
Fifteen unique probes distributed over a 130-kb region that includes the DHFR ori-b were immobilized to nylon filters with a slot blot apparatus and hybridized to
the 32P-labeled early replication intermediates as described previously (19). Relative counts per minute were obtained by phosphorimaging analysis (column labeled
“Relative cpm”). To correct for variation between probes due to differences in probe size, deoxyadenine content, and hybridization efficiency, labeled nascent DNA
from exponentially proliferating CHOC 400 cells was hybridized to the same probes. The relative value for each probe was divided by the corresponding value for
exponentially proliferating nuclei (column labeled “Corrected cpm”). These values were then normalized by adjusting the lowest corrected value to 1.00 (column labeled
“Rel. Early DNA Syn.”). Probe l is a segment of bacteriophage lambda DNA included in each experiment to evaluate the degree of nonspecific hybridization. (B)
Relative early DNA synthesis values, defined above, were plotted versus the map positions of each probe for 1.5-h nuclei (}), 4-h12-AP nuclei (■), and 4-h nuclei (F).
The horizontal axis includes a diagram of the portion of the DHFR locus encompassed by these probes, including the positions of the DHFR and 2BE2121 genes (22).
The vertical shaded line shows the position of the previously mapped origin of bidirectional replication (OBR). (C) Aliquots of nuclei from panels A and B were
introduced into Xenopus egg extracts supplemented with [a-32P]-dATP in either the absence (closed symbols labeled as in panel B) or presence (open symbols) of 3
mM 6-DMAP, and the percentage of input DNA replicated was determined at the indicated times by acid precipitation (e.g., 100% DNA synthesis would indicate
replication of all CHOC 400 genomic DNA). (D) CHOC 400 cells were synchronized in mitosis and released into G1 phase in complete medium. At 1.5 h after
metaphase, cells were transferred to medium containing the indicated concentration of 2-AP and incubated for an additional 2.5 h (until 4 h after mitosis). Intact nuclei
prepared from these cells were incubated in Xenopus egg extract supplemented with aphidicolin, and the sites of initiation of replication were mapped by using the
ELFH assay as for panel A. The relative specificity of initiation was then defined as the average relative early DNA synthesis values for probes B through E (highlighted
by the grey line in panel B) and plotted versus the concentration of 2-AP. Shown are the mean values from four independent experiments and the standard deviation
about the mean.
4316
WU AND GILBERT
MOL. CELL. BIOL.
FIG. 2. The inhibitory effect of 2-AP coincides with the ODP. CHOC 400 cells were synchronized in metaphase and released into complete medium as in Fig. 1.
At the indicated times, cells were transferred to medium containing 10 mM 2-AP, and incubation was continued until 4 h after metaphase. Intact nuclei prepared from
these cells were incubated in Xenopus egg extract supplemented with aphidicolin, and the sites of initiation of replication were mapped by using the ELFH assay as
in Fig. 1. Shown are the mean values of four independent experiments and the standard deviation about the mean (when greater than 1). The vertical shaded line shows
the position of the previously mapped origin of bidirectional replication.
phase (references 25 and 47 and references therein). Levels of
cyclin A protein in synchronized CHOC 400 cells were found
to be induced shortly after Rb phosphorylation (Fig. 3A), consistent with our finding that CHOC 400 cells enter S phase 10
to 11 h after metaphase (reference 45 and Fig. 3B).
The R point for both serum and isoleucine occurs 3 to 5 h
after the ODP. The R point was originally defined as the point
at which entry into S phase and progress through the remainder of the cell cycle becomes independent of serum growth
mitogens and resistant to moderate levels of protein synthesis
inhibitors (38, 48, 49). To examine the R point for growth
mitogens, it is sufficient to deprive cells of serum supplements
in their growth medium. Measuring the point after which cells
become resistant to protein synthesis inhibitors is more complicated, requiring the titration of inhibitors such as cycloheximide to achieve inhibition of 50 to 75% of protein synthesis
during early G1 phase (38). Another method that has proven
effective in arresting many cell lines, including CHOC 400, at
the R point is to inhibit protein synthesis by removing isoleucine from the growth medium (6). Although the removal of
serum and the inhibition of protein synthesis are thought to
measure the same control point, they may not act on the same
regulatory pathway. In fact, murine embryonic fibroblasts isolated from Rb2/2 mice have been shown to lack the amino acid
R point but retain the serum R point, demonstrating that these
arrest points can be uncoupled (24). For this reason, we will
refer to them separately as the serum R point and the amino
acid R point.
Thus, to determine the time during G1 phase at which these
R-point controls are exerted, CHOC 400 cells were synchronized in metaphase and plated into prewarmed medium supplemented with 5% fetal bovine serum (complete growth medium). At hourly intervals thereafter, the culture medium in
separate cultures was replaced with medium lacking either
serum or isoleucine. At 14 h after metaphase, a time when 75
to 85% of cells grown in complete medium are actively syn-
thesizing DNA, all cultures were pulse-labeled with the thymidine analog BrdU, and the percentage of cells in each group
that had proceeded into S phase and incorporated BrdU was
evaluated by indirect immunofluorescence using an anti-BrdU
antibody. Results (Fig. 3B) demonstrated that (i) the serum R
point is several hours after the ODP, (ii) the amino acid R
point occurs very close to the time of serum R-point control,
and (iii) R-point control occurs at or close to the time of Rb
phosphorylation in CHOC 400 cells.
The ODP is isoleucine and serum independent. The data in
Fig. 3 demonstrate that the R point is downstream of the ODP.
However, they do not allow us to evaluate whether the absence
of serum or isoleucine during the early stages of G1 phase has
an effect on origin choice at the ODP. For example, serum or
isoleucine deprivation at early times in G1 phase could arrest
cells at those earlier points. Alternatively, the ODP transition
might require the continued exposure of cells to growth mitogens, even though it occurs upstream of the R point. Thus, we
adopted the following strategy. CHOC 400 cells were synchronized in metaphase and plated directly into medium that
lacked either serum or isoleucine. Nuclei were prepared from
these cells at 4 h after metaphase, previously shown to be the
earliest time during G1 phase at which origin choice is complete (45), and at 14 h after metaphase, when 75 to 85% of cells
will normally have entered into S phase. These nuclei were
then introduced into Xenopus egg extracts to evaluate their
capacity to replicate DNA (replication licensing) as well as the
sites of initiation of replication (origin choice). The expectations from this strategy were as follows. If serum or isoleucine
deprivation affects origin choice at the ODP, then at 4 h, a
difference in origin specificity should be detected relative to
nuclei isolated from control cultures plated into complete
growth medium. By contrast, if these growth conditions do not
affect origin choice, results should be identical between these
cells and control cultures. By 14 h after metaphase, control
cultures will have already entered S phase, and we would
VOL. 17, 1997
REPLICATION ORIGIN DECISION POINT
4317
of 6-DMAP-sensitive DNA synthesis with nuclei from both
serum- and isoleucine-starved cell populations was as efficient
as with nuclei from control cultures, demonstrating that replication licensing occurred in the absence of serum or isoleucine.
At 14 h, nuclei from starved cells were still competent for
replication, although the efficiency of replication (reflected in
the slope of the curve) was slightly less than in control cultures
(Fig. 4D). DNA synthesis within nuclei from starved cultures
initiated after a 20-min lag period and was sensitive to
6-DMAP, consistent with the arrest of these cells in G1 phase,
prior to the establishment of replication forks. By contrast,
DNA synthesis within nuclei from control cultures at 14 h
began immediately upon contact with the extract and was to a
large extent resistant to 6-DMAP, consistent with the entry of
these cells into S phase and the extension of preexisting replication forks in the extract (reference 19 and Fig. 4D).
To evaluate whether the slight reduction in the efficiency of
replication with starved cultures was due to a reduced efficiency of DNA synthesis within all nuclei or to a reduction in
FIG. 3. The growth factor and amino acid R points coincide with Rb phosphorylation and occur several hours after the ODP. (A) CHOC 400 cells were
synchronized in metaphase by mitotic shake-off and plated into complete medium for the indicated lengths of time. Cell lysates were prepared, subjected to
SDS-PAGE, and analyzed by Western blotting using polyclonal antibodies recognizing either hamster Rb (top) or hamster cyclin A (bottom). Phosphorylation
of Rb protein product is detected as a shift in apparent molecular mass from 110
to 116 kDa, as indicated. Cyclin A migrated with an apparent molecular mass of
60 kDa. (B) CHOC 400 cells were synchronized in metaphase by mitotic shakeoff and plated into complete medium. At the indicated times, medium was
removed and replaced with medium lacking either serum (M
y ) or isoleucine (}).
At 14.5 h after metaphase, BrdU was added to the medium for 30 min. Cells were
then fixed, and the percentage of nuclei that had progressed into S phase was
determined by using indirect immunofluorescence with anti-BrdU antibody. As
a control for the normal entry of cells into S phase, duplicate plates of cells grown
in complete medium were pulse-labeled with BrdU for 30 min at each hour, and
the percentage of BrdU-labeled nuclei was determined (F).
expect that replication forks will have emanated away from
their sites of origin. With nuclei that are arrested in G1 phase
due to the lack of serum or isoleucine, Xenopus egg extract
would initiate nonspecifically if cells arrested upstream of the
ODP (pre-ODP) and site specifically if cells arrested downstream of the ODP (post-ODP).
First, it had to be demonstrated that the lack of serum and
isoleucine did not prevent replication licensing. Xenopus egg
extract will not initiate DNA replication within intact nuclei
that are not licensed to replicate (33, 34); hence, it would not
be possible to map the sites of replication initiation. As expected, CHOC 400 cells synchronized in mitosis and released
into G1 phase in the absence of either serum or isoleucine did
not phosphorylate Rb or induce cyclin A (Fig. 4A) and did not
enter S phase (Fig. 4B). Intact nuclei were prepared from these
cells at 4 and 14 h after mitosis and introduced into a Xenopus
egg extract supplemented with [a-32P]dATP, and DNA synthesis was monitored by the accumulation of radiolabel into
acid-precipitable DNA (Fig. 4C and D). At 4 h, the initiation
FIG. 4. Cells arrested upstream of Rb phosphorylation by exposure to serum- or isoleucine-free medium throughout G1 phase are licensed to replicate.
CHOC 400 cells were synchronized in mitosis and plated into medium lacking
either serum or isoleucine. As a control, duplicates were plated into complete
medium. (A) At 4 and 14 h, cells were collected for analysis of the phosphorylation state of Rb and the presence of cyclin A as in Fig. 3A. (B) At 3.5 and
13.5 h, BrdU was added to the cultures, and 30 min later, the frequency of
labeled nuclei was determined as in Fig. 3B. (C and D) Cells were synchronized
in mitosis and plated into medium lacking either serum (M
y ) or isoleucine (}) or
into complete medium (H). At 4 h (C) and 14 h (D) after metaphase, intact
nuclei were introduced into Xenopus egg extracts supplemented with [a-32P]
dATP in either the absence (closed symbols) or presence (open symbols) of 3
mM 6-DMAP, and the percentage of input DNA replicated was determined as
in Fig. 1C. (E) The same intact nuclei as in panels C and D were incubated for
2 h in Xenopus egg extracts supplemented with 500 mM BrdUTP, and the
percentage of nuclei that incorporated BrdU was determined (Intact Nuclei). In
addition, 14-h nuclei were permeabilized by prolonged exposure to digitonin and
analyzed in parallel (Permeable Nuclei).
4318
WU AND GILBERT
the number of nuclei that retained the potential to initiate
replication in the extract (licensed nuclei), 4- and 14-h nuclei
were incubated in Xenopus egg extract supplemented with
BrdUTP, and the percentage of nuclei that incorporated
BrdUTP was evaluated (Fig. 4E). Virtually all 4-h nuclei from
either starved or control cultures were competent for replication. At 14 h about 67% of the nuclei from starved cultures
were competent for replication in the extract. This proportion
of licensed nuclei accounts adequately for the reduction in the
efficiency of DNA synthesis detected in Fig. 4D. The remaining
33% appear to have lost the license to replicate. Permeabilization of 14-h nuclei restored the capacity to replicate in Xenopus egg extract to these nuclei (Fig. 4E, Permeable Nuclei).
Thus, a proportion of serum- or isoleucine-arrested cells lose
licensing, while the majority remain arrested in G1 phase fully
competent to initiate replication.
Next, we wanted to verify our assumption that the ELFH
assay would distinguish between cells arrested in the post-ODP
stages of G1 phase and cells that simply did not arrest at the R
point and entered S phase in culture. We predicted that as cells
enter S phase, replication forks will emanate away from their
sites of origin and specificity will no longer be detectable with
the ELFH assay. Figure 5 shows the results of an experiment in
which nuclei were isolated from cells synchronized at several
time points after metaphase and the sites of earliest DNA
synthesis in Xenopus egg extract mapped by the ELFH assay.
At 1.5 h, no origin specificity is detected, as these pre-ODP
nuclei have not undergone the transition to specific initiation,
consistent with results in Fig. 1 and previous data (45). At 4 h,
the commitment to initiate replication at the DHFR origin was
completed, and it was maintained at least through 6 h after
metaphase (Fig. 5). By 7 and 8 h after metaphase, some cells
(4% and 10%, respectively, in this experiment) have begun to
enter S phase, and the positions of nascent DNA are less
specifically defined, consistent with a fraction of forks having
already moved away from their sites of origin at the time of
incubation in Xenopus egg cytosol. Finally, by 14 h after metaphase, most cells (78% in this experiment) have entered S
phase, and the majority of replication forks appear to have
emanated away from the DHFR replication origin. Thus, mapping the positions of nascent DNA strands after incubation of
nuclei in Xenopus egg cytosol provides a means to determine
whether cells are arrested in G1 phase and whether this arrest
is in the pre-ODP or post-ODP stages of G1 phase.
Having established the above-described procedures to examine the progression of cells through G1 phase and into S
phase, the effect of serum and isoleucine deprivation on the
selection of origins of replication at the ODP was evaluated.
CHOC 400 cells were synchronized in metaphase and released
into medium lacking either serum or isoleucine for 1.5, 4, or
14 h. Control cultures were incubated in complete medium and
collected at those same time points. Nuclei were isolated from
these cells and incubated in Xenopus egg extract, and the
positions of nascent DNA strands were mapped by the ELFH
assay (Fig. 6). At 1.5 h, no origin specificity is detected, demonstrating that these pre-ODP nuclei have not undergone the
transition to specific initiation under any of these growth conditions. By 4 h, cells deprived of serum and isoleucine had
passed through a normal ODP transition into G1 phase. At
14 h, the commitment to initiate at the DHFR origin remained
stable in arrested cells, while in control cultures, replication
forks had emanated away from the DHFR origin. Since cells
arrested in deficient medium contained only hypophosphorylated Rb (Fig. 4A) and did not enter S phase (Fig. 4B and 6),
we conclude that these cells are arrested prior to the R point
and in the post-ODP stage of G1 phase.
MOL. CELL. BIOL.
FIG. 5. Nuclei from pre-ODP, post-ODP, and S-phase CHOC 400 cells can
be distinguished by using the ELFH assay. CHOC 400 cells were synchronized in
mitosis and plated into complete medium for the indicated lengths of time.
Nuclei were then prepared from these cells and incubated for 2 h in Xenopus egg
extract supplemented with aphidicolin, and the positions of the earliest nascent
DNA strands were mapped by the ELFH assay as in Fig. 1A. The relative early
DNA synthesis values for early replication intermediates in nuclei from cells
synchronized at the indicated times after metaphase were plotted versus the map
positions of each probe as in Fig. 1B. The vertical shaded line shows the position
of the previously mapped origin of bidirectional replication.
Origin choice remains stable until R-point arrest leads to a
loss of replication licensing. Results from Fig. 6 raised the
possibility that long-term exposure of exponentially growing
populations of cells to serum- or isoleucine-free medium might
result in synchronization of cells in the post-ODP stage of G1
phase. Alternatively, if the nuclear event that leads to origin
choice at the ODP is the result of the assembly of an unstable
complex, this complex could decay after prolonged periods of
G0 arrest. To distinguish between these two possibilities, exponentially growing populations of CHOC 400 cells were arrested by long-term exposure to deficient medium. Experimental conditions were adjusted to achieve the minimal exposure
time that resulted in the presence of less than 5% S-phase cells,
as measured by the uptake of BrdU in a 30-min pulse-label
(not shown). This was achieved within 36 h in isoleucine-free
medium. Exposure of cells to serum-free medium required
96 h to achieve sufficient synchrony. Thus, we compared exponentially growing populations of cells that were deprived of
isoleucine for 36 h with those of cells exposed to serum-deficient medium for 96 h. Results (Fig. 7) revealed that cells
arrested by serum starvation had completely lost their competence to enter S phase when their nuclei were exposed to
Xenopus egg cytosol, making origin mapping impossible. These
cells were judged to have lost licensing, as (i) permeabilization
of these nuclei rendered them competent for replication (Fig.
7B and C), consistent with previous results (33, 34), and (ii) the
VOL. 17, 1997
REPLICATION ORIGIN DECISION POINT
4319
the potential for nuclei to replicate but does not dictate the
sites of initiation of DNA replication. These results reveal an
order of G1-phase events that leads up to the initiation of DNA
replication, which in CHOC 400 cells includes (i) licensing of
chromatin during or shortly after mitosis without the selection
of specific replication origins, (ii) a distinct ODP at 3 to 4 h
after mitosis that selects origin sites, (iii) phosphorylation of
Rb and passage through the R point at 6 to 8 h after mitosis,
and (iv) initiation of replication and entry into S-phase at 9 to
12 h after mitosis.
Rb phosphorylation and the R point. In the course of the
experiments described in this report, we have simultaneously
determined the timing of Rb phosphorylation and the R point
in cells progressing through a normal G1 phase in culture.
Most studies of Rb phosphorylation have been performed with
quiescent cells that were stimulated to reenter the cell cycle
FIG. 6. Cells deprived of serum or isoleucine throughout early G1 pass
through the ODP on schedule and become arrested in the post-ODP stage of G1
phase. CHOC 400 cells were synchronized in mitosis and plated into medium
lacking either serum (2 Serum) or isoleucine (2 Ile), or into complete medium
(Control) for either 1.5 h (}), 4 h (M
y ), or 14 h (H). Nuclei were then prepared
from these cells and incubated for 2 h in Xenopus egg extract supplemented with
aphidicolin, and the positions of the earliest nascent DNA strands were mapped
by the ELFH assay as in Fig. 1. Shown are the mean values from four independent experiments and the standard deviation about the mean (when greater than
1). The vertical shaded line shows the position of the previously mapped origin
of bidirectional replication.
capacity to replicate in Xenopus egg extract could be restored
in all nuclei after stimulating these quiescent cells to reenter
the cell cycle by returning them to complete medium for 12 h
(Fig. 7C). Thus, we conclude that these were living cells, arrested in a G0 state.
By contrast, cells synchronized by isoleucine starvation for
36 h retained a sufficient number of replication-competent
cells to map the initiation of DNA replication in Xenopus egg
extract. These cells were found to be arrested in the post-ODP
stage of G1 phase (Fig. 7D). Taken together with the results
from Fig. 6, these results lead us to conclude that in the absence of isoleucine or serum, cell nuclei first become licensed
to replicate as they enter G1 phase, they pass through the ODP
on schedule, and then they become arrested at the R point.
After R-point arrest, individual cells gradually lose licensing,
retaining their commitment to initiate at the DHFR origin
until licensing is lost.
DISCUSSION
As part of our ongoing effort to understand the nature of the
ODP, one approach has been to stage the ODP with respect to
various G1-phase events. Results reported here demonstrate
that the commitment to initiate DNA replication specifically at
the CHO DHFR replication origin requires a mitogen-independent protein kinase activity that is activated early in G1
phase, prior to Rb phosphorylation and R-point control. Furthermore, we have provided additional evidence that replication licensing occurs during or shortly after mitosis to provide
FIG. 7. Origin choice remains stable within those nuclei that retain licensing
after R-point arrest. (A) Exponentially growing CHOC 400 cells were exposed to
isoleucine-free medium for 36 h (M
y ) or serum-free medium for 96 h (}). As a
control, cells were synchronized in post-ODP G1 phase (4 h after mitosis [Control]) (H). Nuclei from these cells were then incubated in Xenopus egg extract
supplemented with [a-32P]dATP in either the absence (closed symbols) or presence (open symbols) of 3 mM 6-DMAP, and the percentage of input DNA
replicated was determined as in Fig. 1C. (B) Aliquots of intact nuclei shown in
panel A were incubated for 2 h in Xenopus egg extract supplemented with
BrdUTP, and the percentage of nuclei labeled was evaluated as in Fig. 4E (Intact
Nuclei). Aliquots of these nuclei were permeabilized by prolonged exposure to
digitonin and analyzed in parallel as in Fig. 4E (Permeabilized Nuclei). (C)
CHOC 400 cells arrested in serum-free medium for 96 h were returned to
complete medium for the indicated lengths of time. Both intact (M
y ) and permeabilized (}) nuclei were incubated in BrdUTP supplemented extract, and the
percentage of nuclei labeled was evaluated as in panel B. As a control, aliquots
of the same cells were incubated with BrdU in culture for 30 min at each time
point, and the percentage of cells that had entered S phase (H) was determined
as in Fig. 3B. (D) Nuclei from cells arrested in isoleucine-free medium as in
panels A and B were incubated for 2 h in Xenopus egg extract supplemented with
aphidicolin, and the positions of the earliest nascent DNA strands were mapped
by the ELFH assay as in Fig. 1. Shown are the mean values of five independent
experiments and the standard deviation about the mean (when greater than 1).
Axes are labeled as in Fig. 1B. The vertical shaded line shows the position of the
previously mapped origin of bidirectional replication (OBR). Only experiments
in which arrested cells showed ,5% BrdU-positive cells in a 30-min pulse were
considered for any of the analyses shown in this report.
4320
WU AND GILBERT
(43). Studies that have determined the timing of the R point
have not simultaneously examined Rb phosphorylation. We
have used an experimental scheme to measure the serum and
amino acid R points as well as Rb phosphorylation simultaneously in cells traversing G1 phase after mitotic selection. Our
results demonstrate that Rb phosphorylation coincides temporally with the R point.
G0 and replication licensing. Leno and Munshi (34) previously demonstrated that Xenopus egg extracts will not initiate
replication in intact nuclei isolated from quiescent NIH 3T3
cells. They concluded that proliferating cells lose their capacity
to initiate replication upon growth arrest. However, these experiments were not able to distinguish whether cells exiting
mitosis in deficient growth conditions remain in the postreplicative state and never undergo licensing or whether cells acquire the licensed state and then lose it upon entry into G0.
Our results confirm and extend those of Leno and Munshi,
demonstrating that in the absence of either isoleucine or serum
throughout the early G1-phase period, replication licensing is
first completed and then maintained at least until the R point.
After cells become arrested at the R point, they gradually
begin to lose their capacity to replicate in Xenopus egg extracts
(Fig. 4 and 7). This capacity can be restored by returning
arrested cells to complete medium and allowing the cells to
reenter the cell cycle (Fig. 7).
The capacity to replicate in Xenopus egg extract could also
be restored by permeabilizing nuclei prior to incubation in
extract. This is presumed to allow the entry of one or more
licensing factors that lack a nuclear localization signal and can
access chromatin only during mitosis or when the nuclear
membrane is permeabilized by experimental manipulation (4).
Our observation that quiescent cells can reacquire the licensed
state either through permeabilization of the nuclear membrane
or, in the absence of nuclear membrane breakdown, through
the return of cultured cells to the cell cycle implies that the
original view of licensing is oversimplified. Moreover, Coverley
et al. reported that exposure of G2-phase cells to the general
protein kinase inhibitor 6-DMAP for 30 min rendered unpermeabilized nuclei isolated from these cells competent for replication in Xenopus egg extracts (12), suggesting the existence
of a protein kinase-dependent block to licensing in G2 phase.
We have treated serum-arrested CHOC 400 cells with 1 mM
6-DMAP for 0, 30, and 60 min and have not observed a return
to the replication-competent state (46), suggesting that the loss
of licensing during quiescence may occur by a different mechanism than that established to prevent rereplication under
normal growth conditions. A molecular understanding of how
these various means for nuclei to switch between the licensed
and unlicensed states relate to each other and to the original
hypothesis of replication licensing awaits identification of the
factor that enters permeabilized nuclei.
Mechanism of origin choice at the ODP. The mechanism of
origin choice at the ODP is unknown; however, the studies
presented here rule out a number of possibilities. Two direct
links between Rb and the initiation of replication at the DHFR
chromosomal locus have been proposed. First, transcription of
the DHFR gene is induced by Rb phosphorylation, through
dissociation of Rb from the E2F transcriptional activator (42,
44). It has been proposed that the activation of transcription
from the DHFR gene could focus the initiation of replication
to a region downstream of the transcription termination site
(22). Second, a binding site for another Rb-associated factor,
Pura, is located within a region of bent DNA at the CHO
DHFR origin locus, as well as within other origins of replication (29). Binding of Pura to its recognition site in naked DNA
is prevented by the hypophosphorylated form of Rb (29). Since
MOL. CELL. BIOL.
Pura has been shown to exhibit unwinding activity (29), it was
reasonable to postulate that Pura could be important in the
cell cycle-regulated selection of the DHFR origin. Our results
presented here show that the ODP occurs prior to Rb phosphorylation and hence prior to the activation of DHFR transcription by E2F and the potential binding of Pura to the
DHFR origin. Thus, it is unlikely that either of these events is
related to the ODP transition.
As discussed (see the introduction), the specification of replication origins at the ODP appears similar to the specification
of replication origins that occurs at the blastula stage of Xenopus development (27). A number of changes occur in the structure of nuclei at this time during Xenopus development, and it
has been proposed that these changes could restrict the initiation of replication to specific chromosomal sites (19). We
have recently shown that one aspect of nuclear structure that
can have a profound influence on the sites of initiation of
replication in Xenopus egg extract is chromosome architecture
(32). Chromosome architecture is reorganized after each mitosis as chromosomes are decondensed, find their positions
within the nucleus (14), and perhaps reestablish contacts with
the nuclear matrix (26). It is likely that this structural organization is required for nuclear function. For example, positioning of gene sequences at the surfaces of chromosomal territories is proposed to define an interchromosomal compartment
for gene expression (31). Similar architectural requirements
may be critical to establish functional interactions between
replication origins and initiation complexes.
Not mutually exclusive to the role of chromosome architecture in the selection of origins is the possibility that the ODP
represents the assembly of a preinitiation complex that can be
recognized by Xenopus egg cytosol. The sensitivity of the ODP
to 2-AP may provide one means to investigate the nature of the
ODP. At present, we have no direct evidence that the inhibitory effect of 2-AP on the ODP transition results from the
inactivation of a protein kinase, nor can we say what protein
kinase might be involved. 2-AP does not alter the overall pattern of phosphorylation in HeLa cells (50) and thus does not
appear to be a general inhibitor of cellular kinases. However,
the specificity of 2-AP is not well defined. Other, more specific
protein kinase inhibitors have been isolated and shown to
arrest mammalian cells at various points during G1 phase (17).
Identifying the kinase pathway that is required for origin recognition should provide a valuable clue as to the mechanism of
origin choice at the ODP.
ACKNOWLEDGMENTS
We thank D. Dimitrova, M. L. DePamphilis, and W. C. Burhans for
critical reading of the manuscript.
This work was supported by National Science Foundation grant
MCB-9505907 to D.M.G.
REFERENCES
1. Andreassen, P., and R. Margolis. 1992. 2-Aminopurine overrides multiple
cell cycle checkpoints in BHK cells. Proc. Natl. Acad. Sci. USA 89:2272–
2276.
2. Andreassen, P., and R. Margolis. 1991. Induction of partial mitosis in BHK
cells by 2-aminopurine. J. Cell Sci. 100:299–310.
3. Blow, J. J. 1993. Preventing re-replication of DNA in a single cell cycle:
evidence for a replication licensing factor. J. Cell Biol. 122:993–1002.
4. Blow, J. J., and R. A. Laskey. 1988. A role for the nuclear envelope in
controlling DNA replication within the cell cycle. Nature 332:546–548.
5. Blow, J. J., and A. M. Sleeman. 1990. Replication of purified DNA in
Xenopus egg extract is dependent on nuclear assembly. J. Cell Sci. 95:383–
391.
6. Burhans, W. C., L. T. Vassilev, M. S. Caddle, N. H. Heintz, and M. L.
DePamphilis. 1990. Identification of an origin of bidirectional DNA replication in mammalian chromosomes. Cell 62:955–965.
7. Carpenter, P. B., P. R. Mueller, and W. G. Dunphy. 1996. Role for a
VOL. 17, 1997
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
Xenopus ORC2-related protein in controlling DNA replication. Nature 379:
357–360.
Chong, J. P. J., H. M. Mahbubani, C.-Y. Khoo, and J. J. Blow. 1995.
Purification of an MCM-containing complex as a component of the DNA
replication licensing system. Nature 375:418–424.
Cocker, J. H., S. Piatti, C. Santocanale, K. Nasmyth, and J. F. Diffley. 1996.
An essential role for the Cdc6 protein in forming the pre-replicative complexes of budding yeast. Nature 379:180–182.
Coleman, T. R., P. B. Carpenter, and W. G. Dunphy. 1996. The Xenopus
Cdc6 protein is essential for the initiation of a single round of DNA replication in cell-free extracts. Cell 87:53–63.
Coverley, D., and R. A. Laskey. 1994. Regulation of eukaryotic DNA replication. Annu. Rev. Biochem. 63:745–776.
Coverley, D., H. Wilkinson, and C. Downes. 1996. A protein kinase-dependent block to reinitiation of DNA replication in G2 phase in mammalian
cells. Exp. Cell Res. 225:294–300.
DePamphilis, M. L. 1993. Origins of DNA replication in metazoan chromosomes. J. Biol. Chem. 268:1–4.
Dernburg, A. F., K. W. Broman, J. C. Fung, W. F. Marshall, J. Philips, D. A.
Agard, and J. W. Sedat. 1996. Perturbation of nuclear architecture by longdistance chromosome interactions. Cell 85:745–759.
Diffley, J. F., J. H. Cocker, S. J. Dowell, and A. Rowley. 1994. Two steps in
the assembly of complexes at yeast replication origins in vivo. Cell 78:303–
316.
Dijkwel, P. A., and J. L. Hamlin. 1995. The Chinese hamster dihydrofolate
reductase origin consists of multiple potential nascent-strand start sites. Mol.
Cell. Biol. 15:3023–3031.
Gadbois, D., H. Crissman, R. Tobey, and E. Bradbury. 1992. Multiple kinase
arrest points in the G1 phase of nontransformed mammalian cells are absent
in transformed cells. Proc. Natl. Acad. Sci. USA 89:8626–8630.
Gavin, K., M. Hidaka, and B. Stillman. 1995. Conserved initiator proteins in
eukaryotes. Science 270:1667–1671.
Gilbert, D. M., H. Miyazawa, and M. L. DePamphilis. 1995. Site-specific
initiation of DNA replication in Xenopus egg extract requires nuclear structure. Mol. Cell. Biol. 15:2942–2954.
Gong, J., F. Traganos, and Z. Darzynkiewicz. 1994. Staurosporine blocks cell
progression through G1 between the cyclin D and cyclin E restriction points.
Cancer Res. 54:3136–3139.
Gossen, M., D. Pak, S. Hansen, J. Acharya, and M. Botchan. 1995. A
Drosophila homolog of the yeast origin recognition complex. Science 270:
1674–1677.
Hamlin, J. L., and P. A. Dijkwel. 1995. On the nature of replication origins
in higher eukaryotes. Curr. Opin. Genes Dev. 5:153–161.
Hamlin, J. L., P. J. Mosca, and V. V. Levenson. 1994. Defining origins of
replication in mammalian cells. Biochim. Biophys. Acta 1198:85–111.
Herrera, R., V. Sah, B. Williams, T. Makela, R. Weinberg, and T. Jacks.
1996. Altered cell cycle kinetics, gene expression, and G1 restriction point
regulation in Rb-deficient fibroblasts. Mol. Cell. Biol. 16:2402–2407.
Huet, X., J. Rech, A. Plet, A. Vie, and J. Blanchard. 1996. Cyclin A expression
is under negative transcriptional control during the cell cycle. Mol. Cell. Biol.
16:3789–3798.
Hutchison, C., J. Bridger, L. Cox, and I. Kill. 1994. Weaving a pattern from
disparate threads: lamin function in nuclear assembly and DNA replication.
J. Cell Sci. 107:3259–3269.
Hyrien, O., C. Maric, and M. Mechali. 1995. Transition in specification of
embryonic metazoan DNA replication origins. Science 270:994–997.
Hyrien, O., and M. Mechali. 1993. Chromosomal replication initiates and
terminates at random sequences but at regular intervals in the ribosomal
DNA of Xenopus early embryos. EMBO J. 12:4511–4520.
REPLICATION ORIGIN DECISION POINT
4321
29. Johnson, E., P.-L. Chen, C. Krachmarov, S. Barr, M. Kanovsky, Z.-W. Ma,
and W.-H. Lee. 1995. Association of human Pura with the retinoblastoma
protein, Rb, regulates binding to the single-stranded DNA Pura recognition
element. J. Biol. Chem. 270:1–9.
30. Kubota, Y., S. Mimura, S. Nishimoto, H. Takizawa, and H. Nojima. 1995.
Identification of the yeast MCM3-related protein as a component of Xenopus DNA replication licensing factor. Cell 81:601–609.
31. Kurz, A., S. Lampel, J. Nickolenko, J. Bradl, A. Benner, R. Zirbel, T.
Cremer, and P. Lichter. 1996. Active and inactive genes localize preferentially in the periphery of chromosome territories. J. Cell Biol. 135:1195–1205.
32. Lawlis, S., S. Keezer, J.-R. Wu, and D. M. Gilbert. 1996. Chromosome
architecture can dictate site-specific initiation of DNA replication in Xenopus egg extracts. J. Cell Biol. 135:1–12.
33. Leno, G. H., C. S. Downes, and R. A. Laskey. 1992. The nuclear membrane
prevents replication of human G2 nuclei but not G1 nuclei in Xenopus egg
extract. Cell 69:151–158.
34. Leno, G. H., and R. Munshi. 1994. Initiation of DNA replication in nuclei
from quiescent cells requires permeabilization of the nuclear membrane. J.
Cell Biol. 127:5–14.
35. Ludlow, J., C. Glendening, D. Livingston, and J. DeCaprio. 1993. Specific
enzymatic dephosphorylation of the retinoblastoma protein. Mol. Cell. Biol.
13:367–372.
36. Madine, M. A., C.-Y. Khoo, A. D. Mills, and R. A. Laskey. 1995. MCM3
complex required for cell cycle regulation of DNA replication in vertebrate
cells. Nature 375:421–424.
37. Metheny, L., D. Nelson, and J. Ludlow. 1995. Effect of the kinase inhibitor
2-aminopurine on the cell cycle stage-dependent phosphorylation state of
the retinoblastoma protein. Exp. Cell Res. 218:151–154.
38. Pardee, A. 1989. G1 events and the regulation of cell proliferation. Science
246:603–608.
39. Pardee, A. 1974. A restriction point for control of normal animal cell proliferation. Proc. Natl. Acad. Sci. USA 71:1286–1290.
40. Rowles, A., J. Chong, L. Brown, M. Howell, G. Evan, and J. Blow. 1996.
Interaction between the origin recognition complex and the replication licensing system in Xenopus. Cell 87:287–296.
41. Temin, H. 1971. Stimulation by serum of multiplication of stationary chicken
cells. J. Cell. Physiol. 78:161–170.
42. Weinberg, R. 1996. E2F and cell proliferation: a world turned upsidedown.
Cell 85:457–459.
43. Weinberg, R. 1995. The retinoblastoma protein and cell cycle control. Cell
81:323–330.
44. Wells, J., P. Held, S. Illenye, and N. Heintz. 1996. Protein-DNA interactions
at the major and minor promoters of the divergently transcribed dhfr and
rep3 genes during the Chinese hamster ovary cell cycle. Mol. Cell. Biol.
16:634–647.
45. Wu, J.-R., and D. M. Gilbert. 1996. A distinct G1 step required to specify the
chinese hamster DHFR replication origin. Science 271:1270–1272.
46. Wu, J.-R., and D. M. Gilbert. Unpublished data.
47. Zerfass-Thome, K., A. Schulze, W. Zwerchke, B. Vogt, K. Helin, J. Bartek, B.
Henglein, and P. Jansen-Durr. 1997. p27KIP1 blocks cyclin E-dependent
transactivation of cyclin A gene expression. Mol. Cell. Biol. 17:407–415.
48. Zetterberg, A., and O. Larsson. 1991. Coordination between cell growth and
cell cycle transit in animal cells. Cold Spring Harbor Symp. Quant. Biol.
56:137–147.
49. Zetterberg, A., and O. Larsson. 1995. What is the restriction point? Curr.
Opin. Cell Biol. 7:835–842.
50. Zinn, K., A. Keller, L.-A. Whittemore, and T. Maniatis. 1988. 2-Aminopurine selectively inhibits the induction of b-interferon, c-fos, and c-myc gene
expression. Science 240:210–213.