Oncogene (1997) 15, 649 ± 656
 1997 Stockton Press All rights reserved 0950 ± 9232/97 $12.00
Activation of c-Myc uncouples DNA replication from activation of
G1-cyclin-dependent kinases
Oliver Pusch1, Gerhard Bernaschek1, Martin Eilers2 and Markus HengstschlaÈger1
1
Obstetrics and Gynecology, University of Vienna, Department of Prenatal Diagnosis and Therapy, WaÈhringer GuÈrtel 18 ± 20,
A-1090 Vienna, Austria, 2Zentrum fuÈr Molekulare Biologie Heidelberg (ZMBH), Im Neuenheimer Feld 282, 69120 Heidelberg,
Germany
Proto-oncogenes like c-myc are thought to control exit
from the cell cycle rather than progression through the
cell cycle itself. We now present a di€erent view of Myc
function. Exponentially growing Rat1-MycERTM ®broblasts were size-fractionated by centrifugal elutriation. In
these cells, activation of cyclin E- and cyclin Adependent kinases, degradation of p27, hyperphosphorylation of retinoblastoma protein and activation of E2F
occur sequentially at speci®c cell sizes. Upon activation
of Myc, however, these transitions all occur simultaneously in small cells immediately after exit from
mitosis. In contrast, Myc has no discernible e€ect on
the cell size at which DNA replication is initiated. These
data show ®rst that Myc controls the activity of G1
cyclin-dependent kinases independently from the transition between quiescence and proliferation and from any
e€ect on cell growth in size. These data also provide
evidence of at least one dominant mechanism besides
activation of E2F and of cyclin E/cdk2 kinase, which
prevents DNA replication unless a critical cell size has
been reached.
Keywords: c-Myc; G1-cyclins; cell cycle; initiation of
replication
Introduction
Passage through the cell cycle depends on an increase
in cell size. In eukaryotic cells, initiation of DNA
replication does not occur until cells reach a minimum
size (Killander and Zetterberg, 1965). In yeast, this
control acts at a point in G1 which is called START
(Nurse and Bisset, 1981). Similarly, in mammalian cells
commitment to DNA replication is thought to occur at
a speci®c point, termed restriction point, in mid G1
phase (Rossow et al., 1979). Whether, the size control
of mammalian cells also acts at the restriction point is
not known; however, many events used to de®ne
passage through the restriction point are size-dependent in exponentially growing mammalian cells
(Zetterberg et al., 1995).
Many genes that control mammalian cell proliferation are proto-oncogenes, cellular homologues of the
transduced oncogenes of retroviruses. Induction of
oncogene activity is often strongly mitogenic and can
be sucient to induce proliferation in resting cells. For
example, induced expression of transforming alleles of
Correspondence: M HengstschlaÈger
Received 6 February 1997; revised 25 April 1997; accepted 25 April
1997
ras (Winston et al., 1996) or activation of temperaturesensitive alleles of v-src stimulates cell cycle entry in
resting cells (Welham et al., 1990). One striking
example for this is provided by the c-myc gene. c-myc
encodes a transcription factor (Myc) of the helix ±
loop ± helix/leucine zipper class of proteins (for review,
see Amati and Land, 1994; Bernards, 1995). Myc
protein binds to speci®c DNA sequences (termed Eboxes) as part of a heterodimeric complex with a
partner protein called Max (Blackwell et al., 1990;
Blackwood and Eisenman, 1991; Kerkho€ et al., 1991;
Prendergast et al., 1991; Prendergast and Zi€, 1991).
The complex is a potent activator of transcription
(Amati et al., 1992; Kretzner et al., 1992; Amin et al.,
1993). Induction of conditional alleles of c-myc is
sucient to stimulate cell cycle re-entry and proliferation in resting cells (Eilers et al., 1991). Both the ability
to bind to DNA and to interact with Max are
necessary for the mitogenic properties of Myc,
strongly suggesting that Myc acts by transcriptionally
activating a critical set of target genes (Stone et al.,
1987; Amati et al., 1993a; Amati et al., 1993b).
Several models are conceivable to explain how Myc
might stimulate proliferation. First, Myc might
stimulate growth in cell size by activating transcription
of genes which encode rate-limiting metabolic enzymes.
Potential examples for such target genes encode CAD
(Carbamoyl-phosphate synthase/aspartate carbamoyltransferase/dihydroorotase), a rate limiting enzyme in
pyrimidine biosynthesis (Miltenberger et al., 1995) and
ornithine decarboxylase, the rate limiting enzyme of
polyamine biosynthesis (Bello-Fernandez et al., 1993).
Polyamines stimulate a wide variety of biological
processes (Tabor and Tabor, 1984); thus, induction of
ornithine decarboxylase and CAD expression by Myc
might contribute to cell growth and thus indirectly lead
to cell proliferation. Alternatively, Myc might directly
stimulate genes that encode components of the `cell
cycle machinery' and thus stimulate proliferation
directly rather than indirectly via a possible e€ect on
cell growth. This latter model predicts that Myc should
activate downstream targets in the cell cycle independent of cell size and overt e€ects on cellular
proliferation.
Closely related is the question which molecular
events are used by a cell to ascertain that a critical
cell size for the initiation of DNA replication has been
reached. In exponentially growing mammalian cells,
E2F-dependent transcription, hyperphosphorylation of
retinoblastoma protein (pRb) and transcription and
activation of the cyclin E- and cyclin A- dependent
kinases are periodic and are initiated before the onset
of DNA replication. These events are required for
c-Myc induces G1-CDKs independent of replication
O Pusch et al
650
entry of a cell into S-phase (reviewed in Nigg, 1995;
Weinberg, 1995; Pines, 1995); activation of cyclindependent kinases (cdk) can be rate limiting for
passage through G1 (Ohtsubo and Roberts, 1993;
Resnitzky et al., 1994; Resnitzky and Reed, 1995). But
it is open whether these events are actually sucient to
stimulate DNA synthesis or whether additional
controls ascertain that S-phase is only initiated after
cells have grown in size.
We have addressed both issues using Rat1 fibroblasts which carry a hormone-inducible allele of the cmyc proto-oncogene (MycERTM; Eilers et al., 1989;
Steiner et al., 1995). Previous work has demonstrated
that activation of Myc in resting cells leads to a rapid
activation of cyclin E/cdk2 kinase (Steiner et al., 1995);
cyclin E/cdk2 in turn transactivates the E2F transcription factor and expression of cyclin A (Rudolph et al.,
1996). We have now size-fractionated exponentially
growing Rat1-MycERTM cells by centrifugal elutriation,
either in the presence or absence of activated Myc.
Under these conditions induction of Myc causes
activation of cyclin D, cyclin E- and cyclin Adependent kinases, degradation of the cdk inhibitor
p27, and E2F dependent transcription independent of
cell size. These data show unequivocally that c-myc
controls the activity of cyclin-dependent kinases
independent from any e€ect on exit or entry into the
cell cycle or from cell size. Second, we show that even
in the presence of active cyclin D1, cyclin E- and cyclin
A-dependent kinases and of active E2F, Rat1 cells
grow to normal size before they initiate DNA
replication and replication-dependent transcription of
the histone H4 gene. Thus, at least one dominant
mechanism must exist that monitors the size requirement of DNA replication.
Results
To explore potential e€ects of Myc on cell cycle
progression in exponentially growing cells, Rat1MycERTM cells were seeded at low density, either in
the presence or absence of hydroxy-tamoxifen (4OHT). Twenty-four hours later, the cell cycle
distribution was determined cyto¯uorometrically. Under these conditions, Rat1-MycERTM cells showed a
cell cycle distribution characteristic of growing cells
both in the presence or absence of activated Myc
(Figure 1a). Quantitiation of multiple experiments
showed that induction of Myc caused a marginal
decrease in the number of G1 phase cells (from
53%+3 to 45%+2) with an concomitant increase in
the number of S phase cells. Variation of the
incubation times with 4-OHT between 20 and 40 h
yielded no signi®cant di€erences. 4-OHT had no e€ect
on cell cycle distribution in Rat1 cells stably infected
with the empty retroviral vector (Rat1-pBpuro) or in
cells expressing a non-functional allele of Myc (Rat1DelMycER) (Stone et al., 1987; Eilers et al., 1989) in
which residues 106 ± 143 of Myc are deleted. In
parallel, determination of the cell doubling time by
cell counting showed a small reduction in the doubling
time of induced versus non-induced cells (from 13 h to
about 12.5 h, Figure 1b). We concluded that Myc has
little overt e€ect on cell cycle progression of
exponentially growing Rat1-MycERTM cells and in-
Figure 1 E€ect of c-Myc activation on cell cycle phase
distributions in cycling Rat1 ®broblasts. (a) Logarithmically
growing control cells (Rat1-pBpuro), cells expressing wild type
MycER (Rat1-MycER), and cells expressing mutant (del106 ± 143)
MycER (Rat1-DelMycER) with (+) and without (7) 24 h
activation of Myc by 4-OHT were harvested, ®xed and DNA
was stained with propidium iodide. Cell cycle phase distributions
were analysed on a ¯ow cytometer and the percentage of cells in
each phase of the cell cycle was estimated with the CellFIT
computer program. (b) Doubling times of the cells analysed in (a)
were determined by cell counting and the duration of the distinct
cell cycle phases was calculated by relating doubling times and
percentage of cell cycle phase distributions. The values of three
independent determinations are presented
duced a maximally 8% decrease (corresponding to
30 min of a total length of 7.6 h) in the duration of G1
phase.
Next, we asked whether Myc-induced cells initiate
DNA replication at the same cell size as their noninduced counterparts. To address this question we
separated cells according to cell size phases by
centrifugal elutriation. The advantage of this synchronization procedure is that cells, which have never been
forced to leave the cell cycle, are separated without the
use of drugs. Isolated cells were arbitrarily assigned to
seven narrow size classes and the DNA content of each
class was determined by staining with propidium
iodide. Alternatively, small G1 cells were isolated and
replated in the presence of 10% FCS either with or
without 4-OHT. These cells entered S phase with a
degree of synchrony (about 60%), that is signi®cantly
higher than that observed with other synchronization
procedures (Steiner et al., 1995; HengstschlaÈger et al.,
1994b; 1996). Comparing Myc-induced and noninduced Rat1-MycERTM cells, we found in both types
of experiment that the onset of DNA replication in
Myc-overexpressing cells occurs at the same cell size as
in normal cells (Figure 2).
Previously Myc has been shown to upregulate
expression of cyclin A and, to a lesser extent, cyclin E
c-Myc induces G1-CDKs independent of replication
O Pusch et al
651
Figure 2 c-Myc does not a€ect cell size control of the initiation of DNA replication. (a) Cell size of randomly cycling normal and
c-Myc overexpressing cells were cyto¯uorometrically analysed using the forward scatter (FSC) program (Beckton Dickinson). In
addition, the cell volume was determined in femtoliter (¯) by a Casy cell analysis system (SchaÈrfe System). (b) Logarithmically
growing cells were separated according to cell size by centrifugal elutriation. The elutriated fractions were cyto¯uorometrically
analysed for DNA distribution after staining DNA with propidium iodide. Size in femotliter of the cells in the distinct elutriated
fractions was determined as described above. (c) Non induced Rat1-MycERTM cells were separated by centrifugal elutriation. The
®rst G1 phase fraction was replated either in the presence or absence of 4-OHT. Every 4 h cells were analysed for DNA distribution
and cell size
mRNAs during the G0/G1 transition (Jansen-DuÈrr et
al., 1993). We wondered whether Myc a€ected
expression of cyclins A, E and D in exponentially
dividing Rat1-MycERTM cells. Fractions obtained by
centrifugal elutriation from induced or non-induced
cells, respectively, were analysed for cyclin mRNA
expression by Northern blotting. In untreated cells,
cyclin E mRNA expression increased during G1 and
decreased after S phase. Cyclin A mRNA expression
was induced at the G1/S boundary and remained at a
high level until the next G1 phase. In contrast, no
signi®cant cell cycle ¯uctuation of cyclin D1 and cyclin
D3 mRNA expression was observed (Figure 3;
expression of Cyclin D2 mRNA was not detectable
in these cells). Similar observations were made in other
cell lines (Motokura and Arnold, 1993). Activation of
wild-type Myc triggered strong upregulation of cyclin
A and E mRNA expression in very early G1 cells,
abrogating the cell cycle regulation of either mRNA.
Cyclin D1 expression was slightly downregulated,
whereas cyclin D3 expression was una€ected by Myc
(Figures 3 and 4a). Western blot analyses showed that
the Myc-dependent induction of cyclin mRNA in early
G1 cells was paralleled by an induction of cyclin
protein (Figure 4a). Overexpression of the inactive (del
106 ± 143) MycER protein had no e€ect on the
expression of these cyclins (not shown).
Cyclin A and cyclin E proteins are activating
subunits of cdk2. Cyclin A- as well as cyclin Eassociated kinase activities are low in G1 and are
induced at the G1/S boundary, at a point where cells
have reached a critical size (Dulic et al., 1992; Pagano
et al., 1992, 1993; Ko€ et al., 1992). We determined
cyclin E- and A-kinase activities and the amount of
p27 and Cdc25A protein in parallel (Figure 4). In
control cells, we observed a strong regulation of both
activities throughout the cell cycle: cyclin E-dependent
kinase activity was activated in mid to late G1. Cyclin
A-dependent kinase activity was activated at the G1/S
boundary. In contrast, upon induction of Myc, both
cyclin E- and cyclin A-dependent kinase activities
remained high throughout the cell cycle; at maximum, a two-fold regulation in very early G1 could
be observed. D1-type-cyclin-associated kinase activity
was essentially independent of cell size in our assays; it
was slightly activated by Myc at all cell cycle phases.
Speci®city of the D1-kinase assays was ascertained by
usage of an isotype-matched control antibody. Kinase
signals using control antibodies were about 15-fold
lower than the shown signals for uninduced Rat1MycERTM cells (data not shown; compare Steiner et al.,
1995). Degradation of p27 was initiated in very early
G1 immediately after exit from mitosis in Myctransformed cells; no p27 was detectable in mid G1
cells. In contrast, p27 is present into S-phase of control
cells. We did not detect any e€ects of Myc on the
expression levels of Cdc25A protein (Figure 4a).
To address the biological consequences of these
observations, we analysed potential downstream targets
of cdk2, one of which is the retinoblastoma protein
(Nigg, 1995; Weinberg, 1995; Pines, 1995). Western
blot analysis of the elutriated fractions of uninduced
Rat1-MycERTM cells showed a transition from the
hypophosphorylated to the hyperphosphorylated form
of pRb in mid G1, as previously demonstrated in many
cell lines (Figure 4; Nigg, 1995; Weinberg, 1995; Pines,
1995). In contrast, induced Rat1-MycERTM cells
exhibited hyperphosphorylated pRb throughout the
whole cell cycle, even in the ®rst elutriated fraction
containing early G1 cells (Figure 4). These results
c-Myc induces G1-CDKs independent of replication
O Pusch et al
652
Rat1-MycER –
elutriated fractions
98
1
1
96
3
1
91
6
3
81
16
3
14
74
12
Rat1-MycER +
elutriated fractions
14
22
64
15
11
74
98
1
1
96
2
2
92
6
2
84
14
2
12
76
12
15
28
57
19
10
71
% G1
%S
% G2/M
cyclin E mRNA
cyclin A mRNA
cyclin D1 mRNA
cyclin D3 mRNA
methylene blue
Figure 3 Activation of Myc deregulates cell cycle dependent expression of cyclin A and cyclin E mRNAs. Rat1-MycERTM cells
with and without incubation with 4-OHT were separated by centrifugal elutriation. FACS analyses of DNA distributions of the
di€erent fractions are presented in the upper panel. Lower panels show mRNA levels of cyclins A, E, D1 and D3
strongly suggest that the Myc-activated cyclin-associated kinases in early G1 cells are biologically active.
Hyperphosphorylation of pRb during G1 prevents its
interactions with E2F, thus liberating transcriptionally
active `free' E2F (MuÈller, 1995). We analysed complex
formation on the E2F binding site of the adenovirus
E2 promoter. The formation pattern of the di€erent
obtained complexes were identical to that described
earlier (Figure 5a; Pagano et al., 1992). The upper
bands represent transcriptionally inactive E2F complexes containing pocket proteins (Pagano et al., 1992).
We identi®ed `free' transcriptional active E2F by
treatment with deoxycholate, which dissociates pocket-protein complexes of E2F (arrow in Figure 5a;
Bagchi et al., 1990). We detected a signi®cant Mycinduced increase in the amount of `free' E2F in
logarithmically growing and in early small G1 cells
(Figure 5a and b). There were no di€erences in the
formation of E2F complexes after treatment of the
control cells Rat1-pBabepuro or Rat1-DelMycER with
4-OHT (data not shown). This experiment further
revealed that the total level of E2F activity was
increased upon activation of Myc (Figure 5a).
To test whether the increase in free E2F was
paralleled by an increase in E2F-dependent transcription, we decided to analyse the e€ects of c-Myc
overexpression on cell cycle regulation of the E2F
dependent transcription of the thymidine kinase (TK)
and the dihydrofolate reductase (DHFR) gene. Both
genes have been shown to be expressed speci®cally in
the S-phase of exponentially growing cells and the role
of E2F in the regulation of their promoters has been
demonstrated (Blake and Azizkhan, 1989; Slansky et
al., 1993; Ogris et al., 1993; HengstschlaÈger et al.,
1994b, 1996; Dou et al., 1994). Northern blot
experiments of uninduced Rat1-MycERTM cells after
centrifugal elutriation con®rmed this regulation (Figure
5c and d). Activation of Myc causes high levels of TK
and DHFR mRNA already in early small G1 cells. In
contrast, the cell cycle dependence of histone H4
transcription, which is not mediated by E2F, remained
una€ected by Myc activation (Figure 5c and d).
Discussion
In this paper, we show that activation of Myc relieves
the size requirement for a number of key transitions
during the G1 phase of the mammalian cell cycle,
thereby uncoupling these events from normal cell cycle
progression. In addition, we show that, even in the
presence of active E2F-dependent transcription and of
active cyclin-dependent kinases cells delay initiation of
DNA replication until they have reached a normal cell
size.
Previous work had demonstrated that induction of
Myc in resting cells triggers a rapid induction of cyclin
E/cdk2 kinase activity (Steiner et al., 1995). While these
observations provided tools to understand the mitogenic function of Myc, one of the key questions
remained unresolved: whether the observed e€ects are
an indirect re¯ection of Myc's mitogenic properties or
whether they indeed de®ne a genetic pathway in which
Myc controls the activity of the cyclin E/cdk2 kinase
independent of the status of the cell. The data
presented here resolve the issue. Under the conditions
of the experiments (i.e. in exponentially growing cells)
Myc has no signi®cant e€ect on cell cycle progression,
c-Myc induces G1-CDKs independent of replication
O Pusch et al
Figure 4 (a) Induced or non-induced Rat1-MycERTM cells were
separated by centrifugal elutriation. FACS analyses of DNA
distribution of the di€erent fractions are presented in the upper
panel. Lower panels show western blots probed with antibodies
for cyclin A, E, D1, p27 and Cdc25A. In vitro kinase activities of
cyclin E-, cyclin A-, and cyclin D1-associated kinase as well as
pRb phosphorylation status were determined for each elutriated
fraction. (b) Induced or non-induced Rat1-MycERTM cells were
separated by centrifugal elutriation three times. Fractions
containing similar cell cycle distributions as presented in (a)
were analysed for in vitro cyclin E-, cyclin A-, and cyclin D1associated kinase activities. Radioactivity of the di€erent obtained
bands in the dried gel were counted by an Instant Imager
(Packard, Electronic Autoradiograph). The so obtained data are
presented in percentage relative to the highest value (set 100%)
cell growth and initiation of DNA replication. Nevertheless, activation of Myc almost completely abolishes
the cell cycle regulation of the cyclin E- and cyclin Adependent kinases and results in cell cycle independent
hyperphosphorylation of pRb and E2F-dependent
transcription. Our data predict the existence of direct
targets of Myc in the activation of the cyclin E/cdk2
kinase complex. In resting cells Myc-induced loss of
p27 from cyclin E/cdk2 complexes is the rate limiting
step for activation (Steiner et al., 1995). The precise
mechanism of this reaction and the putative targets are
yet unknown and currently under investigation.
Although it has recently been shown that Cdc25A is
a direct target for activation by Myc (Galaktionov et
al., 1996), our data show that Myc-dependent
induction of cyclin-dependent kinase activities in early
G1 phase is not caused by increased levels of Cdc25A.
One of the hallmarks of Myc-transformed cells is
their often complete insensitivity to growth factor
deprivation for entry into S-phase (Eilers et al., 1991;
Evan et al., 1992). In contrast, non-transformed
counterparts enter this state in mid G1. The transition
from growth factor sensitivity to insensitivity has been
termed restriction point and re¯ects the time when cells
become committed to DNA replication (Rossow et al.,
1979). Passage through the restriction point correlates
closely with the onset of cyclin E and cyclin A
expression and with hyperphosphorylation of pRb
(Dou et al., 1993). Our data suggest that Myctransformed Rat1 cells are equivalent to post-commitment normal cells for the entire duration of their G1
phase and never enter the growth factor sensitive phase
of the cell cycle of their non-transformed counterparts.
Thus, the Myc-induced change in cell cycle progression
reported here would be sucient to explain the
resistance to growth factor deprivation that is
observed in Myc-transformed cells.
Our ®ndings that Myc cells do not start replication
although they already express E2F-dependent transcription are in apparent contrast with recent
observations that ectopic expression of E2F alone is
sucient to drive resting cells into S-phase (Johnson et
al., 1993; Qin et al., 1994; Shan and Lee, 1994;
DeGregori et al., 1995; Kowalik et al., 1995). However,
the size at which these cells initiate DNA synthesis has
not been determined and it is conceivable that arrested
E2F-expressing cells actually grow signi®cantly before
DNA synthesis. In support of this notion, we have
recently analysed Rat1 cells expressing a conditionally
inducible E2F gene and found that, upon induction,
E2F-dependent transcription is active already in very
small G1 cells, yet there is no discernible e€ect on the
size at which replication is initiated (T Soucek, O
Pusch and M HengstschlaÈger, in preparation). Further,
our data suggest that the signi®cant reduction of cell
size at entry into S-phase and length of G1 that is
observed in cells overexpressing both cyclins D1 and E
(Resnitzky et al., 1994; Resnitzky and Reed, 1995) may
not simply be due to premature activation of cyclindependent kinases. If that were the case, we would
expect to observe similar e€ects in Myc-transformed
cells. Rather, it appears that the high levels of
overexpression of both cyclins D1 and E in such cells
at least in part override size control mechanisms that
are still intact under physiological cyclin expression
levels of Myc-transformed cells.
At present, there is no indication as to how this
additional size control might act in mammalian cells.
Assuming that this control is also exerted by kinases,
potential candidates include cdk3. Although cdk3 has
never been found in association with cyclins, because
of high sequence identity with both cdc2 and cdk2 and
the ability to complement cdc28 mutations in yeast, it
was classi®ed as a cyclin-dependent kinase (Meyerson
et al., 1992). It has been demonstrated that cdk3
executes an essential function regulating S-phase entry,
which is distinct from the functions of cdc2 and cdk2
(Van der Heuvel and Harlow, 1993; Hofmann and
Livingston, 1996). In yeast, another kinase, cdc7, and
its activating subunit, dbf4, are required to initiate
DNA replication and interact directly with replication
origins (Dowell et al., 1994). Mammalian homolouges
of these latter genes have not been identi®ed. However,
653
c-Myc induces G1-CDKs independent of replication
O Pusch et al
654
a
b
Rat1-MycER
elutriated fractions
c
Rat1-MycER –
elutriated fractions
Rat1-MycER +
elutriated fractions
1
2
6
4 %S
98 81 14 14 15
98 84 12 15 19 % G1
1
1
3
2 %G2/M
1 16 74 22 11
1
14 76 28 10 % S
–4– OHT
+4– OHT
–4– OHT
+4– OHT
+4– OHT
+4– OHT +DOC
–4– OHT +DOC
+4– OHT
98 97 91 94 %G1
–4– OHT
+4– OHT +mt oligo
+4– OHT +wt oligo
+4– OHT
Rat1-MycER log
1
1
2 12 57 71 % G2/M
3
12 64 74
TK
DHFR
H4
GAPDH
▲
d
Figure 5 c-Myc induces protein binding at the E2F binding site of the adenovirus E2 promoter and E2F-dependent gene expression
in early G1 cells, but does not a€ect the replication-associated histone H4 mRNA expression throughout the cell cycle. (a)
Speci®city of the E2F band shifts. Extracts from logarithmically growing induced Rat1-MycERTM cells were incubated with a 5'
end-labelled DNA fragment corresponding to an E2F binding site from the adenovirus E2 promoter. The speci®city of the resulting
protein complexes was investigated by competition with excess of unlabeled probe DNA of either the wild-type (wt) or the mutated
(mt) sequence. Parallel samples were incubated at 48C in the presence of 0.8% deoxycholate for 20 min and then NP-40 was added
to a ®nal concentration of 1.5%. (b) Rat1-MycERTM cells with and without 4-OHT treatment for 24 h were separated according to
the di€erent cell cycle phases by centrifugal elutriation. Band shift assays using the E2 oligo were performed as described above. (c)
Northern blots documenting the expression of thymidine kinase (TK), dihydrofolate reductase (DHFR), and histone H4 (H4)
mRNA in fractions from uninduced and induced Rat1-MycERTM cells. (d) Autoradiographs of panel C were scanned
densitometrically and TK, DHFR and H4 gene expression was normalised to GAPDH and to the expression in S-phase (open
circles represent uninduced, full circles 4-OHT treated Rat1-MycERTM cells)
forms of control other than by cyclin/cdk complexes
are easily conceivable and thus further work will be
required to identify the responsible mechanisms.
Materials and methods
Cell culture, centrifugal elutriation and ¯ow cytometry.
Establishment and cultivation of the Rat-1 cell lines used
in this study have been described previously (Eilers et al.,
1989; Littlewood et al., 1995). Myc functions were
activated with 100 n M 4-hydroxytamoxifen (4-OHT).
Separation of logarithmically growing cells into distinct
cell cycle phases was accomplished by centrifugal elutriation in a Beckman J2 ± 21 M centrifuge and a JE ± 6B rotor
with a standard separation chamber (HengstschlaÈger et al.,
1994a). The rotor was kept at a speed of 2000 r.p.m.,
temperature was 208C, and medium ¯ow was controlled
with a Cole-Parmer Master¯ex pump. Consecutive fractions of 150 ± 300 ml were collected at increasing ¯ow rates.
Cyto¯uorometric analyses of cell cycle distributions were
performed using a Becton-Dickinson FACScan. DNA was
stained with propidium iodide (Giunta and Pucillo, 1995).
Cell size in femtoliter was determined with a Casy Cell
Counter and Analyser (SchaÈrfe System).
Western blot and Northern blot analyses
Protein extracts were prepared in bu€er containing 20 mM
HEPES pH 7.9, 0.4 M NaCl, 2.5% glycerol, 1 m M EDTA,
1 mM phenylmethylsulfonyl ¯uoride (PMSF), 0.5 mM NaF,
0.5 mM Na3VO4, 0.02 mg/ml leupeptin, 0.02 mg/ml aprotinin, 0.003 mg/ml benzamidinchloride, 0.1 mg/ml trypsin
inhibitor and 0.5 mM DTT. After 20 min on ice, the
extracts were centrifuged and supernatants were stored at
7708C. Western blot analyses were performed as described
(Steiner et al., 1995). The following antibodies were used:
against cyclin E, M20 (sc-481, St. Cruz); cyclin A, H432
(sc-751, St. Cruz); cyclin D1 (DCS-6, Progen); p27, C19
(sc-528, St. Cruz); Cdc25A, 144 (sc-97, St. Cruz); pRb,
14001A (Pharmingen) (see also Steiner et al., 1995).
The used protocol for RNA extraction and Northern blot
analysis is described in HengstschlaÈger et al. (1996).
Hybridizations probes were full length human or mouse
cyclin A, cyclin E, cyclin D1, cyclin D3, dihydrofolate
reductase, thymidine kinase, histone H4 and a 1300 bp
fragment of the rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA.
c-Myc induces G1-CDKs independent of replication
O Pusch et al
Immunoprecipitations and immune complex kinase assays
Protein extracts were prepared as described above. Cyclin
immunoprecipitations and analyses of cyclin-dependent
kinases were performed according to Dulic et al. (1992).
The following antibodies were used: against cyclin E, M20
(sc-481, St. Cruz); cyclin A, H432 (sc-751, St. Cruz); cyclin
D1 (DCS-11, NeoMarkers); substrate for cyclin A- and Eassociated kinase detection was histone H1, for cyclin D1kinase GST-pRb (Steiner et al., 1995).
Band shift analyses
Protein was extracted as described above. Gel mobility
shift analyses were carried out with 5 ± 10 mg extract as
described by Jansen-DuÈrr (1995). Synthetic oligonucleotides corresponding to both strands of the wildtype or
mutant distal E2F binding site of the adenovirus E2
promoter were used (oligos are described in Pagano et al.,
1992). Deoxycholate treatment was performed as described
(Bagchi et al., 1990).
Acknowledgements
The authors wish to thank T Littlewood and G Evans for
providing Rat1 cell lines carrying tamoxifen-inducible
MycER alleles, JM Sedivy, T Soucek and E HengstschlaÈger-Ottnad for helpful discussion and comments and L
Desbarats for critically reading the manuscript. Work in
M.H.'s laboratory is supported by the `Fonds zur
FoÈ rderung der wissenschaftlichen Forschung', by the
`Komission Onkologie' and a PhD-fellowship from the
Medical Faculty of the University of Vienna, and by the
Austrian `Nationalbank'. ME's laboratory is supported by
the Human Frontiers of Science Program and the Hess
Programm of the Deutsche Forschungsgemeinschaft.
References
Amati B, Brooks MW, Levy N, Littlewood TD, Evan GI and
Land H. (1993a). Cell, 72, 233 ± 245.
Amati B, Dalton S, Brooks M, Littlewood T, Evan G and
Land H. (1992). Nature, 359, 423 ± 426.
Amati B and Land H. (1994). Curr. Opin. Genet. Dev., 4,
102 ± 108.
Amati B, Littlewood TD, Evan GI and Land H. (1993b).
EMBO J., 13, 5083 ± 5087.
Amin C, Wagner AJ and Hay N. (1993). Mol. Cell. Biol., 13,
383 ± 390.
Bagchi S, Raychaudhuri P and Nevins JR. (1990). Cell, 62,
659 ± 669.
Bello-Fernandez C, Packham G and Cleveland JL. (1993).
Proc. Natl. Acad. Sci. USA, 90, 7804 ± 7808.
Bernards R. (1995). Curr. Biol., 5, 859 ± 861.
Blackwell TK, Kretzner L, Blackwood EM, Eisenman RN
and Weintraub H. (1990). Science, 250, 1149 ± 1151.
Blackwood EM and Eisenman RN. (1991). Science, 251,
1211 ± 1217.
Blake MC and Azizkhan J. (1989). Mol. Cell. Biol., 9, 4994 ±
5002.
DeGregori J, Leone G, Ohtani K, Miron A and Nevins JR.
(1995). Genes Dev., 9, 2873 ± 2887.
Dou OP, Levin AH, Zhao S and Pardee AB. (1993). Cancer
Res., 53, 1493 ± 1497.
Dou QP, Zhao S, Levin AH, Wang J, Helin K and Pardee
AB. (1994). J. Biol. Chem., 269, 1306 ± 1313.
Dowell SJ, Romanowski P and Di‚ey JF. (1994). Science,
265, 1243 ± 1246.
Dulic V, Lees E and Reed SI. (1992). Science, 257, 1958 ±
1961.
Eilers M, Picard D, Yamamoto KR and Bishop JM. (1989).
Nature, 340, 66 ± 68.
Eilers M, Schirm S and Bishop JM. (1991). EMBO J., 10,
133 ± 141.
Evan GI, Wyllie AH, Gilbert CS, Littlewood TD, Land H,
Brooks M, Waters CM, Penn LZ and Hancock DC.
(1992). Cell, 69, 119 ± 128.
Galaktionov K, Chen X and Beach D. (1996). Nature, 382,
511 ± 517.
Giunta M and Pucillo C. (1995). Cell cycle. Pagano M (ed.).
Springer Verlag, pp. 46 ± 52.
HengstschlaÈger M, HengstschlaÈger-Ottnad E, Pusch O and
Wawra E. (1996). Oncogene, 12, 1635 ± 1643.
HengstschlaÈger M, Knoȯer M, MuÈllner EW, Ogris E,
Wintersberger E and Wawra E. (1994a). J. Biol. Chem.,
269, 13836 ± 13842.
HengstschlaÈger M, Mudrak I, Wintersberger E and Wawra
E. (1994b). Cell Growth Di€., 5, 1389 ± 1394.
Hofmann F and Livingston DM. (1996). Genes Dev., 10,
851 ± 861.
Jansen-DuÈrr P. (1995). Cell cycle Pagano M (ed.). Springer
Verlag, pp. 243 ± 249.
Jansen-DuÈrr P, Meichle A, Steiner P, Pagano M, Finke K,
Botz J, Wessbecher J, Draetta G and Eilers M. (1993).
Proc. Natl. Acad. Sci. USA, 90, 3685 ± 3689.
Johnson DG, Schwarz JK, Cress WD and Nevins JR. (1993).
Nature, 365, 349 ± 352.
Kerkho€ E, Bister K and Klempnauer K-H. (1991), Proc.
Natl. Acad. Sci. USA, 88, 4323 ± 4327.
Killander D and Zetterberg A. (1965). Exp. Cell. Res., 40,
12 ± 20.
Ko€ A, Giordano A, Desai D, Yamashita K, Harper W,
Elledge S, Nishimoto T, Morgan D, Franza R and
Roberts J. (1992). Science, 257, 1689 ± 1693.
Kowalik TF, DeGregori J, Schwarz JK and Nevins JR.
(1995). J. Virol., 69, 2491 ± 2500.
Kretzner L, Blackwood EM and Eisenman RN. (1992).
Nature, 359, 426 ± 429.
Littlewood TD, Hancock DC, Danielian PS, Parker MG and
Evan GI. (1995). Nucl. Acids Res., 23, 1686 ± 1690.
Meyerson M, Enders GH, Wu C-L, Su L-K, Gorka C,
Nelson C, Harlow E and Tsai L-H. (1992). EMBO. J., 11,
2909 ± 2917.
Miltenberger RJ, Sukow KA and Farnham PJ. (1995). Mol.
Cell. Biol., 15, 2527 ± 2535.
Motokura T and Arnold A. (1993). Biochim. Biophys. Acta,
1155, 63 ± 78.
MuÈller R. (1995). Trends Genet., 11, 173 ± 178.
Nigg EA. (1995). BioEssays, 17, 471 ± 480.
Nurse P and Bisset Y. (1981). Nature, 292, 558 ± 560.
Ogris E, Rotheneder H, Mudrak I, Pichler A and
Wintersberger E. (1993). J. Virol., 67, 1765 ± 1771.
Ohtsubo M and Roberts JM. (1993). Science, 259, 1908 ±
1912.
Pagano M, Draetta G and Jansen-DuÈrr P. (1992). Science,
255, 1144 ± 1147.
Pagano M, Pepperkok R, Lukas J, Baldin V, Ansorge W,
Bartek J and Draetta G. (1993). J. Cell Biol., 121, 101 ±
111.
Pines J. (1995). Biochem. J., 308, 697 ± 711.
Prendergast GC, Lawe D and Zi€ EB. (1991). Cell, 65, 395 ±
407.
Prendergast GC and Zi€ EB. (1991). Science, 251, 186 ± 189.
Qin XQ, Livingston DM, Kaelin WG, Jr. and Adams PD.
(1994). Proc. Natl. Acad. Sci. USA, 91, 10918 ± 10922.
Resnitzky D, Gossen M, Bujard H and Reed SI. (1994). Mol.
Cell. Biol., 14, 1669 ± 1679.
655
c-Myc induces G1-CDKs independent of replication
O Pusch et al
656
Resnitzky D and Reed SI. (1995). Mol. Cell. Biol., 15, 3463 ±
3469.
Rossow PW, Riddle VGH and Pardee AB. (1979). Proc.
Natl. Acad. Sci. USA, 76, 4446 ± 4450.
Rudolph B, Zwicker J, Sa€rich R, Henglein B, MuÈller R,
Ansorge W and Eilers M. (1996). EMBO J., 15, 3065 ±
3076.
Shan B. and Lee WH. (1994). Mol. Cell. Biol., 14, 8166 ±
8173.
Slansky JL, Li Y, Kaelin WG and Farnham P. (1993). Mol.
Cell. Biol., 13, 1610 ± 1618.
Steiner P, Philipp A, Lukas J, Godden.Kent D, Pagano M,
Mittnacht S, Bartek J and Eilers M. (1995). EMBO J., 14,
4814 ± 4826.
Stone J, deLange T, Ramsay G, Jakobovits E, Bishop JM,
Varmus H and Lee B. (1987). Mol. Cell. Biol., 7, 1697 ±
1709.
Tabor CW and Tabor H. (1984). Ann. Rev. Biochem., 53,
749 ± 790.
Van der Heuvel S and Harlow E. (1993). Science, 262, 2050 ±
2054.
Weinberg RA. (1995). Cell, 81, 323 ± 330.
Welham MJ, Wyke JA and Wyke AW. (1990). Oncogene, 5,
161 ± 169.
Winston JT, Coats SR, Wang YZ and Pledger WJ. (1996).
Oncogene, 12, 127 ± 134.
Zetterberg A, Larsson O and Wiman KG. (1995). Curr. Op.
Cell. Biol., 7, 835 ± 842.