Content uploaded by Anil Kapoor
Author content
All content in this area was uploaded by Anil Kapoor on May 22, 2014
Content may be subject to copyright.
ATM activity contributes to the tumor-suppressing functions of p14
ARF
Yanxia Li
1,5
, Dongcheng Wu
2,5
, Biao Chen
1
, Alistair Ingram
1,2
, Lizhi He
1
, Lieqi Liu
2
, Dahai Zhu
3
,
Anil Kapoor
2,4
and Damu Tang*
,1,2
1
Division of Nephrology, Department of Medicine, McMaster University, Hamilton, ON, Canada;
2
Father Sean O’Sullivan Research
Institute, St Joseph’s Hospital, Hamilton, ON, Canada;
3
Department of Biochemistry, National Laboratory of Medical Molecular
Biology, Peking Union Medical College, Beijing, People’s Republic of China;
4
Department of Surgery, McMaster University, Canada
P14/p19
ARF
(ARF) plays a major role in the activation of
p53 by oncogenic signals. The biochemical basis of this has
not been fully elucidated. We report here that forced
expression of p14
ARF
enhances phosphorylation of p53
serine 15 (p53S15) in NIH3T3, IMR90 and MCF7 cells.
Ectopic expression of the oncogenes c-myc, E2F1 and E1A,
all of which activate p53 at least partially via ARF, lead to
p53S15 phosphorylation in IMR90 cells. In addition,
ectopic expression of p53 also results in p53S15 phosphor-
ylation, suggesting that this is a common event in the ARF–
p53 tumor suppression system. Furthermore, p53-, p14
ARF
-,
c-myc- and E2F1-, but not E1A-, induced p53S15
phosphorylation was substantially reduced in AT fibroblasts
(GM05823). Downregulation of ATM in MCF7 cells using
RNA interference (RNAi) technology significantly attenu-
ated p14
ARF
- and p53-induced phosphorylation of p53S15.
Ectopically expressed ARF in NIH3T3 cells induced ATM
nuclear foci and activated ATM kinase. Functionally,
ectopic expression of p14
ARF
and c-myc inhibited the
proliferation of IMR90 but not ATM null GM05823 cells,
and p14
ARF
-induced inhibition of MCF7 cell proliferation
was significantly attenuated by downregulation of ATM by
RNAi. Taken together, these data show a functional role
for ATM in ARF-mediated tumor suppression.
Oncogene (2004) 23, 7355–7365. doi:10.1038/sj.onc.1207957
Published online 19 July 2004
Keywords: ARF tumor suppressor; P53 and ATM
Introduction
The INK4A/ARF locus on human chromosome
9p21 encodes two tumor suppressors, p16
INK4A
and
p14
ARF
, by alternative splicing and use of different
reading frames (Quelle et al., 1995). They function in
the retinoblastoma (Rb) and p53 tumor suppressor
pathways, respectively (Serrano et al., 1993; Kamijo
et al., 1997).
The product of the alternative reading frame (ARF)
of p16
INK4A
(p14 in human and p19 in murine systems;
referred to as ARF herein for generic statements)
(Quelle et al., 1995; Stott et al., 1998), stabilizes p53 in
response to oncogenic signals. In primary fibroblasts,
expression of oncogenic Ras activates p53 in an ARF-
dependent manner (Serrano et al., 1997; Palmero et al.,
1998). Expression of oncogenes such as c-myc, E1A,
E2F1, v-Able and b-catenin in normal cells also leads to
p53 activation through ARF (de Stanchina et al., 1998;
Zindy et al., 1998; Cong et al., 1999; Dimri et al., 2000;
Damalas et al., 2001). Furthermore, transgenic mice
heterozygous for p19
ARF
engineered to express c-myc in
B cells exhibit accelerated B-cell lymphomas; 80% of
such tumors had lost the wild-type p19
ARF
allele
(Eischen et al., 1999), lending in vivo support to the
notion that ARF plays an essential role in p53 activation
by hyperproliferation signals. While it remains unclear
how ARF might sense oncogenic hyperproliferation
signals, recent developments have revealed that c-myc
and E2F1 induce upregulation of the death-associated
protein (DAP) kinase, a calcium/calmodulin-regulated
kinase, and DAP kinase activates p53 in an
ARF-dependent manner (Raveh et al., 2001).
ARF stabilizes p53 by interaction with Mdm2, the
major player in p53 destabilization in vivo. Homozygous
deletion of mdm2 results in embryonic lethality in mice,
which is rescued by simultaneous deletion of p53 (Jones
et al., 1995; Montes de Oca et al., 1995). Mdm2 interacts
with the N-terminal region of p53, which contains the
acidic transcriptional activation domain. This interac-
tion results in direct inhibition of the transcriptional
activity of p53 (Momand et al., 1992), suppression of
the p300/CBP-mediated p53 acetylation that stabilizes
p53 (Ito et al., 2001), ligation of ubiqitins on p53
(Honda et al., 1997) and export of p53 from the
nucleus into cytoplasmic proteasomes for degradation
(Roth et al., 1998). ARF interacts directly with Mdm2,
thereby inhibiting the ubiquitin ligase activity of
Mdm2 in vitro (Honda and Yasuda, 1999) and abolish-
ing the inhibitory activity of Mdm2 on p53 acetylation
(Ito et al., 2001). Thus, the net result of ARF action is
neutralization of the multi-faceted inhibitory effects of
Mdm2 on p53.
Received 8 March 2004; revised 24 May 2004; accepted 9 June 2004;
published online 19 July 2004
*Correspondence: D Tang, L305, St Joseph’s Hospital, 50 Charlton
Ave East, Hamilton, ON, Canada L8N 4A6;
E-mail: damut@mcmaster.ca
5
These authors contributed equally to this work
Oncogene (2004) 23, 7355–7365
&
2004 Nature Publishing Group
All rights reserved 0950-9232/04 $30.00
www.nature.com/onc
Biochemically, how ARF antagonizes Mdm2 is
not clear. The observation that ARF localizes to
nucleoli led to the hypothesis that ARF sequesters
Mdm2 into nucleolar structures, resulting in p53
stabilization (Weber et al., 1999; Zhang and Xiong,
1999). Subsequently, however, it was found that
p14
ARF
does not co-localize with Hdm2 (human
Mdm2) inside the nucleoli, but rather outside these
structures (Kashuba et al., 2003), and it is the
non-nucleolar ARF that stabilizes p53 (Llanos et al.,
2001). Thus, nucleolar ARF may have other functions
that are unrelated to p53. Indeed, it was found that
p14
ARF
interacts with topoisomerase I in nucleoli
(Olivier et al., 2003) and that nucleolar ARF inhibits
ribosomal RNA processing, consistent with nucleoli
as sites of ribosomal biosynthesis (Sugimoto et al.,
2003). The observations that p14
ARF
resides together
with Hdm2 and p53 outside nucleolar structures,
and that ARF is capable of forming a ternary
complex with Mdm2 and p53 (Kamijo et al., 1998;
Kashuba et al., 2003) indicate that the interaction
with ARF does not lead to release of Mdm2 from p53. It
is thus unclear how ARF activates p53, while p53
remains bound to Mdm2.
Modulation of the interaction between Mdm2 and
p53 plays a major role in p53 activation in the cellular
DNA damage response. ATM, a member of the PI3-
kinase family, directly phosphorylates p53S15 in re-
sponse to ionizing radiation and activates checkpoint
kinase 2 (Chk2) (Zhou and Elledge, 2000). Chk2
subsequently phosphorylates p53S20 (Hirao et al.,
2000). Phosphorylation of p53S15 and p53S20, two
residues adjacent to and located in the Mdm2-binding
region (residues 17–22), respectively, facilitates the
dissociation of Mdm2 from p53 (Shieh et al., 1997;
Chehab et al., 1999). Furthermore, direct ATM-
mediated phosphorylation on S395 of Mdm2 promotes
its degradation via the ubiquitin system (Michael and
Oren, 2002).
Evidence indicates that interaction between the
ARF–Mdm2–p53 pathway activated by oncogenic
signals and ATM–Mdm2–p53 signaling initiated by
DNA damage exists. Although the cellular DNA
damage response is intact without ARF (Kamijo
et al., 1997), ARF enhances DNA damage-induced
apoptosis (de Stanchina et al., 1998) and is required
for some forms of the DNA damage response
(Khan et al., 2000). Furthermore, p14
ARF
has been
observed in a complex containing HR6 and p53
in response to cisplatin and adriamycin treatment
(Lyakhovich and Shekhar, 2003). However, whether
ATM plays a role in ARF–Mdm2–p53 activation is
not clear.
We report here that ATM contributes to the tumor-
suppressing functions of p14
ARF
. Enforced expression
of p14
ARF
leads to ATM-dependent p53S15 phosphor-
ylation, possibly mediated by inducing formation of
ATM nuclear foci. Downregulation of ATM results
in attenuation of ARF-induced p53S15 phosphoryla-
tion and releases ARF-induced inhibition of cell
proliferation.
Results
Expression of p14
ARF
leads to phosphorylation of p53S15
The fact that both the cellular DNA damage response
and oncogenic stimuli target Mdm2 to stabilize p53
promoted us to investigate whether p14
ARF
could utilize
the DNA damage pathway to facilitate p53 activation.
As phosphorylation of p53S15 is a typical event in the
cellular DNA damage response (Zhou and Elledge,
2000), we first examined if p14
ARF
could induce p53S15
phosphorylation. Since expression of ARF potently
inhibits the proliferation of NIH3T3 cells (Kamijo et al.,
1998), we constructed a tetracycline (Tet)-inducible
p14
ARF
-expressing NIH3T3 cell line, NIH3T3
Teton
/
p14
ARF
, and a control cell line infected with an empty
Tet-responsive vector retrovirus (pRevTet), NIH3T3
Te-
ton
/pRevTet. Addition of doxycycline (Dox), a Tet
analogue, specifically induced p14
ARF
expression in a
time-dependent manner (Figure 1a). As expected,
p14
ARF
stabilized p53 and thereby upregulated p21
CIP1
(Figure 1a). Consistent with previous publications
(Weber et al., 1999; Zhang and Xiong, 1999), p14
ARF
was expressed exclusively in the nuclei (Figure 1b).
Interestingly, ectopically expressed p14
ARF
also en-
hanced the phosphorylation of p53S15, which was
not observed in control NIH3T3
Teton
/pRevTet cells
(Figure 1a). Kinetically, the phosphorylation of
p53S15 peaks between 16 and 24 h of Dox treatment,
decreasing by day 2 of Dox induction (Figure 1a). The
total p53 level follows a similar pattern, while p14
ARF
is
detected at 16 h and remains at peak level after 24–48 h
of Dox induction (Figure 1a). The reason for a decline in
p53 and phosphorylated p53S15 in the context of
ongoing high levels of p14
ARF
is not clear, but may
indicate activation of factors leading to p53 down-
regulation. This is consistent with the theme of network
regulation on p53 protein stability, in that p53 protein
upregulates its negative regulator Mdm2, leading to p53
degradation (Roth et al., 1998).
To determine whether p14
ARF
-induced phosphoryla-
tion of p53S15 is unique to NIH3T3 cells, we also
studied this event in MCF7 and IMR90 cells.
MCF7 cells were infected with either pBabe or
pBabe-p14
ARF
retrovirus for 48 h. Again, expression
of p14
ARF
increases the levels of p53S15 phosphoryla-
tion and total p53 protein (Figure 1d). The same result
was also obtained in IMR90 cells (Figure 2). To
ascertain at the cellular level whether cells expressing
p14
ARF
were those demonstrating phosphorylation of
p53S15, MCF7 cells infected with pBabe or pBabe-
p14
ARF
were double stained with an anti-FLAG
antibody (M2) for p14
ARF
and a polyclonal antibody
specific for phosphorylated p53S15. Cells expressing
high levels of p14
ARF
indeed exhibit enhanced phosphor-
ylation on the serine 15 (Figure 3D, ‘e, f, g, h’ for
MCF7ConRNAi cells). The same results were also
obtained in parental MCF7 cells (data not shown).
Taken together, these data demonstrate that expression
of p14
ARF
induces the phosphorylation of p53S15 in
multiple cells.
ATM contributes to the functions of p14
ARF
YLiet al
7356
Oncogene
Expression of oncogenes induces phosphorylation of
p53S15
The major established function of ARF is the activation
of p53 in response to oncogenic signals. Several
oncogenes, notably E1A,c-myc and E2F1, have been
shown to activate p53 at least partially through ARF
(de Stanchina et al., 1998; Zindy et al., 1998; Dimri et al.,
2000), suggesting that oncogenic signals may also
induce p53S15 phosphorylation. To examine such a
possibility, IMR90 cells were infected with an empty
vector retrovirus or retroviruses expressing p53,
p14
ARF
, E1A,c-myc or E2F1. Ectopic expression of
these transgenes was demonstrated (Figure 2). Com-
pared to the empty vector retrovirus infection, exogen-
ous p53, p14
ARF
, E1A, c-myc and E2F1 all induce the
phosphorylation of p53S15 (Figure 2). Expression of
p53, p14
ARF
and E1A modestly upregulates p21
CIP1
(Figure 2). Taken together, these data support the
concept that oncogenic-stimulated p14
ARF
leads to the
phosphorylation of p53S15.
Expression of p53 induces phosphorylation of p53S15
Oncogenic signals can also activate p53 in ARF-
independent manner. While how this is achieved
remains largely unknown, it has been clearly demon-
strated by several groups that these ARF-independent
oncogenic pathways lead to p53 stabilization (de
Stanchina et al., 1998; Zindy et al., 1998; Russell
et al., 2002). It is thus possible that a high level of
p53 in these settings is sufficient to initiate p53S15
phosphorylation. In order to test this hypothesis,
NIH3T3 cells expressing human p53 (detected by DO-
1 antibody specific for human p53) in a Tet-inducible
manner were constructed (Figure 1c). Ectopic p53
indeed induced p53S15 phosphorylation and upregu-
lated p21
CIP1
(Figure 1c). Kinetically, Dox-induced p53
expression precedes p53S15 phosphorylation, as p53
expression is detectable at 8 h, while the serine 15
phosphorylation is not seen until 16 h following Dox
induction (Figure 1c). Dox treatment of NIH3T3
Teton
/
pRevTet cells (controls) leads to neither p53 stabiliza-
tion nor phosphorylation of p53S15 (Figure 1a). In
addition, expression of p53 in MCF7 and IMR90 cells
also resulted in enhancement of p53S15 phosphorylation
(Figures 1d and 2), indicating that this is not limited to
one cell line.
Figure 1 Expression of p14
ARF
and p53 results in p53S15
phosphorylation. (a) NIH3T3
Teton
/pRevTet and NIH3T3
Teton
/
p14
ARF
cells were induced with Dox (2 mg/ml) for the times
indicated. Expression of p14
ARF
, p53S15 phosphorylation (S15-P),
p21
CIP1
(p21) and actin were determined by Western blot using an
anti-FLAG antibody (M2) for p14
ARF
and other specific antibodies.
(b) NIH3T3
Teton
/pRevTet and NIH3T3
Teton
/p14
ARF
cells were
treated with or without Dox for 1 day. Immunofluorescent staining
with M2 antibody for p14
ARF
was performed. The nuclei were
counterstained with DAPI. The merged images of M2 and DAPI
(overlay) are shown. (c) NIH3T3
Teton
/p53 (human) cells were
induced to express p53 by Dox as indicated. The expression of
p53 (human), S15-P, p21 and actin was determined using DO-1, an
antibody specific for human p53, and other specific antibodies. (d)
MCF7 cells were infected with pBabe, pBabe-p14
ARF
and pBabe-
p53 as indicated, or lanes 1, 2 and 3, respectively (inset). The
expression of p14
ARF
(ARF), p53 and p53S15 phosphorylation
(S15-P) was determined by Western blot using specific antibodies
(inset). The level of p53S15 phosphorylation in cells infected with
p14
ARF
or p53 is standardized against that in cells infected with
pBabe. Experiments were repeated at least thrice
ATM contributes to the functions of p14
ARF
YLiet al
7357
Oncogene
ATM mediates p14
ARF
-induced p53S15 phosphorylation
Since S15 of p53 is phosphorylated by ATM/ATR in the
cellular DNA damage response, we reasoned that ATM
could be a candidate kinase for the phosphorylation of
S15 of p53 in response to p14
ARF
signals. This is
consistent with a recent report indicating a role for
ARF in the cellular DNA damage response (Khan et al.,
2000). To determine if ATM plays a role in p14
ARF
-
induced phosphorylation on p53S15, IMR90 and AT
fibroblasts (GM05823) were infected with an empty
vector retrovirus or retrovirus expressing one of p53,
p14
ARF
, E1A, c-myc or E2F1. Comparable levels of these
transgenes were expressed in IMR90 vs GM05823 cells
(Figure 2). Compared to the empty vector retrovirus
infection, levels of p53S15 phosphorylation were in-
creased in IMR90 cells expressing p53, p14
ARF
,c-myc
and E2F1 (Figure 2), and dramatically reduced in
GM05823 cells (Figure 2). E1A-induced p53S15 phos-
phorylation was equally efficacious in IMR90 and in
GM05823 cells (Figure 2), indicating a role for other
kinases, possibly ATR. Collectively, these observations
support the notion that ATM is a candidate kinase in
phosphorylation of p53S15 in response to ARF signals.
To exclude the possibility that other defects associated
with the human AT fibroblast GM05823 cells might
prevent p14
ARF
from inducing p53S15 phosphorylation,
and to examine whether the involvement of ATM in this
process is unique to human fibroblast cells, we
constructed MCF7 cells in which ATM expression was
specifically downregulated by RNA interference
(RNAi). MCF7 cells were infected with pRIH1 retro-
virus expressing either ATM RNAi fragment 1 or 2.
Blast searching confirmed that both RNAi fragments
match to human ATM only. Compared to the empty
vector (pRIH1) infection, pRIH1-ATMRNAi1 did not
reduce ATM expression (data not shown). We thus used
ATMRNAi1 as a negative control, and designated it as
control RNAi (ConRNAi). In comparison with cells
infected with ConRNAi, those infected with ATMR-
NAi2 (designated as ATMRNAi) display significantly
reduced levels of ATM protein and mRNA (Figure 3A
and B). Densitometry analysis revealed that ATM
protein and mRNA in ATMRNAi cells were expressed
at levels of about 10% of those seen in ConRNAi cells
(data not shown). Consistent with a major role of ATM
in etoposide (ETOP)-induced p53 stabilization and
p53S15 phosphorylation, downregulation of ATM
severely impaired these events (Figure 3A). Taken
together, these data clearly demonstrate the reduction
of ATM in MCF7ATMRNAi cells.
Using MCF7ConRNAi and MCF7ATMRNAi cells,
the role of ATM in p14
ARF
- and p53-induced p53S15
phosphorylation was examined. Both cells were infected
with an empty vector retrovirus or retrovirus expressing
either p14
ARF
or p53 (Figure 3C and D). Both p14
ARF
and p53 were expressed to comparable levels in
MCF7ConRNAi vs MCF7ATMRNAi cells, as detected
by immunofluorescent staining (Figure 3D, e vs q for
p14
ARF
, i vs u for p53) and by Western blot (data not
shown). Although by comparison to the level of p53S15
phosphorylation in pBabe-infected cells the relative
increase of p53S15 phosphorylation (4–5 fold) induced
by p14
ARF
and p53 remains the same in MCF7ConRNAi
vs MCF7ATMRNAi cells, the absolute levels of p53S15
phosphorylation induced by p14
ARF
and p53 are
significantly reduced in MCF7ATMRNAi cells (Figure
3C and D, f vs r and j vs v). Taken together, the above
data demonstrate a major role for ATM in phosphor-
ylation of p53S15 in response to oncogene-p14
ARF
signals.
ARF enhances ATM kinase activity and induces nuclear
ATM foci
ATM and ATR phosphorylate the S15 of p53 in the
cellular DNA damage response. This is mediated by
increasing ATM kinase activity and redistribution of
Figure 2 ATM plays a role in p14
ARF
-induced p53S15 phosphorylation. IMR90 and GM05823 AT fibroblasts were infected with
pBabe-based retroviruses (‘V’ for pBabe vector) as indicated. Expression of p53, p14
ARF
,c-myc, E1A, E2F1, p53S15 phosphorylation
(S15-P), p21
CIP1
and actin were determined by Western blot using specific antibodies as enumerated in Materials and methods. The
second p53 bands are alternative p53 translation products (Yin et al., 2002)
ATM contributes to the functions of p14
ARF
YLiet al
7358
Oncogene
ATR in response to DNA damage (Tibbetts et al.,
2000; Bakkenist and Kastan, 2003). To support the
observation that ATM may phosphorylate p53S15 by
ARF signals, we examined ATM kinase activity in
NIH3T3
Teton
/p14
ARF
cells. These cells were chosen since
ectopic ARF induces p53S15 phosphorylation
(Figure 1a) and inhibits cell proliferation (Kamijo
et al., 1998). Induction of p14
ARF
enhanced ATM kinase
Figure 3 Downregulation of ATM attenuates p14
ARF
-induced p53S15 phosphorylation. (A) MCF7 cells were infected with pRIH1-
based retrovirus expressing a control RNAi fragment (ConRNAi) or ATMRNAi fragment as indicated. Infections were selected in
hygromycin. Cells were treated with or without ETOP (0.1 m
M) for 2 h. The expression of ATM, p53, S15-P and actin was detected by
Western blot using specific antibodies. (B) RT–PCR was carried out on RNA purified from MCF7ConRNAi and MCF7ATMRNAi
cells with specific primers to ATM and actin as indicated. The 400 bp ATM and 241 bp actin products are shown. Amplification of
actin was linear up to 25 cycles for both RNA populations (data not shown). Thus, actin was amplified with 23 cycles and ATM was
amplified with 35 cycles. The experiment was repeated thrice. (C) MCF7ConRNAi and MCF7ATMRNAi cells were infected with
pBabe-based retroviruses as indicated. Phosphorylation on p53S15 (S15-P) was examined by Western blot and standardized against
that observed in pBabe-infected MCF7ConRNAi cells. (D) The expression of p14
ARF
, p53 and S15-P in MCF7ConRNAi and
MCF7ATMRNAi cells infected with pBabe or p14
ARF
or p53 retrovirus was determined by immunostaining using specific antibodies.
Nuclei were counterstained with DAPI. (a, e, q) were stained for ARF; (i, m, u) were stained for p53; (b, f, j, n, r, v) were stained
for p53S15 phosphorylation (S15-P). Reduced DAPI staining in S15-P-positive nuclei may be attributable to apoptotic DNA
degradation. Failure to detect p53 signals in panel (m) was due to the low level of endogenous p53 in these cells compared to cells
expressing ectopic p53
ATM contributes to the functions of p14
ARF
YLiet al
7359
Oncogene
activity in NIH3T3
Teton
/p14
ARF
cells, while Dox alone
induced no changes in ATM kinase activity in
NIH3T3
Teton
/pRevTet (vector) cells (Figure 4a). Com-
pared to ETOP treatment, which induces an approxi-
mately fivefold increase in ATM kinase activity,
induction of p14
ARF
and p53 by Dox in NIH3T3
Teton
/
p14
ARF
and NIH3T3
Teton
/p53 cells, respectively, enhances
ATM kinase activity about twofold (Figure 4b).
To determine the possible mechanisms by which ARF
activates ATM, we examined cellular localization of
ATM in ARF-expressing vs control cells. Interestingly,
induction of ARF in NIH3T3
Teton
/p14
ARF
cells led to
the formation of ATM foci in nuclei (Figure 5). This
was not observed in mock-treated (without Dox)
NIH3T3
Teton
/p14
ARF
cells or in individual Dox-treated
NIH3T3
Teton
/p14
ARF
cells that did not express ARF
(arrow, Figure 5), demonstrating the specificity of ATM
foci for ARF-expressing cells. As radiation induces
phosphorylation on S1981 of ATM, an event leading to
ATM activation by dissociation of its kinase domain
from FAT domain (Bakkenist and Kastan, 2003), the
formation of ATM foci induced by ARF might result
from ATM conformation changes, leading to releasing
ATM from FAT domain and thus ATM activation.
Taken together, the above data demonstrate that ARF
induces ATM activation, and suggest that this may be
mediated by the formation of ATM foci in response to
ARF signals.
ATM contributes to the tumor-suppressing functions
of p14
ARF
The physiological function of ARF–p53 activation by
oncogenic signals is to prevent cell proliferation and
thereby suppress tumorigenesis. To determine whether
ATM contributes to this function, we infected IMR90
and AT GM05823 fibroblasts with an empty retrovirus
or retroviruses expressing p53, p14
ARF
, E1A, c-myc or
E2F1. Expression of these transgenes was confirmed
(Figure 2). Infections were selected in puromycin to
determine surviving cells, which were then stained with
0.1% crystal violet (Figure 6a). Compared to the empty
vector infection, expression of these transgenes substan-
tially inhibited the proliferation of IMR90 cells, as
expected (Figure 6a). Interestingly, the ability of
ectopically expressed p14
ARF
and c-myc to inhibit cell
proliferation, compared to the empty vector-infected
cells, was significantly attenuated in GM05823 cells
(Figure 6a). This result also indirectly reveals that the
reduced survival of IMR90 cells infected with p14
ARF
or
c-myc retrovirus when compared to pBabe infection
(Figure 6a) was not due to low titers of p14
ARF
or c-myc
retrovirus, consistent with our titration experiments
showing comparable titers for pBabe, p53, p14
ARF
,
E2F1, c-myc and E1A retrovirus (for detail, see
Materials and methods). Overall, less GM05823 cells
than IMR90 cells observed in this experiment may be
attributed to the well-recognized reduced proliferation
capacity associated with GM05823 cells. E1A still,
however, potently blocks the proliferation of
GM05823 cells, which agrees well with the observations
that expression of E1A produces comparable levels of
Figure 4 Ectopic p14
ARF
activates ATM. (a) NIH3T3
Teton
/pRevTet
(Vector) and NIH3T3
Teton
/p14
ARF
(ARF) cells were treated with or
without Dox for 24 h as indicated. ATM was immunoprecipitated
(IP) from 1 mg of total cell lysate protein using an anti-ATM
antibody (Ab-3) and used to phosphorylate GST-p53 (2 mg) in vitro.
GST-p53 was detected by a Western blot using an anti-p53
antibody (DO-1). P53S15 phosphorylation (S15-P) and IP ATM
were examined by Western blot using specific antibodies. (b)
NIH3T3
Teton
/pRevTet (Vector) cells were treated with either ETOP
at 100 m
M for 4 h or Dox for 24 h. NIH3T3
Teton
/p14
ARF
(ARF) and
NIH3T3
Teton
/p53 (p53) cells were induced with Dox for 24 h. ATM
kinase assay and detection of p53S15 phosphorylation, GST-p53
and IP ATM were then carried out as described in (a). The levels of
p53S15 phosphorylation in ETOP or Dox-treated cells were
standardized against those in the corresponding controls. Experi-
ments were repeated thrice
Figure 5 Expression of p14
ARF
results in formation of ATM foci.
NIH3T3
Teton
/p14
ARF
cells were induced with Dox for 24 h and then
immunofluorescently stained with an anti-FLAG antibody (M2)
for p14
ARF
(green) or an anti-ATM antibody (Ab-3) (red). Nuclei
were counterstained with DAPI (blue). The merged images (Merge)
of ARF (green) and ATM (red) are shown. Arrow indicates Dox-
treated cell without the ARF expression. In such cells, no ATM
foci were detected
ATM contributes to the functions of p14
ARF
YLiet al
7360
Oncogene
p53S15 phosphorylation in GM05823 cells vs in IMR90
cells (Figure 2). The reason for the inhibition of the
proliferation of GM05823 cells by ectopic p53 and E2F1
is not clear. It is possible that, in this setting, both
proteins may use means other than p53S15 phosphor-
ylation to modulate this inhibition.
To further confirm the role of ATM in p14
ARF
-
mediated tumor suppression, we infected MCF7ConR-
NAi and MCF7ATMRNAi cells with pBabe, p14
ARF
or
p53 retrovirus. Surviving cells were selected in puromy-
cin and stained with crystal violet (Figure 6b). Com-
pared to MCF7ConRNAi cells, MCF7ATMRNAi cells
are dramatically resistant to p14
ARF
- and p53-induced
growth inhibition (Figure 6b). Taken together, these
results demonstrate the functional importance of ATM
in p14
ARF
-mediated tumor suppression.
Attenuation of ATM function by prolonged ARF
expression allows cell survival of ectopic p14
ARF
The observation that downregulation of ATM in AT
fibroblasts (GM05823) and MCF7ATMRNAi cells
reduced p14
ARF
-mediated tumor suppression (Figure 6)
suggested that cells which survive p14
ARF
expression may
have ATM function attenuated. To test this possibility,
MCF7ConRNAi and MCF7ATMRNAi cells were
infected with either an empty vector (pBabe) or
pBabe/p14
ARF
retrovirus and selected in puromycin for
2–3 weeks. To avoid the possible artifacts associated
with individual surviving cell colonies, a pooled
population of surviving cells was harvested. In compar-
ison to pBabe-infected cells, those surviving ectopic
p14
ARF
proliferated at a reduced rate (data not shown),
suggesting the presence of ectopic ARF in the cells.
Indeed, these cells did express ectopic p14
ARF
(Figure 7a).
To examine ATM functions in those cells, we treated
them with 100 m
M ETOP and examined phosphorylation
of p53 S15. As expected, ETOP-induced S15 phosphor-
ylation in pBabe-infected MCF7ATMRNAi cells was
Figure 6 ATM contributes to p14
ARF
-mediated tumor suppres-
sion. IMR90 and GM05823 fibroblasts (a) and MCF7ConRNAi
and MCF7ATMRNAi cells (b) were infected with pBabe-based
retroviruses as indicated. Infections were selected in puromycin and
surviving cells were stained with 0.1% crystal violet. Comparable
titers for individual retrovirus were used (for detail, see Materials
and methods). The experiments were repeated twice
Figure 7 Prolonged expression of ectopic p14ARF attenuates
ATM function. MCF7ConRNAi (ConRNAi) and MCF7ATMR-
NAi (ATMRNAi) cells were infected with pBabe or pBabe/p14
ARF
(p14
ARF
) as indicated. Surviving cells were selected in puromycin
and recovered as a pooled population. (a) Expression of ectopic
p14
ARF
in the corresponding surviving cells was determined by
Western blot using an anti-FLAG antibody (M2). (b) The
corresponding surviving cells were treated with ETOP for 2 h and
were examined for the expression of p53S15 phosphorylation
(S15-P) or total p53 by Western blot using specific antibodies
ATM contributes to the functions of p14
ARF
YLiet al
7361
Oncogene
significantly reduced compared to pBabe-infected
MCF7ConRNAi cells (Figure 7b). Interestingly, the
MCF7ConRNAi cells surviving ectopic p14
ARF
also
displayed substantial reduction in ETOP-induced
p53S15 phosphorylation when compared to pBabe-
infected MCF7ConRNAi cells (Figure 7b), which was
further reduced when the corresponding MCF7ATMR-
NAi cells were analysed (Figure 7b). The basal level
of p53 in MCF7ConRNAi or MCF7ATMRNAi
cells surviving ectopic p14
ARF
did not differ signifi-
cantly from that in both cells infected with pBabe
(Figure 7b). MCF7 cells are known to contain
relatively high basal levels of p53 (Wu and Tang,
unpublished observation). Since ETOP-induced p53S15
phosphorylation is largely reduced in GM05823
(Tang et al., 2002) and MCF7ATMRNAi cells
(Figure 3A), the reduction of ETOP-induced S15
phosphorylation in viable p14
ARF
-expressing MCF7 cells
is consistent with the concept that ATM function was
attenuated in these cells.
Discussion
While p19
ARF
clearly plays a major role in murine
tumorigenesis, there might be some debate regarding the
contribution of p14
ARF
vs p16
INK4A
in human tumorigen-
esis. However, mounting evidence accumulated recently
reveals an important function of p14
ARF
in blocking
tumorigenesis in humans. By assaying deletion of the
exon 1b (specific for p14
ARF
) or methylation of p14
ARF
promoter, inactivation of p14
ARF
has been detected in
33% of colon carcinomas (Burri et al., 2001), 25% of
oligodendrogliomas (Watanabe et al., 2001), 14% of
breast cancers (Ho et al., 2001), 26.5% of oral squamous
cell carcinomas and 32–58% of gliomas (Ichimura et al.,
2000; Shintani et al., 2001). Mechanistically, p14
ARF
bridges the RB and p53 tumor suppressors (Bates et al.,
1998). However, ARF is not the only pathway leading to
p53 activation by oncogenic signals.
The DNA damage and oncogene–ARF pathways
activate p53 by targeting Mdm2. The data reported here
reveal an intriguing link between the ATM–Mdm2–p53
DNA damage pathway and the oncogenic ARF–
Mdm2–p53 cellular response. We have established that
ATM contributes to the tumor suppressor functions of
p14
ARF
. Ectopic p14
ARF
enhanced ATM kinase activity
(Figure 4). Reduction of ATM function attenuated
p14
ARF
mediated tumor suppression on AT fibroblasts
(GM05823) and MCF7ATMRNAi cells (Figure 6), and
cells surviving ectopic p14
ARF
may have ATM function
attenuated (Figure 7). Our finding that ATM facilitates
ARF-mediated tumor-suppressing function is consistent
with a previous report showing that ARF plays a role in
the premature senescence caused by loss of ATM in
MEFs (Kamijo et al., 1999). Since ATM sits far
upstream in the cellular DNA damage response and
coordinates p53 activation by phosphorylation of p53,
Chk2 and Mdm2 (Zhou and Elledge, 2000), utilization
of ATM by p14
ARF
may significantly promote the tumor
suppression functions of ARF.
Phosphorylation of p53S15 by ARF would be
expected to further destabilize the Mdm2–p53 complex.
By direct interaction with p53, Mdm2 inhibits p53’s
transcriptional activity (Momand et al., 1992), indicat-
ing a requirement for separation of p53 from Mdm2 for
full-scale p53 activation. Mdm2 binds to the N-terminus
of p53 between the residues 17 and 22 and phosphor-
ylation at S15 and S20 facilitates p53 dissociation from
Mdm2 (Shieh et al., 1997; Chehab et al., 1999).
However, since an ARF–Mdm2–p53 ternary complex
was observed under certain conditions (Kamijo et al.,
1998; Kashuba et al., 2003), it is unclear biochemically
whether ARF directly destabilizes the Mdm2–p53
complex via interaction with Mdm2. Our findings
suggest a mechanism whereby ARF could promote
dissociation of Mdm2 from p53, in that ARF-induced
ATM activation leads to p53S15 phosphorylation. This
is consistent with the common cell biology phenomenon
of exploitation of the same mechanism by multiple
pathways.
We have shown here that ectopic p53 also leads to
p53S15 phosphorylation, an observation consistent with
a previous report (Russell et al., 2002). As oncogenes
can also activate p53 via ARF-independent pathways,
this result indicates that an initial oncogenic signal may
stabilize p53, through ARF-dependent or -independent
pathways, leading to partial p53 activation. This induces
modifications including S15 phosphorylation on p53,
leading to its full-scale activation. However, this does
not exclude the possibility that ARF or other oncogenic
signal transducers may also directly activate these
modification processes. These two scenarios are not
mutually exclusive.
While both the cellular DNA damage response and
oncogene–ARF signals activate ATM kinase, they may
achieve this through different strategies. DNA damage
activates ATM kinase by its dissociation from the FAT
domain via phosphorylation of S1981 (Bakkenist and
Kastan, 2003). ARF signals, on the other hand, lead to
formation of ATM nuclear foci (Figure 5). This may be
the result of or may result in a change in ATM
conformation, leading to separation of ATM kinase
domain from the FAT domain. Redistribution of a
p53S15 kinase is also a strategy used by the cellular
DNA damage response, in that it is generally believed
that formation of ATR foci plays an essential role in
p53S15 phosphorylation in the cellular DNA damage
response (Tibbetts et al., 2000). It is thus tempting to
propose that the formation of ATM foci may be a major
mechanism underlying ARF-induced ATM activation.
Our observation that the expression of E2F1, c-myc
and E1A results in p53S15 phosphorylation is consistent
with recent reports concerning E2F1 and c-myc (Rogoff
et al., 2002; Vafa et al., 2002). However, oncogenic
stimuli-induced phosphorylation of p53S15 is complex.
Expression of c-myc may induce DNA damage through
generation of reactive oxygen species in normal human
fibroblasts (NHF) (Vafa et al., 2002). Indeed, the
addition of the antioxidant N-acetyl-
L-cysteine (NAC)
significantly reduced the c-myc-induced p53S15 phos-
phorylation. However, NAC does not completely
ATM contributes to the functions of p14
ARF
YLiet al
7362
Oncogene
abolish c-myc-induced p53S15 phosphorylation and
does not rescue c-myc-mediated inhibition of NHF
cell proliferation (Vafa et al., 2002), suggesting a role
of p14
ARF
in c-myc-induced p53S15 phosphorylation
and the inhibition of NHF cell proliferation in this
context. This is consistent with reports showing that
the Myc signals activate p53-dependent tumor
suppression at least partially via ARF (Zindy et al.,
1998; Russell et al., 2002). E2F1 was shown to induce
p53S15 phosphorylation independently of ARF (Rogoff
et al., 2002; Russell et al., 2002). This does not exclude a
role of ARF in E2F1-induced p53S15 phosphorylation
in other situations, as E2F1 also induces ARF
transcription (Bates et al., 1998; Inoue et al., 1999;
Dimri et al., 2000; Jin and Levine, 2001). Our
interpretation that E2F1 and c-myc induce the S15
phosphorylation in our system at least partially via the
action of p14
ARF
is consistent with these reports.
Further, our interpretation is not mutually exclusive
from the concepts of oncogene-induced ATM activation
through DNA damage.
Regardless of what may ultimately be the major
mechanism whereby oncogene expression activates
ATM, we have shown here that ATM contributes to
the tumor-suppressing functions of p14
ARF
. Investiga-
tion of the underlying mechanism of this process will be
of considerable interest in the field of ARF research.
Materials and methods
Materials, cell cultures and plasmids
The Tet-inducible cell line, 293
Teton
, was from Clontech. IMR90
and the AT fibroblast cell line GM05823 were from ATCC and
NIGMS Human Genetic Cell Repository, respectively.
P14
ARF
was cloned by PCR from C33A cervical carcinoma
cells, the sequences confirmed and tagged with a C-terminal
FLAG epitope. The C-terminal FLAG-tagged p14
ARF
was
subcloned into either pTET1 (a Tet-responsive vector) or a
retroviral vector pBabe or a retroviral vector containing a Tet
response element, pRevTet (Clontech). Plasmids expressing
retroviral Gag-Pol (GP) and VSV-G proteins were from
Stratagene.
Retroviral infection
Retroviral infection was performed as we have published, with
some modifications (Wu et al., 2002). Retroviral plasmid DNA
was co-transfected into 293T cells with the GP and VSV-G
plasmids, using a calcium phosphate method for 48 h. A virus-
containing medium was then harvested and used to infect
target cells in the presence of 10 mg/ml of polybrene (Sigma).
To determine the retrovirus titer, a LacZ retrovirus without a
mammalian selection marker was co-packed with a retroviral
vector expressing a cDNA of interest at a ratio of 1 : 5
(LacZ:cDNA of interest). The virus was used to infect
NIH3T3 cells for 36 h before staining for LacZ. Comparable
levels of staining among individually packed retroviruses
indicate comparable titers.
Generation of Tet-inducible p53 and p14
ARF
-expressing cell lines
NIH3T3 cells were infected with pRevTeton retrovirus,
expressing the reverse Tet-controlled transactivator
(Clontech), to generate NIH3T3
Teton
cells, which were then
infected with pRevTet-p53 and pRevTet-p14
ARF
, respectively.
The infections were selected in a medium containing hygro-
mycin. 293
Teton
/p14
ARF
cells were constructed by transfection
of 293
Teton
cells with pTET1-p14
ARF
, using the calcium
phosphate method.
Western blot
Cells lysate was prepared as previously described (Wu et al.,
2002). The following primary antibodies were used: anti-
FLAG (M2 at 3 mg/ml, Sigma), anti-p53 (FL-393 at 1 mg/ml,
DO-1 at 0.5 mg/ml, Santa Cruz), anti-phosphorylated p53S15
(1 mg/ml, Cell Signaling), anti-actin (0.5 mg/ml, Santa Cruz),
anti-p21
CIP1
(1 mg/ml, Santa Cruz) and anti-ATM (Ab-3, 2 mg/
ml, Oncogene Res. Prod).
Immunofluorescent detection of p53, p14
ARF
and ATM
Cells were fixed with 3% paraformaldehyde and incubated for
2 h with affinity-purified monoclonal anti-FLAG antibody
(M2 at 1 : 50), anti-p53, anti-phospho-p53S15, or anti-ATM
(Ab-3), followed by incubation with FITC-conjugated anti-
mouse or Cy5-Donkey anti-rabbit IgG (Jackson Immnuno
Research). Immunofluorescence was detected using a fluor-
escent microscope (Zeiss).
Silencing of endogenous ATM in MCF7 cells by RNAi
A novel retroviral vector, pRIH1, was constructed to direct the
expression of an RNAi fragment by the human RNase P RNA
H1 promoter, an RNA polymerase III promoter (Y Li and D
Tang, unpublished data). The control ATM RNAi fragment
(ConRNAi) used in these experiments was sense: GATCCC
C
GCACCAGTCCAGTATTGGCTTCAAG-AGAGCCAATA
CTGGACTGGTGCTTTTTGGAAA and antisense: AGCTT
TTCCAAAA-AGCACCAGTCCAGTATTGGCTCTCTTGA
AGCCAATACTGGACTGGTGCGGG. The ATM RNAi
fragment (ATMRNAi) was sense: GATCCCC
TCTCAGCAA-
CAGTGGTTAGTTCAAGAGACTAACCACTGTTGCTGA-
GATTTTTGGAAA and antisense: AGCTTTTC-CAAAAA
TCTCAGCAACAGTGGTTAGTCTCTTGAACTAACCAC
TGTTGCTGAGAGGG. Nucleotides bolded and underlined
are those of the sense and antisense sequences that will form a
hairpin structure upon transcription inside cells. High titer
retrovirus was packed and used to infect MCF7 cells.
Reverse transcription (RT)–PCR
Total RNA was isolated using Trizol (Life Technology) from
MCF7ConRNAi and MCF7ATMRNAi cells according to the
manufacturer’s instruction. RT was carried out using an RT
kit (Clontech). Specific primers used for ATM are: 5
0
-
TCTTGATAAATGAGCAGTCAGC-3
0
(upper) and 5
0
-ACA
CGTTCAGCTACTTTGTCG-3
0
(lower); for b-actin: 5
0
-AAC
CCCAAGGCCAACCGCGAGAAG-3
0
(upper) and 5
0
-TCAT
GAGGTAGTCAGTCAG-3
0
(lower). Amplification of actin
up to 25 cycles was in the linear range (data not shown). Thus,
RT–PCR of ATM and actin was achieved with 35 and 23
cycles, respectively.
ATM kinase assay
An N-terminal fragment (residues 1–40) of p53 was expressed
as a GST fusion (GST-p53) protein in E. coli (Bl-21) as
previously described (Tang et al., 1997). ATM was immuno-
precipitated from cell lysate using an anti-ATM antibody (Ab-
3, Oncogene Res Prod) and used to phosphorylate GST-p53
ATM contributes to the functions of p14
ARF
YLiet al
7363
Oncogene
(2 mg) in vitro at 301C for 30 min according to a published
procedure (Chan et al., 2000). After termination of the
reaction by addition of 10 m
M EDTA, protein G bead with
bound ATM was precipitated by centrifugation (5000 g for
5 min). The supernatant was then incubated with glutathione
Sepharose to precipitate GST-p53. The supernatant was
combined with the protein G bead precipitates for analysis
of precipitated ATM by a Western blot. Glutathione
Sepharose-precipitated GST-p53 was analysed for p53S15
phosphorylation by a Western blot using a specific antibody
(Cell Signaling).
Acknowledgements
We thank Drs Joseph R Nevins of Duke University and Frank
Graham of McMaster University for providing E2F1 and
E1A, respectively. This work was supported in part by the
National Cancer Institute of Canada (grant no. 13009 200), a
grant of Kidney Foundation of Canada, Canadian Founda-
tion for Innovation (grant no. 6987) all to DT and a grant
from the Canadian Institutes of Heath Research to AI. We like
to dedicate this work to Dr Vincent J Kidd, member
(professor) of St Jude Children’s Research Hospital, who
suddenly passed away on May 7, 2004.
References
Bakkenist CJ and Kastan MB. (2003). Nature, 421, 499–506.
Bates S, Phillips AC, Clark PA, Stott F, Peters G, Ludwig RL
and Vousden KH. (1998). Nature, 395, 124–125.
Burri N, Shaw P, Bouzourene H, Sordat I, Sordat B, Gillet M,
Schorderet D, Bosman FT and Chaubert P. (2001). Lab.
Invest., 81, 217–229.
Chan DW, So SC, Block W, Ye R, Khanna KK, Wold MS,
Douglas P, Goodarzi AA, Pelley J, Taya Y, Lavin MF and
Lees-Miller SP. (2000). J. Biol. Chem., 275, 7803–7810.
Chehab NH, Malikzay A, Stavridi ES and Halazonetis TD.
(1999). Proc. Natl. Acad. Sci. USA, 96, 13777–13782.
Cong F, Zou X, Hinrichs K, Calame K and Goff SP. (1999).
Oncogene, 18, 7731–7739.
Damalas A, Kahan S, Shtutman M, Ben Ze’ev A and Oren M.
(2001). EMBO J., 20, 4912–4922.
de Stanchina E, McCurrach ME, Zindy F, Shieh SY, Ferbeyre
G, Samuelson AV, Prives C, Roussel MF, Sherr CJ and
Lowe SW. (1998). Genes. Dev., 12, 2434–2442.
Dimri GP, Itahana K, Acosta M and Campisi J. (2000). Mol.
Cell Biol., 20, 273–285.
Eischen CM, Weber JD, Roussel MF, Sherr CJ and Cleveland
JL. (1999). Genes. Dev., 13, 2658–2669.
Hirao A, Kong YY, Matsuoka S, Wakeham A, Ruland J,
Yoshida H, Liu D, Elledge SJ and Mak TW. (2000). Science,
287, 1824–1827.
Ho GH, Calvano JE, Bisogna M, Abouezzi Z, Borgen PI,
Cordon-Cardo C and van Zee KJ. (2001). Breast Cancer
Res. Treat., 65, 225–232.
Honda R, Tanaka H and Yasuda H. (1997). FEBS Lett., 420,
25–27.
Honda R and Yasuda H. (1999). EMBO J., 18, 22–27.
Ichimura K, Bolin MB, Goike HM, Schmidt EE, Moshref A
and Collins VP. (2000). Cancer Res., 60, 417–424.
Inoue K, Roussel MF and Sherr CJ. (1999). Proc. Natl. Acad.
Sci. USA, 96, 3993–3998.
Ito A, Lai CH, Zhao X, Saito S, Hamilton MH, Appella E and
Yao TP. (2001). EMBO J., 20, 1331–1340.
Jin S and Levine AJ. (2001). J. Cell Sci., 114, 4139–4140.
Jones SN, Roe AE, Donehower LA and Bradley A. (1995).
Nature, 378, 206–208.
Kamijo T, van de Kamp E, Chong MJ, Zindy F, Diehl JA,
Sherr CJ and McKinnon PJ. (1999). Cancer Res., 59,
2464–2469.
Kamijo T, Weber JD, Zambetti G, Zindy F, Roussel MF and
Sherr CJ. (1998). Proc. Natl. Acad. Sci. USA, 95, 8292–8297.
Kamijo T, Zindy F, Roussel MF, Quelle DE, Downing JR,
Ashmun RA, Grosveld G and Sherr CJ. (1997). Cell, 91,
649–659.
Kashuba E, Mattsso K, Klein G and Szekel L. (2003). Mol.
Cancer, 2, 18.
Khan SH, Moritsugu J and Wahl GM. (2000). Proc. Natl.
Acad. Sci. USA, 97, 3266–3271.
Llanos S, Clark PA, Rowe J and Peters G. (2001). Nat. Cell
Biol., 3, 445–452.
Lyakhovich A and Shekhar MP. (2003). Mol. Cell Biol., 23,
2463–2475.
Michael D and Oren M. (2002). Curr. Opin. Genet. Dev., 12,
53–59.
Momand J, Zambetti GP, Olson DC, George D and Levine
AJ. (1992). Cell, 69, 1237–1245.
Montes de Oca LR, Wagner DS and Lozano G. (1995).
Nature, 378, 203–206.
Olivier A, Lucie K, Jean-Francois R, Christian-Jacques L and
Paule S. (2003). Oncogene, 22, 1945–1954.
Palmero I, Pantoja C and Serrano M. (1998). Nature, 395,
125–126.
Quelle DE, Zindy F, Ashmun RA and Sherr CJ. (1995). Cell,
83, 993–1000.
Raveh T, Droguett G, Horwitz MS, DePinho RA and Kimchi
A. (2001). Nat. Cell Biol., 3, 1–7.
Rogoff HA, Pickering MT, Debatis ME, Jones S and Kowalik
TF. (2002). Mol. Cell Biol., 22, 5308–5318.
Roth J, Dobbelstein M, Freedman DA, Shenk T and Levine
AJ. (1998). EMBO J., 17, 554–564.
Russell JL, Powers JT, Rounbehler RJ, Rogers PM, Conti CJ
and Johnson DG. (2002). Mol. Cell Biol., 22, 1360–1368.
Serrano M, Hannon GJ and Beach D. (1993). Nature, 366,
704–707.
Serrano M, Lin AW, McCurrach ME, Beach D and Lowe SW.
(1997). Cell, 88, 593–602.
Shieh SY, Ikeda M, Taya Y and Prives C. (1997). Cell, 91,
325–334.
Shintani S, Nakahara Y, Mihara M, Ueyama Y and
Matsumura T. (2001). Oral Oncol., 37, 498–504.
Stott FJ, Bates S, James MC, McConnell BB, Starborg M,
Brookes S, Palmero I, Ryan K, Hara E, Vousden KH and
Peters G. (1998). EMBO J., 17, 5001–5014.
Sugimoto M, Kuo ML, Roussel MF and Sherr CJ. (2003).
Mol. Cell, 11, 415–424.
Tang D, Chun AC, Zhang M and Wang JH. (1997). J. Biol.
Chem., 272, 12318–12327.
Tang D, Wu D, Hirao A, Lahti JM, Liu L, Mazza B, Kidd VJ,
Mak TW and Ingram AJ. (2002). J. Biol. Chem., 277,
12710–12717.
Tibbetts RS, Cortez D, Brumbaugh KM, Scully R, Livingston
D, Elledge SJ and Abraham RT. (2000). Genes Dev., 14,
2989–3002.
Vafa O, Wade M, Kern S, Beeche M, Pandita TK, Hampton
GM and Wahl GM. (2002). Mol. Cell., 9, 1031–1044.
Watanabe T, Nakamura M, Yonekawa Y, Kleihues P
and Ohgaki H. (2001). Acta Neuropathol (Berl)., 101,
185–189.
Weber JD, Taylor LJ, Roussel MF, Sherr CJ and Bar-Sagi D.
(1999). Nat. Cell Biol., 1, 20–26.
ATM contributes to the functions of p14
ARF
YLiet al
7364
Oncogene
Wu D, Ingram AJ, Lahti JH, Mazza B, Grenet J, Kapoor A,
Liu L, Kidd VJ and Tang D. (2002). J. Biol. Chem., 277,
12001–12008.
Yin Y, Stephen CW, Luciani MG and Fahraeus R. (2002).
Nat. Cell Biol., 4, 462–467.
Zhang Y and Xiong Y. (1999). Mol. Cell, 3, 579–591.
Zhou BB and Elledge SJ. (2000). Nature, 408, 433–439.
Zindy F, Eischen CM, Randle DH, Kamijo T, Cleveland JL,
Sherr CJ and Roussel MF. (1998). Genes Dev., 12,
2424–2433.
ATM contributes to the functions of p14
ARF
YLiet al
7365
Oncogene
