|
|
|
|
 p53 in Tumor Suppression and Aging:
The primary goal of this
laboratory is to understand
the molecular mechanisms by
which p53 mediates tumor
suppression, and to elucidate
its role in cell growth
arrest, apoptosis and cell
ageing. In particular, we are
seeking to elucidate the
multiple regulatory pathways
that control activation,
stabilization and subcellular
localization of p53 in
response to various types of
cellular stress.
The p53 protein is often referred
to as the “guardian of the genome”
because of its crucial role in
coordinating cellular responses to
the genotoxic stresses. The anti-proliferative
effects of p53 are imparted
through a variety of mechanisms
that include cell cycle arrest,
apoptosis, and cellular
senescence. p53 can also be
thought of as the critical node in
a regulatory circuitry where
countless signaling pathways
converge from everyday cellular
processes (e.g.,
growth factor stimulation) to
DNA damage responses (e.g., ATM/ATR
activation) and abnormal oncogenic
events (e.g., Myc or Ras
activation). Mutations of the p53
gene have been documented in more
than half of all human tumors.
Moreover, other defects in the p53
pathway are commonly observed in
tumor cells that retain wild-type
p53, suggesting that inactivation
of p53 function may indeed
constitute an obligatory step in
cancer development (Brooks and Gu,
2003). One key molecular function
of p53 that is required for tumor
suppression is its ability to act
as a transcriptional factor that
binds and regulates the expression
of specific target genes. A number
of these genes are critically
involved in either cell growth
arrest or apoptosis, including p21CIP1/WAF1,
Mdm2, GADD45, Cyclin G, 14-3-3s,
Noxa, p53AIP1, and PUMA. The p53
polypeptide is a short-lived
protein whose abundance and
transcriptional activity are
maintained at low levels in normal
cells. In many physiological
settings, this tight regulation of
p53 function is essential for
maintaining normal cell growth.
However, in cells that sustain
various types of stress, the
steady-state levels and
transcriptional activity of p53
are dramatically increased. While
the precise mechanisms by which
p53 is activated are not
completely understood, they are
generally thought to involve
post-translational modifications
of p53, including ubiquitination,
phosphorylation and acetylation
(reviewed in Brooks and Gu, 2003).
Project I: Identification
and elucidation of the p53
acetylation/deacetylation
pathway
The critical role of
acetylation in p53 regulation.
Our early work
demonstrated that CBP/p300, a
histone acetyltransferase
(HAT), acts as a coactivator
of p53 that potentiates its
transcriptional activity and
in vivo functions (Gu
et al., 1997). At the time it
was thought that the enzymatic
activity of protein
acetyltransferases such as CBP/p300
is restricted to histone
acetylation. Nevertheless,
p300 mutations had been found
in several types of tumors,
and germline mutations of CBP
were known to impart a high
cancer risk to patients with
Rubeinstein-Taybi syndrome and
in CBP knockout mice (reviewed
by Goodman and Smolik, 2000).
The functional synergism
between p53 and
CBP/p300
and the apparent involvement
of CBP/p300 in tumorigenesis
led us to test whether p53
serves as a direct enzymatic
substrate of CBP/p300. By
showing that CBP/p300
specifically acetylates p53 at
multiple lysine residues of
its C-terminal regulatory
domain (Lys 370, 371, 372,
381, 382), we provided the
first example of non-histone
protein acetylation. This
finding predicted that
acetylation is in fact a
general protein modification (Gu
and Roeder, 1997), a notion
that has now been verified for
many other important
regulatory factors (reviewed
by Kouzarides, 2000). Further
studies revealed that
acetylation dramatically
stimulates the
sequence-specific DNA-binding
activity of p53, possibly as a
result of an acetylation-induced
conformational change (Gu and
Roeder, 1997). In
addition, CBP/p300-mediated
p53 acetylation was confirmed
in vivo using
antibodies that specifically
recognize acetylated forms of
p53 (Luo et al., 2000, 2003),
and p53 was subsequently found
to be acetylated at Lys320 by
PCAF, another HAT cofactor.
Significantly, the
steady-state levels of
acetylated p53 are stimulated
in response to various types
of cellular stress, indicating
an important role for p53
acetylation in the stress
response.
Mouse thymocytes and ES cells
that express a
transcriptionally defective
p53 mutant (p53Q25S26)
fail to undergo DNA
damage-induced apoptosis.
Interestingly, this mutant
protein is phosphorylated
normally at the N-terminus,
but not acetylated at its
C-terminus, in response to DNA
damage, suggesting that
acetylation is critical for
both p53 transactivation and
p53-dependent apoptosis (Luo
et al., 2000). Furthermore,
both oncogenic Ras and PML can
upregulate the levels of
acetylated p53 in primary
fibroblasts and induce
premature senescence in a
p53-dependent manner. In
addition, we have found that
the acetylation sites of p53
are essential for its
ubiquitination and subsequent
degradation by Mdm2 (Li et
al., 2002). These sites may
also influence the
interactions between p53 and
transcriptional co-activators
such as CBP/p300 and PCAF,
especially since p53
acetylation is required for
efficient recruitment of these
co-activators to the promoter
regions of p53 target genes
and for their transcriptional
activation. Thus, CBP/p300-dependent
acetylation of p53 has now
been implicated in several
p53-mediated processes,
including transcriptional
activation, apoptosis, and
senescence.
Functional consequences of
p53 deacetylation. In
contrast, much less is known
about the role of
deacetylation in modulating
p53 function. Under normal
conditions, the proportion of
acetylated p53 in cells
remains low. This may reflect
a strong constitutive action
of in vivo deacetylases.
Indeed, the acetylation level
of p53 is enhanced when cells
are treated with histone
deacetylase (HDAC) inhibitors
such as Trichostatin A (TSA).
These observations led us to
characterize the HDAC1 complex
that is responsible for p53
deacetylation, and in the
course of these studies we
found that PID/MTA2, a
component of this complex,
acts as an adaptor protein to
enhance HDAC1-mediated p53
deacetylation
(Luo et
al., 2000). Curiously,
though, the deacetylation
activity of p53 is not
completely repressed by TSA in
cells (Luo et al., 2001).
Therefore, we considered a
possible role for the silent
information regulator 2 (Sir2)
proteins, a novel family of
histone deacetylases that
require nicotinamide adenine
dinucleotide (NAD) as an
essential cofactor and are not
affected by TSA. Previous
studies had shown that the
deacetylase activity of these
proteins is essential for
Sir2-mediated gene silencing
and life-span extension in
yeast and C. elegans.
We found that a mammalian
ortholog of Sir2 (Sir2a)
physically interacts with p53
and attenuates p53-mediated
functions (Luo et al., 2001).
Moreover, nicotinamide
(Vitamin B3) inhibits NAD-dependent
p53 deacetylation induced by
Sir2a, and enhances p53
acetylation levels in vivo.
We also observed that Sir2a
represses p53-dependent
apoptosis in response to
either DNA damage or oxidative
stress. Thus, our studies
identified a novel regulatory
pathway mediated by mammalian
Sir2a that controls p53
function and showed that
maximal acetylation of p53, a
critical requirement for p53
activation in response to
cellular stress, requires the
inhibition of two distinct
types of deacetylase enzymes (Luo
et al., 2000, 2001). By
revealing an important role
for Sir2a in modulating the
sensitivity of cells to
p53-dependent apoptosis, these
results have potential
implications for cancer
therapy. Interestingly,
inhibitors of HDAC-mediated
deacetylases, including sodium
butyrate, TSA, SAHA and
others, have already been
proposed as antitumor drugs.
Thus, we propose that a
combination of genotoxic drugs
with both HDAC and Sir2a
deacetylase inhibitors may
have synergistic effects in
cancer therapy, at least for
tumors that retain wild-type
p53.
Figure 1.
Control of p53
transcriptional activity.
DNA damage induces
phosphorylation, acetylation,
tetramerization, and
stabilization of p53
polypeptides, which in turn
lead to an increase in the
transactivation potential of
p53. Transcriptional
activation of p53 can induce a
variety of phenotypic
responses (e.g., growth
arrest, cellular senescence,
apoptosis) depending on the
cell type and nature of the
cellular stress. Deacetylation
of p53 by Sir2 and/or PID/HDAC1
may be especially important
for downregulation of
p53-dependent transcription
once DNA-damage response is
complete. A, acetylation; P,
phosphorylation; TFs,
transcription factors; TSA,
trichostatin A.
The physiological rationale
for p53 deacetylation remains
an open question. One
possibility is that
deacetylation provides a
mechanism to halt p53 function
rapidly once DNA repair is
completed or transcriptional
activation of it target genes
is no longer necessary (Figure
1). Also, since the major
acetylation and ubiquitination
sites of p53 are overlapping,
deacetylation may also be an
important step in
MDM2-mediated p53 degradation
(Li et al., 2002). The
involvement of the Sir2a
family of HDACs suggests an
interesting link between
nicotinamide (vitamin B3),
cellular metabolism, and
p53-mediated cellular
responses to genotoxic stress
(Luo et al., 2001). Indeed,
since transgenic mice
harboring an N-terminal p53
deletion exhibit an
early-ageing phenotype and Sir
proteins have been implicated
in gene silencing and lifespan
extension in both yeast and C.
elegans, Sir2a may
provide an interesting link
between p53 and cellular
longevity. In the future we
will address several
outstanding questions in this
area: (1) How does acetylation
control p53-mediated function?
(e.g., can we identify
acetylation-dependent in
vivo targets of p53
transactivation, and/or
acetylation-mediated protein
interactions involving p53?);
(2) Are there functional
interconnections between the
p53 acetylase/deacetylase
complexes and the DNA repair
machinery? and (3) What is the
role of mammalian Sir2 in cell
senescence and lifespan
extension?
Project II: Elucidating
the p53 ubiquitination pathway
Ubiquitination regulates a
diverse spectrum of cellular
processes by providing a
specific signal for
intracellular protein
degradation. Although p53
ubiquitination was first
discovered in papillomavirus-infected
cells as a process mediated by
the viral E6 protein, in
normal cells, Mdm2 plays a
major role in the
ubiquitination and subsequent
degradation of p53. The
oncoprotein Mdm2 is a
ubiquitin E3 ligase that
physically interacts with the
N-terminus of p53 and
catalyzes p53 ubiquitination
in vivo. As such, Mdm2
counteracts the tumor
suppressor activity of p53 by
inducing nuclear export and
degradation of ubiquitinated
p53 polypeptides.
Interestingly, Mdm2 gene
transcription is activated by
p53, setting up an
autoregulatory loop in which
the increased Mdm2 production
that occurs during a cellular
stress response ultimately
leads to downregulation of p53
activity. 5 to 10% of human
tumors overexpress Mdm2 due to
gene amplification or
increased transcription and
translation. For example, the
Mdm2 gene is amplified in 30%
of osteosarcomas and 20% of
all soft tissue tumors. The
critical role of Mdm2 in
degrading p53 is best
illustrated by studies carried
out in mice where inactivation
of p53 was shown to completely
rescue the embryonic lethality
caused by loss of Mdm2
function.
Figure 2. A
model for regulation of p53
stability by HAUSP in the
cells subjected to DNA damage.
In unstressed cells, p53 is
ubiquitinated by Mdm2 and
consequently degraded by the
26S proteasome. Some activated
oncoproteins (e.g., E1a, c-Myc)
can stabilize p53 by inducing
p14ARF, a tumor-suppressor
that inhibits mdm2-meditated
ubiquitination of p53. In
response to DNA damage, two
complementary mechanisms of
promote p53 activation: On one
hand, the ubiquitin hydrolase
activity of HAUSP effectively
reverses Mdm2-mediated p53
ubiquitination. On the other
hand, stress-induced Mdm2
autoubiquitination induces
degradation of existing Mdm2
polypeptides, forestalling
further Mdm2-mediated p53
ubiquitination.
Stabilization of p53 in response
to DNA damage and other types of
stress is critical for its tumor
suppressor function. However,
the precise mechanism by which
p53 is stabilized is not
completely understood. In
response to DNA damage, p53 is
phosphorylated at multiple sites
(mainly Ser 15 or Ser 20); these
phosphorylation events promote
p53 stabilization by preventing
Mdm2 binding and thereby
rendering p53 more resistant to
Mdm2-mediated degradation.
Interestingly, several groups
have reported that mutations of
these phosphorylation sites do
not significantly inhibit the
ability of DNA damage to
stabilize p53. Furthermore, some
genotoxic drugs, such as
actinomycin D, can stabilize p53
without provoking either Ser 15
or Ser 20 phosphorylation. These
studies imply the existence of
an alternative mechanism for p53
stabilization that may function
even when the Mdm2-mediated
ubiquitination pathway is
intact. By mass spectrometry of
affinity-purified p53-associated
factors, we recently identified
HAUSP (herpesvirus-associated
ubiquitin-specific protease) as
a novel p53-interacting protein
(Li et al., 2002). HAUSP
strongly stabilizes p53–even in
the presence of excess Mdm2–and
induces p53-dependent cell
growth repression and apoptosis.
Significantly, HAUSP has an
intrinsic enzymatic activity
that specifically
deubiquitinates p53 both
in vitro and in vivo.
Indeed, expression of
catalytically inactive HAUSP
increases the cellular levels of
p53 ubiquitination and
attenuates p53 stabilization
induced by DNA damage. Our
findings revealed a novel
mechanism by which p53 can be
stabilized by direct
deubiquitination and showed that
the balance between the
Mdm2-mediated ubiquitination and
HAUSP-mediated deubiquitination
is critical for p53
stabilization (Li et al., 2002).
Protein
deubiquitination is now
recognized as an important
regulatory event that controls
a broad range of cellular
processes. At least four
distinct families of ubiquitin
hydrolases have been
identified, the largest of
which, the UBP proteins,
already consists of more than
80 known members. Despite the
pervasive influence of these
proteins on cell biology,
deubiquitination is still a
poorly understood process that
has not been adequately
characterized. Indeed, HAUSP
represents the first example
of a mammalian protein that
deubiquitinates a specific
cellular factor.
Interestingly, the HAUSP-p53
interaction is significantly
enhanced in cells subjected to
DNA damage (Li et al., 2002).
Thus, the HAUSP-mediated
pathway may be a critical
mechanism for stabilizing p53
during the stress response
(Fig. 2). To address this and
other issues regarding the
relationship between HAUSP and
p53, we will investigate (1)
how the p53-HAUSP interaction
is induced by DNA damage, (2)
whether HAUSP-mediated
regulation of p53 is altered
during tumorigenesis, and (3)
the physiological role of
HAUSP in animal development
(by targeted mutagenesis).
Project III:
Identification of Parc as a
key regulator of p53
subcellular localization
Nuclear localization of p53
is essential for its tumor
suppressor function.
Although post-translation
modifications and
stabilization of p53 are well
accepted as key events in the
p53-mediated stress response
(reviewed in Brooks and Gu,
2003), subcellular
localization also appears to
play a critical role in the
regulation of p53 function.
While p53 is diffusely
distributed in normal
unstressed cells, in response
to DNA damage and other types
of stress, p53 translocates to
the nucleus where it activates
endogenous target genes. Thus,
nuclear localization of p53 is
essential for its function as
a transcription factor in the
stress response. Indeed, in
many tumor types, including
inflammatory breast
carcinomas, undifferentiated
neuroblastomas, colorectal
carcinomas and retinoblastomas,
wild-type p53 is functionally
inactivated by abnormal
cytoplasmic sequestration. In
these cases, constitutive
cytoplasmic localization of
p53 has been linked with tumor
metastasis, poor response to
chemotherapy, and poor
long-term patient survival (Nikolaev
and Gu, 2003). A number of
proteins have been proposed to
serve as cytoplasmic anchor
proteins that block nuclear
localization of p53, including
ribosomal proteins, Hsc70,
tubulin, F-actin and others.
As each of these proteins is
highly abundant, and the
specificity of their
interactions with p53 needs
further verification, none has
received wide acceptation as
the bona fide cytoplasmic
anchor of p53.
The Parcmediated pathway
regulates subcellular
localization of p53. To
reevaluate the hypothesis that
cytoplasmic factors regulate
p53 localization, we purified
and characterized
p53-containing protein
complexes from the cytoplasm
of unstressed cells (Nikolaev
et al, 2003). Our study
revealed the existence of a
novel cytoplasmic protein (Parc)
that controls the subcellular
localization and function of
p53. In particular, we found
that (i) p53 interacts
directly with Parc to form a
large protein complex (~ 1Mda)
in the cytoplasm of unstressed
cells, (ii) Parc contains
amino acid motifs (e.g., the
Ring-IBR-Ring and Cullin
homology domain) common in
proteins that function in the
ubiquitin system, (iii) Parc
has an intrinsic ubiquitin
ligase activity but fails to
induce p53 degradation in
vivo, (iv) inactivation of
endogenous Parc expression
leads to p53 nuclear
localization and p53-dependent
apoptosis while Parc
overexpression induces
cytoplasmic sequestration of
p53, and (v) RNAi-mediated
reduction of Parc expression
sensitizes neuroblastoma cells
to DNA damage (Nikolaev et al,
2003). Therefore, we propose
that Parc serves as an anchor
protein that tethers p53 in
the cytoplasm and thereby
regulates p53 subcellular
localization (see Figure 3).
In addition, since Parc is
highly expressed in
neuroblastoma cells that
feature both abnormal
cytoplasmic retention of p53
and an impaired p53-dependent
DNA-damage response, these
findings offer a promising
therapeutic strategy for these
types of tumors.
Figure 3. Parc controls the
subcellular localization tumor
suppressor function. In
response to DNA damage or other
types of stress, p53
translocates to the nucleus
where it activates endogenous
target genes. While this
presumably requires
disassociation of the p53-Parc
complex, the mechanisms by which
the p53-Parc interaction is
regulated in response to stress
have not been determined. A,
acetylation; P, phosphorylation;
TFs, transcription factors.
Parc is a constitutive
cytoplasmic protein that
strongly interacts with the
C-terminal domain of p53.
Since this domain harbors the
three known nuclear
localization signals (NLSs) of
p53, it is conceivable that
Parc blocks nuclear import of
p53 by concealing its NLS
motifs. Interestingly,
overexpression of the
p53-binding domain of Parc
(residues 1-770) alone is not
sufficient to induce
cytoplasmic retention of p53 (Nikolaev
et al., 2003). Thus, it is
likely that additional Parc
sequences facilitate
cytoplasmic sequestration of
p53, perhaps by linking the
Parc protein to stable
cytoplasmic complexes/or
structures. In any case,
latent p53 is tightly
associated with Parc in the
cytoplasm of unstressed
cells. However, in response
to DNA damage and other types
of stress, p53 is rapidly
stabilized and translocated
into the nucleus (see Figure
3). As such, regulation of the
Parc-p53 interaction in
response to stress is an
extremely important issue that
warrants further study. Since
p53 is subjected to
post-translational
modifications in stressed
cells, it is possible that
phosphorylation and/or
acetylation of the p53 protein
may regulate its interaction
with Parc. To address these
issues we will investigate (1)
the regulation of the Parc-p53
interaction in response to
stress and (2) the
physiological role of Parc
during tumorigenesis and in
normal animal development.
Selected Publications
1. Luo, J,
Su, F., Chen, D., Shiloh, A.,
and Gu, W. (2000)
Deacetylation of p53 modulates
its effect on cell growth and
apoptosis. Nature
408, 377-381.
2. Guo, A.,
Salomoni, P., Luo, J., Shih, A.,
Zhong, S., Gu, W., and
Pandolfi, P.P. (2000) Role of
PML in p53-dependent apoptosis.
Nature Cell Biol.
2, 730-736.
3. Luo,
J., Nikolaev, A. Y., Imai, S.,
Chen, D., Su, F., Shiloh, A.,
Guarente, L., and Gu, W.
(2001) Negative control of p53
by Sir2a promotes cell survival
under stress. Cell.
107, 137-148.
4.
Bereschenko, O. R., Gu, W.,
and Dalla-Favera, R. (2002)
Acetylation inactivates the
transcriptional repressor BCL-6.
Nature Genetics,
32, 606-613.
5. Brooks,
C. L., and Gu, W. (2003)
Ubiquitination, phosphorylation
and acetylation: the molecular
basis for p53 regulation.
Curr. Opin. Cell Biol.
15, 164-171
6. Li, M.,
Chen, D., Shiloh, A., Luo, J.,
Nikolaev, A. Y., Qin, J., and
Gu, W. (2002)
Deubiquitination of p53 by HAUSP
constitutes an important pathway
for p53 stabilization.
Nature 416, 648-653
7. Hu, M.,
Li, P., Li, M., Li., W., Yao,
T., Wu, J-W., Gu, W.,
Cohen, R. E., and Shi, Y. (2002)
Crystal structure of a UBP-family
deubiquiting enzyme: in
isolation and in complex with
ubiquitin aldehyde Cell
111, 1041-1054.
8. Nikolaev,
A. Y., Li, M., Puskas, N., Qin,
J., and Gu, W. (2003)
Parc: a cytoplasmic anchor for
p53. Cell
112, 29-40.
9. Li, M.,
Chris, B., Wu-Baer, F., Chen,
D., Bear, R., and Gu, W.
(2003) Mono- vs.
polyubiquitination: differential
Control of p53 fate by MDM2.
Science
(in press).
|
|
|
