Bdour Alhajjaj
Southern Illinois University
2ed
Review
To begin, p53 gene is known to be the cancer repressor gene, and the gene that
regulate cell cycle and cause cell death “ apoptosis”. To clarify, once the DNA is
damaged, this gene prevents the cell from dividing. The mutation on this protein can
result of number of cancers. It is very important mechanisms for cells in living organisms
to have the ability of destroying the cell than allowing cancer to occur. The normal
human body has to have a specific number of chromosomes and these chromosomes go
through mitosis and meiosis. Genes encode each chromosome. These genes are the
inheritance unit in living organisms. Each person should have 46 chromosomes 23 are
from the father and the other half is from the mother; otherwise, many different
syndromes will be developed. It was found that p53 is located on chromosome 17. The
p53 gene is regulating the cell cycle. It has the capacity that it blocks the cycle of the cell
and promotes the cell death “ apoptosis”, thus prevent cancer from devolving.
Surprisingly, stress signals as a result of hypoxia activate p53 for example. However, a
new study showed that having any type of stress signals is not the only requirement for
activating p53. However, it is believed that each stress signal has its own role as a tumor
repressor.
The concept of p53 as a repressor gene and its capacity to reverse the cell cycle and
stop the tumor from growing is simplified. Some genes would obey the p53 function and
stop while others will grow. If the case is fully understood, cancer will no longer exist.
Some cases require further manipulations of p53 gene activation in order to cause the
death of certain cells.
Leading Edge
Review
Blinded by the Light:
The Growing Complexity of p53
Karen H. Vousden1,* and Carol Prives2,*
The Beatson Institute for Cancer Research, Garscube Estate, Switchback Road, Glasgow G61 1BD, UK
Department of Biological Sciences, Columbia University, 1212 Amsterdam Avenue, New York, NY 10027, USA
*Correspondence: k.vousden@beatson.gla.ac.uk (K.H.V.), clp3@columbia.edu (C.P.)
DOI 10.1016/j.cell.2009.04.037
1
2
While the tumor suppressor functions of p53 have long been recognized, the contribution of p53 to
numerous other aspects of disease and normal life is only now being appreciated. This burgeoning
range of responses to p53 is reflected by an increasing variety of mechanisms through which p53
can function, although the ability to activate transcription remains key to p53’s modus operandi.
Control of p53’s transcriptional activity is crucial for determining which p53 response is activated,
a decision we must understand if we are to exploit efficiently the next generation of drugs that
selectively activate or inhibit p53.
Introduction
If genius is the ability to reduce the complicated to the simple,
then the study of p53 makes fools of us all. Beyond the indisputable importance of p53 as a tumor suppressor, an increasing and sometimes bewildering number of new roles for p53
have recently been reported, including the ability to regulate
metabolism, fecundity, and various aspects of differentiation and development. We are beginning to develop an intimate understanding of at least some of p53’s functions and
the mechanisms by which p53 is regulated. It is very clear,
for example, that p53 is a transcription factor and regulates
the expression of an array of different genes (encoding both
proteins and microRNAs) that then mediate the p53 response.
However, despite this intensive effort, there is still much to
learn, and p53 remains a highly dynamic and rapidly expanding area of study. It is impossible to cover all aspects of p53associated biology in one review, and so we have reluctantly
passed over many fascinating topics that include the mechanisms that signal to p53; the relationships between p53 and
its family members p63 and p73; the regulation of p53 expression, turnover, and localization; the effects of polymorphisms
in components of the p53 pathways; the expression of different p53 isoforms; and the consequences of p53 mutations that
occur during cancer development. Instead, we have chosen a
few emerging themes to give a flavor of some recent advances
in the p53 world.
p53 and Tumor Suppression: All that Is Good
The ability of p53 to efficiently inhibit cell proliferation—by
both blocking cell cycle progression and promoting apoptotic
cell death—provides a clear mechanism to stem tumor cell
growth and so inhibit cancer development. Activation of p53
is driven by a wide variety of stress signals that a cell might
encounter during malignant progression—genotoxic damage, oncogene activation, loss of normal cell contacts, and
hypoxia to name but a few—leading to a model in which the
growth inhibitory functions of p53 are normally held dormant,
to be unleashed only in nascent cancer cells. But the situation is much more complex. We now understand that some
p53 functions do not require activation by acute stress and
that p53 can promote what appear to be entirely contradictory outcomes, although each of them may have a critical role
to play in tumor suppression.
Death: The Final Frontier?
The concept that p53 can kill cancer cells is made even more
pleasing by the idea that p53 might selectively induce apoptosis in developing tumor cells, while driving only a reversible
cell-cycle arrest in their normal counterparts. As we will see,
this is a massive oversimplification of the complex and heterogeneous responses to p53 activation, where some types
of normal cells die while some types of tumor cells survive.
Nevertheless, the possibility that the response to p53 activation can be modulated and that the therapeutic activation of
p53 might be manipulated to promote death more efficiently
in tumor cells than in normal cells is very attractive. Numerous studies have sought to reveal the molecular mechanisms
that underlie the control of the response to p53 activation. But
before we discuss them, let us take a step back and reconsider
the real contribution of p53-induced apoptosis to tumor suppression.
Many of our models for p53 function suppose that induction of programmed cell death is the key mechanism by which
p53 eliminates cancer cells. Indeed, mice that express a p53
mutant protein lacking the ability to induce cell cycle arrest
but retaining apoptosis-inducing functions are still efficiently
protected from spontaneous tumor development (Toledo et
al., 2006). However, a growing body of evidence indicates that
other functions of p53 may be equally important to prevent or
stall cancer development. A clear hint that apoptosis is not
the only weapon in p53’s tumor suppressive arsenal comes
from the identification of PUMA (p53-upregulated modulator of
apoptosis) as a key mediator of p53’s apoptotic activity. PUMA
is a BH3 (Bcl-2 homology domain 3)-only protein that induces
apoptosis through the mitochondrial pathway. The study of
Cell 137, May 1, 2009 ©2009 Elsevier Inc.
413
Figure 1. p53 Responses in Mediating
Tumor Suppression
The control of cell survival, proliferation, and death
by p53 is mediated by the regulation of expression of p53 target genes (some examples shown in
blue) in the nucleus and transcriptionally independent cytoplasmic functions of p53. Most of these
p53 responses have the potential to contribute to
tumor suppression.
PUMA null mice showed that the induction of PUMA by p53 is
necessary for the apoptotic response to p53 activation in many
tissues (Yu and Zhang, 2003). However, PUMA null mice are not
prone to developing cancer (Michalak et al., 2008), although
subsequent studies have shown that the loss of PUMA can
promote tumorigenesis that is driven by the Myc oncogene
(Garrison et al., 2008; Hemann et al., 2004). It therefore seems
clear that p53 can retain tumor suppressive functions even
in the absence of a robust apoptotic response. The analysis
of an unusual mutant p53 protein led to similar conclusions.
Whereas most cancer-associated p53 mutations destroy all
tested activities of p53, a few tumors harbor mutations in p53
that allow the protein to retain its cell cycle arrest function but
selectively lose its ability to induce apoptosis (Rowan et al.,
1996). The generation of mice expressing one such mutant (a
single amino acid substitution of proline for arginine at residue
172 in the mouse—the equivalent of residue 175 in the human
protein) revealed a very interesting phenotype. Despite being
completely deficient in p53-driven apoptosis, these mice are
still reasonably well protected from tumor development (Liu et
al., 2004). Clearly, other functions retained by this mutant p53
protein can, at least partially, impede tumor development.
Not Dying, but Stopping
So what else might p53 be doing to prevent cancer development? There are several interesting options, including a number of different antiangiogenic activities of p53 that could limit
tumor progression (Teodoro et al., 2007). But maybe the most
obvious candidate for another tumor suppressor activity of p53
is the ability to inhibit cell proliferation and growth (Figure 1).
p53 can effectively block cell cycle progression by activating
the transcription of the cyclin-dependent kinase inhibitor p21,
although several other p53-target genes such as 14-3-3 sigma
and GADD45 also contribute to this response (El-Deiry, 1998).
The induction of p21 expression is extremely sensitive to even
low levels of p53 protein, leading to the idea that a temporary
G1 block, as induced by mild damage or stress, allows cells to
survive safely until the damage has been resolved or the stress
removed (we will come back to this idea). However, a transient cell
cycle arrest might be risky, if a cell with oncogenic potential that
414 Cell 137, May 1, 2009 ©2009 Elsevier Inc.
cannot be repaired is allowed to endure
and ultimately resume proliferation. So
how can cancer cells be permanently
restrained, if not through elimination by
apoptotic cell death? The answer seems
to lie in the activation of senescence—
an irreversible cell cycle arrest. A slew of
fascinating studies have highlighted the
importance of senescence in the inhibition of tumor progression and have identified a key role for p53
in this response. Several of these studies show that a pivotal
point in this pathway is the induction of DNA damage, by oncogene activation or in response to telomere dysfunction, which in
turn leads to the activation of p53 (Deng et al., 2008; Halazonetis
et al., 2008). It would appear that precancerous lesions, which
we probably all carry in abundance, are largely held back from
malignant progression by p53-induced senescence. Small surprise, then, that the loss of p53 has such a dramatic effect in
allowing cancer to develop. Furthermore, senescence remains
an important response to p53 activation, even in established
tumors. In mouse models, reactivation of p53 proves to be a
potently effective therapy for cancer, resulting in the regression
of many different tumor types (Martins et al., 2006; Ventura et al.,
2007; Xue et al., 2007). Most interesting is the type of response
to p53 activitation, which in carcinomas and sarcomas triggers
senescence rather than apoptosis. Although tissue culture studies would suggest that senescence is a cytostatic response that
should promote stabilization of the disease rather than induce
its regression, encouraging results from in vivo studies indicate
that the subsequent engagement of the immune system can
result in tumor clearance (Xue et al., 2007).
As with other p53 responses, senescence is likely to result
from changes in the expression of a number of proteins such
as the plasminogen activator inhibitor 1 (PAI-1) (Kortlever et
al., 2006; Leal et al., 2008). Intriguingly, one of the key mediators of p53-induced senescence is p21 (Brown et al., 1997),
and tumor suppression by the apoptosis-defective p53 R172P
mutant is accompanied by the activation of p21 and senescence (Cosme-Blanco et al., 2007; Van Nguyen et al., 2007).
Furthermore, introduction of this p53 mutant into p21 null mice
results in the complete loss of the cell cycle arrest response
and enhanced tumorigenicity (Barboza et al., 2006). Taken
together, these results suggest that p53-dependent activation
of p21 is an important axis in senescence-dependent tumor
suppression. However, a recent study showed that although
p21 plays an important role in mediating the p53-dependent
cellular response to stress, lack of p21 does not strongly promote tumor development (Choudhury et al., 2007). In some
Figure 2. Dual Mechanisms of p53 Function
in Tumor Suppression and Aging
p53 can help to promote the repair and survival of
damaged cells, or it can promote the permanent
removal of damaged cells though death or senescence. The ultimate result of p53 activation depends on many variables, including the extent of
the stress or damage. In this model, basal p53 activity or that induced by low-stress elicits the protector responses that support cell survival, control
glycolysis, and promote the repair of genotoxic
damage. Sustained stress or irreparable damage,
on the other hand, induces the killer functions of
p53 to activate cell death or senescence. Notably,
the protector functions of p53 could contribute to
tumor development if not properly regulated (red,
dashed arrow). Some of p53’s protector functions
may also help to enhance longevity, whereas the
consequences of p53’s killer functions can promote aging.
ways, this result parallels the failure of PUMA null mice to
spontaneously develop cancer, despite a clear defect in p53dependent apoptosis. Perhaps the senescent and apoptotic
responses act as insurance for each other (a sort of tumor
suppressive belt and braces) with the predominant response
depending on cell type and context. An analysis of the tumor
spectrum of mice lacking both p21 and PUMA would be most
interesting.
Choosing Life or Choosing Death
The tumor suppressive effects of p53-dependent induction
of either senescence or cell death are easy to understand, as
both can emphatically prevent further replication of the incipient cancer cell. In a multicellular organism, this seems to be the
sensible choice—better to lose a few rogue cells to death or
senescence rather than risk retaining one cell that might prove
fatal. Despite the inherent logic of this argument, the cellular
response to p53 is not so straightforward. In addition to eliminating damaged cells, p53 can also contribute to cell survival
through a surprisingly large number of mechanisms (Figure 1).
Numerous p53 target proteins function to inhibit apoptosis,
including p21, decoy death receptors such as DcR1 and DcR2,
the transcription factor SLUG (which represses the expression
of PUMA), and several activators of the AKT/PKB (protein kinase
B) survival pathways (Janicke et al., 2008). Another group of
p53-inducible genes have recently also been shown to act as
antioxidants by decreasing the levels of intracellular reactive
oxygen species (Liu et al., 2008; Sablina et al., 2005). Although
this function for p53 would help inhibit tumor progression by
protecting cells against DNA damage and genome instability,
downregulation of reactive oxygen species through these p53dependent mechanisms can also result in decreased susceptibility to apoptosis (Bensaad et al., 2006).
So why would p53 contribute to cell survival when promoting
death seems to be the safest option? The implication is that the
wholesale elimination of every cell that is exposed to some level
of stress, no matter how mild, is not always the most desirable
response. It is possible that survival promotes senescence,
which is simply another way to permanently remove a problematic cell. Indeed, it may be telling that p21, a key mediator
of p53-induced senescence, also plays a role in cell survival.
However, it also seems likely that under some conditions, this
activity of p53 really does protect cells, allowing them to rejoin
the normal population after the resolution of any damage. p53
engages an entire suite of responses that directly contribute to
DNA repair (Gatz and Wiesmuller, 2006), so it seems only logical to assume that under some circumstances these activities
of p53 will be harnessed for use—and there seems little point
in repairing a cell that is doomed to die or senesce.
This brings us to the interesting question of what determines
the outcome of p53 activation. Whether or not an apoptotic
response is elicited is strongly dependent on the type of tissue, the nature of the stress signal, and the cell’s environment.
But there is also some evidence to suggest that the decision between life and death can be determined by the extent
of damage or the duration of stress (Figure 2). In this model,
low levels of transient stress associated with repairable damage elicit the survival response (where p53 acts as a protector), whereas high levels of sustained stress accompanied by
irreparable damage lead to cell death or senescence (where
p53 acts as a killer) (Bensaad and Vousden, 2007). Despite the
clear difference in outcomes, both of these responses to p53
could contribute to tumor suppression, either by preventing the
accumulation of oncogenic lesions or by eradicating damaged
cells through cell death or senescence. This model also suggests that there are activities of p53 (as a protector) that might
be extremely dangerous to sustain under conditions where
repair or recovery is not possible. We will revisit this point later
in the discussion.
Controlling the Engine
Oncogenic changes that promote cancer cell proliferation and
survival are often accompanied by alterations in cell metabolism that also play a vital role in supporting tumor development
Cell 137, May 1, 2009 ©2009 Elsevier Inc.
415
Figure 3. p53 Contributes to Multiple
Normal Processes and Disease Pathologies
In addition to its well-known role as a tumor suppressor, p53 also regulates other cellular (right)
and developmental processes (left). These include
processes that result in positive outcomes (red
arrow) and those that result in diseases or other
unfavorable outcomes (black arrow). Examples of
p53 target genes (blue) that are regulated by p53
to produce the indicated cellular outcomes of p53
induction are shown. Note that, in many cases, numerous targets have been identified to mediate a
specific outcome, even though only one example
target gene is shown here.
(DeBerardinis et al., 2008). This is a complex subject, but in
essence the reprogramming of metabolic pathways provides
cancer cells with numerous benefits, including the ability to
survive under adverse conditions (such as low or variable oxygen availability), the ability to mobilize anabolic pathways that
generate the macromolecules necessary for growth, and the
ability to limit oxidative damage (DeBerardinis et al., 2008).
Indeed, the dependence of cancers on metabolic transformation is highlighted by mouse models of cancer in which interfering with these altered metabolic programs profoundly limits
tumor cell growth (Bonnet et al., 2007; Christofk et al., 2008;
Fantin et al., 2006). A role for p53 in responding to and regulating metabolic changes is therefore an exciting and burgeoning
area of study.
So how does p53 contribute to the regulation of metabolism,
and how might this help it to function as a tumor suppressor?
Not surprisingly, p53 can be activated by metabolic adversity (such as starvation)—a response that can be mediated
through the action of AMP-activated protein kinase (AMPK),
a key component of the cell’s response to bioenergetic stress
(Jones et al., 2005). p53 then promotes a program of gene
expression (including the induction of AMPK expression)
to negatively regulate the kinase mTOR (mammalian target
of rapamycin), a central node in the control of protein synthesis (Budanov and Karin, 2008; Feng et al., 2007a) (Figure
1). This response to p53 helps to ensure the coordination of
cell growth and cell proliferation, which is also regulated by
p53. A role for p53 in the response to starvation and metabolic stress is also reflected in the ability of p53 to regulate
autophagy, a membrane trafficking-mediated “self-eating”
process that results in lysosomal digestion of cellular components. Autophagy promotes short-term cellular survival
under starvation conditions and also helps to eliminate damaged proteins and organelles. The ability of p53 to promote
autophagy through the induction of lysosomal proteins such
as DRAM (damage-regulated autophagy modulator) (Crighton
et al., 2006) or through negative regulation of mTOR signaling
is certainly consistent with an observed role for autophagy in
tumor suppression (Matthew et al., 2007) (Figure 1). But the
relationship between p53 and autophagy remains unclear—
indeed basal levels of cytoplasmic p53 have been shown to
416 Cell 137, May 1, 2009 ©2009 Elsevier Inc.
inhibit autophagy, also by regulating mTOR activity (Tasdemir et al., 2008). Similarly complicated are the consequences
of autophagy for tumor suppression, as p53-associated
autophagy has been reported to both contribute to apoptosis
(Crighton et al., 2006) and help tumor cell survival (Amaravadi
et al., 2007). How these opposing responses to p53-induced
autophagy are coordinated remains to be determined.
In addition to regulating growth through modulating mTOR
signaling, p53 may have even more intricate roles in the regulation of metabolic pathways (Jones and Thompson, 2009).
These recently uncovered functions include modulating glucose uptake (Kawauchi et al., 2008), dampening glycolysis
(Bensaad et al., 2006; Kondoh et al., 2005), and enhancing
mitochondrial respiration (Ma et al., 2007). Intriguingly, some
of these effects of p53 could help to curb the acquisition of
enhanced aerobic glycolysis (the Warburg effect), one of the
most common metabolic changes associated with oncogenic
transformation. Thus, they may provide yet another route by
which p53 can restrain tumor development.
p53 and Tumor Promotion: Crossing to the Dark Side
Our recent appreciation of p53’s role in regulating glycolysis,
oxidative stress, and cell survival leads us to a growing tangle of
complexity within the p53 pathway, where p53 can be involved
in disparate and even contradictory responses. These functions of p53 highlight an interesting paradox touched on earlier: if some activities of p53 that normally contribute to tumor
suppression are not properly regulated, they might “switch
sides” to help promote cancer development (Figure 2). Obvious examples include the prosurvival functions of p53, which
might protect cells undergoing repair following mild stress but
would be extremely counterproductive if maintained in irreparably damaged cells. Similarly, the ability of p53 to impede
glycolysis can help control oncogenic transformation, but
the consequent promotion of alternative metabolic pathways
(such as the pentose phosphate pathway) might also drive the
anabolism necessary for tumor cell growth (DeBerardinis et al.,
2008). Nontranscriptional functions of p53 such as its role in
inhibiting autophagy could also contribute to tumor development under some circumstances. Intriguingly, this particular
function of wild-type p53 is retained by cancer-associated
mutant p53 proteins (Morselli et al., 2008), although whether
this activity helps malignant progression is not yet known. It
would appear that the tight restriction of some p53 responses
may be necessary to prevent one of our principal tumor suppressors from turning from friend to foe.
p53 in Other Pathologies
Although the ability of p53 to drive processes such as cell
death is clearly beneficial in the context of limiting tumor development, engaging these p53-mediated responses under all
stress conditions may not necessarily be advantageous. For
instance, the DNA-damage-induced p53 response is largely
responsible for radiation or chemotherapy-induced sickness.
Although this has often been viewed as an unfortunate but
necessary side effect of activating the important genome protection functions of p53, more recent work has shown, rather
surprisingly, that this initial p53 response is not required for
tumor suppression (Christophorou et al., 2006; Efeyan et al.,
2006). Rather, it seems that the ability of p53 to sense and
respond to oncogene activation is the key to preventing cancer, suggesting that transient inhibition of p53 might be useful in protecting normal tissue from the short-term negative
effects of cancer therapies (Berns, 2006). Beyond cancer, the
fact that p53 responds to a myriad different types of stress
without being able to distinguish when this is helpful and when
not results in its involvement in a number of negative effects
(Figure 3). For example, the induction of p53-driven apoptosis
in response to hypoxia is clearly an asset when the hypoxia
is caused by a lack of blood supply in a rapidly growing
tumor. However, this response is much less desirable when
the hypoxia is due to ischemia following stroke or myocardial
infarction. Indeed, inhibition of p53 seems to be extremely
beneficial during the early stages of ischemia or during subsequent reperfusion injury (Liu et al., 2006). The activation of
p53 by ribosomal stress has been linked to tumor suppression, with abnormalities in the expression of several ribosomal
proteins resulting in an increased tumor susceptibility in disorders such as Diamond Blackfan Anemia, possibly reflecting a failure to properly activate p53 (Montanaro et al., 2008).
On the other hand, perturbation of ribosome biosynthesis
by mutations in TCOF1 (Treacher Collins-Franceschetti syndrome 1) can cause constitutive activation of p53 (Jones et al.,
2008) and result in the congenital disorder known as Treacher
Collins syndrome. Similarly, p53 appears to contribute to the
pathology of various neurodegenerative diseases. Activation
of p53 by mutant forms of huntingtin (which are responsible
for Huntington’s disease) partially mediates the neurodegeneration and neurobehavioral abnormalities observed in
mouse models (Bae et al., 2005). In turn, p53 further induces
the expression of huntingtin (Feng et al., 2006), an effect that
might further exacerbate the progress of the disease. In animal models of Parkinson’s disease, the loss of DJ-1 expression (a gene mutated in early onset Parkinson’s disease in
man) leads to the activation of p53 and the death of dopaminergic neurons (Bretaud et al., 2007). p53 has also been implicated as a mediator of neuronal death in Alzheimer’s disease
(Culmsee and Landshamer, 2006), although recent studies
suggest that expression of amyloid-beta peptides associated
with Alzheimer’s may induce a conformational shift in the p53
protein (Lanni et al., 2007). This effect might provide a useful additional marker for the diagnosis of Alzheimer’s and also
presents the intriguing possibility that the unfolding of p53 into
a “mutant” conformation may contribute to the development
of the disease. While these findings highlight the potentially
detrimental effects of p53 activity in the nervous system, the
ability of p53 to promote neural outgrowth and axon regeneration suggest that it can also have a more positive contribution
to neuronal regeneration after central or peripheral nervous
system injuries (Di Giovanni et al., 2006).
Everyday p53 Functions: No Stress, No Worries?
One of the most interesting shifts in our thinking about p53
is the realization that its remit may be far broader than simply to promote a tumor suppressive response to acute stress.
Indeed, the ability to prevent cancer has been suggested to
be an “evolutionarily late” cooption of p53 activities that had
initially evolved to protect the germline and monitor development (Aranda-Anzaldo and Dent, 2007; Vousden and Lane,
2007) (Figure 3). These primordial functions of p53 as a guardian of the germline appear in lower organisms (such as flies
and worms) that have no clear need for cancer suppression
(Derry et al., 2001; Sutcliffe and Brehm, 2004) and are also
reflected in its ability to protect mouse embryonic stem cells
from DNA damage by inducing their differentiation (Lin et al.,
2005). A role for p53 in differentiation and development is also
observed in the frog Xenopus laevis, where p53 engages in
complex interactions with the Smad transcriptional regulators
to direct embryonic germ layer specification (Piccolo, 2008).
p53 even makes a subtle but important contribution to several aspects of normal growth and development in mice, where
the p53 family members p63 and p73 shoulder the bulk of the
developmental functions (Aranda-Anzaldo and Dent, 2007;
Vousden and Lane, 2007).
It is now becoming apparent that the manifestations of some
of p53’s functions in diverse aspects of health and disease
(Figure 3) do not require acute stress. Rather, these p53 functions depend on basal levels of p53 or the activation of p53
by low levels of constitutive stress. One possible function for
basal p53 activity is in the control of stem cell renewal. Even in
the absence of obvious stress, p53 can limit the self-renewal
of adult neural stem cells (Meletis et al., 2006) and regulate
quiescence in hematopoietic stem cells (Liu et al., 2009). p53
also represses the expression of CD44, cell surface proteins
involved in regulating many aspects of cell migration and survival (Godar et al., 2008). This may represent another contribution to tumor suppression, but the observation that basal levels
of p53 can also control CD44 suggests a mechanism by which
p53 could be involved in regulating normal functions of CD44,
such as epithelial development or stem cell renewal. In other
studies, p53 has been shown to contribute to fecundity in mice
by directly regulating the expression of leukemia inhibitory factor (LIF), a protein required for blastocyte implantation (Hu et
al., 2007). Most interestingly, polymorphisms known to affect
p53 activity are associated with implantation failures in women
(Kay et al., 2006), hinting at the conservation of this function of
p53 in humans.
Cell 137, May 1, 2009 ©2009 Elsevier Inc.
417
The ability of basal p53 activity to modulate metabolism
may also have some interesting consequences beyond the
control of cancer development. For example, the ability of p53
to promote aerobic respiration appears to be critical in mice
to maintain endurance during exercise (Matoba et al., 2006).
But maybe the most intriguing additional role for p53 is in the
regulation of longevity and aging (Figure 2). In lower organisms such as worms and flies, the loss of p53 can be associated with increased longevity (Arum and Johnson, 2007;
Bauer et al., 2005). Initial studies suggested that p53 could
also drive premature aging in mammalian systems, although
there is some indication that this effect may be the consequence of inappropriate, unregulated p53 activity—possibly
reflecting enhanced oxidative stress (Matheu et al., 2008). By
contrast, our more recent understanding of the antioxidant
functions of basal (uninduced) levels of p53 and the ability of
p53 to negatively regulate the IGF-1/mTOR growth regulation
pathways suggest an alternative possibility: that p53 activity
enhances longevity. Indeed, mice engineered to express additional copies of normally regulated p53 showed resistance to
cancer without any accelerated aging (Matheu et al., 2007). In
fact, in combination with increased copies of the tumor suppressor Arf, the enhanced p53 allele even enhances longevity
in mice (Matheu et al., 2007). Thus, just as p53 can function
as both protector and killer in its role as a tumor suppressor, it
may also both promote and prevent aging (Figure 2). Interestingly, p53 function has been shown to decline with age (Feng
et al., 2007b). This could contribute not only to the increased
incidence of cancer with increased age but also possibly to
the process of aging itself. Can p53 activity be somehow harnessed to allow us to both avoid cancer and enjoy increased
longevity? This may be a dream, but it is certainly a question
worth examining.
The Nuts and Bolts of p53 Regulation
As outlined above, the consequences of p53 activation can
be dramatically different depending on numerous factors and
contexts. Differences in stimuli, cellular milieu, or external
environment can result in different p53-dependent outcomes.
How are such decisions in cellular outcomes made? Studies performed in past 5 years have provided deeper insights
into this question. Myriad transcriptional targets mediate
the diverse outcomes of p53 activation (Figure 3). Although
most p53 targets are only induced by types of stress that
lead to increased p53 levels, some targets can be activated
by basally expressed p53. p53 also may have a bona fide
transcription-independent, mitochondria-associated role
in inducing apoptosis (Moll et al., 2005; Schuler and Green,
2005). It is possible that both transcription-dependent and
transcription-independent functions of p53 are required for
promoting apoptosis and limiting tumorigenesis. Indeed,
PUMA, a proapoptotic protein encoded by a p53 target gene,
is required to release cytoplasmic p53 from the antiapoptotic
protein Bcl-XL to facilitate mitochondrial outer membrane
permeabilization (Chipuk et al., 2005). Moreover, deletion of
the p53 binding sites in the endogenous PUMA promoter
in human colorectal cancer cells (by homologous recombination) reduces both PUMA expression and DNA damage418 Cell 137, May 1, 2009 ©2009 Elsevier Inc.
induced apoptosis (Wang et al., 2007a). We discuss below
recent developments in the understanding of transcriptional
regulation by p53, focusing on but a few of the many notable
reports that examine how p53 selects its target genes and the
ensuing cellular outcome.
Thinking Globally
Elucidation of p53’s function as a transcriptional regulator
will require the integration of both macroscopic and microscopic views to evaluate the complete set of p53-regulated
genes and to delve into the mechanisms by which it selects
its target genes in different settings. From the macroscopic
vantage point, it will be important to know the full repertoire
of direct transcriptional targets bound and regulated by p53.
One recent review lists 129 such p53 transcriptional targets
that were identified as a result of either single gene discoveries or multigene screens (Riley et al., 2008). There are likely to
be many more genes specifically bound and activated by p53.
Furthermore, the number of genes whose expression is altered
indirectly upon induction of p53 is likely to be in the thousands.
To find new direct p53 targets, there is well-justified interest
in the identification of sites within the human genome that are
bound by p53. The original consensus site recognized by p53
consisting of two copies of the sequence—RRRCA/TT/AGYYY
(R, purine; Y, pyrimidine)—has been refined by using bioinformatics analysis (Hoh et al., 2002; Miled et al., 2005; Riley
et al., 2008; Sbisa et al., 2007) and experimental approaches
using chromatin immunoprecipitation (chIP) in conjunction with
microarrays (“chIP on chip”) (Cawley et al., 2004; Hearnes et al.,
2005; Smeenk et al., 2008) or chIP-paired-end (PET) sequencing (Wei et al., 2006). Such screens for p53-binding sites have
provided intriguing information regarding p53 target binding.
First, they have revealed that not all “good” p53-binding sites
in the genome are occupied by p53 under the conditions that
were used. Factors such as the spacing between half sites (Jordan et al., 2008), the location of a site within a heterochromatic
locus, or the presence at the site of a p53 dominant-negative
isoform or family member are all likely to be important determinants of p53 association with any particular site. Second, not
all regions bound by p53 have sequences that conform to the
p53 consensus site. This could be the result of p53 association
with other DNA-binding proteins such as nuclear transcription
factor Y (NF-Y) (Imbriano et al., 2005). Third, not all genes in the
vicinity of bound p53 are transcriptionally regulated as a result
of p53 binding. Eukaryotic gene promoters usually require
multiple factors for activation. Moreover, the presence of corepressors at the same promoter could counteract the activity of
p53. It will be important to both refine and extend the current
bioinformatic and experimental approaches to obtain a more
dynamic global view of p53 binding and activation. Although
daunting in terms of effort and cost, it will be crucial to elucidate the extent and kinetics of p53 binding at genomic targets
after different stimuli and in different types of cells.
Surveys of genes whose expression is altered upon induction
of p53 always include large numbers of genes whose expression
is reduced. Indeed, transcriptional repression by p53 is important in promoting cell death. p53 may reduce gene expression
by several mechanisms (Laptenko and Prives, 2006). First, p53
may increase the expression of a protein (such as p21) that pre-
vents phosphorylation of the retinoblastoma protein, thereby
maintaining genes regulated by the E2F transcription factors
in a repressed state. Indeed, the repression of numerous p53
target genes has been shown to be mediated by p21 (Lohr et
al., 2003; Tang et al., 2004; Shats et al., 2004; Baptiste-Okoh et
al., 2008b). Second, p53 transcriptional repression may result
from the direct association of p53 with promoters that possess
binding sites for other transcription factors such as Sp1 (Esteve
et al., 2007; Innocente and Lee, 2005; Sengupta et al., 2005;
Zaky et al., 2008; Zhan et al., 2005), NF-Y (Imbriano et al., 2005;
Matsui et al., 2004), or SMADs (found in combination with a
p53 recognition sequence) (Wilkinson et al., 2008). Promoters
repressed by p53 may also harbor a cell cycle-dependent element/cell cycle gene homology region (CDE/CHR element), a
sequence recognized by several different transcription factors
(Rother et al., 2007; St Clair et al., 2004). Third, gene repression could be mediated by unique p53 “repression” elements
to which p53 binds directly (Godar et al., 2008; Johnson et al.,
2001).
Gene expression microarrays have revealed that p53 regulated genes are not limited to those involved in cell cycle
arrest and apoptosis. Many other gene clusters associated
with diverse processes such as DNA repair, transcription, cell
adhesion, cell mobility, metabolism, and membrane functions
are also affected by p53 activity. The complex repertoire of
p53 regulated genes further highlights the imperative need to
understand how p53 selects its targets.
Acting Locally
Moving from the macroscopic to the microscopic, much attention has been paid to the mechanisms by which p53 selects
some of its key target genes. We will start by discussing a
number of studies that have revealed new facets of how p53
contacts its binding sites in DNA, some of which have also
provided insights into p53-binding site selectivity in vitro and
in cell culture. X-ray crystallographic analyses have revealed
a new interface in a p53 dimer bound to DNA (Ho et al., 2006)
and shown that four p53 core domains bind as a dimer of dimers to two cognate half sites in DNA (Kitayner et al., 2006).
Interestingly, there are DNA-sequence-specific differences in
the contacts made between the p53 protein surfaces, which
could translate into the degree of induction for a given target
gene (Kitayner et al., 2006). Now that Fersht and colleagues
(Tidow et al., 2007) have succeeded in obtaining a structure of
full-length p53 bound to DNA by using a combination of small
angle X-ray scattering and nuclear magnetic resonance, it will
be possible to gain a clearer view of how p53 interacts with different DNA sequences.
How does p53 identify its cognate binding sites in a vast sea
of genomic DNA? Although this question is still unanswered,
two studies have shown that the p53 protein is capable of diffusing two dimensionally on DNA in vitro (McKinney et al., 2004;
Tafvizi et al., 2008). It is not yet known over what distances p53
can slide on DNA, whether p53 can similarly diffuse along DNA
that is wrapped around a histone octamer, and whether the
chromatin state of the DNA regulates how p53 binds or slides.
Experimental and molecular modeling studies have revealed
that the propensity of a p53 cognate binding sequence to
bend has a significant impact on the stability and affinity of
p53 binding (Batta and Kundu, 2007; Pan and Nussinov, 2008).
A fascinating albeit somewhat exotic study has revealed that a
tethered photo-oxidant, anthraquinone, can actually transmit
electrical charge through the DNA to p53 protein bound at a
distance, resulting in the photo-oxidation and specific release
of p53 from some cognate binding sites (for example, sites in
the promoter of the gene encoding Mdm2) but not others (for
example, sites in the p21 promoter) (Augustyn et al., 2007).
These findings may relate to changes in the oxidative state of
p53 after hydrogen peroxide treatment of cells and the ensuing
selective impact on its ability to activate transcription.
Several proteins and small molecules have been shown to
regulate the DNA-binding specificity of p53. Some of these
work through the p53 tetramerization region. For example, a
p53 isoform (p53β), that can form heterotetramers with wildtype p53, stimulates wild-type p53 binding and activation of
the Bax (Bcl-2-associated X protein) gene, but it does not promote p53 activation of p21 expression (Bourdon et al., 2005).
The tyrosine kinase c-Abl, on the other hand, stabilizes p53
tetramerization and augments the binding of p53 to the p21
promoter instead of the Bax promoter (Wei et al., 2005a). Others factors may affect p53 DNA binding by directly interacting
with the central core domain of p53. Among the earliest discovered and still studied of these proteins are the apoptosis
stimulating proteins of p53 (ASPPs), which selectively stimulate
p53 binding and activation of the Bax promoter but not the
p21 promoter (Sullivan and Lu, 2007). In contrast, a zinc-finger
protein called hematopoietic zinc finger (HZF), itself a p53 transcriptional target, interacts with the p53-DNA-binding domain
and promotes the recruitment of p53 to the p21 and 14-3-3
promoters, but not to the Bax promoter or the promoter of the
proapoptotic Noxa gene. These findings are consistent with
the observation that mouse embryonic fibroblasts lacking HZF
exhibit increased Bax expression and decreased p21 expression in comparison to that in wild-type mouse embryonic fibroblasts (Das et al., 2007). Even a small molecule such as nicotinamide adenine dinucleotide (NAD+) can selectively impact p53
binding to DNA in vitro and correspondingly affect the level of
p53-induced Mdm2 expression without impacting p21 expression in vivo (McLure et al., 2004). These are likely only a few of
the ways in which p53 DNA binding and transcriptional activation can be differentially regulated.
Modifying the Regulator
Since the first discoveries revealing that p53 undergoes stressinduced phosphorylation or acetylation, there have been
numerous complicated studies describing these (and other)
modifications to p53 and deciphering how they affect p53
function as a transcriptional regulator. As there are a number
of excellent reviews covering these aspects of p53 regulation
(Appella and Anderson, 2001; Bode and Dong, 2004; Kruse
and Gu, 2008; Olsson et al., 2007), we will focus on discussing
findings relevant to a few key p53 modifications that selectively
impact the cellular outcomes of p53 activation (Figure 4).
The Many Roles of Phosphorylation
Serine 46 (S46), an N-terminal phosphorylation site in
human p53, clearly has discriminatory functions for p53 as
a transcriptional activator. Phosphorylation at this residue is
Cell 137, May 1, 2009 ©2009 Elsevier Inc.
419
Figure 4. Selective Impact of p53
Modifications
p53 protein domains include the transcriptional
activation domain I (TAD 1, residues 20–40), the
transcriptional activation domain II (TAD II, residues 40–60), the proline domain (PP, residues
60–90), the sequence-specific core DNA-binding
domain (DNA-binding core, residues 100–300), the
linker region (L, residues 301–324), the tetramerization domain (Tet, residues 325–356), and the
lysine-rich basic C-terminal domain (++, residues
363–393). A few examples are depicted of residues that when modified by phosphorylation (P),
acetylation (Ac), or ubiquitination (Ub), result in a
specific cellular outcome in response to p53 activation (for example, apoptosis versus cell cycle
arrest) that depends on preferential activation of
the indicated target genes.
correlated with an altered p53 transcriptional program that
includes the induction of p53-regulated apoptosis-inducing
protein 1 (p53AIP1), a proapoptotic factor that promotes the
release of mitochondrial cytochrome c during apoptosis.
One of the protein kinases that phosphorylate this site is
homeodomain interacting protein kinase 2 (HIPK2) (D’Orazi
et al., 2002; Hofmann et al., 2002), a protein regulated by the
tumor suppressor Axin (Rui et al., 2004). Under conditions
of moderate DNA damage, the p53 negative regulator Mdm2
induces HIPK2 degradation (Figure 4). In contrast, severe
DNA damage results in reduced Mdm2 levels, thus allowing
the now stable HIPK2 to phosphorylate p53 at S46 to induce
cell death (Rinaldo et al., 2007). These findings implicate
Mdm2 as a determinant of alternative cell fates that are regulated by p53 (Shmueli and Oren, 2007), a concept that we
will return to later in this Review. The observations also support the previously mentioned view that the extent of cellular
damage may determine whether p53 acts as a survival factor
or a death factor. To make matters more complicated, p53
not only negatively regulates HIPK2 through inducing Mdm2
expression for its degradation, but also positively regulates
HIPK2 by facilitating its caspase-mediated cleavage and
subsequent activation (Gresko et al., 2006). There are other
protein kinases that can either directly phosphorylate S46
or are otherwise required for S46 phosphorylation. These
include dual-specificity tyrosine-phosphorylation-regulated
kinase 2 (DYRK2) (Taira et al., 2007), AMPK (Okoshi et al.,
2008), protein kinase C delta (Yoshida et al., 2006), and p38
mitogen-activated protein kinase (Perfettini et al., 2005).
That several kinases can phosphorylate S46 both supports
the importance of this site in p53 function and makes the
understanding of its regulation more challenging. S58 in the
mouse p53 protein is likely the corresponding residue of S46
in the human protein (Cecchinelli et al., 2006), and it will be
interesting to determine the physiological outcome of mutating this residue in mice.
C-terminal phosphorylation sites in p53 have also been
linked to selective impacts on target gene expression and outcomes. S315 is somewhat unique among these sites in that it
is phosphorylated by growth-promoting kinases. Phosphorylation of S315 regulates the ability of p53 to repress Nanog, a
factor required for stem cell self-renewal, through the recruitment of the transcriptional regulator and corepressor mSin3A.
420 Cell 137, May 1, 2009 ©2009 Elsevier Inc.
Mice harboring mutant p53 where S315 is mutated to alanine
(S315A) are impaired in Nanog repression. However, the interpretation of experiments where S315 is disrupted is complicated by the proximity of S315 to the major nuclear localization
signal sequence of p53. The transcription factor E2F requires
S315 to facilitate nuclear retention of human p53 (Fogal et al.,
2005), whereas the binding of this region by the glycogen synthase kinase-β (GSK-β) after ER stress causes cytoplasmic
retention and destabilization of p53 (Qu et al., 2004). Also within
the C terminus of p53 are S366 and threonine 387 (T387), two
sites that have been shown to be regulated by the checkpoint
kinases Chk1 and Chk2. Downregulation of either Chk kinase
or the mutation of these two p53 phosphorylation sites selectively affects p53 activation, promoter binding, and acetylation
of C-terminal lysines on p53 (Ou et al., 2005). In mice, substitution of S389, a UV-inducible modification, with alanine results
in altered expression of some p53 target genes and generally
reduced repression of p53 targets in UV irradiated cells (Bruins
et al., 2007, 2008). Although there is still much to learn about
how different phosphorylations regulate p53, several sites
clearly have discriminatory impacts on some target genes in
comparison to others.
The Ever-Shifting Functions of p53 Lysines
As complex as the consequences of p53 phosphorylation
are, the roles of p53 lysine residues are even more perplexing. Numerous studies have implicated the lysines within the
extreme C terminus of p53 as being important for the protein’s
transcriptional activities. Yet, confoundingly, two knockin mice
in which either six (Feng et al., 2005) or all seven (Krummel et
al., 2005) of the extreme C-terminal lysines were mutated to
arginines have mild (albeit somewhat different) phenotypes. A
possible explanation for this puzzling result comes from the
identification of two acetylation sites within the p53 central core
domain. The first of these two sites, lysine 120 (K120), is acetylated in vivo in response to DNA damage and increases PUMA
but not Mdm2 expression. Two MYST family histone acetyl
transfereases (HATs)—Tip60 (Tang et al., 2006) and hMOF
(Sykes et al., 2006)—are capable of acetylating K120. Cells
overexpressing a mutant form of p53 where K120 is mutated
to arginine (K120R) show a partial defect in apoptosis. It should
be noted that when the mutant K120A p53 protein is expressed
at physiological levels, the defect in apoptosis is stronger than
when the mutant protein is transiently overexpressed, indicat-
ing that abnormally high levels of p53 can produce misleading
results (Zupnick and Prives, 2006). Furthermore, the loss of one
copy of Tip60 in mice impairs the Myc-induced DNA-damage
response without impacting the p53 transcriptional program,
suggesting that reduced Tip60 levels may not impact p53 in
vivo (Gorrini et al., 2007).
The second core domain acetylation site, K164, is modified
by the transcriptional coactivators p300 and CREB-binding
protein (CBP) and appears to be important for the activation
of the majority of p53 target genes (Tang et al., 2008). When
six of the extreme C-terminal lysines in p53 are mutated in
addition to mutation of K120 and K164, the resulting p53
mutant protein (p53 8KR) is virtually inert. This mutant protein
lacks the transcriptional activation activity required to induce
a plethora of its target genes, including those encoding p21,
PUMA, Bax, and p53-inducible gene 3 (PIG3). The Mdm2
promoter is the one notable exception: Mdm2 expression is
induced by the 8KR p53 protein to a level similar to that seen
with wild-type p53. What is unique about the Mdm2 promoter
and its regulation by p53? In cultured human breast cancer
MCF7 cells, an ATPase component of the SWI/SNF chromatin
remodeling complex called Brg1 is required for p53 binding
and induction of the p21 promoter but not the Mdm2 promoter
(Xu et al., 2007). This suggests that, compared to other p53
targets, the Mdm2 promoter may not be as tightly associated
with nucleosomes. Whether this is relevant to the observation
that unacetylatable p53 can still activate expression of Mdm2
remains to be determined.
Findings implicating p300 and CBP as being critical for
p53’s activities in both arrest and apoptosis are supported
by the observation that the F box protein Skp2 prevents p300
from binding to and acetylating p53 with consequent reduced
expression of p53 targets such as p21 and Puma (Kitagawa et
al., 2008). However, these data will need to be reconciled with
the observation that the targeted deletion of p300 in human
HCT116 colon cancer cells results in reduced p21 expression
but increased Puma expression, with the corresponding cellular outcomes of reduced cell cycle arrest and increased apoptosis (Iyer et al., 2004).
K320 in p53 is a substrate of the transcription coactivator
P300/CBP-associated factor (PCAF). It is another lysine that
plays an interesting role in p53 target gene selection. Cells
ectopically expressing a mutant p53 where K320 was mutated
to glutamine (K320Q; Q is thought to mimic acetylation) display decreased apoptosis after some forms of DNA damage.
Although the K320Q mutant protein is capable of inducing p21
expression, it can neither induce the expression of the gene
encoding apoptotic peptidase-activating factor 1 (APAF1)
nor repress some p53 targets such as the gene encoding
the apoptosis-inhibitor protein survivin (Knights et al., 2006).
Acetylated K320 also preferentially activates two transcriptional targets of p53 (Coronin 1b and Rab13) whose gene products associate with the cellular cytoskeleton and are involved
in neurite outgrowth during axonal regeneration (Di Giovanni et
al., 2006). Treatment of cultured cells with a hypoxia-mimicking
drug (etoposide) leads to increased association of PCAF and
K320-acetylated p53 with the p21 promoter compared to p53
acetylated at K382, despite a global decrease in the amount
of p53 acetylated at K320 (Xenaki et al., 2008). The possibility that, under some conditions, acetylation of K320 predisposes p53 to activate p21 and decreases its ability to induce
proapoptotic targets genes is nicely consistent with the observation that K320R knockin mice harbor several cell types that
display increased apoptosis and higher expression of relevant
p53 target genes (Chao et al., 2006).
Of course, lysine mutations do not exclusively reflect the
loss of p53 acetylation. Other lysine modifications such as
methylation, ubiquitination, sumoylation, and neddylation have
the potential to also alter p53’s transcriptional activity. Experimental results obtained using lysine amino acid substitutions
need to be viewed as only circumstantial and not definitive. A
zinc-finger protein E4F1, first identified as a cellular target of
the Adenoviral E1a protein, ubiquitinates p53 at K320. Interestingly, ubiquitination at K320 does not destabilize p53. Rather,
it selectively facilitates p53 activation of p21 and cyclin G1
expression without affecting the expression of the proapoptotic gene Noxa. This is consistent with the observation that
E4F1 expression markedly reduces UV-dependent p53-mediated cell death (Le Cam et al., 2006). Intriguingly, PCAF, which
has been unequivocally shown to acetylate p53 at K320, has
also been reported to exhibit E3 ubiquitin ligase activity toward
Mdm2 (Linares et al., 2007). Future studies may reveal whether
there is crosstalk between PCAF and E4F1 in the regulation of
p53 and Mdm2.
Regarding alternate modifications of p53 lysines, methylation in particular has been a subject of great interest. Methylation of K372 by the SET domain methyltransferase Set9 leads
to increased p21 expression (Chuikov et al., 2004), but whether
this has a selective impact on p53 target gene activation has
not been determined. Methylation of K382 by Set8, however,
has the interesting effect of suppressing the activation of several strong p53 targets but not others that are normally less
well induced (Shi et al., 2007). K370 is methylated by the methyltransferase Smyd2 (SET and MYND domain containing 2)
and causes the repression of p53 transcriptional activation,
although K370 methylation is itself inhibited by Set9 methylation of K372 (Huang et al., 2006). The demethylase LSD1
removes K370 dimethylation and in doing so prevents p53
from interacting with p53-binding protein 1 (53BP1), a coactivator of p53 (Huang et al., 2007). The functional roles of p53
lysine modifications are further complicated by the crosstalk
that exists between methylation and acetylation (Ivanov et al.,
2007). Specifically, methylation of K372 by Set7/9 is induced
by DNA damage and correlates with increased acetylation of
C-terminal p53 lysines including K382.
All told, a rather daunting set of combinatorial possibilities
can result from p53 lysine modifications. It is anticipated that
once a full set of data has accumulated regarding the possible
combination of modifications, great computational power will
be needed to deconstruct their impact on p53 functions and
the cellular outcomes. In the meantime, new p53 modifications continue to be uncovered. It was recently reported that
the protein arginine methyltransferase 5 (PRMT5) is involved in
methylation of at least two arginine residues (R333 and R335)
within the p53 tetramerization domain (Jansson et al., 2008).
Depletion of PRMT5 by siRNA in human cancer cell lines leads
Cell 137, May 1, 2009 ©2009 Elsevier Inc.
421
Figure 5. p53-Interacting Proteins Exert
Selective Influences on p53 Target Genes
and Outcomes
Different proteins can bind to p53 to induce different cellular outcomes to p53 activity. Proteins that
interact with the p53 DNA-binding core (green),
tetramerization domain (Tet, blue), or C-terminal
basic domain (++, purple) are shown. Brn3A, Hzf,
c-Abl, YB1, and p18/Hamlet selectively induce
p53 activation of genes encoding cell cycle regulators such as p21 to facilitate cell cycle arrest. In
contrast, ASPPs, the zebrafish p53 variant delta113p53, the p53 isoform p53β, Brn3b, NFκB/
p52, and Muc1 selectively activate the expression
of apoptotic regulators such as PUMA, Bax, and
Noxa to promote cell death.
to increased apoptosis along with loss of p21 and a modest
increase in proapoptotic Puma and Noxa proteins (Jansson et
al., 2008). Whether other p53 arginines are methylated remains
to be determined.
Changing Partners
Numerous noncovalent modifiers of p53 can also exert discriminatory effects on its ability to activate or repress its gene
targets (Figure 5). Indeed, as one might expect, there is complex interplay between p53 modifications and its binding partners. The best studied and validated of the p53 interactors are
its negative regulators Mdm2 and MdmX. A wealth of studies
have delved into their interactions with p53 and have been well
reviewed (Marine et al., 2006, 2007; Poyurovsky and Prives,
2006; Toledo and Wahl, 2006). Therefore, we will only highlight
here a few recent findings in this aspect of the p53 field. Mechanistic understanding of how Mdm2 and MdmX repress p53
transcriptional activity is an area still requiring insight. A recent
study showing that the loss of p53 rescues the early lethality in
mice caused by a mutant form of Mdm2 lacking E3 ligase activity indicates the primacy of the role of Mdm2 in degrading p53
(Itahana et al., 2007). Nonetheless, Mdm2 has also been shown
to reduce p53 acetylation by displacing p300 from p53 (Ito et al.,
2001; Kobet et al., 2000; Teufel et al., 2007), as well as by inhibiting and degrading PCAF (Jin et al., 2004). Mdm2 can also recruit
the histone deacetylases HDAC1 (Ito et al., 2002) and KAP1
(Wang et al., 2005), thus providing additional means by which
Mdm2 might function to repress acetylation of either p53 or histones in the vicinity of p53-binding sites. Mdm2 expressed from
its endogenous locus associates with p53 at the p21 promoter
(Arva et al., 2005; Minsky and Oren, 2004; Ohkubo et al., 2006;
Tang et al., 2008; White et al., 2006). Ectopically overexpressed
Mdm2 (and MdmX) can bind to several other p53 target promoters, with the exception of the Mdm2 promoter itself (Tang
et al., 2008). Having about 2-fold higher levels of Mdm2 protein in H1299 cells with tetracycline-regulated p53 expression
leads to lower levels of PIG3 and 14-3-3σ but does not affect
p21 or Bax expression when compared to similar cells without
extra Mdm2 (Ohkubo et al., 2006). Due to its ability to either displace acetylases or directly ubiquitinate histone H2B (Minsky
and Oren, 2004), or possibly through other mechanisms, Mdm2
coassociation with p53 at its target gene promoters allows it to
negatively regulate p53 transactivation in a selective manner.
How and when MdmX negatively regulates p53 are questions
being actively pursued (for example, see Wang et al., 2007b).
Perhaps the most challenging aspect of this field of study is
422 Cell 137, May 1, 2009 ©2009 Elsevier Inc.
integrating the roles of Mdm2 and MdmX in the regulation of
p53 turnover and localization with their respective impacts on
p53 transcriptional activities.
Intrinsic to the p53 protein is its ability to select different
target genes. Within its N terminus, p53 possesses transactivation domains (TADs; TAD I within residues 20–40 and TAD
II within residues 40–60) and a proline-rich domain (spanning
residues 60–90) that can also regulate transcription, possibly
in conjunction with TAD II (Harms and Chen, 2006). Loss of TAD
I function, most frequently achieved by simultaneous mutation
of leucine 22 (L22) and tryptophan 23 (W23) (p53L22Q/W23S),
produces a mutant p53 protein that was originally thought
to be virtually bereft of all transcriptional activity. Yet, recent
studies have shown that this mutant p53 protein can activate a
subset of p53 proapoptotic targets in mouse cells (Johnson et
al., 2005) and in human cells (Jung et al., 2006; Baptiste-Okoh
et al., 2008a). These findings reveal that additional regions
in p53—TAD II, the proline-rich region, or both—possess the
capacity to function autonomously in transcriptional activation.
Although TAD I and TAD II may be able to act independently,
they also work in concert to recruit specific components of the
multisubunit transcriptional activator STAGA complex, namely
GCN5, Taf9, and ADA2b, in order to activate target genes such
as p21, Puma, and GADD45 (Gamper and Roeder, 2008). Furthermore, the p53-mediated transcriptional repression that is
induced by hypoxia requires both TAD I and TAD II (Hammond
et al., 2006). Based on data from studies examining p53L22Q/
W23S, however, it is possible that some p53 targets do not
require the STAGA complex. Other regions in the p53 protein
with transactivation capability may be able to recruit distinct
factors that are necessary to induce gene activation. Small
molecules such as Nutlin (Vassilev et al., 2004) and compound
1d (Ding et al., 2005), which bind to Mdm2 and disrupt the interaction between Mdm2 and TAD I of p53, can greatly increase
p53 expression and activity (Ding et al., 2005; Vassilev et al.,
2004). Intriguingly, in contrast to the effect of Nutlin in inducing
cell cycle arrest, another small molecule (RITA) that also binds
to p53 and prevents it from interacting with Mdm2 causes
apoptosis and downregulation of p21 expression in some cells
(Enge et al., 2009). Whether these different compounds selectively affect different p53 activation regions is a question of
considerable interest.
Although intrinsic features of the p53 protein are critical to
its ability to induce cell cycle arrest or cell death, an increasing panoply of cellular factors have also been identified that
work with p53 to produce a specific cellular response (Figure
5). Not surprisingly, there is an intimate relationship between
the ability of p53 to activate its target genes and the transcription machinery with which p53 interacts. It has been shown in
recent years that the arginine methyl transferases CARM1 and
PRMT1 collaborate with p300 to facilitate p53-mediated transcription from DNA assembled into chromatin in vitro (An et al.,
2004). Additionally, components of a subcomplex (including the
protein MED/TRAP220) of the mediator transcriptional coactivator complex can interact with p53 (Zhang et al., 2005).
A multitude of p53 binding proteins that can redirect p53
toward a specific cellular outcome have also been uncovered.
For example, a well-studied p53 polymorphism at codon
72—where the residue can be either proline (P72) or arginine
(R72)—results in differential cellular outputs depending on the
p53 variant: the R72 variant protein is more proapoptotic than
the P72 allele (Pietsch et al., 2006). The ASPP family member,
iASPP, preferentially binds to the P72 variant of p53 and inhibits its activity, providing a mechanistic explanation for why the
R72 variant protein is more effective at inducing apoptosis
than the P72 variant protein (Bergamaschi et al., 2006). As
one example of the interplay between p53 modification and
its interaction with regulatory factors, the peptidyl-prolyl cis/
trans isomerase Pin1 recognizes p53 phosphorylated at S46,
leading to dissociation of iASPP from p53 and thereby promoting apoptosis (Mantovani et al., 2007). Another example
of a protein that binds to p53 and directs differential cellular
outcomes is the p38-regulated protein p18/Hamlet. p18/Hamlet associates with p53 and increases both p53-mediated
apoptosis and activation of some p53 target gene promoters
(e.g., Noxa) but not others (Puma, Bax, and p21) (Cuadrado
et al., 2007). Intriguingly, cyclin G, itself a p53 target, may
decrease levels of p18/Hamlet, providing another level of regulation of p53 outcomes (Cuadrado et al., 2007). Another protein, Brn3A, binds to p53 and specifically inhibits its ability to
activate the Bax and Noxa promoters to promote apoptosis.
However, Bm3A also cooperates with p53 to activate gene
expression from the p21 promoter (Hudson et al., 2005). It
remains to be seen if there is any direct competition or crossregulation between the activities exerted by p18/Hamlet and
Brn3A. Interestingly, Brn3b, a related factor to Brn3A, functions in the opposite manner as Brn3A by assisting p53 to
activate Bax expression and not p21 expression (BudhramMahadeo et al., 2006).
An interesting regulator of p53 targets genes is the p52
subunit of the transcription factor NFκB, which inhibits p21
expression but cooperates with p53 to increase Puma, DR5,
and Gadd45 expression; p52 also directly associates with the
promoters of these genes (Schumm et al., 2006). In the case of
Muc1, however, an integral membrane glycoprotein that is frequently overexpressed in cancer, it has been found that Muc1
associates with the p21 promoter in a p53-dependent manner
to facilitate p21 transcription. Interestingly, Muc1 also associates with the Bax promoter independent of p53 to repress Bax
expression. Consistent with Muc1’s regulatory functions, the
amount of Muc1 protein in cells shows positive correlations
with cell cycle arrest and cell survival, and negative correlations with cell death (Wei et al., 2005b). The Y-box-binding pro-
tein YB1 has a similar impact on p53. YB1 associates with p53,
blocking its activation of Bax expression, but does not impede
p53 induction of p21 expression (Homer et al., 2005). Finally, in
zebrafish, another selective p53 regulator, itself a p53 variant
(delta113p53), represses full-length p53 activation of arrest but
not apoptosis genes (Chen et al., 2005).
In addition to myriad p53-binding proteins, several new proteins have been identified that can also bind and regulate p53
but have yet to be shown to impart any selectivity to p53 target
gene activation. Among these proteins are Sug1, a component of the 19S proteasome (Zhu et al., 2007), heterogeneous
ribonucleoprotein particle K (hnRNPK), which possibly acts
through Mdm2 (Moumen et al., 2005), Hbo1 (Iizuka et al., 2008),
KLF5 (Zhu et al., 2006), NF-Y (Imbriano et al., 2005), clathrin
heavy chain (Enari et al., 2006), and the orphan receptor TR3
(Zhao et al., 2006). We will be very curious to learn whether
any of these proteins exert promoter-selective effects on p53’s
transactivation capabilities.
Finally, not all proteins that affect p53 transcriptional activities and outcomes interact directly with p53. Notable examples of this include a somewhat mysterious regulator of p53,
a noncoding RNA expressed from the MEG3 gene locus. This
noncoding RNA seems to selectively affect p53 in human cells
by downregulating Mdm2 expression, increasing p53 expression, and stimulating p53 activation of at least one target gene
(growth differentiation factor 15; GDF15), all without affecting p21 transcription (Zhou et al., 2007). Another instance of
these indirect p53 regulators is the transcriptional repressor
Zbt4, which forms a heterotrimeric complex with the Sin3 corepressor and the transcription factor Miz1 in order to repress
p53-mediated p21 induction and cell cycle arrest (Weber et al.,
2008). The transcription repressor Slug is a beautiful case in
point of a factor that profoundly affects p53 outcomes without
directly interacting with p53. The gene encoding Slug is a p53
target, and Slug proteins bind to the p53-inducible Puma promoter to repress both gene expression and irradiation-induced
apoptosis in hematopoietic cells (Wu et al., 2005).
Yet another factor that binds to a subset of p53 target genes
independently of p53 (for example, PIG3 and AIP1 but not p21
or Puma) is the nuclear transport factor hCas/Cse1L that cooperates with p53 to activate only those genes to which it binds.
At those promoters, hCas/Cse1L most likely functions by facilitating downregulation of the repressive histone trimethylation
mark, K27 of histone H3 (Tanaka et al., 2007).
All told, numerous transcriptional mechanisms that direct
p53 toward its different cellular outcomes have been reported.
The levels of p53, its modifications, the proteins that directly
interact with it, and the proteins that interact independent of
p53 with its target promoters can all differentially affect the
outcome of p53 activation (Figure 6).
New Kids on the Block: p53 and MicroRNAs
Protein-encoding genes are not the only transcription targets
of p53. No less than seven groups independently reported in
2007 that p53 can directly regulate the expression of select
microRNAs (miRNAs), most dramatically the miR-34 locus
consisting of miR-34a, miR-34b, and miR-34c (Bommer et al.,
2007; Chang et al., 2007; Corney et al., 2007; He et al., 2007;
Raver-Shapira et al., 2007; Tarasov et al., 2007; Tazawa et al.,
Cell 137, May 1, 2009 ©2009 Elsevier Inc.
423
of p53. It will be interesting for future
2007). Several reports showed that p53
studies to delve into the mechanism of
can bind directly to response elements
its discriminatory regulation.
within the miR-34a and miR-34b/c promoters to stimulate transcription from
Therapeutic Applications of p53
this locus. Certainly, miR-34a expresAlthough there is still much to learn,
sion is physiologically relevant to the
it seems clear that manipulating the
impact of p53 activity on cells: it can
p53 pathway will bring considerable
induce cell cycle arrest and senescence,
therapeutic benefits. As p53 activity is
as well as facilitate cell death. Functional
impaired or defective in most human
ablation of miR-34a reduces these p53cancers, regardless of type or tissue of
mediated effects on cells. Notably, the
origin, one obvious goal is to try and remiR-34 family is conserved in flies and
establish the growth-inhibitory functions
worms, a rather infrequent occurrence
of p53 in cancer cells. This approach
among miRNAs. As is often the case with
is elegantly supported by animal studmiRNAs, the most obvious challenge
ies where reactivation of wild-type p53
is to find the target genes of miR-34
leads to efficient tumor regression (Marwhose downregulation is required for
tins et al., 2006; Ventura et al., 2007; Xue
cell-cycle arrest or cell death. Convincet al., 2007). So how might p53 be reacing candidates identified in some of the
tivated? One line of attack that has been
reports include the cell cycle regulators
successful in the clinic is the use of gene
cyclin-dependent kinase 4 (CDK4) and
therapy to reintroduce p53 into tumor
cyclin E2, as well as the proto-oncogene
cells by means of vectors such as adenmesenchymal-epithelial transition factor
oviruses (Senzer and Nemunaitis, 2009).
(Met) and the antiapoptotic factor Bcl2.
Our growing understanding of how p53
More recently, the sirtuin SIRT1 was
is regulated has also led to the develalso shown to be a target of miR-34a,
opment of small molecule drugs that
and its downregulation correlates with
stabilize and activate the p53 protein.
increased acetylation of p53 (Yamakuchi
Although these drugs mostly function by
et al., 2008). The discovery of miR-34a
interfering with the ability of Mdm2 to taras a key p53 target begs the question
get p53 for degradation, cell-based drug
of whether its levels are altered in canscreens have also identified inhibitors
cer. Satisfyingly, some of the aforemenof sirtuins—protein deacetylases that
tioned studies uncovering p53 regulacan restrain p53 activity—as effective
tion of the miRNA locus report markedly Figure 6. Multiple Mechanisms of
p53-activating agents (Lain et al., 2008).
lower amounts of miR-34 in some human Differential p53 Target Gene Regulation
The cellular response to p53 activation can be
tumors and tumor-derived cell lines determined by differential target gene activation. These types of drugs, some of which are
(Bommer et al., 2007; Chang et al., 2007; Whether a given promoter is activated or re- in advanced stages of preclinical develpressed depends on the amount of p53 protein,
opment or early clinical trials (Shangary
Tazawa et al., 2007).
miR-34 is not the only microRNA to its modification state, and the cofactors present and Wang, 2009), are predicted to show
at the promoter. p53 protein levels or modification
be targeted by p53. Several of the first state can also dictate which genes are targeted efficacy in tumors that retain wild-type
publications identifying the miR-34 for transcriptional activation. The induction or re- p53. It is worth noting that the developlocus also found other possible miRNA pression of p53 target gene transcription can also ment of such drugs has sparked a vigordepend on the presence of numerous coactivators
targets of p53. It has now been reported or additional cooperating factors that enhance or ous debate about the potential toxicity
that a systemic activation of p53 may
that miR-192 and miR-215 are induced repress p53-induced transcription.
cause in normal tissues. In animal modby p53 and promote increased p21
expression (Braun et al., 2008). Moreover, miR-145 has been els, the absence of Mdm2 in normal p53-expressing tissues is
implicated as a p53 target that can repress c-myc expression alarmingly detrimental (Ringshausen et al., 2006), although it
(Sachdeva et al., 2009). In fact, a number of miRNAs that tar- might be hoped that Mdm2-inhibiting drugs will be less effecget antiproliferative genes have been shown to be repressed tive, and so easier to tolerate, than ablation of the Mdm2 gene.
by p53 in a manner that requires E2F, not unlike other tar- Furthermore, because these drugs are not genotoxic, they will
gets of p53-mediated repression (Brosh et al., 2008). Note hopefully avoid some of the damage inflicted by conventional
that although the small molecule Nutlin can induce miR-34a chemotherapies, which of course also activate p53 in all cells.
expression and senescence in some cells such as normal Intriguingly, some studies have suggested that Mdm2-inhibiting
human fibroblasts (Kumamoto et al., 2008), it fails to induce drugs function much better in cells undergoing DNA-damage
miR-34a and p21 expression in at least one tumor cell line signaling, a characteristic that might further distinguish nor(BV173 leukemia cells) that instead undergoes apoptosis upon mal cells from cancer cells (Brummelkamp et al., 2006). If the
Nutlin treatment (Paris et al., 2008). Thus, like other p53 target use of p53 activators in cancer therapy proves to be incredgenes, miR-34a is not universally induced upon the activation ibly successful, we may need to consider the possible proag424 Cell 137, May 1, 2009 ©2009 Elsevier Inc.
ing effects of p53 activation during systemic long-term treatment with these drugs, although at present such concerns will
likely seem trivial to most cancer patients. Restoration of p53
function in those cancers expressing mutant p53 is even more
challenging, although small molecules that refold some mutant
p53 proteins and thus reactivate their wild-type functions have
been described (Selivanova and Wiman, 2007). This approach
is technically difficult, but it may be an excellent way to selectively target cancer cells that express mutant p53 proteins.
The concept that we should be trying to reactivate p53 in cancer
cells is supported by experimental data and many human studies that show a correlation between mutations in p53 and poor
disease prognosis (Petitjean et al., 2007). However, as we have
highlighted in this Review, the cellular response to p53 can range
widely from cell death to cell survival, and the consequences of
retaining p53 activity in tumor cells are similarly difficult to predict. Indeed, the retention of wild-type p53 has been shown to
protect breast cancers from some forms of cytotoxic chemotherapy and so can be associated with a poor response to treatment (Bertheau et al., 2008). Possibly this effect of wild-type p53
could be exploited for therapeutic benefit. In this approach, failure
of tumor cells to mount a p53-mediated response would make
them particularly sensitive to cytotoxic drugs that function during S phase or the G2/M phase transition of the cell cycle. These
drugs would be much less toxic to normal cells that can benefit
from p53’s cell cycle arrest and survival activities (Sur et al., 2009).
An extension of this idea is to use drugs that activate p53 to further protect normal cells or tissue during the treatment of cancers harboring mutant p53 alleles (Carvajal et al., 2005; Kranz and
Dobbelstein, 2006). Somewhat confusingly, the same concept of
chemoprotection of normal tissue has also been proposed for
the use of p53 inhibitors. It is clear that much of the toxicity seen
in response to conventional genotoxic chemotherapies is due to
the activation of p53 and the subsequent p53-induced death of
radiosensitive cells in the hematopoetic system, gut lining, and
other tissues. In this case, inhibition of p53 in normal cells may
protect them from death, thereby increasing the patient’s tolerance to higher and hopefully more effective doses of radiation or
chemotherapy (Gudkov and Komarova, 2005; Strom et al., 2006).
Mouse studies support this strategy, suggesting that many side
effects of acute genotoxic insult might be avoided by a short-term
inhibition of p53, without causing a substantial loss in tumor suppressor activity (Christophorou et al., 2006).
The potential use of p53 in therapy is not limited to cancer.
For some disorders, the inhibition of p53 (or at least the inhibition of the p53-mediated apoptotic response) could be a
desired therapeutic goal. This area of p53-directed therapy is
still underexplored, but p53-inhibitory compounds have been
used with some success in animal models of ischemia and
Parkinson’s disease (Duan et al., 2002; Leker et al., 2004).
As drugs are developed that can reactivate wild-type functions of mutant p53, or turn wild-type p53 on or off, our understanding of how best to use these new tools in therapy will
grow. However, as we learn more about the intricacies of p53
regulation and function, predicting the outcomes of these drug
treatments becomes more difficult. The response of cancer
cells to p53 is clearly complicated enough, but how modulating
p53 might contribute to other aspects of disease and longevity
is only just beginning to be explored. Perhaps in a few years, a
similar review will be able to integrate the complexity of p53 into
a clearer picture to give insights into how this knowledge may
be used to improve the prognosis of not only cancer patients
but sufferers of other diseases as well.
Acknowledgments
We thank K. Ryan and E. Gottlieb for helpful suggestions and acknowledge
funding support from Cancer Research UK (K.H.V.) and NCI grant number
CA77742 (C.P.).
References
Amaravadi, R.K., Yu, D., Lum, J.J., Bui, T., Christophorou, M.A., Evan, G.I.,
Thomas-Tikhonenko, A., and Thompson, C.B. (2007). Autophagy inhibition
enhances therapy-induced apoptosis in a Myc-induced model of lymphoma. J. Clin. Invest. 117, 326–336.
An, W., Kim, J., and Roeder, R.G. (2004). Ordered cooperative functions of
PRMT1, p300, and CARM1 in transcriptional activation by p53. Cell 117,
735–748.
Appella, E., and Anderson, C.W. (2001). Post-translational modifications and activation of p53 by genotoxic stresses. Eur. J. Biochem. 268,
2764–2772.
Aranda-Anzaldo, A., and Dent, M.A. (2007). Reassessing the role of p53 in
cancer and ageing from an evolutionary perspective. Mech. Ageing Dev.
128, 293–302.
Arum, O., and Johnson, T.E. (2007). Reduced expression of the Caenorhabditis elegans p53 ortholog cep-1 results in increased longevity. J. Gerontol.
A Biol. Sci. Med. Sci. 62, 951–959.
Arva, N.C., Gopen, T.R., Talbott, K.E., Campbell, L.E., Chicas, A., White,
D.E., Bond, G.L., Levine, A.J., and Bargonetti, J. (2005). A chromatin-associated and transcriptionally inactive p53-Mdm2 complex occurs in mdm2
SNP309 homozygous cells. J. Biol. Chem. 280, 26776–26787.
Augustyn, K.E., Merino, E.J., and Barton, J.K. (2007). A role for DNA-mediated charge transport in regulating p53: Oxidation of the DNA-bound
protein from a distance. Proc. Natl. Acad. Sci. USA 104, 18907–18912.
Bae, B.I., Xu, H., Igarashi, S., Fujimuro, M., Agrawal, N., Taya, Y., Hayward,
S.D., Moran, T.H., Montell, C., Ross, C.A., et al. (2005). p53 mediates cellular dysfunction and behavioral abnormalities in Huntington’s disease.
Neuron 47, 29–41.
Baptiste-Okoh, N., Barsotti, A.M., and Prives, C. (2008a). A role for caspase 2 and PIDD in the process of p53-mediated apoptosis. Proc. Natl.
Acad. Sci. USA 105, 1937–1942.
Baptiste-Okoh, N., Barsotti, A.M., and Prives, C. (2008b). Caspase 2 is
both required for p53-mediated apoptosis and downregulated by p53 in a
p21-dependent manner. Cell Cycle 7, 1133–1138.
Barboza, J.A., Liu, G., El-Naggar, A.K., and Lozano, G. (2006). p21 delays
tumor onset by preservation of chromosomal stability. Proc. Natl. Acad.
Sci. USA 103, 19842–19847.
Batta, K., and Kundu, T.K. (2007). Activation of p53 function by human transcriptional coactivator PC4: role of protein-protein interaction, DNA bending, and posttranslational modifications. Mol. Cell. Biol. 27, 7603–7614.
Bauer, J.H., Poon, P.C., Glatt-Deeley, H., Abrams, J.M., and Helfand, S.L.
(2005). Neuronal expression of p53 dominant-negative proteins in adult
Drosophila melanogaster extends life span. Curr. Biol. 15, 2063–2068.
Bensaad, K., and Vousden, K.H. (2007). p53: new roles in metabolism.
Trends Cell Biol. 17, 286–291.
Bensaad, K., Tsuruta, A., Selak, M.A., Vidal, M.N., Nakano, K., Bartrons,
R., Gottlieb, E., and Vousden, K.H. (2006). TIGAR, a p53-inducible regulator of glycolysis and apoptosis. Cell 126, 107–120.
Cell 137, May 1, 2009 ©2009 Elsevier Inc.
425
Bergamaschi, D., Samuels, Y., Sullivan, A., Zvelebil, M., Breyssens, H., Bisso,
A., Del Sal, G., Syed, N., Smith, P., Gasco, M., et al. (2006). iASPP preferentially binds p53 proline-rich region and modulates apoptotic function of codon
72-polymorphic p53. Nat. Genet. 38, 1133–1141.
Berns, A. (2006). Cancer biology: can less be more for p53? Nature 443,
153–154.
Bertheau, P., Espie, M., Turpin, E., Lehmann, J., Plassa, L.-F., Varna, M., Janin, A., and de The, H. (2008). TP53 status and response to chemotherapy in
breast cancer. Pathobiology 75, 132–139.
Bode, A.M., and Dong, Z. (2004). Post-translational modification of p53 in
tumorigenesis. Nat. Rev. Cancer 4, 793–805.
Bommer, G.T., Gerin, I., Feng, Y., Kaczorowski, A.J., Kuick, R., Love, R.E.,
Zhai, Y., Giordano, T.J., Qin, Z.S., Moore, B.B., et al. (2007). p53-mediated
activation of miRNA34 candidate tumor-suppressor genes. Curr. Biol. 17,
1298–1307.
Bonnet, S., Archer, S.L., Allalunis-Turner, J., Haromy, A., Beaulieu, C., Thompson, R., Lee, C.T., Lopaschuk, G.D., Puttagunta, L., Bonnet, S., et al. (2007). A
mitochondrial-K+ channel axis is suppressed in cancer and its normalization
promotes apoptosis and inhibits cancer cell growth. Cancer Cell 11, 37–51.
Bourdon, J.C., Fernandes, K., Murray-Zmijewski, F., Liu, G., Diot, A., Xirodimas, D.P., Saville, M.K., and Lane, D.P. (2005). p53 isoforms can regulate p53
transcriptional activity. Genes Dev. 19, 2122–2137.
Braun, C.J., Zhang, X., Savelyeva, I., Wolff, S., Moll, U.M., Schepeler, T., Orntoft, T.F., Andersen, C.L., and Dobbelstein, M. (2008). p53-Responsive micrornas 192 and 215 are capable of inducing cell cycle arrest. Cancer Res.
68, 10094–10104.
Bretaud, S., Allen, C., Ingham, P.W., and Bandmann, O. (2007). p53-dependent neuronal cell death in a DJ-1-deficient zebrafish model of Parkinson’s
disease. J. Neurochem. 100, 1626–1635.
Brosh, R., Shalgi, R., Liran, A., Landan, G., Korotayev, K., Nguyen, G.H., Enerly, E., Johnsen, H., Buganim, Y., Solomon, H., et al. (2008). p53-Repressed
miRNAs are involved with E2F in a feed-forward loop promoting proliferation.
Mol. Syst. Biol 4, 229.
Brown, J.P., Wei, W., and Sedivy, J.M. (1997). Bypass of senescence after
disruption of p21CIP1/WAF1 gene in normal diploid human fibroblasts. Science
277, 831–834.
Bruins, W., Jonker, M.J., Bruning, O., Pennings, J.L., Schaap, M.M., Hoogervorst, E.M., van Steeg, H., Breit, T.M., and de Vries, A. (2007). Delayed expression of apoptotic and cell cycle control genes in carcinogen-exposed bladders
of mice lacking p53.S389 phosphorylation. Carcinogenesis 28, 1814–1823.
Bruins, W., Bruning, O., Jonker, M.J., Zwart, E., van der Hoeven, T.V., Pennings, J.L., Rauwerda, H., de Vries, A., and Breit, T.M. (2008). The absence
of Ser389 phosphorylation in p53 affects the basal gene expression level of
many p53-dependent genes and alters the biphasic response to UV exposure
in mouse embryonic fibroblasts. Mol. Cell. Biol. 28, 1974–1987.
Brummelkamp, T.R., Fabius, A.W., Mullenders, J., Madiredjo, M., Velds, A.,
Kerkhoven, R.M., Bernards, R., and Beijersbergen, R.L. (2006). An shRNA
barcode screen provides insight into cancer cell vulnerability to MDM2 inhibitors. Nat. Chem. Biol. 2, 202–206.
Budanov, A.V., and Karin, M. (2008). p53 target genes sestrin1 and sestrin2
connect genotoxic stress and mTOR signaling. Cell 134, 451–460.
Budhram-Mahadeo, V.S., Bowen, S., Lee, S., Perez-Sanchez, C., Ensor, E.,
Morris, P.J., and Latchman, D.S. (2006). Brn-3b enhances the pro-apoptotic
effects of p53 but not its induction of cell cycle arrest by cooperating in transactivation of bax expression. Nucleic Acids Res. 34, 6640–6652.
Carvajal, D., Tovar, C., Yang, H., Vu, B.T., Heimbrook, D.C., and Vassilev, L.T.
(2005). Activation of p53 by MDM2 antagonists can protect proliferating cells
from mitotic inhibitors. Cancer Res. 65, 1918–1924.
Cawley, S., Bekiranov, S., Ng, H.H., Kapranov, P., Sekinger, E.A., Kampa, D.,
Piccolboni, A., Sementchenko, V., Cheng, J., Williams, A.J., et al. (2004). Unbiased mapping of transcription factor binding sites along human chromo-
426 Cell 137, May 1, 2009 ©2009 Elsevier Inc.
somes 21 and 22 points to widespread regulation of noncoding RNAs. Cell
116, 499–509.
Cecchinelli, B., Porrello, A., Lazzari, C., Gradi, A., Bossi, G., D’Angelo, M.,
Sacchi, A., and Soddu, S. (2006). Ser58 of mouse p53 is the homologue of
human Ser46 and is phosphorylated by HIPK2 in apoptosis. Cell Death Differ.
13, 1994–1997.
Chang, T.C., Wentzel, E.A., Kent, O.A., Ramachandran, K., Mullendore, M.,
Lee, K.H., Feldmann, G., Yamakuchi, M., Ferlito, M., Lowenstein, C.J., et al.
(2007). Transactivation of miR-34a by p53 broadly influences gene expression
and promotes apoptosis. Mol. Cell 26, 745–752.
Chao, C., Wu, Z., Mazur, S.J., Borges, H., Rossi, M., Lin, T., Wang, J.Y., Anderson, C.W., Appella, E., and Xu, Y. (2006). Acetylation of mouse p53 at lysine
317 negatively regulates p53 apoptotic activities after DNA damage. Mol. Cell.
Biol. 26, 6859–6869.
Chen, J., Ruan, H., Ng, S.M., Gao, C., Soo, H.M., Wu, W., Zhang, Z., Wen, Z.,
Lane, D.P., and Peng, J. (2005). Loss of function of def selectively up-regulates Delta113p53 expression to arrest expansion growth of digestive organs
in zebrafish. Genes Dev. 19, 2900–2911.
Chipuk, J.E., Bouchier-Hayes, L., Kuwana, T., Newmeyer, D.D., and Green,
D.R. (2005). PUMA couples the nuclear and cytoplasmic proapoptotic function of p53. Science 309, 1732–1735.
Choudhury, A.R., Ju, Z., Djojosubroto, M.W., Schienke, A., Lechel, A., Schaetzlein, S., Jiang, H., Stepczynska, A., Wang, C., Buer, J., et al. (2007). Cdkn1a
deletion improves stem cell function and lifespan of mice with dysfunctional
telomeres without accelerating cancer formation. Nat. Genet. 39, 99–105.
Christofk, H.R., Vander Heiden, M.G., Harris, M.H., Ramanathan, A., Gerszten, R.E., Wei, R., Fleming, M.D., Schreiber, S.L., and Cantley, L.C. (2008). The
M2 splice isoform of pyruvate kinase is important for cancer metabolism and
tumor growth. Nature 452, 230–233.
Christophorou, M.A., Ringhausen, I., Finch, A.J., Brown Swigart, L., and Evan,
G.I. (2006). The pathological p53-mediated response to DNA damage is distinct from p53-mediated tumor suppression. Nature 14, 214–217.
Chuikov, S., Kurash, J.K., Wilson, J.R., Xiao, B., Justin, N., Ivanov, G.S., McKinney, K., Tempst, P., Prives, C., Gamblin, S.J., et al. (2004). Regulation of p53
activity through lysine methylation. Nature 432, 353–360.
Corney, D.C., Flesken-Nikitin, A., Godwin, A.K., Wang, W., and Nikitin, A.Y.
(2007). MicroRNA-34b and MicroRNA-34c are targets of p53 and cooperate
in control of cell proliferation and adhesion-independent growth. Cancer Res.
67, 8433–8438.
Cosme-Blanco, W., Shen, M.-F., Lazar, A.J.F., Pathak, S., Lozano, G., Multani,
A.S., and Chang, S. (2007). Telomere dysfunction suppresses spontaneous
tumorigeesis in vivo by initiating p53-dependent cellular senescence. EMBO
Rep. 8, 497–503.
Crighton, D., Wilkinson, S., O’Prey, J., Syed, N., Harrison, P.R., Gasco, M.,
Garrone, O., Crook, T., and Ryan, K.M. (2006). DRAM, a p53-induced modulator of autophagy, is critical for apoptosis. Cell 14, 121–134.
Cuadrado, A., Lafarga, V., Cheung, P.C., Dolado, I., Llanos, S., Cohen, P., and
Nebreda, A.R. (2007). A new p38 MAP kinase-regulated transcriptional coactivator that stimulates p53-dependent apoptosis. EMBO J. 26, 2115–2126.
Culmsee, C., and Landshamer, S. (2006). Molecular insights into mechanisms
of the cell death program: role in the progression of neurodegenerative disorders. Curr. Alzheimer Res. 3, 269–283.
D’Orazi, G., Cecchinelli, B., Bruno, T., Manni, I., Higashimoto, Y., Saito, S.,
Gostissa, M., Coen, S., Marchetti, A., Del Sal, G., et al. (2002). Homeodomain-interacting protein kinase-2 phosphorylates p53 at Ser 46 and mediates
apoptosis. Nat. Cell Biol. 4, 11–19.
Das, S., Raj, L., Zhao, B., Kimura, Y., Bernstein, A., Aaronson, S.A., and Lee,
S.W. (2007). Hzf Determines cell survival upon genotoxic stress by modulating
p53 transactivation. Cell 130, 624–637.
DeBerardinis, R.J., Lum, J.J., Hatzivassilou, G., and Thompson, C.B. (2008).
The biology of cancer: metabolic reprogramming fuels cell growth and prolif-
eration. Cell Metab. 7, 11–20.
Deng, Y., Chan, S.S., and Chang, S. (2008). Telomere dysfunction and tumour
suppression: the senescence connection. Nat. Rev. Cancer 8, 450–458.
Derry, W.B., Putzke, A.P., and Rothman, J.H. (2001). Caenorhabditis elegans
p53: role in apoptosis, meiosis and stress resistance. Science 294, 591–595.
Di Giovanni, S., Knights, C.D., Rao, M., Yakovlev, A., Beers, J., Catania, J.,
Avantaggiati, M.L., and Faden, A.I. (2006). The tumor suppressor protein
p53 is required for neurite outgrowth and axon regeneration. EMBO J. 25,
4084–4096.
Ding, K., Lu, Y., Nikolovska-Coleska, Z., Qiu, S., Ding, Y., Gao, W., Stuckey, J.,
Krajewski, K., Roller, P.P., Tomita, Y., et al. (2005). Structure-based design of
potent non-peptide MDM2 inhibitors. J. Am. Chem. Soc. 127, 10130–10131.
Duan, W., Zhu, X., Ladenheim, B., Yu, Q.S., Guo, Z., Oyler, J., Cutler, R.G.,
Cadet, J.L., Greig, N.H., and Mattson, M.P. (2002). p53 inhibitors preserve
dopamine neurons and motor function in experimental parkinsonism. Ann.
Neurol. 52, 597–606.
Efeyan, A., Garcia-Cao, I., Herranz, D., Velasco-Miguel, S., and Serrano, M.
(2006). Policing of oncogene activity by p53. Nature 443, 159.
El-Deiry, W.S. (1998). Regulation of p53 downstream genes. Semin. Cancer
Biol. 8, 345–357.
Miu, K., Watnick, R.S., Reinhardt, F., et al. (2008). Growth-inhibitory and tumor- suppressive functions of p53 depend on its repression of CD44 expression. Cell 134, 62–73.
Gorrini, C., Squatrito, M., Luise, C., Syed, N., Perna, D., Wark, L., Martinato,
F., Sardella, D., Verrecchia, A., Bennett, S., et al. (2007). Tip60 is a haplo-insufficient tumour suppressor required for an oncogene-induced DNA damage
response. Nature 448, 1063–1067.
Gresko, E., Roscic, A., Ritterhoff, S., Vichalkovski, A., del Sal, G., and Schmitz,
M.L. (2006). Autoregulatory control of the p53 response by caspase-mediated
processing of HIPK2. EMBO J. 25, 1883–1894.
Gudkov, A.V., and Komarova, E.A. (2005). Prospective therapeutic applications of p53 inhibitors. Biochem. Biophys. Res. Commun. 331, 726–736.
Halazonetis, T.D., Gorgoulis, V.G., and Barteck, J. (2008). An oncogene-induced
DNA damage model for cancer development. Science 319, 1352–1355.
Hammond, E.M., Mandell, D.J., Salim, A., Krieg, A.J., Johnson, T.M., Shirazi,
H.A., Attardi, L.D., and Giaccia, A.J. (2006). Genome-wide analysis of p53
under hypoxic conditions. Mol. Cell. Biol. 26, 3492–3504.
Harms, K.L., and Chen, X. (2006). The functional domains in p53 family proteins exhibit both common and distinct properties. Cell Death Differ. 13,
890–897.
Enari, M., Ohmori, K., Kitabayashi, I., and Taya, Y. (2006). Requirement of clathrin heavy chain for p53-mediated transcription. Genes Dev. 20, 1087–1099.
He, L., He, X., Lim, L.P., de Stanchina, E., Xuan, Z., Liang, Y., Xue, W., Zender,
L., Magnus, J., Ridzon, D., et al. (2007). A microRNA component of the p53
tumour suppressor network. Nature 447, 1130–1134.
Enge, M., Bao, W., Hedstrom, E., Jackson, S.P., Moumen, A., and Selivanova,
G. (2009). MDM2-dependent downregulation of p21 and hnRNP K provides
a switch between apoptosis and growth arrest induced by pharmacologically
activated p53. Cancer Cell 15, 171–183.
Hearnes, J.M., Mays, D.J., Schavolt, K.L., Tang, L., Jiang, X., and Pietenpol,
J.A. (2005). Chromatin immunoprecipitation-based screen to identify functional genomic binding sites for sequence-specific transactivators. Mol. Cell.
Biol. 25, 10148–10158.
Esteve, P.O., Chin, H.G., and Pradhan, S. (2007). Molecular mechanisms of
transactivation and doxorubicin-mediated repression of survivin gene in cancer cells. J. Biol. Chem. 282, 2615–2625.
Hemann, M.T., Zilfou, J.T., Zhao, Z., Burgess, D.J., Hannon, G.L., and Lowe,
S.W. (2004). Suppression of tumorigenesis by the p53 target PUMA. Proc.
Natl. Acad. Sci. USA 101, 9333–9338.
Fantin, V.R., Syt-Pierre, J., and Leder, P. (2006). Attenuation of LDH-A expression uncovers a link between glycolysis, micochondrial physiology, and tumor
maintenance. Cancer Cell 9, 425–434.
Ho, W.C., Fitzgerald, M.X., and Marmorstein, R. (2006). Structure of the p53
core domain dimer bound to DNA. J. Biol. Chem. 281, 20494–20502.
Feng, L., Lin, T., Uranishi, H., Gu, W., and Xu, Y. (2005). Functional analysis of
the roles of posttranslational modifications at the p53 C terminus in regulating
p53 stability and activity. Mol. Cell. Biol. 25, 5389–5395.
Feng, Z., Jin, S., Zupnick, A., Hoh, J., de Stanchina, E., Lowe, S., Prives, C.,
and Levine, A.J. (2006). p53 tumor suppressor protein regulates the levels of
huntingtin gene expression. Oncogene 25, 1–7.
Feng, Z., Hu, W., de Stanchina, E., Teresky, A.K., Jin, S., Lowe, S., and Levine,
A.J. (2007a). The regulation of AMPK beta1, TSC2, and PTEN expression by
p53: stress, cell and tissue specificity, and the role of these gene products in
modulating the IGF-1-AKT-mTOR pathways. Cancer Res. 67, 3043–3053.
Feng, Z., Hu, W., Teresky, A.K., Hernando, E., Cordon-Cardo, C., and Levine,
A.J. (2007b). Declining p53 function in the aging process: a possible mechanism for the increased tumor incidence in older populations. Proc. Natl. Acad.
Sci. USA 104, 16633–16638.
Hofmann, T.G., Moller, A., Sirma, H., Zentgraf, H., Taya, Y., Droge, W., Will,
H., and Schmitz, M.L. (2002). Regulation of p53 activity by its interaction with
homeodomain-interacting protein kinase-2. Nat. Cell Biol. 4, 1–10.
Hoh, J., Jin, S., Parrado, T., Edington, J., Levine, A.J., and Ott, J. (2002).
The p53MH algorithm and its application in detecting p53-responsive genes.
Proc. Natl. Acad. Sci. USA 99, 8467–8472.
Homer, C., Knight, D.A., Hananeia, L., Sheard, P., Risk, J., Lasham, A., Royds,
J.A., and Braithwaite, A.W. (2005). Y-box factor YB1 controls p53 apoptotic
function. Oncogene 24, 8314–8325.
Hu, W., Feng, Z., Teresky, A.K., and Levine, A.J. (2007). p53 regulates maternal reproduction through LIF. Nature 450, 721–724.
Huang, J., Perez-Burgos, L., Placek, B.J., Sengupta, R., Richter, M., Dorsey,
J.A., Kubicek, S., Opravil, S., Jenuwein, T., and Berger, S.L. (2006). Repression of p53 activity by Smyd2-mediated methylation. Nature 444, 629–632.
Fogal, V., Hsieh, J.K., Royer, C., Zhong, S., and Lu, X. (2005). Cell cycle-dependent nuclear retention of p53 by E2F1 requires phosphorylation of p53 at
Ser315. EMBO J. 24, 2768–2782.
Huang, J., Sengupta, R., Espejo, A.B., Lee, M.G., Dorsey, J.A., Richter, M.,
Opravil, S., Shiekhattar, R., Bedford, M.T., Jenuwein, T., et al. (2007). p53 is
regulated by the lysine demethylase LSD1. Nature 449, 105–108.
Gamper, A.M., and Roeder, R.G. (2008). Multivalent binding of p53 to the STAGA complex mediates coactivator recruitment after UV damage. Mol. Cell.
Biol. 28, 2517–2527.
Hudson, C.D., Morris, P.J., Latchman, D.S., and Budhram-Mahadeo, V.S.
(2005). Brn-3a transcription factor blocks p53-mediated activation of
proapoptotic target genes Noxa and Bax in vitro and in vivo to determine cell
fate. J. Biol. Chem. 280, 11851–11858.
Garrison, S.P., Jeffers, J.R., Yang, C., Nilsson, J.A., Hall, M.A., Rehg, J.E.,
Yue, W., Yu, J., Zhang, L., Onciu, M., et al. (2008). Selection against PUMA
gene expression in Myc-driven B cell lymphomagenesis. Mol. Cell. Biol. 28,
5391–5402.
Iizuka, M., Sarmento, O.F., Sekiya, T., Scrable, H., Allis, C.D., and Smith, M.M.
(2008). Hbo1 Links p53-dependent stress signaling to DNA replication licensing. Mol. Cell. Biol. 28, 140–153.
Gatz, S.A., and Wiesmuller, L. (2006). p53 in recombination and repair. Cell
Death Differ. 13, 1003–1016.
Godar, S., Ince, T.A., Bell, G.W., Feldser, D., Donaher, J.L., Bergh, J., Liu, A.,
Imbriano, C., Gurtner, A., Cocchiarella, F., Di Agostino, S., Basile, V., Gostissa,
M., Dobbelstein, M., Del Sal, G., Piaggio, G., and Mantovani, R. (2005). Direct
p53 transcriptional repression: in vivo analysis of CCAAT-containing G2/M
promoters. Mol. Cell. Biol. 25, 3737–3751.
Cell 137, May 1, 2009 ©2009 Elsevier Inc.
427
Innocente, S.A., and Lee, J.M. (2005). p53 is a NF-Y- and p21-independent, Sp1-dependent repressor of cyclin B1 transcription. FEBS Lett. 579,
1001–1007.
Itahana, K., Mao, H., Jin, A., Itahana, Y., Clegg, H.V., Lindstrom, M.S., Bhat,
K.P., Godfrey, V.L., Evan, G.I., and Zhang, Y. (2007). Targeted inactivation of
Mdm2 RING finger E3 ubiquitin ligase activity in the mouse reveals mechanistic insights into p53 regulation. Cancer Cell 12, 355–366.
Knights, C.D., Catania, J., Di Giovanni, S., Muratoglu, S., Perez, R., Swartzbeck, A., Quong, A.A., Zhang, X., Beerman, T., Pestell, R.G., et al. (2006).
Distinct p53 acetylation cassettes differentially influence gene-expression
pat...
Purchase answer to see full
attachment