Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Article
Gain-of-Function Mutant p53 Promotes
Cell Growth and Cancer Cell Metabolism
via Inhibition of AMPK Activation
Ge Zhou,1,* Jiping Wang,1 Mei Zhao,1 Tong-Xin Xie,1 Noriaki Tanaka,1 Daisuke Sano,1 Ameeta A. Patel,1
Alexandra M. Ward,1 Vlad C. Sandulache,1 Samar A. Jasser,1 Heath D. Skinner,1 Alison Lea Fitzgerald,1
Abdullah A. Osman,1 Yongkun Wei,2 Xuefeng Xia,5,6 Zhou Songyang,7,8 Gordon B. Mills,3 Mien-Chie Hung,2,9
Carlos Caulin,1,4 Jiyong Liang,3 and Jeffrey N. Myers1,5,*
1Department
of Head and Neck Surgery
of Molecular and Cellular Oncology
3Department of Systems Biology
4Department of Genetics
5Department of Cancer Biology
The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
6Center for Diabetes Research, Methodist Hospital Research Institute, Houston, TX 77030, USA
7Verna and Marrs McLean Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, TX 77030, USA
8School of Life Sciences, Sun Yat-sen University, Guangzhou 510275, China
9Graduated Institute of Cancer Biology and Center for Molecular Medicine, China Medical University, Taichung 404, Taiwan
*Correspondence: [email protected] (G.Z.), [email protected] (J.N.M.)
http://dx.doi.org/10.1016/j.molcel.2014.04.024
2Department
SUMMARY
Many mutant p53 proteins (mutp53s) exert oncogenic gain-of-function (GOF) properties, but the
mechanisms mediating these functions remain
poorly defined. We show here that GOF mutp53s
inhibit AMP-activated protein kinase (AMPK)
signaling in head and neck cancer cells. Conversely,
downregulation of GOF mutp53s enhances AMPK
activation under energy stress, decreasing the
activity of the anabolic factors acetyl-CoA carboxylase and ribosomal protein S6 and inhibiting aerobic
glycolytic potential and invasive cell growth. Under
conditions of energy stress, GOF mutp53s, but not
wild-type p53, preferentially bind to the AMPKa
subunit and inhibit AMPK activation. Given the
importance of AMPK as an energy sensor and tumor
suppressor that inhibits anabolic metabolism, our
findings reveal that direct inhibition of AMPK activation is an important mechanism through which
mutp53s can gain oncogenic function.
INTRODUCTION
Mutation of the TP53 tumor suppressor gene is one of the most
frequent genetic alterations in cancer, including head and neck
squamous cell carcinoma (HNSCC) (Agrawal et al., 2011; Stransky et al., 2011). Although mutation of the TP53 gene can result in
loss of wild-type p53 (wtp53) function or exert a dominant-negative effect over the remaining wild-type allele, some mutated
forms of p53 (mutp53s) can lead to a gain of oncogenic properties that promote tumor growth and progression. However, the
mechanisms involved in mutp53 gain of function (GOF) remain
relatively poorly understood (Oren and Rotter, 2010).
Metabolic alterations, particularly the metabolic reprogramming to aerobic glycolysis (i.e., the Warburg effect) and the reprograming of mitochondrial metabolism, represent a hallmark
of cancer that contributes to malignant transformation as well
as the growth and maintenance of tumors (Hanahan and Weinberg, 2011; Vander Heiden et al., 2009; Ward and Thompson,
2012). In vivo dynamic mechanisms such as phosphoinositide
3-kinase (PI3K)/protein kinase B (AKT)/mammalian homolog of
target of rapamycin (mTOR) and adenosine monophosphate
(AMP)-activated protein kinase (AMPK) sense the cellular energy
status and regulate the balance between anabolism (an adenosine triphosphate [ATP]-consuming process that leads to macromolecular synthesis) and catabolism (a process that degrades
marcomolecules to release energy through increased ATP
production) (DeBerardinis and Thompson, 2012). AMPK is a
highly conserved heterotrimeric serine/threonine protein kinase
complex composed of a catalytic a subunit and regulatory b
and g subunits. As a major cellular energy sensor and a master
regulator of metabolic homeostasis, AMPK is sensitive to the
cellular AMP:ATP and adenosine diphosphate:ATP ratios and
is activated by metabolic stresses that inhibit ATP production
or stimulate ATP consumption (Hardie et al., 2012). Once activated, AMPK stimulates catabolism while inhibiting anabolism.
AMPK achieves these effects by targeting many downstream
metabolic enzymes (e.g., acetyl-CoA carboxylase [ACC] and
mTOR) and by phosphorylating transcription factors (e.g., sterol
regulatory element-binding protein 1 [SREBP1]) or cofactors that
regulate gene expression (Hardie et al., 2012; Mihaylova and
Shaw, 2011).
Studies have shown that wtp53 can regulate many metabolic
pathways, such as carbohydrate and lipid metabolism, ROS
regulation, and autophagy (Berkers et al., 2013; Goldstein and
Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc. 1
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
Rotter, 2012). Importantly, stimulation of AMPK leads to the
phosphorylation and activation of wtp53 (Jones et al., 2005;
Okoshi et al., 2008). However, it remains unclear whether
wtp53 is the direct target of AMPK (Fogarty and Hardie, 2010;
Hardie, 2011). Recently, AMPK was shown to promote the
stability of wtp53 indirectly through phosphorylation and inactivation of MDMX (He et al., 2014) and the p53 deacetylase,
SIRT1 (Lee et al., 2012). The activation of wtp53 by AMPK
signaling is believed to establish a metabolic checkpoint to
suppress the growth of cells under conditions of metabolic
stress (Jones et al., 2005). Therefore, AMPK is considered a
tumor suppressor (Faubert et al., 2013; Luo et al., 2010). Moreover, once activated, wtp53 can, in turn, increase AMPK activity
through transcriptional activation of the gene encoding the b
subunit of AMPK (Feng et al., 2007) and sestrin (Budanov
and Karin, 2008), providing a positive feedback effect to AMPK
function. This positive feedback between AMPK and wtp53 is
believed to play an important role in tumor suppression.
The vast majority of mutant p53s arise from missense mutations that can cause significant alterations in tertiary structure
(Xu et al., 2011) which, in turn, can cause changes in p53 function
through altered protein-protein interactomes and/or altered
regulation of gene expression, thereby contributing to mutp53
GOF properties (Freed-Pastor and Prives, 2012; Muller and
Vousden, 2013; Solomon et al., 2012). Recently, mutp53s were
also shown to regulate metabolic pathways, such as steroid
metabolism, via regulation of the transcription factor SREBP
(Freed-Pastor et al., 2012), a downstream target of AMPK that
directly phosphorylates and inhibits SREBP activity (Li et al.,
2011). In the current study, we show that AMPK signaling is
inhibited by GOF mutp53s. Moreover, we show that GOF
mutp53s, but not wtp53, preferentially bind to the AMPKa
subunit and directly inhibit AMPK activation, which increases
anabolic growth and contributes to the GOF properties of
mutp53s.
RESULTS
mutp53s Gain Oncogenic Function to Promote Invasive
Growth of HNSCC Cells Both In Vitro and In Vivo
To study the functional impact of TP53 mutations, we first
selected several human tumor-derived HNSCC cell lines with
various TP53 status (Figures 1A and 1B) with which to test the
function of mutp53s. When treated with the DNA-damaging
agent 5-fluorouracil (5-FU), all the cell lines with mutp53s
appeared to have lost wtp53 transcriptional function, as shown
by the lack of induction of expression of the wtp53 targets,
MDM2 and p21 (Figure 1B). One HNSCC cell line tested,
UMSCC1, was found to lack detectable endogenous p53 protein
due to the instability of splice site mutants (Figures 1A and 1B,
and see Figure S1 online). This cell line was then stably transfected with various mutp53s (Figure 1C) to evaluate whether
they could confer GOF activity. Overexpression of mutp53s
P151S, R175H, G245C, and R282W in UMSCC1 cells promoted
invasive cell growth, as characterized by the formation of larger
tumor spheroids with longer migration and invasion distances in
an inverted three-dimensional (3D) Matrigel culture when
compared with retroviral vector pBabe (Figure 1D), whereas
expression of the C-terminal truncation mutp53, E336X, did
not significantly promote UMSCC1 cell growth in the same assay
(Figure 1D).
In further support of mutp53 GOF, we used lentivirusmediated p53 short hairpin RNA (shp53RNA) interference to
knock down the expression of endogenous mutp53s (Figure 1E).
Suppression of endogenous mutp53 R175H and P151S expression in Detroit 562 and Tu138 cells, respectively, resulted in
smaller tumor spheroids with less migration and invasion
compared with the control (Figure 1F). To further support the
idea that mutp53 P151S is a GOF mutant, we previously demonstrated that overexpression of mutp53 P151S in UMSCC1 cells
led to increased tumor growth and decreased animal survival
in a mouse orthotopic xenograft oral tumor model. Conversely,
downregulation of mutp53 P151S in Tu138 cells suppressed
tumor growth and increased animal survival in the same model
(Sano et al., 2011).
Collectively, our in vitro and in vivo results clearly demonstrated that certain mutp53s, such as P151S, R175H, G245C,
and R282W, gain oncogenic function to promote invasive cell
growth, whereas the C-terminal truncation mutp53, E336X,
exhibits limited or no significant GOF activities in UMSCC1 cells.
Overexpression of GOF mutp53s Inhibits AMPK
Activation
The formation of larger tumor spheroids in 3D Matrigel by overexpression of GOF mutp53s in UMSCC1 cells could have been
due to an increase in cell proliferation, cell growth (cell size), or
both, which is strongly linked to mTOR signaling (Zoncu et al.,
2011). Consistent with this, downregulation of mutp53 P151S
in Tu138 cells resulted in a slightly smaller cell size when cells
were cultured under low-glucose culture condition (Figure S2A).
Because wtp53 is involved in inhibiting mTOR signaling by
modulating both positive (PI3K/AKT) and negative (liver kinase
B1 [LKB1]/AMPK) upstream regulatory pathways of mTOR
signaling (Feng and Levine, 2010), we then examined AKT and
AMPK signaling in our stable cell lines expressing mutp53s.
Interestingly, while no difference in AKT phosphorylation was
observed in UMSCC1 stable cell lines (Figures 2D–2F), we noted
that the induction of AMPKa Thr172 phosphorylation, which
is required for AMPK kinase activation (Hawley et al., 1996),
in response to glucose deprivation, 5-aminoimidazole-4carboxamide-1-b-4-ribofuranoside (AICAR), and metabolic
stress (serum-free, glucose-free, and hypoxia) was inhibited in
UMSCC1 cells expressing mutp53 R175H when compared
with the control cell line lacking mutp53 expression (Figures
2A–2C). However, under the same conditions, AMPK activation
was only minimally affected in cells expressing the C-terminal
truncated, non-GOF mutp53 E336X (Figures 2A–2C).
We then examined whether other GOF mutp53s have similar
inhibitory effects on AMPK signaling in UMSCC1 cells. Under
subconfluent culture condition two mutp53s, P151S and
R175H, but not G245C and R282W, efficiently inhibited phosphorylation of AMPKa as well as ACC, a downstream target of
AMPK (Winder et al., 1997), in response to glucose deprivation
(Figures 2D). However, when cells were cultured under highdensity cell culture condition, all GOF mutp53s, including
G245C and R282W, which did not inhibit AMPK activation in
2 Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc.
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
Figure 1. mutp53s Gain Oncogenic Function to Promote Invasive Growth of HNSCC Cells
(A) Schematic diagram of wtp53 and mutp53s identified in the HNSCC cell lines shown in (B).
(B) Immunoblotting (IB) analysis of lysates from HNSCC cell lines with wtp53 or mutp53s in response to 5-FU treatment (1 mM, 24 hr). Breast cancer cell line
MCF-7 and HNSCC cell line HN30 with wtp53 were used as the positive controls.
(C) IB analysis of UMSCC1 cell lines stably overexpressing various mutp53s after retroviral transfection. Various HNSCC cell lines with corresponding
endogenous mutp53s and UMSCC1 cells expressing retroviral vector pBabe were used as controls.
(D) Inverted in vitro 3D Matrigel culture of various UMSCC1 stable cell lines. Images were acquired using confocal microscopy every 40 mm from the bottom to the
top of the Matrigel culture.
(E) IB analysis of control and lentiviral shp53RNA-transfected Detroit 562 and Tu138 cells. Ctr, control shRNA.
(F) Inverted in vitro 3D Matrigel culture of HNSCC cell lines Detroit 562 and Tu138. Quantitative data are represented as mean ± SD.
Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc. 3
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
Figure 2. Overexpression of GOF mutp53s Suppresses Activation of AMPK Signaling in Response to Energy Stress
(A–F) IB analyses of various UMSCC1 stable cell lines in response to stress: glucose starvation (A, D–F), 1 mM AICAR (B), and metabolic stress (glucose-free,
serum-free, and hypoxia [1% O2]) (C). In (B), cells were serum starved overnight before AICAR treatment in the absence of serum.
(G and H) IB analyses of K-Ras–transformed or pBabe control p53 / and p53(R172H/H) MEFs in response to glucose starvation. See also Figure S2.
subconfluent culture, inhibited AMPK activation in response to
glucose deprivation when compared with the control (Figures
2E and S2B). Importantly, this inhibition appeared to be specific
to GOF mutp53s, since overexpression of neither non-GOF
mutp53 E336X nor wtp53 inhibited AMPK activation under the
same condition (Figures 2A–2C and 2F). To understand why all
GOF mutp53s inhibited AMPK activation (AMPKa Thr172 phos-
phorylation), we further used quantitative reverse transcription
PCR (RT-qPCR) and western blotting to examine the possible
role of transcription control involved in mutp53-mediated
AMPK inhibition. Our results, however, showed no inhibition of
mutp53s to expression of the subunits of AMPK and its upstream
LKB1 kinase complexes (i.e., AMPKa/b/g and LKB1/Mo25a/
b/Strada/b) that are directly involved in AMPK activation in
4 Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc.
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
response to energy stress (Hardie et al., 2012) in UMSCC1 cells
(Figures 2D–2F and S2C).
To further test the role of GOF mutp53s in AMPK signaling, we
used mouse embryonic fibroblasts (MEFs) with or without
mutp53R172H/H (equivalent to human mutp53R175H/H), which
was shown to have GOF activity in a genetically engineered
mouse model (Lang et al., 2004; Olive et al., 2004). Compared
with control p53 / MEF cells, mutp53R172H/H MEFs exhibited
lower levels of induction of AMPK and ACC phosphorylation in
response to glucose starvation under confluent culture condition
(Figures 2G and 2H). Taken together, our results showed that
unlike wtp53, which activates AMPK signaling (Budanov and
Karin, 2008; Feng et al., 2007), GOF mutp53s can actually inhibit
AMPK activation and AMPK signaling.
Downregulation of GOF mutp53s Enhances AMPK
Activation Induced by Energy Stress
As further evidence that GOF mutp53s inhibit AMPK activation,
knockdown of endogenous mutp53 P151S in Tu138 cells (Figures 3A and 3B), and mutp53 R175H in pharyngeal cancer
Detroit 562 cells (Figure 3C) and breast cancer SKBR3 cells (Figure 3D) resulted in increased induction of AMPK and ACC phosphorylation in response to glucose deprivation or metabolic
stress. In contrast, we did not observe any impact of downregulation of wtp53 on AMPK activation in response to glucose deprivation in HN30 cells (Figure S3). Additionally, consistent with the
role of AMPK as a negative regulator of the mTOR/S6 ribosomal
protein pathway (Shackelford and Shaw, 2009), downregulation
of mutp53 P151S in Tu138 cells resulted in decreased phosphorylation of the S6 ribosomal protein in response to glucose
deprivation (Figure 3E), presumably through enhanced AMPK
signaling. Moreover, immunohistochemical (IHC) evaluation of
orthotopically grown oral cancer xenografts derived from
Tu138-control shRNA (Ctr) and Tu138-shp53 cells (Sano et al.,
2011) confirmed that Tu138-shp53 tumors exhibit increased
levels of phosphorylated AMPK (pAMPK) and ACC (pACC)
compared with Tu138-Ctr tumors (Figure 3F).
Finally, introducing an shRNA-resistant version of mutp53
R175H or P151S into Detroit 562 shp53 or Tu138 shp53 cells
by retroviral infection partially restored mutp53 expression in
these cells, which not only led to decreased levels of energy
stress-induced pAMPK (Figures 3G and 3H) but also partly
rescued the invasive growth and prevented the phenotypic
regression of Detroit 562 shp53 cells in 3D Matrigel culture
(Figure 3I). These results, taken together, further supported our
hypothesis that GOF mutp53s negatively regulate AMPK activation and signaling and contribute to the invasive growth phenotype of HNSCC cells both in vitro and in vivo.
Energy Stress Induces the Interaction of GOF mutp53s
and AMPK In Vivo
In vivo, AMPK is shuttled between the nucleus and cytoplasm
(Kodiha et al., 2007). However, cytoplasm is where AMPK is
activated by its upstream kinase in response to energy stress
(Tsou et al., 2011). To investigate the mechanisms involved in
mutp53-mediated AMPK inhibition, we first examined the subcellular localization, especially the cytoplasmic localization, of
mutp53s in UMSCC1 cells. Immunofluorescent staining revealed
that GOF mutp53 P151S was localized in both the nucleus and
cytoplasm of UMSCC1 cells (Figure 4A). In contrast, mutp53
E336X, likely because of its lack of a C-terminal p53 nuclear
export signal (residues 340–351) (Stommel et al., 1999), localized
exclusively in the nucleus of UMSCC1 cells (Figure 4B). Since
this mutp53 E336X could not inhibit AMPK activation in vivo (Figures 2A–2C), the subcellular localization of mutp53s (i.e., the
cytoplasmic localization) might be critical for mutp53-mediated
AMPK inhibition. Consistent with this notion, GOF mutp53
G245C, which could only efficiently inhibit AMPK activation in
high-density cell culture (compare Figures 2E and 2D), was predominantly localized to the nucleus in low-density cell culture
(Figure 4C). However, its increased cytoplasmic localization
was seen when cell density was increased (Figure 4C), suggesting that the stress from high-density cell culture facilitates the
cytoplasmic localization of this mutp53.
Next, we investigated whether mutp53s could interact with
AMPK in vivo. To this end, we first transfected 293FT human embryonic kidney cells with green fluorescent protein (GFP)-tagged
AMPKa2 and Myc-tagged mutp53 R175H. Our results showed
that mutp53 R175H was strongly associated with the GFPAMPKa2 complex but not with control GFP (Figure 4D). Similarly,
when Myc-tagged GOF mutp53s P151S, R175H, G245C, and
R282W were transfected, they were all able to complex with hemagglutinnin (HA)-tagged AMPKa2 expressed in 293FT cells (Figure 4E, lanes 5–8; Figure S4). In contrast, neither transfected wtp53
nor the nucleus-localized mutp53 E336X was able to complex with
HA-tagged AMPKa2 in the same assay (Figure 4E, compare lanes
4 and 9 to lanes 5–8), suggesting that compared to wtp53 and
E336X, GOF mutp53s (i.e., P151S, R175H, G245C, and R282W)
gain the capability to interact with AMPKa2 in 293FT cells.
The interaction of GOF mutp53s P151S, R175H, G245C, and
R282W with endogenous AMPKa was also observed in
UMSCC1 cell lines stably expressing these mutp53s (Figure 4F)
and in different HNSCC cell lines expressing various endogenous mutp53s (Figures 4G–4K), whereas no interaction between
AMPKa and wtp53 or nucleus-localized mutp53 E336X was
detected under the same conditions (Figures 4F and 4I–4K).
Our results further showed that although endogenous GOF
mutp53s P151S and R175H weakly interacted with the endogenous AMPKa in Tu138 and Detroit 562 cells under normal culture
conditions, the enhanced interaction between mutp53s and
AMPKa could actually be induced by glucose deprivation
(Figures 4J and 4K), suggesting that energy stress induces
AMPKa-mutp53s interaction in vivo. Finally, our results
confirmed that in the immunoprecipitated mutp53-AMPKa complex from Detroit 562 cells, the b and g subunits of AMPK were
also detected (Figures 4H and 4L), suggesting that GOF mutp53s
interact not only with endogenous a subunit of AMPK but also
with endogenous AMPK(a/b/g) holoenzyme in vivo.
Taken together, these results clearly showed not only that the
interaction between GOF mutp53s and AMPK (a/b/g) occurs
in vivo but also that it is induced by energy stress.
mutp53s, but Not wtp53, Preferentially Bind to AMPKa
and Inhibit AMPK Activation
To investigate whether mutp53s can directly interact with AMPKa,
we performed glutathione S-transferase (GST) pull-down assays
Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc. 5
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
(legend on next page)
6 Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc.
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
using either AMPKa2-6xHis expressed in bacteria (Figure 5B) or
in vitro transcribed and translated AMPKa (Figures 5C and 5D)
and purified bacterially expressed wild-type and mutant GSTp53 proteins (Figures 5A–5D). Similar to our in vivo results, the
GST pull-down assay demonstrated that mutp53s were indeed
able to directly interact with both AMPKa1 and AMPKa2 in vitro
(Figures 5B–5D) and that both the DNA-binding domain (DBD)
and the C terminus of p53 were required for p53-AMPKa interaction (Figures 5A and 5B). Interestingly, it appeared that all the GSTp53 fusion proteins tested, including GST fusion proteins of wtp53
and mutp53 E336X, that did not interact with AMPKa in vivo (Figures 4E, 4F, and 4I–4K) interacted with AMPKa directly in vitro in a
GST pull-down assay (Figure 5A–5D). Since p53 is a very sensitive
molecule and one amino acid change often converts wtp53 into a
mutp53 with a different conformation, the fusion protein tag may
significantly affect the p53 molecule. To minimize the impact of
the N-terminal GST tag (27KD) on p53 conformation and physical
properties, we further purified and used C-terminal 6xHis (1KD)tagged p53 proteins in an in vitro coimmunoprecipitation (coIP)
assay. The results indicated that wtp53 had a much weaker
binding affinity to AMPKa2 than did GOF mutp53s (Figure 5E).
This observation was confirmed with an enzyme-linked immunosorbent assay (ELISA)-based protein-binding assay (Figure S5A).
To test whether the direct binding of AMPKa can inhibit its
activation (a-Thr172 phosphorylation) by upstream kinases, we
carried out in vitro kinase assays using the active AMPK
upstream kinase, LKB1 or TGF-b-activated kinase-1 (TAK1),
and AMPKa1 or AMPKa2 substrates in the presence or absence
of purified GST-tagged cleaved wtp53 or different mutp53s. Under these experimental conditions, AMPKa-Thr172 was efficiently
phosphorylated by LKB1 or TAK1, and AMPKa phosphorylation
was strongly inhibited by mutp53s, whereas no inhibition was
observed when similar amounts of wtp53 or the N-terminal
deletion D161 mutant were used (Figures 5F, 5G, and S5B).
Given that both GST-wtp53 and GST-D161 bound to AMPKa
(Figures 5A–5D) but the GST-tagged cleaved wtp53 or D161 did
not inhibit AMPKa Thr172 phosphorylation (Figures 5F and 5G),
the binding itself may be insufficient to block the activation of
AMPK, and the inhibition of a-Thr172 phosphorylation by
upstream kinases may require an N-terminally intact p53 with
the ‘‘mutant molecular conformation.’’ In support of this notion,
our in vitro pull-down assay using different purified proteins
showed that although wtp53 and the mutants D161, R175H,
and D161-R175H all bound to AMPKa2, only the GOF mutant
R175H efficiently inhibited AMPKa2-LKB1 interaction (Figure 5H). Therefore, whereas DBD is required for p53-AMPKa
binding (Figures 5A and 5B), the N terminus of mutp53s is
responsible for blocking AMPKa-LKB1 interaction (Figure 5H).
Finally, since active AMPK is a heterotrimeric serine/threonine
protein kinase complex that contains not only an a subunit
(catalytic subunit) but also regulatory b and g subunits (Hardie
et al., 2012), after demonstrating that mutp53s could bind to
the a subunit and inhibit its Thr172 phosphorylation by upstream
kinases (Figures 5A–5H), we investigated whether mutp53s can
directly affect AMPK activation as well as its subsequent kinase
activity. To test this idea, we first cotransfected 293FT with GFPAMPKb1 and HA-AMPKa2, and then used Chromotek-GFPTrap agarose beads to pull down AMPK (a/b/g), which was
further dephosphorylated by phosphatase PP2C. Subsequently,
this dephosphorylated AMPK (inactive kinase) was used in a
linked LKB-mediated AMPKa-Thr172 phosphorylation/AMPK
activity assay in the presence or absence of different wtp53
and mutp53s (Figure 5I, left). The mutp53s R175H, G245C,
P151S, and E336X were also able to inhibit a-Thr172 phosphorylation and suppress activation of AMPK by LKB1, subsequently
suppressing its kinase activity (Figure 5I, right). In contrast,
wtp53 that bound to AMPKa weakly (Figure 5E) and the DDBD
mutp53 that did not bind to AMPKa (Figure 5B) exhibited either
marginal or no impact on a-Thr172 phosphorylation and AMPK
activation by LKB1 (Figure 5I, right).
Taken together, our results demonstrated that mutp53 directly
binds to the AMPKa subunit via the DBD, while its N terminus
blocks the interaction between AMPKa and upstream kinases,
leading to inhibition of AMPK activation and AMPK kinase
activity.
Inhibition of AMPK Rescues Both Decreased Invasive
Cell Growth and Impaired Aerobic Glycolytic Potential
Associated with Loss of GOF mutp53s
Consistent with the tumor-suppressing role of AMPK, knockdown of AMPKa1 in p53-deficient UMSCC1 and PCI-13 cells
promoted cell growth in 3D Matrigel (Figures S6A–S6D). To
further test the role of inhibition of AMPK activation in the GOF
mutp53-associated phenotype, we knocked down AMPKa in
mutp53 downregulated cells. While downregulation of mutp53
P151S in Tu138 cells resulted in decreased S6 phosphorylation in
response to glucose starvation, further knockdown of AMPKa2
in those cells by two independent shRNAs rescued the
decreased phosphorylation of ribosomal protein S6 due to the
loss of mutp53 P151S (Figure 6A).
Because ribosomal protein S6 is a critical downstream
mediator of mTOR regulation of protein synthesis and cell growth
(Zoncu et al., 2011), we expected that GOF mutp53s would
enhance cell growth (i.e., cell size) through S6 activation. Flow
cytometry analysis demonstrated that downregulation of
mutp53 P151S in Tu138 cells resulted in a slight but consistent
Figure 3. Downregulation of Endogenous GOF mutp53s Enhances Activation of AMPK Signaling in Response to Energy Stress in HNSCC
Cells and Breast Cancer SKBR3 Cells
(A–E) IB analyses of various stable cell lines in response to energy stress.
(F) IHC staining of p53, pAMPK, and pACC in tongue tumors of orthotopic xenografts derived from Tu138-Ctr and Tu138-shp53 cells. Tumors were harvested
30 days after 5 3 105 cells were injected into the tongues of nude mice. Shown is one representative staining of tumors from three different mice. (Right panel)
Quantitative data are represented as mean ± SD.
(G and H) IB analyses of various p53 shRNA cell lines rescued by expression of shRNA-resistant mutp53s in response to glucose starvation.
(I) Expression of shRNA-resistant mutp53 R175H rescued the invasive growth of p53 shRNA Detroit 562 cells in an inverted 3D Matrigel culture. Relative migrating
distance for each cell line from one of three representative experiments is shown in right panel. Data are represented as mean ± SD. See also Figure S3.
Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc. 7
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
Figure 4. Energy Stress Induces the Interaction of AMPK and mutp53s In Vivo
(A–C) Immunofluorescent stainings of mutp53s in UMSSC1 stable cells under different culture conditions as indicated. Hypoxia, 1% O2; normoxia
(reoxygenation), 21% O2.
(legend continued on next page)
8 Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc.
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
decrease in the cell size of both G1 and G2/M cells cultured in
low-glucose medium (0.5 mM glucose) (Figure 6B). More
important, the decrease in cell size induced by the loss of
mutp53 was rescued by knockdown of AMPKa2 expression in
Tu138 shp53RNA cells (Figure 6B). Similarly, downregulation of
AMPKa2 in Tu138 shp53RNA and Detroit 562 shp53RNA cells
rescued the impaired invasive cell growth associated with loss
of mutp53s in an in vitro 3D Matrigel culture assay, which
resulted in much larger and more disorganized tumor spheroids
with longer migration/invasion distances (Figures 6C–6E and
S6E–S6H). Furthermore, introducing an shRNA-resistant version
of AMPKa2 into mutp53- and AMPKa1&2-double knockdown
Tu138 and Detroit 562 cells by retroviral infection (Figures 6D
and S6G) reversed the phenotype, leading to smaller cell size
(Figure S6I) and decreased invasive cell growth with less cell
migration and invasion in an inverted 3D Matrigel culture (Figures
6E and S6H). Together, these results strongly suggested that
AMPK is an important downstream target of GOF mutp53s and
a critical mediator of the observed cell growth and invasive
phenotype associated with mutp53 GOF.
Phosphorylation of ACC and SREBP-1c by AMPK leads to
inhibition of ACC and SREBP-1c activity, resulting in increased
fatty acid oxidation and decreased lipogenesis (Hardie and
Pan, 2002; Li et al., 2011). Given the inhibitory role of GOF
mutp53s on AMPK, we expected that GOF mutp53s would
also regulate lipid metabolism. Consistent with this idea,
p53(R172H/H) MEFs treated with a combination of insulin, rosiglitazone, dexamethasone, and isobutylmethylxanthine, which
stimulates lipogenesis (Esfandiari et al., 2007), exhibited much
higher lipid levels than p53 / MEFs did as evaluated by oil red
O staining (Figure 6F).
Finally, we examined the role of mutp53s in aerobic glycolysis
and mitochondrial oxidative phosphorylation (OXPHOS), as
measured by the extracellular acidification rate (ECAR) and the
oxygen consumption rate (OCR) (Marshall et al., 2011). Although
downregulation of mutp53 had no obvious impact on OXPHOS
response (Figures S7A and S7B), it significantly impaired aerobic
glycolytic capacity in response to glucose stimulation in both
Detroit 562 and Tu138 cells (Figures 6G, 6H, S7C, and S7D).
Furthermore, an shRNA-resistant mutp53 P151S rescued the
impaired aerobic glycolytic capacity in Tu138 shp53RNA cells
(Figure S7D). More important, the impaired aerobic glycolytic
capacity due to the loss of mutp53 was partially rescued by
further knockdown of AMPKa (Figures 6G, 6H, and S7E). However, this rescue was reversed by expression of an shRNA-resistant AMPKa2 in mutp53- and AMPKa1&2-double knockdown
Tu138 cells, leading to decreased aerobic glycolytic capacity
in response to glucose stimulation (Figure 6H).
Taken together, these results suggested that GOF mutp53s
are involved in promoting the invasive growth and metabolism
of cancer cells, such as lipid synthesis or accumulation and
aerobic glycolysis, at least partly through the inhibition of AMPK.
GOF mutp53s Promote Tumor Growth and Inhibit AMPK
Activation in Orthotopic Xenografts and GOF mutp53
Genetically Engineered Mouse Models of HNSCC
In support of the tumor-suppressing role of AMPK, downregulation of AMPKa in p53-deficient PCI-13 cells (Figure S6A) resulted
in increased tumor growth in an orthotopic mouse oral cancer
model (Figure S7F). To further gain insight into mutp53 GOF activity in the same animal tumor model, we injected UMSCC1
stable cell lines expressing different WT and different mutp53s
into the tongues of nude mice. GOF mutp53s R175H, G245C,
and R282W, but not mutp53 E336X, significantly accelerated
tumor onset and promoted tumor growth when compared with
wtp53 and pBabe control (Figures 7A and S7G). Moreover, while
pBabe and wtp53 tumors appeared to grow in defined areas and
exhibited a more differentiated morphology, GOF mutp53 tumors
displayed a more aggressive phenotype and often infiltrated and
invaded into adjacent tissues (Figure 7B). Consistently, IHC
staining indicated that AMPK activation (AMPKa-Thr172 phosphorylation) was inhibited in tumors derived from UMSCC1
stable cell lines expressing GOF mutp53s R175H, G245C, and
R282W when compared with the controls (Figure 7C).
Finally, using a genetically defined mouse model in which
oral papillomas arise from the activation of endogenous oncogenic K-ras, in the presence or absence of additional GOF
mutp53R172H or loss-of-function (p53 / ) mutations (Acin et al.,
2011), we evaluated whether GOF mutp53s modulate AMPK
activation. pAMPK was highly expressed in the basal and suprabasal layers of the papilloma of p53 / tumors (black arrows),
whereas in GOF mutp53R172H/- papilloma, mutp53 happened
to be highly expressed in the same layers, but with decreased
levels of pAMPK (red arrows) (Figure 7D), suggesting that
expression of mutp53 R172H in tumors inhibits AMPK activation
and thereby promotes tumor progression.
DISCUSSION
In the current study, we have shown that GOF mutp53s negatively regulate the metabolic effects of AMPK signaling, leading
to increased lipid production, aerobic glycolysis, and invasive
(D) IP and IB analyses of 293FT cells cotransfected with Myc-p53 R175H and GFP-AMPKa2 or control pEGFP vector.
(E) IP and IB analyses of 293FT cells cotransfected with HA-AMPKa2 and Myc-tagged wtp53 or mutp53s. wtp53 has a proline residue at codon 72, whereas all the
mutp53s have an arginine residue at codon 72; thus, wtp53 and mutp53s migrate differently.
(F) IP and IB analyses of the interaction of AMPKa and mutp53 in different mutp53-expressing UMSCC1 stable cell lines. Cells were starved of glucose for 2 hr
before being harvested for IP.
(G) IP and IB analyses of the interaction of endogenous AMPKa and mutp53 P151S in Tu138 cells under subconfluent normal culture condition.
(H) IP and IB analyses of the interaction of endogenous AMPKa/b and mutp53 R175H in Detroit 562 cells.
(I) IP and IB analyses of the interaction of endogenous AMPKa and mutp53 in HNSCC cell lines. Cells were starved of glucose for 2 hr before being harvested
for IP.
(J and K) IP and IB analyses of the interaction of endogenous mutp53 and AMPKa in HNSCC cell lines in response to energy stress.
(L) IP and IB analyses of the interaction of endogenous mutp53 R175H and AMPKa/b/g in Detroit 562 cells in response to energy stress. See also Figure S4.
Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc. 9
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
Figure 5. mutp53s Directly Bind to the AMPKa Subunit and Inhibit AMPK Activation by Upstream Kinases
(A) Schematic presentation of wtp53 and mutp53 proteins used in the GST pull-down assay. DBD, DNA-binding domain.
(B) GST pull-down assay of the interaction between purified AMPKa2-6xHis and GST-wtp53/mutp53 proteins.
(C and D) GST pull-down assay of the interaction between in vitro-translated 35S-labeled AMPKa1 (C) or AMPKa2 (D) and the same amounts of purified GSTwtp53/mutp53 proteins. Shown is a representative result that was conducted simultaneously for both AMPKa1 and AMPKa2. Relative quantitative data (mean ±
SD) from two independent experiments were calculated from gel densitometry using NIH imaging software and are shown at top. The activity of R175H was used
as the referent (100%).
(E) In vitro coIP/IB assay using purified WT/mutp53-6xHis and AMPKa2-6xHis proteins.
(legend continued on next page)
10 Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc.
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
cell growth. Moreover, our results demonstrated that GOF
mutp53s, through their DBDs, preferentially bind to the AMPKa
subunit, whereas their N termini block AMPKa-LKB1 interaction,
leading to inhibition of AMPK activation. On the basis of these
results, we propose a working model whereby mutp53s can
gain functional activity (Figure 7E): GOF mutp53s with high
AMPKa-binding affinity gain oncogenic function to promote
cell growth and cancer cell metabolism through direct inhibition
of AMPK activation.
The majority of p53 missense mutations occur in the DBD,
which usually leads to loss of the specific DNA-binding activity
of p53 as a transcription factor that directly regulates gene
expression. In spite of this, while missense mutations disrupt
the specific p53 DNA-binding activity, they often cause significant alterations in p53 tertiary structure (Xu et al., 2011), which
in turn can cause changes in p53 function through altered protein-protein interactomes and/or altered regulation of gene
expression, contributing to mutp53 GOF properties (FreedPastor and Prives, 2012; Muller and Vousden, 2013; Solomon
et al., 2012). As an excellent example, our results demonstrated
a difference between wtp53 and mutp53s in terms of their ability
to bind AMPKa and inhibit AMPK activation both in vitro and
in vivo (Figures 2F, 4E, 4F, 4J, 5E–5I, 7C, and S3–S5), suggesting
an important molecular mechanism for the difference between
mutp53s and wtp53 in their abilities to interact with other proteins, regulate gene expression, and exert GOF properties.
Most mutp53 GOF mechanisms that have been proposed are
involved in transcriptional control of gene expression, in which
mutp53s function as either transcriptional coactivators or transcriptional corepressors to control the transcriptional machinery
that regulates gene expression (Freed-Pastor and Prives, 2012;
Muller and Vousden, 2013; Solomon et al., 2012). Among many
possible GOF mechanisms, dominant-negative effects over other
p53 family members, p63 and p73, were proposed (Li and Prives,
2007). Interestingly, TAp63 was reported to transcriptionally activate LKB1 and AMPKa2 expression in mouse fat, liver, and
muscle in a TAp63-knockout mouse model (Su et al., 2012),
raising the possibility that GOF mutp53s inhibit LKB1 and AMPKa
gene expression through dominant-negative effects over TAp63.
However, while our evidence demonstrated that GOF mutp53s
inhibit AMPK activation (Figures 2–7), our results using both
real-time reverse-transcriptase PCR (Figure S2C) and western
blotting (Figures 2D–2F and 3C) revealed that expression of
mutp53s had no significant impact on the expression of either
LKB1 or AMPKa or on other subunits of LKB1 and AMPK kinases
in HNSCC cells. Instead, all our evidence has indicated that GOF
mutp53s have a direct role in AMPK activation (Figures 4 and 5).
As another piece of evidence supporting the transcriptionindependent role of GOF mutp53s, our results showed that
although the C-terminal-truncated mutp53 E336X exhibited
both AMPKa-binding and -inhibitory activity in vitro (Figures
5A–5D, 5F, 5G, and 5I), it failed to efficiently bind AMPKa
(Figures 4E, 4F, 4I, and 4K) or inhibit AMPK activation (Figures
2A–2C) in vivo and also failed to promote tumor cell growth
(Figures 1D and S7G). The fact that E336X localized exclusively
in the nucleus (Figure 4B) suggested that the cytoplasmic localization of mutp53s is also important for their GOF activities.
Consistent with this notion, a previous study reported that
some mutp53 proteins localize in the cytoplasm and inhibit
autophagy (Morselli et al., 2008). More important, our results
showed that like AMPK, whose subcellular localization (e.g.,
the cytoplasmic localization) is influenced by cell density (Kodiha
et al., 2007), the subcellular localization of mutp53s was
subjected to the regulation of cell culture conditions such as
cell density (Figure 4C). This finding may be important to understanding the biological functions of mutp53s.
AMPK was recently identified to negatively regulate the
Warburg effect through inhibition of the hypoxia-induced factor
1 (HIF1) pathway (Faubert et al., 2013). Therefore, the inhibition
of AMPK by GOF mutp53s, which relieves the suppression of
HIF1 by AMPK, is expected to increase HIF1 protein expression
and thus lead to increased glucose influx and glycolysis. Indeed,
mutant p53s have been shown to promote HIF1 expression
(Kamat et al., 2007), and HIF1 has been found to be required
for thymic lymphomas to develop in a p53 mutant mouse model
(Bertout et al., 2009).
In summary, our evidence strongly supports a transcriptionindependent mechanism that is important to the GOF activity
of mutp53s. Moreover, since AMPK is a master cellular energy
sensor and a regulator of metabolic homeostasis, our demonstration that mutp53s are actively involved in the regulation of
cancer metabolism through direct inhibition of AMPK further
extends our understanding of the molecular mechanisms of
mutp53s in cancers. Our findings also provide an excellent
example of how a gene mutation in tumor cells can transform
an agonistic relationship (i.e., wtp53 activating AMPK signaling)
into an antagonistic relationship (i.e., mutp53 inhibiting AMPK).
In this case, p53 mutation transforms a signaling network with
tumor-suppressing functions into a network with oncogenic
potential to promote tumor growth and progression.
EXPERIMENTAL PROCEDURES
See Supplemental Experimental Procedures.
Establishment of Stable Cell Lines Over- or Underexpressing
Target Genes
Stable cell lines overexpressing a mutp53 or K-ras were obtained from
pooled cells after puromycin (1–2 mg/ml) selection of retrovirus-infected cells.
Lentiviral shRNA expression systems were used to stably knock down mutp53
or AMPKa expression.
(F and G) LKB1 (F) and TAK1 (G) in vitro kinase assays using 50 ng of purified GST-AMPKa1 as the substrate in the presence of the exact amount of bovine serum
albumin (BSA) (1–4 mg) or various purified GST-wtp53/mutp53 proteins as indicated on Coomassie blue gel for each reaction (30 ml of total assay volume). Kinase
assays were conducted in the presence of Xa factor to cleave the N-terminal GST tag of p53 proteins.
(H) N terminus of mutp53 R175H is required for blocking AMPKa2-LKB1 interaction. In vitro His pull-down assay used different purified AMPKa2, LKB1, and p53
proteins as indicated. Results and experimental procedure are summarized on left. Pull-down assay was conducted in the presence of Xa factor.
(I) Linked LKB1-mediated AMPKa-Thr172 phosphorylation/AMPK activity assay. The experimental procedure is shown at left and the results at right. Kinase assay
was conducted in the presence of Xa factor. Relative AMPK activity is represented as mean ± SD. See also Figure S5.
Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc. 11
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
Figure 6. Inhibition of AMPK Rescues the Decreased Invasive Cell Growth and Impaired Aerobic Glycolytic Potential Associated with Loss
of GOF mutp53s
(A) IB analysis of pAMPK, pACC, and pS6 ribosomal protein in various Tu138 stable cell lines in response to glucose deprivation. NT, nontarget.
(B) Flow cytometry analysis of cell size of various Tu138 stable cell lines. Cells were cultured in 0.5 mM glucose Dulbecco’s modified Eagle’s medium (DMEM)
overnight before being subjected to analysis.
(C) Impaired cell growth due to loss of GOF mutp53 in 3D Matrigel was rescued by inhibition of AMPK. Shown are representative images of 3D culture of Tu138
stable cell lines.
(D) IB analysis of various Tu138 stable cell lines in response to glucose deprivation.
(E) Inverted in vitro 3D Matrigel culture of various stable cell lines.
(F) Oil Red O staining of confluent p53 / and p53(R172H/H) MEFs cultured in the presence of insulin, rosiglitazone, dexamethasone, and isobutylmethylxanthine.
(Top) Oil red O-stained dishes. (Bottom) Staining was quantified by absorbance at OD495 nm. Data are means ± SD (n = 3).
(G and H) Glycolytic rate (ECAR) of various Detroit 562 (G) and Tu138 (H) stable cell lines in response to glucose stimulation. Cells were starved of glucose for 2 hr
prior to initiation of experiment. Drugs were added at time points A (glucose, 25 mM), B (oligomycin, 1 mM), C (FCCP, 300 nM), and D (rotenone and antimycin,
100 nM). Experiments were conducted in the absence of sodium pyruvate. Data are presented as averages of three to five measurements. Error bars represent
standard error of the mean (SEM). See also Figures S6 and S7.
12 Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc.
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
Figure 7. GOF mutp53s Promote Tumor
Growth and Inhibit AMPK activation in
Orthotopic Xenografts and GOF mutp53
Genetically Engineered Mouse Model of
HNSCC
(A) Oral tumor growth of UMSCC1 stable cell lines
expressing different wtp53 and mutp53s. 1 3 105
cells/mouse were injected into the tongue of nude
mice. Error bar, SD.
(B) Hematoxylin and eosin staining of tumor
samples from UMSCC1 wtp53 and mutp53 stable
cell lines injected into the tongues of nude mice.
(C) IHC staining of p53 and pAMPK in tongue
tumor samples from an orthotopic mouse model of
UMSCC1 wtp53 and mutp53 stable cell lines.
Error bar, SD. *p < 0.0001, n = 4.
(D) IHC staining of p53 and pAMPK in papilloma
samples from mice with the indicated genotypes.
Error bar, SD. *p < 0.0001, n = 4.
(E) A working model for GOF mutp53s, in which
wtp53 functions indirectly downstream of AMPK
and GOF mutp53s function directly upstream of
AMPK and inhibit its activation, leading to an
impaired metabolic checkpoint and increased
anabolic tumor growth and progression. See also
Figure S7.
ACKNOWLEDGMENTS
In Vitro Inverted Matrigel Culture
A total of 40,000 cells, mixed and embedded in 50 ml of BD-Matrigel (#354230;
BD Biosciences), were layered on the surface of a BD BioCoat control
insert (#354578; BD Biosciences) and incubated at 37 C for 30–45 min, allowing the Matrigel to solidify. An additional 200 ml of Matrigel was layered on the
surface of the solidified Matrigel. When all the Matrigel was solidified, inserts
were placed in a 24-well culture plate with 0.1 ml of serum-free culture medium
in each lower chamber. On the upper side of the inserts, 0.3–0.5 ml of culture
medium containing 20% fetal bovine serum (FBS) and 50 ng/ml epidermal
growth factor (EGF) was added (see also Supplemental Experimental
Procedures).
We thank Drs. Guillermina Lozano and Yoko Takahashi for providing p53 /
and p53(R172H/H) MEFs and SKBR3, UMSCC1, HN30, and Detroit 562 stable
cell lines. We also thank Drs. Kun-liang Guan, Jianhua Yang, and Xiaolu
Yang for providing the plasmids. This research was supported by the University Cancer Foundation via the Institutional Research Grant program at The
University of Texas MD Anderson Cancer Center to G.Z. (IRG Grant Fred
35608) and by NIH grants RO1 DE015344 to C.C.; CA133249 and
GM095599 to Z.S.; and RO1 DE14613, P50CA097007, and CA016672
to J.N.M.
Glycolysis and OXPHOS Measurements
Glycolysis and OXPHOS were measured as ECAR and OCR, respectively,
using Seahorse XF24 extracellular flux analyzers according to the manufacturer’s instructions.
Received: May 29, 2013
Revised: February 18, 2014
Accepted: April 7, 2014
Published: May 22, 2014
SUPPLEMENTAL INFORMATION
REFERENCES
Supplemental Information includes seven figures and Supplemental Experimental Procedures and can be found with this article at http://dx.doi.org/10.
1016/j.molcel.2014.04.024.
Acin, S., Li, Z., Mejia, O., Roop, D.R., El-Naggar, A.K., and Caulin, C. (2011).
Gain-of-function mutant p53 but not p53 deletion promotes head and neck
cancer progression in response to oncogenic K-ras. J. Pathol. 225, 479–489.
Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc. 13
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
Agrawal, N., Frederick, M.J., Pickering, C.R., Bettegowda, C., Chang, K., Li,
R.J., Fakhry, C., Xie, T.X., Zhang, J., Wang, J., et al. (2011). Exome sequencing
of head and neck squamous cell carcinoma reveals inactivating mutations in
NOTCH1. Science 333, 1154–1157.
Kamat, C.D., Green, D.E., Warnke, L., Thorpe, J.E., Ceriello, A., and Ihnat, M.A.
(2007). Mutant p53 facilitates pro-angiogenic, hyperproliferative phenotype in
response to chronic relative hypoxia. Cancer Lett. 249, 209–219.
Berkers, C.R., Maddocks, O.D., Cheung, E.C., Mor, I., and Vousden, K.H.
(2013). Metabolic regulation by p53 family members. Cell Metab. 18, 617–633.
Kodiha, M., Rassi, J.G., Brown, C.M., and Stochaj, U. (2007). Localization of
AMP kinase is regulated by stress, cell density, and signaling through the
MEK/ERK1/2 pathway. Am. J. Physiol. Cell Physiol. 293, C1427–C1436.
Bertout, J.A., Patel, S.A., Fryer, B.H., Durham, A.C., Covello, K.L., Olive, K.P.,
Goldschmidt, M.H., and Simon, M.C. (2009). Heterozygosity for hypoxia
inducible factor 1alpha decreases the incidence of thymic lymphomas in a
p53 mutant mouse model. Cancer Res. 69, 3213–3220.
Lang, G.A., Iwakuma, T., Suh, Y.A., Liu, G., Rao, V.A., Parant, J.M., ValentinVega, Y.A., Terzian, T., Caldwell, L.C., Strong, L.C., et al. (2004). Gain of
function of a p53 hot spot mutation in a mouse model of Li-Fraumeni syndrome. Cell 119, 861–872.
Budanov, A.V., and Karin, M. (2008). p53 target genes sestrin1 and sestrin2
connect genotoxic stress and mTOR signaling. Cell 134, 451–460.
Lee, C.W., Wong, L.L., Tse, E.Y., Liu, H.F., Leong, V.Y., Lee, J.M., Hardie, D.G.,
Ng, I.O., and Ching, Y.P. (2012). AMPK promotes p53 acetylation via phosphorylation and inactivation of SIRT1 in liver cancer cells. Cancer Res. 72,
4394–4404.
DeBerardinis, R.J., and Thompson, C.B. (2012). Cellular metabolism and
disease: what do metabolic outliers teach us? Cell 148, 1132–1144.
Esfandiari, F., You, M., Villanueva, J.A., Wong, D.H., French, S.W., and
Halsted, C.H. (2007). S-adenosylmethionine attenuates hepatic lipid synthesis
in micropigs fed ethanol with a folate-deficient diet. Alcohol. Clin. Exp. Res. 31,
1231–1239.
Faubert, B., Boily, G., Izreig, S., Griss, T., Samborska, B., Dong, Z., Dupuy, F.,
Chambers, C., Fuerth, B.J., Viollet, B., et al. (2013). AMPK is a negative regulator of the Warburg effect and suppresses tumor growth in vivo. Cell Metab.
17, 113–124.
Feng, Z., and Levine, A.J. (2010). The regulation of energy metabolism and the
IGF-1/mTOR pathways by the p53 protein. Trends Cell Biol. 20, 427–434.
Feng, Z., Hu, W., de Stanchina, E., Teresky, A.K., Jin, S., Lowe, S., and Levine,
A.J. (2007). 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.
Fogarty, S., and Hardie, D.G. (2010). Development of protein kinase activators:
AMPK as a target in metabolic disorders and cancer. Biochim. Biophys. Acta
1804, 581–591.
Freed-Pastor, W.A., and Prives, C. (2012). Mutant p53: one name, many proteins. Genes Dev. 26, 1268–1286.
Freed-Pastor, W.A., Mizuno, H., Zhao, X., Langerød, A., Moon, S.H.,
Rodriguez-Barrueco, R., Barsotti, A., Chicas, A., Li, W., Polotskaia, A., et al.
(2012). Mutant p53 disrupts mammary tissue architecture via the mevalonate
pathway. Cell 148, 244–258.
Goldstein, I., and Rotter, V. (2012). Regulation of lipid metabolism by p53—
fighting two villains with one sword. Trends Endocrinol. Metab. 23, 567–575.
Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next
generation. Cell 144, 646–674.
Hardie, D.G. (2011). AMP-activated protein kinase: an energy sensor that
regulates all aspects of cell function. Genes Dev. 25, 1895–1908.
Li, Y., and Prives, C. (2007). Are interactions with p63 and p73 involved in
mutant p53 gain of oncogenic function? Oncogene 26, 2220–2225.
Li, Y., Xu, S., Mihaylova, M.M., Zheng, B., Hou, X., Jiang, B., Park, O., Luo, Z.,
Lefai, E., Shyy, J.Y., et al. (2011). AMPK phosphorylates and inhibits SREBP
activity to attenuate hepatic steatosis and atherosclerosis in diet-induced
insulin-resistant mice. Cell Metab. 13, 376–388.
Luo, Z., Zang, M., and Guo, W. (2010). AMPK as a metabolic tumor suppressor: control of metabolism and cell growth. Future Oncol. 6, 457–470.
Marshall, M., van der Windt, D., Nicholls, E., Myers, H., and Dziedzic, K. (2011).
Radiographic thumb osteoarthritis: frequency, patterns and associations with
pain and clinical assessment findings in a community-dwelling population.
Rheumatology (Oxford) 50, 735–739.
Mihaylova, M.M., and Shaw, R.J. (2011). The AMPK signalling pathway coordinates cell growth, autophagy and metabolism. Nat. Cell Biol. 13, 1016–1023.
Morselli, E., Tasdemir, E., Maiuri, M.C., Galluzzi, L., Kepp, O., Criollo, A.,
Vicencio, J.M., Soussi, T., and Kroemer, G. (2008). Mutant p53 protein localized in the cytoplasm inhibits autophagy. Cell Cycle 7, 3056–3061.
Muller, P.A., and Vousden, K.H. (2013). p53 mutations in cancer. Nat. Cell Biol.
15, 2–8.
Okoshi, R., Ozaki, T., Yamamoto, H., Ando, K., Koida, N., Ono, S., Koda, T.,
Kamijo, T., Nakagawara, A., and Kizaki, H. (2008). Activation of AMP-activated
protein kinase induces p53-dependent apoptotic cell death in response to
energetic stress. J. Biol. Chem. 283, 3979–3987.
Olive, K.P., Tuveson, D.A., Ruhe, Z.C., Yin, B., Willis, N.A., Bronson, R.T.,
Crowley, D., and Jacks, T. (2004). Mutant p53 gain of function in two mouse
models of Li-Fraumeni syndrome. Cell 119, 847–860.
Oren, M., and Rotter, V. (2010). Mutant p53 gain-of-function in cancer. Cold
Spring Harb. Perspect. Biol. 2, a001107.
Hardie, D.G., and Pan, D.A. (2002). Regulation of fatty acid synthesis and
oxidation by the AMP-activated protein kinase. Biochem. Soc. Trans. 30,
1064–1070.
Sano, D., Xie, T.X., Ow, T.J., Zhao, M., Pickering, C.R., Zhou, G., Sandulache,
V.C., Wheeler, D.A., Gibbs, R.A., Caulin, C., and Myers, J.N. (2011). Disruptive
TP53 mutation is associated with aggressive disease characteristics in an
orthotopic murine model of oral tongue cancer. Clin. Cancer Res. 17, 6658–
6670.
Hardie, D.G., Ross, F.A., and Hawley, S.A. (2012). AMPK: a nutrient and energy
sensor that maintains energy homeostasis. Nat. Rev. Mol. Cell Biol. 13,
251–262.
Shackelford, D.B., and Shaw, R.J. (2009). The LKB1-AMPK pathway:
metabolism and growth control in tumour suppression. Nat. Rev. Cancer 9,
563–575.
Hawley, S.A., Davison, M., Woods, A., Davies, S.P., Beri, R.K., Carling, D., and
Hardie, D.G. (1996). Characterization of the AMP-activated protein kinase kinase from rat liver and identification of threonine 172 as the major site at which
it phosphorylates AMP-activated protein kinase. J. Biol. Chem. 271, 27879–
27887.
Solomon, H., Buganim, Y., Kogan-Sakin, I., Pomeraniec, L., Assia, Y., Madar,
S., Goldstein, I., Brosh, R., Kalo, E., Beatus, T., et al. (2012). Various p53
mutant proteins differently regulate the Ras circuit to induce a cancer-related
gene signature. J. Cell Sci. 125, 3144–3152.
He, G., Zhang, Y.W., Lee, J.H., Zeng, S.X., Wang, Y.V., Luo, Z., Dong, X.C.,
Viollet, B., Wahl, G.M., and Lu, H. (2014). AMP-activated protein kinase
induces p53 by phosphorylating MDMX and inhibiting its activity. Mol. Cell.
Biol. 34, 148–157.
Jones, R.G., Plas, D.R., Kubek, S., Buzzai, M., Mu, J., Xu, Y., Birnbaum, M.J.,
and Thompson, C.B. (2005). AMP-activated protein kinase induces a
p53-dependent metabolic checkpoint. Mol. Cell 18, 283–293.
Stommel, J.M., Marchenko, N.D., Jimenez, G.S., Moll, U.M., Hope, T.J., and
Wahl, G.M. (1999). A leucine-rich nuclear export signal in the p53 tetramerization domain: regulation of subcellular localization and p53 activity by NES
masking. EMBO J. 18, 1660–1672.
Stransky, N., Egloff, A.M., Tward, A.D., Kostic, A.D., Cibulskis, K., Sivachenko,
A., Kryukov, G.V., Lawrence, M.S., Sougnez, C., McKenna, A., et al. (2011).
The mutational landscape of head and neck squamous cell carcinoma.
Science 333, 1157–1160.
14 Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc.
MOLCEL 5043
Please cite this article in press as: Zhou et al., Gain-of-Function Mutant p53 Promotes Cell Growth and Cancer Cell Metabolism via Inhibition of AMPK
Activation, Molecular Cell (2014), http://dx.doi.org/10.1016/j.molcel.2014.04.024
Molecular Cell
Gain-of-Function Mutant p53 Inhibits AMPK
Su, X., Gi, Y.J., Chakravarti, D., Chan, I.L., Zhang, A., Xia, X., Tsai, K.Y., and
Flores, E.R. (2012). TAp63 is a master transcriptional regulator of lipid and
glucose metabolism. Cell Metab. 16, 511–525.
Tsou, P., Zheng, B., Hsu, C.H., Sasaki, A.T., and Cantley, L.C. (2011). A fluorescent reporter of AMPK activity and cellular energy stress. Cell Metab. 13,
476–486.
Vander Heiden, M.G., Cantley, L.C., and Thompson, C.B. (2009).
Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324, 1029–1033.
Ward, P.S., and Thompson, C.B. (2012). Metabolic reprogramming: a cancer
hallmark even warburg did not anticipate. Cancer Cell 21, 297–308.
Winder, W.W., Wilson, H.A., Hardie, D.G., Rasmussen, B.B., Hutber, C.A., Call,
G.B., Clayton, R.D., Conley, L.M., Yoon, S., and Zhou, B. (1997).
Phosphorylation of rat muscle acetyl-CoA carboxylase by AMP-activated protein kinase and protein kinase A. J. Appl. Physiol. 82, 219–225.
Xu, J., Reumers, J., Couceiro, J.R., De Smet, F., Gallardo, R., Rudyak, S.,
Cornelis, A., Rozenski, J., Zwolinska, A., Marine, J.C., et al. (2011). Gain of
function of mutant p53 by coaggregation with multiple tumor suppressors.
Nat. Chem. Biol. 7, 285–295.
Zoncu, R., Efeyan, A., and Sabatini, D.M. (2011). mTOR: from growth
signal integration to cancer, diabetes and ageing. Nat. Rev. Mol. Cell Biol.
12, 21–35.
Molecular Cell 54, 1–15, June 19, 2014 ª2014 Elsevier Inc. 15
MOLCEL 5043